i 
 
A Study of the Anatomy and Physiology 
of Sleep in the Rock Hyrax, Procavia 
capensis. 
 
 
 
Nadine GRAVETT 
 
 
 
This thesis is submitted in fulfilment of the requirements for the degree of Doctor of 
Philosophy 
 
 
 
 
 
 
 
 
SCHOOL OF ANATOMICAL SCIENCES 
UNIVERSITY OF THE WITWATERSRAND 
JOHANNESBURG 
 
July, 2011 
  ii 
DECLARATION 
 
I declare that this thesis is my own unaided work. It is being submitted for the Degree 
of Doctor of Philosophy in the University of the Witwatersrand, Johannesburg. It has 
not been submitted before for any degree or examination in any other University. 
 
_______________________________ 
(Signature of candidate)   
 
______________________day of__________________________2011 
  iii 
ABSTRACT 
The rock hyrax, Procavia capensis, is a social diurnal mammal that typically lives 
in colonies on rocky outcrops and is found throughout most parts of Southern Africa. 
The aim of this thesis was to describe the sleep phenomenology of the rock hyrax. By 
means of immunohistochemistry the location and distribution of the cholinergic, 
catecholaminergic, serotonergic, orexinergic, histaminergic, and the GABAergic 
systems were described. For the most part these systems and their terminal network 
distributions followed the general mammalian organisational plan; however, several 
features, potentially unique to the rock hyrax, were found. These include the presence 
of cholinergic neurons in the AD and AV nuclei of the dorsal thalamus, parvo- and 
magnocellular divisions of the cholinergic LDT and PPT nuclei. A dense orexinergic 
terminal network distribution was noted in the regions coincident with the AD 
nucleus, a feature only observed in other diurnal mammals. Parvalbumin neurons did 
not show any association to the sleep related nuclei, whereas calbindin and calretinin 
neurons were found in all sleep related areas, though with differing densities and 
some homogeneities. The physiological measurable parameters of sleep were 
recorded continuously for 72 h under both solitary and social conditions and 
compared to determine whether differences existed. The results revealed that no major 
differences existed between the social and solitary conditions, but sleep intensity and 
REM duration (particularly REM0) was more pronounced in the social condition. 
REM was ambiguous in these animals, and led to its subdivision into REM1 and 
REM0. It is possible that REM1 could be a form of low voltage slow wave sleep, but 
further investigation is required. If REM1 is a form of NonREM sleep it would imply 
that the rock hyrax has the lowest amount of REM sleep of any terrestrial mammal 
studied to date.  
  iv 
ACKNOWLEGEMENTS 
 
I would like to take this opportunity to thank my PhD supervisor Prof Paul Manger; it 
has been a huge honour and privilege working with you for the last five years. I will 
forever be grateful for all your invaluable advice, guidance and the opportunities you 
presented me with. It has been an awesome five years, thank you for everything you 
have done, your support, encouragement and fantastic attitude. It has been an 
unforgettable journey. Thank you. 
 
To my colleagues and partners in crime, Bugz, Jillani, Constance and Nina, it has 
been an honour to know and work with you guys. Thank you for all your support and 
encouragement through the crazy times. May you all be very successful in all you 
future endeavours. It has been fun. 
  
Thank you, Prof. Jerome Siegel and Dr. Oleg Lyamin, from UCLA, USA for your 
support, invaluable advice and guidance in the understanding and analysis of the sleep 
physiology. 
 
Thank you, Prof Kjell Fuxe, for your enthusiasm, assistance and advice in analysing 
the immunohistochemistry. 
 
Thank you Mr. Jason Hemingway for you assistance with the statistical analysis.  
 
Thanks to Mrs. Christopher Gravett, Simon Perry and Adhil Bhagwandin for their 
help with the capturing of the hyraxes. 
  v 
 
Thank you to Dr. Wim de Wet and the staff of Klein Kariba, Mrs. Andre Momberg, 
Conrad Momberg and the Labuschagne’s for allowing and helping with the capture of 
hyraxes. 
 
Thank you to the staff of the School of Anatomical Sciences, University of the 
Witwatersrand, Johannesburg, for all the technical and administrative support. 
 
Lastly, thanks to the staff of the Central Animal Services, University of the 
Witwatersrand, Johannesburg, for the surgical and technical support. 
  vi 
 DEDICATION  
 
This thesis is dedicated in loving memory to my dearest uncle, Willie Nortje  
 
My Mom (Connie), Dad (Keith) and Brother (Christopher). 
Thank you for all your support, love, guidance and encouragement but most of all 
thank you for never allowing me to quit and believing in me. Everything I am today is 
a direct result of your hard work and sacrifices. No amount of words could ever 
describe how much I appreciate everything you have done for me. I love you with all 
my heart and more than you would ever be able to comprehend. I could not have 
asked for a better family. I am truly blessed!  
 
To my family, my grannies, uncles, aunts and cousins. Thank you for your support, 
interest and love. I love you all. 
 
To Simon Perry, thank you for all your support, help and words of encouragement. 
You mean the world to me and I love you with all my heart. 
 
To my dear friend, Adhil Bhagwandin, you have surely made the last four years 
interesting to say the least. Thank you for all your help, support, and encouragement. 
Most importantly thank you for being such an awesome friend. You have made this a 
journey that I will never forget. 
 
 
 
  vii 
CONTENTS 
 
DECLARATION                  ii 
ABSTRACT                   iii 
ACKNOWLEDGMENTS               iv 
DEDICATION                  vi 
 
Chapter 1: Introduction       
1.1. The rock hyrax, Procavia capensis: An overview        1 
1.2. The sleep phenomenology of the rock hyrax, Procavia capensis    2 
1.3. Specific aims                  5 
1.4. Individual chapters                6 
1.4.1. Chapter 2                  6 
1.4.2. Chapter 3                  7 
1.4.3. Chapter 4                  8 
1.4.4. Chapter 5                  9 
1.4.5. Chapter 6                  10 
 
Chapter 2: Nuclear organization and morphology of cholinergic, putative 
catecholaminergic and serotonergic neurons in the brain of the rock hyrax, 
Procavia capensis 
2.1. Introduction                  11 
2.2. Materials and Methods               13 
2.3. Abbreviations                 17 
2.4. Results                   22 
  viii 
2.4.1. Cholinergic neurons               22 
2.4.1.1. Striatal cholinergic interneurons            23 
2.4.1.1.1. Nucleus accumbens              23 
2.4.1.1.2. Dorsal striatopallidal complex – caudate/putamen globus pallidus  23 
2.4.1.1.3. Islands of Calleja and olfactory tubercle         24 
2.4.1.2. Cholinergic nuclei of the basal forebrain         25 
2.4.1.2.1. Medial septal nucleus                25 
2.4.1.2.2. Diagonal band of Broca             25 
2.4.1.2.3. Nucleus basalis                26 
2.4.1.3. Diencephalic cholinergic nuclei            26 
2.4.1.3.1. Medial habenular nucleus             26 
2.4.1.3.2. Hypothalamic cholinergic nuclei           27 
2.4.1.3.3. Cholinergic neurons in the anterodorsal and anteroventral dorsal thalamic 
nuclei                     27 
2.4.1.4. Pontomesencephalic nuclei             28 
2.4.1.4.1. Parabigeminal nucleus              28 
2.4.1.4.2. Pedunculopontine tegmental nucleus (PPT) – magnocellular and 
parvocellular nuclei                 29 
2.4.1.4.3. Laterodorsal tegmental nucleus (LDT) – magnocellular and parvocellular 
nuclei                     30 
2.4.1.5. Cholinergic cranial nerve motor nuclei          31 
2.4.2. Putative catecholaminergic nuclei            33 
2.4.2.1. Olfactory bulb (A16)               34 
2.4.2.2. Diencephalic nuclei               34 
2.4.2.3. Midbrain nuclei                35 
  ix 
2.4.2.3.1. Ventral tegmental area nuclei (VTA, A10 complex)      35 
2.4.2.3.2. Substantia nigra nuclear complex (A9)         37 
2.4.2.3.3. Retrorubral nucleus (A8)              38 
2.4.2.4. Pontine nuclei – the locus coeruleus (LC) nuclear complex     38 
2.4.2.4.1. Medullary nuclei               40 
2.4.3. Serotonergic nuclei                41 
2.4.3.1. Rostral cluster                42 
2.4.3.1.1. Caudal linear nucleus (CLi)             42 
2.4.3.1.2. Supralemniscal nucleus (B9)             42 
2.4.3.1.3. Median raphe (MnR)               43 
2.4.3.1.4. Dorsal raphe nuclear complex (DR)           43 
2.4.3.2. Caudal cluster                45 
2.4.3.2.1. Raphe magnus nucleus (RMg)            45 
2.4.3.2.2. Rostral and caudal ventrolateral serotonergic medullary columns (RVL and 
CVL)                      45 
2.4.3.2.3. Raphe pallidus nucleus (RPa)             46 
2.4.3.2.4. Raphe obscurus nucleus (ROb)            46 
2.5. Discussion                  72 
2.5.1. Cholinergic neurons in the anterodorsal (AD) and anteroventral (AV) nuclei of 
the dorsal thalamus of the rock hyrax             73 
2.5.2. Magnocellular and parvocellular divisions/nuclei of the pedunculopontine and 
laterodorsal tegmental nuclei in the rock hyrax          74 
2.5.3. Increased numbers of preganglionic cholinergic neurons in the inferior 
salivatory nucleus (pIX) of the rock hyrax           77 
  x 
2.5.4. Lack of TH+ neurons in compact locus coeruleus (A6c) and decreased numbers 
of TH+ neurons in the locus coeruleus proper (A6d) of the rock hyrax    77 
2.5.5. Evolutionary and phylogenetic considerations         79 
 
Chapter 3: Distribution of orexin-A immunoreactive neurons and their terminal 
networks in the brain of the rock hyrax, Procavia capensis 
3.1. Introduction                  81 
3.2. Materials and Methods               84 
3.3. Abbreviations                 87 
3.4. Results                   91 
3.4.1. Orexin-A neuronal cell body distribution          91 
3.4.2. Orexin-A terminal network distribution          93 
3.4.2.1. Telencephalon                 93 
3.4.2.2. Diencephalon                 93 
3.4.2.3. Midbrain and Pons (Mesencephalon and Metencephalon)     94 
3.4.2.4. Medulla Oblongata (Myelencephalon)          95 
3.5. Discussion                  110 
3.5.1. Orexinergic neuronal distribution             110 
3.5.2. Orexinergic terminal network distribution          113 
3.5.3. Orexinergic innervation of the anterodorsal nucleus of the dorsal thalamus and 
diurnality?                    115 
 
Chapter 4: Solitary sleep in the rock hyrax, Procavia capensis 
4.1. Introduction                  117 
4.2. Materials and Methods               119 
  xi 
4.2.1. Surgical Procedure                120 
4.2.2. Sleep recording                 121 
4.2.3. Data analysis                 122 
4.3. Results                   124 
4.3.1. Physiological Data – State Definitions           125 
4.3.1.1. Total state times                126 
4.3.1.2. Number of episodes               127 
4.3.1.3. Duration of episodes               128 
4.3.1.4. REM periodicity and slow wave activity (SWS)       129 
4.3.2. Behavioural Data                130 
4.4. Discussion                  159 
4.4.1. Comparison to previous sleep studies in Hyraxes        162 
4.4.2. Comparison to sleep in other Afrotheria, Xenarthra and other mammals   162 
 
 
Chapter 5: Social contrasted with solitary sleep patterns in the rock hyrax, 
Procavia capensis 
5.1. Introduction                  168 
5.2. Materials and Methods               170 
5.2.1. Surgical Procedure                171 
5.2.2. Sleep recording                 172 
5.2.3. Data analysis                 173 
5.3. Results                   175 
5.3.1. Physiological Data                176 
5.3.1.1. Total state times                177 
  xii 
5.3.1.2. Number of episodes               179 
5.3.1.3. Duration of episodes               180 
5.3.1.4. REM periodicity and slow wave activity (SWA)       182 
5.3.2. Behavioural Data                183 
5.4. Discussion                  208 
5.4.1. Sleep fragmentation in the social condition         209 
5.4.2. Reduced sleep quotas, but better sleep quality in the social condition?  211 
5.4.3. The effects of social sleeping in the rock hyrax        213 
 
Chapter 6: The interrelations of the distribution of sleep associated nuclei and 
terminal networks in the brain of the rock hyrax, Procavia capensis 
6.1. Introduction                  215 
6.2. Materials and Methods               218 
6.3. Abbreviations                 221 
6.4. Results                   227 
6.4.1. Cholinergic nuclei                227 
6.4.2. Catecholaminergic nuclei              230 
6.4.3. Serotonergic nuclei                231 
6.4.4. Orexinergic nuclei                233 
6.4.5. Serotonergic terminal networks related to the sleep-wake nuclei    233 
6.4.6. Orexinergic terminal networks related to the sleep-wake nuclei    234 
6.4.7. Histaminergic terminal networks related to the sleep-wake nuclei   235 
6.4.8. GABAergic inhibitory interneurons related to the sleep-wake nuclei   236 
6.4.8.1. GABAergic interneurons and the cholinergic system      237 
6.4.8.2. GABAergic interneurons and the catecholaminergic system    239 
  xiii 
6.4.8.3. GABAergic interneurons and the serotonergic system      239 
6.4.9.4. GABAergic interneurons and the orexinergic system      240 
6.5. Discussion                  269 
6.5.1 The unique cholinergic and orexinergic features of the hyrax brain   269 
6.5.2 GABAergic interneurons, functional aspects of relevance      270 
6.5.3 GABAergic neurons and the anterior thalamic nuclei       271 
6.5.4 GABAergic neurons and the pontine cholinergic nuclei      272 
6.5.5 GABAergic neurons and the locus coeruleus complex       273 
6.5.6 GABAergic neurons and the serotonergic dorsal raphe complex    274 
6.5.7 Summary                  275 
 
Chapter 7: Concluding Remarks 
7.1. Conclusion                  276 
7.2. The way forward                 281 
 
8. References 
8.1. Chapter 1                   283 
8.2. Chapter 2                   287 
8.3. Chapter 3                   297 
8.4. Chapter 4                   309 
8.5. Chapter 5                   315 
8.6. Chapter 6                   319 
8.7. Chapter 7                   325 
 
Outputs form Thesis                 329 
  xiv 
List of figures 
 
Figure 2.1: Photographs of the rock hyrax brain          48 
Figure 2.2: Diagrams of coronal sections through the rock hyrax brain    50 
Figure 2.3: Diagrams of sagittal section through the rock hyrax brain     54 
Figure 2.4: Photomicrographic montage of the basal forebrain       56 
Figure 2.5: Photomicrographs of cholinergic nuclei         58 
Figure 2.6: Photomicrographs of cholinergic nuclei         60 
Figure 2.7: Photomicrographs of cholinergic nuclei         62 
Figure 2.8: Photomicrographs of catecholaminergic nuclei        64 
Figure 2.9: Photomicrographic montage of catecholaminergic nuclei    66 
Figure 2.10: Photomicrographic montage of serotonergic nuclei     68 
Figure 2.11: Photomicrographs of serotonergic nuclei        70 
 
Figure 3.1: Diagrams of coronal sections through the rock hyrax brain    96 
Figure 3.2: Photomicrograph montage of the hypothalamus      100 
Figure 3.3: Photomicrographs of orexinergic nuclei         102 
Figure 3.4: Photomicrographs of orexinergic terminal networks     104 
Figure 3.5: Photomicrographs of orexinergic terminal networks     106 
Figure 3.6: Photomicrographs of orexinergic terminal networks     108 
 
Figure 4.1: Examples of EEG and EMG polygraphs        131 
Figure 4.2: EEG and EMG polygraphs demonstrating the transitions    133 
Figure 4.3: Scatter plots of instantaneous heart rate (IHR)        135 
Figure 4.4: Diagram illustrating the spectral power of states      137 
  xv 
Figure 4.5: Hypnograms showing transitions occurring over a 24 h    139 
Figure 4.6: Hypnograms showing transitions occurring over a 24 h    141 
Figure 4.7: State transition probabilities            143 
Figure 4.8: Histograms of total time spent in each physiological state    145 
Figure 4.9: Histograms depicting REM periodicity         149 
Figure 4.10: Slow wave activity for the 72 h recording period       151 
Figure 4.11: Histograms of total time spent in each behavioural state    153 
 
Figure 5.1: Diagram illustrating the spectral power of states      186 
Figure 5.2: State transition probabilities            188 
Figure 5.3: Histograms of total time spent in each physiological state    190 
Figure 5.4: Histograms depicting REM periodicity         194 
Figure 5.5: Slow wave activity for the 72 h recording period       196 
Figure 5.6: Histograms of total time spent in each behavioural state    198 
Figure 5.7: Hypnograms showing transitions occurring over a 24 h    200 
Figure 5.8: Hypnograms showing transitions occurring over a 24 h    202 
 
Figure 6.1: Diagrams of coronal sections through the rock hyrax brain    241 
Figure 6.2: Photomicrographs of serotonergic terminal networks     249 
Figure 6.3: Photomicrographs of orexinergic terminal networks     251 
Figure 6.4: Photomicrographs of histaminergic terminal networks     253 
Figure 6.5: Photomicrographs of GABAergic neurons        255 
Figure 6.6: Photomicrographs of the anterior thalamic nuclei      257 
Figure 6.7: Photomicrographs of GABAergic neurons        259 
Figure 6.8: Photomicrographs of GABAergic neurons        261 
  xvi 
Figure 6.9: Photomicrographs of GABAergic neurons        263 
Figure 6.10: Photomicrographs of GABAergic neurons        265 
 
 
LIST OF TABLES  
 
Table 4.1: Total wake, sleep and REM times as a percentage of 24 hours for each 
individual animal as well as the species mean for the solitary condition.   155 
 
Table 4.2: Illustration of where statistically significant differences (p<0.05) were 
noted between the three recording days for each state with regards to the number of 
epochs, episodes and episode duration for each individual animal as well as for the 
group for the solitary condition.              157 
 
Table 5.1: Total wake, sleep and REM times as a percentage of 24 hours for each 
individual animal as well as the species mean for the social condition    204 
 
Table 5.2: Illustration of where statistically significant differences (p<0.05) were 
noted between the three recording days for each state with regards to the number of 
epochs, episodes and episode duration for each individual animal as well as for the 
group for the social condition.              206 
 
Table 6.1: Illustration of different the degrees of GABAergic neuronal and terminal 
network distribution in relation to the nuclei of the cholinergic, catecholaminergic, 
serotonergic and orexinergic systems associated with the sleep-wake cycle.  267 
 1
Chapter 1 
 Introduction 
 
1.1. The rock hyrax, Procavia capensis: An overview on the evolutionary history, 
ecology and behaviour 
The rock hyrax belongs to the cohort Afrotheria that consists of Proboscidea 
(elephant), Sirenia (manatee and dugong), Hyrocoidea (hyracoids), Macroscelidea 
(elephant shrew), Tubulidentata (aardvark), and Tenrecoidea (tenrecs and golden mole) 
(Rübsamen et al., 1982; Kleinschmidt et al., 1986; Carter and Enders, 2004; Greenwood 
et al., 2004; Hakeem et al., 2005; Kellogg et al., 2007; Tabuce, 20008). The rock hyrax is 
a member of the Procavidea family, which is the only living family within the 
Hyracoidea order (Bothma, 1971; Skinner and Smithers, 1990). The family consists of 
five different species grouped into three distinct genera; Procavia capensis (rock hyrax, 
one species), Heterohyrax (bush hyrax, one species) and Dendrohyrax (tree hyrax, three 
species) (Hoeck, 2010). According to the fossil record the appearance of the Afrotheria 
dates back to the Cretaceous with the origin of the Hyrocoidea order older than the early 
Eocene. The fossil record also suggests that this order dominated the Afro-Arabian 
continent as the most diverse herbivores (Tabuce, 2008).  
The rock hyrax is a small, agile, diurnal, herbivorous mammal that typically 
inhabits rocky outcrops or “koppies”, and they are widely distributed throughout Africa 
(Skinner and Chimimba, 2005; Hoeck, 2010). The average weight of the rock hyrax 
varies between 1.5 to 5.6 kg and they are not considered to be sexually dimorphic in 
terms of body mass, as body mass appears to be closely correlated to annual precipitation 
 2
(Klein and Cruz-Uribe, 1996).  Morphological characteristics of the rock hyrax include: 
rubbery plantigrade feet with abundant sweat glands, long vibrissae distributed across the 
body surface, a dorsal gland surrounded by a creamy-yellowish or black circle of hair, 
tusk-like upper incisors, complex digestive system, abdominal testes in males and a 
duplex uterus in females. A low metabolic rate and a poor ability to regulate body 
temperature are some of the physiological characteristic associated with the rock hyrax 
(Hoeck, 2010).  
Rock hyraxes are social animals and typically live in colonies consisting of four 
to sixty individuals (Olds and Shoshani, 1982; Bothma, 1971; Skinner and Smithers, 
1990). Depending on the size of the rocky outcrop these colonies can consist of more 
than one family group. Each family group consists of one territorial male, peripheral 
males, dispersing males, three to seven related females, and juveniles of both sexes 
(Hoeck, 2010). Sexual maturity is reached between 28 to 29 months in males and 16 to 
17 months in females, with seasonal mating being triggered by the photoperiod (Skinner 
and Chimimba, 2005). The average gestation period and number of offspring is 7.5 
months and 1 to 4 respectively. Common predators of the rock hyrax include eagles, 
leopards, jackals and snakes (Hoeck, 2010). 
 
1.2. The sleep phenomenology of the rock hyrax, Procavia capensis 
Many studies deal with the evolutionary history, ecology and behavior of the hyrax, 
but when it comes to the anatomy of the brain and the associated sleep phenomenology, 
very little previous data is available. Previous studies that have focused on the hyrax 
brain have been more comparative than specific. The hyrax brain has been classified as 
 3
being macrosmatic (Hoffmeister, 1967) and from a comparative point it has been found 
to be similar to those of ungulates (Flower and Lydekker, 1891). Furthermore, 
comparative studies between the hyrax and their relatives, elephants, have revealed that 
both these animals have a surprisingly large hippocampus, perhaps associated with a 
good memory and higher cognitive functions (Hakeem et al., 2005); however, a large 
void is still apparent in our understanding of the anatomy and associated physiology of 
the hyrax brain, particularly the structures involved in the production of the sleep-wake 
cycle. A study by Snyder in 1974 on sleep in Hyracoidea revealed that sleep in the rock 
could be classified as being polycyclic, that is when periods of sleep and waking are 
constantly interchanging with one another.  The study revealed that the amount of 
paradoxical and transitional sleep most closely resembled those of other small vegetarian 
mammals and ungulates respectively. There was also no significant difference in the 
distribution of the different sleep stages during the day and night however the study did 
note that feeding predominantly took place during the night. A study by Sherwood et al. 
(2009) describing the neocortical distribution and morphology of calbindin, calretinin, 
and parvalbumin interneurons in the atlantogenatans - which the rock hyrax is a member 
of – in general, reported that calretinin interneurons were less common in the neocortex 
compared to calbindin and parvalbumin. The study also concluded that extant 
afrotherians showed similar morphologies to the stem eutherian mammal in contrast to 
other lineages that favour a larger number of derived traits.     
The study of mammalian sleep evolution has sparked several debates and theories. 
Many theories are centered on phylogeny, and claim that below the class level sleep 
phenomenology may be constrained in its evolution, while others believe that this sleep 
 4
phenomena is purely adaptive brought about by specific selection pressures. However, a 
larger number of these studies on which the theories are based either focus on major 
phylogenetic transitions (e.g. Kavanau et al., 2002; Nicolau et al., 2000) or on 
phylogenetic unusual forms (e.g. Rattenborg et al., 2000; Lyamin et al., 2008) and while 
the adaptive theory might sound appealing it has been found that the factors (e.g. 
environmental, behavioral etc.) considered to be adaptive, only account for a small 
amount of the variance in the sleep phenomena (Siegel, 1995; 2003). Furthermore, 
current data and theories regarding mammalian sleep evolution have been obtained from 
studies comparing mammals of different sizes and orders; however, comparing mammals 
of different orders poses a problem. Several previous studies (reviewed by Manger, 2005) 
have pointed out that the neural systems implicated in the sleep-wake cycle are not 
consistent across different orders, thus by making such comparisons subject to false 
inferences regarding mammalian sleep evolution. Several trends such as the direct 
relationship between the amount of REM (rapid eye movement) sleep and the degree of 
immaturity at birth, as well as the inverse correlation between body size and total sleep 
time have been identified (Siegel, 2003). Yet another problem arises when making these 
deductions, comparisons between the sleep states and factors such as phylogenetic 
relationship, and brain and body size, are not clear-cut and fail to provide the necessary 
explanation as to what exactly changes in the sleep phenomena. Thus, no dependable 
predictability can be assigned to these trends.  
Another issue that needs to be addressed is the dependability of the predictability of 
the anatomy of the somnogenic system. The phenomenon of sleep has been described as 
a particular state of an organism that is hierarchical in its organization. It is made up of 
 5
three distinct yet closely intertwined parts; anatomical, physiological and behavioral. The 
anatomical basis of sleep is comprised of the complex of neurons in the ventral portion of 
the brain, the somnogenic system; the physiological basis is distinguished by the global 
neural activity of the brain and the behavioral aspect characterized by non-responsiveness 
and motoric inactivity. Thus, in order to make correct inferences about evolutionary 
changes in the physiological and behavioral parameters of sleep one requires 
comprehensive insight on the topic of anatomical changes in the somnogenic system. The 
potential predictability of the neuronal systems controlling sleep is a major preliminary 
finding driving the present study (Manger, 2005). Previous studies have shown that a 
distinct evolutionary trend is present in the nuclear complexity of neural systems 
controlling sleep (Manger et al., 2002a-c, 2003). This raises the question of whether 
knowledge of the anatomy of the somnogenic system as well as that of the brain can 
allow accurate predictions of the physiological measurable parameters of sleep. 
 
1.3. Specific aims 
The practical aims of this thesis are twofold: (1) to provide a complete description of 
the subcortical structures of the brain of the rock hyrax that comprise the somnogenic 
system; (2) to determine what may be considered the normal pattern of sleep in the hyrax 
under both solitary and social conditions. This is followed by a theoretical aim, which 
will attempt to place these findings in the broader picture of sleep evolution from an 
adaptive and structuralist standpoint. 
 
 
 6
1.4. Individual chapters 
1.4.1. Chapter 2: Nuclear organization and morphology of cholinergic, putative 
catecholaminergic and serotonergic neurons in the brain of the rock hyrax, 
Procavia capensis. 
The nuclear subdivisions of the cholinergic, putative catecholaminergic and 
serotonergic systems within the brain of the rock hyrax (Procavia capensis) were 
identified following immunohistochemistry for acetylcholinesterase, tyrosine 
hydroxylase and serotonin.  The aim of the study was to investigate possible differences 
in the complement of nuclear subdivisions of these systems by comparing those of the 
rock hyrax to published studies of other mammals. For the most part, the nuclear 
organization of these three systems closely resembled that described for many other 
mammalian species.  The nuclear organization of the serotonergic system was identical to 
that seen in all eutherian mammals. The nuclear organization of the putative 
catecholaminergic system was very similar to that seen in rodents except for the lack of a 
C3 nucleus and the compact division of the locus coeruleus (A6c). In addition, the diffuse 
locus coeruleus (A6d) appeared to contain very few tyrosine hydroxylase 
immunoreactive (TH+) neurons.  The cholinergic system showed many features in 
common with that seen in both rodents and primates; however, there were three 
differences of note: (1) cholinergic neurons were observed in the anterior nuclei of the 
dorsal thalamus; (2) cholinergic parvocellular nerve cells, probably representing 
interneurons, forming subdivisions of the laterodorsal and pedunculopontine tegmental 
nuclei were observed at the midbrain/pons interface; and (3) a large number of 
cholinergic nerve cells in the periventricular gray of the medulla oblongata were 
 7
observed.  Thus, while there are many similarities to other mammalian species, the 
nuclear organization of these systems in the rock hyrax show specific differences to what 
has been observed previously in other mammals.  
 
1.4.2. Chapter 3: Distribution of orexin-A immunoreactive neurons and their 
terminal networks in the brain of the rock hyrax, Procavia capensis. 
This chapter describes the distribution of orexin-A immunoreactive neurons and 
terminal networks in relation to the previously described catecholaminergic, cholinergic 
and serotonergic systems within the brain of the rock hyrax, Procavia capensis. Adult 
female rock hyrax brains were sectioned and immunohistochemically stained with an 
antibody to orexin-A. The staining revealed that the neurons were mainly located within 
the hypothalamus as with other mammals. The orexinergic terminal network distribution 
also resembled the typical mammalian plan. High-density orexinergic terminal networks 
were located within regions of the diencephalon (e.g. paraventricular nuclei), midbrain 
(e.g. serotonergic nuclei) and pons (locus coeruleus), while medium density orexinergic 
terminal networks were evident in the telencephalic (e.g. basal forebrain), diencephalic 
(e.g. hypothalamus), midbrain (e.g. periaqueductal grey matter), pontine (e.g. 
serotonergic nuclei) and medullary regions (e.g. serotonergic and catecholaminergic 
nuclei). Although the distribution of the orexinergic terminal networks was typically 
mammalian, the rock hyrax did show one atypical feature, the presence of a high-density 
orexinergic terminal network within the anterodorsal nucleus of the dorsal thalamus 
(AD). The dense orexinergic innervation of the AD nucleus has only been reported 
previously in the Nile grass rat, Arvicanthis niloticus and Syrian hamster, Mesocricetus 
 8
auratus, both diurnal mammals in nature (Syrian hamster is also known to be nocturnal 
and hibernating in captivity). It is possible that orexinergic innervation of the AD nucleus 
might be a unique feature associated with diurnal mammals. It was also noted that the 
dense orexinergic innervation of the AD nucleus coincided with previously identified 
cholinergic neurons and terminal networks in this particular nucleus of the rock hyrax 
brain. It is possible that this dense orexinergic innervation of the AD nucleus in the brain 
of the rock hyrax may act in concert with the cholinergic neurons and/or the cholinergic 
axonal terminals, which in turn may influence arousal states and motivational processing. 
 
1.4.3. Chapter 4: Solitary sleep in the rock hyrax, Procavia capensis. 
It is well known that the rock hyrax is a social animal that typically lives in 
colonies consisting of few to many individuals. In this chapter, the focus is on the effects 
of solitary sleep in these animals. The rock hyraxes used in this study were captured from 
wild populations at random. A total of three hyraxes were captured at a time and housed 
together in specific recording enclosures. They were allowed to acclimatize for a period 
of one month before any experimentation commenced. One animal was selected from the 
group for recording purposes and implanted with a telemetric recording device. The 
implanted animal was allowed to recover from surgery for a minimum period of one 
week and placed in a recording enclosure by itself. The physiological parameters of sleep 
as well as the accompanying behavior were subsequently recorded continuously for 72 
hours. The aim of the study was to determine the amount of time, as a percentage of 24 
hours, spent in each sleep state, i.e. SWS (slow wave sleep) and REM (rapid eye 
movement), as well as the number and average duration of SWS and REM epochs. 
 9
Furthermore, the study also aimed to determine the impact of sleeping “alone” as 
opposed to sleeping in a group. The questions to be answered was whether a naturally 
social animal being placed in under solitary conditions would show an increase, a 
decrease or no change in the total amount of time spent in each sleep state as well as the 
number and average duration of SWS and REM epochs. 
 
1.4.4. Chapter 5: Social contrasted with solitary sleep conditions in the rock hyrax, 
Procavia capensis. 
In this chapter the difference in social sleep versus solitary sleep in the rock hyrax 
is investigated. Following 72 hours of solitary sleep recording, the non-implanted animals 
were reunited with the implanted animal and allowed to familiarize themselves with each 
other for a period of three days after which 72 hours continuous social sleep recording 
commenced. The data was analyzed and the total amount of time, as a percentage of 24 
hours, spent in each sleep state, i.e. SWS (slow wave sleep) or REM (rapid eye 
movement), as well as the number and average duration of SWS and REM epochs, was 
determined. The social sleep data was compared to solitary sleep data to determine 
whether any differences existed. Furthermore, the behavior of each of the non-implanted 
hyraxes were analyzed and correlated with each other as well as the implanted hyrax, to 
determine whether all hyraxes were in the same state at the same time or whether their 
behavioral states differed from one another. The aim of the study was to determine the 
correlation between social and solitary sleep in the rock hyrax, with regard to total 
amount of sleep time as well as the number and average duration of SWS and REM 
 10
epochs. What effect does the social setting have on the duration, intensity and quality of 
sleep in the rock hyrax and is it significantly different to the solitary scenario?  
  
1.4.5. Chapter 6: Nuclear organization and morphology of GABAergic and 
histaminergic neurons in the brain of the rock hyrax, Procavia capensis, and 
their relation and interplay with the cholinergic, putative catecholaminergic 
and serotonergic neurons. 
 This study examined the interrelations of the distribution of sleep associated 
nuclei and terminal networks in the brain of the rock hyrax, Procavia capensis. These 
systems were identified following immunohistochemistry. A one in ten series of stains 
was made for nissl, myelin, choline acetyltransferase (ChAT) (identification of the 
cholinergic system), tyrosine hydroxylase (TH) (identification of the dopaminergic and 
noradrenergic systems), serotonin (5HT) (identification of the serotonergic system), 
orexin A (identification of the orexinergic system), parvalbumin (identification of a 
subset of the GABAergic system), calretinin (identification of a subset of the GABAergic 
system), calbindin (identification of a subset of the GABAergic system) and histamine 
(identification of the histaminergic system). The aim of the study was to describe the 
general organisation and distribution of the GABAergic and histaminergic systems in the 
brain of the rock hyrax and how it correlates to previous published data of these systems 
in other mammals, and secondly to examine the relation and interplay of the GABAergic 
interneurons to the cholinergic, dopaminergic and noradrenergic, serotonergic, and 
orexinergic systems. 
 11
Chapter 2 
Nuclear organization and morphology of cholinergic, putative catecholaminergic 
and serotonergic neurons in the brain of the rock hyrax, Procavia capensis. 
 
2.1. Introduction 
The rock hyrax, Procavia capensis, is a member of the Procavidea family, which 
is the only living family within the Hyracoidea order (Klein and Cruz-Uribe, 1996). The 
family consists of three genera: Procavia (the rock hyrax, 1 species), Heterohyrax (two 
species) and Dendrohyrax (three species) (Skinner and Chimimba, 2005). The rock hyrax 
is a small, agile, diurnal, social, herbivorous mammal, typically weighing between 2.5 
and 4.6 kg, and lives in rocky outcrops or “koppies” in most parts of Africa (Skinner and 
Chimimba, 2005). Sexual maturity is reached between 28 to 29 months in males and 16 
to 17 months in females, with seasonal mating being triggered by the photoperiod 
(Skinner and Chimimba, 2005). According to the fossil record, hyraxes first appeared 
approximately forty million years ago and they are grouped phylogenetically with the 
African elephant (Proboscidea), and manatee and dugong (Sirenia).  These three orders, 
together with the Chrysochloridea (golden moles), Macroscelidea (elephant shrews), 
Tubulidentata (aardvarks) and the Tenrecidea (tenrecs) form the cohort Afroplacentalia 
(Arnason et al., 2008). 
In the present study the cholinergic, putative catecholaminergic and serotonergic 
systems within the brain of the rock hyrax were examined and described using 
immunohistochemical techniques. It is well known that these systems project to most 
parts of the brain and that they are associated with several functions (Woolf, 1991; 
 12
Smeets and González, 2000; Törk, 1990; Jacobs and Azmitia, 1992); for example, 
cognition (e.g. Bartus et al., 1982; Previc, 1999), the sleep-wake cycle (e.g. Siegel, 2006), 
reproduction (e.g. Tillet, 1995), and sensory-motor (e.g. Pompeiano, 2001; Fuxe et al., 
2007a) functions to name but a few. The cholinergic system has an extensive distribution 
throughout the brain (Woolf, 1991; Reiner and Fibiger, 1995; Manger et al., 2002a; 
Maseko et al., 2007) while the catecholaminergic and serotonergic neuronal systems are 
mainly concentrated within the brainstem (Dahlström and Fuxe, 1964; Andén et al., 
1964; Fuxe et al., 1969, 1970, 2006, 2007a; Diksic and Young, 2001; Manger et al., 
2002b,c; Maseko et al., 2007). The nuclear organization of these systems has been 
studied in several mammalian species (e.g. Maseko et al., 2007; Bhagwandin et al., 2008; 
Limacher et al., 2008); however no studies of these systems have been done in any 
member of the Afroplacentalia cohort. Thus, the aim of this study is to determine the 
nuclear organization of these systems in the brain of the rock hyrax and extend our basis 
for understanding the evolutionary processes associated with the nuclear organization of 
these systems. 
These systems have exhibited some evolutionary trends and even though these 
systems are quite similar across species for the most part, differences due occur. For 
example, the catecholaminergic C3 nucleus has only been reported to be present in the 
rodents (e.g. Smeets and González, 2000; Manger et al., 2002b; Maseko et al., 2007; 
Moon et al., 2007; Badlangana et al., 2007; Dwarika et al., 2008; Bhagwandin et al., 
2008; Limacher et al., 2008). Serotonergic neurons in the mammalian hypothalamus have 
only been reported for monotremes (Manger et al., 2002c). These examples and other 
evidence has led Manger (2005) to propose that members of the same order will show the 
 13
same complement of nuclei, but this complement may differ between orders. Some 
predictions of the current study include: (1) the rock hyrax will have many nuclei in 
common with other mammals; (2) the rock hyrax may have some nuclei unique to this 
species; (3) the rock hyrax may exhibit nuclei found only in members of the 
Afroplacentalia; and (4) the rock hyrax may be missing some nuclei commonly found in 
other mammalian species. 
 
2.2. Materials and Methods 
 A total of six adult female rock hyraxes, Procavia capensis, were used in the 
present study. Permits from the Limpopo and Gauteng Provincial Governments were 
obtained for the capture and transport of the animals from the wild. All animals were 
treated and used according to the guidelines of the University of the Witwatersrand 
Animal Ethics Committee (approval number AESC 2005/8/5), which parallel those of the 
NIH for the care and use of animals in scientific experimentation. Each animal was 
weighed, deeply anaesthetized and subsequently euthanized with weight appropriate 
doses of sodium pentobarbital (200mg sodium pentobarbital/kg, i.p.). Upon cessation of 
respiration the animals were perfused intracardially with 0.9% saline followed by 4% 
paraformaldehyde in 0.1M phosphate buffer (PB, pH 7.4) (approximately 1 l/kg of each 
solution), both solutions having a temperature of 4°C. The brains were then carefully 
removed from the skulls and post-fixed overnight in 4% paraformaldehyde in 0.1M PB 
followed by equilibration in 30% sucrose in 0.1M PB. The brains were then frozen and 
with the aid of a freezing microtome sectioned at 50 µm in either coronal (n = 4) or 
sagittal (n = 2) planes. A one in five series of stains was made for nissl, myelin, choline 
 14
acetyltransferase (ChAT) (identification of the cholinergic system), tyrosine hydroxylase 
(TH) (identification of the dopaminergic and noradrenergic systems), and serotonin 
(5HT) (identification of the serotonergic system). Sections kept for the Nissl series were 
mounted on 0.5% gelatine coated glass slides, cleared in a solution of 1:1 chloroform and 
absolute alcohol, then stained with 1% cresyl violet to reveal cell bodies. Myelin sections 
were stored in 5% formalin for a period of two weeks and were then mounted on 1.5% 
gelatine coated glass slides and subsequently stained with silver solution to reveal myelin 
sheaths (Gallyas, 1979). 
  For immunohistochemical staining each section was treated with endogenous 
peroxidase inhibitor (49.2% methanol: 49.2% 0.1 M PB: 1.6% of 30% H2O2) for 30 min 
and subsequently subjected to three 10 min 0.1 M PB rinses. The sections were then 
preincubated in a solution (blocking buffer) consisting of 3% normal serum (normal 
rabbit serum, NRS, for the ChAT sections and normal goat serum, NGS, for the TH and 
5HT sections), 2% bovine serum albumin (BSA, Sigma) and 0.25% Triton X100 (Merck) 
in 0.1M PB, at room temperature for 2 h. This was followed by three 10 min rinses in 
0.1M PB. The sections were then placed, for 48 h at 4°C under constant gentle shaking, 
in primary antibody solution that contained the appropriately diluted primary antibody in 
blocking buffer (see above). The primary antibodies used were anti-choline 
acetyltransferase for cholinergic neurons (AB144P, Chemicon, raised in goat, at a 
dilution of 1:2000), anti-tyrosine hydroxylase for putative catecholaminergic neurons 
(AB151, Chemicon, raised in rabbit, at a dilution of 1:7500), and anti-serotonin for 
serotonergic neurons (AB938, Chemicon, raised in rabbit, at a dilution of 1:10000). This 
was followed by another three 10 min rinses in 0.1M PB, after which the sections were 
 15
incubated for 2 h at room temperature in secondary antibody solution. The secondary 
antibody solution contained a 1:750 dilution of biotinylated anti-rabbit IgG (BA-1000, 
Vector Labs) in 3% NGS (or anti-goat IgG, BA-5000 in 3% NRS for the ChAT sections), 
and 2% BSA in 0.1M PB. Once this was completed, the sections were again subjected to 
another three 10 min rinses in 0.1M PB, followed by a 1 h incubation in AB solution 
(Vector Labs) and again rinsed. This was followed by a 5 min treatment of the sections in 
a solution consisting of 0.05% diaminobenzidine (DAB) in 0.1M PB, after which, and 
while still in the same solution, 3µl of 30% H2O2 per 0.5ml of solution was added. With 
the aid of a low power stereomicroscope the progression of the staining was visually 
followed and allowed to continue until a level was reached where the background 
staining could assist in reconstruction without obscuring the immunopositive neurons. 
Once this level was reached the reaction was stopped by placing the sections in 0.1M PB, 
followed by a final session of three 10 min rinses in 0.1M PB. All solutions used in the 
immunohistochemical process had a pH of 7.4. The immunohistochemically stained 
sections were mounted on 0.5% gelatine coated slides and left to dry overnight. The 
mounted sections were dehydrated by placing it in 70% alcohol for 2 h at room 
temperature under gentle shaking and then transferred through a series of graded 
alcohols, cleared in xylene and coverslipped with Depex.  
The sections were observed with a low power stereomicroscope, and the 
architectonic borders traced according to the Nissl and myelin stained sections using a 
camera lucida. The corresponding immuno-stained sections were then matched to the 
drawings and the immuno-positive neurons marked. The drawings were scanned and 
redrawn with the aid of the Canvas 8 program. The drawings were made from sections 
 16
obtained from one brain, while photomicrographs were taken from representative 
sections of this and the five remaining brains. The nomenclature used for the cholinergic 
system was adopted from Woolf, (1991), Manger et al. (2002a), Maseko and Manger 
(2007), Maseko et al. (2007), Limacher et al. (2008) and Bhagwandin et al. (2008); the 
catecholaminergic system from Dahlström and Fuxe (1964), Hökfelt et al. (1984), Smeets 
and González (2000), Manger et al. (2002b), Maseko and Manger (2007), Maseko et al. 
(2007), Moon et al. (2007), Dwarika et al. (2008), Limacher et al. (2008) and 
Bhagwandin et al. (2008); and for the serotonergic system from Törk, (1990), Bjarkam et 
al. (1997), Manger et al. (2002c), Maseko and Manger (2007), Maseko et al. (2007), 
Moon et al. (2007), Dwarika et al. (2008), Limacher et al. (2008) and Bhagwandin et al. 
(2008).  While we use the standard nomenclature for the catecholaminergic system in this 
paper, we realize that the neuronal groups we revealed with tyrosine hydroxylase 
immunohistochemistry may not correspond directly with those nuclei that have been 
described in previous studies by Dahlström and Fuxe (1964), Hökfelt et al. (1976), 
Meister et al. (1988), Kitahama et al. (1990, 1996), and Ruggiero et al. (1992); however, 
given the striking similarity of the results of the tyrosine hydroxylase 
immunohistochemistry to that seen in other mammals we feel this terminology is 
appropriate.  Clearly further studies in the rock hyrax with a wider range of antibodies, 
such as those to phenylethanolamine-N-methyltransferase (PNMT), dopamine-β-
hydroxylase (DBH) and aromatic L-amino acid decarboxylase (AADC) would be 
required to fully determine the implied homologies ascribed in this study.  We address 
this potential problem with the caveat of putative catecholaminergic neurons where 
appropriate in the text. 
 17
 
 
2.3. Abbreviations 
III – oculomotor nucleus 
IV – trochlear nucleus 
Vmot – motor division of trigeminal nucleus 
VI – abducens nucleus 
VIId – facial nerve nucleus, dorsal division 
VIIv – facial nerve nucleus, ventral division 
X – dorsal motor vagus nucleus 
XII – hypoglossal nucleus 
3V – third ventricle 
4V – fourth ventricle 
7n – facial nerve 
A1 – caudal ventrolateral medullary tegmental nucleus 
A2 – caudal dorsomedial medullary nucleus 
A4 – dorsal medial division of locus coeruleus 
A5 – fifth arcuate nucleus 
A6c – compact portion of locus coeruleus 
A6d – diffuse portion of locus coeruleus 
A7d – nucleus subcoeruleus, diffuse portion 
A7sc – nucleus subcoeruleus, compact portion 
A8 – retrorubral nucleus 
 18
A9l – substantia nigra, lateral 
A9m – substantia nigra, medial 
A9pc – substantia nigra, pars compacta 
A9v – substantia nigra, ventral or pars reticulata 
A10 – ventral tegmental area 
A10c – ventral tegmental area, central 
A10d – ventral tegmental area, dorsal 
A10dc – ventral tegmental area, dorsal caudal 
A11 – caudal diencephalic group 
A12 – tuberal cell group 
A13 – zona incerta 
A14 – rostral periventricular nucleus 
A15d – anterior hypothalamic group, dorsal division 
A15v – anterior hypothalamic group, ventral division 
A16 – catecholaminergic neurons of the olfactory bulb 
ac – anterior commissure 
AD – anterodorsal nucleus of the dorsal thalamus 
Amyg – amygdala 
AP – area postrema 
AV – anteroventral nucleus of the dorsal thalamus 
B9 – supralemniscal serotonergic nucleus 
C – caudate nucleus 
C1 – rostral ventrolateral medullary tegmental group 
 19
C2 – rostral dorsomedial medullary nucleus 
ca – cerebral aqueduct 
Cb – cerebellum 
cc – corpus callosum 
Cl – claustrum 
CLi – caudal linear nucleus 
CN – cochlear nucleus 
C/P – caudate and putamen nuclei 
CP – cerebral peduncle 
CVL – caudal ventrolateral serotonergic group 
DCN – deep cerebellar nuclei 
Diag.B – diagonal band of Broca 
DR – dorsal raphe 
DRc – dorsal raphe nucleus, caudal division 
DRd – dorsal raphe nucleus, dorsal division 
DRif – dorsal raphe nucleus, interfascicular division 
DRl – dorsal raphe nucleus, lateral division 
DRp – dorsal raphe nucleus, peripheral division 
DRv – dorsal raphe nucleus, ventral division 
DT – dorsal thalamus 
EW – Edinger-Westphal nucleus 
f – fornix 
GC – periaqueductal grey matter 
 20
GLD – dorsal lateral geniculate nucleus 
GP – globus pallidus 
Hbm – medial habenular nucleus 
Hip – hippocampus 
Hyp – hypothalamus 
Hyp.d – dorsal hypothalamic cholinergic nucleus 
Hyp.l – lateral hypothalamic cholinergic nucleus 
Hyp.v – ventral hypothalamic cholinergic nucleus 
IC – inferior colliculus 
ic – internal capsule 
icp – inferior cerebellar peduncle 
io – inferior olivary nuclei 
IP – interpeduncular nucleus 
LDTmc – magnocellular division of the laterodorsal tegmental nucleus 
LDTpc – parvocellular division of the laterodorsal tegmental nucleus 
LRT – lateral reticular nucleus 
LV – lateral ventricle 
mcp – middle cerebellar peduncle 
MnR – median raphe nucleus 
N.Acc – nucleus accumbens 
N.Amb – nucleus ambiguus 
N.Bas – nucleus basalis 
NEO – neocortex 
 21
OB – olfactory bulb 
OC – optic chiasm 
OT – optic tract 
P – putamen 
pVII – preganglionic motor neurons of the superior salivatory nucleus or facial nerve 
pIX – preganglionic motor neurons of the inferior salivatory nucleus 
PBg – parabigeminal nucleus 
PIR – piriform cortex 
PPTmc – magnocellular division of the pedunculopontine nucleus 
PPTpc – parvocellular division of the pedunculopontine nucleus 
Pta – pretectal area 
py – pyramidal tract 
pyx – decussation of the pyramidal tract 
R – thalamic reticular nucleus 
Rmc – red nucleus, magnocellular division 
RMg – raphe magnus nucleus 
ROb – raphe obscurus nucleus 
RPa – raphe pallidus nucleus 
RVL – rostral ventrolateral serotonergic group 
S – septum 
SC – superior colliculus 
scp – superior cerebellar peduncle 
Sep.M – medial septal nucleus 
 22
TOL – olfactory tubercle 
TOL/Is.Call. – olfactory tubercle/Island of Calleja 
vh – ventral horn of spinal cord 
VPO – ventral pontine nucleus 
xscp – decussation of the superior cerebellar peduncle 
ZI – zona incerta 
 
2.4. Results 
The present study was designed to reveal the nuclear organization of the 
cholinergic (ChAT), putative catecholaminergic (TH), and serotonergic systems of the 
rock hyrax, Procavia capensis through immunohistochemical methods. A total of six 
adult female brains were used for this purpose, and the individuals used for this study had 
body masses ranging from 1.14 kg to 1.52 kg and brain masses between 14.4 g and 17.5 g 
(Fig. 2.1). The results revealed that for the most part these systems do not differ 
drastically from those observed in other mammalian species; however the rock hyrax 
does show some unique additional nuclei, specifically within the cholinergic system, that 
have not been noted in other species. 
 
2.4.1. Cholinergic neurons 
The general organization of the cholinergic system encompasses the striatal, basal 
forebrain, diencephalic, and pontomesencephalic groups together with the cranial nerve 
motor nuclei that extend from level of the anterior horn of the lateral ventricle to the 
spinomedullary junction (e.g. Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007).  
 23
These groups were all present in the brain of the rock hyrax and did not show any 
significant differences to the general mammalian group level organization of this system 
(Figs. 2.2, 2.3). Despite this, some novel features, potentially unique to the rock hyrax, 
that have not been observed in other mammals were observed.  These include the 
existence of cholinergic neurons in the anterior nuclei of the dorsal thalamus, the 
existence of magnocellular and parvocellular divisions of both the laterodorsal tegmental 
and pedunculopontine tegmental nuclei, and a large cell group in the dorsomedial 
periventricular grey of the medulla oblongata (Figs. 2.2 – 7). 
 
2.4.1.1. Striatal cholinergic interneurons 
2.4.1.1.1. Nucleus accumbens 
Uniformly distributed choline acetyltransferase immunoreactive (ChAT+) nerve 
cell bodies were located ventrally and slightly anterior to the dorsal striatopallidal 
complex (caudate, putamen and the globus pallidus) (Figs. 2.2 C-F, 2.3). The anterior and 
posterior borders of this nucleus were adjacent to the anterior horn of the lateral ventricle 
and the anterior commissure respectively.  This arrangement is typical to what is 
observed in all mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007).  A 
moderate density of ChAT+ neurons were observed throughout this nucleus, these 
neurons being a mixture of bipolar and multipolar types, but the majority were of the 
multipolar type (Fig. 2.4). The ChAT+ neurons of this nucleus showed no specific 
dendritic organization.  
 
2.4.1.1.2. Dorsal striatopallidal complex – caudate/putamen and globus pallidus 
 24
The caudate/putamen nucleus was located lateral to the lateral ventricle and its 
distribution extended from the level of the anterior horn of the lateral ventricle anteriorly 
to the medial habenular nuclei posteriorly (Figs. 2.2 D-K, 2.3), a location typical of all 
mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007). The boundary 
between the caudate and putamen was clearly defined by the internal capsule at the level 
of the anterior commissure, but not anterior to this level. The location of the globus 
pallidus was found to be ventral and somewhat medial to the putamen and it extended 
from the level of the anterior commissure to the habenular nuclei. A moderate density of 
uniformly distributed ChAT+ neurons was observed within the caudate/putamen (Fig. 
2.4). The globus pallidus only exhibited a small number of ChAT+ neurons and these 
neurons were found to be located mostly where the globus pallidus bordered the putamen 
laterally and the nucleus basalis ventrally.  A similar neuronal morphology was noted for 
the caudate/putamen and the globus pallidus, the ChAT+ neurons being a mixture of 
bipolar and multipolar types, however a multipolar organization predominated.  No 
specific dendritic orientation was observed for these neurons.  
 
2.4.1.1.3. Islands of Calleja and olfactory tubercle 
These nuclei exhibited the typical mammalian organizational plan (Woolf, 1991; 
Manger et al., 2002a; Maseko et al., 2007). They were located in the ventral most portion 
of the cerebral hemisphere at a level ventral to the nucleus accumbens (Figs. 2.2 D-F, 
2.3). These nuclei extended from the level of the anterior horn of the lateral ventricle to 
the level of the anterior commissure. The Islands of Calleja contained a moderate density 
of ChAT+ neuronal clusters and a scattered low to medium density of ChAT+ neurons 
 25
surrounding these clusters were assigned to the olfactory tubercle. The cells in these 
regions were intensely immunoreactive and a mixture of bipolar and multipolar neuron 
types with ovoid shaped somas, were observed. The neurons of the Islands of Calleja as 
well as those of the olfactory tubercle exhibited no specific dendritic orientation.   
 
2.4.1.2. Cholinergic nuclei of the basal forebrain 
2.4.1.2.1. Medial septal nucleus   
This nucleus was identified through the presence of ChAT+ neurons located 
within the septal nuclear complex in the rostral half of the medial wall of the cerebral 
hemisphere (Figs. 2.2 G-H, 2.3). The location of this nucleus is typical of what has been 
observed in other mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007). A 
moderate to high density of ChAT+ neurons, a mixture of both bipolar and multipolar 
types with ovoid cell shapes, was observed. It was also noted that the cells located closer 
to the edge of the hemisphere were smaller than those cells located a small distance from 
the edge. Also, the cells closer to the edge of the hemisphere were mostly bipolar with 
dendrites arranged parallel to the edge, whereas those further away were predominantly 
multipolar and exhibited no specific dendritic orientation. 
 
2.4.1.2.2. Diagonal band of Broca 
The diagonal band of Broca was located anterior to the hypothalamus in the 
ventromedial corner of the cerebral hemisphere (Fig. 2.2F). The division of this nucleus 
into horizontal and vertical bands appeared to be unnecessary as the neurons forming this 
nucleus presented as a continuous, uninterrupted band. A high density of ChAT+ neurons 
 26
was found throughout the extent of this nucleus (Fig. 2.4). The cells were intensely 
immunoreactive, ovoid in shape and multipolar in type with their dendrites orientated 
roughly parallel to the edge of the cerebral hemisphere. It was also noted that the neurons 
of this nucleus were larger than those found in the adjacent olfactory tubercle and islands 
of Calleja, a feature that can be readily used to demarcate this lateroventral boundary of 
the diagonal band.    
 
2.4.1.2.3. Nucleus basalis 
ChAT+ neurons found ventral to the globus pallidus at the level of the anterior 
commissure, and caudal and dorsal to the olfactory tubercle were assigned to nucleus 
basalis (Figs. 2.2 G-H, 2.3). A low to moderate density of ChAT+ neurons was observed 
throughout this nucleus, and these appear to be a continuation of the ChAT+ neurons 
found within the globus pallidus. These ChAT+ neurons were ovoid in shape, a mixture 
of bipolar and multipolar types, and exhibited a rough dorsolateral to ventromedial 
dendritic orientation. 
 
2.4.1.3. Diencephalic cholinergic nuclei 
2.4.1.3.1. Medial habenular nucleus 
The medial habenular nucleus, which forms part of the epithalamus, was located 
contiguous to the third ventricle in the dorsomedial region of the diencephalon (Figs. 2.2 
M, 2.3). The location of this nucleus was typical of what has been observed in other 
mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007). A high density of 
small, round, ChAT+ neurons, was observed.  We could not discern any specific 
 27
dendritic orientation due to the dense packing of neurons within this nucleus.  The 
ChAT+ axons arising from this nucleus clearly outlined the fasciculus retroflexus, which 
was seen to end in a large swirling termination within the interpeduncular nucleus. 
 
2.4.1.3.2. Hypothalamic cholinergic nuclei 
Three distinct ChAT+ neuronal groups (dorsal, ventral and lateral hypothalamic 
nuclei) were identified within the hypothalamus (Figs. 2.2 I-L, 2.3). The dorsal 
hypothalamic nucleus was located between the wall of the third ventricle and the fornix 
within the dorsomedial region of the hypothalamus. A low density of palely stained, 
scattered ChAT+ neurons was observed within this nucleus. The ventral hypothalamic 
nucleus was identified as a cluster of palely stained, widely scattered ChAT+ neurons 
within the ventromedial aspect of the hypothalamus that extended ventrolaterally past the 
level of the fornix. Within the dorsolateral region of the hypothalamus, lateral to the 
fornix, a low density of palely stained ChAT+ neurons were identified as the lateral 
hypothalamic cholinergic nucleus. The neuronal morphology of the ChAT+ neurons 
within these hypothalamic cholinergic nuclei was similar.  These neurons had an ovoid 
shape, were bipolar in type and showed no specific dendritic orientation. 
 
2.4.1.3.3. Cholinergic neurons in the anterodorsal and anteroventral dorsal thalamic 
nuclei 
Within the anterior and dorsal region of the dorsal thalamus the anterodorsal 
nucleus (AD) was seen to exhibit a strong cholinergic neuropil staining. Around the 
margins of the AD nucleus a small number of scattered ChAT+ neurons were observed 
 28
(Figs. 2.2K, 2.3). The ChAT+ neurons within this nucleus were ovoid in shape, a mixture 
of bipolar and multipolar types with the dendrites orientated parallel to the margins of the 
nucleus (Fig. 2.5). Lateral to the anterodorsal nucleus, the anteroventral nucleus (AV) 
exhibited a much paler neuropil ChAT+ neuropil staining, but it was clearly 
distinguishable from the surrounding unreactive tissue of the remainder of the dorsal 
thalamus.  ChAT+ neurons showing a similar morphology to those found in the 
anterodorsal nucleus were observed along the upper medial and lateral borders of the 
anteroventral nucleus. The cholinergic nature of the neuropil and occasional scattered 
ChAT+ neurons within these nuclei might be a unique feature of the hyrax, as these 
observations have not been made in other mammals to date. 
 
2.4.1.4. Pontomesencephalic nuclei 
2.4.1.4.1. Parabigeminal nucleus 
A prominent parabigeminal nucleus was located ventral and slightly anterior to 
the inferior colliculus within the lateral aspect of the midbrain tegmentum (Figs. 2.2 O-P, 
2.3). A dense cluster of strongly reactive ChAT+ neurons, smaller in size to the medially 
located pedunculopontine tegmental nucleus, was observed.  The cell bodies were 
circular in shape but due to the high density of the packing it was difficult to determine 
whether these neurons were bipolar or multipolar and if the dendrites showed any 
specific orientation (Fig. 2.6A).  
 
 
 
 29
 
2.4.1.4.2. Pedunculopontine tegmental nucleus (PPT) – magnocellular and parvocellular 
nuclei 
In all mammals studied to date, the cholinergic neurons of the pedunculopontine 
tegmental nucleus display a homogenous morphology (e.g. Woolf, 1991; Manger et al., 
2002a; Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al., 2008).  As with 
most mammals within the dorsal aspect of the isthmic and pontine tegmental regions, 
anterior to the trigeminal motor nucleus, a group of ChAT+ neurons was identified (Figs. 
2.2 O-Q, 2.3).  This location, and the moderate density distribution of ChAT+ neurons in 
this region, is again typical.  In contrast to other mammals studied, the morphology of the 
cholinergic neurons within this region was not homogenous, and we could readily 
identify magnocellular and parvocellular ChAT+ nuclei within this region (Fig. 2.6B, D).  
The ChAT+ neurons assigned to the magnocellular PPT nucleus were found in the 
location, and evinced the distribution of the more typically described PPT.  These 
magnocellular neurons were found medial and ventral to the superior cerebellar peduncle, 
and in the anterior portion were found to intermingle with the fibres of the superior 
cerebellar peduncle.  The magnocellular ChAT+ neurons were found in a moderate 
density throughout this nucleus, were multipolar in type, showed no specific dendritic 
orientation, and evinced a variety of somal shapes.  In the more caudal aspects of the 
PPT, in a position dorsal and lateral to the magnocellular nucleus, a moderately dense 
cluster of small ChAT+ neurons was observed, termed the parvocellular PPT nucleus.  
The parvocellular PPT nucleus was sometimes clearly distinguished from the 
magnocellular PPT nucleus on topological grounds, as it was found lateral and dorsal to 
 30
the superior cerebellar peduncle in the posterior aspect of the midbrain tegmentum; 
however, the size of the soma was the most reliable distinction between these two nuclei. 
The ovoid shaped, parvocellular ChAT+ neurons were mostly bipolar in type, but there 
were some multipolar neurons in this nucleus.  These neurons had dendrites that were 
predominantly oriented in a dorsomedial to ventrolateral direction.  There was no overlap 
in the distribution of the magnocellular and parvocellular neurons at their border. 
 
2.4.1.4.3. Laterodorsal tegmental nucleus (LDT) – magnocellular and parvocellular 
nuclei 
In all mammals studied to date, the cholinergic neurons of the laterodorsal 
tegmental nucleus display a homogenous morphology (e.g. Woolf, 1991; Manger et al., 
2002a; Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al., 2008). As with 
most mammals within the ventrolateral region of the pontine periventricular- and 
periaqueductal grey matter, a moderate to high density of ChAT+ neurons was identified 
(Figs. 2.2 P-Q, 2.3).  This location and distribution of ChAT+ neurons is again typical of 
LDT in various mammals.  In contrast to other mammals studied, the morphology of the 
cholinergic neurons within this region was not homogenous, and we could readily 
identify magnocellular and parvocellular ChAT+ nuclei within this region (Fig. 2.6B, C).  
At the ventrolateral border of the pontine periventricular- and periaqueductal grey matter 
a densely packed cluster of larger neurons with similar morphology to those constituting 
the magnocellular PPT nucleus was found.  We have termed this cluster of ChAT+ 
neurons the magnocellular LDT nucleus.  This magnocellular LDT nucleus appears to be 
typical of that seen and described as the LDT in most mammals. A second cluster of 
 31
ChAT+ neurons was found medial and dorsal to the magnocellular LDT nucleus, but still 
within the periventricular- and periaqueductal grey matter.  These neurons were 
significantly smaller than those in the magnocellular LDT nucleus, thus we have termed 
this ChAT+ neuronal cluster the parvocellular LDT nucleus.  The morphology and size of 
these neurons was very similar to the ChAT+ neurons forming the parvocellular PPT 
nucleus; however, they did not display any specific dendritic orientation.  These neurons 
were found in a moderate to high density and had soma that were ovoid to circular in 
shape.  The lateral border of the parvocellular LDT nucleus and the medial border of the 
magnocellular LDT nucleus were not clearly defined as the ChAT+ neurons from both 
nuclei showed a region of intermingling.  In this sense, while the magnocellular and 
parvocellular nuclei of the LDT and PPT seem to be contiguous subdivisions of these 
nuclei, the distinct topological parcellation of the two nuclei in the PPT was not as 
strongly expressed in the LDT. 
 
2.4.1.5. Cholinergic cranial nerve motor nuclei 
A number of large, multipolar ChAT+, neurons forming the cranial nerve motor 
nuclei were identified in similar regions to those previously documented for other 
mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007; Limacher et al., 
2008; Bhagwandin et al., 2008).  These nuclei were: the oculomotor nucleus (III) (Fig. 
2.6A), the trochlear nucleus (IV), the motor division of the trigeminal nucleus (Vmot), 
the abducens nucleus (VI), the dorsal and ventral subdivisions of the facial nucleus (VIId 
and VIIv), the nucleus ambiguus, the dorsal motor vagus nucleus (X), the hypoglossal 
nucleus (XII) and the ventral horn of the spinal cord (Figs. 2.2 N-Z, 2.3).  In addition to 
 32
these we were able to locate ChAT+ neurons within the Edinger-Westphal nucleus and 
the preganglionic motor neurons of the superior salivatory nucleus of the facial nerve 
(pVII) and the preganglionic motor neurons of the inferior salivatory nucleus (pIX).  The 
ChAT+ neurons of the Edinger-Westphal nucleus were moderate in number, but were 
strongly immunoreactive. This nucleus primarily gives rise to the preganglionic 
parasympathetic fibres to the eye, constricting the pupil, and is located within the 
periaqueductal grey matter directly contiguous to midline, between and anterior to the 
oculomotor nuclei. The neuronal bodies were ovoid in shape, bipolar and the dendrites 
were orientated dorsoventrally. Within the rostral medullary tegmentum, dorsal to the 
facial nerve nuclei, the ChAT+ neurons of pVII and pIX were identified. The neurons of 
the pVII nucleus were slightly smaller than those of the facial nerve nucleus and 
exhibited a small number of scattered multipolar neurons that demonstrated the typical 
motor neuron morphology, probably representing the preganglionic cholinergic cell 
bodies of the superior salivatory nucleus (see Mitchell and Templeton, 1981; Tóth et al., 
2007). The ChAT+ neurons found medial to X probably represented the ChAT+ 
preganglionic neurons of the inferior salivatory nucleus (pIX) (see Rezek et al., 2008). At 
this level of the medulla oblongata a large number of   ChAT+ nerve cell bodies of 
moderate density were located in the medial periventricular grey (Fig. 2.7). The neurons 
were multipolar but smaller than the motor neurons of X and XII, with no specific 
dendritic orientation.  The extensive number of cholinergic cells in this region appears to 
be unusual in comparison to other mammals. 
 
 
 33
 
2.4.2. Putative catecholaminergic nuclei 
The putative catecholaminergic nuclei, in the current study being those that 
possess neurons that are immunoreactive to tyrosine hydroxylase (TH+), are generally 
divided into several nuclear complexes that extend from the level of the olfactory bulb to 
the spinomedullary junction. These are generally defined as: the olfactory bulb, 
diencephalic, midbrain, pontine, and medullary nuclear complexes. All of the 
aforementioned nuclear complexes were present in the brain of the rock hyrax and the 
location and distribution of these was essentially similar to the typical mammalian 
organizational plan of this particular system previously described in a range of other 
species (Dahlström and Fuxe, 1964; Fuxe et al., 1969, 2007b; Smeets and González, 
2000; Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al., 2008).  No putative 
catecholaminergic nuclei falling outside these defined regions, as is the case in some 
other vertebrates, were identified in the hyrax (Smeets and González, 2000). The standard 
nomenclature of Dahlström and Fuxe (1964) and Hökfelt et al. (1984) was implemented 
in the description of these nuclei.  The putative catecholaminergic nuclei identified in the 
rock hyrax are essentially similar to that seen in other mammals (Figs. 2.2, 2.3); however, 
we did not identify the rodent specific rostral dorsal midline medullary (C3) nucleus 
(Smeets and González, 2000; Moon et al., 2007; Dwarika et al., 2008; Limacher et al., 
2008; Bhagwandin et al., 2008) and the locus coeruleus compact, built of densely packed 
NA cell bodies (see Fuxe et al., 1970) did not exist. Furthermore, only a few single TH+ 
neurons were observed in the adjacent periventricular grey of the pons, termed A6 diffuse 
(A6d) in the current study. 
 34
 
2.4.2.1. Olfactory bulb (A16) 
Within and around the glomerular layer a moderate number of TH+ neurons were 
observed (Figs. 2.2A, 2.3). This position is typical of what has been documented in other 
mammals for the A16 catecholaminergic nucleus (Lichtensteiger, 1966; Lidbrink et el., 
1974; Smeets and González, 2000; Manger et al., 2002b; Maseko et al., 2007; Limacher 
et al., 2008; Bhagwandin et al., 2008). The cells were triangular in shape and a mixture of 
bipolar and multipolar neurons were evident. These neurons also had a weak dendritic 
network surrounding the glomeruli. 
 
2.4.2.2. Diencephalic nuclei 
In the hypothalamus six distinct putative catecholaminergic nuclei were 
identified. These include: the dorsal division of the anterior hypothalamic nucleus 
(A15d); the ventral division of the anterior hypothalamic nucleus (A15v); the rostral 
periventricular nucleus (A14); the zona incerta (A13); the tuberal nucleus (A12); and the 
caudal diencephalic nucleus (A11) (Figs. 2.2 G-L, 2.3). The A15d nucleus was located 
between the third ventricle and the fornix within the anterior portion of the hypothalamus. 
The shape of the neuronal bodies was found to be ovoid, with a mixture of bipolar and 
multipolar types showing no specific dendritic orientation. The A15v nucleus was located 
near the floor of the brain adjacent to the optic tracts within the ventrolateral region of the 
hypothalamus. A low density of TH+ neurons with ovoid cell bodies was observed. A 
mixture of bipolar and multipolar neurons with no specific dendritic orientation was 
evident.  A moderate density of TH+ neurons close to wall of the third ventricle 
 35
throughout the hypothalamus was identified as the A14 nucleus (Fig. 2.8A). The neurons 
had ovoid shaped somas and were mostly of the bipolar type with some being multipolar. 
The dendrites were observed to be orientated roughly parallel to the ventricular wall.  The 
A13 nucleus was identified as a moderate density of TH+ neurons located within the 
dorsal and lateral regions of the hypothalamus, lateral to the fornix, extending into and 
around the zona incerta in the ventral thalamus. The neurons exhibited morphological 
similarity to those within the A15d nucleus.  
In the ventral portion of the hypothalamus, within and in close proximity to the 
arcuate nucleus, a low to moderate density of TH+ neurons was identified as the A12 
nucleus. The neurons of this nucleus were located near the midline and surrounded the 
floor of the third ventricle. The cell bodies were ovoid and the neurons a mixture of the 
bipolar and multipolar types with dendrites orientated roughly parallel to the floor of the 
brain. Around the posterior pole of the third ventricle, in the most caudal part the 
hypothalamus, a low density of TH+ neurons arranged in columns on either side of the 
midline was identified as the A11 nucleus. The neurons of this nucleus were large and 
multipolar, with ovoid to polygonal shaped somas. No specific dendritic orientation was 
observed for these neurons. The cell bodies of the A11 neurons were larger than those of 
all the other neurons of the diencephalic nuclei, with this feature serving as a ready 
marker for delineation of this nucleus. 
 
2.4.2.3. Midbrain nuclei 
2.4.2.3.1. Ventral tegmental area nuclei (VTA, A10 complex) 
 36
The A10 nuclear complex (comprised of the following nuclei: A10 – the ventral 
tegmental area nucleus; A10c – ventral tegmental area, central nucleus; A10d – ventral 
tegmental area, dorsal nucleus; A10dc – ventral tegmental area, dorsal caudal nucleus) 
was located within the medial part of the midbrain tegmentum slightly anterior to the 
level of the oculomotor nerve nucleus (Figs. 2.2 L-O, 2.3). The nuclei forming this 
complex originated from the dorsal and dorsolateral areas around the interpeduncular 
nucleus and extended dorsally to the periaqueductal grey matter, where the aqueduct 
component was formed caudally. The A10 nucleus was identified as a high density of 
TH+ neurons located between the interpeduncular nucleus and the root of the oculomotor 
nerve (Fig. 2.9).  The neurons were distributed dorsally and dorsolaterally to the 
interpeduncular nucleus. The cell bodies were ovoid in shape and the neurons were a 
mixture of the bipolar and multipolar types showing no specific dendritic orientation. 
Immediately dorsal to the interpeduncular and A10 nuclei, a moderate to high density of 
TH+ neurons was identified as the A10c nucleus (Fig. 2.9). The neurons of this nucleus 
formed a characteristically roughly triangular shaped pattern in the midline. A similar 
neuronal morphology as the neurons of the A10 nucleus was observed for the neurons of 
the A10c nucleus; however the neurons of A10c were slightly smaller in size than those 
of A10, this size difference serving as a marker for delineating between these two 
adjacent nuclei. A moderate density of TH+ neurons located dorsal and between the A10c 
nucleus and the periaqueductal grey matter, medial to the oculomotor nerve nucleus, was 
identified as A10d (Fig. 2.9). The cell bodies of these neurons were ovoid in shape, 
bipolar in type and showed a dorsoventral dendritic orientation. The A10dc nucleus was 
located within the periaqueductal grey matter around the lower half of the border of the 
 37
cerebral aqueduct (Fig. 2.8B). The TH+ neurons forming the A10dc nucleus were small, 
ovoid in shape, multipolar and the dendrites were orientated roughly parallel to the edge 
of the cerebral aqueduct. 
  
2.4.2.3.2. Substantia nigra nuclear complex (A9) 
In the ventrolateral region of the midbrain tegmentum, dorsal to the cerebral 
peduncles, four distinct nuclei comprising the substantia nigra nuclear complex were 
identified. These nuclei were: the substantia nigra, pars compacta nucleus (A9pc); 
substantia nigra, pars lateralis nucleus (A9l); substantia nigra, ventral or pars reticulata 
nucleus (A9v); and the substantia nigra, pars medialis nucleus (A9m) (Figs. 2.2 K-N, 
2.3). The A9pc nucleus was identified as a mediolaterally oriented band of TH+ neurons 
of a moderate to high density located dorsal to the cerebral peduncle (Fig. 2.9). The 
neurons were ovoid in shape and bipolar with a medial to lateral dendritic orientation, 
parallel to the direction of the band. The A9l nucleus was located within the ventrolateral 
region of the midbrain tegmentum, dorsolateral to the lateral border of the cerebral 
peduncle in a position lateral to the A9pc nucleus (Figs. 2.8D, 2.9). The somas of the 
TH+ neurons of this nucleus were either polygonal or triangular in shape. The neurons 
were multipolar and did not exhibit any specific dendritic orientation. The A9v nucleus 
was located in the grey matter just dorsal to the cerebral peduncle ventral to the A9pc and 
A9l nuclei (Fig. 2.9). The neurons were found in a low to moderate density in this region 
and showed a rough dorsoventral dendritic orientation. A high density of TH+ neurons 
located between the medial edge of the A9pc nucleus and the root of the oculomotor 
nerve was identified as the A9m nucleus (Fig. 2.9). The neurons of this nucleus showed a 
 38
similar neuronal morphology as A9pc, with the neurons showing no specific dendritic 
orientation.  
 
2.4.2.3.3. Retrorubral nucleus (A8) 
Within the lower half of the midbrain tegmentum a moderate number of TH+ 
neurons were found dorsal to the A9 complex, and ventral and caudal to the 
magnocellular division of the red nucleus.  These neurons, which exhibited a moderate 
density throughout this region, were assigned to the A8 nucleus (Figs. 2.2 N-O, 2.3). The 
TH+ neuronal bodies were ovoid in shape, bipolar and multipolar in type, and showed no 
specific dendritic orientation (Fig. 2.8D).    
 
2.4.2.4. Pontine nuclei – the locus coeruleus (LC) nuclear complex 
The locus coeruleus complex was identified as a large aggregation of TH+ 
neurons within the pontine region. Five nuclei within this complex were identified, being: 
the subcoeruleus compact nucleus (A7sc), subcoeruleus diffuse nucleus (A7d), locus 
coeruleus diffuse nucleus (A6d), fifth arcuate nucleus (A5), and the dorsal medial nucleus 
of the locus coeruleus (A4) (Figs. 2.2 P-Q, 2.3).  A7sc was located within the dorsal 
region of the pontine tegmentum adjacent to the periventricular grey matter. A high 
density of TH+ neurons with ovoid somas of the multipolar type represented this nucleus 
(Fig. 2.8C). The neurons of A7sc showed no specific dendritic orientation. The 
description of this nucleus coincides to what was previously described as the 
subcoeruleus (Dahlström and Fuxe, 1964).  TH+ neurons ventrolateral to A7sc, anterior 
to the trigeminal motor nucleus, in the lateral and dorsolateral region of the pontine 
 39
tegmentum were assigned to the A7d nucleus. Some of the TH+ neurons assigned to A7d 
were also located medial and ventral to the superior cerebellar peduncle. The TH+ 
neurons in this region were found in a moderate to low density, were ovoid in shape, 
multipolar and showed no specific dendritic orientation. A compact A6 built of TH+ 
neurons was absent, and only a small number of TH+ neurons was located within the 
ventrolateral portion of the periventricular grey matter adjacent to A7sc, and identified as 
A6d (Fig. 2.8C). The neurons of A6d had a similar morphology to A7sc and showed no 
specific dendritic orientation. In comparison to other mammalian species, the rock hyrax 
had markedly reduced number of immunopositive neurons within A6d (Dahlström and 
Fuxe, 1964; Fuxe et al., 1969, 2007b; Smeets and González, 2000; Maseko et al., 2007; 
Limacher et al., 2008; Bhagwandin et al., 2008). The TH+ neurons assigned to the A5 
nucleus was located lateral to the superior olivary nucleus, anterior to the facial nerve 
nucleus and ventral to the trigeminal motor nerve nucleus, in the ventrolateral region of 
the pontine tegmentum. The TH+ neurons of this nucleus were few in number and had a 
similar neuronal morphology to the neurons seen in the other locus coeruleus nuclei. The 
TH+ neurons assigned to the A4 nucleus were located adjacent to the wall of the fourth 
ventricle, medial to superior cerebellar peduncle, in the dorsolateral part of the caudal 
periventricular grey matter. This nucleus was represented by a small number of TH+ 
neurons with a similar morphology to the neurons seen in the other locus coeruleus 
nuclei.  
 
 
 
 40
2.4.2.4.1. Medullary nuclei 
TH+ neurons representing five putative catecholaminergic nuclei were found 
within in the medulla of the rock hyrax.  The nuclei identified were: the rostral 
ventrolateral tegmental nucleus (C1), rostral dorsomedial nucleus (C2), caudal 
ventrolateral tegmental nucleus (A1), caudal dorsomedial nucleus (A2) and area postrema 
(AP) (Figs. 2.2 R-X, 2.3). The rostral dorsal midline nucleus (C3), a feature so far only 
identified in rodents (Smeets and González, 2000; Maseko et al., 2007; Limacher et al., 
2008; Bhagwandin et al., 2008) was not present.  The neurons forming the C1 nucleus 
were located in the ventrolateral region of the anterior medulla and were distinguished as 
a column of TH+ neurons that extended from the level of the superior olivary nucleus to 
the level of nucleus ambiguus.  In the course of its distribution this band of neurons was 
located lateral to superior olivary nucleus and facial nerve nucleus, and medial to nucleus 
ambiguus. The neurons of this nucleus were ovoid in shape, multipolar and exhibited a 
mesh-like dendritic network interspersed among the ascending and descending fasciculi 
of the medulla. A low density of TH+ neurons within the dorsal region of the medulla 
was identified as C2. This nucleus was located anterior and dorsal to the vagus motor 
nerve nucleus. The dorsal strip of C2 (Kalia, et al., 1985a, b), which is the part of the C2 
nucleus that is located near the floor of the fourth ventricle above the nucleus of the 
vagus motor nerve, could be identified. The continuation of the dorsal strip, which is 
known as the rostral subdivision of the C2 nucleus (Kalia, et al., 1985a, b), was not 
present.  The neurons of this nucleus were few in number and of low density. The somal 
shape was ovoid and the neurons were bipolar with dendrites orientated parallel to the 
floor of the fourth ventricle.  The A1 nucleus was identified as a column of TH+ neurons 
 41
within the ventrolateral region of the caudal medulla. The neurons of this nucleus were 
distributed from the level of nucleus ambiguus to the level of the spinomedullary 
junction.  In the course of its distribution the neurons of the A1 nucleus were located 
lateral to nucleus ambiguus and the lateral reticular nucleus. The distinguishing factor 
between the partly overlapping caudal and rostral neuronal columns of the A1 and C1 
(caudal part) nucleus, respectively, was their position relative to the nucleus ambiguus. 
The A1 column was located lateral to nucleus ambiguus while the caudal part of the C1 
column was located medial to this nucleus.  The neuronal morphology and dendritic 
organization of the A1 nucleus was similar to that of the C1 nucleus. A number of 
moderately sized TH+ neurons were located between, as well as around, the nuclei of the 
dorsal motor vagus and hypoglossal nerves and these were assigned to the A2 nucleus; 
however, none were found in the nucleus tractus solitarius, which is the major location 
for this cell group in rodents (Dahlström and Fuxe, 1964).  These neurons were ovoid in 
shape, bipolar in type and their dendrites exhibited a medial to lateral orientation. The 
area postrema (AP) was identified as a high-density cluster of small TH+ neurons. This 
nucleus was located just anterior to the spinomedullary junction, adjacent to the floor of 
the fourth ventricle and dorsal to the nucleus tractus solitarius and the central canal. The 
neurons were ovoid shaped, bipolar in type, and showed no specific dendritic orientation.  
 
2.4.3. Serotonergic nuclei 
A number of distinct serotonergic immunoreactive (5HT+) nuclei were found 
throughout the brainstem of the rock hyrax.  These were observed from the level of the 
decussation of the superior cerebellar peduncle through to the spinomedullary junction 
 42
(Figs. 2.2 M-W, 2.3, 2.10, 2.11). These nuclei were readily divisible into rostral and 
caudal nuclear clusters and the location of the nuclei within these clusters were found to 
be similar to what has been described for other eutherian mammals (e.g. Törk, 1990; 
Manger et al., 2002c; Badlangana et al., 2007; Maseko et al., 2007; Moon et al., 2007; 
Dwarika et al., 2008; Limacher et al., 2008; Bhagwandin et al., 2008). 
  
2.4.3.1. Rostral cluster 
2.4.3.1.1. Caudal linear nucleus (CLi) 
A moderate density of 5HT+ neurons was located within the midbrain tegmentum 
of the rock hyrax. A cluster of these neurons were found to be in a position immediately 
dorsal to the interpeduncular nucleus and anterior to the decussation of the superior 
cerebellar peduncle in the ventral midline of the midbrain tegmentum.  This cluster was 
designated as the caudal linear nucleus (Figs. 2.2 M-N, 2.3). At the lateral border of the 
interpeduncular nucleus, the 5HT+ neurons of the CLi also extended to the ventral 
surface of the brain and it was noted that this nucleus was the most rostrally located 
nucleus of all the 5HT+ nuclei. The neurons of this nucleus had ovoid shaped somas, 
were bipolar in nature, and had a rough dorsoventral dendritic orientation (Fig. 2.11A, B).   
 
2.4.3.1.2. Supralemniscal nucleus (B9) 
The B9 nucleus was identified as a loosely packed arc of 5HT+ neurons, lateral to 
the interpeduncular nucleus and caudal to the A9pc nucleus (Figs. 2.2 N, 2.3). This 
nucleus was located superior to the medial lemniscus and was found to be in continuity 
 43
with the ventrolateral neurons of the CLi nucleus.  The neurons had ovoid shaped somas, 
were bipolar in type, and the dendrites had no specific orientation (Fig. 2.11B).   
 
2.4.3.1.3. Median raphe (MnR) 
In a pararaphe position, on either side of the midline, two clear, densely packed 
columns of 5HT+ neurons, extending dorsal to ventral along the midbrain and pontine 
tegmentum, were identified as the MnR nucleus (Figs, 2 N-Q, 3). The neurons forming 
this nucleus extended from the caudal aspect of the superior cerebellar peduncle to the 
level of the anterior most aspect of the trigeminal motor nucleus.  The cell bodies of the 
5HT+ neurons of the MnR nucleus were ovoid, bipolar in type and the dendrites showed 
a rough dorsoventral orientation (Fig. 2.10). 
 
2.4.3.1.4. Dorsal raphe nuclear complex (DR) 
Within the 5HT+ dorsal raphe nuclear complex six distinct nuclei, extending from 
the level of the oculomotor nerve nucleus to the level of the trigeminal motor nerve 
nucleus within the periaqueductal and periventricular grey matter, were identified. These 
nuclei were: the dorsal raphe interfascicular (DRif) nucleus, the dorsal raphe ventral 
(DRv) nucleus, the dorsal raphe dorsal (DRd) nucleus, the dorsal raphe lateral (DRl) 
nucleus, the dorsal raphe peripheral (DRp) nucleus, and the dorsal raphe caudal (DRc) 
nucleus (Figs 2 O-Q, 3).  A dense cluster of 5HT+ neurons within the most ventral 
medial portion of the periventricular grey matter, between the two medial longitudinal 
fasciculi, was identified as the DRif nucleus (Fig. 2.10). The neurons of this nucleus had 
ovoid shaped cell somas, were bipolar in type and had dendrites that were orientated 
 44
roughly dorsoventrally. Immediately dorsal to the DRif nucleus and caudal to the 
oculomotor nerve nuclei, a high density of 5HT+ neurons was identified as the DRv 
nucleus (Fig. 2.10). These neurons were ovoid in shape and bipolar with no specific 
dendritic orientation. The DRd nucleus was identified as a high density of 5HT+ neurons 
located ventral to the inferior border of the cerebral aqueduct and immediately dorsal to 
the DRv nucleus (Fig. 2.10). The neurons of this nucleus had a similar neuronal 
morphology and dendritic orientation as the neurons of the DRv nucleus. A very low 
density of 5HT+ neurons located lateral to the DRd and DRv nuclei, anterior to the 
ChAT+ neurons of the LDT, in the ventrolateral portion of the periaqueductal grey matter 
was identified as the DRp nucleus. In the adjacent midbrain tegmentum, a small number 
of 5HT+ neurons forming part of the DRp nucleus were observed (Fig. 2.10). Of all the 
dorsal raphe nuclei, the tegmental neurons of the DRp nucleus were the only ones not 
located within the periventricular and periaqueductal grey matter. The 5HT+ neurons 
forming the DRp were ovoid to polygonal in shape, multipolar, and showed no specific 
dendritic orientation.  Adjacent to the ventrolateral edge of the cerebral aqueduct, in a 
position dorsolateral to the DRd nucleus, a group of 5HT+ neurons were assigned to the 
DRl nucleus (Fig. 2.10). A low density of large, ovoid, multipolar neurons with no 
specific dendritic orientation was observed within this nucleus. This nucleus was readily 
distinguished from the other dorsal raphe nuclei due to the low neuronal density and the 
large soma of the neurons. An arc of 5HT+ neurons across the dorsal midline of the 
periventricular grey matter, where the cerebral aqueduct opens into the fourth ventricle, 
formed by the caudal coalescences of the two lateralized clusters of the DRl nucleus, was 
identified as the DRc nucleus (Fig. 2.11C). The neurons of this nucleus had a similar 
 45
neuronal morphology and dendritic orientation as those of the DRl nucleus. Due to the 
lack of 5HT+ neurons within this region of the brain in monotremes, the DRc was 
classified as an independent nucleus (Manger et al., 2002c). 
 
 2.4.3.2. Caudal cluster 
2.4.3.2.1. Raphe magnus nucleus (RMg) 
The 5HT+ neurons forming the RMg nucleus were located within the rostral 
medullary tegmentum, extending from the level of the anterior border of the facial nerve 
nucleus to the anterior border of nucleus ambiguus (Figs. 2.2 Q-S, 2.3). These 5HT+ 
neurons were low in density and formed two weakly expressed columns on either side of 
the midline. The neurons of the RMg nucleus were large, multipolar and exhibited a weak 
dorsoventral dendritic orientation. 
 
2.4.3.2.2. Rostral and caudal ventrolateral serotonergic medullary columns (RVL and 
CVL) 
Within the ventrolateral medullary tegmentum a column of 5HT+ neurons were 
identified as the RVL and CVL nuclei. This column extended from the level of the facial 
nerve nucleus to the spinomedullary junction and appeared to be a lateral extension of the 
neurons forming the RMg nucleus (Figs. 2.2 R-W, 2.3). The part of the column extending 
from the facial nerve nucleus to the rostral border of nucleus ambiguus was identified as 
the RVL nucleus, while the part of the column extending from the rostral border of 
nucleus ambiguus to the spinomedullary junction was designated as the CVL nucleus. 
The columns of the RVL nucleus were found immediately dorsal to the pyramidal tracts 
 46
anteriorly, and it was noted that this group of neurons bifurcated around the anterior pole 
of the inferior olivary nucleus, giving rise to two bilateral columns. These columns 
extended, in a position lateral to the inferior olivary nucleus, caudally within the 
ventrolateral medulla. The neurons of the RVL and CVL nuclei had a similar neuronal 
morphology and dendritic orientation as the neurons of the RMg nucleus and decreased 
in density from moderate, rostrally, to low, caudally.  It was observed, as with other 
mammalian species studied to date, that the RVL and CVL columns were continuous 
(e.g. Maseko and Manger, 2007; Maseko et al., 2007; Moon et al., 2007; Dwarika et al., 
2008; Bhagwandin et al., 2008); however, it is possible to make the distinction between 
these two nuclei, as the CVL column has not been found in the opossum or the 
monotremes (Crutcher and Humbertson, 1978; Manger et al., 2002c). 
 
2.4.3.2.3. Raphe pallidus nucleus (RPa) 
A low density and number of 5HT+ neurons located between the pyramidal tracts 
and at the ventral most border of the inferior olivary nucleus, within the ventral surface of 
the midline and medial medulla, was identified as the medial and lateral component of 
the RPa nucleus (Figs. 2.2 R-T, 2.3). The neurons of this nucleus exhibited fusiform-
shaped somas, were bipolar in type with a dorsoventral dendritic orientation parallel to 
the medial border of the pyramidal tracts. 
 
2.4.3.2.4. Raphe obscurus nucleus (ROb) 
On either side of the midline a low density of 5HT+ neurons, extending dorsal to 
ventral, from the level of the nucleus ambiguus to the spinomedullary junction were 
 47
identified as the ROb nucleus (Figs. 2.2 T-W, 2.3). The neurons were arranged in two 
loosely packed columns on either side of the midline. The cell bodies of these neurons 
were fusiform in shape, bipolar, and had a dorsoventral dendritic orientation.  Unlike in 
some other species studied, there were no neurons located a short distance from the 
central columns (e.g. Maseko and Manger, 2007; Maseko et al., 2007; Moon et al., 2007; 
Dwarika et al., 2008; Bhagwandin et al., 2008). 
 48
 
 
 
 
 
Figure 2.1: Photographs of the dorsal (top), lateral (middle) and ventral (bottom) aspects 
of the rock hyrax brain. Scale bar = 1 cm. 
 

 50
 
 
 
 
 
Figure 2.2: Diagrammatic reconstructions of a series of coronal sections through the 
brain of the rock hyrax illustrating the location of neurons immunohistochemically 
reactive for choline acetyltransferase (ChAT, black circles), tyrosine hydroxylase (TH, 
black triangles) and serotonin (open squares). The outlines of the architectonic regions 
were drawn using Nissl and myelin stains and immunoreactive neurons marked on the 
drawings. Drawing A represents the most rostral section, Z the most caudal. The 
drawings are approximately 1500 µm apart. See list for abbreviations. 
 



 54
 
 
 
 
 
Figure 2.3: Diagrams of idealized sagittal sections through the rock hyrax brain showing 
A. cholinergic, B. putative catecholaminergic, and C. serotonergic nuclei or nuclear 
complexes. See list for abbreviations. 
 

 56
 
 
 
 
 
Figure 2.4: Photomicrograph montage of the basal forebrain nuclei anterior to the 
anterior commissure, demonstrating choline acetyletransferase (ChAT) 
immunoreactivity.  Note the strongly ChAT+ densely packed neurons of the diagonal 
band of Broca (Diag.B) and the scattered strongly ChAT+ neurons in nucleus accumbens 
(N.Acc), caudate (C) and putamen (P). A weakly to moderately intense ChAT+ neuropil 
is found in these regions. Scale bar = 1 mm. 
 

 58
 
 
 
 
 
Figure 2.5: Photomicrographs of coronal sections through the anterior nuclei of the 
dorsal thalamus of the rock hyrax demonstrating the location of ChAT+ neurons and 
neuropil in the anteroventral (AV) and anterodorsal nuclei (AD).  A. Nissl stained 
section. B. Adjacent ChAT immunoreacted section. C. Higher power photomicrograph of 
AD showing location of ChAT+ neurons. D. High power photomicrograph showing 
ChAT+ neuron and terminals. Scale bar in B = 1 mm and applies to A and B. Scale bar in 
C = 100 µm Scale bar in D = 50 µm. 3V – third ventricle. 
 

 60
 
 
 
 
 
Figure 2.6: Photomicrographs of ChAT immunoreacted sections through the dorsal 
isthmic/pontine region of the rock hyrax brain. A. The magnocellular division of the 
pedunculopontine nucleus (PPTmc) at the level of the oculomotor nucleus (III) and the 
parabigeminal nucleus (PBg). B. The magnocellular and parvocellular divisions of both 
the laterodorsal (LDTmc, LDTpc) and pedunculopontine (PPTmc, PPTpc) nuclei at a 
level slightly caudal to that shown in A.  Note the distinct clusters of small neurons 
(LDTpc and PPTpc) located medial to the LDTmc and dorsal to the PPTmc. The 
distinct clusters are shown at higher power in C (LDTmc and LDTpc) and D (PPTmc 
and PPTpc). Scale bar in B = 1 mm and applies to A and B. Scale bar in D = 100 µm and 
applies to C and D. ca – cerebral aqueduct. 
 

 62
 
 
 
 
 
Figure 2.7: Photomicrographs of ChAT immunoreacted sections through periventricular 
grey matter in the dorsal caudal medullary region of the rock hyrax brain. A. The greatly 
expanded preganglionic motor neurons of the inferior salivatory nucleus (pIX) at a rostral 
level through this nucleus. B. The same nucleus at a more caudal level where it lies dorsal 
to the dorsal motor vagus (X) and hypoglossal (XII) nuclei. The Scale bar in B = 1 mm, 
applies to A and B. 4V – fourth ventricle. 
 

 64
 
 
 
 
 
Figure 2.8: Photomicrographs of selected neuronal groups immunohistochemically 
reactive for tyrosine hydroxylase within the brain of the rock hyrax. A. The rostral 
periventricular nucleus (A14) located adjacent to the walls of the third ventricle (3V) in 
the hypothalamus. B. The dorsal caudal nucleus of the ventral tegmental complex 
(A10dc) located around the ventral region of the cerebral aqueduct (ca). C. Very few 
TH+ neurons are found in the diffuse portion of the locus coeruleus (A6d) within the 
ventrolateral periaqueductal grey matter, while the compact portion of the nucleus 
subcoeruleus (A7sc) in the adjacent dorsal pontine tegmentum is rich in strongly TH+ 
neurons. D. The retrorubral nucleus (A8) and the lateral nucleus of the substantia nigra 
complex (A9l) in the lateral and ventral midbrain tegmentum. Scale bar in C = 500 µm 
and applies to A, B, C. Scale bar D = 1 mm. 
 

 66
 
 
 
 
 
Figure 2.9: Photomicrograph montage of the nuclear organization of the ventral 
tegmental area (A10, A10c, A10d) and the substantia nigra (A9m, A9pc, A9l, A9v) in 
the rock hyrax as revealed using tyrosine hydroxylase immunohistochemistry.  Scale bar 
= 1 mm. IP – interpeduncular nucleus; PC – cerebral peduncle; Rmc – red nucleus, 
magnocellular division. 
 

 68
 
 
 
 
 
Figure 2.10: Photomicrograph montage of the neuronal groups immunohistochemically 
reactive for serotonin within the dorsal raphe nuclear complex of the rock hyrax brain. 
Dorsal raphe lateral nucleus (DRl), dorsal raphe dorsal nucleus (DRd), dorsal raphe 
ventral nucleus (DRv), dorsal raphe interfascicular nucleus (DRif), dorsal raphe 
peripheral nucleus (DRp) and the median raphe nucleus (MnR). Scale bar = 1 mm. ca – 
cerebral aqueduct. 
 

 70
 
 
 
 
 
Figure 2.11: Photomicrographs of selected neuronal groups immunohistochemically 
reactive for serotonin in the midbrain and brainstem of the rock hyrax. A. The caudal 
linear nucleus (CLi) at its most anterior level, immediately dorsal to the interpeduncular 
nucleus (IP). B. The CLi at a more posterior level, where it is located ventral to the 
decussation of the superior cerebellar peduncle (xscp), and where at its most ventral and 
lateral aspect it becomes continuous with the supralemniscal serotonergic nucleus (B9). 
C. The caudal division of the dorsal raphe nucleus (DRc) lying immediately ventral to 
the most posterior portion of the cerebral aqueduct (ca). Scale bar in C = 1 mm, and 
applies to all. 
 

 72
2.5. Discussion 
The aim of the current study was to reveal the nuclear organization of the 
cholinergic, putative catecholaminergic and serotonergic systems of the rock hyrax, 
Procavia capensis. The results revealed that these systems, for the most part, are similar 
to what has previously been described in other mammals; however, certain specific 
differences were observed regarding the nuclear organization of the cholinergic and 
putative catecholaminergic systems. The nuclear organization of the serotonergic system 
was similar to what has been documented for other eutherian mammals (e.g. Törk, 1990; 
Maseko et al., 2007; Moon et al., 2007; Dwarika et al., 2008; Limacher et al., 2008; 
Bhagwandin et al., 2008). The differences that were noted for the cholinergic system 
included the existence of cholinergic immunoreactive neurons in the anterior dorsal and 
ventral nuclei of the dorsal thalamus, the presence of parvocellular and magnocellular 
divisions/nuclei of both the LDT and PPT, and the presence of a large cholinergic cell 
group in the periventricular grey of the rostral medulla (pIX). The features of the putative 
catecholaminergic system that differed significantly from that seen in other mammals 
were the absence of a dense cluster of TH+ neurons constructing the locus coeruleus 
compact (A6c) and a diffuse locus coeruleus (A6d) with only few TH+ neurons.  We 
discuss each of these unusual features in turn, and then compare the overall organization 
of these systems in the hyrax to prior observations of these systems across various 
mammalian species. 
 
 
 73
2.5.1. Cholinergic neurons in the anterodorsal (AD) and anteroventral (AV) nuclei of the 
dorsal thalamus of the rock hyrax 
The AD and AV nuclei identified in the brain of the rock hyrax form part of the 
anterior nuclei of the dorsal thalamus and are a general feature of the mammalian dorsal 
thalamus (Jones, 2007).  The anterior nuclear group typically consists of four distinct 
nuclei, the anterodorsal (AD), anteromedial (AM), anteroventral (AV), and lateral dorsal 
(LD) nuclei (Jones, 2007).  In the rock hyrax the AD nucleus was located within the 
anterior and dorsal regions of the dorsal thalamus, as is typical of mammals, and 
exhibited a strong cholinergic neuropil staining along with ChAT immunoreactive 
neurons scattered around the margins of the nucleus. Lateral to the AD nucleus in the 
rock hyrax, a second nucleus exhibiting a weaker cholinergic neuropil staining with 
ChAT immunoreactive neurons along the upper medial and lateral borders was identified 
as the AV nucleus. While both the AD and AV nuclei are known for their strong 
histochemical reactivity to acetylcholinesterase, choline acetyltransferase 
immunoreactive neurons have not been reported in either of these nuclei in any other 
mammal (Jones, 2007; Maseko et al., 2007). 
Afferent input to these nuclei originates largely from the hypothalamus and 
hippocampal formation, but also includes corticothalamic and thalamocortical 
connections (Jones, 2007). The AD and AV nuclei have been reported to be involved in 
spatial learning and memory, and navigation; however, these nuclei are not the sole 
contributors, but rather work in conjunction with the other anterior thalamic nuclei to 
achieve these functions (Segal et al., 1988; van Groen et al., 2002; Oda et al., 2003; 
Wolff et al., 2008). It is difficult to hypothesize on the potential effect of the presence of 
 74
cholinergic neurons within and closely surrounding these nuclei in regard to function; but 
given that the cholinergic system in general appears to be related to activity requiring 
both alertness and awareness, these neurons may have the effect of enhancing the 
wakefulness promoting functions of the anterior thalamic nuclei in the hyrax, possibly 
leading also to enhanced learning and memory functions. 
Presently, it appears that these anterior thalamic cholinergic neurons are unique to 
the rock hyrax as they have not been observed in other mammals.  Despite this, we can 
only tentatively conclude that they are a unique feature as no other members of the 
Afroplacentalia have been examined for the presence of these neurons.  Thus, these 
anterior thalamic cholinergic neurons may be unique to the hyrax, or may be found in 
other members of the Afroplacentalia, especially the closely related Proboscideans and 
Sirenians.  Until further Afroplacentalia species have been examined the uniqueness of 
this aspect of the cholinergic system remains uncertain. 
 
2.5.2. Magnocellular and parvocellular divisions/nuclei of the pedunculopontine and 
laterodorsal tegmental nuclei in the rock hyrax 
Within all mammals studied to date, cholinergic neurons within the pontine 
tegmentum (PPT) and periventricular grey matter (LDT) have been reported (e.g. Maseko 
et al., 2007). In all these prior studies the morphology of the cholinergic neurons within 
and between the PPT and LDT nuclei have been reported to be homogeneous (e.g. 
Woolf, 1991; Manger et al., 2002a; Maseko and Manger, 2007; Maseko et al., 2007; 
Limacher et al., 2008; Bhagwandin et al., 2008).  The situation observed in the current 
study in the rock hyrax raises a significant difference in this regard, as the cholinergic 
 75
neurons within these nuclei did not exhibit a homogenous morphology.  Rather, we 
identified two neuronal types within each of these nuclei, and the inner spatial 
distribution and segregation of these two neuronal types has led to the proposal of the 
subdivision of these nuclei into parvocellular and magnocellular PPT and LDT nuclei. 
The magnocellular subdivision of both the PPT and LDT nuclei exhibited a neuronal 
morphology reminiscent of that described for all other mammals (Woolf, 1991; Manger 
et al., 2002a; Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al., 2008).  For 
this reason, plus the topography of the neurons forming these magnocellular divisions 
(LDTmc being in the ventrolateral portion of the grey matter and PPTmc lying in the 
adjacent tegmentum), we propose that the magnocellular divisions of these nuclei in the 
rock hyrax are homologous to the LDT and PPT nuclei described for all other mammals.   
The parvocellular cholinergic subdivisions, on the other hand, have not been reported in 
any other mammal studied to date, although a non-cholinergic parvocellular component 
has at times been included as part of the PPT in rat and cat (Semba and Fibiger, 1989).  
The morphology of the neurons ascribed to the parvocellular divisions, being small, and 
bipolar with oval soma, are reminiscent of cholinergic interneurons found in the cerebral 
cortex (Bhagwandin et al., 2006) or superior colliculus (Tan and Harvey, 1989) in some 
other mammals.  These putative cholinergic interneurons were found medial to the 
LDTmc, and dorsal to the PPTmc.  It is possible that these cholinergic interneurons are 
also immunoreactive for GABA, forming part of the pontine GABAergic system. It has 
been reported by Bayraktar et al. (1997) that nearly all VIP+ and approximately 88% of 
ChAT+ neurons also contain GABA. A similar situation has been observed in the cortical 
cholinergic neurons found in certain species of mammals (Bhagwandin et al., 2006).  
 76
Potentially, a change at the molecular level of organization occurred within these 
neurons, such as molecular interactions, adding multiple transmission lines and thus 
increased functionality to this neuronal group (Agnati et al., 1986). These potentially 
GABAergic neurons may have acquired ChAT proteins resulting in their 
immunoreactivity. 
The above described potential molecular change and increased transmission lines 
in these cholinergic interneuron groups in close apposition to the traditional cholinergic 
LDT and PPT could lead to interesting functional outcomes.  It is well known that 
GABAergic neurons within the pons play a major role in the regulation of the traditional 
cholinergic LDT and PPT with regard to the sleep-wake cycle especially REM sleep 
(Maloney et al., 1999, 2000; Boissard et al., 2003; Pal and Mallick, 2006; Luppi et al., 
2007).  The location of the cholinergic, potentially GABAergic neurons described as the 
PPTpc and LDTpc in the current study makes them prime candidates forming part of a 
local GABAergic network involved in the switching of states during the sleep-wake cycle 
(Luppi et al., 2007).  The added transmission line related to the production of 
acetylcholine in the hyrax may produce some unusual effects on the measurable 
physiological parameters of sleep, especially REM sleep in the hyrax. 
Again it appears that these parvocellular cholinergic LDT and PPT may be unique 
to the rock hyrax as they have not been observed in other mammals (e.g. Maseko et al., 
2007); however, as with the other currently unique aspects of the cholinergic system in 
the hyrax we can only tentatively conclude that they are unique features until other 
members of the Afrotheria have been examined for the presence of these nuclei. 
 
 77
2.5.3. Increased numbers of preganglionic cholinergic neurons in the inferior salivatory 
nucleus (pIX) of the rock hyrax 
Cholinergic neurons within the preganglionic inferior salivatory nucleus have 
been identified in many mammalian species including rodents, carnivores, 
megachiropterans and primates, yet these are absent from the brains of other mammals 
studied (Maseko et al., 2007). Cholinergic neurons were identified in this nucleus in the 
rock hyrax; however, it appears from a qualitative comparison that the numbers of 
cholinergic neurons in this nucleus in the rock hyrax are far greater than that seen in the 
other mammals in which these neurons have been found.  This qualitative impression, 
while impressive, needs quantitative support. The possible projections and function of 
this enlarged periventricular cholinergic cell group of the medulla is unknown and 
comparisons across other species belonging to the Afrotheria are needed to determine 
whether this feature of the cholinergic system of the hyrax is a feature of the Afrotherians 
in general. 
 
2.5.4. Lack of TH+ neurons in compact locus coeruleus (A6c) and decreased numbers of 
TH+ neurons in the locus coeruleus proper (A6d) of the rock hyrax 
The diffuse division of the locus coeruleus (A6d), or the locus coeruleus proper 
(Dahlström and Fuxe, 1964), has been reported in all mammals studied to date, while the 
compact portion has only been reported in rabbits, tree shrew, megabats and primates 
(e.g. Maseko et al., 2007). While the A6c is lacking in the rock hyrax, the A6d nucleus is 
present but was particularly low in the number of TH immunoreactive neurons.  This is a 
qualitative impression, and further work comparing the number of neurons in this nucleus 
 78
across mammals and determining what kind of relationship may exist between neuronal 
number and brain mass (perhaps allometric) is required to confirm this impression; 
however, the qualitative impression is so dramatic as to warrant specific discussion. One 
interesting point to be noted is that even though the A6d nucleus showed an apparent 
reduction in the number of TH+ neurons, the other nuclei of the locus coeruleus complex 
(A7sc, A7d, A5 and A4) do not appear to be relatively more cell rich in comparison to 
other mammals (again a qualitative impression needing quantitative support). 
This potential reduction in the number of neurons within the locus coeruleus may 
lead to a reduction in the amount of noradrenaline released in the specific target regions 
of the A6d in the hyrax. It should be noted that A6d in the rat and other mammals 
projects to all cortical regions of the brain, including the cerebellar cortex, as well as to 
the thalamus, the lower brain stem reticular formation and the dorsal and ventral horns of 
the spinal cord (see Fuxe et al., 1970; Ungerstedt, 1971, 2007). The neurons of the locus 
coeruleus represent a tonic arousal system important for vigilance and attention (see Fuxe 
et al., 2007). The relative absence of the catecholamine component of the locus coeruleus 
proper (A6d) in the hyrax is a highly interesting observation. Has the subcoeruleus (A7sc 
and A7d) taken over the role of the locus coeruleus (A6d) in the hyrax or does the hyrax 
show a lack of tonic arousal? These are major questions to be answered in future work. 
The reduced number of neurons found within the A6d in the hyrax may be a feature 
unique to this species.  Further work on other members of the Afrotheria will determine 
whether this is indeed a species-specific feature or a feature of a broader taxonomic 
grouping. 
 
 79
2.5.5. Evolutionary and phylogenetic considerations 
The current study extends the database that may be used for comparison across 
mammalian species to determine both phylogenetic and evolutionary trends of the 
cholinergic, putative catecholaminergic and serotonergic systems.  Of these systems the 
serotonergic system appears to be the most conservative in terms of changes in the 
nuclear organization associated with different lineages. The rock hyrax, along with all 
other eutherian mammals studied to date and the metatherian wallaby appear to have the 
same nuclear organization (Maseko et al., 2007).  The metatherian opossum (Crutcher 
and Humbertson, 1978) and the prototherian monotremes (Manger et al., 2002c) have 
only small comparative differences in the nuclear organization of the serotonergic 
system.  The catecholaminergic system of the hyrax has a nuclear organization almost 
identical to that seen in rodents (except for the lack of a C3 nucleus, Dahlström and Fuxe, 
1964; Smeets and González, 2000) and megabats and primates (except for the lack of a 
compact division of the locus coeruleus, A6c, Maseko et al., 2007).  This nuclear 
organization aligns the hyrax phylogenetically most closely with the rodents and primates 
studied to date.  The cholinergic system, for the most part, shares a similar plan of 
organization as that described for the rodents, megabats and primates (Maseko et al., 
2007).  Despite this, the hyrax does evince a degree of nuclear organizational uniqueness 
compared with other mammalian species through the possession of cholinergic neurons 
in the anterior nuclei of the dorsal thalamus and the parvocellular LDT and PPT nuclei of 
the pontine region.  In the broader sense, the organization of these systems in the hyrax 
appear most similar to those seen in the rodents, rabbit, tree shrew, megabats and 
primates (Maseko et al., 2007).  In this sense, the nuclear organizations of the neural 
 80
systems studied reflect current concepts of the phylogenetic interrelationships of 
mammalian lineages (Arnason et al., 2008). 
The overall unique complement of nuclei of the three systems studied support the 
proposal of Manger (2005) that indicates each order will possess a unique nuclear 
complement of these systems.  The current study supports the concept of placing the 
hyraxes into the phylogenetic grouping of an order to the exclusion of other mammals.  It 
would be of interest to examine the cholinergic, catecholaminergic and serotonergic 
systems of other species of hyraxes to determine whether or not the “unique” features 
described in the brain of the rock hyrax are shared by all members of the Hyracoidea.  It 
will also be of interest to examine other members of the Afrotheria and members of the 
Xenarthra to determine if these “unique” features are limited to this order (Hyrocoidea), 
cohort (Afroplacentalia), or are part of a broader phylogenetic grouping (such as the 
Notoplacentalia) (Arnason, 2008).  The current study set out to test several potential 
predictions.  The first prediction, that the hyrax will have many nuclei in common with 
other mammalian species, is supported.  The second prediction, that the hyrax will have 
some unique features is, currently, tentatively supported.  The strength of this second 
prediction and the third prediction can be tested by examining further species of the 
Afrotherian supercohort.  The fourth prediction, that the hyrax may lack nuclei found in 
other species, is supported due to the lack of two catecholaminergic nuclei found either 
exclusively in rodents (C3), or seen in rabbits, tree shrews, megabats and primates (A6d).  
The current study then supports the usefulness of using these systems as indicators of 
phylogenetic relationships and supports current concepts regarding the evolution of these 
neural systems (Manger, 2005). 
 81
Chapter 3 
Distribution of orexin-A immunoreactive neurons and their terminal networks in 
the brain of the rock hyrax, Procavia capensis. 
 
3.1. Introduction 
 Orexin or hypocretin is a hypothalamic neuropeptide that consists of 130-131 
amino acids in mammals and was first described in 1998 by two independent research 
groups. The Sutcliffe group called it hypocretin because it shared similar sequencing 
patterns to the secretin group of peptides and because of its hypothalamic localization 
(De Lecea et al., 1998; Kukkonen et al., 2002). The 130 amino acid precursor peptide 
they identified became known as preprohypocretin with hypocretin 1 and hypocretin 2 
being the final peptides. The term orexin was coined by the Yanagisawa group who 
discovered that this peptide was linked to food intake (Sakurai et al., 1998; Kukkonen et 
al., 2002) and noted that injecting this peptide into the lateral ventricle of non-fasted rats 
would increase food intake. The precursor peptide thus became known as preproorexin 
with orexin-A and orexin B being the final peptides (Sakurai et al., 1998; Kukkonen et 
al., 2002).   
Studies on several mammalian species have reported the location and distribution 
of orexinergic neurons and terminal networks (e.g. humans, Moore et al., 2001; cat, 
Wagner et al., 2000; Zhang et al., 2001, 2002; sheep, Iqbal et al., 2001; a variety of 
rodents, e.g. Broberger et al., 1998; Peyron et al., 1998; Chen et al., 1999; Cutler et al., 
1999; Date et al., 1999; Nixon and Smale, 2007; microchiropterans, Kruger et al., 2010; 
and kangaroo, Yamamoto et al., 2006). These studies all concur regarding the location of 
 82
the orexinergic neurons and the general distribution of their major terminal networks. 
Orexinergic neurons have only been observed within the hypothalamus with the majority 
of the neurons found in the lateral hypothalamic area and the perifornical region; 
however, other studies have also reported isolated neurons in the region of the median 
eminence, posterior, anterior, dorsal and dorsomedial hypothalamus, zona incerta and 
ventrolateral hypothalamus. In contrast to the restricted neuronal distribution, orexinergic 
fibres are widely distributed throughout the brain and spinal cord and exhibit different 
densities of terminal networks in different regions of the brain, for example a high 
density network is evident in the areas immediately surrounding the third ventricle and 
areas involved in the regulation of the sleep-wake cycle (Chen et al., 1999; Cutler et al., 
1999; Hagan et al., 1999; Wagner et al., 2000; McGranagham and Piggins, 2001; Zhang 
et al., 2002, 2004; Espana et al., 2005; Kruger et al., 2010). Orexinergic fibres have been 
reported as being both smooth and varicose, with the varicose type predominating 
(Peyron et al., 1998; Chen et al., 1999; Cutler et al., 1999; Kukkonen et al., 2002). 
Vasoactive intestinal peptide, vasopressin and neuropeptide Y terminal networks provide 
the main innervation to orexinergic neurons while a multitude of other neuronal circuits 
receive orexinergic efferents (Peyron et al., 1998; Cutler et al., 1999; Date et al., 1999; 
Horvath et al., 1999; Nambu et al., 1999; Abrahamson et al., 2001; Backberg et al., 2002; 
Kukkonen et al., 2002). Studies performed on rats to measure the release of orexin have 
revealed that release is closely correlated to circadian activity. These studies found that 
orexin concentrations in the hypothalamus were maximal around light onset and minimal 
at the point of light offset. Diurnal variation was also noted with similar measurements 
taken from the pons and intercisternal space; however, the precise time points were 
 83
shifted compared to the hypothalamus (Taheri et al., 2000; Fujiki et al., 2001; Yoshida et 
al., 2001; Kukkonen et al., 2002). 
The diurnal variation of orexin release has an effect on food consumption, thus 
when orexin is released in higher concentrations (i.e. during the morning) food intake is 
promoted, the opposite being true of low orexin release (i.e. during the night) (Cutler et 
al., 1999; Hagan et al., 1999; McGranaghan and Piggins, 2001; Baldo et al., 2003; 
Khorooshi and Klengenspoor, 2005).  Orexin regulates sleep by exciting neurons that are 
involved in the sleep-wake cycle, such as those of the locus coeruleus, dorsal raphe, and 
pontine cholinergic nuclei. This excitation causes the promotion of wakefulness and 
reduces sleep, especially episodes of REM (Hagan et al., 1999; Bourgin et al., 2000; 
Piper et al., 2000; Wagner et al., 2000; Espana et al., 2001; Huang et al., 2001; 
McGranaghan and Piggins, 2001; Methippara et al., 2000; Kukkonen et al., 2002; 
Yamanaka et al., 2002; Zang et al., 2004; Lee at al., 2005).  
Although many studies deal with the location of orexinergic neurons and terminal 
network distribution, very few studies actually investigate the relation and interplay of 
these neurons and terminal networks to other neural systems. The aim of the present 
study is, thus, to describe the location and distribution of orexinergic neurons and 
terminal networks in the brain of the rock hyrax, Procavia capensis, based on orexin-A 
immunocytochemistry and to compare these to the known cholinergic, catecholaminergic 
and serotonergic systems (Gravett et al., 2009). This may shed light on the possible 
relation and interplay between the orexinergic system and these neural systems. 
Currently, the majority of studies reporting on orexin have made use of nocturnal 
laboratory rodents. Thus, the rock hyrax makes for an interesting candidate for the study 
 84
of the orexinergic system as these mammals are part of the unusual and morphologically 
diverse Afrotheria (Tabuce et al., 2008), for which no report of the orexinergic system is 
available. In addition to this the hyraxes are diurnally active and their sleep has been 
described as being polycyclic (Snyder, 1974). Moreover, they are poor thermoregulators 
(McNairn and Fairall, 1984; Brown and Downs, 2006) and orexins are known to be 
involved in thermoregulation and energy balance as they stimulate energy expenditure, 
increase food intake and suppress metabolic rate, which in turn my lead to hypothermia 
(Szekely et al., 2010; Teske et al., 2010). 
 
3.2. Materials and Methods 
 The brains of six adult female rock hyraxes, P. capensis, were used in the present 
study. Permits from the Limpopo and Gauteng Provincial Governments were obtained for 
the capture and transport of the animals from the wild. All animals were treated and used 
according to the guidelines of the University of the Witwatersrand Animal Ethics 
Committee (approval number AESC 2005/8/5), which parallel those of the NIH for the 
care and use of animals in scientific experimentation. Each animal was weighed, deeply 
anaesthetized and subsequently euthanized with weight appropriate doses of sodium 
pentobarbital (200 mg sodium pentobarbital/kg, i.p.). Upon cessation of respiration the 
animals were perfused intracardially with 0.9% saline followed by 4% paraformaldehyde 
in 0.1M phosphate buffer (PB, pH 7.4) (approximately 1 l/kg of each solution), both 
solutions having a temperature of 4°C. The brains were then carefully removed from the 
skulls and post-fixed overnight in 4% paraformaldehyde in 0.1M PB followed by 
equilibration in 30% sucrose in 0.1M PB. The brains were then frozen in dry ice and with 
 85
the aid of a freezing microtome sectioned at 50 µm in the coronal plane. A one in three 
series of stains was made for Nissl, myelin, and orexin-A.  Sections kept for the Nissl 
series were mounted on 0.5% gelatine coated glass slides, cleared in a solution of 1:1 
chloroform and absolute alcohol, then stained with 1% cresyl violet to reveal cell bodies. 
Sections used for myelin staining were stored in 5% formalin for a period of two weeks 
and were then mounted on 1% gelatine coated glass slides and subsequently stained with 
silver solution to reveal myelin sheaths (Gallyas, 1979). 
  For immunohistochemical staining each section was treated with an endogenous 
inhibitor peroxidase (49.2% methanol: 49.2% 0.1 M PB: 1.6% of 30% H2O2) for 30 min 
and subsequently subjected to three 10 min 0.1 M PB rinses. The sections were then 
preincubated in a solution (blocking buffer) consisting of 3% normal goat serum (NGS, 
Chemicon), 2% bovine serum albumin (BSA, Sigma) and 0.25% Triton X100 (Merck) in 
0.1M PB, at room temperature for 2 h. This was followed by three 10 min rinses in 0.1M 
PB. The sections were then placed for 48 h at 4°C under constant gentle shaking, in 
primary antibody solution, that contained the appropriately diluted primary antibody in 
blocking buffer (see above). The primary antibody used was anti-orexin-A (AB 3704, 
Millipore, raised in rabbit, dilution 1:1500). This was followed by another three 10 min 
rinses in 0.1M PB, after which the sections were incubated for 2 h at room temperature in 
secondary antibody solution. The secondary antibody solution contained a 1:1000 
dilution of biotinylated anti-rabbit IgG (BA-1000, Vector Labs) in 3% normal goat serum 
(NGS), and 2% bovine serum albumin (BSA) in 0.1M PB. Once this was completed, the 
sections were again subjected to another three 10 min rinses in 0.1M PB, followed by a 1 
h incubation in avidin-biotin solution (Vector Labs) and again rinsed. This was followed 
 86
by a 5 min treatment of the sections in a solution consisting of 0.05% diaminobenzidine 
tetrahydrochloride (DAB) in 0.1M PB, after which, and while still in the same solution, 3 
µl of 30% H2O2 per 0.5 ml of solution was added. With the aid of a low power 
stereomicroscope the progression of the staining was visually followed and allowed to 
continue until a level was reached where the background staining could assist in 
reconstruction without obscuring the immunopositive structures. Once this level was 
reached the reaction was stopped by placing the sections in 0.1M PB, followed by a final 
session of three 10 min rinses in 0.1M PB. All solutions used in the 
immunohistochemical process had a pH of 7.4. The immunohistochemically stained 
sections were mounted on 0.5% gelatine coated slides and left to dry overnight. The 
mounted sections were dehydrated by placing it in 70% alcohol for 2 h at room 
temperature under gentle shaking and then transferred through a series of graded 
alcohols, cleared in xylene and coverslipped with Depex.  
The sections were observed with a low power stereomicroscope, and the 
architectonic borders traced according to the Nissl and myelin stained sections using a 
camera lucida. The immunostained sections were then matched to the drawings and the 
immunopositive neurons marked, in addition the density of axon terminal staining was 
graded from low to high for each immunostained section and medium and high marked 
on the drawings. The drawings were scanned and redrawn with the aid of the Canvas 8 
program. Location and distribution of orexin immunopositive (Orx+) neurons and 
terminal networks were described in relation to the general neuroanatomy of the rock 
hyrax brain and previously described cholinergic, catecholaminergic, and serotonergic 
systems (Gravett et al., 2009).  
 87
3.3. Abbreviations 
III – oculomotor nucleus 
3V – third ventricle 
4V – fourth ventricle 
A1 – caudal ventrolateral medullary tegmental nucleus 
A2 – caudal dorsomedial medullary nucleus 
A6 – locus coeruleus 
A7 – subcoeruleus  
A7d – nucleus subcoeruleus, diffuse portion 
A7sc – nucleus subcoeruleus, compact portion 
A9 – substantia nigra 
A10 – ventral tegmental area 
A10c – ventral tegmental area, central 
A10d – ventral tegmental area, dorsal 
A10dc – ventral tegmental area, dorsal caudal 
A11 – caudal diencephalic group 
A12 – tuberal cell group 
A13 – zona incerta 
A14 – rostral periventricular nucleus 
A15d – anterior hypothalamic group, dorsal division 
A15v – anterior hypothalamic group, ventral division 
ac – anterior commissure 
AD – anterodorsal nucleus of the dorsal thalamus 
 88
Amyg – amygdala 
AP – area postrema 
AV – anteroventral nucleus of the dorsal thalamus 
B9 – supralemniscal serotonergic nucleus 
C – caudate nucleus 
C1 – rostral ventrolateral medullary tegmental group 
C2 – rostral dorsomedial medullary nucleus 
ca – cerebral aqueduct 
Cb – cerebellum 
cc – corpus callosum 
Cl – claustrum 
CLi – caudal linear nucleus 
CN – cochlear nucleus 
CP – cerebral peduncle 
CVL – caudal ventrolateral serotonergic group 
DCN – deep cerebellar nuclei 
Diag.B – diagonal band of Broca 
DR – dorsal raphe 
DRc – dorsal raphe nucleus, caudal division 
DRd – dorsal raphe nucleus, dorsal division 
DRif – dorsal raphe nucleus, interfascicular division 
DRl – dorsal raphe nucleus, lateral division 
DRp – dorsal raphe nucleus, peripheral division 
 89
DRv – dorsal raphe nucleus, ventral division 
DT – dorsal thalamus 
EW – Edinger-Westphal nucleus 
f – fornix 
fr – fasciculus retroflexus 
GC – periaqueductal grey matter 
GLD – dorsal lateral geniculate nucleus 
GP – globus pallidus 
Hb – habenular nuclei 
Hip – hippocampus 
Hyp – hypothalamus 
Hyp.d – dorsal hypothalamic cholinergic nucleus 
Hyp.l – lateral hypothalamic cholinergic nucleus 
Hyp.v – ventral hypothalamic cholinergic nucleus 
IC – inferior colliculus 
ic – internal capsule 
icp – inferior cerebellar peduncle 
IGL – intergeniculate leaflet 
io – inferior olivary nuclei 
IP – interpeduncular nucleus 
LDT – laterodorsal tegmental nucleus 
LOT – lateral olfactory tract 
LRT – lateral reticular nucleus 
 90
LV – lateral ventricle 
Mc – main cluster of orexinergic immunoreactive neurons 
mcp – middle cerebellar peduncle 
MnR – median raphe nucleus 
N.Acc – nucleus accumbens 
N.Bas – nucleus basalis 
NEO – neocortex 
OC – optic chiasm 
OT – optic tract 
OTc – optict tract cluster of orexinergic immunoreactive neurons  
P – putamen 
PV – paraventricular nucleus 
PBg – parabigeminal nucleus 
PIR – piriform cortex 
PPT – pedunculopontine nucleus 
Pta – pretectal area 
py – pyramidal tract 
pyx – decussation of the pyramidal tract 
R – thalamic reticular nucleus 
RMg – raphe magnus nucleus 
RN – red nucleus 
ROb – raphe obscurus nucleus 
RPa – raphe pallidus nucleus 
 91
RVL – rostral ventrolateral serotonergic group 
S – septum  
SC – superior colliculus 
SpV – spinal trigeminal tract 
TOL – olfactory tubercle 
vh – ventral horn of spinal cord 
VPO – ventral pontine nucleus 
ZI – zona incerta 
ZIc – zona incerta cluster of orexinergic immunoreactive neurons 
 
3.4. Results 
The results revealed that the distribution of orexin-A immunopositive (Orx+) 
neuron cell bodies and terminal networks did not, for the most part, differ significantly 
from what has been described in other mammals. Despite this, an interesting difference 
was noted that has not been reported in most mammals studied to date, namely a strong 
Orx+ terminal network in the anterior dorsal nucleus of the dorsal thalamus (AD). Orx+ 
neurons and terminal networks are described in relation to either the general 
neuroanatomy of the rock hyrax brain and the cholinergic, catecholaminergic, and 
serotonergic systems (as described previously for this species in Gravett et al., 2009). 
 
3.4.1. Orexin-A neuronal cell body distribution 
 Orx+ neuronal cell bodies were identified mainly within the hypothalamus of the 
brain of the rock hyrax. These cell bodies were grouped into three distinct clusters; the 
 92
main cluster, the zona incerta cluster and the optic tract cluster (Figs. 3.1-3). The main 
cluster was identified as a large group of densely packed Orx+ neuronal cell bodies 
located lateral to the third ventricle in the perifornical region, with a large number of 
neuronal cell bodies extending medially from this area, as well as into the dorsomedial 
and lateral hypothalamic areas. The majority of Orx+ neurons belonged to this cluster and 
they were found to be ovoid in shape, a mixture of both bipolar and multipolar types and 
showed no apparent regular dendritic orientation. The Orx+ neuronal cell bodies of this 
main cluster were intermingled with the A11 and A14 nuclear groups of the 
catecholaminergic system. From the main cluster a group of Orx+ neuronal cell bodies 
extended laterally into the region of the zona incerta. This cluster had a very low density 
of Orx+ neurons that were mixed with neurons of the lateral hypothalamic cholinergic 
nucleus and the A13 nucleus of the catecholaminergic system. The cell bodies were ovoid 
in shape, a mixture of bipolar and multipolar type and had no specific dendritic 
orientation. The third cluster extended ventrolaterally from the main cluster to the 
ventrolateral region of the hypothalamus adjacent to the optic tract. This cluster exhibited 
a low density of Orx+ neuronal cell bodies that were mostly of the bipolar type with the 
same cell morphology as described for the other clusters. These neurons were found to 
intermingle with those of the ventral hypothalamic cholinergic nucleus (Figs. 3.1 E-G, 
3.2). 
 
 
 
 
 93
3.4.2. Orexin-A terminal network distribution 
3.4.2.1. Telencephalon  
 Within the telencephalon of the rock hyrax brain orexinergic terminal networks of 
medium density were noted. These terminal networks were mainly distributed along the 
medial and ventrolateral border of the telencephalon and located in and around the 
following structures: the nucleus accumbens, olfactory tubercle and islands of Calleja, 
diagonal band of Broca and the septal nuclear complex (Figs. 3.1 A-C, 3.4). It is of 
interest to note than in all these structures a number of cholinergic neurons were also 
located (Gravett et al., 2009). The remaining regions of the telencephalon had either a 
low density Orx+ terminal network, or no Orx+ structures could be observed. 
 
3.4.2.2. Diencephalon  
 Both high and medium density orexinergic terminal networks were observed 
throughout the diencephalon. The high density terminal networks were located in the 
anterior hypothalamic area, the periventricular areas adjacent to the third ventricle in the 
hypothalamus (Fig. 3.2), in the paraventricular nuclei of the epithalamus (Fig. 3.5A) 
extending to a region surrounding the habenular nuclei, and dorsolaterally in the AD 
(Figs. 3.1 D-J, 3.5). The medium density terminal networks surrounded these areas of 
high density and were distributed through most parts of the hypothalamus (Fig. 3.2), the 
midline of the dorsal thalamus in the intralaminar nuclei and extended laterally to the 
intergeniculate leaflet (Fig. 3.5C). The high-density terminal networks overlapped with 
the A12, A14, and A15v nuclei of the catecholaminergic system, and the AD (Fig. 3.5B) 
that forms part of the cholinergic system of the rock hyrax (Gravett et al., 2009). The 
 94
Orx+ terminal networks of medium density were found to overlap the A11 and A13 
nuclei of the catecholaminergic system and the dorsal, lateral, and ventral hypothalamic 
nuclei of the cholinergic system. The remaining regions of the diencephalon had either a 
low density Orx+ terminal network, or no Orx+ structures could be observed (Fig. 3.2). 
 
3.4.2.3. Midbrain and Pons (Mesencephalon and Metencephalon) 
 Both high and medium dense orexinergic terminal networks characterized the 
midbrain and pontine regions; however, the medium density Orx+ terminal networks 
were found to dominate in these regions. The terminal networks of high density were 
located ventral to the cerebral aqueduct within the midline of the midbrain and pontine 
regions and overlapped the A10d and A10dc nuclei of the catecholaminergic system and 
most serotonergic nuclei of the dorsal raphe (Fig. 3.6A). The medium dense Orx+ 
terminal networks were coincident with the dopaminergic ventral tegmental area, the 
noradrenergic locus coeruleus complex (Fig. 3.6B), the cholinergic nuclei of the 
laterodorsal tegmental and pedunculopontine nuclei (both magno and parvocellular 
divisions of these nuclei, Gravett et al., 2009), the cholinergic parabigeminal and 
Edinger-Westphal nuclei, and the caudal linear (CLi), supralemniscal (B9), caudal 
nucleus of the dorsal raphe (DRc) and median raphe (MnR) serotonergic nuclei (Fig. 
3.6C). In addition to this, the entire periaqueductal gray matter was observed to have a 
medium density Orx+ terminal network that extended outward through all layers of the 
superior colliculus (Figs. 3.1 K-N, 3.6). The remaining regions of the midbrain and pons 
had either a low density Orx+ terminal network, or no Orx+ structures could be observed. 
 95
It is worth noting that we could only find a low density Orx+ network in the inferior 
colliculus. 
  
3.4.2.4. Medulla Oblongata (Myelencephalon) 
 Only medium and low-density Orx+ terminal networks were present within the 
medulla oblongata of the brain of the rock hyrax. The medium density networks were 
found in all regions where serotonergic neurons were located (i.e. the raphe magnus, 
rostral and caudal ventrolateral columns, raphe pallidus and raphe obscurus).   Similarly, 
all regions that contained catecholaminergic neurons were found to have a medium 
density Orx+ terminal network, these being the area postrema, C1, C2, A1 and A2 nuclei. 
Anterior in the medulla a medium density Orx+ terminal network was found in the region 
of the parabrachial nuclear complex. Towards the spinal cord these medium density Orx+ 
networks were located in the areas surrounding the central canal and the outer edge of the 
dorsal horn in the spinal trigeminal tract (SpV) (Fig. 3.1 O-V). The remaining regions of 
the medulla had either a low density Orx+ terminal network, or no Orx+ structures could 
be observed. 
 
 96
 
 
 
 
 
Figure 3.1: Drawings of coronal sections through one half of the brain of the rock hyrax, 
illustrating orexin-A immunoreactive neurons and terminal network distribution. A single 
black dot indicates a single cell body, black and grey shaded areas indicate regions in the 
brain of the rock hyrax where high and medium density, respectively, orexin-A terminal 
networks were observed. Areas where no shading is evident represent regions of low or 
no orexinergic innervation. Drawing A represents the most rostral section, V the most 
caudal, and each drawing is approximately 1500 µm apart. See list for abbreviations.  
 



 100
 
 
 
 
 
Figure 3.2: Photomicrograph montage demonstrating the location of orexin-A 
immunoreactive neurons and terminal networks within the hypothalamus of the rock 
hyrax showing the main cluster (Mc), the zona incerta cluster (ZIc) and the optic tract 
cluster (OTc). Scale = 1 mm. 3V – third ventricle; f – fornix; OT – optic tract. 
 

 102
 
 
 
 
 
Figure 3.3: Photomicrographs illustrating the neuronal morphology of the three clusters 
of orexin-A immunoreactive nuclear groups within the hypothalamus of the rock hyrax. 
(A) Orexin-A immunoreactive neurons of the main cluster, perifornical region. (B) 
Orexin-A immunoreactive neurons of the optic tract cluster, lateral ventral hypothalamic 
area. (C) Orexin-A immunoreactive neurons of the zona incerta cluster, lateral 
hypothalamic area. In all images, medial is to the left, lateral to the right, and dorsal to 
the top. Scale in C = 500 µm and applies to all.  
 

 104
 
 
 
 
 
Figure 3.4: Photomicrographs illustrating the distribution of orexin-A immunoreactive 
terminal networks within certain parts of the forebrain of the rock hyrax. (A) Low density 
orexin-A immunoreactive terminal networks within the cerebral cortex. (B) Medium 
density orexin-A immunoreactive terminal networks within the diagonal band of Broca 
(Diag. B). (C) Medium density orexin-A immunoreactive terminal networks within the 
nucleus accumbens (N. Acc). (D) Medium density orexin-A immunoreactive terminal 
networks within the olfactory tubercle (TOL). Scale in D = 500 µm and applies to all. 
Medial is to the left in all images.  
 

 106
 
 
 
 
 
Figure 3.5: Photomicrograph illustrating the distribution of orexin-A immunoreactive 
terminal networks within the midbrain and pontine regions of the rock hyrax. (A) High 
density orexin-A immunoreactive terminal networks within the serotonergic dorsal raphe 
(DR) nuclear complex. Scale bar = 1 mm. (B) Medium density orexin-A immunoreactive 
terminal networks within the locus coeruleus nuclear complex. (C) Medium density 
orexin-A immunoreactive terminal networks within the serotonergic caudal linear 
nucleus (CLi) and supralemniscal nucleus (B9). Scale in C = 500 µm and applies to B 
and C. A6 – locus coeruleus; A7d – subcoeruleus diffuse region; A7sc, subcoeruleus 
compact region; ca – cerebral aqueduct; DRd – dorsal division of dorsal raphe; DRl – 
lateral division of dorsal raphe; DRp – peripheral division of dorsal raphe; DRv – ventral 
division of dorsal raphe. Medial is to the left in all images. 
 

 108
 
 
 
 
 
Figure 3.6: Photomicrograph illustrating orexin-A immunoreactive terminal networks 
within the diencephalon of the rock hyrax. (A) High density orexin-A immunoreactive 
terminal networks within the paraventricular thalamic nucleus (PV) and the lateral 
habenular nuclei along the border of the stria medullaris (Hb). Scale bar = 1 mm. (B) 
High density orexin-A immunoreactive terminal networks within the anterodorsal 
nucleus of the dorsal thalamus (AD). (C) Medium density orexin-A immunoreactive 
terminal networks within the intergeniculate leaflet (IGL) mainly oriented in the ventro-
dorsal direction. Scale in C = 500 µm and applies to B and C. AV – anteroventral nucleus 
of the dorsal thalamus. Medial is to the left in all images 
 

 110
3.5. Discussion 
The aim of the present study was to identify and describe the neuronal location, 
nuclear organization and terminal network distribution of the orexinergic system in the 
brain of the rock hyrax, Procavia capensis by means of orexin-A immunocytochemistry. 
The study also examined the topological relationships of the Orx+ neurons and terminal 
networks with the previously described cholinergic, catecholaminergic and serotonergic 
systems in the brain of the rock hyrax (Gravett et al., 2009). The results revealed that the 
distribution and nuclear organization of the Orx+ neurons of the rock hyrax conforms 
closely to what has been described previously for other mammals (Broberger et al., 1998; 
Peyron et al., 1998; Moore et al., 2001; Zang et al., 2001, 2002; Nixon and Smale, 2007; 
Yamamoto., 2006; Kruger et al., 2010). The orexin-A immunoreactive terminal network 
distribution also showed a similar architecture as described in other mammalian species; 
however the anterodorsal nucleus of the thalamus was observed to have a dense 
orexinergic terminal network. Thus, the orexinergic system, as demonstrated with orexin-
A immunoreactivity, within the rock hyrax brain is similar to what has been described in 
other mammals, but there is a feature of the terminal network distribution that has only 
been reported in a small number of other mammals.  
 
3.5.1. Orexinergic neuronal distribution  
 The location of orexinergic neuronal cell bodies has been described in 
representative species of all vertebrate classes – mammals (e.g. Wagner et al., 2000; Iqbal 
et al., 2001; Moore et al., 2001; Yamamoto et al., 2006; Nixon and Smale, 2007; Kruger 
et al., 2010; Stoyanova at al., 2010), birds (e.g. Ohkubo et al., 2002), reptiles (Domínguez 
 111
et al., 2010), amphibians (e.g. Shibahara et al., 1999; Galas et al., 2001; Singletary et al., 
2005; Suzuki et al., 2008; López et al., 2009) and fish (e.g. Kaslin et al., 2004; Huesa et 
al., 2005; Yokogawa et al., 2007). All these studies have revealed that for the most part 
orexinergic neurons were located within the hypothalamus. It is clear from previous 
studies that this system shows a degree of variability within and between different 
vertebrate classes, but also a strong degree of similarity (e.g. Suzuki et al., 2008; 
Domínguez et al., 2010). In mammals the majority of orexinergic neurons are located 
within the lateral hypothalamic area and the perifornical region (which we term the main 
cluster, Kruger et al., 2010); however, other studies in mammals have also reported 
isolated neurons in the region of the median eminence, posterior, dorsal and dorsomedial 
hypothalamus (Chen et al., 1999; Cutler et al., 1999; Hagan et al., 1999; Wagner et al., 
2000; McGranaghan and Piggins, 2001; Zhang et al., 2002, 2004; Espana et al., 2005). In 
the rock hyrax we could subdivide the orexinergic neurons into three distinct clusters: the 
main cluster, the zona incerta cluster and the optic tract cluster. While these clusters have 
been reported present in most mammals studied to date it has been found that 
microchiropterans and hamsters lack the optic tract cluster (Wagner et al., 2000; 
McGranaghan and Piggins, 2001; Kruger et al., 2010). The nuclear organization of the 
orexinergic neurons in the rock hyrax is therefore what may be described as very 
typically mammalian.  It would seem that, apart from minor differences, the nuclear 
organization of the orexinergic neurons is strongly conserved across mammals. An 
important aspect of orexinergic nuclear parcellation relates to the connectivity of this 
system.  For example, the zona incerta cluster specifically projects to the intergeniculate 
leaflet (Vidal et al., 2005) and what we term the main orexinergic cluster may be 
 112
subdivided into perifornical and lateral hypothalamic nuclei (Yoshida et al., 2006). Thus, 
in the hyrax we may be observing 4 specific nuclei within this system, however, further 
study on the specific connectivity of this system would be required to justify further 
nuclear parcellation than that presented here. 
In conjunction with neuronal location and distribution, the present study also 
examined the topographical interrelationships of the orexinergic neurons with the 
previously described cholinergic, catecholaminergic and serotonergic systems in the brain 
of the rock hyrax (Gravett et al., 2009).  The observations revealed that orexinergic 
neurons of the main cluster were topographically coincident with the A11 and A14 
nuclear groups of the catecholaminergic system; the zona incerta cluster was 
topographically coincident with the A13 nucleus of the catecholaminergic system and the 
lateral hypothalamic nucleus of the cholinergic system; the optic tract cluster was 
topographically coincident with the cholinergic ventral hypothalamic nucleus. It has been 
reported that the membrane potential of orexinergic neurons are hyperpolarized by 
noradrenaline and serotonin and depolarized by acetylcholine (Yamanaka et al., 2003). 
Thus, orexinergic neurons are possibly rapidly depolarized where they intermingle with 
neurons from the cholinergic system. The potential intrinsic hypothalamic circuitry of 
these systems, due to the topological congruency, would be an avenue of interesting 
further study given the roles these systems play in the sleep-wake cycle, satiety and 
motivational states.    
 
 
 
 113
3.5.2. Orexinergic terminal network distribution 
The orexinergic terminal network distribution within the brain of the rock hyrax 
was found to be similar to that described in other mammalian species studied to date, 
especially in the manner that these terminal networks are found primarily along the 
midline and ventricular surfaces; however, the anterodorsal nucleus of the dorsal 
thalamus (AD) of the rock hyrax exhibited a high-density orexinergic terminal network 
which is an apparent variance to most other mammals (see below).  
The majority of the mammalian brain is in receipt of either very low or no 
orexinergic innervation, and this pattern is evident in the cerebral cortex, striatopallidal 
complex, hippocampal complex, amygdalar complex, the majority of the dorsal thalamus, 
brainstem and cerebellum of the rock hyrax brain. Medium dense orexinergic terminal 
networks were observed within regions of the telencephalon, diencephalon, midbrain and 
the medulla of the brain of the rock hyrax, whereas a high density orexinergic innervation 
was only evident in certain regions of the diencephalon and midbrain. This distribution of 
medium to high dense orexinergic terminal networks once again resembled the typical 
mammalian organizational plan for this system as reported in previous studies. 
Throughout the telencephalon, diencephalon, midbrain and medullary regions the 
orexinergic terminal networks were found intermingled with several nuclear complexes 
of the catecholaminergic, cholinergic and serotonergic systems (Gravett et al., 2009).  
The regions within the telencephalon that exhibited a medium density of 
orexinergic terminal networks included the nucleus accumbens, the olfactory tubercle, the 
diagonal band of Broca, and the septal nuclear complex. This distribution of medium 
density orexinergic terminal networks within the telencephalon of the rock hyrax 
 114
coincides with previous reports for this system in other mammals. Within the 
diencephalon a medium density of orexinergic innervation was observed in the region of 
the intergeniculate leaflet, whereas a high density was observed in regions immediately 
surrounding the third ventricle, the midline of the dorsal thalamus as well as the AD 
nucleus (see below). The medium density orexinergic terminal networks overlapped the 
medial septal nucleus, dorsal, lateral and ventral hypothalamic nuclei of the cholinergic 
system and the A11 and A13 nuclear complexes of the catecholaminergic system. The 
distribution of medium to high density orexinergic terminal networks in the regions of the 
diencephalon of the rock hyrax brain did not differ significantly to published reports in 
other mammals for this region. However, it did differ to other mammals with respect to 
the orexinergic innervation of the AD nucleus, which has only been previously reported 
to occur in the Nile grass rat, Arvicanthis niloticus (Novak and Albers, 2002; Nixon and 
Smale, 2007) and the Syrian hamster, Mesocricetus auratus (Mintz et al., 2001). A 
medium density orexinergic terminal network distribution dominated the midbrain and 
pontine regions in the rock hyrax brain, however high density terminal networks were 
observed within the dopaminergic ventral tegmental area (VTA) and serotonergic dorsal 
raphe nuclei. A study by Nakamura et al. (2000) has reported that the calcium 
concentration in isolated A10 neurons is increased by orexin-A and this in turn induces 
hyperlocomotion, stereotypy and grooming in rats. Orexin-A has further been reported to 
have an excitatory effect on the serotonergic neurons of the dorsal raphe nucleus, which 
play an important role in the sleep-wake cycle as well as regulation of food intake and 
mood control (Brown et al., 2001; Matsuzaki et al., 2002; Tao et al., 2006). A moderate 
density of orexinergic terminal networks was identified in the noradrenergic locus 
 115
coeruleus and subcoeruleus, and cholinergic laterodorsal tegmental (LDT) and 
pedunculopontine tegmental (PPT) nuclei. These structures are known to play a crucial 
part in the sleep-wake cycle and have been reported to be activated by orexin. 
Orexinergic input to the cells of the locus coeruleus facilitate the waking state, whereas 
orexinergic innervation to the LDT and PPT play a role in the regulation of REM sleep 
(Ohno and Sakurai, 2008). Finally, the medullary region exhibited a medium dense 
orexinergic terminal network distribution in the areas ventral to the fourth ventricle and 
regions surrounding the central canal, whereas the remainder of this region was 
characterized by a low or absent orexinergic terminal network distribution. These 
terminal networks furthermore overlapped the RPa, RVL, RMg, ROb and CVL nuclei of 
the serotonergic system, the C1-2 and A1-2 nuclei of the catecholaminergic system, the 
nucleus ambiguus, the dorsal motor vagus and hypoglossal nuclei of the cholinergic 
system. Thus, for the most part the orexinergic terminal network distribution within the 
brain of the rock hyrax typically resembles the general mammalian organizational plan 
for this system.  
 
3.5.3. Orexinergic innervation of the anterodorsal nucleus of the dorsal thalamus and 
diurnality? 
The majority of studies that describe the distribution of orexinergic terminal 
networks have examined nocturnal mammals. None of these studies reported either a 
dense or moderate orexinergic innervation of the AD. Two previous reports of moderate 
to dense orexinergic innervation of the AD examined the Nile grass rat, Arvicanthis 
niloticus (Novak and Albers, 2002; Nixon and Smale, 2007) and the Syrian hamster, 
 116
Mesocricetus auratus (Mintz et al., 2001).  Novak and Albers (2002) reported a low 
density of orexinergic fibers in the AD nucleus and a high density in the anteroventral 
(AV) nucleus of the dorsal thalamus of the grass rat when the orexin B antibody was 
used, while Nixon and Smale (2007) reported a low density for the AV nucleus with the 
orexin-A antibody and a medium density in the AD nucleus with both orexin-A and B 
antibodies. The rock hyrax and the Syrian hamster (Mintz et al., 2001) both showed a 
high density of orexinergic terminals within the AD nucleus with an orexin-A antibody.  
Of potential functional correlation to this is the fact that the rock hyrax (Olds and 
Shoshani, 1982), Nile grass rat (Novak et al., 2000) and Syrian hamster (Gatterman et al., 
2008) are all diurnal species. Thus, it is possible that orexinergic innervation of the AD 
nucleus, the AV nucleus, or the anterior nuclear complex as a whole, might be a unique 
feature associated with diurnal mammals. This possibility needs to be tested across a 
range of mammalian species to determine whether this potentially very interesting 
finding is correct. 
 117
Chapter 4 
Solitary sleep in the rock hyrax, Procavia capensis. 
 
4.1. Introduction 
 Sleep is typically defined as a homeostatically regulated, easily reversible state of 
sustained quiescence accompanied by reduced sensory responsiveness. Mammalian sleep 
has been physiologically divided into non rapid eye movement sleep (NREM), often 
called slow wave sleep (SWS), and rapid eye movement sleep (REM) (Nicolau et al., 
2000; Zepelin et al., 2005; Cirelli and Tononi, 2008; Lesku et al., 2008; Siegel, 2008). No 
consensus has been reached as to the functions of REM and NREM. One major clue to 
understanding these functions is an examination of how sleep amounts vary across 
species with differing physiological and ecological specializations (Tobler, 2005; Zepelin 
et al., 2005; Siegel, 2008; Horne, 2009; Rial et al., 2010).  
A recent review (McNamara et al., 2008) detailed that sleep studies have been 
undertaken for approximately 127 different mammalian species representing 46 families 
from 17 different orders. They concluded that on average mammals sleep for 
approximately 12 h, with REM occupying approximately 17% of total sleep time; 
however, this is a generalization, as total sleep time varies significantly among different 
mammalian species. For instance, total sleep time (TST) in the rabbit, Oryctolaus 
cuniculus, is approximately 8.8 h (Spiess et al., 1970), TST in the laboratory mouse, Mus 
musculus, amounts to 12.8 h (Valatx and Bugat, 1974) compared to the horse, Equus 
caballus, where the TST is 2.9 h (Ruckebusch, 1972). In addition to this variation in 
sleep times, some mammals show some unusual forms of sleep – cetaceans exhibit 
 118
unihemispheric SWS (Lyamin et al., 2008), the ferret, armadillo and possum show 
unusually high amounts of REM sleep (Siegel, 2005), the fur seal has very little REM 
sleep while in the water compared to REM amounts when they are on land (Lyamin et al., 
2008), and the platypus which has up to 8h/day of REM sleep, but lacks cortical 
desynchronization during much of this sleep phase (Siegel et al., 1999).   
The rock hyrax belongs to the order Hyracoidea, which in turn is part of the clade 
Afrotheria (Tabuce et al., 2008). The Afrotherian clade is an unusual grouping, as 
although a strong molecular support for this clade exists, the morphological support is 
equivocal. The Afrotherian species show a vast range of body sizes, morphologies, life 
histories and ecological niches. For members of the Afrotheria, sleep has been recorded 
behaviourally in captive Asian and African Elephants, with TST in these animals being 4-
6.5 h and 3.3 h respectively (Kurt, 1960; Tobler, 1992). EEG recordings of a fully aquatic 
Amazonian manatee, Trichechus inungius, showed that this animal spent 27% of the 
recorded time in SWS, which could be unihemispheric in nature, and 1% in paradoxical, 
or REM sleep (Mukhametov et al., 1992); however, the study was performed on a single 
animal and the sleep recording commenced directly after surgery and lasted only a few 
days. Within the Hyracoidea Snyder (1974) reports a TST of 4.9 h for Procavia johnstoni 
(rock hyrax), 5.7 h for Heterohyrax brucei (bush hyrax), and 4.9 h for Dendrohyrax 
validus (tree hyrax); however, only an abstract has been published for the aforementioned 
study and no EEG spectral evidence have been provided to support the TST reported.  
Reports of sleep in other Afrotherian species are not known to the authors. In the present 
study the physiological measurable parameters of sleep and the accompanying behaviour 
 119
were recorded using telemetry in the rock hyrax, Procavia capensis, a previously 
unstudied hyrax species.  
 
4.2. Materials and Methods  
 A total of five adult rock hyraxes, P. capensis, (4 male and 1 female) with body 
weights ranging between 1.74 – 4.3 kg (Table 4.1), were used in the present study. 
Permits from the Limpopo and Gauteng Provincial Governments were obtained for the 
capture and transport of the animals from the wild. All animals were treated and used 
according to the guidelines of the University of the Witwatersrand Animal Ethics 
Committee (approval number AESC 2005/8/5), which parallel those of the NIH for the 
care and use of animals in scientific experimentation. The animals were captured at 
random from wild populations and thereafter allowed to acclimatize for a period of one 
month to the recording enclosures that had a 12-12 lighting schedule with temperature 
maintained between 19-21°C. Each animal was implanted with a telemetric recording 
devise (Data Sciences International) that allowed for the recording of physiologically 
measurable parameters of sleep without cables or restraint.  The chamber in which 
recording occurred was 1.8 m x 1.5 m with a painted concrete surface that was covered 
with straw. The height of the chamber was approximately 1.5 m and steel mesh was 
placed over the chamber to prevent the animals from escaping. A wooden box (90 x 90 x 
30 cm) with a Perspex roof and two entrances was placed inside the chamber and food 
(combinations of cucumber, tomato, sweet potato, pumpkin, apples and rabbit pellets as a 
source of roughage) and fresh water was supplied daily. Behaviour was also recorded 
with a low light CCD digital camera connected to a DVD recorder.   
 120
4.2.1. Surgical procedure 
After acclimatization, surgical implantation of the telemetric recoding device was 
performed. The animals were weighed before surgery and anesthetized with weight 
appropriate doses of a 2:1 mixture of ketamine and xylazine (Anaket-V and Chanazine 
2% Injection, Bayer HealthCare). The head and neck, left thoracic (two 2 cm x 1 cm) and 
abdominal (10 cm x 10 cm) regions were shaved and cleaned with CHX Chlorhexidine 
Disinfectant (0.5% chlorhexidine diglucunate in 75% alcohol, Kyron Laboratories (PTY) 
LTD) before surgery commenced. These areas correspond, respectively, to the regions 
were the EEG (electroencephalogram), EMG (electromyogram) and ECG 
(electrocardiogram) electrodes and telemeter would be implanted. The animal was placed 
on a heated blanket in order to maintain a constant body temperature throughout the 
surgery and the head was securely placed in a stereotaxic apparatus to prevent movement 
and allow for the accurate placement of the EEG and EMG electrodes. During the 
surgical procedure the animal was kept under a constant state of anaesthesia by means of 
isoflurane ventilation (1-2% in an oxygen/70% nitrous oxide mixture, Isofor Inhalation 
Anaesthetic – Safe Line Pharmaceuticals (Pty) Ltd). The animal’s heart rate, body 
temperature and percentage oxygen saturation was monitored.    
Under aseptic conditions, a mid-sagittal incision was made over the skull and the 
skin and temporal muscle were reflected to expose the part of the skull overlying the 
motor cortex. Using a dental drill, three 2 mm diameter holes were made in the cranial 
vault to expose the underlying dura mater. The first hole was drilled anterior to the 
olfactory bulbs for the placement of the indifferent electrode, while two holes were 
drilled approximately 5 mm apart just lateral to the sagittal sinus over the left motor 
 121
cortex for the placement of the stainless steel recording electrodes (gauge of electrode 
0.457 mm, silastic outside diameter 0.9 mm and inside diameter 0.508 mm, PhysioTel ® 
Multiplus Transmitter, Data Sciences International). The electrodes were placed in such a 
manner that the tips rested firmly on the surface of the cortex but did not pierce the dura 
mater and were secured in place with dental cement. Two stainless steel EMG electrodes 
(gauge of electrode 0.457 mm, silastic outside diameter 0.9 mm and inside diameter 
0.508 mm, PhysioTel ® Multiplus Transmitter, Data Sciences International) (1.5 cm 
apart) were sutured into the dorsal nuchal musculature, while two ECG electrodes (3 cm 
apart) were sutured into the subdermal tissue overlying the rib cage in the left thoracic 
region. A subcutaneous pocket was created (10 cm x 10 cm) over the left abdominal 
region, which allowed for the implantation of the telemetry unit. All skin incisions were 
sutured following implantation. After surgery was complete the animal was given an 
intramuscular analgesic (0.1 ml Tamgestic, Schering-Plough, mixed with 0.9 ml sterile 
water, 1 ml mixture/kg) and returned to the recording enclosure. Recovery was monitored 
every half hour until it could be established that the animal was able to move freely and 
eat/drink normally.  
 
4.2.2. Sleep recording 
After the surgical procedure the animal was allowed a recovery period of one 
week before the recording of sleep commenced. The animals were housed in the same 
recording enclosure for recovery as well as recording. A receiver was mounted and 
secured to one wall of the enclosure while a low light CCD digital camera was mounted 
above the enclosure. The telemetric recording system (Data Sciences International, DSI, 
 122
PhysioTel ® Multiplus Transmitter, model TL10M3-D70-EEE implant) consisted of a 
DEM multiplex interface to which the receiver was connected. The signal from the 
implanted transmitter (round, 13cm² with stainless steel electrodes, weight – 37 g, 
volume – 25 ml, 3 channels) detected by the receiver was relayed to the input amplifier of 
the Data Sciences computer system, after which it was digitally recorded (in DSI format) 
for analysis. After the recording was completed, data digitally saved in the DSI format 
was converted to text format and these files were in turn converted into the appropriate 
format needed for recognition and analysis by the Spike 2 computer program (version 
4.2, Cambridge Electronic Design). Sleep (physiologically measurable parameters and 
the DVD recorded associated behaviour) was recorded continuously for a period of 72 
hours. The recording enclosure was in a sound attenuating room. The animals were 
disturbed only once a day for approximately 5 min at the same time during each of the 
recording days for feeding.   
 
4.2.3. Data analysis 
Version 4.2.2 of the Spike 2 software (Cambridge Electronic Designs, UC) was 
used in order to convert the recorded data into the appropriate format, i.e. Spike 2 data 
format, for offline analysis. The EEG data was subsequently scored in 5 s epochs as 
either waking characterized by low voltage high frequency EEG and high voltage EMG, 
SWS characterized by high voltage low frequency EEG and EMG lower in amplitude 
than waking, and REM (rapid eye movement). REM in the rock hyrax was found to be 
ambiguous and did not resemble the classical definition of REM sleep, thus REM0, was 
characterized by low voltage high frequency EEG, an EMG that was almost atonic, and 
 123
an irregular ECG; whilst REM1 was characterized by a low voltage high frequency EEG, 
EMG amplitude characteristic of SWS and a regular ECG (Figs. 4.1-6).  An epoch was 
only assigned to a particular state if the state occupied at least 50% of the epoch. The data 
obtained from the 5 s epoch scoring was analyzed to determine the modal state per 
minute to generate the data for the 1 min epoch scoring. An unpaired two-tailed t-test 
(parametric data) or Mann-Whitney (nonparametric data) test (p<0.05) was employed to 
determine if a statistically significant differences existed between the 5 s and 1 min epoch 
scoring methods. Behaviour was scored in 1 min epochs as either immobile – animal was 
completely immobile for >30 s, quiet waking – animal was immobile and only moving its 
head or made minor movements in the same place for >30 s, active waking – animal was 
actively moving around for >30 s, or eating/drinking – animal was eating and/or dinking 
for >30 s (Figs. 4.5-6).  
All the physiologically and behaviourally scored data was tested for normality 
prior to any statistical testing. The Kolmogrov and Smirnov test was conducted to 
determine whether the data was parametric or nonparametric in nature (p>0.05). A one-
way ANOVA was used in all comparisons where the data was parametric and a 
statistically significant difference between sample means were obtained in cases where 
p<0.05. Where it was found that the data was nonparametric the Kruskal-Wallis test was 
employed and a statistically significant difference between sample medians were 
obtained in all cases where p<0.05. Comparisons between light and dark were made by 
means of contingency tables, where in the case of parametric data the unpaired two-tailed 
T-test and in the case of nonparametric data the Mann-Whitney test was employed to 
 124
determine whether statistically significant differences existed (p<0.05). Microsoft Excel 
and GraphPad InStat computer programs were used in the analysis of the data. 
 
4.3. Results 
The physiologically measurable parameters of sleep, as well as the associated 
behaviours, were recorded in total of five hyraxes, P. capensis, continuously for a period 
of 72 h. The analysis of the data pooled across the five individuals recorded showed that 
66.2% (16 h/day) of the time was spent awake, 25.1%  (6 h/day) in SWS, 3% (1 h/day) in 
REM1 and 0.4% (6 min/day) in REM0 (the 0.9h unaccounted for can be attributed to 
periods where signal loss occurred). There was no statistically significant difference in 
the distribution of these stages between light and dark periods (Table 4.1, Figs. 4.5, 4.6, 
4.8). REM in the rock hyrax was ambiguous which led to its subdivision into REM1 and 
REM0. REM0 was considered the definitive form of this state whereas REM1 showed 
mixed characteristics of both low voltage slow wave sleep as well as REM. Waking most 
often transitioned to SWS and SWS to waking in these animals, thus making this the 
most common sleep state transition. There was a 10% probability that SWS will be 
followed by an episode of REM1 and a 0.2% chance that it would be followed by a 
REM0 episode.  REM1 on the other hand was most likely to transition to waking, but was 
also occasionally followed by REM0. In most cases REM0 was followed by waking, but 
it also transitioned to REM1. Both REM1 and REM0 exhibited an equal chance of 
transitioning to SWS. Thus the three most common state transitional pathways would be, 
waking → SWS → waking, waking → SWS → REM1 → waking, and waking → SWS 
→ REM1 → REM0 → waking (Figs. 4.2 and 4.7) 
 125
4.3.1. Physiological Data – State Definitions 
The polygraphic data was scored in 5 s epochs as waking, SWS, or REM (which 
was subdivided into REM0 and REM1). Waking was characterized by a low voltage, 
high frequency EEG with a high amplitude EMG (EMG during active waking being 
higher in amplitude compared to quiet waking). The ECG for this state was more regular 
during periods of quiet waking and more irregular during periods of active waking. SWS 
episodes were characterized by a high voltage, low frequency EEG, an EMG of lower 
amplitude when compared to waking and a regular ECG (Figs. 4.1-2). Instantaneous 
heart rate during waking and SWS was on average 154 bpm (SER 0.44, STDEV ± 7.79) 
and 158 bpm (SER 0.62, STDEV ± 10.99) respectively (Fig. 4.3). Identification of REM 
episodes was more ambiguous than the identification of waking and SWS episodes, due 
to the fact that both physiologically and behaviourally it did not follow the “textbook” 
definition of this state in all cases. REM is characterized as typically following a period 
of SWS, having an EEG that is similar to the low voltage high frequency waves of 
waking, an atonic EMG, an irregular ECG, muscle twitches and rapid eye movements 
(Desiraju et al., 1966; Elgar et al., 1988; Tobler, 1995; Siegel, 2005; Steriade, 2005; 
Horne, 2009). In most instances REM in the rock hyrax did not meet all of these criteria, 
leading to the subdivision of this stage into REM0 and REM1. In some animals the 
muscle twitches characteristic of REM were observed. A loss of muscle tone was evident 
in all of hyraxes examined where the animal would be immobile and lying down with its 
head on the ground and then slowly fall to one side before repositioning itself. Head 
dropping was also noted in cases where the animal’s head was not initially placed on the 
ground. REM0 episodes were characterized by a low voltage, high frequency EEG 
 126
(similar to waking EEG), a noticeable drop in the EMG (in some cases it was almost 
completely atonic) and an irregular ECG; whereas REM1 episodes exhibited low voltage, 
high frequency EEG, an unaltered EMG that stayed at the same amplitude as the 
preceding SWS episode, and both regular and irregular ECG rates (Figs. 4.1-2). 
Instantaneous heart rate during REM0 and REM1 was on average 163 bpm (SER 1.52, 
STDEV ± 27.34) and 153 bpm (SER 0.61, STDEV ± 10.56) respectively (Fig. 4.3).  
 
4.3.1.1. Total state times 
 The results of the 5 s epoch scoring method revealed that 66.2% (16 h/day) of the 
time was spent awake, 25.1%  (6 h/day) in SWS, 3% (43 min/day) in REM1 and 0.4%  (6 
min/day) in REM0. The 1 min epoch scoring method showed similar results, 66.7% (16 
h/day) of the time was spent awake, 24.8% (6 h/day) in SWS, 3.1% (45 min/day) in 
REM1 and 0.4% (6 min/day) in REM0 (unaccounted time can be attributed to signal loss) 
(Fig. 4.8A). During the light period waking was found to occupy 66.5% of time in the 5 s 
scoring method and 67.3% in the 1 min scoring method, SWS occupied 24.7% of the 
time in the 5 s scoring method and 25% in the 1 min scoring method, REM1 occupied 
2.8% and 3% of the time for the 5 s and 1 min scoring method respectively and REM0 
occupied 0.4% of the time for both scoring methods (Fig. 4.8B). During the dark period 
waking accounted for 66.1% of the time for both scoring methods, SWS 25.4% (5 s) and 
24.4% (1 min), REM1 3.3% (same for both methods) and REM0 0.5% (5 s) and 0.4% (1 
min) (Fig. 4.8C). Thus, no statistically significant difference was noted between the 5 s 
and 1 min scoring methods with regard to TST and TWT over 24 h and during the light 
and dark periods. No statistically significant difference was noted between the light and 
 127
dark periods with regard to TST and TWT for both scoring methods as well (Unpaired t-
test for parametric data, Mann Whitney test for nonparametric data, significant p<0.05).   
 
4.3.1.2. Number of episodes 
 The number of episodes per state over 24 h from each individual animal subjected 
to experimentation was tested for continuity between the different recording days. No 
statistically significant differences in the number of episodes for each particular state 
(Waking, SWS, REM1, and REM0) were noted overall (One-way ANOVA for 
parametric data, Kruskal-Wallis test for nonparametric data, significant p<0.05). One 
animal did show a difference in the number of waking, SWS and REM1 episodes, with 
day 1 showing more episodes of both these states compared to the remaining two days. 
When the group was considered as one entity no statistically significant difference was 
noted in the distribution of the number of episodes between the days for all stages except 
SWS. A greater number of SWS episodes were noted during day 1 as opposed to days 2 
and 3 (Table 4.2). The average number of waking, SWS, REM1 and REM0 episodes in 
24 h for the 5 s epoch scoring method amounted to 638, 559, 37 and 8 respectively 
whereas for the 1 min epoch scoring method it amounted to 129, 116, 27 and 4 
respectively. No statistically significant difference was noted with regard to number of 
REM1 and REM0 episodes in 24 h, however a significant difference was noted for the 
number of waking and SWS episodes in 24 h when the 5 s and 1 min scoring methods 
were compared (average number of waking and SWS episodes in 24 h were significantly 
more for the 5 s scoring method) (Fig. 4.8A). The aforementioned was also observed 
during the light and dark periods when these two scoring methods were compared (Fig. 
 128
4.8B, C). No statistically significant difference was noted with regard to the number of 
episodes per state when the light and dark periods were compared for both scoring 
methods (Unpaired t-test for parametric data, Mann Whitney test for nonparametric data, 
significant p<0.05). 
 
4.3.1.3. Duration of episodes 
 The average duration of each episode for a particular state was compared between 
the three different recording days to test for equal variances. The results revealed that 
there was no statistically significant difference in the average duration of an episode for a 
particular state between the three recording days for most of the animals tested (One-way 
ANOVA for parametric data, Kruskal-Wallis test for nonparametric data, significant 
p<0.05). Two animals did show a difference in the average duration of waking and SWS 
episodes, with the average duration of a SWS episode during the third day being much 
shorter than that of the two preceding days. When the entire group was compared it was 
found that the average duration of REM1 and REM0 episodes did not differ significantly 
between days; however, waking and SWS did show a statistically significant difference 
in their duration between days. The average duration of a waking episode was shorter 
during the first day of recording, whereas the average duration of a SWS episode was 
longer during the first day, compared to the remaining two days of recording (Table 4.2). 
The average duration per state over 24 h amounted to 90 s for waking, 39 s for SWS, 70 s 
for REM1 and 46 s for REM0 for the 5 s scoring methods whereas for the 1 min scoring 
method it amounted to 448 s (7 min), 185 s (3 min), 99 s (2 min) and 78 s respectively 
(Fig. 4.8A). A statistically significant difference for all states was noted when these two 
 129
scoring methods were compared, average episode duration for each of the states was 
significantly longer for the 1 min data compared to the 5 s data.  The aforementioned 
observations were also observed during the light and dark periods when these two scoring 
methods were compared (Fig. 4.8B, C). No statistically significant difference was noted 
between the light and dark periods with regard to average episode duration per state for 
both of the scoring methods. No difference was noted in the average duration of REM1 
episodes between the light and dark periods for the 5 s scoring method however the 1 min 
scoring method did show that the average REM1 episode duration was longer during the 
light period compared to the dark period. For both scoring methods the average REM0 
episode duration was longer during the light period compared to the dark period 
(Unpaired t-test for parametric data, Mann Whitney test for nonparametric data, 
significant p<0.05). 
 
4.3.1.4. REM periodicity and slow wave activity (SWA) 
REM periodicity was calculated from the onset of one episode to the onset of the 
following episode and was based on the 72 h recording period for both the 5 s and 1 min 
scoring methods. A statistically significant difference was noted with regard to REM 
periodicity when the 5 s and 1 min scoring methods were compared. The modal REM 
periodicity for the 5 s data was 5 min whereas it was 10 min for the 1 min data. The 
average REM sleep cycle interval for the 5 s scoring method was 24 min whereas for the 
1 min scoring method it was 33 min.  When all waking episodes that lasted longer that 10 
min were removed from the 1 min data, the modal sleep cycle length was found to be 15 
min (Fig. 4.9).  
 130
 Slow wave activity based on two-hour intervals was calculated for all states 
(SWA-all) and for SWS (SWA-SWS). The results revealed that the SWA was highest 
during SWS and showed a gradual reduction in intensity from waking to REM1 to 
REM0. SWA for both SWA-all and SWA-SWS remained relatively constant between the 
three days of recording. No statistically significant difference was noted with regard to 
SWA during light and dark periods for either SWA-all or SWA-SWS. A statistically 
significant difference (Unpaired t-test, significant p<0.05) was noted when the SWA-
SWS was compared to SWA-all. SWA in SWA-SWS was greater than SWA-all over 24 
h, as well as during both dark and light periods (Fig. 4.10).      
 
4.3.2. Behavioural Data 
Behaviour was recorded simultaneously with the physiological recordings and 
scored in 1 min epochs as immobile (animal not moving for >30 s), quiet waking (animal 
moving head or making minor movements in the same place for >30 s), active waking 
(animal was actively moving around for >30 s) or eating/drinking (animal was 
eating/drinking for >30 s). On average 68% (16 h/day) of the time was spent in a state of 
immobility, 23% (6 h/day) in quiet waking, 5% (1 h/day) in active waking and 3% (43 
min/day) of the time was spent eating/drinking. When the distribution of these 
behavioural states was compared for dark and light periods, no statistically significant 
differences were noted; however, on average these animals were more active (moving, 
eating and drinking) during the dark compared to the light period (Unpaired t-test for 
parametric data, Mann Whitney test for nonparametric data, significant p<0.05)  (Fig. 
4.11). 
 131
 
 
 
 
 
Figure 4.1: Examples of EEG and EMG polygraphs demonstrating a 2 min episode of 
active wake, quiet wake, slow wave, REM (consisting of both REM type 1 and REM type 
0), REM1 and REM0 states in the rock hyrax. Waking episodes were characterised by a 
low voltage high frequency EEG. The EMG for active waking exhibited higher voltages 
compared to the quiet waking state, and high voltage spikes that likely correspond with 
movements, were evident during this state. SWS was characterized by a high voltage, low 
frequency EEG and an EMG that was lower in amplitude than during the waking states. 
The EEG for both REM1 and REM0 resembled that of waking; however, during REM1 
the EMG remained at the same amplitude as the preceding SWS episode, while during 
REM0 the EMG was reduced in amplitude. 
 

 133
 
 
 
 
 
Figure 4.2: EEG and EMG polygraphs demonstrating the transitions that occur between 
the different sleep states in the rock hyrax. The first set of polygraphs is an example of 
SWS transitioning to REM1 which then leads into REM0 followed by waking. Note that 
the EMG during REM1 remains at the same amplitude as during the preceding SWS 
episode, and that when REM1 transitions to REM0 the EMG reduces in amplitude. The 
second set of polygraphs demonstrates transition from SWS to REM0 followed by 
waking.  
 

 135
 
 
 
 
 
Figure 4.3: Scatter plots representing the instantaneous heart rate (IHR) during each of 
the sleep states and waking. The IHR rate during waking was irregular and more widely 
scattered compared to SWS that exhibited a more regular pattern. The IHR of REM1 
closely resembled that of waking while REM0 was even more irregular and widely 
scattered compared to waking. 
 

 137
 
 
 
 
 
Figure 4.4: Diagram illustrating the spectral power and associated frequency bands 
characteristics of waking, SWS and REM1 and REM0 in the rock hyrax.  
 

 139
 
 
 
 
 
Figure 4.5: Hypnograms showing the physiological and correlated behavioral state 
transitions occurring over a 24 h period for H06 starting at 9 am. The shaded area is 
representative of the dark period. The polycyclic nature of sleep is visible in the 
physiological hypnogram, with SWS equally represented during both the light and dark 
periods; however, both REM1 and REM0 were more prevalent during the dark period.  
The animal also appeared to be more active and was eating more during the dark periods 
compared to the light periods for the 24 hours represented here. 
 

 141
 
 
 
 
 
Figure 4.6: Hypnogram showing the physiological and correlated behavioral state 
transitions occurring over a 24 hour period for H07 starting at 9 am. The shaded area is 
representative of the dark period. SWS is equally represented during both the light and 
dark periods which is indicative of the polycyclic nature of sleep in these animals. REM1 
was more prevalent during the dark period, and REM0 did not occur as frequently as in 
H06 (see Fig. 4.5). During this particular 24 hours active waking was distributed equally 
during both light and darks periods, while feeding behaviour predominated during the 
light period. 
 

 143
 
 
 
 
 
Figure 4.7: State transition probabilities in the rock hyrax during 24 hours based on the 
data of two randomly selected animals. In most cases SWS was followed by waking, thus 
making this the most common sleep pathway observed in the rock hyrax. The next most 
common pathway would be when SWS is followed by r1 (REM1) which then 
predominantly transitions to waking; however in some cases it also transitions to r0 
(REM0) which in turn is most commonly followed by waking. 
 

 145
 
 
Figure 4.8:  Histograms depicting the total amount of time spent in each of the defined 
physiological states as well as the average number of episodes and episode duration for 
24 hours (A, first set of graphs), the light period (B, second set of graphs) and the dark 
period (C, third set of graphs) for both the 5 s (left column) and 1 min epoch (right 
column) scoring methods. The percentage of time occupied by each state over 24 hours 
and the dark period remains the same for both methods of scoring; however, a significant 
difference was noted in the percentage of time occupied by REM1. The number of 
waking and SWS episodes are significantly less in the 1 min compared to the 5 s scoring 
method. The average episode duration for all states was also significantly different; 
however, there was no significant difference in the average duration of REM0 episodes. 
Light shaded bars represents each of the animals studied (from left to right - H107, H03, 
H04, H05, H07) and the dark bars represents the species mean. The graphs also show the 
standard error of the mean (represented on the species mean) for the total amount of time 
spent in each of the defined physiological states as well as the average number of 
episodes and episode duration for 24 hours (A, first set of graphs), the light period (B, 
second set of graphs) and the dark period for each of the physiologically defined states 
for both the 5 s and 1 min epoch scoring method. An unpaired two-tailed t-test was used 
in all statistical analyses where the data was normally distributed whereas a Mann 
Whitney test was used in cases where the data was not normally distributed and 
significance was obtained in all instances where p<0.05. 
 



 149
 
 
 
 
 
Figure 4.9: Histograms depicting REM periodicity for 5 s and 1 min epoch scoring 
methods based on a 72 h recording period as well as the REM periodicity based on the 1 
min epoch data when waking episodes that lasted longer than 10 min were eliminated 
from the data. The modal sleep cycle for the 5 s epoch scoring method was 5 min 
whereas it was 10 min for the 1 min epoch scoring method. When waking episodes that 
lasted longer than 10 min were eliminated from the 1 min epoch data, the modal sleep 
cycle length was 15 min. 
 

 151
 
 
 
 
 
Figure 4.10: Slow wave activity (based on 2-h intervals) in three animals for the 72 h 
recording period during all states (A) and SWS (B). SWA remained constant during all 
three recording days and did not differ significantly between animals (Unpaired two-
tailed t-test, p>0.05). (C) Histogram illustrating the average spectral power for the 72 h 
recording period for all states and SWS during the light and dark period. No significant 
difference (Unpaired two-tailed t-test, p>0.05) is evident between the light and dark 
periods for all states and SWS. (D) Line graph showing the average SWA during wake 
and sleep stages. SWA was the greatest during SWS and was consistently greater 
throughout the recording period. 
 

 153
 
 
 
 
 
Figure 4.11: Histogram illustrating the percentage of time occupied by each behavioral 
state for the 72 h recording period during the light and dark period respectively. There is 
no significant difference in the distribution of immobile and quiet waking between the 
light and dark period; however, there is a tendency towards more active waking and 
eating/drinking behaviour during the dark period (Parametric data - Unpaired t-test, 
significant p<0.05; Nonparametric data – Mann Whitney test, significant p<0.05). Light 
shaded bars represents each of the animals studied (from left to right - H107, H03, H04, 
H05, H07) and the dark bars represents the species mean for each of the behavioral states.  
 

 155
 
 
 
 
 
Table 4.1:  Total wake, sleep and REM times as a percentage of 24 hours for each 
individual animal as well as the species mean. TWT – total waking time, TST – total 
sleep time, TSWS – total SWS time, TREM – total REM time, Tr1 – total REM1 time, 
Tr0 – total REM0 time. 
 
 156
 
 
 
 
 
 
 
 
 
Animal ID 
 
Body Weight 
(Kg) 
Brain Weight 
(g) Sex 
 
TWT 
 
 
TST   
 
 
TSWS 
 
 
TREM 
 
 
Tr1 
 
 
Tr0 
 
          
H107 3.1 19.6 Female 71.6 25.6 23.5 2.1 1.6 0.5 
H03 3.14 27.1 Male 69 28.9 27.4 1.5 1.4 0.1 
H05 4.3 20.4 Male 73.4 20.4 16.7 3.7 3.4 0.3 
H06 1.74 19.5 Male 60.4 35 29 6 4.7 1.3 
H07 2.05 17.9 Male 56.4 32.8 28.7 4.1 4.1 0.04 
          
Species 
Averages 2.87 20.9  66.2 28.5 25.1 3.5 3 0.4 
 
 
 157
 
 
 
Table 4.2: Illustration of where statistically significant differences (p<0.05) were noted 
between the three recording days for each state with regards to the number of epochs, 
episodes and episode duration for each individual animal as well as for the group. Each of 
the animals subjected to experimentation was tested individually to determine whether 
statistically significant differences existed between each of the recording days with 
regard to the number of epochs and episodes as well as the episode duration for each of 
the physiologically defined states. Following this, the data obtained from all the animals 
were pooled together to determine whether any statistically significant differences existed 
between each of the recording days for the group. All data was tested for normality prior 
to any statistical analyses (Kolmogorov and Smirnov test, p>0.05).  A one-way ANOVA 
was used for comparisons between parametric data (statistical significance where p<0.05) 
and a Kruskal-Wallis test was used for comparisons between nonparametric data 
(statistical significance where p<0.05). Italisized values indicate statistically significant 
difference between the recordings. 
 
 158
 
 
 
 
 
 
 
Animal ID 
Duration Episodes Epochs 
Waking SWS r1 r0 Waking SWS r1 r0 Waking SWS r1 r0 
 
            
H107 0.8736 0.0197 0.221 0.7002 0.5239 0.0692 0.1894 0.8531 0.01192 0.0048 0.074 0.8088 
H03 0.0257 1.37E-09 0.3069 0.6581 0.0412 3.04E-05 0.0109 0.3105 4.94E-15 2.87E-16 0.3069 0.5211 
H05 0.6861 0.5999 0.5173 0.1828 0.974 0.9506 0.6132 0.8317 0.2673 0.7823 0.5375 0.9144 
H06 0.6068 0.8088 0.4852 0.264 0.1315 0.1861 0.2254 0.795 0.484 0.245 0.1374 0.4525 
H07 0.2338 0.7018 0.403 0.304 0.0332 0.3182 0.8984 0.364 0.3012 0.7307 0.793 0.352 
             
All 0.0192 0.0117 0.5138 0.8684 0.4016 0.0128 0.1914 0.6802 0.0006 2.42E-05 0.129 0.651 
 
 159
4.4. Discussion 
 The aim of the present study was to describe sleep in the rock hyrax, Procavia 
capensis. On average these animals spent 16 h/day in a state of waking, 6 h/day in SWS, 
and 50 min/day in REM (of which REM1 contributed 43 min/day and REM0 6 min/day). 
The polygraphic aspects of waking and SWS in the rock hyrax were characteristic of 
what has been described for these physiological states in many mammals (Desiraju et al., 
1966; Elgar et al., 1988; Tobler, 1995; Siegel, 2005; Steriade, 2005; Horne, 2009). In this 
study of the rock hyrax, REM was subdivided into REM1 and REM0, as the majority of 
the REM episodes did not fit the conventional criteria. REM1 was characterized by an 
EEG similar to waking, an EMG that was the same as during SWS, and a mostly regular 
ECG. REM0 was characterized by an EEG similar to waking, EMG lower in amplitude 
than during SWS and an irregular ECG (Figs. 4.1-4). Of these two, REM0 most closely 
resembled what would be considered classical REM sleep (Desiraju et al., 1966; Steriade, 
2005; Nicolau et al.; 2000). At present, with the data we currently have, it is difficult to 
determine whether REM1, which was found exhibiting all the same physiological 
parameters in all hyraxes studied, is a type of NREM or a type of REM.  It is possible 
that REM1, as defined in this study, could be considered a form of low voltage slow 
wave sleep, a type of NREM. According to Bergmann et al. (1987) low voltage slow 
wave sleep (LS) is characterized by behavioural quiescence, desynchronized EEG, EMG 
that is at the same amplitude as during SWS and theta activity that resembles waking and 
is lower than during REM. In the rock hyrax REM1 exhibited some of these 
characteristics such as behavioural quiescence, desynchronized EEG and EMG similar to 
SWS. In addition to this, the ECG was not as irregular as seen during unambiguous REM 
 160
sleep in the hyrax.  Thus, the data can be interpreted in such a way as to consider REM1 a 
type of low voltage NREM similar in structure to LS as defined by Bergmann et al. 
(1987) for the rat. 
Despite this, it is possible that REM1 is a type of REM. The desynchronized 
EEG, ECG less regular in comparison to SWS, behavioural quiescence, greater amount 
of theta activity, and that REM1 always followed a SWS episode, all indicate that REM1 
could be considered a type of REM. Clearly, this is not what would be described as 
typical mammalian REM sleep, but REM sleep in certain mammalian species, such as the 
platypus which has REM with cortical synchronization (Siegel et al., 1999), EMG in the 
rat and rabbit that are not always atonic in REM (e.g. Pivik et al., 1981; Gottesmann, 
1992), and Tupaia in which REM has a higher amplitude EMG than in SWS (Berger and 
Walker, 1972), does not always fit the standard criteria determined in studies of typical 
laboratory animals and humans. Thus, with the currently available data REM1 might also 
be interpreted as a form of REM sleep. 
While we do not currently have the data to resolve this issue, future studies of the 
rock hyrax, measuring such variables as the hippocampal theta activity, electro-
oculogram, as well as single unit recordings from the sleep related brainstem nuclei could 
resolve whether REM1 in the rock hyrax is a form of NREM or REM. Whatever the 
result of such future studies, the implications of REM1 being a type of NREM or REM 
are of interest. If REM1 is a type of NREM, then, like many other mammals, the rock 
hyrax would have more than one form of NREM; however, this would mean that the rock 
hyrax exhibits the lowest amount of REM sleep of any terrestrial mammal studied to 
date, approximately 6 min/day. This is of course, an interesting possibility and may make 
 161
the hyrax an interesting animal model for the study of the function of REM sleep due to 
its possible paucity in this species. With the alternative view, if REM1 is a type of REM 
state, then the hyrax exhibits more than one type of REM, which would be unusual, and 
NREM of the hyrax would consist purely of SWS. This again makes the hyrax an 
interesting model animal for study of REM function, as if REM1 is a type of REM, then 
perhaps the potential benefits of REM sleep for organismal function can be derived 
without the expression of the full range of typical REM sleep polygraphic features. This 
is clearly an area that needs further investigation. 
The most common sleep state transition pathway was found between waking and 
SWS (waking → SWS → waking), the next most common was found to be waking 
which leads to SWS which then transitions to REM1 followed by waking (waking → 
SWS → REM1 → waking) (Figs. 4.2, 4.7). When the 5 s and 1 min epoch scoring 
techniques were compared, no statistically significant difference was noted with regard to 
TST and waking times for the 24 h as well as the light and dark periods. A statistically 
significant difference was noted with regard to the number of episodes and episode 
duration for each of the states during the 24 h, light and dark periods (Fig. 4.8A-C). Thus, 
when drawing conclusions regarding total times over the 24h period for the different 
states, either the 5 s or 1 min epoch data will provide the same answers, but when 
determining the number of episodes and episode duration it appears to make more 
biological sense to use the 1 min epoch scoring method. 
  
 
 
 162
4.4.1. Comparison to previous sleep studies in Hyraxes 
A previous abstract by Snyder (1974) reported sleep in three other species of 
hyrax, Procavia johnstoni, Heterohyrax brucei and Dendrohyrax validus. Snyder 
reported that TST for: P. johnstoni was 4.9 h (SWS contributing 4.66 h and REM 0.24 h), 
H. brucei was 5.7 h (5.24 h consisting of SWS and 0.45 h REM), and D. vaidus was 4.9 h 
(SWS contributing 4.66 h and REM 0.24 h). There was no statistically significant 
difference with regard to TST between light and dark periods for P. johnstoni and H. 
brucei, but a light/dark difference in TST was noted for D. validus, which was the only 
solitary nocturnal species studied (Kingdon, 1971; Campbell and Tobler, 1984; Elgar et 
al., 1988); however, as stated previously only an abstract was published for the TST 
reported for these species of hyrax and no spectral power analysis was performed on the 
data to corroborate these findings. Our results showed a 1.9 h and 1.1 h increase in TST 
when compared to P. johnstoni and H. brucei respectively. Total REM time, that is if we 
can consider REM1 to be a form of REM, in our study was also considerably greater 
compared to the reported values stated previously in these two species (P. capensis – 50 
min/day, P. johnstoni – 14.4 min/day, H. brucei – 27 min/day).  No statistically 
significant difference was noted between the light and dark periods with regard to TST in 
the current study as well as in the Snyder study (P. johnstoni and H. brucei only), 
underlining the polycyclic nature of sleep in this mammalian group.  
 
4.4.2. Comparison to sleep in other Afrotheria, Xenarthra and other mammals 
Hyraxes belong to the superorder Afrotheria, which includes the African and 
Asian elephant, the manatee and dugong, aardvark, golden moles, tenrecs, and sengis. 
 163
Observational studies have reported that TST in the Asian and African elephant amounts 
of approximately 3.8 h (Campbell and Tobler, 1984; Kurt, 1960), with sleep mainly 
occurring during the dark periods in the early morning hours (Wyatt and Eltringham, 
1974). Moss (1982) reported that wild African elephant sleep a total of 3 – 5 h at night, 
whereas captive Asian elephants slept for approximately 4 – 6.5 h (Tobler 1992; Wilson 
et al., 2006). An electrophysiological study on an Amazonian manatee, Trichechus 
inungius, revealed that this animal spent approximately 27% (6.5 h) in SWS and 1% (14 
min) in REM sleep over a 24 h period and that 25% of SWS was occupied by 
interhemispheric asymmetry (Mukhametov et al., 1992); however, it must be noted that 
the TST reported was based on study of a single animal and the recording of sleep was 
done directly after surgery, thus the TST obtained may not be representative of this 
species. As for the insectivoran-like afrotheres, TST in the tenrecs has been reported to be 
15.6 h, with SWS contributing 13.26 h and REM 2.34 h (Campbell and Tobler, 1984; 
Elgar et al., 1988; Snyder, 1972).  
It appears, as is clear from the aforementioned examples that sleep in the 
Afrotherians is just as diverse and unusual as the composition of this group, and this sleep 
diversity could possibly be attributed to the different lifestyles and behaviors of each of 
its members (Siegel, 2009). For example, some members are diurnal (e.g. elephant, 
manatee, rock hyrax, sengis) where others are nocturnal (e.g. tenrecs, golden moles, 
aardvark). Some member’s offspring are precocial (e.g. rock hyrax) whereas others are 
altricial (e.g. stenrecs). The degree of sociality also differs greatly between the different 
members of this superorder, with elephants and rock hyraxes being social mammals, the 
manatee semi-social, some tenrecs being social, whereas the sengis live in monogamous 
 164
pairs. TST has been correlated to body size, but is only one of many factors that have 
been reported to be correlated to TST in mammals, and it has been reported that a 
negative relationship exists between these two variables (Tobler, 1995; Zepelin et al., 
2005). Thus, at first glance the Afrotheres seem to adhere to this relationship, with the 
elephants, which has the largest body mass in this superorder, having the lowest TST, 
whereas the tenrecs and sengis, which have smallest body mass in this superorder, 
appears to have the highest TST. One would expect, purely based on body mass that the 
average TST for hyraxes would lie somewhat in the middle, i.e. between the TST for the 
manatees and the insectivore-like Afrotherians; however, we found that the TST in the 
rock hyrax closely resembles what has been reported in the manatee (TST in the rock 
hyrax – 28.5% and manatee – 28%) (Mukhametov et al., 1992). Despite this, the manatee 
does show a greater percentage of TSWS (27%, rock hyrax – 25.1%), whereas the rock 
hyrax shows a greater percentage of TREM (3.5%, manatee – 1%). This is not in 
accordance with the claimed inverse relation between body mass and sleep time, as the 
manatee has a body mass approximately 100 fold that of the rock hyrax, it leads a semi-
social lifestyle whereas the rock hyrax is known to be social, their offspring are born 
altricially whereas the hyrax offspring are born precocially, and most probably the 
greatest difference between these mammals is that the manatee is an aquatic mammal 
whereas the rock hyrax is a terrestrial mammal. Although these two mammals share a 
common ancestor, they are vastly different in many respects, but spend roughly the same 
amount of time asleep each day. In addition, the manatee has unusual asymmetrical SWS 
patterns. These differences would indicate other factors, apart from body mass and state 
at birth (i.e. precocial vs. altricial) as being responsible for generating the differing sleep 
 165
times and patterns seen in these mammals. The above list of similarities and differences 
would appear to suggest a combination of phylogenetic constraint and adaptation to 
environmental factors would need to be evoked to fully explain the observed results.  
Sleep has also been studied both behaviorally and electrophysiologically in the 
Afrotherian sister group Xenarthra which consists of armadillos, sloths and anteaters 
(Hallström et al., 2007). TST in the giant armadillo, Priodontes giganteus, has been 
reported to be 18.1 h (Affanni, 1972) and in the nine banded armadillo, Dasypus 
novemcinctus, 17.4 h (Van Twyver and Allison, 1974). For both species no difference 
was observed with regard to the distribution of TST between the light and dark periods 
(Campbell and Tobler, 1984). Behavioral studies conducted on the two toed sloth, 
Choloepus hoffmanni, showed that TST in these animals amounted to 16.4 h (Sunquist 
and Montgomery, 1973). The brown throated three toed sloth, Bradypypus variegatus, 
was found to have TST of 15. 84 h when it was recorded in captivity (Campbell and 
Tobler, 1984; Galvăo de Moura Filho, 1983) and a TST of 9.63 h when recorded in the 
wild (Rattenborg et al., 2008). The rock hyrax and three toed sloth are both herbivores, 
whereas the giant armadillo and tenrec are insectivorous or carnivorous. When body mass 
is compared to the number of hours of sleep a day, the tenrec and giant armadillo sleep 
for a larger part of the day compared to the rock hyrax and three toed sloth. An analysis 
of all reported sleep times found TST in carnivores are significantly greater than that of 
herbivores (Siegel, 2005).  
Physiological parameters such as body and brain mass, metabolic rate and state of 
birth have been correlated to TST in mammals; however, no consensus can be reached 
regarding these correlations as different studies all have opposing views. For example, 
 166
when just comparing body mass, mammals with the same dietary preference that fall 
within the same body mass range have different TST, the rock hyrax with an average 
body mass of 4 kg has a TST of approximately 7 h each day, whereas the tree toed sloth 
which has an average body mass of 3.5 – 4.5 kg sleeps for approximately 10 h a day in 
the wild and 16 h in captivity (Siegel, 2005; Rattenborg et al., 2008). It has also been 
shown that risk of predation as well as relative exposure of the sleeping site correlates 
negatively with TST (Capellini et al., 2008). Rock hyraxes have been reported to have a 
danger factor of 0.5 (Allison and Cicchetti, 1976) and are species that are typically 
heavily predated, but they sleep in relative unexposed sites. State of birth is another 
physiological factor that has been shown to correlate with TST. Rock hyrax offspring are 
born precocially, thus these animals should have a reduction in TREM time, but when 
compared to the manatee, a mammal whose offspring are born altricially and that has a 
larger body mass and lower metabolic rate, TREM is greater. Considering the evidence it 
is thus possible that ethological rather than physiological factors are responsible for the 
variations seen in TST between different mammalian species. 
The current data also poses some interesting questions. The results showed that 
these animals were awake on average for 16 h/day, and the behavioural data suggested 
that active waking and eating was more predominant during the dark periods as opposed 
to the light periods. Thus they are not strongly diurnal or nocturnal, but can best be 
characterized as polycyclic without a strong circadian difference in activity or sleep 
patterns.  In the wild rock hyraxes are rarely seen out of their dens during the night, so 
could the bursts of nocturnal activity that have been observed in our study be replicated 
when these animals are in “hiding” during the night in the wild, and what would the 
 167
relevance of this activity be when these animals sleep socially? It is furthermore also 
possible that the activity patterns observed in these animals in the current study could be 
attributed to the standardized laboratory conditions (i.e. constant ambient temperature, 12 
h light/dark cycle as well as light intensity). In the wild these animals are social, they do 
not always experience 12 h of light followed by 12 h of dark, they are not always exposed 
to same light intensity, and the ambient temperature can vary greatly between day and 
night. It has also been shown that rock hyraxes have poor thermoregulative abilities and 
are often seen basking in the sun and typically sleep huddled up in groups, a possible 
mechanism for conserving body heat. It is therefore possible that the biology of the 
species can account for differences observed with regard to the phasing of sleep and 
activity budgets seen in the current study and that the standardized conditions might just 
represent a unusual condition that the will only be experienced a limited amount of time 
in the wild. Further studies on rock hyraxes in the wild would be required to determine to 
what extent these standardized laboratory conditions alter the natural phasing of sleep and 
activity budgets in these animals. 
 168
Chapter 5 
Social contrasted with solitary sleep patterns in the rock hyrax, Procavia capensis. 
 
5.1. Introduction 
 Sleep is a homeostatically regulated process and it is characterized by its easy 
reversibility, immobility and reduced responsiveness to sensory stimuli. Species specific 
sleep postures and sleep sites, as well as occlusion of the eyes, are typically regarded as 
signs of behavioral sleep. Mammalian sleep is divided into non rapid eye movement sleep 
(NREM) which is often but not always synonymous to slow wave sleep (SWS) and rapid 
eye movement sleep (REM) (Nicolau et al., 2000; Zepelin et al., 2005; Cirelli and 
Tononi, 2008; Lesku et al., 2008; Siegel, 2008). Many studies performed in mammals in 
laboratories to date have been conducted on animals housed in solitary, or asocial, 
conditions even though that particular animal may be social in the natural state 
(McNamara et al., 2008). It has been suggested that social animals sleep less, exhibit 
more fragmented sleep patterns and have lower NREM and REM quotas (Capellini et al., 
2008). A possible reason for this is that species that sleep socially can apparently enter 
deeper stages of sleep as they have the security of sleeping in a group and are thus able to 
sleep more efficiently and acquire the benefits of sleep in a shorter time frame. It has also 
been hypothesized that social species have to invest more time in social interactions and 
relationships which in effect leave them with less time to sleep (Capellini et al., 2008).  
In primates, it has been proposed that hierarchy may play a role in the manifestation of 
sleep patterns that are observed in social mammals (Noser et al., 2003). Male Gelada 
baboons show no correlation between sleep duration and social rank, whereas females 
 169
and juveniles exhibit an increase in sleep duration with decreasing rank. Furthermore, 
dominant males have an increased amount of transitional sleep, which indicates that 
increasing rank leads to a decreased amount of relaxed sleep, which in effect causes an 
increased degree of alertness enabling these animals to react swiftly to nocturnal dangers. 
This study also indicated that no correlation existed between sleep fragmentation and 
social rank (Noser et al., 2003). It has also been shown that group or network size in 
primates does not correlate with sleep times (Nunn et al., 2010); however, when 
Drosophila were placed in a socially enriched environment they exhibited an increase in 
daytime, but not nighttime sleep (Ganguly-Fitzgerald et al., 2006). Studies by Meerlo et 
al. (1997, 2001) have also shown that in rodents social conflict affects NREM sleep by 
increasing electroencephalographic (EEG) slow wave activity (SWA) during the 
subsequent sleeping bout. Sleep duration was not affected by the social stimulus, but the 
rodents appeared to compensate for the socially induced sleep debt by increasing SWA 
during NREM. It was also noted that this increase in SWA was not only the result of the 
length of the preceding waking episode, but also the social nature of the preceding 
waking episode.    
Despite these previous studies, to the authors’ knowledge, no comparative 
physiologically monitored sleep studies have been undertaken on the same animal under 
freely interacting social and solitary, or asocial, conditions. Thus, in the present study 
sleep was telemetrically recorded in the rock hyrax, P. capensis, under both solitary, or 
asocial, and freely interacting social conditions. The aim of the present study was to 
determine whether a significant difference existed in wake and sleep states between the 
social and solitary conditions to test hypotheses regarding the effect of sleeping socially.  
 170
5.2. Materials and Methods  
 A total of five adult rock hyraxes, P. capensis, (4 male and 1 female) with body 
weights ranging between 1.74 – 4.3 kg (Table 5.1), were used in the present study. 
Permits from the Limpopo and Gauteng Provincial Governments were obtained for the 
capture and transport of the animals from the wild. All animals were treated and used 
according to the guidelines of the University of the Witwatersrand Animal Ethics 
Committee (approval number AESC 2005/8/5), which parallel those of the NIH for the 
care and use of animals in scientific experimentation. The animals were captured at 
random from wild populations and thereafter allowed to acclimatize for a period of one 
month in a social setting to the recording enclosures that had a 12-12 lighting schedule 
with temperature maintained between 19-21ºC. Each animal was implanted with a 
telemetric recording devise (Data Sciences International) that allowed for the recording 
of physiologically measurable parameters of sleep without cables or restraint. The 
chamber in which the recording occurred was 1.8 m x 1.5 m with a painted concrete 
surface that was covered with straw. The height of the chamber was approximately 1.5 m 
and steel mesh was placed over the chamber to prevent the animals from escaping. A 
wooden box (90 x 90 x 30 cm) with a Perspex roof and two entrances was placed inside 
the chamber and food (combinations of cucumber, tomato, sweet potato, pumpkin, apples 
and rabbit pellets as a source of roughage) and fresh water was supplied daily. Behaviour 
was also recorded with a low light CCD digital camera connected to a DVD recorder.   
 
 
 
 171
5.2.1. Surgical procedure 
After acclimatization, surgical implantation of the telemetric recoding device was 
performed. The animals were weighed before surgery and anesthetized with weight 
appropriate doses of a 2:1 mixture of ketamine and xylazine (Anaket-V and Chanazine 
2% Injection, Bayer HealthCare). The head and neck, left thoracic (two 2cm x 1cm) and 
abdominal (10cm x 10cm) regions were shaved and cleaned with CHX Chlorhexidine 
Disinfectant (0.5% chlorhexidine diglucunate in 75% alcohol, Kyron Laboratories (PTY) 
LTD) before surgery commenced. These areas correspond, respectively, to the regions 
were the EEG (electroencephalogram), EMG (electromyogram) and ECG 
(electrocardiogram) electrodes and telemeter would be implanted. The animal was placed 
on a heated blanket in order to maintain a constant body temperature throughout the 
surgery and the head was securely placed in a stereotaxic apparatus to prevent movement 
and allow for the accurate placement of the EEG and EMG electrodes. During the 
surgical procedure the animal was kept under a constant state of anaesthesia by means of 
isoflurane ventilation (1-2% in an oxygen/70% nitrous oxide mixture, Isofor Inhalation 
Anaesthetic – Safe Line Pharmaceuticals (Pty) Ltd). The animal’s heart rate, body 
temperature and percentage oxygen saturation was monitored.    
Under aseptic conditions, a mid-sagittal incision was made over the skull and the 
skin and temporal muscle were reflected to expose the part of the skull overlying the 
motor cortex. Using a dental drill, three 2 mm diameter holes were made in the cranial 
vault to expose the underlying dura mater. The first hole was drilled anterior to the 
olfactory bulbs for the placement of the indifferent electrode, while two holes were 
drilled approximately 5 mm apart just lateral to the sagittal sinus over the left motor 
 172
cortex for the placement of the recording electrodes (gauge of electrode 0.457 mm, 
silastic outside diameter 0.9 mm and inside diameter 0.508 mm, PhysioTel ® Multiplus 
Transmitter, Data Sciences International). The electrodes were placed in such a manner 
that the tips rested firmly on the surface of the cortex but did not pierce the dura mater 
and were secured in place with dental cement. Two EMG electrodes (gauge of electrode 
0.457 mm, silastic outside diameter 0.9 mm and inside diameter 0.508 mm, PhysioTel ® 
Multiplus Transmitter, Data Sciences International) (1.5 cm apart) were sutured into the 
nuchal musculature, while two ECG electrodes (3 cm apart) were sutured into the 
subdermal tissue overlying the rib cage in the left thoracic region. A subcutaneous pocket 
was created (10 cm x 10 cm) over the left abdominal region, which allowed for the 
implantation of the telemeter. All skin incisions were sutured following electrode and 
telemeter implantation. After surgery was complete the animal was given an 
intramuscular analgesic (0.1 ml Tamgestic, Schering-Plough, mixed with 0.9 ml sterile 
water, 1 ml mixture/kg) and returned to the recording enclosure. Recovery was monitored 
every half hour until it could be established that the animal was fully lucid and able to 
move freely and eat/drink normally.  
 
5.2.2. Sleep recording 
After the surgical procedure the animals were allowed a recovery period of one 
week before the recording of sleep commenced. The animals were housed in the same 
recording enclosure for recovery as well as recording and during the recovery period the 
implanted animal was housed under solitary conditions. A receiver was mounted and 
secured to one wall of the enclosure while a low light CCD digital camera was mounted 
 173
above the enclosure. The telemetric recording system (Data Sciences International, DSI, 
PhysioTel ® Multiplus Transmitter, model TL10M3-D70-EEE implant) consisted of a 
DEM multiplex interface to which the receiver was connected. The signal from the 
implanted transmitter (round, 13cm² with stainless steel electrodes, weight – 37g, volume 
– 25 ml, 3 channels) detected by the receiver was relayed to the input amplifier of the 
Data Sciences computer system, after which it was digitally recorded (in DSI format) for 
analysis. Data digitally saved in the DSI format was converted to text format and these 
files were in turn converted into the appropriate format needed for recognition and 
analysis by the Spike 2 computer program (version 4.2, Cambridge Electronic Design). 
Asocial sleep (physiological measurable parameters and the associated behaviour) was 
subsequently recorded continuously for a period of 72 hours after which one or two other 
hyraxes were introduced to enclosure with the implanted hyrax. The hyraxes were 
allowed to acclimatize to their new social environment for a period of three days that was 
followed by the recording of sleep in this social condition for a continuous period of 72 
hours.  The recording enclosure was in a room that was acoustically isolated. The animals 
were disturbed only once a day for approximately 5 min at the same time during each of 
the recording days for feeding.   
 
5.2.3. Data analysis 
Version 4.2.2 of the Spike 2 software (Cambridge Electronic Designs, UC) was 
used in order to convert the recorded data into the appropriate format, i.e. Spike 2 data 
format, for offline analysis. The EEG data was subsequently scored in 5 s epochs as 
either waking characterized by low voltage high frequency EEG and high voltage EMG, 
 174
SWS characterized by high voltage low frequency EEG and EMG lower in amplitude 
than waking, and REM (rapid eye movement). REM in the rock hyrax was found to be 
ambiguous and did not resemble the classical definition of REM sleep, thus REM0, was 
characterized by low voltage high frequency EEG, an EMG that was almost atonic, and 
an irregular ECG; whilst REM1 was characterized by a low voltage high frequency EEG, 
EMG amplitude characteristic of SWS and a regular ECG. An epoch was only assigned 
to a particular state if the state occupied at least 50% of the epoch. The data obtained 
from the 5 s epoch scoring was analyzed to determine the modal state per minute to 
generate the data for the 1 min epoch scoring. An unpaired two-tailed t-test (parametric 
data) or Mann-Whitney (nonparametric data) test (p<0.05) was employed to determine if 
a statistically significant differences existed between the 5 s and 1 min epoch scoring 
methods. Behaviour was scored in 1 min epochs as either immobile – animal was 
completely immobile for >30 s, quiet waking – animal was immobile and only moving its 
head or made minor movements in the same place for >30 s, active waking – animal was 
actively moving around for >30 s, or eating/drinking – animal was eating and/or dinking 
for >30 s.  
All the physiologically and behaviourally scored data was tested for normality 
prior to any statistical testing. The Kolmogrov and Smirnov test was conducted to 
determine whether the data was parametric or nonparametric in nature (p>0.05). A one-
way ANOVA was used in all comparisons where the data was parametric and a 
statistically significant difference between sample means were obtained in cases where 
p<0.05. Where it was found that the data was nonparametric the Kruskal-Wallis test was 
employed and a statistically significant difference between sample medians were 
 175
obtained in all cases where p<0.05. Comparisons between light and dark were made by 
means of contingency tables, where in the case of parametric data the unpaired two-tailed 
t-test and in the case of nonparametric data the Mann-Whitney test was employed to 
determine whether statistically significant differences existed (p<0.05). Microsoft Excel 
and GraphPad InStat computer programs were used in the analysis of the data. 
 
5.3. Results 
The physiologically measurable parameters of sleep as well as the associated 
behaviour were recorded in five hyraxes, P. capensis, continuously for a period of 72 h 
under solitary, or asocial, conditions. This was followed by the introduction of one or two 
other non-implanted hyraxes to the enclosure of the implanted hyrax. The animals were 
allowed to acclimatize to the new social setting for a period of three days after which the 
physiological parameters of sleep and the accompanying behaviour was recorded 
continuously for 72 h. The results revealed that in the social condition 66.8% (16 h/day) 
of 24 h was spent awake, 24.8% (6h/day) in SWS, 2.7% (39 min/day) in REM1 and 0.6% 
(9 min/day) in REM0 (the 1.2h unaccounted for can be attributed to periods where signal 
loss occurred) (Table 5.1). These results did not differ significantly from that during the 
solitary, or asocial, condition in these animals (Table 5.2); however, the duration of 
REM0 episodes appeared to be longer under social compared to solitary conditions. Over 
the 24 h period REM0 was on average 20 s longer during the social condition, similar 
during the light period and approximately 40 s longer that the solitary condition during 
the dark period (Fig. 5.3). The modal and average REM periodicity was also greater 
during social conditions, however when waking episodes that lasted longer than 10 min 
 176
were eliminated from the data, the modal REM periodicity for both the social and solitary 
conditions was found to be similar (15 min) (Fig. 5.4). The average SWA was also 
greater during the social condition during the first two days of recording, but during the 
last day of recording the SWA during the social condition was lower than that recorded 
for the solitary condition (Fig. 5.5). The most common state transition pathway during 
both conditions was waking → SWS → waking. During the social condition, REM1 and 
REM0 never transitioned to SWS, however during the solitary condition this did occur. 
Furthermore waking only transitioned to SWS during the social condition whereas during 
the solitary condition waking was also seen to transition to REM1. REM1 transitioned 
more readily to REM0 during the solitary condition, whereas REM0 transitioned more 
readily to REM1 during the social condition (Fig. 5.2). The second most common state 
transition pathway during both conditions was waking → SWS → REM1 → waking.  
 
5.3.1. Physiological Data  
The polygraphic data was scored in 5 s epochs as waking, SWS, or REM (which 
was subdivided into REM0 and REM1) (see Chapter 4 for a full description of each the 
physiological states). Waking was characterized by a low voltage, high frequency EEG 
with a high amplitude EMG (EMG during active waking being higher in amplitude 
compared to quiet waking). The ECG for this state was more regular during periods of 
quiet waking and more irregular during periods of active waking. SWS episodes were 
characterized by a high voltage, low frequency EEG, an EMG of lower amplitude when 
compared to waking and a regular ECG. Identification of REM episodes were more 
ambiguous than the identification of waking and SWS episodes (see chapter 4) and were 
 177
subsequently subdivided into REM0 and REM1. REM0 episodes were characterized by a 
low voltage, high frequency EEG (similar to waking EEG), a noticeable drop in the EMG 
(in some cases it was almost completely atonic) and an irregular ECG; whereas REM1 
episodes exhibited low voltage, high frequency EEG, an unaltered EMG that stayed at the 
same amplitude as the preceding SWS episode, and both regular and irregular ECG rates. 
The polygraphic characteristics of all states were similar during both the social and 
asocial condition. Instantaneous heart rate during the social condition was on average 174 
bpm (STDEV ± 25.16) and 155 bpm (STDEV ± 6.85) for waking and SWS respectively 
whereas during REM0 and REM1 it was on average 110 bpm (STDEV ± 14.18) and 156 
bpm (STDEV ± 9.18) respectively. During the solitary condition the instantaneous heart 
rate during waking, SWS, REM0 and REM1 was 154 bpm (STDEV ± 7.79), 158 bpm 
(STDEV ± 10.99), 163 bpm (STDEV ± 27.34) and 153 bpm (STDEV ± 10.56) 
respectively. The instantaneous heart rate during SWS and REM1 did not differ 
significantly between the social and solitary conditions whereas waking and REM0 did 
show a statistically significant difference. The instantaneous heart rate for waking during 
the social condition was greater compared to the solitary condition and showed the 
greatest amount of variation compared to all other states. The instantaneous heart rate in 
REM0 during the solitary condition was greater than during the social condition and 
showed the greatest amount of variability compared to all other states.  
  
5.3.1.1. Total state times 
  Form the pooled data of all five hyraxes it was noted that in 24 h under social 
conditions, 66.8% (16 h/day) of the time was spent awake, 24.8% (6 h/day) in SWS, 
 178
2.7% (40 min/day) in REM1 and 0.6% (7min/day) in REM0. Under solitary conditions 
these animals spent approximately 66.2 % in 24 h (16 h/day) awake, 25.1% (6 h/day) in 
SWS, 3% in REM1 (43 min/day) and 0.4% in REM0 (6 min/day). No statistically 
significant differences exist in terms of total state times when comparing social and 
solitary conditions in the rock hyrax. The one state that approached statistical 
significance when the social and solitary conditions were compared was REM0 – the 
total time spent in REM0 was found to be greater during the dark period during the social 
condition (One-way ANOVA for parametric data, Kruskal-Wallis test for nonparametric 
data, significant p<0.05). 
When social and solitary conditions were compared in each individual animal, 
statistically significant differences were noted between these two conditions for waking 
and SWS in four of the five animals examined. Half of the animals that exhibited a 
difference showed a greater amount of time spent awake and less time in SWS when in 
the social condition whereas the other half of the animals showed the same tendency but 
during solitary conditions. Two animals also showed a difference in the time spent in 
REM1, one showed an increase during the social condition whereas the other showed an 
increase during the solitary condition. Finally, three out of the five animals tested showed 
a difference in the time spent in REM0, two out of these three animals showed an 
increase during social conditions whereas only one showed an increase during solitary 
conditions (One-way ANOVA for parametric data, Kruskal-Wallis test for nonparametric 
data, significant p<0.05) 
 
 
 179
5.3.1.2. Number of episodes 
 When the data from all five animals was grouped to create a species average, it 
was found that the number of waking episodes was greater during the solitary condition 
(129 episodes over 24 h) compared to the social condition (121 episodes over 24 h). The 
number of SWS (solitary 116, social 111) and REM1 (solitary 27, social 26) episodes 
were increased, but not statistically significantly, during the solitary condition, whereas 
the number of REM0 episodes was greater during the social condition (solitary 4, social 
5), but again this difference did not reach statistical significance. This tendency was also 
observed over 24 h as well as during the light period; however during the dark period the 
number of episodes for all states was increased during the solitary condition. The average 
number of waking episodes during the 12 h dark period was 61 and 59, SWS 55 and 51, 
REM1 14 and 13, and REM0 3 and 2 for the solitary and social conditions respectively. 
More waking and SWS episodes were evident during the light period for both solitary 
and social conditions and a statistically significant difference in the number of waking 
episodes were noted during the light period during the solitary condition. The number of 
REM1 episodes was greater during the dark period for both social and solitary conditions, 
whereas the number of REM0 episodes was greater during the light period during the 
social condition and greater during the dark period for the solitary condition (Fig. 5.3). 
When comparing the number of episodes of each state during the social and 
solitary conditions for each animal as an individual, not all animals showed differences in 
the number of episodes for the same state. For example, three animals showed a 
statistically significant difference in the number of waking episodes, two in the number of 
SWS episodes, two in the number of REM1 episodes, and three in the number of REM0 
 180
episodes. In the majority of the cases, it was found that the number of episodes for 
waking was higher during the solitary condition. Of the two animals that showed a 
statistically significant difference in the number of SWS episodes, one animal showed an 
increase in the number of episodes of this particular state during the social conditions 
whereas the other animal showed an increase during the solitary condition. The same 
tendency was noted with regard to the two animals that exhibited a difference in the 
number of REM1 episodes, whereas in the majority of the animals that showed a 
statistically significant difference in the number of REM0 episodes it was found that 
during the social condition this state occurred more frequently compared to the solitary 
condition. No statistically significant differences were noted in the distribution of the 
number of episodes per state between dark and light times when the social and solitary 
conditions were compared within each individual animal (In all statistical analyses an 
ANOVA/unpaired t-test for parametric data or Kruskal-Wallis/Mann Whitney test for 
nonparametric data was used, significant p<0.05) (Table 5.2). 
 
5.3.1.3. Duration of episodes 
 When the data from all five animals were pooled to create a species average, the 
average episode duration during social and solitary conditions for waking over a 24 h 
period was shown to be 476 s and 448 s, SWS 196 s and 185 s, REM1 94 s and 99 s, and 
REM0 98 s and 78 s respectively. No statistically significant differences existed between 
the social and solitary conditions concerning the average episode duration of the different 
states. While not statistically significant, the average duration of a waking, SWS and 
REM0 episode was greater during the social condition, but the opposite was found for 
 181
REM1. This tendency was also observed during the light as well as during the dark 
periods. The only state that did approach a statistically significant difference was REM0. 
The average REM0 episode was greater during the social condition over 24 h as well as 
during the dark period, but no difference was noted during the light period. On average 
REM0 was approximately 20 s longer during social sleep over 24 h and 40 s longer 
during the dark period (Fig. 5.3) 
When the animals were tested separately to determine whether individuals 
showed a difference in the average episode duration of a state between social and solitary 
conditions the results revealed that all animals exhibited statistically significant 
differences in the average episode duration for all of the physiologically defined states; 
however, the differences noted for each animal did not reveal a discernable trend, as not 
all animals showed the same differences with regard to the average episode duration for a 
particular state. For example, two animals showed a difference in the average duration of 
waking episodes, in one case the average duration of the waking episodes were longer in 
the social conditions, whereas in the other case the opposite was found. Four animals 
showed a difference in the average duration of SWS episodes, but in this instance three 
out of four showed that the average duration of a SWS episode was longer during the 
solitary condition. Two animals exhibited a difference in the average duration of a REM1 
episode, once again in one case it was found that duration of this state was longer during 
the social condition whereas the other case showed the opposite. Finally, three animals 
showed a difference in the average duration of a REM0 episode, in two of the three 
animals it was found that the average duration of this state was longer in the social 
compared to the solitary condition (In all statistical analyses an ANOVA/unpaired t-test 
 182
for parametric data or Kruskal-Wallis/Mann Whitney test for nonparametric data was 
used, significant p<0.05) (Table 5.2). 
 
5.3.1.4. REM periodicity and slow wave activity (SWA) 
REM periodicity was calculated from the onset of one REM episode to the onset 
of the following REM episode and was based on the 72 h recording period for the 1 min 
epoch scoring method. A statistically significant difference was noted when the modal 
and average REM periodicity was compared between the social and solitary conditions.  
The modal REM periodicity for the social condition ranged between 5-15 min whereas 
for the solitary condition it was 10 min. The mean REM sleep cycle interval for the social 
condition was 43 min whereas during the solitary condition it was 33 min. When waking 
episodes that lasted longer that 10 min were eliminated from the data, the REM 
periodicity for both the social and solitary conditions was found to be 15 min (Fig. 5.4).   
 SWA based on two hour intervals was calculated during waking, SWS, REM1 
and REM0. The average SWA was highest during SWS and gradually declined in 
intensity from waking to REM1 to REM0. The SWA for all these states remained mostly 
constant between all three recording days (Fig. 5.5A).  
During the light period, SWA during SWS (SWA-SWS) was greater than SWA in 
all states (SWA-all) for the solitary condition, whereas SWA-all was greater than SWA-
SWS during the social condition. No statistically significant differences were noted 
between SWA-SWS and SWA-all for the social condition; however, in the solitary 
condition, SWA-SWS was statistically significantly greater than SWA-all. No 
statistically significant difference was noted between conditions for SWA-SWS; 
 183
however, a statistically significant difference was noted for SWA-all, where SWA-all 
was greater during the social condition compared to the solitary condition.  
During the dark period SWA-SWS was greater than SWA-all for both social and 
solitary conditions. A statistically significant difference existed between SWA-SWS and 
SWA-all for both the social and solitary condition; however, when SWA-SWS and 
SWA-all was compared between conditions no statistically significant difference existed 
(Fig. 5.5B). 
When the average SWA per day was compared between social and solitary 
conditions, SWA during day 1 and day 2 for SWA-SWS as well as SWA-all were higher 
in the social condition, but on the third day the opposite was found. Furthermore, the 
average SWA-SWS and SWA-all remained mostly constant between the three recording 
days for the solitary condition whereas during the social condition SWA-SWS and SWA 
during waking was significantly greater during the first recording day compared to the 
two remaining recording days (In all statistical analyses an unpaired t-test for parametric 
data or Mann Whitney test for nonparametric data was used, significant p<0.05) (Fig. 
5.5C).   
 
5.3.2. Behavioural Data 
Behaviour was recorded simultaneously with the physiologically measurable 
parameters of sleep and scored in 1 min epochs as either immobile (animal not moving 
for >30 s), quiet waking (animal moving head or making minor movements in the same 
place for >30 s), active waking (animal is actively moving around for >30 s) or 
eating/drinking (animal is eating/drinking for >30 s). During the social condition, over a 
 184
period of 24 h, on average 71% (17 hrs) of the time was spent in a state of immobility, 
21% (5 h) in quiet waking, 3% (43 min) in active waking and 3% (43 min) of the time 
was spent eating/drinking and during the solitary condition they spent on average 68% 
(16 hrs) of the time immobile, 23% (5.5 hrs) in quiet waking, 5% (1 hr) in active waking 
and 3% (43 min) eating/drinking. When comparing the social and solitary conditions no 
statistically significant difference existed, however these animals spent on average about 
1 h extra a day immobile in the social condition and were more active when in the 
solitary condition. Quiet waking showed a marginal increase during the solitary 
compared to social condition, whereas the time devoted to eating/drinking remained the 
same for both conditions.  
When the distribution of these behavioural states over 24 h during the social 
condition was compared for the light and dark periods it was found that the animals were 
more immobile during the light period. Active waking and eating/drinking behaviour 
doubled during the dark period compared to that in the light, whereas only a slight 
increase was noted for the distribution of quiet waking in the dark compared to the light 
periods. An increased amount of quiet waking was observed during the dark period. The 
aforementioned light/dark distribution of these behavioural states was found to be similar 
to the pattern observed in the solitary condition, but a significant difference was noted for 
the distribution of active waking. In the solitary condition the animals were more active 
compared to the social condition and this active waking occurred more during the dark 
period than the light period (In all statistical analyses an unpaired t-test for parametric 
data or Mann Whitney test for nonparametric data was used, significant p<0.05) (Fig. 
5.6).  
 185
In the social setting the behaviour of the non-implanted animals was also scored 
and analyzed. The results revealed that a positive correlation existed between the 
behaviour of the implanted and non-implanted animal/s, and in all cases the probability of 
the behaviour of the implanted vs. non-implanted animals being unrelated was found to 
approach zero. Figures 5.7 and 5.8 show behavioural hypnograms over 24 h of the same 
animal under social and solitary conditions as well as the behaviour of the non-implanted 
animal during the social condition. It is clear from these hypnograms that, in the majority 
of the cases, the implanted and non-implanted animals enter the same behavioural state at 
the same time and remain in that particular state for roughly the same amount of time.  
 186
 
 
 
 
 
Figure 5.1: Diagram illustrating the spectral power and associated frequency bands 
characteristics of waking, SWS and REM in the rock hyrax for both the solitary and 
social conditions. The spectral power for all states was slightly more increased during the 
solitary condition.  
 

 188
 
 
 
 
 
Figure 5.2: State transition probabilities in the rock hyrax during solitary (left) and social 
(right) sleep for 24 hours. During social sleep SWS will be followed in most cases by 
waking, thus making this the most common sleep pathway observed in the rock hyrax. 
The next most common pathway would be when SWS is followed by r1 (REM1) which 
then predominantly transitions to waking, however in some cases it does also transition to 
r0 (REM0) which in turn is most commonly followed by REM1 but it can also be 
followed by waking. Compared to the solitary condition REM1 and REM0 never 
transitioned to SWS during the social condition. 
 

 190
 
 
 
Figure 5.3:  Histograms depicting the total amount of time spent in each of the defined 
physiological states as well as the average number of episodes and episode duration for 
during solitary (left) and social sleep (right) for 24 hours (A, first set of graphs), the 
light period (B, second set of graphs) and the dark period (C, third set of graphs). No 
statistically significant difference between solitary and social sleep was noted for any of 
the aforementioned periods. The number wake, SWS and REM1 episodes were slightly 
more during solitary sleep whereas the duration of wake, SWS and REM0 was greater 
during social sleep (Parametric - Unpaired t-test, significant p<0.05; Non-parametric – 
Mann Whitney test, significant p<0.05); this pattern was observed for the 24 h, light and 
dark periods. Light shaded bars represents each of the animals studied (from left to right - 
H107, H03, H04, H05, H07) and the dark bars represents the species mean for each of the 
physiological states. The graphs also show the standard error of the mean (represented on 
the species mean) for the total amount of time spent in each of the defined physiological 
states as well as the average number of episodes and episode duration for 24 hours (A, 
first set of graphs), the light period (B, second set of graphs) and the dark period for 
each of the physiologically defined states during both solitary and social sleep. 
 



 194
 
 
 
 
 
Figure 5.4: Histograms depicting REM periodicity for the solitary (top) and social 
(middle) conditions as well as the REM periodicity during the social conditions after 
waking episodes that lasted longer than 10 min were eliminated from the data (bottom). 
The REM periodicity was calculated based on a 72 hour recording period. The modal 
sleep cycle for the solitary sleep was 10 minutes and for social sleep it ranged between 5-
15 minutes. The average REM periodicity during solitary sleep was 33min and 43 min 
during social sleep. A statistically significant difference was noted with regard to REM 
periodicity between solitary and social sleep (Unpaired t-test, significant p<0.05). After 
waking episodes that lasted longer than 10 minutes were eliminated from the data the 
modal REM periodicity during the social condition was 15 minutes. 
 

 196
 
 
 
 
 
Figure 5.5: Slow wave activity (based on 2-h intervals) in three animals for the 72 hour 
recording period during SWS for the solitary (A) and social (B) conditions.  SWA was 
greater during the light period during the social condition, whereas during the dark period 
it was greater when the animals under solitary conditions. No statistically significant 
difference was noted between the light and dark period for the solitary condition, 
however such a difference was noted during the social condition (C). SWA during the 
social condition was greater than that of the solitary condition for the first two days of 
recording, however during the last day of recording SWA was slightly higher during the 
solitary condition. Solitary SWA between the three recording days remained fairly 
constant, whereas during the social condition a greater variability, especially SWA during 
waking and SWS, between the three recording days was noted. (D). Unpaired t-test used 
in all statistical analysis, significant p<0.05. 
 

 198
 
 
 
 
 
Figure 5.6: Histogram illustrating the percentage of time occupied by each behavioral 
state during the light and dark period respectively for the solitary (left) as well as social 
(right) conditions. For both the solitary and social conditions there is no significant 
difference in the distribution of waking and quiet waking between the light and dark 
period, however it is clear that more active waking and eating/drinking behaviour occurs 
during the dark period. No statistically significant difference exists between solitary and 
social conditions for all states except for active waking, between the light and dark 
periods. More active waking occurs during the dark period during the solitary condition 
compared to the social condition (Parametric - Unpaired t-test, significant p<0.05; Non-
parametric – Mann Whitney test, significant p<0.05). Light shaded bars represents each 
of the animals studied (from left to right - H107, H03, H04, H05, H07) and the dark bars 
represents the species mean for each of the behavioral states.  
 

 200
 
 
 
 
 
Figure 5.7: Hypnograms showing the behavioral state transitions occurring over a 24 
hour period for H06 starting at 9 am for the same animal during solitary and social sleep 
as well the non-implanted animal during the social setting. The shaded area is 
representative of the dark period. No statistically significant difference exists in the 
behaviour between the solitary and social setting, active waking appears to be more 
frequent in the solitary  during the dark period compared to the social setting and no 
significant difference exists between the implanted and non-implanted animal in the 
social setting with regards to the distribution of each behavioral state over 24 h 
 

 202
 
 
 
 
 
Figure 5.8: Hypnograms showing the behavioral state transitions occurring over a 24 
hour period for H07 starting at 9 am for the same animal during solitary and social sleep 
as well the non-implanted animal during the social setting. The shaded area is 
representative of the dark period. No statistically significant difference exists in the 
behaviour between the solitary and social setting, active waking appears to be more 
frequent in the solitary compared to the social setting and no significant difference exists 
between the implanted and non-implanted animal in the social setting with regards to the 
distribution of each behavioral state over 24 h.  
 

 204
 
 
 
 
 
Table 5.1:  Total wake, sleep and REM times as a percentage of 24 hours for each 
individual animal as well as the species mean. TWT – total waking time, TST – total 
sleep time, TSWS – total SWS time, TREM – total REM time, Tr1 – total REM1 time, 
Tr0 – total REM0 time. 
 
 205
 
 
 
 
 
 
 
 
 
Animal ID 
 
Body Weight 
(Kg) 
Brain Weight 
(g) Sex 
 
TWT 
 
 
TST   
 
 
TSWS 
 
 
TREM 
 
 
Tr1 
 
 
Tr0 
 
          
H107 3.1 19.6 Female 44.7 32 29 3 2 1 
H03 3.14 27.1 Male 78.3 16.2 14.3 1.9 1.5 0.4 
H05 4.3 20.4 Male 65.3 31.5 26.6 4.9 4.2 0.7 
H06 1.74 19.5 Male 55.8 32.4 28.4 4 3.6 0.4 
H07 2.05 17.9 Male 74.7 21.8 20 1.8 1.6 0.2 
    
      
Species 
Averages 2.87 20.9  66.8 28.1 24.8 3.3 2.7 0.6 
 
 
 
 
 206
 
 
Table 5.2:  Illustration of where statistically significant differences (p<0.05) were noted 
between the three recording days for each state with regards to the number of epochs, 
episodes and episode duration during the social conditions for each individual animal, for 
the group as well as for the social vs. the solitary condition. Each of the animals subjected 
to experimentation was tested individually to determine whether statistically significant 
differences existed between each of the recording days with regard to the number of 
epochs and episodes as well as the episode duration for each of the physiologically 
defined states during the social setting. Following this, the data obtained from all the 
animals were pooled together to determine whether any statistically significant 
differences existed between each of the recording days for the group during the social 
setting. Finally the data obtained from all the individuals for the social and solitary 
setting were pooled together respectively and tested to determine whether statistically 
significant difference existed between the social and solitary conditions. All data was 
tested for normality prior to any statistical analyses (Kolmogorov and Smirnov test, 
p>0.05).  A one-way ANOVA was used for comparisons between parametric data 
(statistical significance where p<0.05) and a Kruskal-Wallis test was used for 
comparisons between nonparametric data (statistical significance where p<0.05). 
Italisized values indicate statistically significant difference between the recordings. 
 
 
 207
 
 
 
 
 
 
Animal ID 
Duration Episodes Epochs 
Waking SWS r1 r0 Waking SWS r1 r0 Waking SWS r1 r0 
 
            
H107 0.3195 0.0601 0.8977 0.9488 0.0867 0.0124 0.2309 0.1328 0.5257 0.5021 0.8495 0.7861 
H03 0.6861 0.5999 0.5173 0.1828 0.6893 0.7271 0.6914 0.1481 0.5697 0.951 0.4233 0.1163 
H05 0.6012 0.3912 0.3947 0.4624 0.3106 0.4872 0.4056 0.0493 0.7261 0.8449 0.2467 0.1014 
H06 0.66 0.457 0.6263 0.4922 0.4746 0.9651 0.8269 0.3429 0.0925 0.661 0.0871 0.4641 
H07 0.9414 0.0005 0.0723 0.4579 0.03661 0.7911 0.0109 0.3492 0.0006 0.0034 0.02152 0.3624 
             
All 0.8835 0.54 0.2892 0.2465 0.4034 0.0612 0.8025 0.2223 0.1827 0.1386 0.452 0.1487 
Social vs. Solitary 0.0897 0.1163 0.4207 0.0642 0.0057 0.1826 0.6241 0.0649 0.4323 0.6818 0.3958 0.0719 
 
 
                                  
 
 208
5.4. Discussion 
The aim of the present study was to describe the sleep-wake cycle in the rock 
hyrax, Procavia capensis, a diurnal social mammal when subjected to normal social 
conditions and to compare it to the results obtained during solitary, or asocial, conditions. 
Waking and SWS times were the same during both social and solitary conditions, but 
REM1 time was decreased (39 min/day – social, 43 min/day – asocial) and REM0 time 
was increased (9 min/day – social, 6 min/day – asocial) during the social condition (Table 
5.1). No statistically significant difference was noted between conditions with regard to 
the number of episodes as well as the average episode duration for each of the defined 
physiological sates for the 24 h, light and dark periods. The number of waking, SWS and 
REM1 episodes was increased in the solitary condition whereas the number of REM0 
episodes was greater during the social condition. The average duration of waking, SWS 
and REM0 episodes was greater when the animals were in the social condition, whereas 
the duration of a REM1 episode was greater when the animals were in the solitary 
condition, The aforementioned tendencies were observed during the light as well as the 
dark periods, with the exception of the number of REM0 episodes, which were found to 
be greater during the dark period under solitary conditions. The modal REM periodicity 
when waking episodes that lasted longer than 10 min were eliminated from the data was 
found to be 15 min for both conditions. The most common state transition pathway was 
found to be similar for both social and asocial sleep, which was waking → SWS → 
waking. During the solitary condition REM1 and REM0 transitioned to SWS and waking 
to REM1, whereas this was not the case during the social condition.  
 209
Behaviourally no statistically significant difference was noted between social and 
solitary conditions. Active waking and eating/drinking behaviour was more prevalent 
during the dark period as opposed to the light period during both conditions; however the 
animals were significantly more active during the dark periods when they were in the 
solitary condition (Fig. 5.6). The non-implanted animals also expressed the same 
behaviour as the implanted animal in the social condition. The results revealed a positive 
correlation between the implanted and non-implanted animal behaviour and the 
probability of the two data sets not being equal was zero in all cases (Figs. 5.7, 5.8).  
 
5.4.1. Sleep fragmentation in the social condition 
  It has been suggested that species that sleep socially express more fragmented 
sleep patterns and that sleep cycle length should be shorter in solitary species based on 
the belief that predation risk is reduced in larger group due to detection and dilution 
effects (Caro, 2005; Lima et al., 2005; Capellini et al., 2008). It has also been shown that 
sleep cycle length and phasing of sleep were not significantly correlated with social sleep 
behaviour (Capellini et al., 2008); however, no studies to the authors’ knowledge have 
been performed in the same species to examine the effects of these two sleeping 
conditions. In the current study no statistically significant difference was noted with 
regard to the number of waking, SWS, REM1 or REM0 episodes over 24 h as well as 
during the light and dark times when the social and solitary conditions were compared. 
However, during the light period a statistically significant difference was noted with 
regard to just the number of waking episodes during the solitary condition. This 
corresponds to the behavioural data that also showed that these animals exhibited more 
 210
quiet and active waking episodes during the light period in the solitary compared to the 
social condition. Furthermore, the average duration of SWS episodes were greater than 
REM1 and REM0 episodes for both the social and solitary condition, during the 24h, 
light and dark periods. No statistically significant difference was noted with regard to 
sleep cycle length or duration of episodes for all states, with the exception of REM0 (see 
below) between social and solitary conditions. Thus it appears that sleep does not become 
more fragmented during social conditions in the rock hyrax.   
 
5.4.2. Reduced sleep quotas, but better sleep quality in the social condition? 
In addition to sleep fragmentation, it has been proposed that NREM and REM 
quotas are lower in species that sleep in a social condition as they potentially sleep more 
efficiently and obtain all the benefits of sleep in a shorter time frame (Capellini et al., 
2008). This is thought to be due to the fact that sleeping in a group provides a greater 
sense of security allowing the animals to enter “deeper” states of sleep. It has also been 
hypothesized that the shorter sleep times could be attributed to a greater amount of time 
that has to be devoted to social interactions which in effect results in less time available 
for sleep (Capellini et al., 2008). In the current study of the rock hyrax we found that no 
statistically significant differences existed between social and solitary conditions. Total 
sleep and waking times were similar during both condition, the average number of 
episodes for all states was not different nor was the average episode duration for all 
states. The one exception to this was REM0, which was on average 20 s longer during the 
social condition compared to the solitary condition over 24 h, and 40 s longer during the 
dark period.  
 211
In the present study it was also revealed that SWS during SWA-SWS as well as 
SWA-all was significantly greater during the social condition compared to the solitary 
condition (Fig. 5.5 A-C).  The average SWA per day over the three recording days was 
also greater during the social condition during the first two days, but during the last day 
SWA was slightly higher during the solitary condition (Fig. 5.5C). SWA during NREM is 
considered a measure of sleep intensity and sleep debt as well as an indicator of NREM 
homeostasis (Borbély and Neuhaus, 1980; Tobler and Borbély, 1986; Daan et al., 1984; 
Meerlo et al., 1997, 2001; Borbély and Achermann, 1999) The studies by Meerlo et al. 
(1997, 2001) showed that a correlation exists between social conflict and sleep intensity. 
In their studies rats were subjected to a single social defeat for a 1 h period and 
subsequent recording revealed that SWA increased during the following NREM episode. 
Interestingly, the rats did not rectify the sleep debt caused by the social conflict by 
increasing their sleep duration, instead SWA increased and duration remained unchanged. 
This led to the conclusion that both the duration and the nature of the prior waking bout 
influence the need for sleep. It is possible that increased SWA seen during the social 
condition in the rock hyrax could be a result of social interactions, as an increase in this 
activity is only seen during the first two days of recording (although the animals were 
acclimatized to each other for three days prior to recording).  
When comparing SWA during the light versus the dark periods between the social 
and solitary conditions, the results showed that SWA was highest during the light period 
when in the social condition, whereas during the dark period it was highest when in the 
solitary condition. Meerlo et al. (1997, 2001) suggested that SWA during NREM sleep is 
determined by the duration and nature of the previous waking experience. As the results 
 212
of the present study revealed no significant difference in the duration of waking episodes 
between the light and dark period in these animals, it is possible that during social sleep 
the nature of the waking episodes preceding NREM sleep during the light and dark 
periods may be different; however, no physical social conflict was noted behaviourally 
for all animals examined. Despite this, it is interesting though that the SWA for all states 
during the social condition dropped to just slightly below the SWA during the solitary 
condition on the third day of recording. It would have been interesting to investigate 
whether this trend continued if the recording period was extended for a few days longer. 
Extended recording would ultimately resolve the question of whether the increase in 
SWA during the first two days could be attributed to the animals still acclimatizing to 
their social environment, or whether sleep quality actually improves as a result of 
sleeping socially.  
The average REM0 duration was also found to be significantly longer during the 
social compared to solitary condition over 24 h and more so during the dark period (Fig. 
5.3). This increase in REM0 episode duration, plus a very slight increase in total time 
spent in this state, could be a result of stress induced by the social condition. REM0 is the 
form of REM in the rock hyrax that most closely resembles all the characteristic traits of 
REM sleep that has been described in most mammals studied to date. It has been 
hypothesized that on a behavioural level the drive for sleep might be increased by stress 
and many studies have shown that after restraint stress REM sleep time increased 
(Bonnet et al., 1997; Bouyer et al., 1997; Meerlo et al., 2001; Rampin et al., 1991). It is 
also possible that the increase in REM0 duration could be linked to thermoregulation. 
Rock hyraxes are known to be poor thermoregulators and typically sleep huddled in 
 213
groups to conserve body heat (Taylor and Sale, 1969). Horne (2009) proposed that when 
sleeping in a warmer environment, in the case of the rock hyrax this being several 
animals huddled together, body heat loss, and therefore energy expenditure and 
potentially feeding behaviour is reduced. Furthermore it is hypothesized that 
hypothalamic temperature increases, within limits, create a positive feedback loop that 
may result in an increase in REM.  
 
5.4.3. The effects of social sleeping in the rock hyrax 
The present study provides four results of interest regarding the effects of sleeping 
in a social setting for the rock hyrax: (1) total time for all states is similar in both social 
and solitary conditions; (2) the number of episodes for each state is similar in both social 
and solitary conditions; (3) REM0 episode duration is longer, especially during the dark 
period in the social condition; (4) SWA is generally greater in the social setting. These 
results lead to the conclusion that when rock hyraxes sleep in a social setting, they do not 
sleep more or less than when they sleep alone, nor is sleep more fragmented in one 
condition compared to the other. The increase in REM0 duration and the increase in 
overall level of SWA, both indicate a better sleep quality when in the social condition. 
Thus, it appears possible to conclude that sleep quality may be improved when in the 
social setting for the rock hyrax. Whether this conclusion and the observations made in 
this study can be generalized to all social mammals remains an open question for further 
investigation of the kind undertaken here. 
 It is also possible that the increase in SWA seen during social sleep in the rock 
hyrax could be the result of social interactions creating a sleep debt alleviated by 
 214
increasing the number of slow waves. Increased REM0 duration, also seen during social 
sleep, could possibly be the manifestation of better thermoregulative strategies employed 
when sleeping socially. Thus, the differences observed in the current study between a 
social and a solitary condition and the effect on sleep may be related to better sleep 
quality, or may be related to the social and physical conditions experienced during the 
different states. 
 215
Chapter 6 
The interrelations of the distribution of sleep associated nuclei and terminal 
networks in the brain of the rock hyrax, Procavia capensis 
 
6.1. Introduction 
The basal forebrain and hypothalamus as well as the pontomedullary junction of 
the brainstem has been identified as the two main regions within the brain of mammals 
that house the neuronal groups involved in the control and regulation of the sleep-wake 
cycle. The systems of interest in the current study that are involved in the sleep-wake 
cycle include the cholinergic, putative catecholaminergic, serotonergic, orexinergic, 
histaminergic and GABAergic systems. These systems have a broad distribution 
throughout the brain however the nuclei of these systems that are known to be involved 
in the sleep-wake cycle have very specific innervation patterns (Lyamin et al., 2008).  
The cholinergic sleep related nuclei are mainly located within the basal forebrain, 
anterior hypothalamus, midbrain and the dorsal pons. These nuclei form the main arousal 
system of the forebrain and are responsible for the desynchronized electroencephalogram 
(EEG) of waking and rapid eye movement (REM) sleep. They discharge maximally 
during the aforementioned states however during slow wave sleep (SWS) they discharge 
minimally as GABAergic neurons in the vicinity of these cholinergic sleep related nuclei 
inhibit their excitatory actions (Steriade et al., 1990; Siegel, 2000). The cholinergic sleep 
related nuclei of the midbrain and dorsal pons, such as the PPT (pedunculopontine 
tegmental) and LDT (laterodorsal tegmental) nuclei, are selectively active during REM 
sleep and has been reported as being responsible for the suppression  of muscle tone 
 216
during this state (Sakai and Koyama, 1996). A subgroup of these nuclei have also been 
identified as the driving force behind the generation of pontine-geniculate-occipital 
(PGO) spikes, one of the characteristics features associated with  REM sleep in many 
mammals (Lyamin et al., 2008; Steriade et al., 1990).  
One of the most well described sleep related catecholaminergic nuclear group is 
the locus ceroeleus (LC). The neurons of the LC complex are mainly active and fire 
tonically during waking, have a reduced discharge rate during the initial stages of SWS 
and are for the most part inactive during REM. It has also been suggested that the loss of 
muscle tone observed during REM sleep could be attributed to the inactivity of these 
neurons during this particular state (John et al., 2004; Lai et al., 2001; Lyamin et al., 
2008; Siegel et al., 1991, Siegel, 2000). 
The serotonergic brainstem raphe sleep related nuclei have a similar discharge 
pattern as that described for the LC complex and are mainly active during waking 
(Lyamin et al., 2008; Siegel, 2004). It has been reported that these neurons also maintain 
arousal, regulate muscle tone and suppress phasic waking events (Wu et al., 2004). It has 
also been shown that the rostral cluster of serotonergic sleep related nuclei play a role in 
the regulation of PGO spikes and gating the actions of REM sleep (Lyamin et al., 2008; 
Siegel, 2000). 
Between the sleep active nuclei of the anterior hypothalamus and the wake active 
histamine neurons of the posterior hypothalamus the orexinergic sleep related nuclei are 
situated. It has been reported that these neurons have a driving effect on the other arousal 
systems of the brain and through their projections to the brainstems also play a role in the 
 217
regulation of muscle tone (Gerashchenko and Shiromani, 2004; Kiyashchenko et al., 
2002; Lai and Siegel, 1990; Lyamin et al., 2008; Peyron et al., 1998). 
As mentioned previously the histaminerigic sleep related nuclei mainly promote 
wakefulness and are located within the posterior hypothalamus. GABAergic inhibition of 
these nuclei results in sleepiness, and the desynchronized EEG associated with waking as 
well as arousal maintenance of the forebrain is accomplished by the tonic effects of the 
histaminergic, serotonergic and catecholaminergic systems in combination with the 
activity of the cholinergic system (Jonh et al., 2004; Lyamin et al., 2008; Saper et al., 
2001).     
The most potent sleep promoting neuronal groups are those of the GABAergic 
system located within the basal forebrain and anterior hypothalamus. Many of these 
neurons are maximally active during NREM sleep, less active during REM and 
minimally active during waking. At the onset of NREM sleep the discharge rate of these 
neurons is increased which also then have an inhibitory effect on the neurons involved in 
arousal (Lyamin et al., 2008; Szymusiak, 1995; Szymusiak et al., 2001; Siegel, 2004). 
The GABAergic sleep related neurons that are located in the midbrain and dorsal pontine 
regions have also been reported to have an inhibitory effect on the brainstem serotonergic 
and catecholaminergic neuronal groups (Lyamin et al., 2008; Nitz and Siegel, 1997a, b). 
The aim of the present study is to describe the location and the distribution of the 
cholinergic, catecholaminergic, serotonergic, and orexinergic sleep related nuclear groups 
and the interrelation of these groups to the calcium binding GABAergic interneurons, 
parvalbumin (PV), calbindin (CB) and calretinin (CR), and their terminal network 
distributions within the brain of the rock hyrax, Procavia capensis  
 218
6.2. Materials and Methods 
 The brains of six adult rock hyraxes, P. capensis, were used in the present study. 
Permits from the Limpopo and Gauteng Provincial Governments were obtained for the 
capture and transport of the animals from the wild. All animals were treated and used 
according to the guidelines of the University of the Witwatersrand Animal Ethics 
Committee (approval number AESC 2005/8/5), which parallel those of the NIH for the 
care and use of animals in scientific experimentation. Each animal was weighed and then 
euthanized with weight appropriate doses of sodium pentobarbital (200 mg sodium 
pentobarbital/kg, i.p., Euthanase). Upon cessation of respiration the animals were 
perfused intracardially with 0.9% saline followed by 4% paraformaldehyde in 0.1M 
phosphate buffer (PB, pH 7.4) (approximately 1 l/kg of each solution), both solutions 
having a temperature of 4°C. The brains were then carefully removed from the skulls and 
post-fixed overnight in 4% paraformaldehyde in 0.1M PB followed by equilibration in 
30% sucrose in 0.1M PB. The brains were then frozen in dry ice and with a freezing 
microtome sectioned at 50 µm in the coronal plane. A one in ten series of stains was 
made for nissl, myelin, acetylcholine esterase, tyrosine hydroxlase, serotonin, 
hypocretin/orexin, histamine, parvalbumin, calbindin, and calretinin. Sections kept for the 
Nissl series were mounted on 0.5% gelatine coated glass slides, cleared in a solution of 
1:1 chloroform and absolute alcohol, then stained with 1% cresyl violet to reveal cell 
bodies. Myelin sections were stored in 5% formalin for a period of two weeks and were 
then mounted on 1% gelatine coated glass slides and subsequently stained with silver 
solution to reveal myelin sheaths (Gallyas, 1979). 
 219
  For immunohistochemical staining each section was treated with an endogenous 
peroxidase inhibitor (49.2% methanol: 49.2% 0.1 M PB: 1.6% of 30% H2O2) for 30 min 
and subsequently subjected to three 10 min 0.1 M PB rises. The sections were then 
preincubated in a solution (blocking buffer) consisting of 3% normal serum (normal 
rabbit serum, NRS, for the sections to be stained with the CHAT antibody, or normal goat 
serum, NGS, for the remaining sections), 2% bovine serum albumin (BSA, Sigma) and 
0.25% Triton X100 (Merck) in 0.1M PB, at room temperature for 2 h. The sections were 
then placed for 48 h at 4°C under constant gentle shaking, in the primary antibody 
solution that contained the appropriately diluted primary antibody in blocking buffer (see 
above). The primary antibodies used were anti-cholineacetyltransferase for cholinergic 
(ChAT) neurons (AB144P, Chemicon, raised in goat, at a dilution of 1:2000), anti-
tyrosine hydroxylase for catecholaminergic (TH) neurons (AB151, Chemicon, raised in 
rabbit, at a dilution of 1:10000), anti-serotonin for serotonergic (5-HT) neurons and 
terminal networks (AB938, Chemicon, raised in rabbit, at a dilution of 1:10000), anti-
orexin-A for orexinergic neurons and terminal networks (AB 3704, Millipore, raised in 
rabbit, dilution 1:3000), anti-histamine for histaminergic terminal networks (AB 5508, 
Millipore, raised in rabbit, dilution 1:5000, unfortunately while this antibody labels the 
terminal networks, it fails to labels the neurons), anti-parvalbumin for parvalbumin 
containing interneurons and terminal networks (PV28, Swant, raised in rabbit, dilution 
1:30000), anti- calbindin for calbindin containing interneurons and terminal networks 
(CB38a, Swant, raised in rabbit, dilution 1:20000) and anti-calretinin for calretinin 
containing interneurons and terminal networks (7699/3H, Swant, raised in rabbit, dilution 
1:20000). This was followed by three 10 min rinses in 0.1M PB, after which the sections 
 220
were incubated for 2 h at room temperature in secondary antibody solution. The 
secondary antibody solution contained a 1:1000 dilution of biotinylated anti-goat/anti-
rabbit IgG (BA-5000/BA-1000, Vector Labs) in 3% NRS/NGS, and 2% BSA in 0.1M 
PB. Once this was completed, the sections underwent three 10 min rinses in 0.1M PB, 
followed by a 1 h incubation in AB solution (Vector Labs) and again rinsed. This was 
followed by a 5 min treatment of the sections in a solution consisting of 0.05% 
diaminobenzidine (DAB) in 0.1M PB, after which, and while still in the same solution, 
3µl of 30% H2O2 per 1 ml of solution was added. With the aid of a low power 
stereomicroscope the staining was followed visually and allowed to continue until a level 
was reached where the background staining could assist in reconstruction without 
obscuring the immunopositive structures. Once this level was reached the reaction was 
stopped by placing the sections in 0.1M PB, followed by two 10 min rinses in 0.1M PB. 
All solutions used in the immunohistochemical process had a pH of 7.4. The 
immunohistochemically stained sections were mounted on 0.5% gelatine coated slides 
and left to dry overnight. The mounted sections were dehydrated by placing them in 70% 
alcohol for 2 h at room temperature under gentle shaking and then transferred through a 
series of graded alcohols, cleared in xylene and coverslipped with Depex.  
The sections were observed with a low power stereomicroscope, and the 
architectonic borders traced according to the Nissl and myelin stained sections using a 
camera lucida. The corresponding immuno-stained sections were then matched to the 
drawings and the immuno-positive neurons (for ChAT, TH, 5-HT and orexin-A) and 
medium/high terminal networks (for orexin-A, 5-HT and histamine) were marked. The 
location and distribution of the immuno-positive neurons for parvalbumin, calbindin and 
 221
calretinin were described in relation to the previously identified nuclear groups of the 
cholinergic, catecholaminergic, serotonergic and orexinergic systems (Gravett et al., 
2009, 2011). The drawings were scanned and redrawn with the aid of the Canvas 8 
program.  
 
6.3 Abbreviations 
III – oculomotor nucleus 
Vmot – trigeminal nucleus, motor division 
3n – oculomotor nerve 
3V – third ventricle 
4n – trochlear nerve 
5m – motor division of trigeminal nerve 
5s – sensory division of trigeminal nerve  
4V – fourth ventricle 
A5 – fifth arcuate nucleus  
A6d – locus coeruleus, diffuse portion 
A7d – nucleus subcoeruleus, diffuse portion 
A7sc – nucleus subcoeruleus, compact portion 
A8 – retrorubral nucleus 
A9 – substantia nigra 
A9l – substantia nigra, lateral 
A9m – substantia nigra, medial 
A9pc – substantia nigra, pars compacta 
 222
A9v – substantia nigra, ventral  
A10 – ventral tegmental area 
A10c – ventral tegmental area, central 
A10d – ventral tegmental area, dorsal 
A10dc – ventral tegmental area, dorsal caudal 
A11 – caudal diencephalic group 
A12 – tuberal cell group 
A13 – zona incerta 
A14 – rostral periventricular nucleus 
A15d – anterior hypothalamic group, dorsal division 
A15v – anterior hypothalamic group, ventral division 
APT – anterior pretectal nucleus  
ac – anterior commissure 
AD – anterodorsal nucleus of the dorsal thalamus 
Amyg – amygdala 
AV – anteroventral nucleus of the dorsal thalamus 
B9 – supralemniscal serotonergic nucleus 
C – caudate nucleus 
ca – cerebral aqueduct 
cic – commissure of the inferior colliculus 
CL – central lateral thalamic nucleus 
CLi – caudal linear nucleus 
CM – central medial thalamic nucleus 
 223
CP – cerebral peduncle 
csc – commissure of the superior colliculus 
Diag.B – diagonal band of Broca 
DLG – dorsal lateral geniculate nucleus 
DRc – dorsal raphe nucleus, caudal division 
DRd – dorsal raphe nucleus, dorsal division 
DRif – dorsal raphe nucleus, interfascicular division 
DRl – dorsal raphe nucleus, lateral division 
DRp – dorsal raphe nucleus, peripheral division 
DRv – dorsal raphe nucleus, ventral division 
DTg – dorsal tegmental nucleus 
EW – Edinger-Westphal nucleus 
F – nucleus of the fields of Forel 
f – fornix 
fr – fasciculus retroflexus 
GC – central grey matter 
GI – gigantocellular reticular nucleus 
GP – globus pallidus 
Hbm – medial habenular nucleus 
Hbl – lateral habenular nucleus 
Hyp – hypothalamus 
Hyp.d – dorsal hypothalamic cholinergic nucleus 
Hyp.l – lateral hypothalamic cholinergic nucleus 
 224
Hyp.v – ventral hypothalamic cholinergic nucleus 
IC – inferior colliculus 
ic – internal capsule 
IGL – intergeniculate leaflet 
ILL – intermediate nucleus of the lateral lemniscus 
IML – internal medullary lamina 
IP – interpeduncular nucleus 
Is.Call – island of Calleja 
LDTpc – laterodorsal tegmental nucleus, parvocellular division 
LDTmc – laterodorsal tegmental nucleus, magnocellular division  
lfp – longitudinal fasciculus of pons 
LL – lateral lemniscus 
LP – lateral posterior thalamic nucleus 
lp – lateral parabrachial nucleus 
LT – lateral terminal nucleus of the accessory optic tract 
LV – lateral ventricle 
mcp – middle cerebellar peduncle 
Mc – main orexinergic cluster 
MD – mediodorsal thalamic nucleus 
meV – mesencephalic trigeminal nucleus 
MGB – medial geniculate body 
ML – medial lemniscus 
mlf – medial longitudinal fasciculus 
 225
MM – medial mammillary nucleus, medial part 
MnR – median raphe nucleus 
N.Acc – nucleus accumbens 
N.Bas – nucleus basalis 
oc – optic chiasm 
OR – optic radiation 
OT – optic tract 
OTc – optic tract orexinergic cluster 
P – putamen 
PC – parvicellular reticular nucleus 
PBg – parabigeminal nucleus 
PIR – piriform cortex 
Po – posterior thalamic nuclear group 
PP – peripeduncular nucleus 
PPT – posterior pretectal nucleus  
PPTpc – pedunculopontine nucleus, parvocellular division 
PPTmc – pedunculopontine nucleus, magnocellular division 
PR – prerubral field 
PrC – precommissural nucleus 
PF – parafascicular thalamic nucleus 
PVP – paraventricular thalamic nucleus, posterior part 
py – pyramidal tract 
R – thalamic reticular nucleus 
 226
Re – reuniens thalamic nucleus 
Rmc – red nucleus, magnocellular division  
RtTg – reticulotegmental nucleus of the pons 
S – septal nuclear complex 
SC – superior colliculus 
SCN – suprachiasmatic nucleus 
scp – superior cerebellar peduncle 
Sep.L – lateral septal nucleus 
Sep.M – medial septal nucleus 
sm – stria medullaris of the thalamus 
SON – supraoptic nucleus 
SPO – superior olivary nuclear complex 
SubB – subbrachial nucleus 
TOL – olfactory tubercle 
VA – ventral anterior thalamic nucleus 
VLG – ventral lateral geniculate nucleus 
Vmb – ventromedial basal thalamic nucleus 
VPM – ventral posteromedial thalamic nucleus 
VPL – ventral posterolateral thalamic nucleus 
VPO – ventral pontine nucleus 
xtz – decusation of the trapezoid body 
xscp – decussation of the superior cerebellar peduncle 
ZI – zona incerta 
 227
ZIC – zona incerta, caudal part 
ZIc – zona incerta orexinergic cluster 
 
6.4. Results 
The aim of the present study was to reveal the locations, distributions and 
interactions of the neural systems controlling sleep/wake in the brain of the rock hyrax, P. 
capensis. This was achieved by immunohistochemical staining for 
cholineacetyltransferase (identification of cholinergic neurons), tyrosine hydroxylase 
(identification of catecholaminergic neurons), serotonin (identification of serotonergic 
neurons and terminal networks), orexin-A (identification of oerxinergic neurons and 
terminal networks), histamine (identification of histaminergic terminal networks) 
parvalbumin (identification of parvalbumin containing GABAergic interneurons and 
terminal networks), calbindin (identification of calbindin containing GABAergic 
interneurons and terminal networks) and calretinin (identification of calretinin containing 
GABAergic interneurons and terminal networks). The location of the GABAergic 
interneurons and terminal network distribution was described only in relation to the 
previously identified cholinergic, catecholaminergic, serotonergic and orexinergic nuclei, 
known to be involved in the regulation of the sleep-wake cycle, in the brain of the rock 
hyrax (Gravett et al., 2009, 2011, Chapters 2 and 3).  
 
6.4.1. Cholinergic nuclei 
 The cholinergic nuclei involved in sleep/wake in the brain of the rock hyrax were 
located in the basal forebrain, diencephalon and pontomesencephalon (Fig. 6.1) (Gravett 
 228
et al., 2009). These groups followed the general mammalian organizational plan for this 
system, however some novel features were observed that have not been described in other 
mammals, those being cholinergic neurons in the anterior nuclei of the dorsal thalamus, 
parvocellular and magnocellular divisions of the laterodorsal tegmental and 
pedunculopontine tegmental nuclei (Figs. 6.1 D and E, 6.6B, 6.9A)  (Gravett et al., 2009).  
 The cholinergic neurons of the Islands of Calleja and the olfactory tubercle were 
located in the ventral most portion of the cerebral hemisphere at a level ventral to the 
nucleus accumbens. The medial septal nucleus was identified through the presence of 
ChAT+ neurons located within the septal nuclear complex in the rostral half of the medial 
wall of the cerebral hemisphere. The diagonal band of Broca was located anterior to the 
hypothalamus in the ventromedial corner of the cerebral hemisphere. ChAT+ neurons 
found ventral to the globus pallidus at the level of the anterior commissure, and caudal 
and dorsal to the olfactory tubercle were assigned to nucleus basalis (Fig. 6.1 A and B).  
Within the diencephalon, the dorsal hypothalamic nucleus was located between 
the wall of the third ventricle and the fornix within the dorsomedial region of the 
hypothalamus. The ventral hypothalamic nucleus was identified as a cluster of palely 
stained, widely scattered ChAT+ neurons within the ventromedial aspect of the 
hypothalamus that extended ventrolaterally past the level of the fornix. Within the 
dorsolateral region of the hypothalamus, lateral to the fornix, a low density of palely 
stained ChAT+ neurons were identified as the lateral hypothalamic cholinergic nucleus 
(Fig 6.1 D and E) (Gravett et al., 2009). Within the anterior and dorsal region of the 
dorsal thalamus the anterodorsal nucleus (AD) was seen to exhibit a strong cholinergic 
neuropil staining. Around the margins of the AD nucleus a small number of scattered 
 229
ChAT+ neurons were observed. Lateral to the anterodorsal nucleus, the anteroventral 
nucleus (AV) exhibited a number of ChAT+ neurons along the upper medial and lateral 
borders, plus a much paler ChAT+ neuropil staining (Fig. 6.1 D and E, 6.6B).  
 In the pontomesencephalon cholinergic nuclei included the parabigeminal nucleus 
(PBg), the pedunculopontine tegmental nucleus (parvo- and magnocelluar divisions, 
PPTpc and PPTmc) and the laterodorsal tegmental nucleus (parvo- and magnocellular 
divisions, LDTpc and LDTmc) (Figs. 6.1 K-M, 6.9). A prominent parabigeminal nucleus 
was located ventral and slightly anterior to the inferior colliculus within the lateral aspect 
of the midbrain tegmentum.. Within the dorsal aspect of the isthmic and pontine 
tegmental regions, anterior to the trigeminal motor nucleus, a group of ChAT+ neurons 
was identified as the PPT.  In contrast to other mammals studied, the morphology of the 
cholinergic neurons within this region was not homogenous, and we could readily 
identify magnocellular and parvocellular ChAT+ neurons within this region. The ChAT+ 
neurons assigned to the magnocellular PPT nucleus, PPTmc, were found medial and 
ventral to the superior cerebellar peduncle, and in the anterior portion were found to 
intermingle with the fibres of the superior cerebellar peduncle.  In the more caudal 
aspects of the PPT, in a position dorsal and lateral to the magnocellular neurons, a 
moderately dense cluster of small ChAT+ interneurons, the PPTpc, was observed.  The 
PPTpc nucleus was sometimes clearly distinguished from the PPTmc nucleus on 
topological grounds, as it was found lateral and dorsal to the superior cerebellar peduncle 
in the posterior aspect of the midbrain tegmentum; however, the size of the soma was the 
most reliable distinction between these two nuclei (Fig. 6.1 K and L). The LDT was 
identified as a moderate to high density of ChAT+ neurons within the ventrolateral region 
 230
of the pontine periventricular- and periaqueductal grey matter. The magnocellular 
division of the LDT, LDTmc, was located at the ventrolateral border of the pontine 
periventricular- and periaqueductal grey matter. The parvocellular division of the LDT, 
LDTpc, was found medial and dorsal to the magnocellular LDT nucleus, but still within 
the periventricular- and periaqueductal grey matter (Fig. 6.1 L and M) (Gravett et al., 
2009). 
 
6.4.2. Catecholaminergic nuclei 
 The catecholaminergic nuclei involved in the sleep-wake cycle were located 
within the diencephalic and pontine regions of the brain and consisted of: the dorsal 
division of the anterior hypothalamic nucleus (A15d); the ventral division of the anterior 
hypothalamic nucleus (A15v); the rostral periventricular nucleus (A14); the zona incerta 
(A13); the tuberal nucleus (A12); and the caudal diencephalic nucleus (A11); and the 
locus coeruleus complex. The A15d nucleus was located between the third ventricle and 
the fornix within the anterior portion of the hypothalamus. The A15v nucleus was located 
near the floor of the brain adjacent to the optic tracts within the ventrolateral region of the 
hypothalamus. A moderate density of TH+ neurons close to wall of the third ventricle 
throughout the hypothalamus was identified as the A14 nucleus. The A13 nucleus was 
identified as a moderate density of TH+ neurons located within the dorsal and lateral 
regions of the hypothalamus, lateral to the fornix, extending into and around the zona 
incerta in the ventral thalamus. In the ventral portion of the hypothalamus, within and in 
close proximity to the arcuate nucleus, a low to moderate density of TH+ neurons was 
identified as the A12 nucleus. Around the posterior pole of the third ventricle, in the most 
 231
caudal part the hypothalamus, a low density of TH+ neurons arranged in columns on 
either side of the midline was identified as the A11 nucleus (Fig. 6.1 A-H).  
A7sc (the subcoeruleus compact nucleus) was located within the dorsal region of 
the pontine tegmentum adjacent to the periventricular grey matter. TH+ neurons 
ventrolateral to A7sc, anterior to the trigeminal motor nucleus, in the lateral and 
dorsolateral region of the pontine tegmentum were assigned to the A7d (subcoeruleus 
diffuse nucleus) nucleus. Some of the TH+ neurons assigned to A7d were also located 
medial and ventral to the superior cerebellar peduncle. Only a small number of TH+ 
neurons were located within the ventrolateral portion of the periventricular grey matter 
adjacent to the A7sc, and these were identified as A6d (locus coeruleus diffuse nucleus) 
(Fig. 6.1 M and D) (Gravett et al., 2009). 
 
6.4.3. Serotonergic nuclei 
The serotonergic nuclei in all eutherian mammals are typically divided into a 
rostral and caudal cluster (e.g. Törk, 1990). The serotonergic nuclei associated with 
sleep/wake states are mainly located within the rostral cluster and included the caudal 
linear nucleus (CLi), the supralemniscal nucleus (B9), the median raphe nucleus (MnR) 
and the dorsal raphe nuclear complex (DR). A moderate density of 5HT+ neurons was 
located immediately dorsal to the interpeduncular nucleus, and anterior to the decussation 
of the superior cerebellar peduncle in the ventral midline of the midbrain tegmentum was 
designated as the caudal linear nucleus (CLi) (Fig. 6.1 J and K). The B9 nucleus was 
identified as a loosely packed arc of 5HT+ neurons, lateral to the interpeduncular nucleus 
and caudal to the A9pc nucleus. This nucleus was located superior to the medial 
 232
lemniscus and was found to be continuous with the ventrolateral neurons of the CLi 
nucleus (Fig. 6.1 J and K). In a pararaphe position, two densely packed columns of 5HT+ 
neurons, extending dorsal to ventral along the midbrain and pontine tegmentum, were 
identified as the MnR nucleus. The neurons forming this nucleus extended from the 
caudal aspect of the superior cerebellar peduncle to the level of the anterior most aspect 
of the trigeminal motor nucleus (Fig. 6.1 K and L).  
A dense cluster of 5HT+ neurons within the most ventral medial portion of the 
periaqueductal grey matter, between the two medial longitudinal fasciculi, was identified 
as the DRif nucleus (interfascicular division of the dorsal raphe) (Fig. 6.1 K and L). 
Immediately dorsal to the DRif nucleus and caudal to the oculomotor nerve nuclei, a high 
density of 5HT+ neurons was identified as the DRv nucleus (ventral division of the 
dorsal raphe). The DRd nucleus (dorsal division of the dorsal raphe) was identified as a 
high density of 5HT+ neurons located ventral to the inferior border of the cerebral 
aqueduct and immediately dorsal to the DRv nucleus. A very low density of 5HT+ 
neurons located lateral to the DRd and DRv nuclei, anterior to the ChAT+ neurons of the 
LDT, in the ventrolateral portion of the periaqueductal grey matter was identified as the 
DRp nucleus (peripheral division of the dorsal raphe). In the adjacent midbrain 
tegmentum, a small number of 5HT+ neurons forming part of the DRp nucleus were 
observed. Adjacent to the ventrolateral edge of the cerebral aqueduct, in a position 
dorsolateral to the DRd nucleus, a group of 5HT+ neurons were assigned to the DRl 
nucleus (lateral division of the dorsal raphe. An arc of 5HT+ neurons across the dorsal 
midline of the periventricular grey matter, where the cerebral aqueduct opens into the 
fourth ventricle, formed by the caudal coalescence of the two lateralized clusters of the 
 233
DRl nucleus, was identified as the DRc nucleus (lateral division of the dorsal raphe) 
(Gravett et al., 2009) (Fig. 6.1 K-M). 
 
6.4.4. Orexinergic nuclei 
 Orx+ neurons were identified mainly within the hypothalamus of the brain of the 
rock hyrax. These neurons were grouped into three distinct clusters; the main cluster 
(Mc), the zona incerta cluster (ZIc) and the optic tract cluster (OTc). The main cluster 
was identified as a large group of dense Orx+ neurons located lateral to the third ventricle 
in the perifornical region, with a large number of neurons extending medially from this 
area, as well as into the dorsomedial and lateral hypothalamic areas. The majority of 
Orx+ neurons belonged to this cluster. From the main cluster a group of Orx+ neurons 
extended laterally into the region of the zona incerta which was termed the zona incerta 
cluster. The third cluster extended ventrolaterally from the main cluster to the 
ventrolateral region of the hypothalamus adjacent to the optic tract (Gravett et al., 2011) 
(Fig. 6.1 D and E). 
 
6.4.5. Serotonergic terminal networks related to the sleep-wake nuclei 
 A serotonergic terminal network distribution of medium density was observed 
projecting to all the cholinergic sleep related nuclei (see above) except for PPTpc and 
PPTmc where the 5HT+ terminal network was found to be very low in density (Fig. 6.1 
K and L). Most of the catecholaminergic sleep nuclei were coincident with a medium 
density serotonergic terminal network distribution with the exception of the A7d, A7sc 
and A15d nuclei. The density of the serotonergic terminal networks in the region of the 
 234
subcoeruleus (A7) was found to be low to absent, whereas the serotonergic terminal 
network coincident with the A15d nucleus was found to be high-density around the 
neurons located closest to the third ventricle and medium-density around those extending 
laterally from the third ventricle. A medium-density serotonergic terminal network 
distribution was also noted in the region of the three orexinergic nuclei (Fig. 6.1 A-N, 
6.2). 
 
6.4.6. Orexinergic terminal networks related to the sleep-wake nuclei 
 Orexinergic terminal networks of medium density were found projecting to the 
basal forebrain and pontomesencepalic cholinergic nuclei, with the exception of PPTpc 
and PPTmc, where a low to very low-density orexinergic terminal network was observed 
(Fig. 6.1 K and L). A high density of orexinergic terminal networks was evident in the 
areas surrounding the third ventricle while the areas extending laterally from the ventricle 
and thus the majority of the hypothalamus received a medium density of orexinergic 
terminal networks (this gradient of density was coincident with all cholinergic, 
catecholaminergic and orexinergic nuclei found within the hypothalamus). The pontine 
catecholaminergic A6d and A7d nuclei were coincident with a medium density 
orexinergic terminal network, while the A7sc was coincident with a low-density 
orexinergic terminal network. Serotonergic nuclei coincident with orexinergic terminal 
networks of medium density included the MnR, DRl, CLi nuclei, the medial half of the 
B9 nucleus and the part of the DRc nucleus that extended laterally from the midline (Fig. 
6.1 A-N) A high density of orexinergic terminal networks was observed in the midline of 
the pontine region and encapsulated the serotonergic DRc, DRd, DRv, and DRif nuclei. 
 235
The only serotonergic sleep related nucleus coincident with a low-density orexinergic 
terminal network was DRp (Fig. 6.1 K-M, 6.3, 6.6C). 
      
6.4.7. Histaminergic terminal networks related to the sleep-wake nuclei 
 A medium density histaminergic terminal network was observed in the regions 
where the basal forebrain, diencephalic and pontomesencaphalic (excluding PPTpc and 
PPTmc) cholinergic nuclei were found. Within both divisions of the PPT a low-density to 
near absent histaminergic terminal network was observed. Areas of high density 
histaminergic terminal networks overlapping the cholinergic sleep related nuclei included 
the dorsal and ventral hypothalamic cholinergic nuclei, and parts of LDTpc closest to the 
midline. A mixture of both medium and high-density histaminergic terminal networks 
were observed in the areas coincident with the diencephalic catecholaminergic sleep 
related nuclei. Medium densities were coincident with the A11 and A13 nuclei and the 
lateral parts of the A15d nucleus. High densities were observed overlapping the A11 at 
the level of the fasciculus retroflexus, A12, A14, A15v nuclei, and the part of the A15d 
nucleus that was located closest to the ventricle. Within the pontine region a 
predominantly medium to low density histaminergic terminal network was seen. A 
medium density network was observed overlapping the majority of the A6d nucleus, but 
a high density was observed in the rostral portion of this nucleus. A low to absent 
histaminergic terminal network distribution was observed in the A7d and A7sc nuclei.  
Serotonergic nuclei coincident with a medium density histaminergic terminal 
network included the B9, CLi, DRl, MnR (only part of this nucleus immediately dorsal to 
the interpeduncular nucleus), DRif nuclear groups and the lateral parts of the DRd, DRv, 
 236
and DRc nuclei. High density histaminergic terminal networks were observed 
overlapping the DRc (part located closest to the fourth ventricle) and the DRd and DRv 
nuclei (parts located closest to the midline). The medial half and dorsolateral region of 
the main cluster of orexinergic immunoreactive neurons were located in a region of high 
density histaminergic terminal networks while the ventrolateral region was in a region of 
medium density. A high density histaminergic terminal network was seen overlapping the 
medial two thirds of the zona incerta cluster while the remaining third of this cluster as 
well as the optic tract cluster were coincident with a medium density terminal network 
(Fig. 6.1 A-N, 6.4).  
 
6.4.8. GABAergic inhibitory interneurons related to the sleep-wake nuclei 
 The location and distribution of the calcium binding inhibitory interneurons, 
containing parvalbumin (PV), calbindin (CB) and calretinin (CR), as well as their 
terminal network distribution in relation to the previously identified sleep related nuclei 
was been identified. This was done to examine the extent of the inhibitory actions 
produced by these GABAergic interneurons on the nuclei known to be involved in the 
regulation of the sleep-wake cycle. Inhibitory interneurons were found throughout all 
regions of the brain of the rock hyrax. Parvalbumin was found to be very specific in its 
distribution compared to calbindin and calretinin that generally showed a more global and 
homogenous distribution. These interneurons were all small, oval shaped and mostly 
bipolar, showing the typical appearance of GABAergic inhibitory interneurons (Table 
6.1).  
 
 237
6.4.8.1. GABAergic interneurons and the cholinergic system 
 A low to moderate number of PV+ cells were seen in the regions of the Island of 
Calleja and the olfactory tubercle, but a strong PV+ terminal network was observed. No 
CB+ containing cells were observed within the Islands of Calleja, however within the 
areas surrounding this nucleus a moderate density of CB+ cells as well as a strong CB+ 
terminal network was noted. A moderate to high density of evenly distributed CR+ cells 
and a moderate CR+ terminal network was observed in the areas overlapping the Island 
of Calleja and the olfactory tubercle (Fig. 6.5). A few PV+ stained cells accompanied by 
a strongly labeled PV+ terminal network was seen in the areas overlapping the medial 
septal nucleus. Very few CB+ stained cells were observed in the medial septal region and 
the terminal network distribution was also far weaker than that observed for PV+. A 
moderate number of CR+ cells, accompanied by a strong CR+ terminal network, were 
associated with the medial septal nucleus. A low number of PV+ cells, but a strong PV+ 
terminal network, were observed in the region of the diagonal band of Broca. Within this 
region CB+ and CR+ cells had similar densities and terminal network densities as that 
described for PV. The areas coincident with the nucleus basalis exhibited few to no PV+ 
cells and a PV+ terminal network of low density. CB+ cells were moderate to low in this 
region and a moderate density CB+ terminal network was observed. A low number of 
CR+ cells and low density CR+ terminal network was noted in the region of the nucleus 
basalis (Table 6.1).  
 Numerous PV+ cells, accompanied by a dense PV+ terminal network were noted 
within the AD and AV nuclei where the ChAT+ neurons were located. Within the 
ventrolateral part of the AD nucleus no ChAT+ neurons were observed, however a 
 238
moderate number of PV+ cells and a moderate density terminal networks was noted. A 
dorsomedial band of CB+ cells accompanied by a moderate to high density terminal 
network was noted throughout the AD nucleus. Scattered CB+ cells and a low density 
terminal network were seen throughout the AV nucleus. A mixture of both intensely and 
weakly stained CB+ cells within the ventrolateral half of both the AD and AV nuclei 
were observed. The location and density of CR+ cells within the AD nucleus was similar 
to that described for CB. Within the AV nucleus very few to no CR+ cells and a low 
density CR+ terminal network was noted (Fig. 6.6 D-F). No PV+ cells or PV+ terminal 
networks were observed in any of the hypothalamic cholinergic nuclei. A moderate 
density of CB+ cell accompanied by intensely stained CB+ terminal networks were seen 
overlapping all of the hypothalamic cholinergic nuclei, whereas a moderate density of 
CR+ cells and medium density CR+ terminal network was observed (Table 6.1). 
No PV+ cells or terminal networks were seen in the areas overlapping the PPT 
and LDT nuclei. A moderate number of CB+ cells and a moderate density CB+ terminal 
network were observed in the region of the PPT and LDT, whereas a slightly lower 
number of CR+ cells and low to moderate density CR+ terminal network was observed. 
No distinction with regard to cell number and terminal network densities for both CB and 
CR could be made between the parvo- and magnocellular divisions of both the PPT and 
LDT (Fig. 6.9). No PV+ cells or terminal networks were seen the region overlapping the 
PBg nucleus. A moderate number of CB+ cells and a moderate density CB+ terminal 
network were observed in this region. The number of CR+ cells was slightly less than 
that observed for CB, and the density of the CR+ terminal network was also slightly less 
dense than the CB+ terminal network (Table 6.1).  
 239
 6.4.8.2. GABAergic interneurons and the catecholaminergic system 
 No PV+ cells or terminal networks were observed in the areas coincident with the 
hypothalamic catecholaminergic nuclei. A moderate density of CB+ cells and a high 
density CB+ terminal network was observed throughout the hypothalamus, whereas a 
moderate density of CR+ cells and a moderate density CR+ terminal network was 
observed. No PV+ cells, but a low-density PV+ terminal network were observed to 
overlap with the locus coeruleus complex. No CB+ cells were seen within the A6d 
nucleus, but in the areas immediately surrounding this nucleus a high density of CB+ 
cells were noted and a high density CB+ terminal network was seen within this nucleus. 
A moderate to high number of CB+ cells were found intermingled with the cells of the 
A7 nucleus and a high-density CB+ terminal network was observed. A moderate to high 
density of CR+ cells terminal network projections were observed throughout the entire 
locus coeruleus complex (Table 6.1, Figs 6.7 and 6.10)  
 
6.4.8.3. GABAergic interneurons and the serotonergic system 
 No PV+ cells were observed in any of the regions overlapping the locations of the 
serotonergic nuclei. A very low PV+ terminal network was observed in the regions 
overlapping the CLi and B9 nuclei. A number of PV+ cells that were seen lateral to CLi 
did not appear to project into this nucleus. A moderate to high density of CB+ cells and 
terminal networks were seen overlapping all the areas corresponding to the serotonergic 
nuclei. A low to moderate density of CR+ cells and terminal networks were observed in 
the regions overlying the DRl and DRp nuclei. Within the regions overlapping the DRd, 
DRv and DRif nuclei a very low number of CR+ cells and a low density CR+ terminal 
 240
network was observed. A moderate number of CR+ cells and moderate density CR+ 
terminal network was observed coincident with the CLi and B9 nuclei. A moderate to 
high number of CB+ cells and a moderate to high density CB+ terminal networks were 
observed in the region of both the DRc and MnR nuclei, whereas a low to moderate 
number of CR+ cells and low to moderate density CR+ terminal network was observed 
(Table 6.1, Fig. 6.8).  
 
6.4.9.4. GABAergic interneurons and the orexinergic system 
No PV+ cells or terminal networks were observed in the areas coincident with the 
hypothalamic orexinergic sleep related nuclei. A moderate density of CB+ and CR+ cells 
were observed throughout the hypothalamus and were coincident with the distribution of 
the orexinergic neurons. A high-density CB+ terminal network and a moderate-density 
CR+ terminal network were seen in these areas (Table 6.1). 
 
 241
 
 
 
 
 
Figure 6.1: Diagrammatic reconstructions of a series of coronal sections through the 
brain of the rock hyrax illustrating from left to the right the architectonic subdivisions 
(based on Nissl and myelin staining), location of cholinergic immunoreactive neurons, 
catecholaminergic neurons, serotonergic neurons and terminal networks, orexinergic 
neurons and terminal networks and histaminergic terminal networks. Each dot represents 
a single immunoreactive neuron. Areas shaded black indicate high-density terminal 
networks and areas shaded grey indicate medium-density terminal networks. Drawing A 
represents the most rostral section, N the most caudal. The coronal sections are 
approximately 1 mm apart. See list for abbreviations. 
 







 249
 
 
 
 
 
Figure 6.2: Photomicrographs of coronal sections through: (A) the ChAT 
immunoreactive neurons of the medial septal nucleus and (B) the corresponding 
serotonergic terminal network; (C) the orexinergic immunoreactive neurons of the main 
cluster in the hypothalamus and (D) the corresponding serotonergic terminal network; (E) 
the tyrosine hydroxylase immunoreactive neurons of the compact portion of the 
subcoeruleus (A7sc) and (F) the corresponding serotonergic terminal network. Note the 
medium serotonergic terminal network in B and D and the low serotonergic terminal 
network in F. In all figures medial is to the left and dorsal towards the top. Scale bar in F 
= 100 µm and applies to all. 
 

 251
 
 
 
 
 
Figure 6.3: Photomicrographs of coronal sections through: (A) the ChAT 
immunoreactive neurons of the nucleus basalis and (B) the corresponding orexinergic 
terminal network; (C) the tyrosine hydroxylase immunoreactive neurons of the dorsal 
division of the anterior hypothalamic group (A15d) and (D) the corresponding 
orexinergic terminal network; (E) the serotonergic immunoreactive neurons of the 
peripheral division of the dorsal raphe nucleus (DRp) and (F) the corresponding 
orexinergic terminal network; (G) the tyrosine hydroxylase immunoreactive neurons of 
the diffuse portion of the locus coeruleus and (H) the corresponding orexinergic terminal 
network. Note the medium-density orexinergic terminal network in D, F and H and the 
low-density orexinergic terminal network in B. In all figures medial is to the left and 
dorsal towards the top. Scale bar in H = 100 µm and applies to all. 
 

 253
 
 
 
 
 
Figure 6.4: Photomicrographs of coronal sections through: (A) the ChAT 
immunoreactive neurons of the diagonal band of Broca and (B) the corresponding 
histaminergic terminal network; (C) the orexinergic immunoreactive neurons of the main 
cluster of the hypothalamus and (D) the corresponding histaminergic terminal network. 
Note the medium-density histaminergic terminal network distribution in B, and the high-
density histaminergic terminal network in D. In all figures medial is to the left and dorsal 
towards the top. Scale bar in D = 100 µm and applies to all. 
 

 255
 
 
 
 
 
Figure 6.5: Photomicrographs of coronal sections through: (A) the ChAT 
immunoreactive neurons of the olfactory tubercle (TOL) and the corresponding (B) 
calbindin (CB), (C) calretinin (CR) and (D) parvalbumin (PV) immunoreactive neurons. 
A low-density of CB and PV immunoreactive neurons and a moderate density of CR 
immunoreactive neurons were observed in the TOL.  In all figures medial is to the left 
and dorsal towards the top. Scale bar in D = 500 µm and applies to all. 
 

 257
 
 
 
 
 
Figure 6.6: Photomicrographs of coronal sections through the anterodorsal (AD) and 
anteroventral (AV) nuclei of the dorsal thalamus of the rock hyrax stained for (A) Nissl 
(B) ChAT (C) Orexin-A (D) Calbindin (CB) (E) Calretinin (CR) and (F) Parvalbumin 
(PV). Note the division of the AV (A) into a dorsomedial and ventrolateral part. A 
moderate number of ChAT+ neurons are located within AD whereas only a few ChAT+ 
neurons mainly located on the dorsolateral periphery are seen in AV. Also note the dense 
orexinergic terminal network in AD, the similar morphology of the peripheral PV+ and 
ChAT+ cells within the AV. A moderate density of CB+ and PV+ neurons are seen 
within the ventrolateral region of the AV whereas few to no CR interneurons are evident. 
In all figures medial is to the left and dorsal towards the top. Scale bar in F = 1 mm and 
applies to all. 
 

 259
 
 
 
 
 
Figure 6.7: Photomicrographs of coronal sections through: (A) the tyrosine hydroxylase 
immunoreactive neurons of the caudal diencephalic group (A11) and the corresponding 
(B) calbindin (CB) (C) calretinin (CR) and (D) parvalbumin (PV) immunoreactive 
neurons in adjacent sections. Moderate to high-density of CB and CR immunoreactive 
neurons are evident, but no PV immunoreactive interneurons were observed intermingled 
with the TH+ neurons of the A11.  In all figures medial is to the left and dorsal towards 
the top. Scale bar in D = 500 µm and applies to all. 
 

 261
 
 
 
 
 
Figure 6.8: Photomicrographs of coronal sections through: (A) the serotonergic 
immunoreactive neurons of the dorsal raphe nuclear complex and the corresponding (B) 
calbindin (CB) (C) calretinin (CR) and (D) parvalbumin (PV) immunoreactive neurons. 
The midline raphe nuclei (DRd, DRv and DRif) exhibited a high density of CB+ neurons, 
a very low density of CR+ neurons and no PV+ neurons. A moderate density of both 
CB+ and CR+ neurons, but no PV+ neurons, were seen in the peripheral nuclei (DRl and 
DRp) of the dorsal raphe complex.  In all figures medial is to the left and dorsal towards 
the top. Scale bar in D = 1 mm and applies to all.  
 

 263
 
 
 
 
 
Figure 6.9: Photomicrographs of coronal sections through: (A) the ChAT 
immunoreactive neurons of the laterodorsal tegmental nucleus illustrating both the parvo- 
(LDTpc) and magnocellular (LDTmc) divisions and the corresponding (B) calbindin 
(CB) (C) calretinin (CR) and (D) parvalbumin (PV) immunoreactive neurons. A 
moderate density of CB+ neurons, a high density of CR+ neurons, but no PV+ neurons, 
were observed in the regions coincident with the LDT. Note that the CB+ and CR+ 
neurons are evenly distributed between the parvo- and magnocellular divisions. Also note 
the difference in the neuronal morphology between CB+, CR+ and ChAT+ neurons of 
the LDTpc, indicating that these smaller ChAT+ neurons are unlikely to be interneurons 
of a GABAerigic nature. In all figures medial is to the left and dorsal towards the top. 
Scale bar in D = 500 µm and applies to all. 
 

 265
 
 
 
 
 
Figure 6.10: Photomicrographs of coronal sections through: (A) the tyrosine hydroxylase 
immunoreactive neurons of the diffuse portion of the locus coeruleus (A6d) and the 
compact portion of the nucleus subcoeruleus (A7sc) and the corresponding (B) calbindin 
(CB), (C) calretinin (CR) and (D) parvalbumin (PV) immunoreactive neurons. A 
moderate density of CB+ neurons was observed in the region of the A6d, but a low 
density was seen in region of A7sc. CR+ neurons were seen in the region of the A6d and 
given the similarity in morphology it would appear that the TH+ neurons of the A6d are 
also CR+. A low to moderate density of CR+ neurons were seen in region of A7sc, but no 
PV+ neurons were observed in the regions that coincide to the A6d or A7sc. Note the 
high density PV+ immunoreactive terminal network overlapping the A7sc indicating 
PV+ input to this nucleus from a distant location. In all figures medial is to the left and 
dorsal towards the top. Scale bar in D = 500 µm and applies to all.  
 

 267 
 
 
 
 
 
 
 
Table 6.1: Illustration of the degree of GABAergic (calbindin, calretinin and parvalbumin) 
neuronal and terminal network distribution in the regions of the cholinergic, 
catecholaminergic, serotonergic and orexinergic nuclei related to the sleep-wake cycle.  In the 
regions where no GABAergic neurons and terminal networks were observed a (-) has been 
used in the table, however regions that exhibited  low densities of neurons and terminal 
networks have been illustrated by (+), moderate densities by (++) and high densities by 
(+++). 
 268 
 
Calbindin Calretinin Parvalbumin 
Cells Terminals Cells Terminals Cells Terminals 
 
Cholinergic nuclei 
 
 
Sep. M + + ++ +++ + +++ 
Diag. B + + ++ +++ -/+ +++ 
TOL  ++ +++ ++/+++ ++ +/++ +++ 
Is. Call - - ++/+++ ++ +/++ +++ 
N. Bas +/++ ++ + + -/+ + 
AD +/++ ++/+++ +/++ ++/+++ ++ ++/+++ 
AV + + -/+ + ++ +++ 
Hypothalamic 
nuclei ++ +++ 
++ 
 
++ - - 
PPT ++ ++ ++ ++ - - 
LDT ++ ++ ++ ++ - - 
PBg ++ ++ ++ ++ - - 
 
      
 
Catecholaminergic nuclei 
 
Hypothalamic 
nuclei ++ +++ ++ ++ - - 
A6d - +++ ++/+++ ++/+++ - + 
A7 ++/+++ +++ ++/+++ ++/+++ - + 
 
      
 
Serotonergic nuclei 
 
CLi ++/+++ ++/+++ ++ ++ - + 
B9 ++/+++ ++/+++ ++ ++ - + 
MnR ++/+++ ++/+++ +/++ +/++ - -/+ 
DRif ++/+++ ++/+++ + + - + 
DRv ++/+++ ++/+++ + + - + 
DRd ++/+++ ++/+++ + + - + 
DRp ++/+++ ++/+++ +/++ +/++ - + 
DRl ++/+++ ++/+++ +/++ +/++ - + 
DRc ++/+++ ++/+++ +/++ +/++ - + 
 
      
 
Orexinergic nuclei 
 
Hypothalamic 
nuclei ++ +++ ++ ++ - - 
 
      
 
 269
6.5. Discussion 
The aim of the present study was to describe the location and the distribution 
of the cholinergic, catecholaminergic, serotonergic, and orexinergic sleep related 
nuclear groups and the interrelation of these groups to the calcium binding 
GABAergic interneurons, parvalbumin (PV), calbindin (CB) and calretinin (CR), and 
their terminal network distributions within the brain of the rock hyrax, Procavia 
capensis. The location and distribution of the nuclei involved in the control and 
regulation of the sleep-wake cycle was found to be similar to that previously 
described by Gravett et al., (2009, 2011) and followed the typical mammalian 
organisational plan for these systems.  
 
6.5.1 The unique cholinergic and orexinergic features of the hyrax brain 
The presence of cholinergic neurons within the anterodorsal (AD) and 
anteroventral (AV) nuclei of the dorsal thalamus as well as the parvo- and 
magnocellular subdivisions of the laterodorsal tegmental (LDT) and 
pedunculopontine tegmental (PPT) nuclei appears to be unique to the rock hyrax and 
have not been described in any other mammal studied to date (Gravett et al, 2009) 
(Figs. 6.6 B and D, 6.9). The serotonergic, orexinergic and histaminergic terminal 
network distributions were similar to what has been previously described in other 
mammals (Leger at al., 2001, Airaksinen and Panula, 1988; Gravett et al., 2011; 
Inagaki et al., 1988), with the exception of the dense orexinergic innervation of the 
AD nucleus that has only been observed previously in the Nile grass rat, Arvicanthis 
niloticus (Novak and Albers, 2002; Nixon and Smale, 2007) and Syrian hamster, 
Mesocricetus auratus, both diurnal mammals (Mintz et al., 2001). It is possible that 
orexinergic innervation of the AD nucleus might be a unique feature associated with 
 270
diurnal mammals. It was also noted that the dense orexinergic innervation of the AD 
nucleus overlapped with cholinergic neurons and terminal networks in this nucleus. It 
is possible that this dense orexinergic innervation of the AD nucleus may act in 
concert with the cholinergic neurons and/or the cholinergic axonal terminals, which in 
turn may influence arousal states and motivational processing (Gravett et al., 2011).  
 
6.5.2 GABAergic interneurons, functional aspects of relevance 
The location and distribution of the calcium binding GABAergic interneurons 
in relation to the sleep related cholinergic, catecholaminergic, serotonergic and 
orexinergic nuclei was for the most part similar to that described in other mammals 
(Celio, 1990; Charara and Parent, 1998; Rogers, 1992). Of these GABAergic 
inhibitory interneurons PV appeared to have no direct overlap with the nuclei 
involved in the sleep-wake cycle although it may project to them from a distance (see 
Figure 6.10D), whereas CB was ubiquitous in the sleep related nuclei in both its 
neuronal and terminal network distribution and CR was observed to be mostly 
ubiquitous, but had a lower cellular number and less dense terminal network 
expression than CB. CB and CR inhibitory interneurons are generally involved in 
local processing and it is possible that those located within and adjacent to the sleep 
related nuclei may be involved in their circuitry. CB interneurons have a tonic or 
regular firing rate, whereas CR interneurons depolarize in sudden burst of activity and 
thus have irregular firing rates. It is possible that the regular firing rate of CB 
interneurons correlates to state maintenance, whereas the irregular firing rate of CR 
interneurons correlates to state changes (Cauli et al., 1997; Ascoli et al., 2010). Some 
of the functions that have been assigned to these interneuron types include the 
regulation of intracellular calcium homeostasis (Braun, 1990; Baimbridge et al., 1992; 
 271
Charara and Parent, 1998), while CB and CR have been reported to protect neurons 
against neurodegenerative processes and calcium overload (Braun, 1990; Celio, 1990; 
Baimbridge et al., 1992; Parent et al., 1996; Charara and Parent, 1998). Within the 
basal forebrain of the rock hyrax, an even distribution of both CB and CR neurons 
and terminal networks was observed; however, within the regions of the cholinergic 
AD, AV, PPT and LDT nuclei, the catecholaminergic locus coeruleus complex and 
the serotonergic dorsal raphe nuclear complex a heterogeneous distribution of CB and 
CR neurons and terminals networks was observed.  
 
6.5.3 GABAergic neurons and the anterior thalamic nuclei 
Within the AD nucleus a dorsomedial band of CB+ and CR+ neurons and a 
moderate to high-density terminal network distribution was observed. The AV 
nucleus contained a moderate density of scattered CB+ neurons with a low-density 
terminal network, whereas very few CR+ neurons and a low-density terminal network 
distribution were observed. It also appears from the Nissl stained sections that the AV 
nucleus can be architectonically subdivided into dorsomedial and ventrolateral 
subnuclei. Within the dorsomedial AV subnucleus, where the ChAT+ neurons were 
identified, PV+ neurons were also located. Both the ChAT+ and PV+ neurons in this 
region have similar morphologies and it is possible that these ChAT+ neurons are also 
PV+ neurons – It has been reported by Bayraktar et al. (1997) that nearly all VIP+ 
and approximately 88% of ChAT+ neurons also contain GABA. The ventrolateral AV 
subnucleus did not contain any ChAT+ neurons but a moderate density of smaller 
PV+ neurons with a different morphology was evident (Fig. 6.6). It is interesting that 
both these anterior thalamic nuclei exhibit a moderate density of CB+ and PV+ 
interneurons as both CB+ and PV+ interneurons have been described as fast spiking 
 272
cells that discharge tonically and that this discharge pattern is possibly correlated to 
state maintenance (Cauli et al., 1997). GABAergic neurons located within the anterior 
hypothalamus have also been described as being maximally active during NREM 
sleep and that the neurons located in this region as well as the basal forebrain could be 
considered as the most potent sleep promoting neurons (Szymusiak, 1995; Szymusiak 
et al., 2001; Siegel, 2004). Furthermore given that both the cholinergic and 
orexinergic systems in general appear to be related to activity requiring both alertness 
and awareness, the ChAT+ neurons and ChAT+ and Orx+ terminal networks 
observed within the AD and AV nuclei may have the effect of enhancing the 
wakefulness promoting functions of the anterior thalamic nuclei in the rock hyrax. It 
is possible that the presence of the CB+ and CR+ GABAergic interneurons within the 
anterior thalamic nuclei may act against the action of these wakefulness promoting 
cholinergic neurons and in doing so initiate and maintain NREM sleep, while the 
dense orexinergic terminal network (Gravett et al., 2011) may be involved in exciting 
the neurons and in doing so promote wakefulness (Gerashchenko and Shiromani, 
2004). Double-immunohistochemical labelling as well as single unit recording studies 
would be required before the aforementioned possibilities could be confirmed for 
these anterior thalamic nuclei in the brain of the rock hyrax. 
 
6.5.4 GABAergic neurons and the pontine cholinergic nuclei 
An even distribution of both CB+ and CR+ neurons were observed in the 
regions coincident with the cholinergic parvo- and magnocellular LDT and PPT (Fig. 
6.9). The ChAT+ neurons located within these nuclei had a different morphology than 
the intermingled CB+ and CR+ neurons, the ChAT+ neurons were larger and 
multipolar whereas the CB+ and CR+ neurons were smaller and bipolar. Based on the 
 273
clear difference in neuronal morphology between the cholinergic neurons and the 
aforementioned calcium binding inhibitory interneurons, it is likely that the 
parvocellular divisions of both the LDT and PPT are not inhibitory interneurons; 
however double-immunohistochemical labelling would be required to confirm this. If 
the double-immunohistochemical labelling revealed that these neurons are not 
inhibitory interneurons, the implication would be that the LDT and PPT within the 
brain of the rock hyrax would be more complexly organized than that seen in any 
other mammals studied to date. The complexity in the organization seen in the LDT 
and PPT could also possibly provide clues as to the reasons behind the peculiar 
aspects of REM sleep observed in these animals (see Chapters 4 and 5). Another 
unusual difference observed within the region of LDT was the absence of PV+ 
interneurons, but a dense PV terminal network within the magnocellular division. It is 
possible that the dense PV+ terminal network is a result of PV+ projections 
originating from PV+ neuronal populations located a distance away from the 
magnocellular division of the LDT nucleus. 
 
6.5.5 GABAergic neurons and the locus coeruleus complex 
A moderate to high density of CB+ and CR+ neurons were observed in the 
region of the locus coeruleus. The CR+ neurons appeared to have a similar neuronal 
morphology and were located in the same position as the TH+ neurons that define the 
locus coeruleus; this was not the case with the CB+ neurons (Fig. 6.10). It is possible 
that TH+ neurons of locus coeruleus might also be CR+ neurons, but double-
immunohistochemical labelling would be needed to confirm this possibility. A low 
density of CB+ and CR+ neurons were also observed in the regions overlapping the 
subcoeruleus. The location of CB+ and CR+ neurons within the locus coeruleus 
 274
complex of the rock hyrax is thus different to that described for the rat locus coeruleus 
complex, where both these interneuron types were reported to be absent (Rogers, 
1992).   
 
6.5.6 GABAergic neurons and the serotonergic dorsal raphe complex 
A heterogeneous CB+ neuronal distribution was seen in the regions of the 
serotonergic dorsal raphe nuclei. The median raphe nuclear groups (i.e. DRd, DRv, 
DRif) exhibited a higher density of CB+ neurons, whereas those located laterally (i.e. 
DRl and DRp) exhibited a more moderate density. CR+ neurons were present in 
moderate density in the lateral dorsal raphe nuclei (DRl and DRp), whereas they were 
observed in a low to almost absent density in the regions coincident to the midline 
nuclear groups (DRd, DRv and DRif). No PV+ neurons were observed overlapping 
the serotonergic dorsal raphe nuclear complex (Fig. 6.8). These observations are 
different to that described in other mammals. In primates PV+ neurons have been 
reported to present in all the subnuclei of the dorsal raphe, whereas CB+ and CR+ 
neurons were not present (Charara and Parent, 1998). The rat brain has been described 
as being devoid of PV+ neurons in this region, with only CB+ neurons being present 
in the caudal part of the dorsal raphe and CR+ neurons in the regions of the DRl and 
DRc nuclei (Celio, 1990; Arai et al., 1991). It has been hypothesized that in the rat 
some CR+ neurons in the dorsal raphe project selectively to the striatum, globus 
pallidus and ventrobasal thalamus (Krzywkowski et al., 1995); however, it is 
unknown whether CB+ and PV+ neurons are part of these projections (Charara and 
Parent, 1998). It is possible that some of the GABAergic interneurons overlapping 
these serotonergic dorsal raphe nuclei might have an inhibitory effect on the activity 
of the serotonergic neurons and in doing so promote sleepiness. It has also been 
 275
reported that the rostral cluster of serotonergic sleep related nuclei play a role in the 
gating, inhibiting or disinhibiting aspects of REM sleep (Siegel, 2000). It is possible 
that the lack of some the characteristic features associated with REM sleep and the 
peculiarity of REM sleep in particular that has been observed in the rock hyrax could 
be attributed to the differences noted with regard to the distribution and location of 
these interneuronal types and terminal networks and their specific actions in the 
regions coincident to the serotonergic nuclei.    
 
6.5.7 Summary 
In the present study the location and distribution of the classical nuclei 
involved in the control and regulation of the sleep-wake cycle in the rock hyrax 
(cholinergic, catecholaminergic, orexinergic) was not significantly different to that 
described in other mammals. There appears to be a conservancy, to a degree, across 
most species with regard to the systems involved in the sleep-wake cycle both 
anatomically and physiologically. The distribution and interrelation of the 
GABAergic interneuron types were also, for the most part, not significantly different 
to that described for other mammals; however, the rock hyrax did show some unusual 
distributions patterns with regard to these GABAergic neurons both from a neuronal 
and terminal network distribution aspect. These differences could possibly provide 
clues as to reasons behind the unusual sleep reported, particularly REM, for the rock 
hyrax. 
 276
Chapter 7 
Concluding Remarks 
 
7.1. Conclusion 
The aim of the present thesis was to examine the sleep phenomenology of the 
rock hyrax, Procavia capensis. The rock hyrax is a social diurnal mammal that 
typically lives in colonies on rocky outcrops and is found throughout most parts of 
Southern Africa (Olds and Shoshani, 1982; Skinner and Chimimba, 2005). They share 
an interesting phylogenetic relationship with the African elephant and form part of the 
Afrotherian cohort which is considered the sister group to Xenarthra (Tabuce et al., 
2008). Although many studies deal with the evolutionary history, ecology and 
behavior of these mammals, no study to date has been undertaken to examine the 
physiological measurable parameters of sleep as well as the accompanying 
neuroanatomy of the systems involved in the control and regulation of the sleep-wake 
cycle in the rock hyrax. An abstract by Snyder (1974) however did report total sleep 
times in three other species of hyrax, namely Procavia johnstoni, Hetrohyrax brucei 
and Dendrohyrax validus. Unfortunately no further information is given regarding the 
methodological approach used in this particular study, no polygraphic evidence is 
supplied to confirm the state assignation nor is there any mention of any spectral 
power analysis that have been performed on the data. The current study is thus to date 
the most complete report of the sleep phenomenology of the rock hyrax. 
The first study undertaken in the process of understanding sleep in these 
mammals was the immunohistochemical identification of the location and distribution 
of the cholinergic, catecholaminergic and serotonergic systems within the brain of the 
rock hyrax. For the most part, the nuclear organization of these three systems closely 
 277
resembled that described for many other mammalian species (Woolf, 1991; Manger et 
al., 2002a; Maseko and Manger, 2007; Maseko et al., 2007; Limacher et al., 2008; 
Bhagwandin et al., 2008). The nuclear organization of the serotonergic system was 
identical to that seen in all eutherian mammals (Törk, 1990; Maseko et al., 2007; 
Moon et al., 2007; Dwarika et al., 2008; Limacher et al., 2008; Bhagwandin et al., 
2008) while the catecholaminergic system was very similar to that seen in rodents 
except for the lack of a C3 nucleus and the compact division of the locus coeruleus 
(A6c). In addition, the diffuse locus coeruleus (A6d) appeared to contain very few 
tyrosine hydroxylase immunoreactive (TH+) neurons.  The cholinergic system 
showed many features in common with that seen in both rodents and primates; 
however, there were three differences of note: (1) cholinergic neurons were observed 
in the anterior nuclei of the dorsal thalamus; (2) cholinergic parvocellular nerve cells 
forming subdivisions of the laterodorsal (LDT) and pedunculopontine (PPT) 
tegmental nuclei were observed at the midbrain/pons interface; and (3) a large number 
of cholinergic nerve cells in the periventricular gray of the medulla oblongata were 
observed.  Thus, while there are many similarities to other mammalian species, the 
nuclear organization of these systems in the rock hyrax show specific differences to 
what has been observed previously in other mammals.  
The second study undertaken focused on describing the distribution of orexin-
A immunoreactive neurons and terminal networks in relation to the previously 
described catecholaminergic, cholinergic and serotonergic systems within the brain of 
the rock hyrax. The staining revealed that the neurons were mainly located within the 
hypothalamus as with other mammals (Wagner et al., 2000; McGranaghan and 
Piggins, 2001; Kruger et al., 2010). The orexinergic terminal network distribution also 
resembled the typical mammalian plan however the rock hyrax did show one atypical 
 278
feature, the presence of a high-density orexinergic terminal network within the 
anterodorsal nucleus of the dorsal thalamus (AD). The dense orexinergic innervation 
of the AD nucleus has only been reported previously in the Nile grass rat, Arvicanthis 
niloticus (Novak and Albers, 2002; Nixon and Smale, 2007) and Syrian hamster, 
Mesocricetus auratus, both diurnal mammals in nature (Mintz et al., 2001) (Syrian 
hamster is also known to be nocturnal and hibernating when in captivity). It is 
possible that orexinergic innervation of the AD nucleus might be a unique feature 
associated with diurnal mammals. It was also noted that the dense orexinergic 
innervation of the AD nucleus coincided with previously identified cholinergic 
neurons and terminal networks in this particular nucleus of the rock hyrax brain. It is 
possible that this dense orexinergic innervation of the AD nucleus in the brain of the 
rock hyrax may act in concert with the cholinergic neurons and/or the cholinergic 
axonal terminals, which in turn may influence arousal states and motivational 
processing. 
The above studies were followed by the telemetric recording of the 
physiological measurable parameters of sleep in both solitary and social conditions. 
Sleep was firstly recorded under solitary conditions continuously for 72 hours from 
five wild captured rock hyraxes. The results revealed that these animals spent on 
average, 66.2% awake, 25.1 % in SWS and 3.5% in REM. REM in the rock hyrax 
was found to be ambiguous, which led to the subdivision thereof into REM1 and 
REM0. REM0 was considered the definitive form of this state whereas REM1 showed 
mixed characteristics of both low voltage slow wave sleep (LS) as well as normal 
REM. According to Bergmann et al. (1987) low voltage slow wave/NREM sleep (LS) 
is characterized by behavioural quiescence, desynchronized EEG, EMG that is at the 
same amplitude as during SWS and theta activity that resembles waking and lower 
 279
than during REM. In the rock hyrax REM1 did exhibit some of these characteristics 
such as behavioural quiescence, desynchronized EEG and EMG similar to SWS; 
however, theta activity could not be accurately calculated due to the placement of our 
EEG electrodes. If REM1 can be considered low voltage slow wave sleep, the 
implications would be that the rock hyrax exhibits the lowest amount of REM sleep of 
any terrestrial mammal studied to date. No significant differences were noted with 
regard to total sleep time (TST), number of episodes and episode duration for all 
states between the light and dark periods. Thus prior classification of the rock hyrax 
as strongly diurnal does not appear to apply to controlled laboratory conditions.   
After the solitary sleep recording commenced sleep was recorded from the 
same individual continuously for 72 hours under social conditions. No statistically 
significant differences between social and solitary conditions were observed in terms 
of total sleep time (TST), time in waking, number of episodes of different states or 
episode duration for most states. The average REM0 (true REM sleep for the hyrax) 
episode duration was found to be greater during social conditions, and this increase in 
episode duration was noted across the 24 h period, but was most pronounced during 
the dark period. Slow wave activity (SWA) was greater during the social sleep 
condition, especially during SWS, for the first two days recording, but this difference 
was not noted for the last day of recording in the social condition. The results are thus 
in agreement with the notion that neither sleep cycle length nor phasing of sleep is 
correlated to sleeping in a social setting; however, the increased intensity of SWA 
during SWS and the increase in episode duration of true REM sleep (REM0) are both 
suggestive of better sleep quality during social conditions than solitary conditions in 
the rock hyrax. 
 280
Finally by means of immunohistochemistry the interrelations of the 
distribution of sleep associated nuclei and terminal networks in the brain of the rock 
hyrax were examined. The results revealed that the location and distribution of the 
nuclei involved in the sleep-wake cycle were typically mammalian with the exception 
of the cholinergic AD and AV nuclei, and the parvo- and magnocellular divisions of 
the PPT and LDT which appears to be specific to the rock hyrax. The terminal 
network distribution of the serotonergic, orexinergic and histaminergic systems was 
also similar to that previously described in other mammals, however a dense 
orexinergic terminal network distribution of the AD nucleus was noted, which as 
stated previously might be a specific feature of the rock hyrax and potentially diurnal 
mammals. Of the GABAergic calcium binding interneurons parvalbumin did not 
appear to be strongly related to any of the sleep related nuclei, calbindin was found to 
be ubiquitous in the areas immediately overlapping and surrounding the sleep related 
nuclei whereas calretinin was found to be less ubiquitous than calbindin. Within the 
basal forebrain calbindin and calretinin showed an even distribution whereas in areas 
such as the serotonergic dorsal raphe and catecholaminergic locus coeruleus complex 
these neurons were more unusual in their distribution. Other atypical features 
observed was the gradient like distribution of calbindin and calretinin neurons within 
the cholinergic AD and AV nuclei and a denser parvalbumin terminal network 
distribution overlapping the magnocellular division of the LDT. It is thus possible that 
the parvocellular divisions of both the LDT and PPT are probably not inhibitory 
interneurons as initially thought. It is furthermore also possible that the LDT and PPT 
are more complex in their organisation than any other mammal studied to date and 
that this difference in complexity could possibly provide clues to the reasons behind 
the differences observed in REM sleep in these animals.  
 281
7.2. The way forward 
 In order to put to rest the issue of the possible functions of the AD and AV 
nuclei as well as the parvo- and magnocellular divisions of the LDT and PPT double 
immuno labelling as well as single unit recording at the targeted sites would have to 
be performed. The double immuno labelling will enable one to determine whether the 
overlap observed between the different systems, for example the dense orexinergic 
terminal network distribution observed in the region of the cholinergic AD nucleus 
and the dense parvalbumin terminal network distribution overlapping the 
magnocellular LDT, is indicative of connectivity, whereas single unit recording at the 
sites of the aforementioned nuclei would provide clues to their possible functions.  
 REM has proven to be very unique in the rock hyrax. In order to determine 
whether REM1 can be classified as a REM or NREM state more sophisticated sleep 
recording studies have to be performed where the electrodes would be placed directly 
above the hippocampus in order to obtain the purest form of theta activity. In concert 
with the aforementioned study would have to be thermoregulative studies measuring 
brain (hypothalamic) as well as body temperature. The theta activity obtained from 
these studies will enable one to determine whether REM1 is indeed low voltage slow 
wave sleep or whether it can be considered a form of REM sleep. The 
thermoregulative studies would provide better clues as to the reasons why REM0 
episode duration increases during the social condition, and whether this increase 
observed could be attributed to better thermoregulative strategies employed when 
these animals sleep socially or whether other factors such as the nature of the social 
interactions could be responsible as proposed. Furthermore by extending the 
recording period during the social conditions one would be able to determine whether 
the increase seen in SWA during the social condition can be attributed to social stress 
 282
or improved sleep quality. The effects of increasing group sizes would be another way 
forward. Does increasing group size improve sleep quality in the social condition or 
would it have no effect? Furthermore when group size increases would co-operative 
vigilance become apparent and would there be a difference in the total sleep times and 
fragmentation of sleep of those individuals sleeping on the periphery as opposed to 
those sleeping in the centre? Finally, it would be great to be able to record sleep from 
rock hyraxes under their natural wild conditions and to compare those results to the 
results obtained from the current captive studies.    
 283
8. References 
 
8.1. Chapter 1: Introduction 
 
Bothma, J. du P., 1971. Order Hyracoidea, Part 12. In:  Meester, J., and Setzer, H.W., 
The mammals of Africa: An identification manual. Smithsonian Institution Press, 
Washington, D.C., pp. 1-8. 
 
Carter, A.M., and Enders, A.C., 2004. Comparative aspects of trophoblast 
development and placentation. Reproductive Biology and Endocrinology, 2(46). 
 
Flower, W. H., and Lydekker, R., 1891. An introduction to the study of mammals, 
living and extinct. Adam and Charles Black, London.  
 
Greenwood, A.D., Englbrecht, C.C., MacPhee, R.D.E., 2004. Characterization of an 
endogenous retrovirus class in elephants and their relatives. BMC Evolutionary 
Biology, 4(38). 
 
Hakeem, A.Y., Hof, P.R., Sherwood, C.C., Switszer III, R.C., Rasmussen, L.E.L., 
Allman, J.M., 2005. Brain of the African elephant (Loxodonta africana): 
Neuroanatomy from magnetic resonance images. The anatomical record part A, 
287A:1117-1127. 
 
Hoeck, H., 2010. Hyraxes order Hyracoidea. IUCN/SSC Afrotheria Specialist Group. 
http://afrotheria.net/Hyracoidea. 
 284
Hoffmeister, D. F., 1967. Tubulidentates, proboscideans, and hyracoideans. In: 
Anderson, S., and Jones, J.K., Recent mammals of the world: A synopsis of families. 
The Ronald Press Co., New York, p. 355-365. 
 
Kavanau, J.L., 2002. REM and NREM sleep as natural accompaniments of the 
evolution of warm-bloodedness. Neuroscience Biobehaviour Review, 26:889-906. 
 
Kellogg, M.E., Burkett, S., Dennis, T.R., Stone, G., Gray, B.A., McGuire, P.M., Zori, 
R.T., Stanyon, R., 2007. Chromosome painting in the manatee supports Afrotheria 
and Paenungulata. BMC Evolutionary Biology, 7(6). 
 
Klein, R. G., and Cruz-Uribe, K., 1996. Size Variation in the Rock Hyrax (Procavia 
capensis) and Late Quaternary Climatic Changes in South Africa. Quaternary 
Research, 46:193-207. 
 
Kleinschmidt, T., Czelusniak, J., Goodman, M., Braunitzer, G., 1986. Paenungulata: 
A comparison of the hemoglobin sequences from elephant, hyrax and manatee. 
Molecular Biology and Evolution, 3(5):427-435.  
 
Lyamin, O.I., Manger, P.R., Ridgway, S.H., Mukhametov, L.M., Siegel, J.M., 2008. 
Cetacean sleep: an unusual for of mammalian sleep. Neuroscience and Biobehavioral 
Reviews, 32(8):1451-84.  
 
 285
Manger, P.R., Fahringer, H., Pettigrew, J.D., Siegel, J.M., 2002a. Distribution and 
morphology of cholinergic neurons in the brain of the monotremes as revealed by 
ChAT immunohistochemistry. Brain Behavior and Evolution, 60:275-297. 
 
Manger, P.R., Fahringer, H., Pettigrew, J.D., Siegel, J.M., 2002b. The distribution and 
morphological characteristics of catecholaminergic cells in the brain of monotremes 
as revealed by tyrosine hydroxylase immunohistochemistry.  Brain Behavior and 
Evolution, 60:298-314. 
 
Manger, P.R., Fahringer, H., Pettigrew, J.D., Siegel, J.M., 2002c. Distribution and 
morphology of serotonergic neurons in the brain of monotremes.  Brain Behavior and 
Evolution, 60:315-332. 
 
Manger, P.R., Ridgway, S.H., Siegel, J.M., 2003. The locus coeruleus complex of the 
bottlenose dolphin (Tursiops truncatus) as revealed by tyrosine hydroxylase 
immunohistochemistry. Journal of sleep research, 12:149-155. 
 
Manger, P.R., 2005. Establishing order at the systems level in mammalian brain 
evolution. Brain Research Bulletin, 66:282-289. 
 
Nicolau, M. C., Akaârir, M., Gamundí, A., Gonzáles, J., and Rial, R. V., 2000. Why 
we sleep: the evolutionary pathway to the mammalian sleep, Progress in 
Neurobiology, 62(4):379-406. 
 
Olds, N., and Shoshani, J., 1982. Procavia capensis. Mammalian Species, 171:1-7. 
 286
Rattenborg, N.C., Amlaner, C.J., Lima, S.L., 2000. Behavioral, neurophysiological 
and evolutionary perspectives on unihemispheric sleep. Neuroscience and 
Biobehavioral Review, 24:817-842. 
 
Rübsamen, K., Hume, I.D., Engelhardt, W.V., 1982. Physiology of the rock hyrax. 
Comparative Biochemistry and Physiology, 72A(2): 217-277. 
 
Sherwood, C.C., Stimpson, C.D., Butti, C., Bonar, C.J., Newton, A.L., Allman, J.M., 
Hof, P.R., 2009. Neocortical neuron types in Xenarthra and Afrotheria: implications 
for brain evolution in mammals. Brain Structure and Function, 213:301-328. 
 
Siegel, J.M., 1995. Phylogeny and the function of REM sleep. Behavioral Brain 
Research, 69:29-34. 
 
Siegel, J.M., 2003. Why we sleep. Scientific American, November: 92-97.  
 
Skinner, J.D., and Smithers, R.H.N., 1990. The Mammals of the Southern African 
Sub Region. University of Pretoria, Pretoria. 
 
Skinner, J.D., Chimimba, C.T., 2005. The Mammals of the Southern African 
Subregion, 3rd Edition. Cambridge University Press, Cape Town. 
 
Snyder, F., 1974. Sleep–waking patterns of Hyracoidea. Sleep Research, 3:87. 
 
 287
Tabuce, R., Asher, R.J., Lehmann, T., 2008. Afrotherian mammals: a review of 
current data. Mammalia, 72:2-14. 
 
8.2. Chapter 2: Nuclear organization and morphology of cholinergic, putative 
catecholaminergic and serotonergic neurons in the brain of the Rock Hyrax, 
Procavia capensis. 
 
Agnati, L.F., Fuxe, K., Zoli, M., Pich, E.M., Benfenati, F., Zini, I., Goldstein, M., 
1986. Aspects on the information handling by the central nervous system: focus on 
cotransmission in the aged rat brain. Progress in Brain Research, 68:291-301. 
 
Andén, N.E., Carlsson, A., Dahlström, A., Fuxe, K., Hillarp, N.Å., Larsson, K., 1964. 
Demonstration and mapping out of nigro-neostriatal dopamine neurons. Life 
Sciences, 15:523-530. 
 
Arnason, U., Adegoke, J., Gullberg, A., Harley, E.H., Janke, A., Kullberg, M., 2008. 
Mitogenomic relationships of placental mammals and molecular estimates of their 
divergences. Gene, 421:37–51. 
 
Badlangana, N.L., Bhagwandin, A., Fuxe. K., Manger, P.R., 2007. Distribution and 
morphology of putative catecholaminergic and serotonergic neurons in the medulla 
oblongata of a sub-adult giraffe, Giraffa camelopardalis.  Journal of Chemical 
Neuroanatomy, 34:69-79. 
 
 288
Bartus, R.T., Dean, R.L., Beer, B., 1982. Neuropeptide effect on memory in aged 
monkeys. Neurobiology of Aging, 3:61-68. 
 
Bayraktar, T., Staiger, J.F., Acsady, L., Cozzari, C., Freund, T.F., Zilles, K., 1997. 
Co-localization of vasoactive intestinal polypeptide, gamma-amino-butyric acid and 
choline acetyltransferase in the neocortical interneurons of the adult rat. Brain Res. 
757, 209–217.  
 
Bhagwandin, A., Fuxe, K., Manger, P.R., 2006. Choline acetyltransferase 
immunoreactive cortical interneurons do not occur in all rodents: a study of the 
phylogenetic occurrence of this neural characteristic. Journal of Chemical 
Neuroanatomy, 32:208-216. 
 
Bhagwandin, A., Fuxe, K., Bennett, N.C., Manger, P.R., 2008. Nuclear organization 
and morphology of cholinergic, putative catecholaminergic and serotonergic neurons 
in the brains of two species of African mole-rat. Journal of Chemical Neuroanatomy, 
35:371-387. 
 
Bjarkam, C.R., Sorensen, J.C., Geneser, F.A., 1997. Distribution and morphology of 
serotonin-immunoreactive neurons in the brainstem of the New Zealand white rabbit. 
The Journal of Comparative Neurology, 380:507–519. 
 
Boissard, R., Fort, P., Gervasoni, D., Barbagli, B., Luppi, P.H., 2003. Localization of 
the GABAergic and non-GABAergic neurons projecting to the sublaterodorsal 
 289
nucleus and potentially gating paradoxical sleep onset. European Journal of 
Neuroscience, 18:1627-1639. 
 
Crutcher, K.A., Humbertson, A.O., 1978. The organization of monoamine neurons 
within the brainstem of the North American opossum (Didelphis virginiana). The 
Journal of Comparative Neurology, 179:195–222. 
 
Dahlström, A., Fuxe, K., 1964. Evidence for the existence of monoamine-containing 
neurons in the central nervous system. I. Demonstration of monoamine in the cell 
bodies of brainstem neurons. Acta Physiologica Scandinavica, 62:1-52. 
 
Diksic, M., Young, S.N., 2001. Study of the brain serotonergic system with labelled 
alpha-methyl-L-tryptophan. Journal of Neurochemistry, 78:1185-1200. 
 
Dwarika, S., Maseko, B.C., Ihunwo, A.O., Fuxe, K., Manger, P.R., 2008. Distribution 
and morphology of putative catecholaminergic and serotonergic neurons in the brain 
of the greater canerat, Thryonomys swinderianus. Journal of Chemical Neuroanatomy, 
35:108-122. 
 
Fuxe, K., Hökfelt, T., Ungerstedt, U., 1969. Distribution of monoamines in the 
mammalian central nervous system by histochemical studies. In: Hooper, G., 
Metabolism of Amines in the Brain. Macmillan, London, pp. 10–22. 
 
Fuxe, K., Hökfelt, T., Ungerstedt, U., 1970. Morphological and functional aspects of 
central monoamine neurons. International Review of Neurobiology, 13:93–126. 
 290
Fuxe, K., Manger, P., Genedani, S., Agnati, L., 2006. The nigrostriatal DA pathway 
and Parkinson's disease. Journal of Neural Transmission Supplement, 70:71-83. 
 
Fuxe, K., Marcellino, D., Genedani, S., Agnati, L., 2007a. Adenosine A(2A) 
receptors, dopamine D(2) receptors and their interactions in Parkinson's disease. 
Movement Disorders, 22:1990-2017. 
 
Fuxe, K., Dahlström, A., Höistad, M., Marcellino, D., Jansson, A., Rivera, A., Diaz-
Cabiale, Z., Jacobsen, K., Tinner-Staines, B., Hagman, B., Leo, G., Staines, W., 
Guidolin, D., Kehr, J., Genedani, S., Belluardo, N., Agnati, L.F., 2007b. From the 
Golgi-Cajal mapping to the transmitter-based characterization of the neuronal 
networks leading to two modes of brain communication: wiring and volume 
transmission. Brain Research Reviews, 55:17-54. 
 
Gallyas, F., 1979. Silver staining of myelin by means of physical development. 
Neurological Research, 1:203–209. 
 
Hökfelt, T., Johansson, O., Fuxe, K., Goldstein, M., Park, D., 1976. 
Immunohistochemical studies on the localization and distribution of monoamine 
neuron systems in the rat brain. I. Tyrosine hydroxylase in the mes- and diencephalon. 
Medical Biology, 54:427-53. 
 
Hökfelt, T., Martenson, R., Björklund, A., Kleinau, S., Goldstein, M., 1984. 
Distributional maps of tyrosine-hydroxylase-immunoreactive neurons in the rat brain. 
 291
In: Björklund, A., Hökfelt, T., Handbook of Chemical Neuroanatomy, vol. 2. 
Classical Neurotransmitters in the CNS, part 1. Elsevier, Amsterdam, pp. 277–379. 
 
Jacobs, B.L., Azmitia, E.C., 1992. Structure and function of the brain serotonin 
system.  Physiological Reviews, 72:165–229. 
 
Jones, E.G., 2007. The Thalamus, 2nd Edition. Cambridge University Press, 
Cambridge, UK. 
 
Kalia, M., Fuxe, K., Goldstein, M., 1985a. Rat medulla oblongata. II. Noradrenergic 
neurons, nerve fibers and preterminal processes. The Journal of Comparative 
Neurology, 233:308-332. 
 
Kalia, M., Fuxe, K., Goldstein, M., 1985b. Rat medulla oblongata. III. Adrenergic 
(C1 and C2) neurons, nerve fibers and presumptive terminal processes. The Journal of 
Comparative Neurology, 233:333-349. 
 
Kitahama, K., Geffard, M., Okamura, H., Nagatsu, I., Mons, N., Jouvet, M., 1990. 
Dopamine- and dopa-immunoreactive neurons in the cat forebrain with reference to 
tyrosine hydroxylase-immunohistochemistry. Brain Research, 518:83-94. 
 
Kitahama, K., Sakamoto, N., Jouvet, A., Nagatsu, I., Pearson, J., 1996. Dopamine-
beta-hydroxylase and tyrosine hydroxylase immunoreactive neurons in the human 
brainstem. Journal of Chemical Neuroanatomy, 10:137-146. 
 
 292
Klein, R.G., Cruz-Uribe, K., 1996. Size variation in the rock hyrax (Procavia 
capensis) and late quarternary climatic change in South Africa. Quaternary Research, 
46:193-207. 
 
Lichtensteiger, W., 1966. Uptake of norepinepherine in periglomerular cells of the 
olfactory bulb of the mouse. Nature, 210:955-956. 
 
Lidbrink, P., Jonsson, G., Fuxe, K., 1974. Selective reserpine-resistant accumulation 
of catecholamines in central dopamine neurones after dopa administration. Brain 
Research, 67:439-456. 
 
Limacher, A.M., Bhagwandin, A., Fuxe, K., Manger, P.R., 2008. Nuclear 
organization and morphology of cholinergic, putative catecholaminergic and 
serotonergic neurons in the brain of the Cape porcupine (Hystrix africaeaustralis): 
Increased brain size does not lead to increased organizational complexity. Journal of 
Chemical Neuroanatomy, 36:33-52. 
 
Luppi, P.H., Gervasoni, D., Verret, L., Goutagny, R., Peyron, C., Salvert, D., Leger, 
L., Fort, P., 2007. Paradoxical (REM) sleep genesis: the switch from an aminergic-
cholinergic to a GABAergic-glutamatergic hypothesis. Journal of Physiology (Paris), 
100:271-283. 
 
Maloney, K.J., Mainville, L., Jones, B.E., 1999. Differential c-Fos expression in 
cholinergic, monoaminergic, and GABAergic cell groups of the pontomesencephalic 
 293
tegmentum after paradoxical sleep deprivation and recovery. Journal of Neurocience, 
19: 3057-3072. 
 
Maloney, K.J., Mainville, L., Jones, B.E., 2000. c-Fos expression in GABAergic, 
serotonergic, and other neurons of the pontomedullary reticular formation and raphe 
after paradoxical sleep deprivation and recovery. Journal of Neuroscience, 20:4669-
4679. 
 
Manger, P.R., 2005. Establishing order at the systems level in mammalian brain 
evolution. Brain Research Bulletin, 66:282-289. 
 
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002a. Distribution and 
morphology of cholinergic neurons in the brain of the monotremes as revealed by 
ChAT immunohistochemistry. Brain Behavior and Evolution, 60:275-297. 
 
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002b. Distribution and 
morphology of catecholaminergic neurons in the brain of monotremes as revealed by 
tyrosine hydroxylase immunohistochemistry. Brain Behavior and Evolution, 60:298-
314. 
 
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002c. Distribution and 
morphology of serotonergic neurons in the brain of the monotremes. Brain Behavior 
and Evolution, 60:315-332. 
 
 294
Maseko, B.C., Manger, P.R., 2007. Distribution and morphology of cholinergic, 
catecholaminergic and serotonergic neurons in the brain of Schreiber’s long-fingered 
bat, Miniopterus schreibersii. Journal of Chemical Neuroanatomy, 34:80-94. 
 
Maseko, B.C., Bourne, J.A., Manger, P.R., 2007. Distribution and morphology of 
cholinergic, putative catecholaminergic and serotonergic neurons in the brain of the 
Egyptian Rousette flying fox, Rousettus aegyptiacus. Journal of Chemical 
Neuroanatomy, 34:108-127. 
 
Meister, B., Hökfelt, T., Steinbusch, H.W., Skagerberg, G., Lindvall, O., Geffard, M., 
Joh, T.H., Cuello, A.C., Goldstein, M., 1988. Do tyrosine hydroxylase-
immunoreactive neurons in the ventrolateral arcuate nucleus produce dopamine or 
only L-dopa? Journal of Chemical Neuroanatomy, 1:59-64. 
 
Mitchell, J., Templeton, D., 1981. The origin of the preganglionic parasympathetic 
fibers to the mandibular and sublingual salivary glands in the rat: a horseradish 
peroxidase study. Journal of Anatomy, 132:513–518. 
 
Moon, D.J., Maseko, B.C., Ihunwo, A., Fuxe, K., Manger, P.R., 2007. Distribution 
and morphology of catecholaminergic and serotonergic neurons in the brain of the 
highveld gerbil, Tatera brantsii. Journal of Chemical Neuroanatomy, 34:134-144. 
 
Oda, S., Kuroda, M., Kakuta, S., Tanihata, S., Ishikawa, Y., Kishi, K., 2003. 
Ultrastructure of ascending cholinergic terminals in the anteroventral thalamic 
 295
nucleus of the rat: a comparison with the mammillothalamic terminals. Brain 
Research Bulletin, 59:473-483. 
 
Pal, D., Mallick, B.N., 2006. Role of noradrenergic and GABA-ergic inputs in 
pedunculopontine tegmentum for regulation of rapid eye movement sleep in rats. 
Neuropharmocology, 51:1-11. 
 
Pompeiano, O., 2001. Role of the locus coeruleus in the static and dynamic control of 
posture. Archives Italiennes de Biologie, 139:109-124. 
 
Previc, F.H., 1999. Dopamine and the origins of human intelligence. Brain and 
Cognition, 41:299-350. 
 
Reiner, P.B., Fibiger, H.C., 1995. Functional heterogeneity of central cholinergic 
systems. In: Bloom, F.E., Kupfer, D.J., Psychopharmacology: The Fourth Generation 
of Progress. Raven, New York. pp 147-153. 
 
Rezek, O., Boldogkoi, Z., Tombácz, D., Kovágó, C., Gerendai, I., Palkovits, M., Tóth, 
I.E., 2008. Location of parotid preganglionic neurons in the inferior salivatory nucleus 
and their relation to the superior salivatory nucleus of rat. Neuroscience Letters, 
440:265-269. 
 
Ruggiero, D.A., Anwar, M., Gootman, P.M., 1992. Presumptive adrenergic neurons 
containing phenylethanolamine N-methyltransferase immunoreactivity in the medulla 
oblongata of neonatal swine. Brain Research, 583:105-119. 
 296
Segal, M., Greenberger, V., Israeli, M., Biegon, A., 1988. A correlation between 
regional acetylcholinesterase activity in rat brain and performance in a spatial task. 
Behavioral Brain Research, 30:215-219. 
 
Semba, K., Fibiger, H.C., 1989. Organization of central cholinergic systems. 
Progress in Brain Research, 79:37-63. 
 
Siegel, J.M., 2006. The stuff dreams are made of: anatomical substrates of REM 
sleep. Nature Neuroscience, 9:721-722. 
 
Skinner, J.D., Chimimba, C.T., 2005. The Mammals of the Southern African 
Subregion, 3rd Edition. Cambridge University Press, Cape Town. 
 
Smeets, W.J.A.J., González, A., 2000. Catecholamine systems in the brain of 
vertebrates: new perspectives through a comparative approach. Brain Research 
Reviews, 33:308-379. 
 
Tan, M.M.L., Harvey, A.R., 1989. The cholinergic innervation of normal and 
transplanted superior colliculus in the rat: an immunohistochemical study. 
Neuroscience, 32:511-520. 
 
Tillet, Y., 1995. Distribution of neurotransmitters in the sheep brain. Journal of 
Reproduction and Fertility Supplement, 49:199-220. 
 
 297
Törk, I., 1990. Anatomy of the serotonergic system. Annals of the New York 
Academy of Sciences, 600:9–35. 
 
Tóth, I.E., Boldogkõi, Z., Medveczky, I., Palkovits, M., 1999. Lacrimal preganglionic 
neurons form a subdivision of the superior salivatory nucleus of rat: transneuronal 
labeling by pseudorabies virus. Journal of the Autonomic Nervous System, 77:45–54. 
 
van Groen, T., Kadish, I., Micheal Wyss, J., 2002. Role of the anterodorsal and 
anteroventral nuclei of the thalamus in spatial memory in the rat. Behavioral Brain 
Research, 132:19-28. 
 
Wolff, M., Loukavenko, E.A., Will, B.E., Dalrymple-Alford, J.C., 2008. The 
extended hippocampal-diencephalic memory system: enriched housing promotes 
recovery of the flexible use of spatial representations after anterior thalamic lesions. 
Hippocampus, 18:996-1007. 
 
Woolf, N.J., 1991. Cholinergic systems in mammalian brain and spinal cord. Progress 
in Neurobiology, 37:475–524. 
 
8.3. Chapter 3: Distribution of orexin-A immunoreactive neurons and their 
terminal networks in the brain of the rock hyrax, Procavia capensis. 
 
Abrahamson, E.E., Leak, R.K., Moore, R.Y., 2001. The suprachiasmatic nucleus 
projects to posterior hypothalamic arousal systems. Neuroreport, 12:435-440. 
 
 298
Backberg, M., Hervieu, G., Wilson, S., Meister, B., 2002. Orexin receptor-1 (OX-R1) 
immunoreactivity in chemically identified neurons of the hypothalamus: focus on 
orexin targets involved in control of food and water intake. European Journal of 
Neuroscience, 15:315-328. 
 
Baldo B.A., Daniel R.A., Berridge C.W., Kelley A.E., 2003. Overlapping 
distributions of orexin/hypocretin- and dopamine-β-hydroxylase immunoreactive 
fibers in rat brain regions mediating arousal, motivation and stress. Journal of 
Comparative Neurology, 464:220–237. 
 
Bourgin, P., Huitron-Resendiz, S., Spier, A.D., Fabre, V., Morte, B., Criado, J.R., 
Sutcliffe, J.G., Henriksen, S.J., De Lecea, L., 2000. Hypocretin-1 modulates rapid eye 
movement sleep through activation of locus coeruleus neurons. Journal of 
Neuroscience, 20:7760-7765. 
 
Broberger, C., De Lecea, L., Sutcliffe, J.G., Hokfelt, T., 1998. Hypocretin/orexin-And 
melanin-concentrating hormone-expressing cells form distinct populations in the 
rodent lateral hypothalamus: relationship to the neuropeptide Y and agouti gene-
related protein systems. Journal of Comparative Neurology, 402:460–474. 
 
Brown, K.J., Downs, C.T., 2006. Seasonal patterns in body temperature of free-living 
hyrax (Procavia capensis). Comparative Biochemistry and Physiology part A: 
Molecular and Integrative Physiology, 143:42-49. 
 299
Brown, R.E., Sergeeva, O., Eriksson, K.S., Haas, H.L., 2001. Orexin-A excites 
serotonergic neurons in the dorsal raphe nucleus of the rat. Neuropharmacology, 
40:457–459. 
 
Chen, C.T., Dun, S.L., Kwok, E.H., Dun, N.J., Chang, J.K., 1999. Orexin-A-like 
immunoreactivity in the rat brain. Neuroscience Letters, 260:161-164. 
 
Cutler, D.J., Morris, R., Sheridhar, V., Wattam, T.A., Holmes, S., Patel, S., Arch, 
J.R., Wilson, S., Buckingham, R.E., Evans, M.L., Leslie, R.A., Williams, G., 1999. 
Differential distribution of orexin-A and orexin-B immunoreactivity in the rat brain 
and spinal cord. Peptides, 20:455-1470.   
 
Date, Y., Ueta, Y., Yamashita, H., Yamaguchi, H., Matsukura, S., Kangawa, K., 
Sakurai, T., Yanagisawa, M., Nakazato, M., 1999. Orexins, orexigenic hypothalamic 
peptides, interact with autonomic, neuroendocrine and neuroregulatory systems.  
Proceedings of the National Academy of Sciences U.S.A., 96:748-753. 
 
De Lecea, L., Kilduff, T.S., Peyron, C., Gao, X., Foye, P.E., Danielson, P.E., 
Fukuhara, C., Battenberg, E.L., Gautvik, V.T., Bartlett, F.S., Frankel, W.N., Van den 
Pol, A.N., Bloom, F.E., Gautvik, K.M., Sutcliffe, J.G., 1998. The hypocretins: 
hypothalamus-specific peptides with neuroexcitatory activity.  Proceedings of the 
National Academy of Sciences U.S.A., 95:322-327. 
 
 300
Domínguez, L., Morona, R., Joven, A., González, A., López, J.M., 2010. 
Immunohistochemical localization of orexins (hypocretins) in the brain of reptiles and 
its relation to monoaminergic systems. Journal of Chemical Neuroanatomy, 39:20–34. 
 
Espana, R.A., Baldo, B.A., Kelley, A.E., Berridge, C.W., 2001. Wake-promoting and 
sleep-suppressing actions of hypocretin (orexin): basal forebrain sites of action. 
Neuroscience, 106:699-715. 
 
Espana, R.A., Reis, K.M., Valentino, R.J., Berridge, C.W., 2005. Organization of 
hypocretin/orexin efferents to locus coeruleus and basal forebrain arousal-related 
structures. Journal of Comparative Neurology, 481:160-178. 
 
Fujiki, N., Yoshida, Y., Ripley, B., Honda, K., Mignot, E., Nishino, S., 2001. 
Changes in CSF hypocretin-1 (orexin-A) levels in rats across 24 hours and in 
response to food deprivation. Neuroreport, 12:993-997.  
 
Galas, L., Vaudry, H., Braun, B., Van den Pol, A.N., De Lecea, L., Sutcliffe, J.G., 
Chartrel, N., 2001. Immunohistochemical localization and biochemical 
characterization of hypocretin/orexin-related peptides in the central nervous system of 
the frog Rana ridibunda. Journal of Comparative Neurology, 429:242–252. 
 
Gallyas, F., 1979. Silver staining of myelin by means of physical development. 
Neurological Research, 1:203–209. 
 301
Gatterman, R., Johnston, R.E., Yigit, N., Fritzsche, P., Larimer, S., Ozkurt, S., 
Neumann, K., Song, Z., Colak, E., Johnston, J., McPhee, M.E., 2008. Golden 
hamsters are nocturnal in captivity but diurnal in nature. Biology Letters, 4:253-255. 
 
Gravett, N., Bhagwandin, A., Fuxe, K., Manger, P.R., 2009. Nuclear organization and 
morphology of cholinergic, putative catecholaminergic and serotonergic neurons in 
the brain of the rock hyrax, Procavia capensis. Journal of Chemical Neuroanatomy, 
38:57-74. 
 
Hagan, J.J., Leslie, R.A., Patel, S., Evans, M.L., Wattam, T.A., Holmes, S., Benham, 
C.D., Taylor, S.G., Routledge, C., Hemmati, P., Munton, R.P., Ashmeade, T.E., Shah, 
A.S., Hatcher, J.P., Hatcher, P.D., Jones, D.N., Smith, M.I., Piper, D.C., Hunter, A.J., 
Porter, R.A., Upton, N., 1999. Orexin-A activates locus coeruleus cell firing and 
increases arousal in the rat.  Proceedings of the National Academy of Sciences 
U.S.A., 96:10911–10916. 
 
Horvath, T.L., Diano, S., Van den Pol, A.N., 1999. Synaptic interaction between 
hypocretin (orexin) and neuropeptide Y cells in the rodent and primate hypothalamus: 
a novel circuit implicated in metabolic and endocrine regulations. Journal of 
Neuroscience, 19:1072-1087. 
 
Huang, Z.L., Qu, W.M., Li, W.D., Mochizuki, T., Eguchi, N., Watanabe, T., Urade, 
Y., Hayaishi, O., 2001. Arousal effect of orexin-A depends on activation of the 
histaminergic system.  Proceedings of the National Academy of Sciences U.S.A., 98: 
9965-9970. 
 302
Huesa, G., Van den Pol, A.N., Finger, T.E., 2005. Differential distribution of 
hypocretin (orexin) and melanin-concentrating hormone in the goldfish brain. Journal 
of Comparative Neurology, 488:476–491. 
 
Iqbal, J., Pompolo, S., Sakurai, T., Clarke, I.J., 2001. Evidence that orexin-containing 
neurones provide direct input to gonadotropin-releasing hormone neurones in the 
ovine hypothalamus. Journal of Neuroendocrinology, 13:1033-41. 
 
Kaslin, J., Nystedt, J.M., Östergård, M., Peitsaro, N., Panula, P., 2004. The 
orexin/hypocretin system in zebrafish is connected to the aminergic and cholinergic 
systems. Journal of Neuroscience, 24:2678–2689. 
 
Khorooshi, R.M., Klingenspor, M., 2005. Neuronal distribution of melanin- 
concentrating hormone, cocaine- and amphetamineregulated transcript and orexin B 
in the brain of the Djungarian hamster (Phodopus sungorus). Journal of Chemical 
Neuroanatomy, 29:137-148. 
 
Kruger, J-L., Dell, L-A., Pettigrew, J.D., Manger, P.R., 2010. Cellular location and 
major terminal networks of the orexinergic system in the brains of five 
microchiropteran species. Journal of Chemical Neuroanatomy, 40:256-262. 
 
Kukkonen, J.P., Holmqvist, T., Ammoun, S., Åkermam, K.E.O., 2002. Functions of 
the orexinergic/hypocretinergic system. American Journal of Physiology: Cell 
Physiology, 283:C1567-C1591. 
 
 303
Lee, M.E., Hassani, O.K., Jones, B.E., 2005. Discharge of identified 
orexin/hypocretin neurons across the sleep-wake cycle. Journal of Neuroscience, 
25:6716-6720. 
 
López, J.M., Domínguez, L., Morena, N., González, A., 2009. Comparative 
immunohistochemical analysis of the distribution of orexins (hypocretins) in the brain 
of amphibians. Peptides, 30:873–887. 
 
Matsuzaki, I., Sakurai, T., Kunii, K., Nakamura, T., Yanagisawa, M., Goto, K., 2002. 
Involvement of the serotonergic system in orexin-induced behavioral alterations in 
rat. Regulatory Peptides, 104:119–123. 
 
McGranaghan, P.A., Piggins, H.D., 2001. Orexin-A-like immunoreactivity in the 
hypothalamus and thalamus of the Syrian hamster (Mesocricetus auratus) and 
Siberian hamster (Phodopus sungorus), with special reference to circadian structures. 
Brain Research, 904:234-244. 
 
McNairn, I.S., Fairall, N., 1984. Metabolic rate and body temperature of adult and 
juvenile hyrax (Procavia capensis). Comparative Biochemistry and Physiology part 
A: Comparative Physiology, 79:539-545.  
Methippara, M.M., Alam, M.N., Szymusiak, R., McGinty, D., 2000. Effects of lateral 
preoptic area application of orexin-A on sleep-wakefulness. Neuroreport, 11:3423-
3426. 
 
 304
Mintz, E.M., Van den Pol, A.N., Casano, A.A., Albers, H.E., 2001. Distribution of 
hypocretin-(orexin) immunoreactivity in the central nervous system of Syrian 
hamsters (Mesocricetus auratus). Journal of Chemical Neuroanatomy, 21:225–238. 
 
Moore, R.Y., Abrahamson, E.A., Van den Pol, A., 2001. The hypocretin neuron 
system: an arousal system in the human brain. Archives Italiennes de Biologie, 
139:195-205. 
 
Nakamura, T., Uramura, K., Nambu, T., Yada, T., Goto, K.,Yanagisawa, M., Sakurai, 
T., 2000. Orexin-induced hyperlocomotion and stereotypy are mediated by the 
dopaminergic system. Brain Research, 873:181–187. 
 
Nambu, T., Sakurai, T., Mizukami, K., Hosoya, Y., Yanagisawa, M., Goto, K., 1999. 
Distribution of orexin neurons in the adult rat brain. Brain Research, 827:243-260. 
 
Nixon, J.P., Smale, L., 2007. A comparative analysis of the distribution of 
immunoreactive orexin-A and B in the brains of nocturnal and diurnal rodents. 
Behavioral and Brain Functions, 13:28. 
 
Novak, C.M., Smale, L., Nunez, A.A., 2000. Rhythms in Fos expression in brain 
areas related to the sleep-wake cycle in the diurnal Arvicanthis niloticus. American 
Journal of Physiology- Regulatory, Integrative and Comparative Physiology, 
278:R1267-R1274. 
 
Novak, C.M., Albers, H.E., 2002. Localization of hypocretin-like immunoreactivity 
 305
in the brain of the diurnal rodent, Arvicanthis niloticus. Journal of Chemical 
Neuroanatomy, 23:49-58. 
 
Ohkubo, T., Boswell, T., Lumineau, S., 2002. Molecular cloning of chicken prepro-
orexin cDNA and preferential expression in the chicken hypothalamus. Biochimica et 
Biophysica Acta, 1577:476–480. 
 
Olds N, Shoshani J, 1982. Procavia capensis. Mammalian Species, 171:107. 
 
Ohno, K., and Sakurai, T., 2008. Orexin neuronal circuitry: role in the regulation of 
sleep and wakefulness. Frontiers in Neuroendocrinology, 29:70–87. 
 
Peyron, C., Tighe, D.K., Van den Pol, A.N., De Lecea, L., Heller, H.C., Sutcliffe, 
J.G., Kilduff, T.S., 1998. Neurons containing hypocretin (orexin) project to multiple 
neuronal systems. Journal of Neuroscience, 18:9996-10015. 
 
Piper, D.C., Upton, N., Smith, M.I., Hunter, A.J., 2000. The novel brain neuropeptide, 
orexin-A, modulates the sleep-wake cycle of rats. European Journal of Neuroscience, 
12:726-730. 
 
Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R.M., Tanaka, H., 
Williams, S.C., Richardson, J.A., Kozlowski, G.P., Wilson, S., Arch, J.R., 
Buckingham, R.E., Haynes, A.C., Carr, S.A., Annan, R.S., McNulty, D.E., Liu, W.S., 
Terrett, J.A., Elshourbagy, N.A., Bergsma, D.J., Yanagisawa, M., 1998. Orexins and 
 306
orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled 
receptors that regulate feeding behavior. Cell, 92:573-585.   
 
Shibahara, M., Sakurai, T., Nambu, T., Takenouchi, T., Iwaasa, H., Egashira, S., 
Ihara, M., Goto, K., 1999. Structure, tissue distribution, and pharmacological 
characterization of Xenopus orexins. Peptides, 20:1169–1176. 
 
Singletary, K.G., Delville, Y., Farrell, W.J., Wilczynski, W., 2005.  Distribution of 
orexin/hypocretin immunoreactivity in the nervous system of the green treefrog, Hyla 
cinerea. Brain Research, 1041:231–236. 
 
Snyder, F., 1974. Sleep–waking patterns of Hyracoidea. Sleep Research, 3:87. 
 
Stoyanova, I.I., Rutten, W.L., le Feber, J., 2010. Orexin-A and orexin-B during the 
postnatal development of the rat brain. Cellular and Molecular Neurobiology, 30:81-
89. 
 
Suzuki, H., Kubo, Y., Yamamoto, T., 2008. Orexin-A immunoreactive cells and 
fibers in the central nervous system of the axolotl brain and their association with 
tyrosine hydroxylase and serotonin immunoreactive somata. Journal of Chemical 
Neuroanatomy, 35:295-305.  
 
Szekely, M., Petervari, E., Balasko, M., 2010. Thermoregulation, energy balance, 
regulatory peptides: recent developments. Frontiers in Bioscience, 2:1009-1046. 
 
 307
Tabuce, R., Asher, R.J., Lehmann, T., 2008. Afrotherian mammals: a review of 
current data. Mammalia, 72:2-14. 
 
Taheri, S., Sunter, D., Dakin, C., Moyes, S., Seal, L., Gardiner, J., Rossi, M., Ghatei, 
M., Bloom, S., 2000. Diurnal variation in orexin-A immunoreactivity and prepro-
orexin mRNA in the rat central nervous system. Neuroscience Letters, 279:109-112. 
 
Tao, R., Ma, Z., McKenna, J. T., Thakkar, M. M., Winston, S., Strecker, R. E., 
McCarley, R. W., 2006. Differential effect of orexins (hypocretins) on serotonin 
release in the dorsal and median raphe nuclei of freely behaving rats. Neuroscience, 
141:1101–1105. 
 
Teske, J.A., Billington, C.J., Kotz, C.M., 2010. Hypocretin/orexin and energy 
expenditure. Acta Physiologica, 198:303-312. 
 
Vidal, L., Blanchard, J., Morin, L.P., 2005. Hypothalamic and zona incerta neurons 
expressing hypocretin, but not melanin concentrating hormone, project to the hamster 
intergeniculate leaflet. Neuroscience, 134:1081-1090. 
 
Wagner, D., Salin-Pascual, R., Greco, M.A., Shiromani, P.J., 2000. Distribution of 
hypocretin-containing neurons in the lateral hypothalamus and c-fos-immunoreactive 
neurons in the VLPO. Sleep Research, 3:35-42. 
 
Yamamoto, Y., McKinley, M.J., Nakazato, M., Yamashita, H., Shirahata, A., Ueta, 
Y., 2006. Postnatal development of orexin-A and orexin-B like immunoreactivities in 
 308
the Eastern grey kangaroo (Macropus giganteus) hypothalamus. Neuroscience 
Letters, 392:124-128. 
 
Yamanaka, A., Tsujino, N., Funahashi, H., Honda, K., Guan, J.L., Wang, Q.P., 
Tominaga, M., Goto, K., Shioda, S., Sakurai, T., 2002. Orexins activate histaminergic 
neurons via the orexin 2 receptor. Biochemical and Biophysical Research 
Communications, 290:1237-1245. 
 
Yamanaka, A., Muraki,Y.,Tsujino, N., Goto, K., Sakurai, T., 2003. Regulation of 
orexin neurons by the monoaminergic and cholinergic systems. Biochemical and 
Biophysical Research Communications, 303:120–129. 
 
Yokogawa, T., Marin, W., Faraco, J., Pe´zeron, G., Appelbaum, L., Zhang, J., Rosa, 
F., Mourrain, P., Mignot, E., 2007. Characterization of sleep in zebrafish and 
insomnia in hypocretin receptor mutants. PLoS Biology, 5(10):e277.  
 
Yoshida, Y., Fujiki, N., Nakajima, T., Ripley, B., Matsumura, H., Yoneda, H., 
Mignot, E., Nishino, S., 2001. Fluctuation of extracellular hypocretin-1 (orexin-A) 
levels in the rat in relation to the light-dark cycle and sleep-wake activities. European 
Journal of Neuroscience, 14:1075-1081. 
 
Yoshida, K., McCormack, S., Espana, R.A., Crocker, A., Scammell, T.E., 2006. 
Afferents to the orexin neurons of the rat brain. Journal of Comparative Neurology, 
494:845-861. 
 
 309
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2001. Orexin 
(hypocretin)-like immunoreactivity in the cat hypothlamus: a light and electron 
microscopic study. Sleep, 24:67–76. 
 
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2002. Co-localization of 
hypocretin-1 and hypocretin-2 in the cat hypothalamus and brainstem. Peptides, 
23:1479–1483. 
 
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2004. Distribution of 
hypocretin (orexin) immunoreactivity in the feline pons and medulla. Brain Research, 
995:205-217.  
 
8.4. Chapter 4: Solitary sleep in the rock hyrax, Procavia capensis. 
 
Allison, T., and Cicchetti, P.V., 1976. Sleep in mammals: Ecological and 
constitutional correlates. Science, 194(4266): 732-734. 
 
Affanni, J. M., 1972. Observations on the sleep of some South American marsupials 
and edentates. In: The Sleeping Brain, edited by Chase, M., Los Angeles: Brain 
Information Service/Brain Research Institute, UCLA, pp. 21-23. 
 
Berger, R.J., Walker, J.M., 1972. A polygraphic study of sleep in the tree shrew 
(Tupaia glis). Brain Behavior and Evolution 5:54-69. 
 
 310
Bergmann, B.M., Winter, J.B., Rosenberg, R.S., Rechtschaffen, A., 1987. NREM 
Sleep with Low-Voltage EEG n the Rat. Sleep, 10(1):1-11. 
  
Campbell, S.S., and Tobler, I., 1984. Animal Sleep: A Review of Sleep Duration 
Across Phylogeny. Neuroscience and Behavioral Reviews, 8:296-300. 
 
Capellini, I., Barton, R.A., McNamara, P., Preston, B.T., Nunn, C.L., 2008. 
Phylogenetic analysis of the ecology and evolution of mammalian sleep. Evolution, 
62(7): 1764–1776. 
 
Cirelli, C., and Tononi, G., 2008. Is sleep essential? PLOS Biology, 6(8), e216. 
 
Desiraju, T., Anand, B.K., Singh, B., 1966. Electrographic Studies on the Nature of 
Sleep and Wakefulness. Physiology and Bahvior, 1:285-291. 
 
Elgar, M.A., Pagel, M.D., Harvey, P.H., 1988. Sleep in mammals. Animal Behavior, 
36:1407-1419. 
 
Galvăo de Moura Filho, A.G., Huggins, S.E., Lines, S.G., 1983. Sleep and waking in 
the three-toed sloth Bradypus tridactylus. Comparative Biochemistry and Physiology 
A., 76:345–355. 
 
Gottesman, C., 1992. Detection of seven sleep-waking stages in the rat. Neuroscience 
Biobehavioral Review 16:31-38. 
 
 311
Hallström, B.M., Kullberg, M., Nilsson, M.A., Janke, A., 2007. Phylogenomic data 
analyses provide evidence that Xenarthra and Afrotheria are sister groups. Molecular 
Biology and Evolution, 24(9): 2059-2068. 
 
Horne, J., 2009. Neuroscience and Biobehavioral Reviews REM sleep, energy 
balance and ‘optimal foraging ’. Biobehavioral Reviews, 33:466-474. 
 
Kingdon, J., 1971. East African Mammals. An Atlas of evolution in Africa. Academic 
Press, London. 
 
Kurt, F., 1960. Le sommeil des éléphants. Mammalia, 24: 259-272. 
 
Lesku, J. A., Bark, R. J., Martinez-Gonzalez, D., Rattenborg, N. C., Amlaner, C. J., 
Lima, S. L., Planck, M., and Group, F., 2008. Predator-induced plasticity in sleep 
architecture in wild-caught Norway rats (Rattus norvegicus). Behavioural Brain 
Research, 189:298-305. 
 
Lyamin, O.I., Lapierre, J.L., Kosenko, P.O., Mukhametov, L., Siegel, J.M., 2008. 
EEG asymmetry and spectra power during sleep in the northern fur seal. Journal of 
Sleep Research, 17: 154-165. 
 
Lyamin, O.I., Manger, P.R., Ridgway, S.H., Mukhametov, L.M., Siegel, J.M., 2008. 
Cetacean sleep: an unusual form of mammalian sleep. Neuroscience and 
Biobehavioral Reviews, 32(8):1451-84.  
 
 312
Lyamin, O.I., Peter, O., Kosenko, P.O., Lapierre, J.L., Mukhametov, L.M., Siegel, 
J.M., 2008. Fur seals display a strong drive for bilateral slow wave sleep while on 
land. Journal of Neuroscience, 28(48):12614-12621. 
 
McNamara, P., Capellini, I., Harris, E., Nunn, C.L., Barton, R.A., Preston, B., 2008. 
The Phylogeny of Sleep Database: A New Resource for Sleep Scientists. The Open 
Sleep Journal, 1:11-14. 
 
Moss, C., 1982. Portraits in the wild. Chicago: University of Chicago Press, p371. 
Mukhametov, L. M., Lyamin, O. I., Chetyrbok,  I. S., Vassilyev,  A. A., Diaz, R. P., 
1992. Sleep in an Amazonian manatee, Trichechus inunguis. Experientia, 48:417-419. 
 
Nicolau, M.C., Akaârir, M., Gamundí, A., Gonzáles, J., Rial, R.V., 2000. Why we 
sleep: the evolutionary pathway to the mammalian sleep, Progress in Neurobiology, 
62: 379-406. 
 
Pivik, R.T., Sircar, S., Braun, C., 1981. Nuchal muscle tonus during sleep, 
wakefulness and tonic immobility in the rabbit. Physiology and Behavior 26:13-20. 
 
Rattenborg, N.C, Voirin, B., Vyssotski, A.L., Kays, R.W., Spoelstra, K., Kuemmeth, 
F., Heidrich, W., Wikelski, M. 2008. Sleeping outside the box: 
electroencephalographic measures of sleep in sloths inhabiting a rainforest. Biology 
Letters, 4:402-405. 
 
 313
Rial, R.V., Akaârir, M., Gamundí, A., Nicolau, C. Garau, C., Aparicio, S., Tejada, S., 
Gené, L., González, J., De Vera, L.M., Coenen, A.M.L., Barceló, P., Esteban, S., 
2010. Evolution of wakefulness, sleep and hibernation: From reptiles to mammals. 
Neuroscience and Behavioral Reviews, 34:1144-1160. 
 
Ruckebusch, Y., 1972. Comparative aspects of sleep and wakefulness in farm 
animals. In: M. Chase., The Sleeping Brain. Perspectives in the Brain Sciences, Los 
Angeles: Brain Information Service/Brain Research Institute, UCLA, pp. 23-28. 
 
Siegel, J.M., Manger, P.R., Nienhuis, R., Fahringer, H.M., Pettigrew, J.D., 1999. 
Sleep in the platypus. Neuroscience, 91:391-400. 
 
Siegel, J.M, 2005. REM Sleep. In: Kryger, M.H., Roth, R., and Dement, W.C., 
Principles and Practices of Sleep Medicine, 4th edition. Saunders, New York. 
 
Siegel, J.M., 2005. Clues to the function of mammalian sleep. Nature, 437:1264-1271. 
 
Siegel, J.M., 2008. Do all animals sleep? Trends in Neuroscience, 31(4):208-213. 
 
Siegel, J.M., 2009. Sleep viewed as a state of adaptive inactivity. Nature Reviews 
Neuroscience, 10:747-753. 
 
Snyder, F., Bugbee, N., Douthitt, T., 1972.Telemetric studies of 24-hour sleep-waking 
patterns in some primitive mammals. Psychophysiology, 9:122. 
 
 314
Snyder, F., 1974. Sleep–waking patterns of Hyracoidea. Sleep Research, 3:87. 
 
Spiess, H. G., Whitmayer, D.I., Sawyer, C.H., 1970. Patterns of spontaneous and 
induced paradoxical sleep in intact and hypophysectomized rabbits. Brain Research,  
18:155-164. 
 
Steriade, M, 2005. Brain Electrical Activity and Sensory Processing during Waking 
and Sleeping States. In: Kryger, M.H., Roth, R., and Dement, W.C., Principles and 
Practices of Sleep Medicine, 4th edition. Saunders, New York. 
 
Sunquist, M., and Montgomery, G., 1973. Activity patterns and rates of movement of 
two-toed and three-toed sloths, Choloepus hoffmani and Bradypus infuscatus. Journal 
of Mammalogy., 54: 946-954. 
 
Tabuce, R., Asher, R.J., Lehmann, T., 2008. Afrotherian mammals: a review of 
current data. Mammalia, 72:2-14. 
 
Tobler, I., 1992. Behavioral sleep in the Asian elephant in captivity. Sleep, 15:1–12. 
 
Tobler, I., 1995. Is sleep fundamentally different between mammalian species? 
Behavioral Brain Research, 69:35-41. 
 
Tobler, I., 2005. Phylogeny of Sleep Regulation. In: Kryger, M.H., Roth, R., and 
Dement, W.C., Principles and Practices of Sleep Medicine, 4th edition. Saunders, New 
York. 
 315
Valatx, J. L. and Bugat, R., 1974. Facteurs génétiques dans le déterminisme du cycle 
veille-sommeil chez la souris. Brain Research, 69:315-330. 
 
Van Twyver, H., and Allison, T., 1974. Sleep in the armadillo Dasypus novemcinctus 
at moderate and low ambient temperatures. Brain Behavior and Evolution, 9:107-120.  
 
Wilson, M.L., Bashaw, M.J., Fountain, K., Kieschnick, S., Maple, T.L., 2006. 
Nocturnal Behavior in a Group of Female African Elephants. Zoo Biology, 25:173–
186. 
 
Wyatt, J.R., and Eltringham, S.K., 1974. The daily activity of the elephant in the 
Rwenzori National Park, Uganda. East African Wildlife Journal, 12:213-289.  
 
Zeplin, H., Siegel, J.M., Tobler, I., 2005. Mammalian Sleep. In: Kryger, M.H., Roth, 
R., and Dement, W.C., Principles and Practices of Sleep Medicine, 4th edition. 
Saunders, New York. 
 
8.5. Chapter 5: Social contrasted with solitary sleep patters in the rock hyrax, 
Procavia capensis 
 
Bonnet, C., Leger, L., Baubet, V., Debilly, G., Cespuglio, R., 1997. Influence of a 1h 
immobilization stress on sleep states ad corticotropin-like intermediate lobe peptide 
(CLIP or ACTH18-39, Ph-ACTH18-39) brain contents in the rat. Brain Research, 751:54-
63. 
 
 316
Borbély. A.A., and Achermann, P., 1999. Sleep homeostasis and models of sleep 
regulation. Journal of Biological Rhythms, 14(6):557-68. 
 
Borbély, A.A., and Neuhaus, H.U., 1980. Sleep-deprivation: effects on sleep and EEG 
in the rat. Journal of Comparative Physiology, 133:71–87. 
 
Bouyer, J.J., Deminiere, J.M., Mayo, W., Le Moal, M., 1997. Inter-individual 
differences in the effects of acute stress on the sleep-wakefulness cycle in the rat. 
Neuroscience Letters, 225:193-6. 
 
Capellini, I., Nunn, C.L., McNamara, P., Preston, B.T., Barton, R.A., 2008. Energetic 
constraints, not predation, influence the evolution of sleep patterning in mammals. 
Functional Ecology, 22:847-853. 
 
Capellini, I., Barton, R.A., McNamara, P., Preston, B.T., Nunn, C.L., 2008.  
Phylogenetic analysis of the ecology and evolution of mammalian sleep. Evolution, 
62:1764–1776. 
 
Caro, T., 2005. Antipredator defenses in birds and mammals. University of Chicago 
Press, Chicago. 
 
Cirelli, C., and Tononi, G. (2008). Is sleep essential? PLOS Biology, 6(8), e216. 
 
Daan, S., Beersma, D.G.M., Borbély, A.A., 1984.  Timing of sleep: recovery process 
gated by a circadian pacemaker. American Journal of Physiology, 246:R161–R187. 
 317
Ganguly-Fitzgerald, I., Donlea, J., Shaw, P. J., 2006. Waking experience affects sleep 
need in Drosophila. Science, 313:1775-1781. 
 
Horne, J., 2009. Neuroscience and Biobehavioral Reviews REM sleep, energy 
balance and ‘optimal foraging ’. Biobehavioral Reviews, 33:466-474. 
 
Lesku, J. A., Bark, R. J., Martinez-Gonzalez, D., Rattenborg, N. C., Amlaner, C. J., 
Lima, S. L., Planck, M., and Group, F., 2008. Predator-induced plasticity in sleep 
architecture in wild-caught Norway rats (Rattus norvegicus). Behavioural Brain 
Research, 189:298-305. 
 
Lima, S.L., Rattenborg, N.C., Lesku, J.A., Amlaner, C.J., 2005. Sleeping under the 
risk of predation. Animal Behaviour, 70:723–736. 
 
McNamara, P., Capellini, I., Harris, E., Nunn3, C.L., Barton, R.A., Preston, B., 2008. 
The Phylogeny of Sleep Database: A New Resource for Sleep Scientists. The Open 
Sleep Journal, 1:11-14. 
 
Meerlo, P., Pragt, B.J., Daan, S., 1997. Social stress induces high intensity sleep in 
rats. Neuroscience Letters, 225:41–44. 
 
Meerlo, P., de Bruin, E., Strijkstra, A.M., Daan, S., 2001. Social conflict increases 
EEG slow-wave activity during subsequent sleep. Physiology and Behavior, 73:331-
335.  
 318
Meerlo, P., and Turek, F.W., 2001. Effects of social stimuli on sleep in mice: non-
rapid-eye-movement (NREM) sleep is promoted by aggressive interaction but not by 
sexual interaction. Brain Research, 907:84–92. 
 
Noser, R., Gygax, L., Tobler, I., 2003. Sleep and social status in captive gelada 
baboons (Theropithecus gelada). Behavioural Brain Research, 147:9–15. 
 
Nicolau, M.C., Akaârir, M., Gamundí, A., Gonzáles, J., Rial, R.V., 2000. Why we 
sleep: the evolutionary pathway to the mammalian sleep, Progress in Neurobiology, 
62: 379-406. 
 
Nunn, C.L., McNamara, P., Capellini, I., Preston, B.T., Barton, R.A., 2010. Primate 
sleep in phylogenetic perspective. In: McNamara, P., Barton, R.A., Nunn, C.L., 
Evolution of sleep: Phylogenetic and functional perspectives, Cambridge, UK; New 
York: Cambridge University Press. 
 
Rampin, C., Cespuglio, R., Chastrette, N., Jouvet, M., 1991. Immobilization stress 
induces a paradoxical sleep rebound in the rat. Neuroscience Letters, 126:113-8.  
 
Siegel, J.M., 2008. Do all animals sleep? Trends in Neuroscience, 31(4):208-213. 
 
Taylor, C.R., and Sale, J.B., 1969. Temperature regulation in the hyrax. Comparative 
Biochemistry and Physiology, 31:903-907. 
 
 319
Tobler, I., and Borbély, A.A., 1986. Sleep EEG in the rat as a function of prior 
waking. Electroencephalographic clinical Neurophysiology, 64:74–76.  
 
Zepelin, H., 1989. Mammalian sleep. In: Kryger, M.H., Roth, T., Dement, W.C., 
Principles and Practices of Sleep Medicine, pp. 30–49. Saunders, 
Philadelphia. 
 
8.6. Chapter 6: The interrelations of the distribution of sleep associated nuclei 
and terminal networks in the brain of the rock hyrax, Procavia capensis 
 
Airaksinen, M.S., and Panula, P., 1988. The Histaminergic System in the Guinea Pig 
Central Nervous System: An Immunocytochemical Mapping Study Using an 
Antiserum Against Histamine. The Journal of Comparative Neurology, 273:163-186. 
 
Arai, R., Winsky, L., Arai, M., Jacobowitz, D.M., 1991. Immunohistochemical 
localization of calretinin in the rat hindbrain. Journal of  Comparative Neurology, 
310:21-44.  
 
Ascoli, G.A., Alonso-Nanclares, L., Anderson, S.A., Barrionuevo, G., Benavides-
Piccione, R., Burkhalter, A., Buzsáki, G., Cauli, B., DeFelipe, J., Fairén, A., 
Feldmeyer, D., Fishell, G., Fregnac, Y., Freund, T.F., Gardner, D., Gardner, E.P., 
Goldberg, J.H., Helmstaedter, M., Hestrin, S., Karube, F., Kisvárday, Z.F., Lambolez, 
B., Lewis, D.A., Marin, O., Markram, H., Muñoz, A., Packer, A., Petersen, C.C.H., 
Rockland, K.S., Rossier, J., Rudy, B., Somogyi, P., Staiger, J.F., Tamas, G., 
Thomson, A.M., Toledo-Rodriguez, M., Wang, Y., West, D.C., Yuste, R., 2008. 
 320
Petilla terminology: nomenclature of features of GABAergic interneurons of the 
cerebral cortex. Nature Reviews Neuroscience, 9:557-568. 
 
Baimbridge, K.G., Celio, M.R., Rogers, J.H., 1992. Calcium-binding proteins in the 
nervous system. Trends in Neuroscience, 15:303-308.  
 
Bayraktar, T., Staiger, J.F., Acsady, L., Cozzari, C., Freund, T.F., Zilles, K., 1997. 
Co-localization of vasoactive intestinal polypeptide, gamma-amino-butyric acid and 
choline acetyltransferase in the neocortical interneurons of the adult rat. Brain Res. 
757, 209–217.  
 
Braun, K., 1990. Calcium-binding proteins in avian and mammalian central nervous 
system: localization development and possible functions. Progress in Histochemistry 
and Cytochemistry, 21:1-64.  
 
Cauli, B.,  Audinat,E., Lambolez, B., Angulo, M.C., Ropert, N., Tsuzuki, K., Hestrin, 
S., Rossier, J., 1997. Molecular and Physiological Diversity of Cortical Nonpyramidal 
Cells. The Journal of Neuroscience, 17: 3894-3906. 
 
Celio, M.R., 1990. Calbindin D-28k and parvalbumin in the rat nervous system. 
Neuroscience 35, 375-475.  
 
Charara, A., and Parent, A., 1998. Chemoarchitecture of the primate dorsal raphe 
nucleus. Journal of Chemical Neuroanatomy, 15:111-127. 
 
 321
Gallyas, F., 1979. Silver staining of myelin by means of physical development. 
Neurological Research, 1:203–209. 
 
Gerashchenko, D., Shiromani, P.J., 2004. Different neuronal phenotypes in the lateral 
hypothalamus and their role in sleep and wakefulness. Molecular Neurobiology, 
29:41–59. 
 
Gravett, N., Bhagwandin, A., Fuxe, K., Manger, P.R., 2009. Nuclear organization and 
morphology of cholinergic, putative catecholaminergic and serotonergic neurons in 
the brain of the rock hyrax, Procavia capensis. Journal of Chemical Neuroanatomy, 
38:57-74. 
 
Gravett, N., Bhagwandin, A., Fuxe, K., Manger, P.R., 2011. Dustribution of orexin-A 
immunoreactive neurons and their terminal networks in the brain of the rock hyrax, 
Procavia capensis.  Journal of Chemical Neuroanatomy, Article in Press 
doi.10.1016/j.jchemneu.2010.11.005. 
 
Inagaki, N., Yamatodani, A., Ando-Yamamoto, M., Tohyama, M., Watanabe, T., 
Wada, H., 1988. Organization of Histaminergic Fibers in the Rat Brain. The Journal 
of Comparative Neurology, 273:283-300. 
 
John, J., Wu, M.F., Boehmer, L.N., Siegel, J.M., 2004. Cataplexy-active neurons in 
the hypothalamus: implications for the role of histamine in sleep and waking 
behavior. Neuron, 42:619–634. 
 322
 
Kiyashchenko, L.I., Mileykovskiy, B.Y., Maidment, N., Lam, H.A., Wu, M.F., John, 
J.,Peever, J., Siegel, J.M., 2002. Release of hypocretin (orexin) during waking and 
sleep states. Journal of Neuroscience, 22:5282–5286. 
 
Krzywkowski, P., Jacobowitz, D.M., Lamour, Y., 1995. Calretinin-containing 
pathways in the rat forebrain. Brain Res. 705, 273-294. 
 
Lai, Y.Y., Siegel, J.M., 1990. Muscle tone suppression and stepping produced by 
stimulation of midbrain and rostral pontine reticular formation. Journal of 
Neuroscience, 10:2727–2738. 
 
Lai, Y.Y., Kodama, T., Siegel, J.M., 2001. Changes in monoamine release in the 
ventral horn and hypoglossal nucleus linked to pontine inhibition of muscle tone: an 
in vivo microdialysis study. Journal of Neuroscience, 21:7384–7391. 
 
Leger, L., Charnay, Y., Hof, P.R., Bouras, C., Cespuglio, R., 2001. Anatomical 
Distribution of Serotonin-Containing Neurons and Axons in the Central Nervous 
System of the Cat. The Journal of Comparative Neurology, 433:157-182. 
 
Lyamin, O.I., Manger, P.R., Ridgway, S.H., Mukhametov, L.M., Siegel, J.M., 2008. 
Cetacean sleep: an unusual form of mammalian sleep. Neuroscience and 
Biobehavioral Reviews, 32(8):1451-84.  
 
 323
Mintz, E.M., Van den Pol, A.N., Casano, A.A., Albers, H.E., 2001. Distribution of 
hypocretin-(orexin) immunoreactivity in the central nervous system of Syrian 
hamsters (Mesocricetus auratus). Journal of Chemical Neuroanatomy, 21:225–238. 
 
Nitz, D., Siegel, J.M., 1997a. GABA release in the locus coeruleus as a function of 
sleep/wake state. Neuroscience, 78:795–801. 
 
Nitz, D., Siegel, J.M., 1997b. GABA release in the dorsal raphe nucleus: role in the 
control of REM sleep. American Journal of Physiology, 273:R451–R455. 
 
Nixon, J.P., Smale, L., 2007. A comparative analysis of the distribution of 
immunoreactive orexin-A and B in the brains of nocturnal and diurnal rodents. 
Behavioral and Brain Functions, 13:28. 
 
Novak, C.M., Albers, H.E., 2002. Localization of hypocretin-like immunoreactivity 
in the brain of the diurnal rodent, Arvicanthis niloticus. Journal of Chemical 
Neuroanatomy, 23:49-58. 
 
Parent, A, Fortin, M., Côté, P.-Y., Cicchetti, F., 1996. Calcium-binding proteins in 
primate basal ganglia. Neuroscience Research, 25:309-334. 
 
Peyron, C., Tighe, D.K., van den Pol, A.N., De Lecea, L., Heller, H.C., Sutcliffe, J.G., 
Kilduff, T.S., 1998. Neurons containing hypocretin (orexin) project to multiple 
neuronal systems. Jornal of Neuroscience, 18:9996–10015. 
 324
 
Rogers, J.H., 1992. Immunohistochemical markers in the rat brain: colocalization of 
calretinin and calbindin-D28k with tyrosine hydroxylase. Brain Research, 587:203-
210.  
 
Sakai, K., Koyama, Y., 1996. Are there cholinergic and non-cholinergic paradoxical 
sleep-on neurons in the pons. Neuroreport, 7:2449–2453. 
 
Saper, C.B., Chou, T.C., Scammell, T.E., 2001. The sleep switch: hypothalamic 
control of sleep and wakefulness. Trends in Neuroscience, 24:726–731. 
 
Siegel, J.M., 2000. Brainstem mechanisms generating REM sleep. In: Kryger, M.H., 
Roth, T., Dement, W.C. (Eds.), Principles and Practices of Sleep Mechanisms. WB 
Saunders, Philadelphia, PA, pp. 112–133. 
 
Siegel, J.M., 2004. The neurotransmitters of sleep. Journal of Clinical Psychiatry, 65 
(suppl. 16):4–7. 
 
Siegel, J.M., Nienhuis, R., Fahringer, H.M., Paul, R., Shiromani, P., Dement, W.C., 
Mignot, E., Chiu, C., 1991. Neuronal activity in narcolepsy: identification of 
cataplexy related cells in the medial medulla. Science, 252:1315–1318. 
 
 325
Steriade, M., Pare, D., Datta, S., Oakson, G., Curro Dossi, R., 1990. Different cellular 
types in mesopontine cholinergic nuclei related to ponto-geniculo-occipital waves. 
Journal of Neuroscience, 10:2560–2579. 
Szymusiak, R., 1995. Magnocellular nuclei of the basal forebrain: substrates of sleep 
and arousal regulation. Sleep, 18:478–500. 
 
Szymusiak, R., Steininger, T., Alam, N., McGinty, D., 2001. Preoptic area 
sleepregulating mechanisms. Archives Italiennes de Biologie, 139:77–92. 
Törk, I., 1990. Anatomy of the serotonergic system. Annals of the New York 
Academy of Sciences, 600:9–35. 
 
Wu, M.F., John, J., Boehmer, L.B., Nguyen, G.B., Siegel, J.M., 2004. Activity of 
dorsal raphe cells across the sleep-waking cycle and during cataplexy in narcoleptic 
dogs. Journal of Physiology, 554:202–215. 
 
8.7. Chapter 7: Concluding Remarks 
 
Bergmann, B.M., Winter, J.B., Rosenberg, R.S., Rechtschaffen, A., 1987. NREM 
Sleep with Low-Voltage EEG n the Rat. Sleep, 10(1):1-11. 
 
Bhagwandin, A., Fuxe, K., Bennett, N.C., Manger, P.R., 2008. Nuclear organization 
and morphology of cholinergic, putative catecholaminergic and serotonergic neurons 
 326
in the brains of two species of African mole-rat. Journal of Chemical Neuroanatomy, 
35:371-387. 
 
Dwarika, S., Maseko, B.C., Ihunwo, A.O., Fuxe, K., Manger, P.R., 2008. Distribution 
and morphology of putative catecholaminergic and serotonergic neurons in the brain 
of the greater canerat, Thryonomys swinderianus. Journal of Chemical Neuroanatomy, 
35:108-122. 
 
Kruger, J-L., Dell, L-A., Pettigrew, J.D., Manger, P.R., 2010. Cellular location and 
major terminal networks of the orexinergic system in the brains of five 
microchiropteran species. Journal of Chemical Neuroanatomy, 40:256-262. 
 
Limacher, A.M., Bhagwandin, A., Fuxe, K., Manger, P.R., 2008. Nuclear 
organization and morphology of cholinergic, putative catecholaminergic and 
serotonergic neurons in the brain of the Cape porcupine (Hystrix africaeaustralis): 
Increased brain size does not lead to increased organizational complexity. Journal of 
Chemical Neuroanatomy, 36:33-52. 
 
Manger, P.R., Fahringer, H., Pettigrew, J.D., Siegel, J.M., 2002a. Distribution and 
morphology of cholinergic neurons in the brain of the monotremes as revealed by 
ChAT immunohistochemistry. Brain Behavior and Evolution, 60:275-297. 
Maseko, B.C., Manger, P.R., 2007. Distribution and morphology of cholinergic, 
catecholaminergic and serotonergic neurons in the brain of Schreiber’s long-fingered 
bat, Miniopterus schreibersii. Journal of Chemical Neuroanatomy, 34:80-94. 
 
 327
Maseko, B.C., Manger, P.R., 2007. Distribution and morphology of cholinergic, 
catecholaminergic and serotonergic neurons in the brain of Schreiber’s long-fingered 
bat, Miniopterus schreibersii. Journal of Chemical Neuroanatomy, 34:80-94. 
 
McGranaghan, P.A., Piggins, H.D., 2001. Orexin-A-like immunoreactivity in the 
hypothalamus and thalamus of the Syrian hamster (Mesocricetus auratus) and 
Siberian hamster (Phodopus sungorus), with special reference to circadian structures. 
Brain Research, 904:234-244. 
 
Mintz, E.M., Van den Pol, A.N., Casano, A.A., Albers, H.E., 2001. Distribution of 
hypocretin-(orexin) immunoreactivity in the central nervous system of Syrian 
hamsters (Mesocricetus auratus). Journal of Chemical Neuroanatomy, 21:225–238. 
 
Moon, D.J., Maseko, B.C., Ihunwo, A., Fuxe, K., Manger, P.R., 2007. Distribution 
and morphology of catecholaminergic and serotonergic neurons in the brain of the 
highveld gerbil, Tatera brantsii. Journal of Chemical Neuroanatomy, 34:134-144. 
 
Nixon, J.P., Smale, L., 2007. A comparative analysis of the distribution of 
immunoreactive orexin-A and B in the brains of nocturnal and diurnal rodents. 
Behavioral and Brain Functions, 13:28. 
 
Novak, C.M., Albers, H.E., 2002. Localization of hypocretin-like immunoreactivity 
in the brain of the diurnal rodent, Arvicanthis niloticus. Journal of Chemical 
Neuroanatomy, 23:49-58. 
 
 328
Olds, N., and Shoshani, J., 1982. Procavia capensis. Mammalian Species, 171:1-7. 
 
Skinner, J.D., Chimimba, C.T., 2005. The Mammals of the Southern African 
Subregion, 3rd Edition. Cambridge University Press, Cape Town. 
Snyder, F., 1974. Sleep–waking patterns of Hyracoidea. Sleep Research, 3:87. 
 
Tabuce, R., Asher, R.J., Lehmann, T., 2008. Afrotherian mammals: a review of 
current data. Mammalia, 72:2-14. 
 
Törk, I., 1990. Anatomy of the serotonergic system. Annals of the New York 
Academy of Sciences, 600:9–35. 
 
Wagner, D., Salin-Pascual, R., Greco, M.A., Shiromani, P.J., 2000. Distribution of 
hypocretin-containing neurons in the lateral hypothalamus and c-fos-immunoreactive 
neurons in the VLPO. Sleep Research, 3:35-42. 
 
Woolf, N.J., 1991. Cholinergic systems in mammalian brain and spinal cord. Progress 
in Neurobiology, 37:475–524. 
 
Journal of Chemical Neuroanatomy 38 (2009) 57–74
Contents lists available at ScienceDirect
Journal of Chemical NeuroanatomyNuclear organization and morphology of cholinergic, putative catecholaminergic
and serotonergic neurons in the brain of the rock hyrax, Procavia capensis
Nadine Gravett a, Adhil Bhagwandin a, Kjell Fuxe b, Paul R. Manger a,*
a School of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown, 2193 Johannesburg, South Africa
bDepartment of Neuroscience, Karolinska Institutet, Retzius va¨g 8, S-171 77 Stockholm, Sweden
A R T I C L E I N F O
Article history:
Received 17 January 2009
Received in revised form 18 February 2009
Accepted 27 February 2009
Available online 14 March 2009
Keywords:
Afrotheria
Immunohistochemistry
Hyrax
Evolution
Mammalia
A B S T R A C T
The nuclear subdivisions of the cholinergic, putative catecholaminergic and serotonergic systemswithin
the brain of the rock hyrax (Procavia capensis) were identified following immunohistochemistry for
acetylcholinesterase, tyrosine hydroxylase and serotonin. The aim of the present study was to
investigate possible differences in the complement of nuclear subdivisions of these systems by
comparing those of the rock hyrax to published studies of other mammals. The rock hyrax belongs to the
order Hyracoidea and forms part of the Afroplacentaliamammalian cohort. For themost part, the nuclear
organization of these three systems closely resembled that described for many other mammalian
species. The nuclear organization of the serotonergic system was identical to that seen in all eutherian
mammals. The nuclear organization of the putative catecholaminergic system was very similar to that
seen in rodents except for the lack of a C3 nucleus and the compact division of the locus coeruleus (A6c).
In addition, the diffuse locus coeruleus (A6d) appeared to contain very few tyrosine hydroxylase
immunoreactive (TH+) neurons. The cholinergic system showed many features in common with that
seen in both rodents and primates; however, there were three differences of note: (1) cholinergic
neurons were observed in the anterior nuclei of the dorsal thalamus; (2) cholinergic parvocellular nerve
cells, probably representing interneurons, forming subdivisions of the laterodorsal and pedunculo-
pontine tegmental nuclei were observed at the midbrain/pons interface; and (3) a large number of
cholinergic nerve cells in the periventricular grey of the medulla oblongata were observed. Thus, while
there aremany similarities to othermammalian species, the nuclear organization of these systems in the
rock hyrax shows specific differences to what has been observed previously in other mammals. These
differences are discussed in both a functional and phylogenetic perspective.
 2009 Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: +27 11 717 2497; fax: +27 11 717 2422.
E-mail address: Paul.Manger@wits.ac.za (P.R. Manger).
Abbreviations: III, oculomotor nucleus; IV, trochlear nucleus; Vmot, motor division of trigeminal nucleus; VI, abducens nucleus; VIId, facial nerve nucleus, dorsal division; VIIv,
facial nerve nucleus, ventral division; X, dorsal motor vagus nucleus; XII, hypoglossal nucleus; 3V, third ventricle; 4V, fourth ventricle; 7n, facial nerve; A1, caudal ventrolateral
medullary tegmental nucleus; A2, caudal dorsomedial medullary nucleus; A4, dorsal medial division of locus coeruleus; A5, fifth arcuate nucleus; A6c, compact portion of locus
coeruleus; A6d, diffuse portion of locus coeruleus; A7d, nucleus subcoeruleus, diffuse portion; A7sc, nucleus subcoeruleus, compact portion; A8, retrorubral nucleus; A9l,
substantia nigra, lateral; A9m, substantia nigra, medial; A9pc, substantia nigra, pars compacta; A9v, substantia nigra, ventral or pars reticulata; A10, ventral tegmental area; A10c,
ventral tegmental area, central; A10d, ventral tegmental area, dorsal; A10dc, ventral tegmental area, dorsal caudal; A11, caudal diencephalic group; A12, tuberal cell group; A13,
zona incerta; A14, rostral periventricular nucleus; A15d, anterior hypothalamic group, dorsal division; A15v, anterior hypothalamic group, ventral division; A16,
catecholaminergic neurons of the olfactorybulb; ac, anterior commissure;AD, anterodorsal nucleus of the dorsal thalamus; Amyg, amygdala; AP, areapostrema;AV, anteroventral
nucleus of the dorsal thalamus; B9, supralemniscal serotonergic nucleus; C, caudate nucleus; C1, rostral ventrolateral medullary tegmental group; C2, rostral dorsomedial
medullary nucleus; ca, cerebral aqueduct; Cb, cerebellum; cc, corpus callosum; Cl, claustrum; CLi, caudal linear nucleus; CN, cochlear nucleus; C/P, caudate and putamen nuclei;
CP, cerebral peduncle; CVL, caudal ventrolateral serotonergic group; DCN, deep cerebellar nuclei; Diag.B, diagonal band of Broca; DR, dorsal raphe; DRc, dorsal raphe nucleus,
caudal division; DRd, dorsal raphe nucleus, dorsal division; DRif, dorsal raphe nucleus, interfascicular division; DRl, dorsal raphe nucleus, lateral division; DRp, dorsal raphe
nucleus, peripheral division; DRv, dorsal raphe nucleus, ventral division; DT, dorsal thalamus; EW, Edinger–Westphal nucleus; f, fornix; GC, periaqueductal grey matter; GLD,
dorsal lateral geniculate nucleus; GP, globus pallidus; Hbm, medial habenular nucleus; Hip, hippocampus; Hyp, hypothalamus; Hyp.d, dorsal hypothalamic cholinergic nucleus;
Hyp.l, lateral hypothalamic cholinergic nucleus; Hyp.v, ventral hypothalamic cholinergic nucleus; IC, inferior colliculus; ic, internal capsule; icp, inferior cerebellar peduncle; io,
inferior olivar nuclei; IP, interpeduncular nucleus; LDTmc, magnocellular division of the laterodorsal tegmental nucleus; LDTpc, parvocellular division of the laterodorsal
tegmental nucleus; LRT, lateral reticular nucleus; LV, lateral ventricle; mcp, middle cerebellar peduncle; MnR,median raphe nucleus; N.Acc, nucleus accumbens; N.Amb, nucleus
ambiguus; N.Bas, nucleus basalis; NEO, neocortex; OB, olfactory bulb; OC, optic chiasm; OT, optic tract; P, putamen; pVII, preganglionic motor neurons of the superior salivatory
nucleus or facial nerve; pIX, preganglionicmotor neurons of the inferior salivatory nucleus; PBg, parabigeminal nucleus; PIR, piriformcortex; PPTmc,magnocellular division of the
pedunculopontine nucleus; PPTpc, parvocellular division of the pedunculopontine nucleus; Pta, pretectal area; py, pyramidal tract; pyx, decussation of the pyramidal tract; R,
thalamic reticular nucleus; Rmc, red nucleus, magnocellular division; RMg, raphe magnus nucleus; ROb, raphe obscurus nucleus; RPa, raphe pallidus nucleus; RVL, rostral
ventrolateral serotonergic group; S, septum; SC, superior colliculus; scp, superior cerebellar peduncle; Sep.M,medial septal nucleus; TOL, olfactory tubercle; TOL/Is.Call., olfactory
tubercle/island of Calleja; vh, ventral horn of spinal cord; VPO, ventral pontine nucleus; xscp, decussation of the superior cerebellar peduncle; ZI, zona incerta.
journal homepage: www.e lsev ier .com/ locate / jchemneu
0891-0618/$ – see front matter  2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jchemneu.2009.02.005
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74581. Introduction
The rock hyrax, Procavia capensis, is a member of the
Procaviidae family, which is the only living family within the
Hyracoidea order (Klein and Cruz-Uribe, 1996). The family
consists of three genera: Procavia (the rock hyrax, one species),
Heterohyrax (two species) and Dendrohyrax (three species)
(Skinner and Chimimba, 2005). The rock hyrax is a small, agile,
diurnal, social, herbivorousmammal, typicallyweighing between
2.5 and 4.6 kg, and lives in rocky outcrops or ‘‘koppies’’ in most
parts of Africa (Skinner and Chimimba, 2005). Sexual maturity is
reached between 28–29 months in males and 16–17 months in
females, with seasonal mating being triggered by the photoperiod
(Skinner and Chimimba, 2005). According to the fossil record,
hyraxes first appeared approximately 40 million years ago and
they are grouped phylogenetically with the African elephant
(Proboscidea), and manatee and dugong (Sirenia). These three
orders, together with the Chrysochloridea (golden moles),
Macroscelidea (elephant shrews), Tubulidentata (aardvarks)
and the Tenrecidea (tenrecs) form the cohort Afroplacentalia
(Arnason et al., 2008).
In the present study the cholinergic, putative catecholami-
nergic and serotonergic systems within the brain of the rock
hyrax were examined and described using immunohistochem-
ical techniques. It is well known that these systems project to
most parts of the brain and that they are associated with several
functions (Woolf, 1991; Smeets and Gonza´lez, 2000; To¨rk, 1990;
Jacobs and Azmitia, 1992); for example, cognition (e.g. Bartus
et al., 1982; Previc, 1999), the sleep-wake cycle (e.g. Siegel,
2006), reproduction (e.g. Tillet, 1995), and sensory-motor (e.g.
Pompeiano, 2001; Fuxe et al., 2007a) functions to name but a
few. The cholinergic system has an extensive distribution
throughout the brain (Woolf, 1991; Reiner and Fibiger, 1995;
Manger et al., 2002a; Maseko et al., 2007) while the catecho-
laminergic and serotonergic neuronal systems are mainly
concentrated within the brainstem (Dahlstro¨m and Fuxe,
1964; Ande´n et al., 1964; Fuxe et al., 1969, 1970, 2006,
2007a; Diksic and Young, 2001; Manger et al., 2002b,c; Maseko
et al., 2007). The nuclear organization of these systems has been
studied in several mammalian species (e.g. Maseko et al., 2007;
Bhagwandin et al., 2008; Limacher et al., 2008); however no
studies of these systems have been done in any member of the
Afroplacentalia cohort. Thus, the aim of this study is to
determine the nuclear organization of these systems in the
brain of the rock hyrax and extend our basis for understanding
the evolutionary processes associated with the nuclear organi-
zation of these systems.
These systems have exhibited some evolutionary trends and
even though these systems are quite similar across species for
the most part, differences due occur. For example, the
catecholaminergic C3 nucleus has only been reported to be
present in the rodents (e.g. Smeets and Gonza´lez, 2000; Manger
et al., 2002b; Maseko et al., 2007; Moon et al., 2007; Badlangana
et al., 2007; Dwarika et al., 2008; Bhagwandin et al., 2008;
Limacher et al., 2008). Serotonergic neurons in the mammalian
hypothalamus have only been reported for monotremes
(Manger et al., 2002c). These examples and other evidence
have led Manger (2005) to propose that members of the same
order will show the same complement of nuclei, but this
complement may differ between orders. Some predictions of the
current study include: (1) the rock hyrax will have many nuclei
in common with other mammals; (2) the rock hyrax may have
some nuclei unique to this species; (3) the rock hyrax may
exhibit nuclei found only in members of the Afroplacentalia; and
(4) the rock hyrax may be missing some nuclei commonly found
in other mammalian species.2. Materials and methods
A total of six adult female rock hyraxes, P. capensis, were used in the present
study. Permits from the Limpopo and Gauteng Provincial Governments were
obtained for the capture and transport of the animals from the wild. All animals
were treated and used according to the guidelines of the University of the
Witwatersrand Animal Ethics Committee, which parallel those of the NIH for the
care and use of animals in scientific experimentation. Each animal was weighed,
deeply anaesthetized and subsequently euthanized with weight appropriate doses
of sodium pentobarbital (200 mg sodium pentobarbital/kg, i.p.). Upon cessation of
respiration the animals were perfused intracardially with 0.9% saline followed by
4% paraformaldehyde in 0.1 M phosphate buffer (PB) (approximately 1 l/kg of each
solution), both solutions having a temperature of 4 8C. The brains were then
carefully removed from the skulls and post-fixed overnight in 4% paraformaldehyde
in 0.1 M PB followed by equilibration in 30% sucrose in 0.1 M PB. The brains were
then frozen and with the aid of a freezing microtome sectioned at 50mm in either
coronal (n = 4) or sagittal (n = 2) planes. A one in five series of stains was made for
nissl, myelin, choline acetyltransferase (ChAT) (identification of the cholinergic
system), tyrosine hydroxylase (TH) (identification of the dopaminergic and
noradrenergic systems), and serotonin (5HT) (identification of the serotonergic
system). Sections kept for the Nissl series were mounted on 0.5% gelatine-coated
glass slides, cleared in a solution of 1:1 chloroform and absolute alcohol, then
stained with 1% cresyl violet to reveal cell bodies. Myelin sections were stored in 5%
formalin for a period of two weeks and were then mounted on 1.5% gelatine-coated
glass slides and subsequently stained with silver solution to reveal myelin sheaths
(Gallyas, 1979).
For immunohistochemical staining each section was treated with endogenous
peroxidase inhibitor (49.2% methanol:49.2% 0.1 M PB:1.6% of 30% H2O2) for 30 min
and subsequently subjected to three 10 min 0.1 M PB rinses. The sections were then
preincubated in a solution (blocking buffer) consisting of 3% normal serum (normal
rabbit serum, NRS, for the ChAT sections and normal goat serum, NGS, for the TH
and 5HT sections), 2% bovine serum albumin (BSA, Sigma) and 0.25% Triton X100
(Merck) in 0.1 M PB, at room temperature for 2 h. This was followed by three 10 min
rinses in 0.1 M PB. The sections were then placed, for 48 h at 4 8C under constant
gentle shaking, in primary antibody solution, that contained the appropriately
diluted primary antibody in blocking buffer (see above). The primary antibodies
used were anti-cholineacetyltransferase for cholinergic neurons (AB144P, Chemi-
con, raised in goat, at a dilution of 1:2000), anti-tyrosine hydroxylase for putative
catecholaminergic neurons (AB151, Chemicon, raised in rabbit, at a dilution of
1:7500), and anti-serotonin for serotonergic neurons (AB938, Chemicon, raised in
rabbit, at a dilution of 1:10000). Thiswas followed by another three 10 min rinses in
0.1 M PB, after which the sections were incubated for 2 h at room temperature in
secondary antibody solution. The secondary antibody solution contained a 1:750
dilution of biotinylated anti-rabbit IgG (BA-1000, Vector Labs) in 3% NGS (or anti-
goat IgG, BA-5000 in 3% NRS for the ChAT sections), and 2% BSA in 0.1 M PB. Once
this was completed, the sections were again subjected to another three 10 min
rinses in 0.1 M PB, followed by a 1 h incubation in AB solution (Vector Labs) and
again rinsed. This was followed by a 5 min treatment of the sections in a solution
consisting of 0.05% diaminobenzidine (DAB) in 0.1 M PB, after which, and while still
in the same solution, 3ml of 30%H2O2 per 0.5 ml of solutionwas added.With the aid
of a low power stereomicroscope the progression of the staining was visually
followed and allowed to continue until a level was reached where the background
staining could assist in reconstruction without obscuring the immunopositive
neurons. Once this level was reached the reaction was stopped by placing the
sections in 0.1 M PB, followed by a final session of three 10 min rinses in 0.1 M PB.
The immunohistochemically stained sections were mounted on 0.5% gelatine-
coated slides and left to dry overnight. The mounted sections were dehydrated by
placing it in 70% alcohol for 2 h at room temperature under gentle shaking and then
transferred through a series of graded alcohols, cleared in xylene and coverslipped
with Depex.
The sections were observed with a low power stereomicroscope, and the
architectonic borders traced according to the Nissl and myelin stained sections
using a camera lucida. The corresponding immuno-stained sections were then
matched to the drawings and the immunopositive neurons marked. The drawings
were scanned and redrawnwith the aid of the Canvas 8 program. The nomenclature
used for the cholinergic system was adopted from Woolf (1991), Manger et al.
(2002a), Maseko and Manger (2007), Maseko et al. (2007), Limacher et al. (2008)
and Bhagwandin et al. (2008); the catecholaminergic system from Dahlstro¨m and
Fuxe (1964), Ho¨kfelt et al. (1984), Smeets and Gonza´lez (2000), Manger et al.
(2002b), Maseko and Manger (2007), Maseko et al. (2007), Moon et al. (2007),
Dwarika et al. (2008), Limacher et al. (2008) and Bhagwandin et al. (2008); and for
the serotonergic system from To¨rk (1990), Bjarkam et al. (1997), Manger et al.
(2002c), Maseko and Manger (2007), Maseko et al. (2007), Moon et al. (2007),
Dwarika et al. (2008), Limacher et al. (2008) and Bhagwandin et al. (2008).Whilewe
use the standard nomenclature for the catecholaminergic system in this paper, we
realize that the neuronal groups we revealed with tyrosine hydroxylase
immunohistochemistry may not correspond directly with those nuclei that have
been described in previous studies by Dahlstro¨m and Fuxe (1964), Ho¨kfelt et al.
(1976), Meister et al. (1988), Kitahama et al. (1990, 1996), and Ruggiero et al.
(1992); however, given the striking similarity of the results of the tyrosine
hydroxylase immunohistochemistry to that seen in other mammals we feel this
terminology is appropriate. Clearly further studies in the rock hyrax with a wider
range of antibodies, such as those to phenylethanolamine-N-methyltransferase
(PNMT), dopamine-b-hydroxylase (DBH) and aromatic L-amino acid decarboxylase
(AADC) would be required to fully determine the implied homologies ascribed in
this study. We address this potential problem with the caveat of putative
catecholaminergic neurons where appropriate in the text.
3. Results
The present study was designed to reveal the nuclear
organization of the cholinergic, putative catecholaminergic, and
serotonergic systems of the rock hyrax, P. capensis through
immunohistochemical methods. A total of six adult female brains
were used for this purpose, and the individuals used for this study
had body masses ranging from 1.14 to 1.52 kg and brain masses
between 14.4 and 17.5 g (Fig. 1). The results revealed that for the
most part these systems do not differ drastically from those
observed in other mammalian species; however the rock hyrax
does show some unique additional nuclei, specifically within the
cholinergic system, that have not been noted in other species.
3.1. Cholinergic neurons
The general organization of the cholinergic system encom-
passes the striatal, basal forebrain, diencephalic, and pontome-
sencephalic groups together with the cranial nerve motor nuclei
that extend from level of the anterior horn of the lateral ventricle to
the spinomedullary junction (e.g. Woolf, 1991; Manger et al.,
2002a; Maseko et al., 2007). These groups were all present in the
brain of the rock hyrax and did not show any significant differences
to the general mammalian group level organization of this system
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 59Fig. 1. Photographs of the dorsal (top), lateral (middle) and ventral (bottom) aspects
of the rock hyrax brain. Scale bar = 1 cm.(Figs. 2 and 3). Despite this, some novel features, potentially
unique to the rock hyrax, that have not been observed in other
mammals were observed. These include the existence of choli-
nergic neurons in the anterior nuclei of the dorsal thalamus, the
existence of magnocellular and parvocellular divisions of both the
laterodorsal tegmental and pedunculopontine tegmental nuclei,
and a large cell group in the dorsomedial periventricular grey of the
medulla oblongata (Figs. 2–7).
3.1.1. Striatal cholinergic interneurons
3.1.1.1. Nucleus accumbens. Uniformly distributed choline acetyl-
transferase immunoreactive (ChAT+) nerve cell bodies were
located ventrally and slightly anterior to the dorsal striatopallidal
complex (caudate, putamen and the globus pallidus) (Figs. 2C–F
and 3). The anterior and posterior borders of this nucleus were
adjacent to the anterior horn of the lateral ventricle and the
anterior commissure, respectively. This arrangement is typical to
what is observed in all mammals (Woolf, 1991; Manger et al.,
2002a; Maseko et al., 2007). A moderate density of ChAT+ neurons
was observed throughout this nucleus, these neurons being a
mixture of bipolar and multipolar types, but the majority were of
the multipolar type (Fig. 4). The ChAT+ neurons of this nucleus
showed no specific dendritic organization.
3.1.1.2. Dorsal striatopallidal complex – caudate/putamen and globus
pallidus. The caudate/putamen nucleus was located lateral to the
lateral ventricle and its distribution extended from the level of the
anterior horn of the lateral ventricle anteriorly to the medial
habenular nuclei posteriorly (Figs. 2D–K and 3), a location typical of
all mammals (Woolf, 1991; Manger et al., 2002a; Maseko et al.,
2007). The boundary between the caudate and putamenwas clearly
defined by the internal capsule at the level of the anterior
commissure, but not anterior to this level. The location of the
globus palliduswas found tobe ventral and somewhatmedial to the
putamen and it extended from the level of the anterior commissure
to thehabenular nuclei. Amoderate densityof uniformlydistributed
ChAT+ neurons was observed within the caudate/putamen (Fig. 4).
Theglobuspallidus onlyexhibiteda small numberofChAT+neurons
and theseneuronswere foundtobe locatedmostlywhere theglobus
pallidus bordered the putamen laterally and the nucleus basalis
ventrally. A similar neuronalmorphologywasnoted for the caudate/
putamen and the globus pallidus, the ChAT+ neurons being a
mixture of bipolar and multipolar types, however a multipolar
organization predominated. No specific dendritic orientation was
observed for these neurons.
3.1.1.3. Islands of Calleja and olfactory tubercle. These nuclei
exhibited the typical mammalian organizational plan (Woolf,
1991;Manger et al., 2002a;Maseko et al., 2007). Theywere located
in the ventral most portion of the cerebral hemisphere at a level
ventral to the nucleus accumbens (Figs. 2D–F and 3). These nuclei
extended from the level of the anterior horn of the lateral ventricle
to the level of the anterior commissure. The islands of Calleja
contained a moderate density of ChAT+ neuronal clusters, and a
scattered low to medium density of ChAT+ neurons surrounding
these clusters were assigned to the olfactory tubercle. The cells in
these regions were intensely immunoreactive and a mixture of
bipolar andmultipolar neuron types with ovoid shaped somas was
observed. The neurons of the islands of Calleja as well as those of
the olfactory tubercle exhibited no specific dendritic orientation.
3.1.2. Cholinergic nuclei of the basal forebrain
3.1.2.1. Medial septal nucleus. This nucleus was identified through
the presence of ChAT+ neurons located within the septal nuclear
complex in the rostral half of the medial wall of the cerebral
hemisphere (Figs. 2G–H and 3). The location of this nucleus is
typical of what has been observed in othermammals (Woolf, 1991;
Manger et al., 2002a; Maseko et al., 2007). A moderate to high
density of ChAT+ neurons, comprising a mixture of both bipolar
and multipolar types with ovoid cell shapes was observed. It was
also noted that the cells located closer to the edge of the
hemisphere were smaller than those cells located at a small
distance from the edge. Also, the cells closer to the edge of the
hemisphere were mostly bipolar with dendrites arranged parallel
to the edge, whereas those further away were predominantly
multipolar and exhibited no specific dendritic orientation.
3.1.2.2. Diagonal band of Broca. The diagonal band of Broca was
located anterior to the hypothalamus in the ventromedial corner
of the cerebral hemisphere (Fig. 2F). The division of this nucleus
into horizontal and vertical bands appeared to be unnecessary as
the neurons forming this nucleus presented as a continuous,
uninterrupted band. A high density of ChAT+ neurons was found
throughout the extent of this nucleus (Fig. 4). The cells were
intensely immunoreactive, ovoid in shape and multipolar in
type with their dendrites orientated roughly parallel to the edge
of the cerebral hemisphere. It was also noted that the neurons of
this nucleus were larger than those found in the adjacent
olfactory tubercle and islands of Calleja, a feature that can be
of
an
in
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7460Fig. 2. Diagrammatic reconstructions of a series of coronal sections through the brain
for cholineacetyltransferase (ChAT, black circles), tyrosine hydroxylase (TH, black tri
drawn using Nissl and myelin stains and immunoreactive neurons marked on the draw
are approximately 1500 mm apart. See list for abbreviations.the rock hyrax illustrating the location of neurons immunohistochemically reactive
gles) and serotonin (open squares). The outlines of the architectonic regions were
gs. Drawing A represents the most rostral section, Z the most caudal. The drawings
readily used to demarcate this lateroventral boundary of the
diagonal band.
3.1.2.3. Nucleus basalis. ChAT+ neurons found ventral to the globus
pallidus at the level of the anterior commissure, and caudal and
dorsal to the olfactory tubercle were assigned to nucleus basalis
(Figs. 2G–H and 3). A low to moderate density of ChAT+ neurons
was observed throughout this nucleus, and these appear to be a
continuation of the ChAT+ neurons found within the globus
pallidus. These ChAT+ neurons were ovoid in shape, a mixture of
bipolar and multipolar types, and exhibited a rough dorsolateral to
ventromedial dendritic orientation.
3.1.3. Diencephalic cholinergic nuclei
3.1.3.1. Medial habenular nucleus. The medial habenular nucleus,
which forms part of the epithalamus, was located contiguous to
the third ventricle in the dorsomedial region of the diencepha-
lon (Figs. 2M and 3). The location of this nucleus was typical of
what has been observed in other mammals (Woolf, 1991;
Manger et al., 2002a; Maseko et al., 2007). A high density of
small, round, ChAT+ neurons, was observed. We could not
discern any specific dendritic orientation due to the dense
packing of neurons within this nucleus. The ChAT+ axons arising
from this nucleus clearly outlined the fasciculus retroflexus,
which was seen to end in a large swirling termination within the
interpeduncular nucleus.
3.1.3.2. Hypothalamic cholinergic nuclei. Three distinct ChAT+
neuronal groups (dorsal, ventral and lateral hypothalamic nuclei)
were identified within the hypothalamus (Figs. 2I–L and 3). The
dorsal hypothalamic nucleus was located between the wall of the
third ventricle and the fornix within the dorsomedial region of the
hypothalamus. A low density of palely stained, scattered ChAT+
neurons was observed within this nucleus. The ventral hypotha-
lamic nucleus was identified as a cluster of palely stained, widely
scattered ChAT+ neurons within the ventromedial aspect of the
hypothalamus that extended ventrolaterally past the level of the
fornix. Within the dorsolateral region of the hypothalamus, lateral
to the fornix, a low density of palely stained ChAT+ neurons was
identified as the lateral hypothalamic cholinergic nucleus. The
neuronal morphology of the ChAT+ neurons within these
hypothalamic cholinergic nuclei was similar. These neurons had
an ovoid shape, were bipolar in type and showed no specific
dendritic orientation.
3.1.3.3. Cholinergic neurons in the anterodorsal and anteroventral
dorsal thalamic nuclei. Within the anterior and dorsal region of the
dorsal thalamus the anterodorsal nucleus (AD) was seen to exhibit
a strong cholinergic neuropil staining. Around the margins of the
Con
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 61Fig. 2. ( tinued ).
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7462AD nucleus a small number of scattered ChAT+ neurons were
observed (Figs. 2K and 3). The ChAT+ neurons within this nucleus
were ovoid in shape, amixture of bipolar andmultipolar typeswith
the dendrites orientated parallel to the margins of the nucleus
(Fig. 5). Lateral to the anterodorsal nucleus, the anteroventral
nucleus (AV) exhibited a much paler neuropil ChAT+ neuropil
staining, but it was clearly distinguishable from the surrounding
unreactive tissue of the remainder of the dorsal thalamus. ChAT+
neurons showing a similar morphology to those found in the
anterdorsal nucleus were observed along the upper medial and
lateral borders of the anteroventral nucleus. The cholinergic nature
of the neuropil and occasional scattered ChAT+ neurons within
Fig. 2. (Conthese nuclei might be a unique feature of the hyrax, as these
observations have not been made in other mammals to date.
3.1.4. Pontomesencephalic nuclei
3.1.4.1. Parabigeminal nucleus. A prominent parabigeminal
nucleus was located ventral and slightly anterior to the inferior
colliculus within the lateral aspect of the midbrain tegmentum
(Figs. 2O–P and 3). A dense cluster of strongly reactive ChAT+
neurons, smaller in size to the medially located pedunculopontine
tegmental nucleus, was observed. The cell bodies were circular in
shape but due to the high density of the packing it was difficult to
tinued ).
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 63determinewhether these neuronswere bipolar ormultipolar and if
the dendrites showed any specific orientation (Fig. 6A).
3.1.4.2. Pedunculopontine tegmental nucleus (PPT) – magnocellular
and parvocellular nuclei. In all mammals studied to date, the
cholinergic neurons of the pedunculopontine tegmental nucleus
display a homogenous morphology (e.g. Woolf, 1991; Manger
et al., 2002a; Maseko et al., 2007; Limacher et al., 2008;
Bhagwandin et al., 2008). As with most mammals within the
dorsal aspect of the isthmic and pontine tegmental regions,
anterior to the trigeminalmotor nucleus, a group of ChAT+ neurons
was identified (Figs. 2O–Q and 3). This location, and the moderate
density distribution of ChAT+ neurons in this region, is again
typical. In contrast to other mammals studied, the morphology of
the cholinergic neurons within this region was not homogenous,
and we could readily identify magnocellular and parvocellular
ChAT+ nuclei within this region (Fig. 6B and D). The ChAT+ neurons
assigned to the magnocellular PPT nucleus were found in the
location, and evinced the distribution of the more typically
described PPT. These magnocellular neurons were found medial
Fig. 3. Diagrams of idealized sagittal sections through the rock hyrax brain showing
(A) cholinergic, (B) putative catecholaminergic, and (C) serotonergic nuclei or
nuclear complexes. See list for abbreviations.and ventral to the superior cerebellar peduncle, and in the anterior
portion were found to intermingle with the fibres of the superior
cerebellar peduncle. Themagnocellular ChAT+ neuronswere found
in a moderate density throughout this nucleus, were multipolar in
type, showed no specific dendritic orientation, and evinced a
variety of somal shapes. In the more caudal aspects of the PPT, in a
position dorsal and lateral to the magnocellular nucleus, a
moderately dense cluster of small ChAT+ neurons was observed,
termed parvocellular PPT nucleus. The parvocellular PPT nucleus
was sometimes clearly distinguished from the magnocellular PPT
nucleus on topological grounds, as it was found lateral and dorsal
to the superior cerebellar peduncle in the posterior aspect of the
midbrain tegmentum; however, the size of the soma was the most
reliable distinction between these two nuclei. The ovoid shaped,
parvocellular ChAT+ neurons were mostly bipolar in type, but
therewere somemultipolar neurons in this nucleus. These neurons
had dendrites that were predominantly oriented in a dorsomedial
to ventrolateral direction. There was no overlap in the distribution
of the magnocellular and parvocellular neurons at their border.
3.1.4.3. Laterodorsal tegmental nucleus (LDT) – magnocellular and
parvocellular nuclei. In all mammals studied to date, the choliner-
gic neurons of the laterodorsal tegmental nucleus display a
homogenous morphology (e.g. Woolf, 1991; Manger et al., 2002a;
Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al.,
2008). As with most mammals within the ventrolateral region of
the pontine periventricular- and periaqueductal-grey matter, a
Fig. 4. Photomicrograph montage of the basal forebrain nuclei anterior to the
anterior commissure, demonstrating cholineacetyletransferase (ChAT)
immunoreactivity. Note the strongly ChAT+ densely packed neurons of the
diagonal band of Broca (Diag.B) and the scattered strongly ChAT+ neurons in
nucleus accumbens (N.Acc), caudate (C) and putamen (P). A weakly to moderately
intense ChAT+ neuropil is found in these regions. Scale bar = 1 mm.
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7464moderate to high density of ChAT+ neurons was identified
(Figs. 2P–Q and 3). This location and distribution of ChAT+
neurons are again typical of LDT in variousmammals. In contrast to
other mammals studied, the morphology of the cholinergic
neurons within this region was not homogenous, and we could
readily identify magnocellular and parvocellular ChAT+ nuclei
within this region (Fig. 6B and C). At the ventrolateral border of the
pontine periventricular- and periaqueductal-greymatter a densely
packed cluster of larger neurons with similar morphology to those
constituting the magnocellular PPT nucleus was found. We have
termed this cluster of ChAT+ neurons the magnocellular LDT
nucleus. This magnocellular LDT nucleus appears to be typical of
that seen and described as the LDT in most mammals. A second
cluster of ChAT+ neurons was found medial and dorsal to the
magnocellular LDT nucleus, but still within the periventricular-
Fig. 5. Photomicrographs of coronal sections through the anterior nuclei of the dorsal thal
the anteroventral (AV) and anterodorsal (AD) nuclei. (A) Nissl stained section. (B) Adjace
location of ChAT+ neurons. (D) High power photomicrograph showing ChAT+ neuron and
(D) = 50mm. 3V – third ventricle.and periaqueductal-grey matter. These neurons were significantly
smaller than those in themagnocellular LDT nucleus, thus we have
termed this ChAT+ neuronal cluster the parvocellular LDT nucleus.
The morphology and size of these neurons were very similar to the
ChAT+ neurons forming the parvocellular PPT nucleus; however,
they did not display any specific dendritic orientation. These
neurons were found in a moderate to high density and had soma
that were ovoid to circular in shape. The lateral border of the
parvocellular LDT nucleus and the medial border of the magno-
cellular LDT nucleuswere not clearly defined as the ChAT+ neurons
from both nuclei showed a region of intermingling. In this sense,
while the magnocellular and parvocellular nuclei of the LDT and
PPT seem to be contiguous subdivisions of these nuclei, the distinct
topological parcellation of the two nuclei in the PPT was not as
strongly expressed in the LDT.
amus of the rock hyrax demonstrating the location of ChAT+ neurons and neuropil in
nt ChAT immunoreacted section. (C) Higher power photomicrograph of AV showing
terminals. Scale bar in (B) = 1 mmand applies to (A and B). Scale bar in (C) = 100mm,
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 653.1.5. Cholinergic cranial nerve motor nuclei
A number of large, multipolar ChAT+ neurons forming the
cranial nerve motor nuclei were identified in similar regions to
those previously documented for other mammals (Woolf, 1991;
Manger et al., 2002a; Maseko et al., 2007; Limacher et al., 2008;
Bhagwandin et al., 2008). These nuclei were: the oculomotor
nucleus (III) (Fig. 6A), the trochlear nucleus (IV), themotor division
of the trigeminal nucleus (Vmot), the abducens nucleus (VI), the
dorsal and ventral subdivisions of the facial nucleus (VIId and VIIv),
Fig. 6. Photomicrographs of ChAT immunoreacted sections through the dorsal isthm
pedunculopontine nucleus (PPTmc) at the level of the oculomotor nucleus (III) and the pa
the laterodorsal (LDTmc, LDTpc) and pedunculopontine (PPTmc, PPTpc) nuclei at a level sl
and PPTpc) locatedmedial to the LDTmc and dorsal to the PPTmc. The distinct clusters are
bar in (B) = 1 mm and applies to (A and B). Scale bar in (D) = 100mm and applies to (Cthe nucleus ambiguus, the dorsal motor vagus nucleus (X), the
hypoglossal nucleus (XII) and the ventral horn of the spinal cord
(Figs. 2N–Z and 3). In addition to these we were able to locate
ChAT+ neurons within the Edinger–Westphal nucleus and the
preganglionic motor neurons of the superior salivatory nucleus of
the facial nerve (pVII) and the preganglionic motor neurons of the
inferior salivatory nucleus (pIX). The ChAT+ neurons of the
Edinger–Westphal nucleus were moderate in number, but were
strongly immunoreactive. This nucleus primarily gives rise to the
ic/pontine region of the rock hyrax brain. (A) The magnocellular division of the
rabigeminal nucleus (PBg). (B) Themagnocellular and parvocellular divisions of both
ightly caudal to that shown in (A). Note the distinct clusters of small neurons (LDTpc
shown at higher power in (C) (LDTmc and LDTpc) and (D) (PPTmc and PPTpc). Scale
and D). ca – cerebral aqueduct.
y
ra
)
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7466preganglionic parasympathetic fibres to the eye, constricting the
pupil, and is locatedwithin the periaqueductal greymatter directly
contiguous to midline, between and anterior to the oculomotor
nuclei. The neuronal bodies were ovoid in shape, bipolar and the
dendrites were orientated dorsoventrally. Within the rostral
medullary tegmentum, dorsal to the facial nerve nuclei, the ChAT+
neurons of pVII and pIX were identified. The neurons of the pVII
nucleus were slightly smaller than those of the facial nerve nucleus
and exhibited a small number of scatteredmultipolar neurons that
demonstrated the typical motor neuron morphology, probably
Fig. 7. Photomicrographs of ChAT immunoreacted sections through periventricular gre
expanded preganglionicmotor neurons of the inferior salivatory nucleus (pIX) at a rost
dorsal to the dorsal motor vagus (X) and hypoglossal (XII) nuclei. The Scale bar in (Brepresenting the preganglionic cholinergic cell bodies of the
superior salivatory nucleus (see Mitchell and Templeton, 1981;
To´th et al., 1999). The ChAT+ neurons found medial to X probably
represented the ChAT+ preganglionic neurons of the inferior
salivatory nucleus (pIX) (see Rezek et al., 2008). At this level of the
medulla oblongata a large number of ChAT+ nerve cell bodies of
moderate density were located in the medial periventricular grey
(Fig. 7). The neurons were multipolar but smaller than the motor
neurons of X and XII, with no specific dendritic orientation. The
extensive number of cholinergic cells in this region appears to be
unusual in comparison to other mammals.
3.2. Putative catecholaminergic nuclei
The putative catecholaminergic nuclei, in the current study
being those that possess neurons that are immunoreactive to
tyrosine hydroxylase (TH+), are generally divided into several
nuclear complexes that extend from the level of the olfactory bulb
to the spinomedullary junction. These are generally defined as: the
olfactory bulb, diencephalic, midbrain, pontine, and medullary
nuclear complexes. All of the aforementioned nuclear complexes
were present in the brain of the rock hyrax and the location and
distribution of these were essentially similar to the typical
mammalian organizational plan of this particular system pre-
viously described in a range of other species (Dahlstro¨m and Fuxe,
1964; Fuxe et al., 1969, 2007b; Smeets and Gonza´lez, 2000;
Maseko et al., 2007; Limacher et al., 2008; Bhagwandin et al.,
2008). No putative catecholaminergic nuclei falling outside thesedefined regions, as is the case in some other vertebrates, were
identified in the hyrax (Smeets and Gonza´lez, 2000). The standard
nomenclature of Dahlstro¨m and Fuxe (1964) and Ho¨kfelt et al.
(1984) was implemented in the description of these nuclei. The
putative catecholaminergic nuclei identified in the rock hyrax are
essentially similar to that seen in other mammals (Figs. 2 and 3);
however, we did not identify the rodent specific rostral dorsal
midline medullary (C3) nucleus (Smeets and Gonza´lez, 2000;
Moon et al., 2007; Dwarika et al., 2008; Limacher et al., 2008;
Bhagwandin et al., 2008) and the locus coeruleus compact, built of
matter in the dorsal caudal medullary region of the rock hyrax brain. (A) The greatly
l level through this nucleus. (B) The same nucleus at amore caudal level where it lies
= 1 mm, applies to (A and B). 4V – fourth ventricle.densely packed NA cell bodies (see Fuxe et al., 1970) did not exist.
Furthermore, only a few single TH+ neurons were observed in the
adjacent periventricular grey of the pons, termed A6 diffuse (A6d)
in the current study.
3.2.1. Olfactory bulb (A16)
Within and around the glomerular layer a moderate number of
TH+ neuronswere observed (Figs. 2A and 3). This position is typical
of what has been documented in other mammals for the A16
catecholaminergic nucleus (Lichtensteiger, 1966; Lidbrink et al.,
1974; Smeets and Gonza´lez, 2000; Manger et al., 2002b; Maseko
et al., 2007; Limacher et al., 2008; Bhagwandin et al., 2008). The
cells were triangular in shape and a mixture of bipolar and
multipolar neurons were evident. These neurons also had a weak
dendritic network surrounding the glomeruli.
3.2.2. Diencephalic nuclei
In the hypothalamus six distinct putative catecholaminergic
nuclei were identified. These include: the dorsal division of the
anterior hypothalamic nucleus (A15d); the ventral division of the
anterior hypothalamic nucleus (A15v); the rostral periventricular
nucleus (A14); the zona incerta (A13); the tuberal nucleus (A12);
and the caudal diencephalic nucleus (A11) (Figs. 2G–L and 3). The
A15d nucleus was located between the third ventricle and the
fornix within the anterior portion of the hypothalamus. The shape
of the neuronal bodies was found to be ovoid, with a mixture of
bipolar and multipolar types showing no specific dendritic
orientation. The A15v nucleus was located near the floor of the
brain adjacent to the optic tracts within the ventrolateral region of
the hypothalamus. A low density of TH+ neurons with ovoid cell
bodies was observed. A mixture of bipolar and multipolar neurons
with no specific dendritic orientation was evident. A moderate
density of TH+ neurons close to the wall of the third ventricle
throughout the hypothalamus was identified as the A14 nucleus
(Fig. 8A). The neurons had ovoid shaped somas and were mostly of
the bipolar type with some being multipolar. The dendrites were
observed to be orientated roughly parallel to the ventricular wall.
The A13 nucleus was identified as a moderate density of TH+
neurons located within the dorsal and lateral regions of the
hypothalamus, lateral to the fornix, extending into and around the
zona incerta in the ventral thalamus. The neurons exhibited
morphological similarity to those within the A15d nucleus.
In the ventral portion of the hypothalamus, within and in close
proximity to the arcuate nucleus, a low tomoderate density of TH+
neurons was identified as the A12 nucleus. The neurons of this
nucleus were located near the midline and surrounded the floor of
the third ventricle. The cell bodies were ovoid and the neurons a
mixture of the bipolar and multipolar types with dendrites
orientated roughly parallel to the floor of the brain. Around the
posterior pole of the third ventricle, in the most caudal part the
hypothalamus, a low density of TH+ neurons arranged in columns
on either side of the midline was identified as the A11 nucleus. The
acti
) in
w TH
sub
tia n
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 67Fig. 8. Photomicrographs of selected neuronal groups immunohistochemically re
periventricular nucleus (A14) located adjacent to the walls of the third ventricle (3V
(A10dc) located around the ventral region of the cerebral aqueduct (ca). (C) Very fe
ventrolateral periaqueductal grey matter, while the compact portion of the nucleus
neurons. (D) The retrorubral nucleus (A8) and the lateral nucleus of the substan
(C) = 500mm and applies to (A–C). Scale bar in (D) = 1 mm.ve for tyrosine hydroxylase within the brain of the rock hyrax. (A) The rostral
the hypothalamus. (B) The dorsal caudal nucleus of the ventral tegmental complex
+ neurons are found in the diffuse portion of the locus coeruleus (A6d) within the
coeruleus (A7sc) in the adjacent dorsal pontine tegmentum is rich in strongly TH+
igra complex (A9l) in the lateral and ventral midbrain tegmentum. Scale bar in
neurons of this nucleus were large and multipolar, with ovoid to
polygonal shaped somas. No specific dendritic orientation was
observed for these neurons. The cell bodies of the A11 neurons
were larger than those of all the other neurons of the diencephalic
nuclei, with this feature serving as a ready marker for delineation
of this nucleus.
3.2.3. Midbrain nuclei
3.2.3.1. Ventral tegmental area nuclei (VTA, A10 complex). The A10
nuclear complex (comprised of the following nuclei: A10 – the
ventral tegmental area nucleus; A10c – ventral tegmental area,
central nucleus; A10d – ventral tegmental area, dorsal nucleus;
A10dc – ventral tegmental area, dorsal caudal nucleus) was located
within the medial part of the midbrain tegmentum slightly
anterior to the level of the oculomotor nerve nucleus (Figs. 2L–O
and 3). The nuclei forming this complex originated from the dorsal
and dorsolateral areas around the interpeduncular nucleus and
extended dorsally to the periaqueductal grey matter, where the
aqueduct component was formed caudally. The A10 nucleus was
identified as a high density of TH+ neurons located between the
interpeduncular nucleus and the root of the oculomotor nerve
(Fig. 9). The neurons were distributed dorsally and dorsolaterally
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7468to the interpeduncular nucleus. The cell bodieswere ovoid in shape
and the neuronswere amixture of the bipolar andmultipolar types
showing no specific dendritic orientation. Immediately dorsal to
the interpeduncular and A10 nuclei, a moderate to high density of
TH+ neurons was identified as the A10c nucleus (Fig. 9). The
neurons of this nucleus formed a characteristically roughly
triangular shaped pattern in the midline. A similar neuronal
morphology as the neurons of the A10 nucleus was observed for
the neurons of the A10c nucleus; however the neurons of A10c
were slightly smaller in size than those of A10, this size difference
serving as a marker for delineating between these two adjacent
nuclei. A moderate density of TH+ neurons located dorsal to and
between the A10c nucleus and the periaqueductal grey matter and
medial to the oculomotor nerve nucleus, were identified as A10d
(Fig. 9). The cell bodies of these neurons were ovoid in shape,
bipolar in type and showed a dorsoventral dendritic orientation.
The A10dc nucleus was located within the periaqueductal grey
matter around the lower half of the border of the cerebral aqueduct
(Fig. 8B). The TH+ neurons forming the A10dc nucleus were small,
Fig. 9. Photomicrograph montage of the nuclear organization of the ventral
tegmental area (A10, A10c, A10d) and the substantia nigra (A9m, A9pc, A9l, A9v) in
the rock hyrax as revealed using tyrosine hydroxylase immunohistochemistry.
Scale bar = 1 mm. IP – interpeduncular nucleus; PC – cerebral peduncle; Rmc – red
nucleus, magnocellular division.ovoid in shape, multipolar and the dendrites were orientated
roughly parallel to the edge of the cerebral aqueduct.
3.2.3.2. Substantia nigra nuclear complex (A9). In the ventrolateral
region of the midbrain tegmentum, dorsal to the cerebral
peduncles, four distinct nuclei comprising the substantia nigra
nuclear complexwere identified. These nuclei were: the substantia
nigra, pars compacta nucleus (A9pc); substantia nigra, pars
lateralis nucleus (A9l); substantia nigra, ventral or pars reticulata
nucleus (A9v); and the substantia nigra, pars medialis nucleus
(A9m) (Figs. 2K–N and 3). The A9pc nucleus was identified as a
mediolaterally oriented band of TH+ neurons of a moderate to high
density located dorsal to the cerebral peduncle (Fig. 9). The
neurons were ovoid in shape and bipolar with a medial to lateral
dendritic orientation, parallel to the direction of the band. The A9l
nucleus was located within the ventrolateral region of the
midbrain tegmentum, dorsolateral to the lateral border of the
cerebral peduncle in a position lateral to the A9pc nucleus
(Figs. 8D, 9). The somas of the TH+ neurons of this nucleus were
either polygonal or triangular in shape. The neurons were
multipolar and did not exhibit any specific dendritic orientation.
The A9v nucleus was located in the grey matter just dorsal to the
cerebral peduncle ventral to the A9pc and A9l nuclei (Fig. 9). The
neuronswere found in a low tomoderate density in this region and
showed a rough dorsoventral dendritic orientation. A high density
of TH+ neurons located between the medial edge of the A9pc
nucleus and the root of the oculomotor nerve was identified as the
A9m nucleus (Fig. 9). The neurons of this nucleus showed a similar
neuronal morphology as A9pc, with the neurons showing no
specific dendritic orientation.
3.2.3.3. Retrorubal nucleus (A8). Within the lower half of the
midbrain tegmentum a moderate number of TH+ neurons were
found dorsal to the A9 complex, and ventral and caudal to the
magnocellular division of the red nucleus. These neurons, which
exhibited a moderate density throughout this region, were
assigned to the A8 nucleus (Figs. 2N–O and 3). The TH+ neuronal
bodies were ovoid in shape, bipolar and multipolar in type, and
showed no specific dendritic orientation (Fig. 8D).
3.2.4. Pontine nuclei – the locus coeruleus (LC) nuclear complex
The locus coeruleus complex was identified as a large
aggregation of TH+ neurons within the pontine region. Five nuclei
within this complex were identified, being: the subcoeruleus
compact nucleus (A7sc), subcoeruleus diffuse nucleus (A7d), locus
coeruleus diffuse nucleus (A6d), fifth arcuate nucleus (A5), and the
dorsal medial nucleus of the locus coeruleus (A4) (Figs. 2P–Q, 3).
A7sc was located within the dorsal region of the pontine
tegmentum adjacent to the periventricular grey matter. A high
density of TH+ neurons with ovoid somas of the multipolar type
represented this nucleus (Fig. 8C). The neurons of A7sc showed no
specific dendritic orientation. The description of this nucleus
coincides to what was previously described as the subcoeruleus
(Dahlstro¨m and Fuxe, 1964). TH+ neurons ventrolateral to A7sc,
anterior to the trigeminal motor nucleus, in the lateral and
dorsolateral region of the pontine tegmentumwere assigned to the
A7d nucleus. Some of the TH+ neurons assigned to A7d were also
locatedmedial and ventral to the superior cerebellar peduncle. The
TH+ neurons in this region were found in a moderate to low
density, were ovoid in shape, multipolar and showed no specific
dendritic orientation. A compact A6 built of TH+ neurons was
absent, and only a small number of TH+ neurons were located
within the ventrolateral portion of the periventricular grey matter
adjacent to A7sc, and identified as A6d (Fig. 8C). The neurons of
A6d had a similar morphology to A7sc and showed no specific
dendritic orientation. In comparison to other mammalian species,
postrema was identified as a high-density cluster of small TH+
neurons. This nucleus was located just anterior to the spinome-
dullary junction, adjacent to the floor of the fourth ventricle and
dorsal to the nucleus tractus solitarius and the central canal. The
neuronswere ovoid shaped, bipolar in type, and showed no specific
dendritic orientation.
3.3. Serotonergic nuclei
A number of distinct serotonergic immunoreactive (5HT+)
nuclei were found throughout the brainstem of the rock hyrax.
These were observed from the level of the decussation of the
superior cerebellar peduncle through to the spinomedullary
junction (Figs. 2M–W, 3, 10, 11). These nuclei were readily
divisible into rostral and caudal nuclear clusters and the location of
the nucleiwithin these clusterswas found to be similar towhat has
been described for other eutherian mammals (e.g. To¨rk, 1990;
Manger et al., 2002c; Badlangana et al., 2007; Maseko et al., 2007;
Moon et al., 2007; Dwarika et al., 2008; Limacher et al., 2008;
Bhagwandin et al., 2008).
3.3.1. Rostral cluster
3.3.1.1. Caudal linear nucleus (CLi). A moderate density of 5HT+
neurons was located within the midbrain tegmentum of the rock
hyrax. A cluster of these neurons was found to be in a position
immediately dorsal to the interpeduncular nucleus, and anterior to
the decussation of the superior cerebellar peduncle in the ventral
midline of the midbrain tegmentum. This cluster was designated
as the caudal linear nucleus (Figs. 2M–N, 3). At the lateral border of
the interpeduncular nucleus, the 5HT+ neurons of the CLi also
extended to the ventral surface of the brain and it was noted that
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 69the rock hyrax had markedly reduced number of immunopositive
neurons within A6d (Dahlstro¨m and Fuxe, 1964; Fuxe et al., 1969,
2007b; Smeets and Gonza´lez, 2000; Maseko et al., 2007; Limacher
et al., 2008; Bhagwandin et al., 2008). The TH+ neurons assigned to
the A5 nucleus was located lateral to the superior olivary nucleus,
anterior to the facial nerve nucleus and ventral to the trigeminal
motor nerve nucleus, in the ventrolateral region of the pontine
tegmentum. The TH+ neurons of this nucleus were few in number
and had a similar neuronal morphology to the neurons seen in the
other locus coeruleus nuclei. The TH+ neurons assigned to the A4
nucleus were located adjacent to the wall of the fourth ventricle,
medial to superior cerebellar peduncle, in the dorsolateral part of
the caudal periventricular grey matter. This nucleus was repre-
sented by a small number of TH+ neurons with a similar
morphology to the neurons seen in the other locus coeruleus
nuclei.
3.2.5. Medullary nuclei
TH+ neurons representing five putative catecholaminergic
nuclei were found within in the medulla of the rock hyrax. The
nuclei identified were: the rostral ventrolateral tegmental nucleus
(C1), rostral dorsomedial nucleus (C2), caudal ventrolateral
tegmental nucleus (A1), caudal dorsomedial nucleus (A2) and
area postrema (AP) (Figs. 2R–X, 3). The rostral dorsal midline
nucleus (C3), a feature so far only identified in rodents (Smeets and
Gonza´lez, 2000; Maseko et al., 2007; Limacher et al., 2008;
Bhagwandin et al., 2008) was not present. The neurons forming the
C1 nucleus were located in the ventrolateral region of the anterior
medulla and were distinguished as a column of TH+ neurons that
extended from the level of the superior olivary nucleus to the level
of nucleus ambiguus. In the course of its distribution this band of
neurons was located lateral to superior olivary nucleus and facial
nerve nucleus, and medial to nucleus ambiguus. The neurons of
this nucleuswere ovoid in shape,multipolar and exhibited amesh-
like dendritic network interspersed among the ascending and
descending fasciculi of the medulla. A low density of TH+ neurons
within the dorsal region of the medulla was identified as C2. This
nucleus was located anterior and dorsal to the vagus motor nerve
nucleus. The dorsal strip of C2 (Kalia et al., 1985a,b), which is the
part of the C2 nucleus that is located near the floor of the fourth
ventricle above the nucleus of the vagus motor nerve, could be
identified. The continuation of the dorsal strip, which is known as
the rostral subdivision of the C2 nucleus (Kalia et al., 1985a, b), was
not present. The neurons of this nucleuswere few in number and of
low density. The somal shape was ovoid and the neurons were
bipolar with dendrites orientated parallel to the floor of the fourth
ventricle. The A1 nucleus was identified as a column of TH+
neurons within the ventrolateral region of the caudal medulla. The
neurons of this nucleus were distributed from the level of nucleus
ambiguus to the level of the spinomedullary junction. In the course
of its distribution the neurons of the A1 nucleus were located
lateral to nucleus ambiguus and the lateral reticular nucleus. The
distinguishing factor between the partly overlapping caudal and
rostral neuronal columns of the A1 and C1 (caudal part) nucleus,
respectively, was their position relative to the nucleus ambiguus.
The A1 column was located lateral to nucleus ambiguus while the
caudal part of the C1 column was located medial to this nucleus.
The neuronal morphology and dendritic organization of the A1
nucleus were similar to that of the C1 nucleus. A number of
moderately sized TH+ neurons were located between, as well as
around, the nuclei of the dorsal motor vagus and hypoglossal
nerves and these were assigned to the A2 nucleus; however, none
was found in the nucleus tractus solitarius, which is the major
location for this cell group in rodents (Dahlstro¨m and Fuxe, 1964).
These neurons were ovoid in shape, bipolar in type and their
dendrites exhibited a medial to lateral orientation. The areathis nucleus was the most rostrally located nucleus of all the 5HT+
Fig. 10. Photomicrograph montage of the neuronal groups immunohistochemically
reactive for serotonin within the dorsal raphe nuclear complex of the rock hyrax
brain. Dorsal raphe lateral nucleus (DRl), dorsal raphe dorsal nucleus (DRd), dorsal
raphe ventral nucleus (DRv), dorsal raphe interfascicular nucleus (DRif), dorsal
raphe peripheral nucleus (DRp) and the median raphe nucleus (MnR). Scale
bar = 1 mm. ca – cerebral aqueduct.
ve
u
nd
ral
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7470nuclei. The neurons of this nucleus had ovoid shaped somas, were
bipolar in nature, and had a rough dorsoventral dendritic
orientation (Fig. 11A and B).
3.3.1.2. Supralemniscal nucleus (B9). The B9 nucleus was identified
as a loosely packed arc of 5HT+ neurons, lateral to the
interpeduncular nucleus and caudal to the A9pc nucleus
(Figs. 2N and 3). This nucleus was located superior to the medial
lemniscus andwas found to be in continuity with the ventrolateral
neurons of the CLi nucleus. The neurons had ovoid shaped somas,
were bipolar in type, and the dendrites had no specific orientation
(Fig. 11B).
3.3.1.3. Median raphe (MnR). In a pararaphe position, on either side
of themidline, two clear, densely packed columns of 5HT+ neurons,
extending dorsal to ventral along the midbrain and pontine
tegmentum, were identified as the MnR nucleus (Figs. 2N–Q and
3). The neurons forming this nucleus extended from the caudal
Fig. 11. Photomicrographs of selected neuronal groups immunohistochemically reacti
nucleus (CLi) at its most anterior level, immediately dorsal to the interpeduncular n
decussation of the superior cerebellar peduncle (xscp), and where at its most ventral a
(B9). (C) The caudal division of the dorsal raphe nucleus (DRc) lying immediately vent
and applies to all.aspect of the superior cerebellar peduncle to the level of the
anterior most aspect of the trigeminal motor nucleus. The cell
bodies of the 5HT+ neurons of theMnR nucleuswere ovoid, bipolar
in type and the dendrites showed a rough dorsoventral orientation
(Fig. 10).
3.3.1.4. Dorsal raphe nuclear complex (DR). Within the 5HT+ dorsal
raphe nuclear complex six distinct nuclei, extending from the
level of the oculomotor nerve nucleus to the level of the trigeminal
motor nerve nucleus within the periaqueductal and periventri-
cular grey matter, were identified. These nuclei were: the dorsal
raphe interfascicular (DRif) nucleus, the dorsal raphe ventral
(DRv) nucleus, the dorsal raphe dorsal (DRd) nucleus, the dorsal
raphe lateral (DRl) nucleus, the dorsal raphe peripheral (DRp)
nucleus, and the dorsal raphe caudal (DRc) nucleus (Figs 2O–Qand
3). A dense cluster of 5HT+ neurons within the most ventral
medial portion of the periventricular grey matter, between the
two medial longitudinal fasciculi, was identified as the DRif
nucleus (Fig. 10). The neurons of this nucleus had ovoid shaped
cell somas, were bipolar in type and had dendrites that were
orientated roughly dorsoventrally. Immediately dorsal to the DRif
nucleus and caudal to the oculomotor nerve nuclei, a high density
of 5HT+ neuronswas identified as the DRv nucleus (Fig. 10). These
neuronswere ovoid in shape andbipolarwith no specific dendriticorientation. The DRd nucleus was identified as a high density of
5HT+ neurons located ventral to the inferior border of the cerebral
aqueduct and immediately dorsal to theDRv nucleus (Fig. 10). The
neurons of this nucleus had a similar neuronal morphology and
dendritic orientation as theneurons of theDRvnucleus. A very low
density of 5HT+ neurons located lateral to theDRd andDRv nuclei,
anterior to the ChAT+ neurons of the LDT, in the ventrolateral
portion of the periaqueductal grey matter was identified as the
DRp nucleus. In the adjacent midbrain tegmentum, a small
number of 5HT+ neurons forming part of the DRp nucleus were
observed (Fig. 10). Of all the dorsal raphe nuclei, the tegmental
neurons of the DRp nucleus were the only ones not located within
the periventricular and periaqueductal grey matter. The 5HT+
neurons forming the DRp were ovoid to polygonal in shape,
multipolar, and showed no specific dendritic orientation. Adja-
cent to the ventrolateral edge of the cerebral aqueduct, in a
position dorsolateral to the DRd nucleus, a group of 5HT+ neurons
was assigned to the DRl nucleus (Fig. 10). A low density of large,
for serotonin in the midbrain and brainstem of the rock hyrax. (A) The caudal linear
cleus (IP). (B) The CLi at a more posterior level, where it is located ventral to the
lateral aspect it becomes continuous with the supralemniscal serotonergic nucleus
to the most posterior portion of the cerebral aqueduct (ca). Scale bar in (C) = 1 mm,ovoid, multipolar neurons with no specific dendritic orientation
was observed within this nucleus. This nucleus was readily
distinguished from the other dorsal raphe nuclei due to the low
neuronal density and the large somaof theneurons. An arc of 5HT+
neurons across the dorsal midline of the periventricular grey
matter, where the cerebral aqueduct opens into the fourth
ventricle, formedby the caudal coalescences of the two lateralized
clusters of the DRl nucleus, was identified as the DRc nucleus
(Fig. 11C). The neurons of this nucleus had a similar neuronal
morphology and dendritic orientation as those of the DRl nucleus.
Due to the lack of 5HT+ neurons within this region of the brain in
monotremes, the DRc was classified as an independent nucleus
(Manger et al., 2002c).
3.3.2. Caudal cluster
3.3.2.1. Raphe magnus nucleus (RMg). The 5HT+ neurons forming
the RMg nucleus were located within the rostral medullary
tegmentum, extending from the level of the anterior border of the
facial nerve nucleus to the anterior border of nucleus ambiguus
(Figs. 2Q–S and 3). These 5HT+ neurons were low in density and
formed two weakly expressed columns on either side of the
midline. The neurons of the RMg nucleus were large, multipolar
and exhibited a weak dorsoventral dendritic orientation.
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 713.3.2.2. Rostral and caudal ventrolateral serotonergic medullary
columns (RVL and CVL). Within the ventrolateral medullary
tegmentum a column of 5HT+ neurons was identified as the
RVL and CVL nuclei. This column extended from the level of the
facial nerve nucleus to the spinomedullary junction and appeared
to be a lateral extension of the neurons forming the RMg nucleus
(Figs. 2R–W, 3). The part of the column extending from the facial
nerve nucleus to the rostral border of nucleus ambiguus was
identified as the RVL nucleus, while the part of the column
extending from the rostral border of nucleus ambiguus to the
spinomedullary junction was designated as the CVL nucleus. The
columns of the RVL nucleus were found immediately dorsal to the
pyramidal tracts anteriorly, and it was noted that this group of
neurons bifurcated around the anterior pole of the inferior olivary
nucleus, giving rise to two bilateral columns. These columns
extended, in a position lateral to the inferior olivary nucleus,
caudally within the ventrolateral medulla. The neurons of the RVL
and CVL nuclei had a similar neuronal morphology and dendritic
orientation as the neurons of the RMg nucleus and decreased in
density from moderate, rostrally, to low, caudally. It was observed,
as with other mammalian species studied to date, that the RVL and
CVL columns were continuous (e.g. Maseko and Manger, 2007;
Maseko et al., 2007; Moon et al., 2007; Dwarika et al., 2008;
Bhagwandin et al., 2008); however, it is possible to make the
distinction between these two nuclei, as the CVL column has not
been found in the opossum or the montremes (Crutcher and
Humbertson, 1978; Manger et al., 2002c).
3.3.2.3. Raphe pallidus nucleus (RPa). A low density and number of
5HT+ neurons located between the pyramidal tracts and at the
ventral most border of the inferior olivary nucleus, within the
ventral surface of the midline and medial medulla, were identified
as the medial and lateral component of the RPa nucleus (Figs. 2R–T
and 3). The neurons of this nucleus exhibited fusiform-shaped
somas, were bipolar in type with a dorsoventral dendritic
orientation parallel to the medial border of the pyramidal tracts.
3.3.2.4. Raphe obscurus nucleus (ROb). On either side of themidline
a low density of 5HT+ neurons, extending dorsal to ventral, from
the level of the nucleus ambiguus to the spinomedullary junction
was identified as the ROb nucleus (Figs. 2T–W, 3). The neurons
were arranged in two loosely packed columns on either side of the
midline. The cell bodies of these neurons were fusiform in shape,
bipolar, and had a dorsoventral dendritic orientation. Unlike in
some other species studied, there were no neurons located a short
distance from the central columns (e.g. Maseko and Manger, 2007;
Maseko et al., 2007; Moon et al., 2007; Dwarika et al., 2008;
Bhagwandin et al., 2008).
4. Discussion
The aim of the current study was to reveal the nuclear
organization of the cholinergic, putative catecholaminergic and
serotonergic systems of the rock hyrax, P. capensis. The results
revealed that these systems, for the most part, are similar to what
has previously been described in othermammals; however, certain
specific differences were observed regarding the nuclear organiza-
tion of the cholinergic and putative catecholaminergic systems.
The nuclear organization of the serotonergic system was similar to
what has been documented for other eutherian mammals (e.g.
To¨rk, 1990; Maseko et al., 2007; Moon et al., 2007; Dwarika et al.,
2008; Limacher et al., 2008; Bhagwandin et al., 2008). The
differences that were noted for the cholinergic system included
the existence of cholinergic immunoreactive neurons in the
anterior dorsal and ventral nuclei of the dorsal thalamus, the
presence of parvocellular and magnocellular divisions/nuclei ofboth the LDT and PPT, and the presence of a large cholinergic cell
group in the periventricular grey of the rostral medulla (pIX). The
features of the putative catecholaminergic system that differed
significantly from that seen in other mammals were the absence of
a dense cluster of TH+ neurons constructing the locus coeruleus
compact (A6c) and a diffuse locus coeruleus (A6d) with only few
TH+ neurons. We discuss each of these unusual features in turn,
and then compare the overall organization of these systems in the
hyrax to prior observations of these systems across various
mammalian species.
4.1. Cholinergic neurons in the anterodorsal (AD) and anteroventral
(AV) nuclei of the dorsal thalamus of the rock hyrax
The AD and AV nuclei identified in the brain of the rock hyrax
form part of the anterior nuclei of the dorsal thalamus and are a
general feature of the mammalian dorsal thalamus (Jones, 2007).
The anterior nuclear group typically consists of four distinct nuclei,
the anterodorsal (AD), anteromedial (AM), anteroventral (AV), and
lateral dorsal (LD) nuclei (Jones, 2007). In the rock hyrax the AD
nucleus was located within the anterior and dorsal regions of the
dorsal thalamus, as is typical of mammals, and exhibited a strong
cholinergic neuropil staining along with ChAT immunoreactive
neurons scattered around themargins of the nucleus. Lateral to the
AD nucleus in the rock hyrax, a second nucleus exhibiting aweaker
cholinergic neuropil staining with ChAT immunoreactive neurons
along the uppermedial and lateral borders was identified as the AV
nucleus. While both the AD and AV nuclei are known for their
strong histochemical reactivity to acetylcholinesterase, choline
acetyltransferase immunoreactive neurons have not been reported
in either of these nuclei in any othermammal (Jones, 2007;Maseko
et al., 2007).
Afferent input to these nuclei originates largely from the
hypothalamus and hippocampal formation, but also includes
corticothalamic and thalamocortical connections (Jones, 2007).
The AD and AV nuclei have been reported to be involved in spatial
learning and memory, and navigation; however, these nuclei are
not the sole contributors, but rather work in conjunction with the
other anterior thalamic nuclei to achieve these functions (Segal
et al., 1988; van Groen et al., 2002; Oda et al., 2003; Wolff et al.,
2008). It is difficult to hypothesize on the potential effect of the
presence of cholinergic neurons within and closely surrounding
these nuclei in regards to function; but given that the cholinergic
system in general appears to be related to activity requiring both
alertness and awareness, these neurons may have the effect of
enhancing the wakefulness promoting functions of the anterior
thalamic nuclei in the hyrax, possibly leading also to enhanced
learning and memory functions.
Presently, it appears that these anterior thalamic cholinergic
neurons are unique to the rock hyrax as they have not been
observed in other mammals. Despite this, we can only tentatively
conclude that they are a unique feature as no othermembers of the
Afroplacentalia have been examined for the presence of these
neurons. Thus, these anterior thalamic cholinergic neuronsmay be
unique to the hyrax, or may be found in other members of the
Afroplacentalia, especially the closely related Proboscideans and
Sirenians. Until further Afroplacentalia species have been exam-
ined the uniqueness of this aspect of the cholinergic system
remains uncertain.
4.2. Magnocellular and parvocellular divisions/nuclei of the
pedunculopontine and laterodorsal tegmental nuclei in the rock hyrax
Within all mammals studied to date, cholinergic neurons
within the pontine tegmentum (PPT) and periventricular grey
matter (LDT) have been reported (e.g. Maseko et al., 2007). In all
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7472these prior studies the morphology of the cholinergic neurons
within and between the PPT and LDT nuclei has been reported to
be homogeneous (e.g. Woolf, 1991; Manger et al., 2002a; Maseko
and Manger, 2007; Maseko et al., 2007; Limacher et al., 2008;
Bhagwandin et al., 2008). The situation observed in the current
study in the rock hyrax raises a significant difference in this
regard, as the cholinergic neurons within these nuclei did not
exhibit a homogenous morphology. Rather, we identified two
neuronal types within each of these nuclei, and the inner spatial
distribution and segregation of these two neuronal types have led
to the proposal of the subdivision of these nuclei into parvocel-
lular and magnocellular PPT and LDT nuclei. The magnocellular
subdivision of both the PPT and LDT nuclei exhibited a neuronal
morphology reminiscent of that described for all other mammals
(Woolf, 1991; Manger et al., 2002a; Maseko et al., 2007; Limacher
et al., 2008; Bhagwandin et al., 2008). For this reason, plus the
topography of the neurons forming thesemagnocellular divisions
(LDTmc being in the ventrolateral portion of the grey matter and
PPTmc lying in the adjacent tegmentum), we propose that the
magnocellular divisions of these nuclei in the rock hyrax are
homologous to the LDT and PPT nuclei described for all other
mammals.
The parvocellular cholinergic subdivisions, on the other hand,
have not been reported in any other mammal studied to date,
although a non-cholinergic parvocellular component has at
times been included as part of the PPT in rat and cat (Semba and
Fibiger, 1989). The morphology of the neurons ascribed to the
parvocellular divisions, being small, and bipolar with oval soma,
is reminiscent of cholinergic interneurons found in the cerebral
cortex (Bhagwandin et al., 2006) or superior colliculus (Tan and
Harvey, 1989) in some other mammals. These putative
cholinergic interneurons were found medial to the LDTmc,
and dorsal to the PPTmc. It is possible that these cholinergic
interneurons are also immunoreactive for GABA, forming part of
the pontine GABAergic system. A similar situation has been
observed in the cortical cholinergic neurons found in certain
species of mammals (Bhagwandin et al., 2006). Potentially, a
change at the molecular level of organization occurred within
these neurons, such as molecular interactions, adding multiple
transmission lines and thus increased functionality to this
neuronal group (Agnati et al., 1986). These potentially GABAer-
gic neurons may have acquired ChAT proteins resulting in their
immunoreactivity.
The above described potential molecular change and increased
transmission lines in these cholinergic interneuron groups in close
apposition to the traditional cholinergic LDT and PPT could lead to
interesting functional outcomes. It is well known that GABAergic
neurons within the pons play a major role in the regulation of the
traditional cholinergic LDT and PPT with regard to the sleep-wake
cycle especially REM sleep (Maloney et al., 1999, 2000; Boissard
et al., 2003; Pal and Mallick, 2006; Luppi et al., 2007). The location
of the cholinergic, potentially GABAergic neurons described as the
PPTpc and LDTpc in the current study makes them prime
candidates forming part of a local GABAergic network involved
in the switching of states during the sleep-wake cycle (Luppi et al.,
2007). The added transmission line related to the production of
acetylcholine in the hyrax may produce some unusual effects on
the measurable physiological parameters of sleep, especially REM
sleep in the hyrax.
Again it appears that these parvocellular cholinergic LDT and
PPT may be unique to the rock hyrax as they have not been
observed in other mammals (e.g. Maseko et al., 2007); however, as
with the other currently unique aspects of the cholinergic system
in the hyrax we can only tentatively conclude that they are unique
features until other members of the Afrotheria have been
examined for the presence of these nuclei.4.3. Increased numbers of preganglionic cholinergic neurons in the
inferior salivatory nucleus (pIX) of the rock hyrax
Cholinergic neuronswithin the preganglionic inferior salivatory
nucleus have been identified in many mammalian species
including rodents, carnivores, megachiropterans and primates,
yet these are absent from the brains of other mammals studied
(Maseko et al., 2007). Cholinergic neurons were identified in this
nucleus in the rock hyrax; however, it appears from a qualitative
comparison that the numbers of cholinergic neurons in this
nucleus in the rock hyrax are far greater than that seen in the other
mammals in which these neurons have been found. This
qualitative impression, while impressive, needs quantitative
support. The possible projections and function of this enlarged
periventricular cholinergic cell group of the medulla are unknown
and comparisons across other species belonging to the Afrotheria
are needed to determine whether this feature of the cholinergic
system of the hyrax is a feature of the Afrotherians in general.
4.4. Lack of TH+ neurons in compact locus coeruleus (A6c) and
decreased numbers of TH+ neurons in the locus coeruleus proper (A6d)
of the rock hyrax
The diffuse division of the locus coeruleus (A6d), or the locus
coeruleus proper (Dahlstro¨m and Fuxe, 1964), has been reported in
all mammals studied to date, while the compact portion has only
been reported in rabbits, tree shrew, megabats and primates (e.g.
Maseko et al., 2007). While the A6c is lacking in the rock hyrax, the
A6d nucleus is present but was particularly low in the number of
TH immunoreactive neurons. This is a qualitative impression, and
further work comparing the number of neurons in this nucleus
across mammals and determining what kind of relationship may
exist between neuronal number and brain mass (perhaps
allometric) is required to confirm this impression; however, the
qualitative impression is so dramatic as to warrant specific
discussion. One interesting point to be noted is that even though
the A6d nucleus showed an apparent reduction in the number of
TH+ neurons, the other nuclei of the locus coeruleus complex
(A7sc, A7d, A5 and A4) do not appear to be relatively more cell rich
in comparison to other mammals (again a qualitative impression
needing quantitative support).
This potential reduction in the number of neurons within the
locus coeruleus may lead to a reduction in the amount of
noradrenaline released in the specific target regions of the A6d
in the hyrax. It should be noted that A6d in the rat and other
mammals projects to all cortical regions of the brain, including the
cerebellar cortex, as well as to the thalamus, the lower brainstem
reticular formation and the dorsal and ventral horns of the spinal
cord (see Fuxe et al., 1970). The neurons of the locus coeruleus
represent a tonic arousal system important for vigilance and
attention (see Fuxe et al., 2007a,b). The relative absence of the
catecholamine component of the locus coeruleus proper (A6d) in
the hyrax is a highly interesting observation. Has the subcoeruleus
(A7sc and A7d) taken over the role of the locus coeruleus (A6d) in
the hyrax or does the hyrax show a lack of tonic arousal? These are
major questions to be answered in future work. The reduced
number of neurons found within the A6d in the hyrax may be a
feature unique to this species. Further work on other members of
the Afrotheria will determine whether this is indeed a species-
specific feature or a feature of a broader taxonomic grouping.
4.5. Evolutionary and phylogenetic considerations
The current study extends the database that may be used for
comparison across mammalian species to determine both phylo-
genetic and evolutionary trends of the cholinergic, putative
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–74 73catecholaminergic and serotonergic systems. Of these systems the
serotonergic system appears to be the most conservative in terms
of changes in the nuclear organization associated with different
lineages. The rock hyrax, along with all other eutherian mammals
studied to date and the metatherian wallaby appear to have the
same nuclear organization (Maseko et al., 2007). The metatherian
opossum (Crutcher and Humbertson, 1978) and the proTo´therian
monotremes (Manger et al., 2002c) have only small comparative
differences in the nuclear organization of the serotonergic system.
The putative catecholaminergic system of the hyrax has a nuclear
organization almost identical to that seen in rodents (except for the
lack of a C3 nucleus, Dahlstro¨m and Fuxe, 1964; Smeets and
Gonza´lez, 2000) and megabats and primates (except for the lack of
a compact division of the locus coeruleus, A6c,Maseko et al., 2007).
This nuclear organization aligns the hyrax phylogenetically most
closely with the rodents and primates studied to date. The
cholinergic system, for the most part, shares a similar plan of
organization as that described for the rodents, megabats and
primates (Maseko et al., 2007). Despite this, the hyrax does evince a
degree of nuclear organizational uniqueness compared with other
mammalian species through the possession of cholinergic neurons
in the anterior nuclei of the dorsal thalamus and the parvocellular
LDT and PPT nuclei of the pontine region. In the broader sense, the
organization of these systems in the hyrax appears most similar to
those seen in the rodents, rabbit, tree shrew, megabats and
primates (Maseko et al., 2007). In this sense, the nuclear
organizations of the neural systems studied reflect current
concepts of the phylogenetic interrelationships of mammalian
lineages (Arnason et al., 2008).
The overall unique complement of nuclei of the three systems
studied supports the proposal ofManger (2005) that indicates each
order will possess a unique nuclear complement of these systems.
The current study supports the concept of placing the hyraxes into
the phylogenetic grouping of an order to the exclusion of other
mammals. It would be of interest to examine the cholinergic,
catecholaminergic and serotonergic systems of other species of
hyraxes to determine whether or not the ‘‘unique’’ features
described in the brain of the rock hyrax are shared by all members
of the Hyracoidea. It will also be of interest to examine other
members of the Afrotheria and members of the Xenarthra to
determine if these ‘‘unique’’ features are limited to this order
(Hyrocoidea), cohort (Afroplacentalia), or are part of a broader
phylogenetic grouping (such as the Notoplacentalia) (Arnason
et al., 2008). The current study sets out to test several potential
predictions. The first prediction, that the hyrax will have many
nuclei in common with other mammalian species, is supported.
The second prediction, that the hyrax will have some unique
features, is currently, tentatively supported. The strength of this
second prediction and the third prediction can be tested by
examining further species of the Afrotherian supercohort. The
fourth prediction, that the hyrax may lack nuclei found in other
species, is supported due to the lack of two catecholaminergic
nuclei found either exclusively in rodents (C3), or seen in rabbits,
tree shrews, megabats and primates (A6d). The current study then
supports the usefulness of using these systems as indicators of
phylogenetic relationships and supports current concepts regard-
ing the evolution of these neural systems (Manger, 2005).
Acknowledgments
This work was supported by funding from the South African
National Research Foundation (PRM) and SIDA (KF). The authors
wish to extend their gratitude to the members of the Central
Animal Services of the University of the Witwatersrand for their
enthusiastic and expert assistance with animal handling and care
during this project.References
Agnati, L.F., Fuxe, K., Zoli, M., Pich, E.M., Benfenati, F., Zini, I., Goldstein, M., 1986.
Aspects on the information handling by the central nervous system: focus on
cotransmission in the aged rat brain. Prog. Brain Res. 68, 291–301.
Ande´n, N.E., Carlsson, A., Dahlstro¨m, A., Fuxe, K., Hillarp, N.A˚., Larsson, K., 1964.
Demonstration and mapping out of nigro-neostriatal dopamine neurons. Life
Sci. 15, 523–530.
Arnason, U., Adegoke, J., Gullberg, A., Harley, E.H., Janke, A., Kullberg, M., 2008.
Mitogenomic relationships of placental mammals and molecular estimates of
their divergences. Gene 421, 37–51.
Badlangana, N.L., Bhagwandin, A., Fuxe, K., Manger, P.R., 2007. Distribution and
morphology of putative catecholaminergic and serotonergic neurons in the
medulla oblongata of a sub-adult giraffe, Giraffa camelopardalis. J. Chem. Neu-
roanat. 34, 69–79.
Bartus, R.T., Dean, R.L., Beer, B., 1982. Neuropeptide effect on memory in aged
monkeys. Neurobiol. Aging 3, 61–68.
Bhagwandin, A., Fuxe, K., Manger, P.R., 2006. Choline acetyltransferase immunor-
eactive cortical interneurons do not occur in all rodents: a study of the
phylogenetic occurrence of this neural characteristic. J. Chem. Neuroanat. 32,
208–216.
Bhagwandin, A., Fuxe, K., Bennett, N.C., Manger, P.R., 2008. Nuclear organization and
morphology of cholinergic, putative catecholaminergic and serotonergic neu-
rons in the brains of two species of African mole-rat. J. Chem. Neuroanat. 35,
371–387.
Bjarkam, C.R., Sorensen, J.C., Geneser, F.A., 1997. Distribution and morphology of
serotonin-immunoreactive neurons in the brainstem of the New Zealand white
rabbit. J. Comp. Neurol. 380, 507–519.
Boissard, R., Fort, P., Gervasoni, D., Barbagli, B., Luppi, P.H., 2003. Localization of the
GABAergic and non-GABAergic neurons projecting to the sublaterodorsal
nucleus and potentially gating paradoxical sleep onset. Eur. J. Neurosci. 18,
1627–1639.
Crutcher, K.A., Humbertson, A.O., 1978. The organization of monoamine neurons
within the brainstem of the North American opossum (Didelphis virginiana). J.
Comp. Neurol. 179, 195–222.
Dahlstro¨m, A., Fuxe, K., 1964. Evidence for the existence of monoamine-containing
neurons in the central nervous system. I. Demonstration of monoamine in the
cell bodies of brainstem neurons. Acta Physiol. Scand. 62, 1–52.
Diksic, M., Young, S.N., 2001. Study of the brain serotonergic system with labelled
alpha-methyl-L-tryptophan. J. Neurochem. 78, 1185–1200.
Dwarika, S., Maseko, B.C., Ihunwo, A.O., Fuxe, K., Manger, P.R., 2008. Distribution and
morphology of putative catecholaminergic and serotonergic neurons in the
brain of the greater canerat, Thryonomys swinderianus. J. Chem. Neuroanat. 35,
108–122.
Fuxe, K., Ho¨kfelt, T., Ungerstedt, U., 1969. Distribution of monoamines in the
mammalian central nervous system by histochemical studies. In: Hooper, G.
(Ed.), Metabolism of Amines in the Brain. Macmillan, London, pp. 10–22.
Fuxe, K., Ho¨kfelt, T., Ungerstedt, U., 1970. Morphological and functional aspects of
central monoamine neurons. Int. Rev. Neurobiol. 13, 93–126.
Fuxe, K., Manger, P., Genedani, S., Agnati, L., 2006. The nigrostriatal DA pathway and
Parkinson’s disease. J. Neural. Transm. Suppl. 70, 71–83.
Fuxe, K., Marcellino, D., Genedani, S., Agnati, L., 2007a. Adenosine A(2A) receptors,
dopamine D(2) receptors and their interactions in Parkinson’s disease. Mov.
Disord. 22, 1990–2017.
Fuxe, K., Dahlstro¨m, A., Ho¨istad, M., Marcellino, D., Jansson, A., Rivera, A., Diaz-
Cabiale, Z., Jacobsen, K., Tinner-Staines, B., Hagman, B., Leo, G., Staines, W.,
Guidolin, D., Kehr, J., Genedani, S., Belluardo, N., Agnati, L.F., 2007b. From the
Golgi–Cajal mapping to the transmitter-based characterization of the neuronal
networks leading to two modes of brain communication: wiring and volume
transmission. Brain Res. Rev. 55, 17–54.
Gallyas, F., 1979. Silver staining of myelin by means of physical development.
Neurolog. Res. 1, 203–209.
Ho¨kfelt, T., Johansson, O., Fuxe, K., Goldstein, M., Park, D., 1976. Immunohisto-
chemical studies on the localization and distribution of monoamine neuron
systems in the rat brain. I. Tyrosine hydroxylase in the mes- and diencephalon.
Med. Biol. 54, 427–453.
Ho¨kfelt, T., Martenson, R., Bjo¨rklund, A., Kleinau, S., Goldstein, M., 1984. Distribu-
tional maps of tyrosine-hydroxylase-immunoreactive neurons in the rat brain.
In: Bjo¨rklund, A., Ho¨kfelt, T. (Eds.), Handbook of Chemical Neuroanatomy, Vol.
2. Classical Neurotransmitters in the CNS, Part 1. Elsevier, Amsterdam, pp. 277–
379.
Jacobs, B.L., Azmitia, E.C., 1992. Structure and function of the brain serotonin
system. Physiol. Rev. 72, 165–229.
Jones, E.G., 2007. 2nd ed. The Thalamus, 2nd ed., Cambridge University Press,
Cambridge, UK.
Kalia, M., Fuxe, K., Goldstein, M., 1985a. Rat medulla oblongata. II. Noradrenergic
neurons, nerve fibers and preterminal processes. J. Comp. Neurol. 233, 308–332.
Kalia, M., Fuxe, K., Goldstein, M., 1985b. Rat medulla oblongata. III. Adrenergic (C1
and C2) neurons, nerve fibers and presumptive terminal processes. J. Comp.
Neurol. 233, 333–349.
Kitahama, K., Geffard, M., Okamura, H., Nagatsu, I., Mons, N., Jouvet, M., 1990.
Dopamine- and dopa-immunoreactive neurons in the cat forebrain with refer-
ence to tyrosine hydroxylase-immunohistochemistry. Brain Res. 518, 83–94.
Kitahama, K., Sakamoto, N., Jouvet, A., Nagatsu, I., Pearson, J., 1996. Dopamine-beta-
hydroxylase and tyrosine hydroxylase immunoreactive neurons in the human
brainstem. J. Chem. Neuroanat. 10, 137–146.
Klein, R.G., Cruz-Uribe, K., 1996. Size variation in the rock hyrax (Procavia capensis)
and late quarternary climatic change in South Africa. Quat. Res. 46, 193–207.
Lichtensteiger, W., 1966. Uptake of norepinepherine in periglomerular cells of the
olfactory bulb of the mouse. Nature 210, 955–956.
Lidbrink, P., Jonsson, G., Fuxe, K., 1974. Selective reserpine-resistant accumulation
of catecholamines in central dopamine neurones after dopa administration.
Brain Res. 67, 439–456.
Limacher, A.M., Bhagwandin, A., Fuxe, K., Manger, P.R., 2008. Nuclear organization
and morphology of cholinergic, putative catecholaminergic and serotonergic
neurons in the brain of the Cape porcupine (Hystrix africaeaustralis): increased
brain size does not lead to increased organizational complexity. J. Chem.
Neuroanat. 36, 33–52.
Luppi, P.H., Gervasoni, D., Verret, L., Goutagny, R., Peyron, C., Salvert, D., Leger, L., Fort,
P., 2007. Paradoxical (REM) sleep genesis: the switch from an aminergic-choli-
nergic to a GABAergic-glutamatergic hypothesis. J. Physiol. (Paris) 100, 271–283.
Maloney, K.J., Mainville, L., Jones, B.E., 1999. Differential c-Fos expression in
cholinergic, monoaminergic, and GABAergic cell groups of the pontomesence-
phalic tegmentum after paradoxical sleep deprivation and recovery. J. Neurosci.
19, 3057–3072.
Maloney, K.J., Mainville, L., Jones, B.E., 2000. c-Fos expression in GABAergic, seroto-
nergic, and other neurons of the pontomedullary reticular formation and raphe
after paradoxical sleep deprivation and recovery. J. Neurosci. 20, 4669–4679.
Manger, P.R., 2005. Establishing order at the systems level in mammalian brain
evolution. Brain Res. Bull. 66, 282–289.
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002a. Distribution and
morphology of cholinergic neurons in the brain of the monotremes as revealed
by ChAT immunohistochemistry. Brain Behav. Evol. 60, 275–297.
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002b. Distribution and
morphology of catecholaminergic neurons in the brain of monotremes as
revealed by tyrosine hydroxylase immunohistochemistry. Brain Behav. Evol.
60, 298–314.
Manger, P.R., Fahringer, H.M., Pettigrew, J.D., Siegel, J.M., 2002c. Distribution and
morphology of serotonergic neurons in the brain of the monotremes. Brain
Oda, S., Kuroda, M., Kakuta, S., Tanihata, S., Ishikawa, Y., Kishi, K., 2003. Ultra-
structure of ascending cholinergic terminals in the anteroventral thalamic
nucleus of the rat: a comparison with the mammillothalamic terminals. Brain
Res. Bull. 59, 473–483.
Pal, D., Mallick, B.N., 2006. Role of noradrenergic and GABA-ergic inputs in ped-
unculopontine tegmentum for regulation of rapid eye movement sleep in rats.
Neuropharmocology 51, 1–11.
Pompeiano, O., 2001. Role of the locus coeruleus in the static and dynamic control of
posture. Arch. Ital. Biol. 139, 109–124.
Previc, F.H., 1999. Dopamine and the origins of human intelligence. Brain Cogn. 41,
299–350.
Reiner, P.B., Fibiger, H.C., 1995. Functional heterogeneity of central cholinergic
systems. In: Bloom, F.E., Kupfer, D.J. (Eds.), Psychopharmacology: The Fourth
Generation of Progress. Raven, New York, pp. 147–153.
Rezek, O., Boldogkoi, Z., Tomba´cz, D., Kova´go´, C., Gerendai, I., Palkovits, M., To´th, I.E.,
2008. Location of parotid preganglionic neurons in the inferior salivatory
nucleus and their relation to the superior salivatory nucleus of rat. Neurosci.
Lett. 440, 265–269.
Ruggiero, D.A., Anwar, M., Gootman, P.M., 1992. Presumptive adrenergic neurons
containing phenylethanolamine N-methyltransferase immunoreactivity in the
medulla oblongata of neonatal swine. Brain Res. 583, 105–119.
Segal, M., Greenberger, V., Israeli, M., Biegon, A., 1988. A correlation between
regional acetylcholinesterase activity in rat brain and performance in a spatial
task. Behav. Brain Res. 30, 215–219.
Semba, K., Fibiger, H.C., 1989. Organization of central cholinergic systems. Prog.
Brain Res. 79, 37–63.
Siegel, J.M., 2006. The stuff dreams aremade of: anatomical substrates of REM sleep.
Nat. Neurosci. 9, 721–722.
Skinner, J.D., Chimimba, C.T., 2005. 3rd ed. The Mammals of the Southern African
Subregion, 3rd ed., Cambridge University Press, Cape Town.
Smeets, W.J.A.J., Gonza´lez, A., 2000. Catecholamine systems in the brain of verte-
brates: new perspectives through a comparative approach. Brain Res. Rev. 33,
308–379.
Tan, M.M.L., Harvey, A.R., 1989. The cholinergic innervation of normal and trans-
N. Gravett et al. / Journal of Chemical Neuroanatomy 38 (2009) 57–7474Behav. Evol. 60, 315–332.
Maseko, B.C., Manger, P.R., 2007. Distribution and morphology of cholinergic,
catecholaminergic and serotonergic neurons in the brain of Schreiber’s long-
fingered bat, Miniopterus schreibersii. J. Chem. Neuroanat. 34, 80–94.
Maseko, B.C., Bourne, J.A., Manger, P.R., 2007. Distribution and morphology of
cholinergic, putative catecholaminergic and serotonergic neurons in the brain
of the Egyptian Rousette flying fox, Rousettus aegyptiacus. J. Chem. Neuroanat.
34, 108–127.
Meister, B., Ho¨kfelt, T., Steinbusch, H.W., Skagerberg, G., Lindvall, O., Geffard, M., Joh,
T.H., Cuello, A.C., Goldstein, M., 1988. Do tyrosine hydroxylase-immunoreactive
neurons in the ventrolateral arcuate nucleus produce dopamine or only L-dopa?
J. Chem. Neuroanat. 1, 59–64.
Mitchell, J., Templeton, D., 1981. The origin of the preganglionic parasympathetic
fibers to themandibular and sublingual salivary glands in the rat: a horseradish
peroxidase study. J. Anat. 132, 513–518.
Moon, D.J., Maseko, B.C., Ihunwo, A., Fuxe, K., Manger, P.R., 2007. Distribution and
morphology of catecholaminergic and serotonergic neurons in the brain of the
highveld gerbil, Tatera brantsii. J. Chem. Neuroanat. 34, 134–144.planted superior colliculus in the rat: an immunohistochemical study. Neu-
roscience 32, 511–520.
Tillet, Y., 1995. Distribution of neurotransmitters in the sheep brain. J. Reprod. Fertil.
Suppl. 49, 199–220.
To¨rk, I., 1990. Anatomy of the serotonergic system. Ann. N. Y. Acad. Sci. 600, 9–35.
To´th, I.E., Boldogko˜i, Z., Medveczky, I., Palkovits, M., 1999. Lacrimal preganglionic
neurons form a subdivision of the superior salivatory nucleus of rat: transneur-
onal labeling by pseudorabies virus. J. Auton. Nerv. Syst. 77, 45–54.
van Groen, T., Kadish, I., Micheal Wyss, J., 2002. Role of the anterodorsal and
anteroventral nuclei of the thalamus in spatial memory in the rat. Behav. Brain
Res. 132, 19–28.
Wolff, M., Loukavenko, E.A., Will, B.E., Dalrymple-Alford, J.C., 2008. The extended
hippocampal-diencephalic memory system: enriched housing promotes recov-
ery of the flexible use of spatial representations after anterior thalamic lesions.
Hippocampus 18, 996–1007.
Woolf, N.J., 1991. Cholinergic systems in mammalian brain and spinal cord. Prog.
Neurobiol. 37, 475–524.
Distribution of orexin-A immunoreactive neurons and their terminal networks in
the brain of the rock hyrax, Procavia capensis
Nadine Gravett a, Adhil Bhagwandin a, Kjell Fuxe b, Paul R. Manger a,*
a School of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown 2193, Johannesburg, South Africa
bDepartment of Neuroscience, Karolinska Institutet, Retzius va¨g 8, S-171 77 Stockholm, Sweden
Journal of Chemical Neuroanatomy 41 (2011) 86–96
A R T I C L E I N F O A B S T R A C T
the distribution of orexin-A immunoreactive neurons and terminal networks
d
x,
Contents lists available at ScienceDirect
Journal of Chemical Neuroanatomy
journal homepage: www.e lsev ier .com/ locate / jchemneuAccepted 21 November 2010
Available online 30 November 2010
Keywords:
Afrotheria
Orexin
Hypocretin
Diurnal
Comparative neuroanatomy
immunohistochemically stained with an antibody to orexin-A. The staining revealed that the neurons
weremainly locatedwithin the hypothalamus aswith othermammals. The orexinergic terminal network
distribution also resembled the typical mammalian plan. High-density orexinergic terminal networks
were located within regions of the diencephalon (e.g. paraventricular nuclei), midbrain (e.g. serotonergic
nuclei) and pons (locus coeruleus), while medium density orexinergic terminal networks were evident in
the telencephalic (e.g. basal forebrain), diencephalic (e.g. hypothalamus), midbrain (e.g. periaqueductal
gray matter), pontine (e.g. serotonergic nuclei) and medullary regions (e.g. serotonergic and
catecholaminergic nuclei). Although the distribution of the orexinergic terminal networks was typically
mammalian, the rock hyrax did show one atypical feature, the presence of a high-density orexinergic
terminal network within the anterodorsal nucleus of the dorsal thalamus (AD). The dense orexinergic
innervation of the AD nucleus has only been reported previously in the Nile grass rat, Arvicanthis niloticus
and Syrian hamster, Mesocricetus auratus, both diurnal mammals. It is possible that orexinergic
innervation of the AD nucleus might be a unique feature associated with diurnal mammals. It was also
noted that the dense orexinergic innervation of the AD nucleus coincided with previously identified
cholinergic neurons and terminal networks in this particular nucleus of the rock hyrax brain. It is possible
that this dense orexinergic innervation of the AD nucleus in the brain of the rock hyraxmay act in concert
with the cholinergic neurons and/or the cholinergic axonal terminals, which in turn may influence
arousal states and motivational processing.
 2010 Elsevier B.V. All rights reserved.
Abbreviations: III, oculomotor nucleus; 3V, third ventricle; 4V, fourth ventricle; A1, caudal ventrolateral medullary tegmental nucleus; A2, caudal dorsomedial medullary
nucleus; A6, locus coeruleus; A7, subcoeruleus; A7d, nucleus subcoeruleus, diffuse portion; A7sc, nucleus subcoeruleus, compact portion; A9, substantia nigra; A10, ventral
tegmental area; A10c, ventral tegmental area, central; A10d, ventral tegmental area, dorsal; A10dc, ventral tegmental area, dorsal caudal; A11, caudal diencephalic group;
A12, tuberal cell group; A13, zona incerta; A14, rostral periventricular nucleus; A15d, anterior hypothalamic group, dorsal division; A15v, anterior hypothalamic group,
ventral division, acanterior commissure; AD, anterodorsal nucleus of the dorsal thalamus; Amyg, amygdala; AP, area postrema; AV, anteroventral nucleus of the dorsal
thalamus; B9, supralemniscal serotonergic nucleus; C, caudate nucleus; C1, rostral ventrolateral medullary tegmental group; C2, rostral dorsomedial medullary nucleus; ca,
cerebral aqueduct; Cb, cerebellum; cc, corpus callosum; Cl, claustrum; CLi, caudal linear nucleus; CN, cochlear nucleus; CP, cerebral peduncle; CVL, caudal ventrolateral
serotonergic group; DCN, deep cerebellar nuclei; Diag.B, diagonal band of Broca; DR, dorsal raphe; DRc, dorsal raphe nucleus, caudal division; DRd, dorsal raphe nucleus,
dorsal division; DRif, dorsal raphe nucleus, interfascicular division; DRl, dorsal raphe nucleus, lateral division; DRp, dorsal raphe nucleus, peripheral division; DRv, dorsal
raphe nucleus, ventral division; DT, dorsal thalamus; EW, Edinger-Westphal nucleus; f, fornix; fr, fasciculus retroflexus; GC, periaqueductal grey matter; GLD, dorsal lateral
geniculate nucleus; GP, globus pallidus; Hb, habenular nuclei; Hip, hippocampus; Hyp, hypothalamus; Hyp.d, dorsal hypothalamic cholinergic nucleus; Hyp.l, lateral
hypothalamic cholinergic nucleus; Hyp.v, ventral hypothalamic cholinergic nucleus; IC, inferior colliculus; ic, internal capsule; icp, inferior cerebellar peduncle; IGL,
intergeniculate leaflet; io, inferior olivary nuclei; IP, interpeduncular nucleus; LDT, laterodorsal tegmental nucleus; LOT, lateral olfactory tract; LRT, lateral reticular nucleus;
LV, lateral ventricle; Mc, main cluster of orexinergic immunoreactive neurons; mcp, middle cerebellar peduncle; MnR, median raphe nucleus; N.Acc, nucleus accumbens;
N.Bas, nucleus basalis; NEO, neocortex; OC, optic chiasm; OT, optic tract; OTc, optict tract cluster of orexinergic immunoreactive neurons; P, putamen; PV, paraventricular
nucleus; PBg, parabigeminal nucleus; PIR, piriform cortex; PPT, pedunculopontine nucleus; Pta, pretectal area; py, pyramidal tract; pyx, decussation of the pyramidal tract; R,
thalamic reticular nucleus; RMg, raphe magnus nucleus; RN, red nucleus; ROb, raphe obscurus nucleus; RPa, raphe pallidus nucleus; RVL, rostral ventrolateral serotonergic
group; S, septum; SC, superior colliculus; SpV, spinal trigeminal tract; TOL, olfactory tubercle; vh, ventral horn of spinal cord; VPO, ventral pontine nucleus; ZI, zona incerta;
ZIc, zona incerta cluster of orexinergic immunoreactive neurons.
* Corresponding author. Tel.: +27 11 717 2497; fax: +27 11 717 2422.
E-mail address: Paul.Manger@wits.ac.za (P.R. Manger).
0891-0618/$ – see front matter  2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jchemneu.2010.11.005Article history:
Received 9 October 2010
Received in revised form 21 November 2010
The present study describes
in relation to the previously
the brain of the rock hyraescribed catecholaminergic, cholinergic and serotonergic systems within
Procavia capensis. Adult female rock hyrax brains were sectioned and
N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96 871. Introduction
Orexin or hypocretin is a hypothalamic neuropeptide that
consists of 130–131 amino acids in mammals and was first
described in1998by two independent research groups. The Sutcliffe
group called it hypocretin because it shared similar sequencing
patterns to the secretin group of peptides and because of its
hypothalamic localization (De Lecea et al., 1998; Kukkonen et al.,
2002). The 130 amino acid precursor peptide they identifiedbecame
known as preprohypocretin with hypocretin 1 and hypocretin 2
being the final peptides. The term orexin was coined by the
Yanagisawa group who discovered that this peptide was linked to
food intake (Sakurai et al., 1998; Kukkonen et al., 2002) and noted
that injecting this peptide into the lateral ventricleof non-fasted rats
would increase food intake. The precursor peptide thus became
known as preproorexin with orexin-A and orexin B being the final
peptides (Sakurai et al., 1998; Kukkonen et al., 2002).
Studies on several mammalian species have reported the
location and distribution of orexinergic neurons and terminal
networks (e.g. humans, Moore et al., 2001; cat,Wagner et al., 2000;
Zhang et al., 2001, 2002; sheep, Iqbal et al., 2001; a variety of
rodents, e.g. Broberger et al., 1998; Peyron et al., 1998; Chen et al.,
1999; Cutler et al., 1999; Date et al., 1999; Nixon and Smale, 2007;
microchiropterans, Kruger et al., 2010; and kangaroo, Yamamoto
et al., 2006). These studies all concur regarding the location of the
orexinergic neurons and the general distribution of their major
terminal networks. Orexinergic neurons have only been observed
within the hypothalamuswith themajority of the neurons found in
the lateral hypothalamic area and the perifornical region; however,
other studies have also reported isolated neurons in the region of
the median eminence, posterior, anterior, dorsal and dorsomedial
hypothalamus, zona incerta and ventrolateral hypothalamus. In
contrast to the restricted neuronal distribution, orexinergic fibres
are widely distributed throughout the brain and spinal cord and
exhibit different densities of terminal networks in different regions
of the brain, for example a high density network is evident in the
areas immediately surrounding the third ventricle and areas
involved in the regulation of the sleep–wake cycle (Chen et al.,
1999; Cutler et al., 1999; Hagan et al., 1999; Wagner et al., 2000;
McGranaghan and Piggins, 2001; Zhang et al., 2002, 2004; Espana
et al., 2005; Kruger et al., 2010). Orexinergic fibres have been
reported as being both smooth and varicose, with the varicose type
predominating (Peyron et al., 1998; Chen et al., 1999; Cutler et al.,
1999; Kukkonen et al., 2002). Vasoactive intestinal peptide,
vasopressin and neuropeptide Y terminal networks provide the
main innervation to orexinergic neuronswhile amultitude of other
neuronal circuits receive orexinergic efferents (Peyron et al., 1998;
Cutler et al., 1999; Date et al., 1999; Horvath et al., 1999; Nambu
et al., 1999; Abrahamson et al., 2001; Backberg et al., 2002;
Kukkonen et al., 2002). Studies performed on rats to measure the
release of orexin have revealed that release is closely correlated to
circadian activity. These studies found that orexin concentrations
in the hypothalamus were maximal around light onset and
minimal at the point of light offset. Diurnal variation was also
noted with similar measurements taken from the pons and
intercisternal space; however, the precise time points were shifted
compared to the hypothalamus (Taheri et al., 2000; Fujiki et al.,
2001; Yoshida et al., 2001; Kukkonen et al., 2002).
The diurnal variation of orexin release has an effect on food
consumption, thus when orexin is released in higher concentra-
tions (i.e. during the morning) food intake is promoted, the
opposite being true of low orexin release (i.e. during the night)
(Cutler et al., 1999; Hagan et al., 1999; McGranaghan and Piggins,
2001; Baldo et al., 2003; Khorooshi and Klingenspor, 2005). Orexin
regulates sleep by exciting neurons that are involved in the sleep–
wake cycle, such as those of the locus coeruleus, dorsal raphe, andpontine cholinergic nuclei. This excitation causes the promotion of
wakefulness and reduces sleep, especially episodes of REM (Hagan
et al., 1999; Bourgin et al., 2000; Piper et al., 2000; Wagner et al.,
2000; Espana et al., 2001; Huang et al., 2001; McGranaghan and
Piggins, 2001; Methippara et al., 2000; Kukkonen et al., 2002;
Yamanaka et al., 2002; Zhang et al., 2004; Lee et al., 2005).
Although many studies deal with the location of orexinergic
neurons and terminal network distribution, very few studies
actually investigate the relation and interplay of these neurons and
terminal networks to other neural systems. The aim of the present
study is, thus, to describe the location and distribution of
orexinergic neurons and terminal networks in the brain of the
rock hyrax, Procavia capensis, based on orexin-A immunocyto-
chemistry and to compare these to the known cholinergic,
catecholaminergic and serotonergic systems (Gravett et al.,
2009). This may shed light on the possible relation and interplay
between the orexinergic system and these neural systems.
Currently, the majority of studies reporting on orexin have made
use of nocturnal laboratory rodents. Thus, the rock hyraxmakes for
an interesting candidate for the study of the orexinergic system as
these mammals are part of the unusual and morphologically
diverse Afrotheria (Tabuce et al., 2008), for which no report of the
orexinergic system in available. In addition to this the hyraxes are
diurnally active and their sleep has been described as being
polycyclic (Snyder, 1974). Moreover, they are poor thermoregu-
lators (McNairn and Fairall, 1984; Brown and Downs, 2006) and
orexins are clearly involved in thermoregulation and energy
balance (Szekely et al., 2010; Teske et al., 2010).
2. Materials and methods
The brains of six adult female rock hyraxes, P. capensis, were used in the present
study.Permits fromtheLimpopoandGautengProvincialGovernmentswereobtained
for the capture and transport of the animals from the wild. All animals were treated
and used according to the guidelines of the University of the Witwatersrand Animal
Ethics Committee, which parallel those of the NIH for the care and use of animals in
scientific experimentation. Each animal was weighed, deeply anaesthetized and
subsequently euthanized with weight appropriate doses of sodium pentobarbital
(200 mg sodium pentobarbital/kg, i.p.). Upon cessation of respiration the animals
were perfused intracardially with 0.9% saline followed by 4% paraformaldehyde in
0.1 M phosphate buffer (PB) (pH 7.4, approximately 1 l/kg of each solution), both
solutions having a temperature of 4 8C. The brains were then carefully removed from
the skulls and post-fixed overnight in 4% paraformaldehyde in 0.1 M PB followed by
equilibration in 30% sucrose in 0.1 M PB. The brains were then frozen in dry ice and
with the aid of a freezingmicrotome sectioned at 50mmin the coronal plane. A one in
three series of stains was made for Nissl, myelin, and orexin-A. Sections kept for the
Nissl seriesweremounted on 0.5% gelatine coated glass slides, cleared in a solution of
1:1 chloroform and absolute alcohol, then stained with 1% cresyl violet to reveal cell
bodies. Sectionsused formyelinstainingwerestored in5% formalin foraperiodof two
weeks and were then mounted on 1% gelatine coated glass slides and subsequently
stained with silver solution to reveal myelin sheaths (Gallyas, 1979).
For immunohistochemical staining each sectionwas treatedwith an endogenous
inhibitor peroxidase (49.2%methanol: 49.2% 0.1 M PB: 1.6% of 30% H2O2) for 30 min
and subsequently subjected to three 10 min 0.1 M PB rinses. The sections were then
preincubated in a solution (blocking buffer) consisting of 3% normal goat serum
(NGS, Chemicon), 2% bovine serum albumin (BSA, Sigma) and 0.25% Triton X100
(Merck) in 0.1 M PB, at room temperature for 2 h. This was followed by three 10 min
rinses in 0.1 M PB. The sections were then placed for 48 h at 4 8C under constant
gentle shaking, in primary antibody solution, that contained the appropriately
diluted primary antibody in blocking buffer (see above). The primary antibody used
was anti-orexin-A (AB 3704, Millipore, raised in rabbit, dilution 1:1500). This was
followed by another three 10 min rinses in 0.1 M PB, after which the sections were
incubated for 2 h at room temperature in secondary antibody solution. The
secondary antibody solution contained a 1:1000 dilution of biotinylated anti-rabbit
IgG (BA-1000, Vector Labs) in 3% normal goat serum (NGS), and 2% bovine serum
albumin (BSA) in 0.1 M PB. Once this was completed, the sections were again
subjected to another three 10 min rinses in 0.1 M PB, followed by a 1 h incubation in
avidin-biotin solution (Vector Labs) and again rinsed. This was followed by a 5 min
treatment of the sections in a solution consisting of 0.05% diaminobenzidine
tetrahydrochloride (DAB) in 0.1 M PB, after which, and while still in the same
solution, 3ml of 30% H2O2 per 0.5 ml of solution was added. With the aid of a low
power stereomicroscope the progression of the staining was visually followed and
allowed to continue until a level was reached where the background staining could
assist in reconstruction without obscuring the immunopositive structures. Once
this level was reached the reaction was stopped by placing the sections in 0.1 M PB,
followed by a final session of three 10 min rinses in 0.1 M PB. The immunohis-
tochemically stained sections were mounted on 0.5% gelatine coated slides and left
to dry overnight. The mounted sections were dehydrated by placing it in 70%
alcohol for 2 h at room temperature under gentle shaking and then transferred
through a series of graded alcohols, cleared in xylene and coverslipped with Depex.
The sections were observed with a low power stereomicroscope, and the
architectonic borders traced according to the Nissl and myelin stained sections
[()TD$FIG]
tr
t
o
re
N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–9688Fig. 1.Drawings of coronal sections through one half of the brain of the rock hyrax, illus
black dot indicates a single cell body, black and gray shaded areas indicate regions in
terminal networks were observed. Areas where no shading is evident represent regions
the most caudal, and each drawing is approximately 1500 mm apart. See list for abbating orexin-A immunoreactive neurons and terminal network distribution. A single
he brain of the rock hyrax where high and medium density, respectively, orexin-A
f low or no orexinergic innervation. Drawing A represents themost rostral section,V
viations.
[()TD$FIG] N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96 89using a camera lucida. The immunostained sections were then matched to the
drawings and the immunopositive neurons marked, in addition the density of axon
terminal staining was graded from low to high for each immunostained section and
medium and high marked on the drawings. The drawings were scanned and
redrawn with the aid of the Canvas 8 program. Location and distribution of orexin
immunopositive (Orx+) neurons and terminal networks were described in relation
to the general neuroanatomy of the rock hyrax brain and previously described
cholinergic, catecholaminergic, and serotonergic systems (Gravett et al., 2009).
3. Results
The results revealed that the distribution of orexin-A immu-
nopositive (Orx+) neuron cell bodies and terminal networks did
not, for the most part, differ significantly from what has been
described in othermammals. Despite this, an interesting difference
Fig. 1 (Conwas noted that has not been reported in most mammals studied to
date, namely a strong Orx+ terminal network in the anterior dorsal
nucleus of the dorsal thalamus (AD). Orx+ neurons and terminal
networks are described in relation to either the general neuro-
anatomy of the rock hyrax brain and the cholinergic, catechol-
aminergic, and serotonergic systems (as described previously for
this species in Gravett et al., 2009).
3.1. Orexin-A neuronal cell body distribution
Orx+ neuronal cell bodies were identified mainly within the
hypothalamus of the brain of the rock hyrax. These cell bodieswere
grouped into three distinct clusters; the main cluster, the zona
incerta cluster and the optic tract cluster (Figs. 1–3). The main
tinued )
[()TD$FIG] N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–9690cluster was identified as a large group of densely packed Orx+
neuronal cell bodies located lateral to the third ventricle in the
perifornical region, with a large number of neuronal cell bodies
extending medially from this area, as well as into the dorsomedial
and lateral hypothalamic areas. The majority of Orx+ neurons
belonged to this cluster and theywere found to be ovoid in shape, a
mixture of both bipolar and multipolar types and showed no
apparent regular dendritic orientation. The Orx+ neuronal cell
bodies of this main cluster were intermingled with the A11 and
A14 nuclear groups of the catecholaminergic system. From the
main cluster a group of Orx+ neuronal cell bodies extended
laterally into the region of the zona incerta. This cluster had a very
low density of Orx+ neurons that were mixed with neurons of the
lateral hypothalamic cholinergic nucleus and the A13 nucleus of
the catecholaminergic system. The cell bodies were ovoid in shape,
Fig. 1 (Conta mixture of bipolar and multipolar type and had no specific
dendritic orientation. The third cluster extended ventrolaterally
from the main cluster to the ventrolateral region of the
hypothalamus adjacent to the optic tract. This cluster exhibited
a low density of Orx+ neuronal cell bodies that were mostly of the
bipolar type with the same cell morphology as described for the
other clusters. These neurons were found to intermingle with
those of the ventral hypothalamic cholinergic nucleus (Figs. 1E–G
and 2).
3.2. Orexin-A terminal network distribution
3.2.1. Telencephalon
Within the telencephalon of the rock hyrax brain orexinergic
terminal networks of medium density were noted. These terminal
inued ).
[()TD$FIG] N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96 91networks were mainly distributed along the medial and ventro-
lateral border of the telencephalon and located in and around the
following structures: the nucleus accumbens, olfactory tubercle
and islands of Calleja, diagonal band of Broca and the septal nuclear
complex (Figs. 1A–C and 4). It is of interest to note than in all these
structures a number of cholinergic neurons were also located
(Gravett et al., 2009). The remaining regions of the telencephalon
had either a low density Orx+ terminal network, or no Orx+
structures could be observed.
3.2.2. Diencephalon
Both high and medium density orexinergic terminal networks
were observed throughout the diencephalon. The high density
terminal networks were located in the anterior hypothalamic area,
the periventricular areas adjacent to the third ventricle in the
hypothalamus (Fig. 2), in the paraventricular nuclei of the
epithalamus (Fig. 5A) extending to a region surrounding the
habenular nuclei, and dorsolaterally in the AD (Figs. 1D–J and 5).
The medium density terminal networks surrounded these areas of
high density and were distributed through most parts of the
hypothalamus (Fig. 2), the midline of the dorsal thalamus in the
intralaminar nuclei and extended laterally to the intergeniculate
leaflet (Fig. 5C). The high-density terminal networks overlapped
with the A12, A14, and A15v nuclei of the catecholaminergic
system, and the AD (Fig. 5B) that forms part of the cholinergic
system of the rock hyrax (Gravett et al., 2009). The Orx+ terminal
networks of medium density were found to overlap the A11 and
A13 nuclei of the catecholaminergic system and the dorsal, lateral,
and ventral hypothalamic nuclei of the cholinergic system. The
Fig. 2. Photomicrograph montage demonstrating the location of orexin-A immunorea
showing the main cluster (Mc), the zona incerta cluster (ZIc) and the optic tract clusteremaining regions of the diencephalon had either a low density
Orx+ terminal network, or no Orx+ structures could be observed
(Fig. 2).
3.2.3. Midbrain and Pons (Mesencephalon and Metencephalon)
Both high and medium dense orexinergic terminal networks
characterized the midbrain and pontine regions; however, the
medium density Orx+ terminal networks were found to domi-
nate in these regions. The terminal networks of high density were
located ventral to the cerebral aqueduct within the midline of the
midbrain and pontine regions and overlapped the A10d and
A10dc nuclei of the catecholaminergic system and most
serotonergic nuclei of the dorsal raphe (Fig. 6A). The medium
dense Orx+ terminal networks were coincident with the
dopaminergic ventral tegmental area, the noradrenergic locus
coeruleus complex (Fig. 6B), the cholinergic nuclei of the
laterodorsal tegemental and pedunculopontine nuclei (both
magno and parvocellular divisions of these nuclei, Gravett
et al., 2009), the cholinergic parabigeminal and Edinger-
Westphal nuclei, and the caudal linear (CLi), supralemniscal
(B9), caudal nucleus of the dorsal raphe (DRc) and median raphe
(MnR) serotonergic nuclei (Fig. 6C). In addition to this, the entire
periaqueductal gray matter was observed to have a medium
density Orx+ terminal network that extended outward through
all layers of the superior colliculus (Figs. 1K–N and 6). The
remaining regions of the midbrain and pons had either a low
density Orx+ terminal network, or no Orx+ structures could be
observed. It is worth noting that we could only find a low density
Orx+ network in the inferior colliculus.
ctive neurons and terminal networks within the hypothalamus of the rock hyrax
r (OTc). Scale = 1 mm. 3V: third ventricle; f: fornix; OT: optic tract.
[()TD$FIG] N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96923.2.4. Medulla oblongata (Myelencephalon)
Only medium and low-density Orx+ terminal networks were
present within the medulla oblongata of the brain of the rock
hyrax. The medium density networks were found in all regions
where serotonergic neurons were located (i.e. the raphe magnus,
rostral and caudal ventrolateral columns, raphe pallidus and raphe
obscurus). Similarly, all regions that contained catecholaminergic
neurons were found to have a medium density Orx+ terminal
network, these being the area postrema, C1, C2, A1 and A2 nuclei.
Anterior in the medulla a medium density Orx+ terminal network
was found in the region of the parabrachial nuclear complex.
Towards the spinal cord these medium density Orx+ networks
were located in the areas surrounding the central canal and the
outer edge of the dorsal horn in the spinal trigeminal tract (SpV)
Fig. 3. Photomicrographs illustrating the neuronal morphology of the three clusters
of orexin-A immunoreactive nuclear groups within the hypothalamus of the rock
hyrax. (A) Orexin-A immunoreactive neurons of the main cluster, perifornical
region. (B) Orexin-A immunoreactive neurons of the optic tract cluster, lateral
ventral hypothalamic area. (C) Orexin-A immunoreactive neurons of the zona
incerta cluster, lateral hypothalamic area. In all images, medial is to the left, lateral
to the right, and dorsal to the top. Scale in C = 500 mm and applies to all.(Fig. 1O–V). The remaining regions of the medulla had either a low
density Orx+ terminal network, or no Orx+ structures could be
observed.
4. Discussion
The aim of the present study was to identify and describe the
neuronal location, nuclear organization and terminal network
distribution of the orexinergic system in the brain of the rock
hyrax, Procavia capensis by means of orexin-A immunocytochem-
istry. The study also examined the topological relationships of the
Orx+ neurons and terminal networks with the previously
described cholinergic, catecholaminergic and serotonergic systems
in the brain of the rock hyrax (Gravett et al., 2009). The results
revealed that the distribution and nuclear organization of the Orx+
neurons of the rock hyrax conforms closely to what has been
described previously for other mammals (Broberger et al., 1998;
Peyron et al., 1998; Moore et al., 2001; Zhang et al., 2001, 2002;
Nixon and Smale, 2007; Yamamoto et al., 2006; Kruger et al., 2010).
The orexin-A immunoreactive terminal network distribution also
showed a similar architecture as described in other mammalian
species; however the anterodorsal nucleus of the thalamus was
observed to have a dense orexinergic terminal network. Thus, the
orexinergic system, as demonstrated with orexin-A immunoreac-
tivity, within the rock hyrax brain is similar to what has been
described in other mammals, but there is a feature of the terminal
network distribution that has only been reported in a small
number of other mammals.
4.1. Orexinergic neuronal distribution
The location of orexinergic neuronal cell bodies has been
described in representative species of all vertebrate classes –
mammals (e.g. Wagner et al., 2000; Iqbal et al., 2001; Moore et al.,
2001; Yamamoto et al., 2006; Nixon and Smale, 2007; Kruger et al.,
2010; Stoyanova et al., 2010), birds (e.g. Ohkubo et al., 2002),
reptiles (Domı´nguez et al., 2010), amphibians (e.g. Shibahara et al.,
1999; Galas et al., 2001; Singletary et al., 2005; Suzuki et al., 2008;
Lo´pez et al., 2009) and fish (e.g. Kaslin et al., 2004; Huesa et al.,
2005; Yokogawa et al., 2007). All these studies have revealed that
for the most part orexinergic neurons were located within the
hypothalamus. It is clear from previous studies that this system
shows a degree of variability within and between different
vertebrate classes, but also a strong degree of similarity (e.g.
Suzuki et al., 2008; Domı´nguez et al., 2010). In mammals the
majority of orexinergic neurons are located within the lateral
hypothalamic area and the perifornical region (which we term the
main cluster, Kruger et al., 2010); however, other studies in
mammals have also reported isolated neurons in the region of the
median eminence, posterior, dorsal and dorsomedial hypothala-
mus (Chen et al., 1999; Cutler et al., 1999; Hagan et al., 1999;
Wagner et al., 2000; McGranaghan and Piggins, 2001; Zhang et al.,
2002, 2004; Espana et al., 2005). In the rock hyrax we could
subdivide the orexinergic neurons into three distinct clusters: the
main cluster, the zona incerta cluster and the optic tract cluster.
While these clusters have been reported present inmostmammals
studied to date it has been found that microchiropterans and
hamsters lack the optic tract cluster (Wagner et al., 2000;
McGranaghan and Piggins, 2001; Kruger et al., 2010). The nuclear
organization of the orexinergic neurons in the rock hyrax is
therefore what may be described as very typically mammalian. It
would seem that, apart from minor differences, the nuclear
organization of the orexinergic neurons is strongly conserved
across mammals. An important aspect of orexinergic nuclear
parcellation relates to the connectivity of this system. For example,
the zona incerta cluster specifically projects to the intergeniculate
[()TD$FIG] N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96 93leaflet (Vidal et al., 2005) and what we term the main orexinergic
cluster may be subdivided into perifornical and lateral hypotha-
lamic nuclei (Yoshida et al., 2006). Thus, in the hyrax we may be
observing 4 specific nuclei within this system, however, further
study on the specific connectivity of this systemwould be required
to justify further nuclear parcellation than that presented here.
In conjunction with neuronal location and distribution, the
present study also examined the topographical interrelationships
of the orexinergic neurons with the previously described
cholinergic, catecholaminergic and serotonergic systems in the
brain of the rock hyrax (Gravett et al., 2009). The observations
revealed that orexinergic neurons of the main cluster were
topographically coincident with the A11 and A14 nuclear groups
of the catecholaminergic system; the zona incerta cluster was
topographically coincident with the A13 nucleus of the catechol-
aminergic system and the lateral hypothalamic nucleus of the
cholinergic system; the optic tract cluster was topographically
coincident with the cholinergic ventral hypothalamic nucleus. It
has been reported that the membrane potential of orexinergic
neurons are hyperpolarized by noradrenaline and serotonin and
depolarized by acetylcholine (Yamanaka et al., 2003). Thus,
orexinergic neurons are possibly rapidly depolarized where they
intermingle with neurons from the cholinergic system. The
potential intrinsic hypothalamic circuitry of these systems, due
to the topological congruency, would be an avenue of interesting
further study given the roles these systems play in the sleep–wake
cycle, satiety and motivational states.
Fig. 4. Photomicrographs illustrating the distribution of orexin-A immunoreactive termin
orexin-A immunoreactive terminal networks within the cerebral cortex. (B) Medium den
(Diag. B). (C) Medium density orexin-A immunoreactive terminal networks within the n
networks within the olfactory tubercle (TOL). Scale in D = 500mm and applies to all. M4.2. Orexinergic terminal network distribution
The orexinergic terminal network distribution within the brain
of the rock hyrax was found to be similar to that described in other
mammalian species studied to date, especially in the manner that
these terminal networks are found primarily along themidline and
ventricular surfaces; however, the anterodorsal nucleus of the
dorsal thalamus (AD) of the rock hyrax exhibited a high-density
orexinergic terminal network which is an apparent variance to
most other mammals (see below).
Themajority of the mammalian brain is in receipt of either very
low or no orexinergic innervation, and this pattern is evident in the
cerebral cortex, striatopallidal complex, hippocampal complex,
amygdalar complex, the majority of the dorsal thalamus,
brainstem and cerebellum of the rock hyrax brain. Medium dense
orexinergic terminal networkswere observedwithin regions of the
telencephalon, diencephalon, midbrain and the medulla of the
brain of the rock hyrax, whereas a high density orexinergic
innervation was only evident in certain regions of the diencepha-
lon and midbrain. This distribution of medium to high dense
orexinergic terminal networks once again resembled the typical
mammalian organizational plan for this system as reported in
previous studies. Throughout the telencephalon, diencephalon,
midbrain and medullary regions the orexinergic terminal net-
works were found intermingled with several nuclear complexes of
the catecholaminergic, cholinergic and serotonergic systems
(Gravett et al., 2009).
al networks within certain parts of the forebrain of the rock hyrax. (A) Low density
sity orexin-A immunoreactive terminal networks within the diagonal band of Broca
ucleus accumbens (N. Acc). (D) Medium density orexin-A immunoreactive terminal
edial is to the left in all images.
[()TD$FIG] [()TD$FIG]N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–9694The regions within the telencephalon that exhibited a medium
density of orexinergic terminal networks included the nucleus
accumbens, the olfactory tubercle, the diagonal band of Broca, and
the septal nuclear complex. This distribution of medium density
orexinergic terminal networks within the telencephalon of the
rock hyrax coincide with previous reports for this system in other
mammals. Within the diencephalon a medium density of
orexinergic innervation was observed in the region of the
intergeniculate leaflet, whereas a high density was observed in
Fig. 5. Photomicrograph illustrating orexin-A immunoreactive terminal networks
within the diencephalon of the rock hyrax. (A) High density orexin-A
immunoreactive terminal networks within the paraventricular thalamic nucleus
(PV) and the lateral habenular nuclei along the border of the stria medullaris (Hb).
Scale bar = 1 mm. (B) High density orexin-A immunoreactive terminal networks
within the anterodorsal nucleus of the dorsal thalamus (AD). (C) Medium density
orexin-A immunoreactive terminal networks within the intergeniculate leaflet
(IGL) mainly oriented in the ventro-dorsal direction. Scale in C = 500mm and
applies toB and C.AV: anteroventral nucleus of the dorsal thalamus.Medial is to the
left in all images.regions immediately surrounding the third ventricle, the midline
of the dorsal thalamus as well as the AD nucleus (see below). The
medium density orexinergic terminal networks overlapped the
medial septal nucleus, dorsal, lateral and ventral hypothalamic
nuclei of the cholinergic system and the A11 and A13 nuclear
complexes of the catecholaminergic system. The distribution of
medium to high density orexinergic terminal networks in the
Fig. 6. Photomicrograph illustrating the distribution of orexin-A immunoreactive
terminal networks within the midbrain and pontine regions of the rock hyrax. (A)
High density orexin-A immunoreactive terminal networks within the serotonergic
dorsal raphe (DR) nuclear complex. Scale bar = 1 mm. (B) Medium density orexin-A
immunoreactive terminal networkswithin the locus coeruleus nuclear complex. (C)
Medium density orexin-A immunoreactive terminal networks within the
serotonergic caudal linear nucleus (CLi) and supralemniscal nucleus (B9). Scale
in C = 500 mm and applies to B and C. A6: locus coeruleus; A7d: subcoeruleus
diffuse region; A7sc, subcoeruleus compact region; ca: cerebral aqueduct; DRd:
dorsal division of dorsal raphe; DRl: lateral division of dorsal raphe; DRp:
peripheral division of dorsal raphe; DRv: ventral division of dorsal raphe. Medial is
to the left in all images.
N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–96 95regions of the diencephalon of the rock hyrax brain did not differ
significantly to published reports in othermammals for this region.
However, it did differ to other mammals with respect to the
orexinergic innervation of the AD nucleus, which has only been
previously reported to occur in the Nile grass rat, Arvicanthis
niloticus (Novak and Albers, 2002; Nixon and Smale, 2007) and the
Syrian hamster, Mesocricetus auratus (Mintz et al., 2001). A
medium density orexinergic terminal network distribution domi-
nated the midbrain and pontine regions in the rock hyrax brain,
however high density terminal networkswere observedwithin the
dopaminergic ventral tegmental area (VTA) and serotonergic
dorsal raphe nuclei. A study byNakamura et al. (2000) has reported
that the calcium concentration in isolated A10 neurons is increased
by orexin-A and this in turn induces hyperlocomotion, stereotypy
and grooming in rats. Orexin-A has further been reported to have
an excitatory effect on the serotonergic neurons of the dorsal raphe
nucleus, which play an important role in the sleep–wake cycle as
well as regulation of food intake and mood control (Brown et al.,
2001; Matsuzaki et al., 2002; Tao et al., 2006). A moderate density
of orexinergic terminal networks was identified in the noradren-
ergic locus coeruleus and subcoeruleus, and cholinergic later-
odorsal tegmental (LDT) and pedunculopontine tegmental (PPT)
nuclei. These structures are known to play a crucial part in the
sleep–wake cycle and have been reported to be activated by orexin.
Orexinergic input to the cells of the locus coeruleus facilitate the
waking state, whereas orexinergic innervation to the LDT and PPT
play a role in the regulation of REM sleep (Ohno and Sakurai, 2008).
Finally, the medullary region exhibited a medium dense orex-
inergic terminal network distribution in the areas ventral to the
fourth ventricle and regions surrounding the central canal,
whereas the remainder of this region was characterized by a
low or absent orexinergic terminal network distribution. These
terminal networks furthermore overlapped the RPa, RVL, RMg, ROb
and CVL nuclei of the serotonergic system, the C1-2 and A1-2
nuclei of the catecholaminergic system, the nucleus ambiguus, the
dorsal motor vagus and hypoglossal nuclei of the cholinergic
system. Thus, for the most part the orexinergic terminal network
distribution within the brain of the rock hyrax typically resembles
the general mammalian organizational plan for this system.
4.3. Orexinergic innervation of the anterodorsal nucleus of the dorsal
thalamus and diurnality?
The majority of studies that describe the distribution of
orexinergic terminal networks have examined nocturnal mam-
mals. None of these studies reported either a dense or moderate
orexinergic innervation of the AD. Two previous reports of
moderate to dense orexinergic innervation of the AD examined
the Nile grass rat, Arvicanthis niloticus (Novak and Albers, 2002;
Nixon and Smale, 2007) and the Syrian hamster, Mesocricetus
auratus (Mintz et al., 2001). Novak and Albers (2002) reported a
low density of orexinergic fibres in the AD nucleus and a high
density in the anteroventral (AV) nucleus of the dorsal thalamus of
the grass rat when the orexin B antibody was used, while Nixon
and Smale (2007) reported a low density for the AV nucleus with
the orexin-A antibody and a medium density in the AD nucleus
with both orexin-A and B antibodies. The rock hyrax and the Syrian
hamster (Mintz et al., 2001) both showed a high density of
orexinergic terminals within the AD nucleus with an orexin-A
antibody.
Of potential functional correlation to this is the fact that the
rock hyrax (Olds and Shoshani, 1982), Nile grass rat (Novak et al.,
2000) and Syrian hamster (Gatterman et al., 2008) are all diurnal
species. Thus, it is possible that orexinergic innervation of the AD
nucleus, the AV nucleus, or the anterior nuclear complex as a
whole, might be a unique feature associated with diurnalmammals. This possibility needs to be tested across a range of
mammalian species to determine whether this potentially very
interesting finding is correct.
References
Abrahamson, E.E., Leak, R.K., Moore, R.Y., 2001. The suprachiasmatic nucleus
projects to posterior hypothalamic arousal systems. Neuroreport 12, 435–440.
Backberg, M., Hervieu, G., Wilson, S., Meister, B., 2002. Orexin receptor-1 (OX-R1)
immunoreactivity in chemically identified neurons of the hypothalamus: focus
on orexin targets involved in control of food and water intake. Eur. J. Neurosci.
15, 315–328.
Baldo, B.A., Daniel, R.A., Berridge, C.W., Kelley, A.E., 2003. Overlapping distributions
of orexin/hypocretin- and dopamine-b-hydroxylase immunoreactive fibers in
rat brain regionsmediating arousal, motivation and stress. J. Comp. Neurol. 464,
220–237.
Bourgin, P., Huitron-Resendiz, S., Spier, A.D., Fabre, V., Morte, B., Criado, J.R.,
Sutcliffe, J.G., Henriksen, S.J., De Lecea, L., 2000. Hypocretin-1 modulates rapid
eye movement sleep through activation of locus coeruleus neurons. J. Neurosci.
20, 7760–7765.
Broberger, C., De Lecea, L., Sutcliffe, J.G., Hokfelt, T., 1998. Hypocretin/orexin-and
melanin-concentrating hormone-expressing cells form distinct populations in
the rodent lateral hypothalamus: relationship to the neuropeptide Y and agouti
gene-related protein systems. J. Comp. Neurol. 402, 460–474.
Brown, K.J., Downs, C.T., 2006. Seasonal patterns in body temperature of free-living
hyrax (Procavia capensis). Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 143,
42–49.
Brown, R.E., Sergeeva, O., Eriksson, K.S., Haas, H.L., 2001. Orexin-A excites seroto-
nergic neurons in the dorsal raphe nucleus of the rat. Neuropharmacology 40,
457–459.
Chen, C.T., Dun, S.L., Kwok, E.H., Dun, N.J., Chang, J.K., 1999. Orexin-A-like immu-
noreactivity in the rat brain. Neurosci. Lett. 260, 161–164.
Cutler, D.J., Morris, R., Sheridhar, V., Wattam, T.A., Holmes, S., Patel, S., Arch, J.R.,
Wilson, S., Buckingham, R.E., Evans, M.L., Leslie, R.A., Williams, G., 1999.
Differential distribution of orexin-A and orexin-B immunoreactivity in the
rat brain and spinal cord. Peptides 20, 1455–1470.
Date, Y., Ueta, Y., Yamashita, H., Yamaguchi, H., Matsukura, S., Kangawa, K., Sakurai,
T., Yanagisawa, M., Nakazato, M., 1999. Orexins, orexigenic hypothalamic
peptides, interact with autonomic, neuroendocrine and neuroregulatory sys-
tems. Proc. Natl. Acad. Sci. U.S.A. 96, 748–753.
De Lecea, L., Kilduff, T.S., Peyron, C., Gao, X., Foye, P.E., Danielson, P.E., Fukuhara, C.,
Battenberg, E.L., Gautvik, V.T., Bartlett, F.S., Frankel, W.N., Van den Pol, A.N.,
Bloom, F.E., Gautvik, K.M., Sutcliffe, J.G., 1998. The hypocretins: hypothalamus-
specific peptides with neuroexcitatory activity. Proc. Natl. Acad. Sci. U.S.A. 95,
322–327.
Domı´nguez, L., Morona, R., Joven, A., Gonza´lez, A., Lo´pez, J.M., 2010. Immunohisto-
chemical localization of orexins (hypocretins) in the brain of reptiles and its
relation to monoaminergic systems. J. Chem. Neuroanat. 39, 20–34.
Espana, R.A., Baldo, B.A., Kelley, A.E., Berridge, C.W., 2001. Wake-promoting and
sleep-suppressing actions of hypocretin (orexin): basal forebrain sites of action.
Neuroscience 106, 699–715.
Espana, R.A., Reis, K.M., Valentino, R.J., Berridge, C.W., 2005. Organization of
hypocretin/orexin efferents to locus coeruleus and basal forebrain arousal-
related structures. J. Comp. Neurol. 481, 160–178.
Fujiki, N., Yoshida, Y., Ripley, B., Honda, K., Mignot, E., Nishino, S., 2001. Changes in
CSF hypocretin-1 (orexin-A) levels in rats across 24 hours and in response to
food deprivation. Neuroreport 12, 993–997.
Galas, L., Vaudry, H., Braun, B., Van den Pol, A.N., De Lecea, L., Sutcliffe, J.G., Chartrel,
N., 2001. Immunohistochemical localization and biochemical characterization
of hypocretin/orexin-related peptides in the central nervous system of the frog
Rana ridibunda. J. Comp. Neurol. 429, 242–252.
Gallyas, F., 1979. Silver staining of myelin by means of physical development.
Neurol. Res. 1, 203–209.
Gatterman, R., Johnston, R.E., Yigit, N., Fritzsche, P., Larimer, S., Ozkurt, S., Neumann,
K., Song, Z., Colak, E., Johnston, J., McPhee, M.E., 2008. Golden hamsters are
nocturnal in captivity but diurnal in nature. Biol. Lett. 4, 253–255.
Gravett, N., Bhagwandin, A., Fuxe, K., Manger, P.R., 2009. Nuclear organization and
morphology of cholinergic, putative catecholaminergic and serotonergic neu-
rons in the brain of the rock hyrax, Procavia capensis. J. Chem. Neuroanat. 38,
57–74.
Hagan, J.J., Leslie, R.A., Patel, S., Evans, M.L., Wattam, T.A., Holmes, S., Benham, C.D.,
Taylor, S.G., Routledge, C., Hemmati, P., Munton, R.P., Ashmeade, T.E., Shah, A.S.,
Hatcher, J.P., Hatcher, P.D., Jones, D.N., Smith, M.I., Piper, D.C., Hunter, A.J.,
Porter, R.A., Upton, N., 1999. Orexin-A activates locus coeruleus cell firing and
increases arousal in the rat. Proc. Natl. Acad. Sci. U.S.A. 96, 10911–10916.
Horvath, T.L., Diano, S., Van den Pol, A.N., 1999. Synaptic interaction between
hypocretin (orexin) and neuropeptide Y cells in the rodent and primate hypo-
thalamus: a novel circuit implicated in metabolic and endocrine regulations. J.
Neurosci. 19, 1072–1087.
Huang, Z.L., Qu, W.M., Li, W.D., Mochizuki, T., Eguchi, N., Watanabe, T., Urade, Y.,
Hayaishi, O., 2001. Arousal effect of orexin-A depends on activation of the
histaminergic system. Proc. Natl. Acad. Sci. U.S.A. 98, 9965–9970.
Huesa, G., Van den Pol, A.N., Finger, T.E., 2005. Differential distribution of hypocretin
(orexin) and melanin-concentrating hormone in the goldfish brain. J. Comp.
Neurol. 488, 476–491.
Iqbal, J., Pompolo, S., Sakurai, T., Clarke, I.J., 2001. Evidence that orexin-containing
neurones provide direct input to gonadotropin-releasing hormone neurones in
the ovine hypothalamus. J. Neuroendocrinol. 13, 1033–1041.
Kaslin, J., Nystedt, J.M., O¨sterga˚rd, M., Peitsaro, N., Panula, P., 2004. The orexin/
hypocretin system in zebrafish is connected to the aminergic and cholinergic
systems. J. Neurosci. 24, 2678–2689.
Khorooshi, R.M., Klingenspor, M., 2005. Neuronal distribution of melanin- concen-
trating hormone, cocaine- and amphetamineregulated transcript and orexin B
in the brain of the Djungarian hamster (Phodopus sungorus). J. Chem. Neuroanat.
29, 137–148.
Kruger, J.-L., Dell, L.-A., Pettigrew, J.D., Manger, P.R., 2010. Cellular location and
major terminal networks of the orexinergic system in the brains of five
microchiropteran species. J. Chem. Neuroanat. 40, 256–262.
Kukkonen, J.P., Holmqvist, T., Ammoun, S., A˚kermam, K.E.O., 2002. Functions of the
orexinergic/hypocretinergic system. Am. J. Physiol. Cell Physiol. 283, C1567–
C1591.
Lee, M.E., Hassani, O.K., Jones, B.E., 2005. Discharge of identified orexin/hypocretin
neurons across the sleep–wake cycle. J. Neurosci. 25, 6716–6720.
Haynes, A.C., Carr, S.A., Annan, R.S., McNulty, D.E., Liu, W.S., Terrett, J.A.,
Elshourbagy, N.A., Bergsma, D.J., Yanagisawa, M., 1998. Orexins and orexin
receptors: a family of hypothalamic neuropeptides and G protein-coupled
receptors that regulate feeding behavior. Cell 92, 573–585.
Shibahara, M., Sakurai, T., Nambu, T., Takenouchi, T., Iwaasa, H., Egashira, S., Ihara,
M., Goto, K., 1999. Structure, tissue distribution, and pharmacological charac-
terization of Xenopus orexins. Peptides 20, 1169–1176.
Singletary, K.G., Delville, Y., Farrell, W.J., Wilczynski, W., 2005. Distribution of
orexin/hypocretin immunoreactivity in the nervous system of the green tree-
frog, Hyla cinerea. Brain Res. 1041, 231–236.
Snyder, F., 1974. Sleep–waking patterns of Hyracoidea. Sleep Res. 3, 87.
Stoyanova, I.I., Rutten, W.L., le Feber, J., 2010. Orexin-A and orexin-B during the
postnatal development of the rat brain. Cell Mol. Neurobiol. 30, 81–89.
Suzuki, H., Kubo, Y., Yamamoto, T., 2008. Orexin-A immunoreactive cells and fibers
in the central nervous system of the axolotl brain and their association with
tyrosine hydroxylase and serotonin immunoreactive somata. J. Chem. Neuroa-
nat. 35, 295–305.
Szekely, M., Petervari, E., Balasko, M., 2010. Thermoregulation, energy balance,
regulatory peptides: recent developments. Front. Biosci. 2, 1009–1046.
Tabuce, R., Asher, R.J., Lehmann, T., 2008. Afrotherianmammals: a review of current
N. Gravett et al. / Journal of Chemical Neuroanatomy 41 (2011) 86–9696Lo´pez, J.M., Domı´nguez, L., Morena, N., Gonza´lez, A., 2009. Comparative immuno-
histochemical analysis of the distribution of orexins (hypocretins) in the brain
of amphibians. Peptides 30, 873–887.
Matsuzaki, I., Sakurai, T., Kunii, K., Nakamura, T., Yanagisawa, M., Goto, K., 2002.
Involvement of the serotonergic system in orexin-induced behavioral altera-
tions in rat. Regul. Pept. 104, 119–123.
McGranaghan, P.A., Piggins, H.D., 2001. Orexin-A-like immunoreactivity in the
hypothalamus and thalamus of the Syrian hamster (Mesocricetus auratus)
and Siberian hamster (Phodopus sungorus), with special reference to circadian
structures. Brain Res. 904, 234–244.
McNairn, I.S., Fairall, N., 1984. Metabolic rate and body temperature of adult and
juvenile hyrax (Procavia capensis). Comp. Biochem. Physiol. A. Comp. Physiol.
79, 539–545.
Methippara, M.M., Alam, M.N., Szymusiak, R., McGinty, D., 2000. Effects of lateral
preoptic area application of orexin-A on sleep–wakefulness. Neuroreport 11,
3423–3426.
Mintz, E.M., Van den Pol, A.N., Casano, A.A., Albers, H.E., 2001. Distribution of
hypocretin-(orexin) immunoreactivity in the central nervous system of Syrian
hamsters (Mesocricetus auratus). J. Chem. Neuroanat. 21, 225–238.
Moore, R.Y., Abrahamson, E.A., Van den Pol, A., 2001. The hypocretin neuron system:
an arousal system in the human brain. Arch. Ital. Biol. 139, 195–205.
Nakamura, T., Uramura, K., Nambu, T., Yada, T., Goto, K., Yanagisawa, M., Sakurai, T.,
2000. Orexin-induced hyperlocomotion and stereotypy are mediated by the
dopaminergic system. Brain Res. 873, 181–187.
Nambu, T., Sakurai, T., Mizukami, K., Hosoya, Y., Yanagisawa, M., Goto, K., 1999.
Distribution of orexin neurons in the adult rat brain. Brain Res. 827, 243–260.
Nixon, J.P., Smale, L., 2007. A comparative analysis of the distribution of immuno-
reactive orexin-A and B in the brains of nocturnal and diurnal rodents. Behav.
Brain Funct. 13, 28.
Novak, C.M., Albers, H.E., 2002. Localization of hypocretin-like immunoreactivity in
the brain of the diurnal rodent Arvicanthis niloticus. J. Chem. Neuroanat. 23, 49–
58.
Novak, C.M., Smale, L., Nunez, A.A., 2000. Rhythms in Fos expression in brain areas
related to the sleep–wake cycle in the diurnal Arvicanthis niloticus. Am. J.
Physiol. Regul. Integr. Comp. Physiol. 278, R1267–R1274.
Ohkubo, T., Boswell, T., Lumineau, S., 2002. Molecular cloning of chicken prepro-
orexin cDNA and preferential expression in the chicken hypothalamus. Bio-
chim. Biophys. Acta 1577, 476–480.
Ohno, K., Sakurai, T., 2008. Orexin neuronal circuitry: role in the regulation of sleep
and wakefulness. Front. Neuroendocrinol. 29, 70–87.
Olds, N., Shoshani, J., 1982. Procavia capensis. Mammalian Species 171, 107.
Peyron, C., Tighe, D.K., Van den Pol, A.N., De Lecea, L., Heller, H.C., Sutcliffe, J.G.,
Kilduff, T.S., 1998. Neurons containing hypocretin (orexin) project to multiple
neuronal systems. J. Neurosci. 18, 9996–10015.
Piper, D.C., Upton, N., Smith, M.I., Hunter, A.J., 2000. The novel brain neuropeptide,
orexin-A, modulates the sleep–wake cycle of rats. Eur. J. Neurosci. 12, 726–730.
Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R.M., Tanaka, H.,Williams,
S.C., Richardson, J.A., Kozlowski, G.P., Wilson, S., Arch, J.R., Buckingham, R.E.,data. Mammalia 72, 2–14.
Taheri, S., Sunter, D., Dakin, C., Moyes, S., Seal, L., Gardiner, J., Rossi, M., Ghatei, M.,
Bloom, S., 2000. Diurnal variation in orexin-A immunoreactivity and prepro-
orexin mRNA in the rat central nervous system. Neurosci. Lett. 279, 109–
112.
Tao, R., Ma, Z., McKenna, J.T., Thakkar, M.M., Winston, S., Strecker, R.E., McCarley,
R.W., 2006. Differential effect of orexins (hypocretins) on serotonin release in
the dorsal and median raphe nuclei of freely behaving rats. Neuroscience 141,
1101–1105.
Teske, J.A., Billington, C.J., Kotz, C.M., 2010. Hypocretin/orexin and energy expendi-
ture. Acta Physiol. 198, 303–312.
Vidal, L., Blanchard, J., Morin, L.P., 2005. Hypothalamic and zona incerta neurons
expressing hypocretin, but not melanin concentrating hormone, project to the
hamster intergeniculate leaflet. Neuroscience 134, 1081–1090.
Wagner, D., Salin-Pascual, R., Greco, M.A., Shiromani, P.J., 2000. Distribution of
hypocretin-containing neurons in the lateral hypothalamus and c-fos-immu-
noreactive neurons in the VLPO. Sleep Res. 3, 35–42.
Yamamoto, Y., McKinley, M.J., Nakazato, M., Yamashita, H., Shirahata, A., Ueta, Y.,
2006. Postnatal development of orexin-A and orexin-B like immunoreactivities
in the Eastern grey kangaroo (Macropus giganteus) hypothalamus. Neurosci.
Lett. 392, 124–128.
Yamanaka, A., Tsujino, N., Funahashi, H., Honda, K., Guan, J.L., Wang, Q.P., Tominaga,
M., Goto, K., Shioda, S., Sakurai, T., 2002. Orexins activate histaminergic neurons
via the orexin 2 receptor. Biochem. Biophys. Res. Commun. 290, 1237–
1245.
Yamanaka, A., Muraki, Y., Tsujino, N., Goto, K., Sakurai, T., 2003. Regulation of orexin
neurons by themonoaminergic and cholinergic systems. Biochem. Biophys. Res.
Commun. 303, 120–129.
Yokogawa, T., Marin, W., Faraco, J., Pe´zeron, G., Appelbaum, L., Zhang, J., Rosa, F.,
Mourrain, P., Mignot, E., 2007. Characterization of sleep in zebrafish and
insomnia in hypocretin receptor mutants. PLoS Biol. 5 (10), e277.
Yoshida, Y., Fujiki, N., Nakajima, T., Ripley, B., Matsumura, H., Yoneda, H., Mignot, E.,
Nishino, S., 2001. Fluctuation of extracellular hypocretin-1 (orexin-A) levels in
the rat in relation to the light-dark cycle and sleep–wake activities. Eur. J.
Neurosci. 14, 1075–1081.
Yoshida, K., McCormack, S., Espana, R.A., Crocker, A., Scammell, T.E., 2006. Afferents
to the orexin neurons of the rat brain. J. Comp. Neurol. 494, 845–861.
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2001. Orexin (hypocretin)-like
immunoreactivity in the cat hypothlamus: a light and electron microscopic
study. Sleep 24, 67–76.
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2002. Co-localization of
hypocretin-1 and hypocretin-2 in the cat hypothalamus and brainstem. Pep-
tides 23, 1479–1483.
Zhang, J.H., Sampogna, S., Morales, F.R., Chase, M.H., 2004. Distribution of hypo-
cretin (orexin) immunoreactivity in the feline pons andmedulla. Brain Res. 995,
205–217.