INVESTIGATING LIFE HISTORY CHARACTERISTICS OF 

ANOPHELES ARABIENSIS INFECTED WITH 

MICROSPORIDIA MB, A PLASMODIUM FALCIPARUM 

BLOCKING SYMBIONT 

 

Godfred Yaw Boanyah 

 

 

 

A thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, in 

fulfilment of the requirements for the degree 

DOCTOR OF PHILOSOPHY 

 

 

 

May 2025 

 

 

 



ii 
 

 

PLAGIARISM DECLARATION 

Thesis title 

Investigating life history characteristics of Anopheles arabiensis infected with Microsporidia 

MB, a Plasmodium falciparum blocking symbiont 

Declaration 

1. I understand what plagiarism entails and am aware of the University’s policy in this 

regard. 

2. I declare that this thesis is my own, original work. Where someone else’s work was 

used (whether from a printed source, the internet or any other source) due 

acknowledgment was given and reference was made according to departmental 

requirements. 

3. I did not make use of another student’s previous work and submit it as my own. 

4. I did not allow and will not allow anyone to copy my work with the intention of 

presenting it as his or her own work. 

Signature: ____ __    Date: May 2025___ 

                  Godfred Yaw Boanyah 

                   (Student ID: 2624674) 

 

 

 

 

 

 



iii 
 

DEDICATION 

I dedicate this work to my late Mum (Deaconess Mary Smith) and Dad (Elder Andrew Kwesi 

Boanyah) whose immense contribution and unflinching support have brought me this far.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



iv 

 

 

RESEARCH OUTPUTS FROM THIS THESIS 

 

I. PUBLICATIONS IN INTERNATIONAL PEER-REVIEWED JOURNALS 

Role: For the following manuscripts, I co-designed the study, performed the laboratory 

experiments, analysed the data, wrote the original draft and contributed to revived versions. 

1. Boanyah GY, Koekemoer LL, Herren JK, Bukhari T. Effect of Microsporidia MB 

infection on the development and fitness of Anopheles arabiensis under different diet 

regimes. Parasites & Vectors. 2024;17(1):294. 

Impact Factor: 3.3 

2. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. Monitoring the capacity of 

Microsporidia MB to spread through Anopheles arabiensis populations under 

laboratory conditions. Malaria Journal (In preparation). 

II. INTERNATIONAL CONFERENCES 

Role: For the following presentations, I performed the experiments, analysed the data, and 

prepared the slides or poster, incorporated editorial input from my supervisors and presented. 

1. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. “Investigating the effect of diet 

on life history characteristics of Anopheles arabiensis infected with Microsporidia MB, 

a Plasmodium falciparum blocking symbiont”. Animal-Microbe Symbioses, Gordon 

Research Seminar and Conference at Renaissance Tuscany Il Ciocco in Lucca, Italy, 

17- 23 June 2023. (Oral and Poster presentation). 

 

2. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. “Investigating the effect of diet 

on life history characteristics of Anopheles arabiensis infected with Microsporidia MB, 

a Plasmodium falciparum blocking symbiont”. 2023 Malaria Gordon Research Seminar 



v 

 

and Conference at Rey Don Jaime Grand Hotel in Barcelona, Spain, 27 May-2 June 

2023. (Poster presentation). 

 

3. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. “Investigating the effect of diet 

on life history characteristics of Anopheles arabiensis infected with Microsporidia MB, 

a Plasmodium falciparum blocking symbiont”. Anti-Vec Network meeting at Kilifi, 

Kenya, 26-28 February 2023. (Poster presentation). 

 

4. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. “Investigating the effect of diet 

on life history characteristics of Anopheles arabiensis infected with Microsporidia MB, 

a Plasmodium falciparum blocking symbiont”. Academic Exchange Session, 2023 

SSPH+ Lugano Summer School, Swiss School of Public Health (SSPH+), Lugano, 

Switzerland, 21-26 August 2023. (Oral presentation). 

III. LOCAL CONFERENCE 

Role: For this presentation, I performed the experiments, analysed the data, and prepared 

the slides or poster, incorporated editorial input from my supervisors and presented. 

1. Boanyah GY, Koekemoer LL, Bukhari T, Herren JK. “Investigating the effect of diet 

on life history characteristics of Anopheles arabiensis infected with Microsporidia MB, 

a Plasmodium falciparum blocking symbiont”. South African Society of Biochemistry 

and Molecular Biology 2024 Congress at the Protea Hotel Ranch Resort, Polokwane, 

South Africa, 7- 10 July 2024 (Poster presentation). 

 

 

 



vi 
 

 
ABSTRACT 

Introduction: Microsporidia MB is a naturally occurring symbiont found in Anopheles 

arabiensis that blocks the transmission of the Plasmodium parasite without any fitness effect 

on the mosquito. Microsporidia MB proliferates in mosquito gonads leading to its high 

intensity, which is linked to horizontal (sexual) and vertical (transovarial) transmission from 

one mosquito to another. Developing this mosquito-symbiont into a control tool is highly 

needed to complement the core malaria control tools. How diet affects Microsporidia MB 

intensity and the Microsporidia MB-infected mosquitoes, leading to the identification of diet 

regimes for mass production of high Microsporidia MB intensity mosquitoes is unknown. 

Furthermore, how Microsporidia MB spread in the infected mosquito population over 

generations is also unknown.  The first aim of this study was to investigate how diet type and 

quantity affect the Microsporidia MB-An. arabiensis life table parameters, namely, larval 

development and mortality, adult size and survival, as well as Microsporidia MB intensity in 

both larvae and adults.  The second aim was to monitor the spread of Microsporidia MB in 

infected An. arabiensis populations. 

Methods: The research was conducted using Microsporidia MB-infected and uninfected first 

filial generation (F1) larvae or adults from engorged females collected from the field (G0). The 

F1 larvae were reared together (group lines, GLs) or as individual families (eggs from one 

female also called isofemale lines, IMLs). GLs were fed on 0.3mg/larva/day of Tetramin, 

GoCat, and Cerelac respectively as well as 0.07 mg/larva/day of Tetramin. Adult GLs were fed 

on 1% or 6% glucose and their survival was determined. IMLs, on the other hand, were fed on 

Tetramin 0.07 and 0.3 mg/larva/d for the larvae and 1% or 6% glucose for the adult experiments 

respectively. Diet choices were selected based on commonly used diets and previous literature. 

Finally, three colonies (biological replicates) of Microsporidia MB-infected mosquitoes were 

established from F1 larvae and maintained until the sixth filial generation (F6) to observe the 



vii 
 

symbiont spread throughout the Microsporidia MB colony. Prevalence and intensity of 

Microsporidia MB, wing length of mosquitoes, humidity and temperature of the laboratory 

were recorded.  

Results: Microsporidia MB infected An. arabiensis fed on Tetramin at a dose of 0.3 

mg/larva/day had the quickest larval growth, greatest adult emergence, largest mosquito body 

size, highest prevalence, and highest density of Microsporidia MB. In contrast to 6% glucose, 

1% glucose did not prolong the lifespan of adult Microsporidia MB-infected mosquitoes. 

Microsporidia MB-infected and uninfected An. arabiensis fed on Tetramin at a dose of 0.07 

mg/larva/day and 1% glucose had similar development time and adult survival respectively, 

showing that the fitness benefit conferred to the mosquito host was diet-dependent. 

Furthermore, the results on the Microsporidia MB colony showed two replicates progressed to 

F6 successfully while the one collapsed (died out) at F2. Prevalence of Microsporidia MB 

increased from F1 to F3 generation and declined subsequently. Furthermore, the intensity of 

Microsporidia MB was not significantly different between the F1 and F6 in both replicates. 

There was a slight positive correlation between Microsporidia MB prevalence and temperature. 

The size of the male mosquitoes was not reduced across the generations while that of females 

fluctuated. 

Conclusion: Even on restricted diets, Microsporidia MB did not adversely affect An. 

arabiensis developmental growth or fitness. Tetramin 0.3 mg/larva/day and 6% glucose are the 

best diet regimes for rearing large number of Microsporidia MB-infected mosquitoes. These 

findings are crucial for mass rearing of high-intensity Microsporidia MB-infected An. 

arabiensis mosquitoes in the laboratory for the purposes of experiments and field releases. 

Furthermore, this first successful Microsporidia MB colony establishment attempt shows the 

pattern of Microsporidia MB prevalence across subsequent generations, and this understanding 

offers a pathway for enhancing rearing protocols to sustain high Microsporidia MB prevalence 



viii 
 

mosquito colonies. The maintenance of Microsporidia MB intensity in the mosquitoes between 

the starting and the last generation gives credence to the potential of the Microsporidia MB 

malaria control strategy. Lastly, the fitness of the male Microsporidia MB-infected An. 

arabiensis across generations observed in this study demonstrates that a male release strategy 

is feasible. 

 

 

 

 

 

 

 

 

 

 

 

 

 



ix 

 

 

ACKNOWLEDGEMENTS 

My greatest appreciation goes to the Almighty God for giving me the ability to go through and 

complete this work. It has been challenging but he has seen me through. 

 

I would like to express my gratitude to my supervisors, Dr. Jeremy Herren, Dr. Tulu Bukhari, 

and Prof. Lizette Koekemoer. Their support, guidance, expertise and patience were vital in 

facilitating my successful completion of my doctoral degree. Their mentorship and the 

experience gained from my research have enhanced my career as an upcoming scientist.  

 

I appreciate Dr Jeremy Herren for giving me the opportunity to attend and present my findings 

at international conferences and visit to the University. In addition to being readily accessible. 

 

I also thank Dr Tulu Bukhari for being very supportive by providing direct primary contact for 

discussions of challenges related to my studies and encouragement when needed. 

 

I would like to express my profound gratitude to Prof Lizette Koekemoer for her swift 

responses and assistance during my annual registrations as an out-of-seat student doing 

research in Kenya. Her gracious hospitality during my visits to Wits is unforgettable. 

 

I am very much grateful to SymbioVector project team for the support received during my 

research work. 

 

I am profoundly grateful for the scholarship awarded me by DAAD through the ARPPIS-PHD 

programme at icipe, which has enabled me to pursue my PhD. I am also appreciative of the 

financial support provided by the Bill and Melinda Gates Foundation, Open Philanthropy, and 



x 

 

Children's Investment Fund Foundation Through SymbioVector, with Dr. Jeremy Herren 

serving as the principal investigator. This research would not have been possible without 

funding. 

 

I am deeply grateful to my exceptional family. Throughout the years it has taken me to achieve 

this degree, my siblings, Abigail, Victor, Victoria, Peter, Hanah, and James, as well as my late 

father and mother, have consistently been there for me. I am grateful for their patience and 

unwavering affection. Additionally, to my dear wife, Eunice. Lastly, I want to thank my friends 

who have been very supportive during my PhD journey. God richly bless you. 

 

 

 

 

 

 

 

 

 

 

 

 

 



xi 
 

TABLE OF CONTENTS 

PLAGIARISM DECLARATION........................................................................................... ii 

DEDICATION........................................................................................................................ iii 

RESEARCH OUTPUTS FROM THIS THESIS ................................................................. iv 

I. PUBLICATIONS IN INTERNATIONAL PEER-REVIEWED JOURNALSiv 

II. INTERNATIONAL CONFERENCES ............................................................. iv 

III. LOCAL CONFERENCE ................................................................................... v 

ABSTRACT …………………………………………………………………………………vi 

ACKNOWLEDGEMENTS ................................................................................................... ix 

LIST OF FIGURES .............................................................................................................. xiv 

LIST OF TABLES ............................................................................................................. xviii 

LIST OF ABBREVIATIONS .............................................................................................. xix 

CHAPTER 1 ............................................................................................................................. 1 

INTRODUCTION.................................................................................................................... 1 

1.1 MALARIA ........................................................................................................... 1 

1.2 MAIN AFRICAN MALARIA VECTORS ....................................................... 3 

1.3 MICROSPORIDIA ............................................................................................. 5 

1.4 PROBLEM STATEMENT ................................................................................ 7 

1.5 RATIONAL OF STUDY .................................................................................. 11 

1.4 AIM AND OBJECTIVES ................................................................................. 11 

CHAPTER 2 ........................................................................................................................... 13 

EFFECT OF MICROSPORIDIA MB INFECTION ON THE DEVELOPMENT AND 

FITNESS OF ANOPHELES ARABIENSIS UNDER DIFFERENT DIET REGIMES .. 13 

2.1 INTRODUCTION ............................................................................................. 13 

2.2 AIM AND OBJECTIVES ................................................................................. 18 

2.3 MATERIALS AND METHODS...................................................................... 18 

2.3.1 Sampling site and Mosquito collection ................................................. 18 

2.3.2 Mosquito maintenance at the insectary................................................. 19 

2.3.3 DNA extraction and Molecular species identification .......................... 20 

2.3.4 Microsporidia MB screening and intensity determination ................... 22 

2.3.5 Grouping of mosquitoes for rearing ...................................................... 23 

2.3.6.1 Effect of Microsporidia MB on developmental fitness under different 

larval diet regimes ........................................................................................... 23 



xii 
 

2.3.6.2 Effect of Microsporidia MB on adult mosquito survival under different 

adult diet regimes ............................................................................................ 26 

2.3.7 Effect of larval and adult diet quantity on Microsporidia MB intensity 

in Isofemale lines (IMLs) after vertical transmission ................................... 27 

2.3.7.1 Effect of larval diet quantity on Microsporidia MB intensity in IMLs

 ………………………………………………………………………….27 

2.3.8 Ethics ...................................................................................................... 30 

2.3.9 Data analysis .......................................................................................... 30 

2.4 RESULTS ........................................................................................................... 31 

2.4.1 Species identification of G0  mosquitoes ............................................... 31 

2.4.2 Effect of Microsporidia MB on developmental fitness under different 

larval diet regimes ........................................................................................... 32 

2.4.2.1 Prevalence and intensity of Microsporidia MB under different larval 

diet regimes ...................................................................................................... 32 

2.4.2.2 Effect of Microsporidia MB on larval mortality, pupation and adult 

emergence under different larval diet regimes .............................................. 33 

2.4.2.3 Effect of Microsporidia MB on adult mosquito (GLs) survival under 

different adult diet regimes ............................................................................. 36 

2.4.3 Effect of larval and adult diet quantity on Microsporidia MB intensity 

and wing size in Isofemale lines (IMLs) after vertical transmission2.4.3.1a 

Effect of Larval diet quantity on the Microsporidia MB intensity of MB+ 

IMLs………………………………………………………………………….37 

2.4.3.1b Effect of Larval diet quantity on the wing length of MB+ and MB- 

IMLs………………………………………………………………………….38 

2.4.3.2 Adult diet quantity on the Microsporidia MB intensity of MB+ IMLs

 ………………………………………………………………………….39 

2.5 DISCUSSION .................................................................................................... 41 

CHAPTER 3 ........................................................................................................................... 46 

MONITORING THE CAPACITY OF MICROSPORIDIA MB TO SPREAD 

THROUGH ANOPHELES ARABIENSIS POPULATIONS ACROSS GENERATIONS

 ………………………………………………………………………………………...46 

3.1 INTRODUCTION ............................................................................................. 46 

3.2 AIM AND OBJECTIVES ................................................................................. 49 

3.3 MATERIALS AND METHODS...................................................................... 49 

3.3.1 Sampling site and Mosquito collection ................................................. 49 

3.3.2 Mosquito maintenance at the insectary................................................. 50 

3.3.3 Molecular species identification ............................................................ 50 

3.3.4 Microsporidia MB screening and intensity determination ................... 50 



xiii 
 

3.3.5 Optimizing colonisation of Microsporidia MB mosquitoes under 

different rearing conditions ............................................................................ 50 

i) Large cage in screen house without vegetation .................................... 51 

i) Large cage in screen house with vegetation ......................................... 53 

iii)  Standard 30x30x30cm cages under laboratory conditions .................. 54 

3.3.6 Data analysis .......................................................................................... 57 

3.4 RESULTS ........................................................................................................... 58 

3.4.1 Optimizing colonisation of Microsporidia MB mosquitoes under 

different rearing conditions ............................................................................ 58 

i) Large cage in screen house without vegetation .......................................... 58 

ii) Large cage in screen house with vegetation .............................................. 58 

iii) Rearing in small cages under insectary conditions ................................. 59 

3.4.2 Microsporidia MB prevalence across generations ............................... 60 

3.4.3 Correlation of Microsporidia MB prevalence and temperature ........... 61 

3.4.4 Microsporidia MB intensity ................................................................... 62 

i. Microsporidia MB intensity across generations ................................... 62 

ii. Microsporidia MB intensity by gender for the combined generations ..... 63 

3.4.5 Wing size of mosquitoes across generations ......................................... 64 

3.5 DISCUSSION .................................................................................................... 65 

CHAPTER 4 ........................................................................................................................... 69 

CONCLUDING SUMMARY ............................................................................................... 69 

4.1 RE-STATEMENT OF THE PROBLEM ........................................................ 69 

4.2 KEY FINDINGS ................................................................................................ 70 

4.2.1 Microsporidia MB does not adversely impact the development and 

fitness of An. arabiensis, even under limited dietary conditions ................... 70 

4.2.2 Optimal diet regime for mass rearing high intensity of Microsporidia 

MB mosquitoes under laboratory conditions ................................................. 70 

4.2.3 Spread of Microsporidia MB in mosquito populations under laboratory 

conditions ........................................................................................................ 70 

4.3 RECOMMENDATION FOR FUTURE MICROSPORIDIA MB 

MOSQUITO COLONY ESTABLISHMENT ...................................................... 71 

5.0 REFERENCES ................................................................................................................. 72 

APPENDICES ...................................................................................................................... 107 

 

 



xiv 

 

 

LIST OF FIGURES 

CHAPTER 1 

Figure 1: Death rate from malaria, 2021 [Adapted from WHO,( 2021)] ...................................................... 1 

Figure 2: Microsporidia MB in the gonads of An. arabiensis under confocal microscope using 

Fluorescence In Situ Hybridization (FISH) (A) MB in female gonad (B) High 

Microsporidia MB intensity in the oocytes (C) Microsporidia MB in the male 

gonad (D) High Microsporidia MB intensity in male testis and ejaculatory duct. The 

images show the Microsporidia MB probe-CY5 FISH probe in red and the Sytox-

Green general DNA stain in green (Adapted from Makhulu et al., (2024). ............... 7 

CHAPTER 2 

Figure 3: Larval trays with dimensions 21 × 15 × 8.5 cm ............................................................................. 20 

Figure 4: Screen house for maintaining mosquito larvae .............................................................................. 20 

Figure 5: Experimental design to determine the effect of different larval diet regimes on larval 

development, mortality and adult emergence of Microsporidia MB infected An. 

arabiensis as well as the prevalence and intensity of Microsporidia MB. MB+ and 

MB- stands for Microsporidia MB infected and uninfected larvae respectively. .... 25 

Figure 6: Experimental design to determine the effect of Microsporidia MB on adult mosquito survival 

under different adult diet regimes .................................................................................. 27 

Figure 7: (A) Experimental design to determine the effect of different larval diet quantity on 

Microsporidia MB intensity in the isofemale lines of An. arabiensis after vertical 

transmission (MB+). (B) Experimental design to determine the effect of adult diet 

quantity on Microsporidia MB intensity in isofemale line of An. arabiensis after 

vertical transmission. MB+ stands for An. arabiensis infected with Microsporidia 

MB. .................................................................................................................................... 29 



xv 

 

Figure 8: Microsporidia MB prevalence and intensity under different larval diet regimes of MB+ GLs. 

(A) Prevalence (%) of Microsporidia MB in 3-day-old F1 adults reared on different 

diet regimes. Error bars represent standard deviation.  (B) Relative Microsporidia 

MB intensity in 3-day-old F1 adults reared on different diet regimes using qPCR. 

Bar represents significant difference between the diet regimes. The number of 

independent biological replicates was three with each replicate made up of 60 

larvae.  Asterisks show the level of significance (*P < 0.05, **P <0 .001, and ***P 

< 0.0001), while ns indicates no significance............................................................... 33 

Figure 9: Effect of Microsporidia MB on mortality and larval development under different larval diet 

regimes of GLs. (A): Effect of Microsporidia MB on larval mortality under 

different diet regimes and (B) Microsporidia MB’s effect on median larval 

development time under different diet regimes. Total number of larvae were 180 in 

three biological replicates (n=60) for each diet regime. Error bars represent standard 

deviations.......................................................................................................................... 35 

Figure 10: Microsporidia MB’s effect on adult emergence under different larval diet regimes of GLs. 

The data shown is the percentage of larvae that developed into adult mosquitoes 

from the various diet regimes. This is the total number of adults that emerged from 

the larvae from each diet. Error bars represent standard deviation. ........................... 36 

Figure 11: Effect of Microsporidia MB on GLs adult mosquito survival under different adult diet 

quantity regimes (A) 1% glucose diet & (B) 6% glucose diet. The dashed lines are 

the 95% CL interval while the solid lines are the survival curve for each regime. .. 37 

Figure 12: Microsporidia MB intensity under different larval diet quantity of MB+ IML. (A) 

Microsporidia MB intensity of adult males and (B) Microsporidia MB intensity of 

adult females that emerged from larvae reared on Tetramin 0.07 mg/larva/d and 0.3 

mg/larva/d. The dotted lines represent the mean intensity of each diet regime. ....... 38 



xvi 
 

Figure 13: Effects of Microsporidia MB on wing length of MB+ & MB- IMLs under larval diet 

quantity regimes. Microsporidia MB significantly increased the length of female 

wings both under Tetramin 0.07 and 0.3 mg/larva/d respectively. The line and the 

bars in the middle of each treatment indicate the mean with 95% CI respectively. . 39 

Figure 14: Microsporidia MB intensity under different adult diet quantities of MB+ IMLs (A) Effect of 

low adult diet (1% glucose) on Microsporidia MB intensity in male and female 

IMLs mosquitoes. (B) Effect of high adult diet (6% glucose) on Microsporidia MB 

intensity in male and female IMLs mosquitoes. (C) Effect of low (1% glucose) and 

high (6% glucose) diet on the intensity of Microsporidia MB of male mosquitoes. 

(D) Effect of low (1% glucose) and high (6% glucose) diet on the intensity of 

Microsporidia MB of female mosquitoes. Bars represent the mean with a 95% 

confidence interval of the Microsporidia MB intensities of each diet regime. ......... 40 

CHAPTER 3 

Figure 15: Picture of population cage (1 m by 70cm by 60 cm): A: clay pot; B: cup containing pupae, C: 

sugar vial containing 6% glucose solution and tissue paper. ...................................... 52 

Figure 16: The schematic diagram of experimental design for the spread of Microsporidia MB in 

infected An. arabiensis population in large cages (Population cage) under semi-field 

condition. Alphabetical letters indicate the chronological steps conducted for 

establishing a large cage. ................................................................................................ 53 

Figure 17: Picture of the Screen house with plants and thatched mud-house in which population cages 

were kept to rear mosquitoes. ......................................................................................... 54 

Figure 18: Picture of the insectary with uncontrolled temperature and humidity ....................................... 56 

Figure 19: Picture of a rack carrying the 30cmby30cm by 30cm cages in the insectary ........................... 56 



xvii 
 

Figure 20: The schematic diagram of the experimental design of the spread of Microsporidia MB in 

infected An. arabiensis in small cages (30x30x30cm) in the laboratory. 

Alphabetical letters indicate the chronological steps ................................................... 57 

Figure 21: Overview of Microsporidia MB spread across generations in An. arabiensis in the insectary 

setting. Total number of adult mosquitoes in each replicate is represented by n. 

Replicate indicate an independent attempt to colonisation. ........................................ 59 

Figure 22: Prevalence of Microsporidia MB across generations in An. arabiensis .................................... 60 

Figure 23: Prevalence of Microsporidia MB and temperature across generations in the two replicates (1 

& 3). .................................................................................................................................. 61 

Figure 24: Microsporidia MB intensities across generations (A) Replicate 1 intensities (B) Replicate 3 

intensities. Asterisks indicate the level of significance (*P < 0.05, **P < 0 .001, 

***P < 0.0001); ns indicates no significance. .............................................................. 62 

Figure 25: Effect of Microsporidia MB intensity on sex across generations. The line and the bars on the 

x-axis indicate the mean with 95% CI, respectively. ................................................... 63 

Figure 26: Male and female wing size of mosquitoes across generations for replicate 1. The line and the 

bars in the middle of each group indicate the mean with 95% CI, respectively. 

Asterisks indicate the level of significance (*P < 0.05, **P < 0 .001, 

***P < 0.0001); ns indicates no significance. ............................................................. 64 

 

 

 

 

 

 



xviii 
 

 

LIST OF TABLES  

Table 1: Nutritional components of Tetramin, GoCat and Cerelac diets ..................................................... 26 

Table 2: G0  mosquitoes collected from the field and the number of females confirmed to be An. 

arabiensis and Microsporidia MB infected. ................................................................. 31 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



xix 

 

 

LIST OF ABBREVIATIONS 

An.  Anopheles  

Ae.  Aedes 

 s.l.  sensu lato   

 s.s.  sensu stricto  

CI  Confidence interval 

COVID-19  Coronavirus disease 2019  

d  Day 

℃  Degree Celsius 

Colonisation   Self-replicating generations of Microsporidia MB infected Anopheles 

arabiensis 

Et al    Et alia 

F (1,2,3,4,5 & 6)        1st, 2nd, 3rd, 4th, 5th or 6th Filial generation 

g             gram 

G0     Wild engorged female mosquitoes from the field 

GLs    Group lines  

hr  Hour 

IMLs  Isofemale lines 

IRS  Indoor residual spraying  

JAK- STAT  The Janus kinase-signal transducer and activator of transcription 

L1  First instar larva 

LLINs  Long-lasting insecticide-treated be nets 



xx 

 

MB+  Microsporidia MB infected mosquitoes 

MB      Microsporidia MB 

MB-     Microsporidia MB uninfected mosquitoes 

mg   Milligram 

ml   Millilitres 

P.   Plasmodium 

PCR   Polymerase Chain Reaction 

qPCR    Quantitative Polymerase Chain Reaction 

SIT  Sterile insect technique  

WHO   World Health Organization 

 

 



1 

 

CHAPTER 1 

INTRODUCTION  

 

1.1 MALARIA 

The spread of malaria is caused by the infectious bite of female Anopheles mosquitoes whiles 

the males on the other hand, pose no threat for disease transmission (Filler et al., 2006; Sinka 

et al., 2010). The infection is caused by a parasite called Plasmodium (Larson, 2019). Although 

there are five species of Plasmodium parasite that can infect humans only two of these species, 

Plasmodium falciparum and Plasmodium vivax, are considered to be dangerous due to major 

medical complications associated with them (Rougeron et al., 2022; Sato, 2021). Plasmodium 

falciparum causes the most lethal form of malaria, the commonest on the African continent, 

and is responsible for the majority of the global mortality (Bousema & Drakeley, 2011; Neveu 

& Lavazec, 2021; Snow, 2015). On the other hand, P. vivax is the most widely distributed  

malaria causing parasite (Bousema & Drakeley, 2011; Kevin Baird, 2013). 

 

 

Figure 1: Death rate from malaria, 2021 [Adapted from WHO,( 2021)] 



2 

 

 

Despite significant efforts and investments in malaria control, including enhancements to 

health systems, vector control measures, and pharmacological treatments, malaria continues to 

pose a significant public health challenge globally, with a particular emphasis on Africa 

(Wilson et al., 2020; WHO, 2020). According to the 2023 world malaria report published by 

the World Health Organization (WHO), there were 608,000  fatalities attributed to malaria 

worldwide, despite the implementation of interventions by international and national malaria 

control programmes (WHO, 2023). The use of long-lasting insecticide-treated bed nets (LLIN) 

and indoor residual spraying (IRS) has led to a significant reduction in malaria transmission in 

recent years (Abeyasinghe et al., 2012; Rek et al., 2020). Notably, children were identified as 

the demographic group most severely impacted by this disease (WHO, 2023). The COVID-19 

pandemic led to a 12% rise in malaria-related fatalities in 2020, primarily as a result of the 

limitations imposed on malaria control initiatives (WHO, 2021). The consequences of climate 

change (Giesen et al.., 2020; Leal Filho et al.., 2023)  and the development of insecticide 

resistance in regions where the disease is prevalent (Glunt et al.., 2018; Hemingway et al.., 

2016; Lindsay et al.., 2021) have impacted the proliferation of mosquito vectors. Furthermore, 

the influence of chemical insecticides (specifically LLIN/IRS) or repellents have resulted in 

alterations in mosquito behaviour and physiology, including shift in the time of females biting 

and outdoor biting, shifts in vector composition, delayed mortality, and other factors that pose 

a significant challenge to these WHO endorsed primary control strategies (Sougoufara et al., 

2020; The malERA Consultative Group on Vector Control, 2011; Wilson et al., 2020). Hence, 

it is imperative to develop other methods for vector control. 



3 

 

 

1.2 MAIN AFRICAN MALARIA VECTORS 

To determine the likelihood of malaria occurring in a certain location or region, it is necessary 

to take into account the presence of Anopheles vector populations. Out of the 465 species of 

Anopheles mosquitoes that have been formally identified, around 70 of them are capable of 

transmitting malaria parasites to humans (Warrell &  Watkins, 2017). These are often 

categorised as either primary or secondary vectors via the use of a categorisation system that 

is not strictly defined, and it is based on the contributions that they have been measured or 

inferred to the transmission of malaria (Brooke, 2022). The primary vectors of malaria in Africa 

are classified into four taxonomic groups: the Anopheles gambiae complex, the Anopheles 

funestus group, the Anopheles nili group, and the Anopheles moucheti group (Sinka et al., 2010, 

2012). The ranges, habits, and ecological systems of these species complexes or groups are 

distinct from one another (Sinka et al., 2010, 2012). The acquisition of malaria by blood 

transfusion is the one and only conceivable exception to infection through an infectious 

mosquito bite,, an event that is exceedingly uncommon (Mangano et al., 2019). 

 

The Anopheles gambiae complex also known as Anopheles gambiae sensu lato (s. l.) is 

responsible for the majority of the transmission of malaria in sub-Saharan Africa (Warrell & 

Watkins, 2017). Anopheles gambiae senso stricto (An. gambiae sensu s.s.), An. coluzzii, An. 

arabiensis, An. melas, An. merus, An. bwambae, An. quadriannulatus, An. amharicus, and An. 

fontaneilli are the nine subspecies that make up this complex.  These subspecies are linked to 

one another and are not identifiable from one another based on their physical characteristics 

(Barrón et al., 2019). Out of this complex, An. gambiae s.s., An. coluzzii and An. arabiensis are 

identified as the major malaria vectors in Africa including An. funestus from the An. funestus 

group (Brooke, 2022). 



4 

 

 

The environmental conditions of a region have a significant impact on the distribution and 

abundance of these species (Oyewole et al., 2006). In the driest regions, An. arabiensis is the 

most common mosquito species. It is distributed across dry areas south of the Sahara, Horn of 

Africa, and Namibian savannah. The larvae live in small, transitory, sunny clear and shallow 

fresh water ponds, stagnant and murky water bodies with or without vegetation, rice fields, and 

other natural and manmade environments (Ashine et al., 2024; Sinka, 2013). Anopheles 

arabiensis' host preference and biting behaviour vary widely according to location, host 

availability, and genotypes. It is partially endophagic and exophilic because it bites indoors at 

night (typically between 19:00 and 03:00 h) and has early-exit behaviour regardless of its meal 

(Ashine et al., 2024). Despite it being anthropophilic, it has a robust tendency to be zoophilic. 

An. arabiensis is becoming more prevalent in major cities as a result of the fact that the human 

environment is always evolving and that African cities are expanding (Dabiré et al., 2012; 

Jones et al., 2012). Furthermore, An. arabiensis has been reported as dominant vector for 

malaria transmission in Western, Eastern, Central and Southern Africa respectively (Dabiré et 

al., 2012; Jones et al., 2012; Kaiser et al., 2021; Sinka et al., 2010; Zanga et al., 2024). 

A recent investigation on the microbiota of two Anopheles species revealed that the 

predominant bacteria in both An. funestus and An. arabiensis were Serratia oryzae and 

Elizabethkingia anopheles (Silva et al., 2021). Anopheles arabiensis Patton was found to 

harbour a symbiotic acetic acid bacterium known as Asaia spp. (Epis et al., 2012). In addition 

to An. arabiensis, Asaia spp. has been detected in the microbiota of four other mosquito species, 

An. gambiae, An. stephensi, Aedes albopictus, and Aedes aegypti (Chouaia et al., 2010). Asaia 

spp. is present in various organs of mosquitoes and is transferred either vertically or 

horizontally between mosquitoes (Rami et al., 2018). Furthermore, this symbiont in An. 



5 

 

arabiensis modulates the immune response of the mosquito against malaria parasite and hence 

has been proposed as a potential vector control tool (Capone et al., 2013).  

 

 

1.3 MICROSPORIDIA  

Cryptomycota, Microsporidia and Aphelida phyla make up the Opisthosporidia superphylum, 

which are basic Fungi. Despite sharing the evolutionary roots of genuine Fungus (Eumycota), 

sister phyla are distinguished from genuine Fungi by their common morphological 

characteristics, such as dual-layered chitinous spore walls and electron-dense anchoring discs 

(Bass et al., 2018). Microsporidia differ from other Opisthosporidia phyla by having distinct 

organelles like the polar tube and major metabolic pathway reductions (Bass et al., 2018). The 

phylum Microsporidia has over 1,300 identified species across 220 genera (Park & Poulin, 

2021). This likely reflects just a portion of the actual variety of microsporidia. The placement 

of the Microsporidia lineage within the tree of life has seen significant alterations (Corradi & 

Keeling, 2009). The first discovered Microsporidium, Nosema bombycis was characterised as 

a yeast-like fungus and classified within the Schizomycetes, which closely aligns with the 

contemporary classification of Microsporidia (Corradi & Keeling, 2009). As a result of a 

taxonomy study that was based on physical characteristics, Microsporidia were subsequently 

moved to a primitive amitochondrian group termed Archezoa. This was done because of the 

absence of cell organelles, particularly mitochondria, in the microsporidia. The idea that 

Microsporidia are, in reality, close relative of Fungi has gained widespread acceptance as a 

result of the accumulation of molecular data (Bass et al., 2018).  

 

Microsporidia are eukaryotic microorganisms that must always remain inside their host cells 

(obligate intracellular microbe), the only way for them to live outside of cells is through spore 



6 

 

formation. Morphological identification is often based on spores, which represent the matured 

infectious stage of the life cycle. It is possible for some species of Microsporidia to generate 

many kinds of spores during their life cycle (Becnel et al., 2005). As an example, the Aedes 

aegypti parasite Edhazardia aedis produces four distinct spore types (Becnel et al., 2005). 

Microsporidia have a meront proliferation phase and chitinous cell wall spores that are 

transmitted from host to host if ingested. Horizontal transmission (mating) species have higher 

virulence and lower host specificity, whereas horizontal and vertical transovarial transmission 

species have lower virulence and higher host specificity (Zilio et al., 2018).  

 

Exploitation of the symbiotic association of inherited non-pathogenic Microsporidia by a major 

African malaria vector, Anopheles arabiensis, evaluated in Kenya has shown blocking potency 

against the transmission of the human Plasmodium parasite (Herren et al., 2020). This 

microsporidium was named Microsporidia MB by Herren et al., (2020). Moreover, a follow-

up investigation showed a high intensity of Microsporidia MB in the gonads of An. arabiensis. 

Microsporidia MB are transmitted from one mosquito to another through two routes: 

Horizontal transmission occurs through mating and vertical transmission from mother to 

offspring  (Herren et al., 2020; Nattoh et al., 2021). Vertical transmission of Microsporidia MB 

was shown to be intensity-dependent (Herren et al., 2020). Microsporidia MB has also been 

found in Anopheles coluzzii and An. gambiae s.s mosquitoes  (Ahouandjinou et al., 2024; 

Akorli et al., 2021; Nattoh et al., 2023). 



7 

 

 

Figure 2: Microsporidia MB in the gonads of An. arabiensis under confocal microscope 

using Fluorescence In Situ Hybridization (FISH) (A) MB in female gonad (B) High 

Microsporidia MB intensity in the oocytes (C) Microsporidia MB in the male gonad (D) 

High Microsporidia MB intensity in male testes and ejaculatory duct. The images show 

the Microsporidia MB probe-CY5 FISH probe in red and the Sytox-Green general DNA 

stain in green (Adapted from Makhulu et al., (2024). 

 

1.4 PROBLEM STATEMENT  

Several promising non-chemical-based mosquito vector control research discoveries have been 

made, ranging from paratransgenesis, symbiotic microbes, genetic modification approaches, 

sterile insect technique (SIT), and many more in the laboratory (Derua et al., 2019; Kaura et 

al., 2023; W. L. Liu et al., 2022; Marshall & Vásquez, 2022; Tyagi, 2022). However, a few of 

these have  been studied thoroughly for a complete understanding needed for operational 

implementation (Wilson et al., 2020). SIT uses either irradiation (Ernawan et al., 2022) or a 

genetic approach (Dilani et al., 2021; Leftwich et al., 2014) to render males sterile. Sterilised 

males are then released into the environment, where they will mate with wild females, produce 

infertile eggs and subsequently result in the reduction of the targeted mosquito population 



8 

 

(Lance & McInnis, 2020; Schetelig & Wimmer, 2011). Another example is the existence of 

symbiosis in mosquitoes that reduces or blocks the transmission of Plasmodium to humans 

(Herren et al., 2020).  

 

Symbiosis refers to a close physical or ecological interaction between two distinct species (Bao 

& Roossinck, 2013). This association can result in various outcomes for the organisms 

involved, including commensalism (beneficial to one species), mutualism (beneficial to both 

species), parasitism (detrimental to one species), or neutrality (no discernible effects) (Smith, 

2017). Endosymbionts are diminutive symbiotic organisms that reside within a host organism 

and reside inside the host's cells (Gray, 2017; Martin et al., 2022). In insects, endosymbionts 

play a vital role by enhancing nutritional provision and defence mechanisms (Hansen et al., 

2020; Waterworth et al., 2020). Additionally, these endosymbionts can have positive or 

negative effects on the fitness of the insects (Anbutsu & Fukatsu, 2011). However, the stability 

and density of these symbionts within the insect host, as well as their ability to provide 

protective or harmful effects, are influenced by environmental factors such as temperature 

availability of nutrients (Doremus et al., 2019; X. D. Liu et al., 2019; Mouton et al., 2006; B. 

Zhang et al., 2019). 

 

Vector-pathogen interaction can be manipulated through the symbiotic microbial association 

of certain microbes, leading to the prevention of transmission of disease pathogens to humans. 

This phenomenon of spreading symbionts in vector populations can be developed into a novel 

vector control strategy to mitigate their human disease transmission capacity (Bian et al., 

2010). Wolbachia is an intracellular bacterium harboured by many insects (Landmann, 2019) 

and is vertically or maternally transmitted. Its’ adoption in Ae. aegypti mosquitoes successfully 

controlled dengue transmission (Dorigatti et al., 2018). According to research, Wolbachia 



9 

 

symbionts inhibit viral replication in mosquitoes by modulating the immune system, thereby 

substantially reducing the prevalence of the human viral disease (dengue) (Dorigatti et al., 

2018; Johnson, 2015). In addition to the above, the symbiont uses a reproductive manipulative 

mechanism called cytoplasmic incompatibility, where the sperms of infected male mosquitoes 

render eggs from non-carriers females unviable after mating (Werren et al., 2008). This 

ultimately leads to a dramatic decrease in the size of the mosquito population (Crawford et al., 

2020; Dobson et al., 2002). The control of mosquito populations using Wolbachia symbionts 

employs two primary mechanisms: population replacement and population suppression 

(Alphey, 2009; Yen & Failloux, 2020). The population replacement strategy involves releasing 

Wolbachia-infected female mosquitoes that, upon mating with either infected or non-infected 

mosquitoes, yield viable offspring (Yen & Failloux, 2020). This facilitates the proliferation of 

Wolbachia within the field population, which contains less competent individuals for disease 

transmission, while the overall mosquito population size remains constant (Yen & Failloux, 

2020). The goal of population suppression is to reduce the overall number of mosquitoes by 

releasing male mosquitoes that are infected with Wolbachia and, after mating with wild 

females, do not produce viable offspring (Beebe et al., 2021; Werren et al., 2008). Several 

mathematical models have been established that can be employed in the effective 

implementation of these two Wolbachia-based mosquito population control methods (Lim et 

al., 2024; Yan et al., 2025; Zhang & Zheng, 2022).  

 

Globally, the method of Wolbachia suppression and replacement of mosquito populations has 

been implemented in countries such as Singapore, Thailand, Mexico, and Australia to 

significantly reduce arboviral diseases, including dengue and Zika (CDC, 2024). For instance, 

the release of Aedes aegypti mosquitoes infected with the wMel strain of Wolbachia pipientis 

in Indonesia has demonstrated a significant protective efficacy, leading to a significant 



10 

 

reduction in dengue symptomatic patients and hospitalizations (Utarini et al., 2021). The 

implementation of a Wolbachia-Aedes aegypti control strategy in 480,000 households in 

Singapore resulted in an approximately 80-90% reduction in the mosquito population (National 

Environmental Agency, 2022). Therefore, in 2022, Singapore increased the household sites for 

the field release of the symbiotic mosquitoes to an additional 100,000 due to the success of 

previous releases (National Enviromental Agency, 2022). Furthermore, cytoplasmic 

incompatibility and SIT have been used in combination with evidence of great success in 

mosquito population suppression in Mexico, Australia and Indonesia (Beebe et al., 2021; 

Martín-Park et al., 2022; Tantowijoyo et al., 2020). 

 

Apart from targeting the adults, it is also possible to use non-chemical-based interventions to 

target the larvae. The implementation of bacterial larvicides for larval source management can 

reduce the population density of mosquito vectors (Dambach et al., 2020). This methodology 

employs microbial larvicides, specifically Bacillus thuringiensis israelensis (Bti) and Bacillus 

sphaericus (Bs), to effectively target and eliminate mosquito larvae while minimizing any 

adverse impact on non-target organisms (Dambach et al., 2019). Bacterial larvicides were 

easily accepted by communities, such as in Burkina Faso (Dambach et al., 2020; Derua et al., 

2019). 

 

Malaria control programmes, including the Roll Back Malaria initiative, achieved notable 

progress with the distribution of insecticide treated bed nets in Africa, but these initiatives have 

encountered stagnation in recent years due to insecticide resistance and climate change among 

other factors (WHO, 2018; Yamey, 2004). Hence, it is important to assess the viability of 

potential strategies such as biological control through natural symbionts in mosquitoes for 

integration into currently employed strategies (Kamareddine, 2012). 



11 

 

 

1.5 RATIONALE OF STUDY 

Anopheles arabiensis serves as a prominent malaria vector in Africa, including East Africa 

(Eligo et al., 2024; Hemming-Schroeder et al., 2020; Mustafa et al., 2021). The occurrence of 

Microsporidia MB in the indigenous Anopheles mosquito population in Kenya varies from 0 

to 9% (Herren et al., 2020). The primary objective of the Microsporidia MB malaria control 

approach is to optimise techniques for the bulk production of Microsporidia MB-infected 

mosquitoes for future field trial releases.  

 

However, how nutrition and environmental conditions affect the intensity and transmission of 

Microsporidia MB in infected An. arabiensis, are unknown. Furthermore, how this will impact 

the life history characteristics of Microsporidia MB infected mosquitoes is also unknown. 

Under standard laboratory mosquito rearing methods, vertical and/or horizontal transmission 

do not lead to 100% transmission of Microsporidia MB from one An. arabiensis to another 

(Nattoh et al., 2021). By understanding how different mosquito diets and rearing regimes affect 

Microsporidia MB prevalence (percentage of individuals carrying Microsporidia MB) and 

intensity, and therefore transmission, optimal rearing protocols can be developed to sustain or 

increase the prevalence of Microsporidia MB in infected An. arabiensis colonies. Mosquitoes 

from these infected colonies could be released into the environment to increase the natural 

prevalence of Microsporidia MB in nature and thereby reducing the spread of malaria. 

 

1.6 AIM AND OBJECTIVES 

The aim of this study was to investigate the effect of diet on the life history characteristics of 

An. arabiensis infected with Microsporidia MB, a P.  falciparum blocking symbiont. ‘It was 

necessary to assess the effect of different diets on Microsporidia MB intensity in mosquito 



12 

 

population without compromising the mosquitoes’ life table parameters and other aspects of 

their biology, and as well establish a Microsporidia MB infected mosquito colony. The outcome 

of this study provides a basis for the formulation of protocols for the mass rearing of 

Microsporidia MB mosquitoes. To accomplish the overarching goal, the study examined the 

three primary objectives in the following manner: 

1) To determine the effect of diet regimes on An. arabiensis infected with Microsporidia 

MB as well as the effects of diet regimes on Microsporidia MB intensity in An. 

arabiensis. 

2)  To determine the effects of mosquito diet quantity after vertical transmission of 

Microsporidia MB-infected An. arabiensis.   

3) To determine the prevalence of Microsporidia MB in infected An. arabiensis under 

different rearing conditions.  

 

 

 

 

 

 

 

 

 

 

 

 

 



13 

 

CHAPTER 2 

 EFFECT OF MICROSPORIDIA MB INFECTION ON THE 

DEVELOPMENT AND FITNESS OF ANOPHELES 

ARABIENSIS UNDER DIFFERENT DIET REGIMES 

 

This chapter was published as: 

Boanyah GY, Koekemoer LL, Herren JK, Bukhari T. Effect of Microsporidia 

MB infection on the development and fitness of Anopheles arabiensis under 

different diet regimes. Parasites & Vectors. 2024;17(1):294 

 

2.1 INTRODUCTION  

The nutritional adaptations and requirements of the aquatic larvae stage of Anopheles 

mosquitoes differ significantly from those of terrestrial adult-stage mosquitoes (Clements, 

1992). The mouth brushes of the omnivorous scavenger larvae are used to collect suspended 

food particles from submerged surfaces (Clements, 1992). Mosquito larvae in the wild 

consume organic waste from their surroundings, namely microorganisms such as bacteria, 

protozoa, and algae, along with crustaceans, plant debris, and insect exuviae (Souza et al., 

2019). Adult mosquitoes, on the other hand, gain energy mainly from plant sources such as the 

nectar of flowers (Müller & Schlein, 2006). Studies have shown that the reproductive capacity 

of mosquitoes is affected by the type and quantity of diet that either the larval or adult 

mosquitoes feed on (Nayar & Sauermann, 1975; Zeller & Koella, 2016). Several studies have 

investigated the effect of diet on Anopheles mosquitoes (Carvajal-Lago et al., 2021; Müller & 

Schlein, 2006; Zeller & Koella, 2016).  

 

One notable finding is that a limited larval diet regime (0.2 mg/larva/day) led to prolonged 

development time, reduced pupation and adult emergence rates, and diminished adult female 



14 

 

body size in comparison to higher diet regimes (0.3 and 0.6 mg/larva/day), illustrating the 

influence of larval diet quantity on mosquito life history traits (Dodson et al., 2011). 

Furthermore, the larval diet's content significantly affects both larval development and adult 

body size (Damiens et al., 2012).  A study on An. arabiensis found that fatty acid profiles in 

mosquitoes were modified by the larval diet they fed on, which controlled mosquito size, 

phosphorus nutrition, and population size (Hood-Nowotny et al., 2012). Larval diets in nature 

and the laboratory contain amino acids, carbohydrates and fats essential for larval development 

and emergence into an adult mosquito (Timmermann & Briegel, 1999; van Schoor et al., 2020). 

A transcription study on five-day-old female Aedes albopictus adult fed with two larval feeding 

regimes: a low quantity diet (1 mg/larva/day of Tetramin (fish food)) and a high quantity diet 

(2 mg/larva/day of Tetramin) indicated that the expression of specific immune genes linked to 

the Toll, JAK-STAT, and Imd pathways was elevated in those subjected to a high larval quantity 

diet (Mackay et al., 2023). These pathways played very critical roles in the mosquito immune 

system through the modulation of cytokine receptors and T helper cells (Seif et al., 2017). The 

inference could be that starvation suppressed gene expression of the immune system to preserve 

mosquito survival, and when they have optimal or high quantity of diet there is no need to 

decrease transcription of genes. 

 

Adult An. stephensi mosquitoes, emerging from larvae that were fed on a reduced larval diet 

regime (0.2 mg/larva/day of Tetramin), exhibited a delay in P. falciparum parasite development 

compared to those fed on 0.6 mg/larva/day of same diet (Shapiro et al., 2016). Additionally, 

the rate at which parasites entered the salivary glands of mosquitoes slowed down, which 

resulted in an extension of the amount of time it needed for mosquitoes to become infectious 

(Shapiro et al., 2016). A separate study involving three distinct larval diets—Dr. Clarke's Pool 

Pellets, Tetramin Fish-Flakes, and Nishikoi Fish Pellets, with protein contents of 47%, 34%, 

https://www.sciencedirect.com/topics/agricultural-and-biological-sciences/amino-acid


15 

 

and 26% respectively, revealed a significantly elevated prevalence and intensity of P. berghei 

in An. coluzzii adults that were fed with the larval diet characterised by the lowest protein 

content (Nishikoi Fish Pellets) during their larval phase, thereby concluding the influence of a 

low-protein larval diet on the vectorial competence of adult mosquitoes (Linenberg et al., 

2016). However, the role of other elements might also play a role and these were not 

investigated by the authors. In another study, in Anopheles darlingi, an increase in larval feed 

amount was positively correlated with a higher frequency of biting, extended blood meal 

duration, and increased wing length, hence enhancing vectorial capacity. Anautogenous 

mosquitoes need blood to lay eggs. Starvation during larval growth resulted small adult 

mosquitoes that required numerous blood meals for egg production and consequently increased 

host–vector interactions and disease transmission (Mitchell-Foster et al., 2012). Measuring the 

wing length of insects is crucial since it serves as a proxy for the determination of body size, 

which in turn can be used to predict fecundity, the amount of larval food consumed longevity, 

number of eggs laid and development time (Armbruster & Hutchinson, 2002; Gutiérrez et al., 

2020; Mackay et al., 2023; Norry & Loeschcke, 2002) 

 

In contrast to the studies mentioned in the aforementioned paragraph,  the adult mosquitoes in 

nature feed  on natural sugar sources from nectar that contain mainly simple carbohydrates for  

metabolic energy (Kessler et al., 2015; Reyes et al., 2021). The level of metabolic energy 

obtained by the mosquito through feeding affects the host-seeking and biting behaviour, which 

consequently impacts the vectorial capacity or disease transmission potency of the adult 

mosquito (Carvajal-Lago et al., 2021; Zirbel & Alto, 2018). For instance, An. gambiae 

mosquitoes that previously consumed glucose exhibited a heightened biting frequency 

compared to those fed on a sucrose diet (Kessler et al., 2015). Longevity, body size, and biting 



16 

 

frequency are determined by the quality and quantity of food consumed by the mosquito in its 

lifetime (Carvajal-Lago et al., 2021).  

 

In addition, adults fed on 10% sucrose or glucose or fructose after 48 hours of starving showed 

a significant expression of p400, piwi4 and ppo8 genes in Ae. aegypti mosquitoes when 

compared with unfed ones (Almire et al., 2021). The expression of these immune genes after 

the sugar feeding of the primary arbovirus vector, Ae. aegypti, demonstrated an enhancement 

in gastrointestinal tract integrity and antiviral immunity, consequently inhibiting the 

development of arbovirus infection and its spread to other tissues, leading ultimately to a 

decrease in disease prevalence (Almire et al., 2021). This is evidence that sugar diet has an 

influence on mosquito immunity. 

 

The influence of adult diet affects male and female mosquitoes differently (Vrzal et al., 2010). 

For example, research showed that mosquitoes consuming nectar rich in both carbohydrates 

and amino acids did not increase the survival of males when compared to those fed on 

carbohydrate only, nonetheless, the survival rate of female mosquitoes increased by 5% (Vrzal 

et al., 2010). Apart from the differences observed in adult survival between males and females, 

diet affects the microbiota in the mosquito.  

 

Furthermore, diet was found to have a significant influence not just on the mosquito’s gut 

microbiota, but also on the gut microbiota of other insects (Luo et al., 2021; Mugo-Kamiri et 

al., 2024). When mirid bugs, Adelphocoris suturalis were fed with balanced (mungbean sprout 

only and combined mungbean sprout and aphids) and more uniform (exclusive feeding on 

aphids) diet, a significant variation of gut microbiota was shown with single diet resulting in 



17 

 

high mortalities of bugs (Luo et al., 2021). Beetles from the family Chrysomelidae that fed on 

generalist plants had diverse microbiota compared to specialist ones, which could simply be 

attributed to they picking up diversity through contact with a more diverse set of hosts and 

thereby demonstrating the effects of nutrition on insect symbionts (Brunetti et al., 2022).  

 

With regards to mosquito rearing in the laboratory, a sugar concentration of 5% or lower was 

classified as a low-stress adult diet, but that of 20% sugar is categorised as a high-stress diet 

with 10% sugar diet used mostly for standard laboratory adult rearing regime (Caragata et al., 

2016). Some other laboratories also use 6% glucose for normal mosquito maintenance (Herren 

et al., 2020; Lambrechts et al., 2006; Nattoh et al., 2021).  

 

Nutrition is well recognised as a key factor in influencing the interactions between a host and 

its symbiotic organisms (Herren et al., 2014; Ponton et al., 2014). Diet composition, for 

example, protein to carbohydrate ratio has been shown to significantly influence the interaction 

between Wolbachia (which are a maternally inherited bacterial endosymbionts) and it's host 

insect   (Camacho et al., 2017; Caragata et al., 2016; Ponton et al., 2014). This, in turn, 

influences Wolbachia intensity and its impact on the mosquito host's longevity and fertility 

(Whittle et al., 2021). Three main processes govern the host regulation of endosymbionts. The 

first suggests that the symbiont intensity is frequently regulated by the dietary requirements of 

the host: when nutrient availability declines, symbiont intensity may increase to synthesise 

necessary nutrients for the host (Snyder et al., 2012). 

 

A higher intensity of Microsporidia MB infection in Anopheles mosquitoes is associated with 

increased vertical transmission (Herren et al., 2020). Therefore, understanding the factors that 



18 

 

maintain high prevalence and intensity of Microsporidia MB in mosquito populations is 

essential for developing it as a malaria control tool. Additionally, the spread of Microsporidia 

MB may be influenced by its impact on host fitness in various natural environments. My 

research explored how different diet regimes affect Microsporidia MB infection parameters in 

An. arabiensis and investigated how these regimes influence life history traits of An. arabiensis 

in the presence of Microsporidia MB. 

 

2.2 AIM AND OBJECTIVES 

The main aim in this chapter was to determine the effect of Microsporidia MB infection on the 

development and fitness of An. arabiensis under different diet regimes. The three main 

objectives below encompass experiments to answer this aim: 

1. To evaluate the effects of diet regimes on Microsporidia MB prevalence and 

intensity in An. arabiensis Group line (GLs). 

2. To determine the effect of diet regimes on larval development, mortality, adult 

emergence and adult survival of An. arabiensis Group lines (GLs) infected with 

Microsporidia MB. 

3.  To determine the effects of mosquito diet quantity after vertical transmission of 

Microsporidia MB in An. arabiensis isofemale lines (IMLs). 

 

2.3 MATERIALS AND METHODS 

 

2.3.1 Sampling site and mosquito collection  

The first filial generation (F1) originates from engorged wild-caught, also called field-collected 

females (Generation zero [G0]) within the period of 2022 and 2023 was utilized for this 

experiment. Wild blood-fed female An. gambiae complex females were collected resting 



19 

 

indoors using aspirators from residential dwellings in the vicinity of the Ahero (–34.9190W, –

0.1661N) irrigation scheme, located in Kisumu County, Kenya. In this location, more than 97% 

of the An. gambiae complex was previously identified as An. arabiensis mosquitoes (Herren et 

al., 2020). Females were morphologically identified (Coetzee, 2020). 

 

2.3.2 Mosquito maintenance at the insectary 

This experiment was undertaken at the International Centre of Insect Physiology and Ecology 

(icipe), Thomas Odhiambo Campus (ITOC), Kenya, from August 2023 to July 2024. The 

mosquitoes were transported to the laboratory located at icipe. For transportation ease, the G0 

females were placed in custom-built cages measuring 30 × 30 × 30 cm, which were covered 

with a damp towel. The mosquitoes had access to a 6% (w/v) glucose solution. In the laboratory 

setting, oviposition was induced by placing individual female mosquitoes (G0) into 1.5 ml 

microcentrifuge tube. The tube was lined with filter paper and 100 µl of water was added for 

moisture (Nepomichene et al., 2017). After oviposition, the eggs of each female were placed 

into water within larval trays measuring 21 × 15 × 8.5 cm (Figure 3). These trays were 

maintained under semi-field settings (Figure 4) until pupation. The female G0  individuals that 

oviposited were identified with PCR assay, as described in section 2.3.4.(Santolamazza et al., 

2008).  Female G0 samples that were not identified as An. arabiensis were discarded and not 

utilised in the experiments. Anopheles arabiensis F1 eggs were reared through to adults. 

Emerging F1 adults were maintained on 6% glucose solution that was replaced every two days 

in 30 × 30 × 30 cm cage. 



20 

 

 

Figure 3: Larval trays with dimensions 21 × 15 × 8.5 cm 

 

Figure 4: Screen house for maintaining mosquito larvae 

 

2.3.3 DNA extraction and molecular species identification 

DNA extraction was done using the ammonium acetate protein precipitation technique used by 

Herren et al., (2020).  One whole mosquito was homogenised in a 300 μl cell lysis solution on 

ice. The sample was incubated at 65°C for 2 hrs. At room temperature, 100 μl of Protein 

Precipitation Buffer (PPS) prepared from 8 M Ammonium acetate from 1 M stock (Sigma-



21 

 

Aldrich Co., St. Louis, MO [Catalog number 09689]) and 1 mM Ethylenediaminetetraacetic 

acid (EDTA) from 0.5 M stock (Sigma-Aldrich Co., St. Louis, MO [Catalog number 03677]) 

was added and vortexed for 30 s. The sample was placed on ice for 15 mins and then centrifuged 

at 15,000 RPM for 15 mins. The supernatant was removed and placed in a fresh 

microcentrifuge tube with 300 μl absolute isopropanol, and the pellets discarded. The 

supernatant was thoroughly mixed by inverting gently 100 times and centrifuging at maximum 

speed for 1 hr. The supernatant was pipetted off, 300 μl of ice cold 70% ethanol was added and 

mixed by inverting several times. This was centrifuged for 30 mins and the ethanol pipetted 

off. The microcentrifuge tube was inverted on tissue paper overnight. The DNA was eluted 

with 50 μl double distilled water and left in the fridge for 3 hrs and then used or stored in -

20°C. DNA extraction was performed for G0 and F1 samples. For G0 samples, 2 μl was utilised 

for molecular species identification and 2 μl for Microsporidia MB screening and intensity, 

whereas for F1 samples, 2 μl of DNA was used for Microsporidia MB screening and intensity 

(Section 2.3.4). 

 

A modified method described by Santolamazza et al., (2008)  was used to confirm species 

identity. Molecular species identification was conducted only on G0 female samples.  A forward 

primer called LikeZianniF, 5’-TCGCCTTAGACCTTGCGTTA-3’ and reverse primer called 

LikeZianniR, 5’-CGCTTCAAGAATTCGAGATAC-3’ (Macrogen Europe, Amsterdam, the 

Netherlands) was used. PCR reactions were carried out in a 10 μl reaction, which contained 

0.5 μl of each primer, 2 μl of Blend master mix Ready-to-load, 5 μl of Nuclease-free PCR water 

(Life Technologies Corporation, 2130 Woodward St. Austin, TX 78744 USA [LOT number 

2107464]), and 2 μl of template DNA extracted from a single G0 mosquito. Thermocycler 

conditions were 95°C for 15 mins followed by 40 cycles of 95°C for 30 s, 60°C for 45 s and 

72°C for 45 s, with a final elongation at 72°C for 5 mins, and a 4°C hold. The resulting products 



22 

 

were analysed on 1.5% agarose gel (1.5 g of agarose in 100 ml of Tris-acetate-EDTA buffer) 

stained with ethidium bromide (5 μl of stock solution per 100 ml gel), with low and high 

molecular weight bands corresponding to fragments containing or lacking Gel score: An. 

arabiensis amplicon size =164 bp (Figure of Gel image in Appendix 1). No template control 

(NTC), two positive known controls, one for An. arabiensis and the other for An. Gambiae, 

were included. 

 

2.3.4 Microsporidia MB screening and intensity determination 

The term Microsporidia MB intensity in this thesis, refers to the relative abundance or density 

of the symbiont in the mosquito. Quantitative PCR (qPCR) was performed using DNA 

extracted from mosquitoes as described in section 2.3.3. The DNA samples of G0  and F1 were 

analysed to identify if they were infected with Microsporidia MB using specific primers 

(MB18SF: 5’-CGCCGG CCGTGAAAAATTTA-3’ and MB18SR: 5’-CCTTGGACGTG 

GGAGCTATC-3’) that target the Microsporidia MB 18S rRNA region (Herren et al., 2020). 

The PCR detection process volume was 10 µl. This solution included 2 µl of HOT FIREPol 

Blend Master mix Ready-To-Load (Solis Biodyne, Estonia with Catalogue number: 04-27-

00115), 0.5 µl of forward and reverse primers at a concentration of 5 pmol/µl, 2 µl of the DNA 

template, and 5 µl of nuclease-free water. The thermocycling protocol consisted of an initial 

denaturation step at 95°C for 15 minutes, followed by 35 cycles of denaturation at 95°C for 1 

minute, annealing at 62°C for 90 seconds, and extension at 72°C for 60 seconds. The final 

autoextension was performed at a temperature of 72°C for 5 minutes. Microsporidia MB-

positive samples underwent relative qPCR analysis to measure Microsporidia MB relative 

intensity (Herren et al., 2020). The presence of Microsporidia MB in each sample was 

confirmed with a distinctive melt curve related to the Microsporidia MB, MB18SF/ MB18SR 

primers. The qPCR utilised the MB18SF/ MB18SR primers, with normalisation performed 

using the Anopheles ribosomal protein S7 gene (primers, S7F: 5’-



23 

 

TCCTGGAGCTGGAGATGAAC-3’ and S7R: 5’-GACGGGTCTGTACCTTCTGG-3’) as the 

reference gene. The qPCR was performed using a MIC qPCR cycler (BioMolecular Systems, 

Australia). The qPCR technique was employed to determine both the prevalence and intensity 

of Microsporidia MB in the experimental mosquitoes. The following controls were used, no 

template control (NTC), one known Microsporidia MB infected (positive control), double 

distilled water (negative control).  

 

2.3.5 Grouping of mosquitoes for rearing 

The F1 larvae and adults originating from confirmed An. arabiensis G0 females and infected 

with Microsporidia MB were either combined (Group lines (GLs)) called MB+ GLs or reared 

separately (Isofemale lines (IMLs) called MB+ IMLs depending on the experiment designs 

(described below). Similarly, uninfected larvae and adults without Microsporidia MB were 

combined (called MB- GLs) or reared separately to serve as controls (called MB- IMLs) for 

the experiments. It must be noted that MB- IMLs were uninfected lines from an infected G0 

used as controls. All experiments were carried out on either F1 larvae or adults. 

 

2.3.6.1 Effect of Microsporidia MB on developmental fitness under different larval diet 

regimes  

The selection of diet regimes was obtained based on a preliminary dose-response experiment 

(Dr Tullu Bukhari, personal communication). Four treatments or diet regimes (details given 

below) was used on both MB+ and MB- GLs- (Figure 5). Hence, as indicated in grouping 

naming in section 2.3.5, the larvae were either Microsporidia MB-infected (called MB+ GLs) 

or uninfected Group lines (MB- GLs).  Every regime had three biological replicates. Each 

biological replicate consisted of 60 unfed 24 h-old An. arabiensis larvae that were placed in a 

larval tray (21 × 15 × 8.5 cm) filled with 1L of distilled water. There were three types of diets  

(Takken et al., 2013): Tetramin® baby fish diet (Tetra GmbH, Germany), Cerelac® baby diet is 



24 

 

a powdered milk (Nestle Co. Ltd.), and GoCat® diet ( Purina®, United Kingdom) that was 

formulated into four diet regimes. The nutritional content of each diet is available in Table 1. 

Two doses of Tetramin were tested i.e., 0.3 mg per larva per day (mg/larva/d)  and 0.07 

mg/larva/d, and one dose of Cerelac and GoCat was tested i.e., 0.3mg/larva/d. Tetramin flakes 

were ground to reduce the size of the flakes into granules, a routine larval-rearing practice in 

the laboratory (Linenberg et al., 2016). Tetramin 0.3 mg/larva/d diet is the reference and served 

as a positive control diet regime and 0.07 mg/larva/d Tetramin diet was tested to determine how 

low quantity diet availability during larval development (from first instar to pupae) influences 

the effect of Microsporidia MB on mosquito life table parameters ( Yan et al., 2021). This is it 

because larval diet in nature is most likely to be a reduced quantity diet. The diet regime added 

to the larval trays was adjusted to the number of larvae that remained in the larval trays (There 

were 60 larvae per tray and hence the initial measurement was made for 60. However, as the 

larvae numbers reduce upon pupation or mortality, the amount of food was adjusted so that the 

diet regime remains fixed: Either 0.3 or 0.07 mg/larva/d). The following lifetable parameters 

of Microsporidia MB or the mosquito host were recorded for the GLs:  

i) Microsporidia MB prevalence and intensity: Prevalence was calculated by the 

number of mosquitoes with Microsporidia MB presence divided by the total number 

of larvae in each regime.  All the mosquitoes in each diet regime were screened for 

Microsporidia MB presence and intensity (described in section 2.3.4). Emerging 

adults from each treatment group were fed on 6% glucose solution ad libitum) and 

harvested on day 3 for Microsporidia MB screening. The three days were allocated 

to assess the influence of the larval diet regimes of the GLs on Microsporidia MB 

intensity, as the intensity during the larval stages (first, second, third, and fourth 

instar stages [L1, L2, L3, and L4]) to pupae is highly variable (Nattoh et al., 2021). 



25 

 

ii) Larval mortality: The number of dead larvae was recorded and removed daily and 

data used to record larval mortality 

iii) Larval developmental time: Time taken for larvae to develop from first instar to 

pupae in days was recorded daily. 

iv) Adult emergence: The number of pupae that emerged as adults was recorded daily. 

Adult emergence rate was calculated using the formula below 

number of  emerged adults

Total number of larvae per treatment
𝑥 100).  

 

 

 
Figure 5: Experimental design to determine the effect of different larval diet regimes on 

larval development, mortality and adult emergence of Microsporidia MB infected An. 

arabiensis as well as the prevalence and intensity of Microsporidia MB. MB+ and MB- 

stands for Microsporidia MB infected and uninfected larvae respectively. 

 

 

 

 

 



26 

 

 

 

Table 1: Nutritional components of Tetramin, GoCat and Cerelac diets 

Main Nutritional Composition of diet 

Cerelac GoCat Tetramin 

Protein         14% Protein          30% Protein          46% 

Fat content   9.6% Fat content    10% Fat content    11% 

Fiber total    1.5% Crude fibre     3% Crude fiber     2% 

Sodium        0.12   

 Linoleic acid 1.8%  

  Moisture         6% 

Vitamin A, B, D & K   

 Arachidonic acid 0.1%  

  Phosphorus 1% 

Zinc (trace amount)   

  Vitamin C (446 mg/kg) 

  Omega-3-fatty acid 500 

mg/kg 

 

 

2.3.6.2 Effect of Microsporidia MB on adult mosquito survival under different adult diet 

regimes 

Larvae from the MB+ GLs were reared on Tetramin 0.3 mg/larva/d for this experiment. This 

diet regime was selected based on results from previous experiments (Figure 5). At least 80 

adults from either Microsporidia MB infected or uninfected GLs were used in the experiment. 

The adult mosquitoes (<1 day old) were divided into two treatment groups (n=40 per cage) and 

placed in separate cages (15 × 15 × 15 cm). In one cage, mosquitoes had ad libitum access to 

1% glucose solution and in the second cage, mosquitoes had ad libitum access to 6 % glucose 

solution (Figure 6). The glucose solution was provided in a vial and mosquitoes were fed 

through a paper towel wick. The glucose vial with a paper wick was replaced every two days. 

The above procedure was conducted concurrently for both MB+ and MB- GLs. Daily adult 

mortality was recorded. The removed dead mosquitoes were preserved in 70% isopropanol and 

then frozen at -20°C. All the dead mosquitoes were screened for Microsporidia MB (section 

2.3.). The prevalence and intensities of Microsporidia MB infection in adults (combined for 



27 

 

males and females) from each diet were recorded. This experiment was conducted in six 

biological replicates for each diet regimes  

 

 

Figure 6: Experimental design to determine the effect of Microsporidia MB on adult 

mosquito survival under different adult diet regimes 

 

2.3.7 Effect of larval and adult diet quantity on Microsporidia MB intensity in Isofemale 

lines (IMLs) after vertical transmission  

To determine if the effect of larval diet (different quantities of the same diet) and adult diet on 

Microsporidia MB intensity was not influenced by the genetic background of GLs, the two 

experiments were conducted with isofemale lines (offspring from the same G0) referred to as 

IMLs (Figure 7). 

 

2.3.7.1 Effect of larval diet quantity on Microsporidia MB intensity in IMLs  

For this experiment (Figure 7A), the two larval diet regimes from Tetramin diet: 0.07 mg and 

0.3 mg per larva/d were used (Yan et al., 2021). Larvae from IMLs (Those infected with 

Microsporidia MB called MB+ IMLs while uninfected larvae were called MB- IMLs [being 

uninfected lines from an infected G0] were divided into two trays and reared on either a low 

quantity diet (Tetramin 0.07 gm/larva/d) or high quantity diet (Tetramin 0.3 mg/larva/d). 



28 

 

 

This experiment was carried out in nine biological replicates, with each replicate consisting of 

at least twenty unfed 24 h-old IMLs larvae (first instar larvae [L1s]) in a larval tray (21 × 15 × 

8.5 cm) with 1 L of distilled water. The diet regime added to the larval trays was adjusted to 

the number of larvae that remained in the larval trays. Dead larvae were removed daily and 

recorded. Pupae were collected daily and placed in a 15 × 15 × 15 cm cage. The adults that 

emerged were fed on 6% glucose solution for three days before being harvested for the 

following records: 

a) Microsporidia MB intensity: The intensity of Microsporidia MB was determined as 

described in section 2.3.4. 

b) Wing lengths: Twenty adults (10 males and 10 females) were harvested on day 3 

from each group and used to determine wing length as a proxy for body size 

(Huestis et al., 2011; Maïga et al., 2012). The mosquito was placed on a microscope 

slide under a dissecting microscope. The left wing was cut using a surgical blade 

and fine-tipped dissection forceps. If the left wing was damaged in the process of 

cutting, the right wing of the same mosquito was then used. The wing length was 

measured in mm from the aluta to the most apical point using a Dino-Lite® Premier 

handheld microscope at a magnification of 32.1 (Huatang Optical Industry Co., Ltd, 

Taiwan). This was done to determine the difference between MB+ and MB- IMLs 

males and females respectively. 

 

 



29 

 

 

Figure 7: (A) Experimental design to determine the effect of different larval diet quantity 

on Microsporidia MB intensity in the isofemale lines of An. arabiensis after vertical 

transmission (MB+). (B) Experimental design to determine the effect of adult diet 

quantity on Microsporidia MB intensity in isofemale line of An. arabiensis after vertical 

transmission. MB+ stands for An. arabiensis infected with Microsporidia MB. 

 

 

2.3.7.2 Effect of adult diet quantity on Microsporidia MB intensity in adult MB+ IMLs 

Similar to larval diet quantity, this experiment aimed to determine if the effect of adult diet on 

Microsporidia MB intensity was influenced by the group lines. The adults originated from 

IMLs larvae that were reared on 0.3 mg Tetramin. The IMLs that produced at least 40 adults 

were used for this experiment (Figure 7B). The adult MB+ were split in two groups of 20 

mosquitoes per cage (15 × 15 × 15 cm). Mosquitoes in one cage were fed on 1% glucose 

solution and in the second cage on 6 % glucose solution ad libitum respectively. Same was 

done for the adult MB- IMLs which served as controls. The glucose solution was replaced with 

a fresh one every two days. On day 14, the adult mosquitoes were harvested to quantify 

Microsporidia MB with qPCR (as per methods described in section 2.3.4). The 14 days is 

optimal for  measurement of the impact of the adult diet on MB intensity in mosquito (Makhulu 

et al., 2024).  



30 

 

 

2.3.8 Ethics 

A research permit was sought from the National Commission for Science, Technology and 

Innovation, Kenya before conducting the research (Ref No. 643670). Ethical approval for the 

collection of mosquitoes from households was obtained from The Scientific and Ethics Review 

Unit-Kenya Medical Research Institute (Non-KEMRI protocol number 4520). Ethics 

certificate number M220622 from the University of the Witwatersrand HREC, South Africa 

was also obtained (Appendix 2). 

 

2.3.9 Data analysis 

Kaplan Meier survival analysis and Cox regression were used to determine the effect of 

different diet regimes on larval development and adult survival (Kaplan & Meier, 1992). 

Prevalence and adult emergence data was arcsine transformed and analysed using the Tukey’s 

honestly significant difference (HSD) test of Analysis of Variance (ANOVA) to determine the 

best-fit diet regime (KendallL, 1938). Non-parametric Mann-Whitney U and Kruskal–Wallis 

tests (Kruskal & Wallis, 1952) were used to compare the Microsporidia MB intensities of the 

adult and larval diet treatment groups after the data  did not pass the normality test ( Shapiro & 

Wilk, 1965). Phenotypic characteristics such as length of wing was tested for significant 

variation within and across treatment groups using the Generalised Linear Model following 

gamma distribution (Nelder & Wedderburn, 1972). R software version 4.1.2 (Giorgi et al., 

2022) was used for analysis with a p-value < 0.05 considered significant. 

 



31 

 

 

2.4 RESULTS 

 

2.4.1 Species identification of G0 mosquitoes 

The table below shows the summary of the three weeks G0 mosquitoes collected from the 

field whose F1offspring were used for all four experiments (2 for GLs and 2 for IMLs). 

Table 2: G0 mosquitoes collected from the field and the number of females confirmed to 

be An. arabiensis and Microsporidia MB infected. 

Experiment 

designated for 

Number of G0 

mosquitoes that laid 

eggs (Out of a total of 

900 G0 collected for 

each week) 

Number of 

individuals 

confirmed as An. 

arabiensis (%) 

Number of G0 

An. arabiensis 

Microsporidia 

MB infected 

mosquitoes (%) 

 

GLs for larvae  500 450 (90) 52 (11.5) 

GLs for adult 

survival  

303 300 (99) 38 (12.6) 

IMLs (Both larvae 

and adult) 

410 410 (100) 43 (10.5) 

 

 

 

 



32 

 

 

2.4.2 Effect of Microsporidia MB on developmental fitness under different larval diet 

regimes 

There were two main experiments under this section as described in section 2.3.6.1 for different 

larval diet regimes (for MB+ GLs and MB- GLs) and 2.3.6.2 for adult diet impact on survival 

of MB+ GLs and MB- GLs. 

 

2.4.2.1 Prevalence and intensity of Microsporidia MB under different larval diet regimes 

Larval diet regimes did not affect the prevalence of Microsporidia MB in emerging MB+ GLs 

adults. There was no statistically significant difference in Microsporidia MB prevalence across 

the four diet regimes (ANOVA after arcsine transformation of the data, F = 2.655, df = 

3, P = 0.185) (Figure 8A). In contrast, the intensity of Microsporidia MB was affected by 

different larval diet regimes (Kruskal Wallis test, X2 = 22.85, df = 3, P < 0.0001). Multiple 

pairwise comparisons showed that three-day-old adults emerging from the larvae feeding on 

Tetramin 0.3 mg/larva/d had the highest intensity of Microsporidia MB when compared to 

those feeding on Tetramin 0.07 mg/larva/d (P= 0.007) and Cerelac 0.3 mg/larva/d (P< 0.001) 

(Figure 8B). There was no significant difference between Microsporidia MB intensity under 

Tetramin 0.3 and GoCat 0.3 mg/larva/d regimes. Furthermore, there was a significant 

difference between GoCat 0.3 mg/larva/d and Cerelac 0.3 mg/larva/d (P<0.006), or between 

Tetramin 0.07 mg/larva/d and Cerelac 0.3 mg/larva/d (P> 0.254).  



33 

 

 

Figure 8: Microsporidia MB prevalence and intensity under different larval diet regimes 

of MB+ GLs. (A) Prevalence (%) of Microsporidia MB in 3-day-old F1 adults reared on 

different diet regimes. Error bars represent standard deviation.  (B) Relative 

Microsporidia MB intensity in 3-day-old F1 adults reared on different diet regimes using 

qPCR. Bar represents significant difference between the diet regimes. The number of 

independent biological replicates was three with each replicate made up of 60 larvae.  The 

corresponding average number of infected individuals are below the x-axis out of the total 

mosquitoes that emerged in the biological replicates of 60 each. Asterisks show the level 

of significance (*P < 0.05, **P <0 .001, and ***P < 0.0001), while ns indicates no 

significance. 

 

2.4.2.2 Effect of Microsporidia MB on larval mortality, pupation and adult emergence 

under different larval diet regimes  

The highest larval mortality was observed when the larvae were fed on Tetramin 0.07 

mg/larva/d. The MB+ GLs cohort showed a 26% larval mortality compared to the MB- GLs 

(28.33% mortality) (Figure 9A). MB+ GLs larvae fed on GoCat 0.3 mg/larva/d showed a lower 

larval mortality of 4.45% mortality, while MB- GLs showed a 6.67% mortality. Furthermore, 

MB+ GLs fed on Cerelac (0.3 mg/larva/d) showed much lower larval mortality of 1.70% and 

MB- GLs with 2.78% mortality (Figure 9A). Tetramin 0.3 mg/larva/d resulted in a statistically 

significantly higher median survival with the lowest larval mortality of 1.67 % in MB+ GLs 



34 

 

larvae compared to 15 % mortality in MB- GLs (Log Rank [Mantel-Cox], X2 = 15, df = 

1, P < 0.001). Microsporidia MB presence in the larva significantly reduced MB+ GLs larval 

mortality when the larvae were fed 0.3 mg/larva/d diet regime for both Tetramin and GoCat 

(Log Rank [Mantel-Cox] X2 = 15, df = 1, P < 0.001 and X2 = 5.43, df = 1, P = 0.012 

respectively).  

 

The larval development results showed that MB+ GLs larvae developed faster than their MB- 

GLs counterparts under certain diet regimes (Figure 9B). The Microsporidia MB growth rate 

enhancement advantage to the mosquito host was diet dependent. MB+ GLs larvae only 

developed significantly faster than their MB- GLs counterparts when fed on the Tetramin 0.3 

and GoCat 0.3 mg/larva/d diet regimes (Hazard ratio (HR) = 1.7, 95% CI = 1.4–2.2, P < 0.001 

and HR = 1.4, 95% CI = 1.1–1.7, P < 0.01, respectively). MB+ GLs larvae fed on Tetramin 

(0.3 mg/larva/d) diet regime had a shorter median development time (days) ± (standard 

deviation) of 9 ± 0.06 d while the MB- GLs controls was 10 ± 0.15 d (Figure 9B). The hazard 

ratio of 1.7 showed that MB+ GLs larvae developed 1.7 d faster than MB- GLs. The median 

development time for MB+ GLs larvae was shorter than MB- GLs when fed on GoCat 0.3 

mg/larva/d as well (10 ± 0.12 d and 11 ±0.14 d respectively). The median time for Cerelac 0.3 

mg/larva/d was 12 ± 0.12 d and 11± 0.21 d for MB+ GLs and MB- GLs larvae, respectively. 

However, with Tetramin 0.07 mg/larva/d, the median development time between the MB+ GLs 

and MB- GLs larvae were 15 ± 0.52 d and 15 ± 0.72 d, respectively. 

 



35 

 

 

Figure 9: Effect of Microsporidia MB on mortality and larval development under different 

larval diet regimes of GLs. (A): Effect of Microsporidia MB on larval mortality under 

different diet regimes and (B) Microsporidia MB’s effect on median larval development 

time under different diet regimes. Total number of larvae were 180 in three biological 

replicates (n=60) for each diet regime. Error bars represent standard deviations. 

 

 

Finally, under this section, the different diets significantly affected the number of GLs larvae 

that ultimately emerged as adults (F = 4.66, df = 3, P = 0.005, Figure 10). However, regardless 

of the treatment, there was no significant difference between MB+ verses MB- GLs mosquitoes 

(ANOVA after arcsine transformation of data F = 4.66, df = 3, P = 0.99). 



36 

 

 

Figure 10: Microsporidia MB’s effect on adult emergence under different larval diet 

regimes of GLs. The data shown is the percentage of larvae that developed into adult 

mosquitoes from the various diet regimes. This is the total number of adults that emerged 

from the larvae from each diet. Error bars represent standard deviation. 

 

2.4.2.3 Effect of Microsporidia MB on adult mosquito (GLs) survival under different adult 

diet regimes  

Adults used for this experiment originate from F1 larvae fed on the same larval diet regime of 

Tetramin (0.3 mg/larva/d). The median survival time (days) ± (standard deviation) for MB+ 

and MB- GLs adults fed on 1% glucose diet (Figure 11A), was 4 ±0.17 d and 4 ± 0.18 d, 

respectively. There was no statistically significant difference in the median survival between 

MB+ and MB- GLs adults (Log-rank, [Mantel-Cox] test, X2 = 2.95, df = 1, P = 0.073). 

On the other hand, those fed on 6% glucose (Figure 11B) showed that MB+ GLs mosquitoes 

survived significantly longer than their MB-GLs counterparts (Log-rank, [Mantel-Cox] test, 

X2 = 5.84, df = 1, P = 0.007). The median survival time for the MB+ GLs adults was 12 ± 0.49 

d while that for uninfected was 10 ±0.47 d.   



37 

 

 

Figure 11: Effect of Microsporidia MB on GLs adult mosquito survival under different 

adult diet quantity regimes (A) 1% glucose diet & (B) 6% glucose diet. The dashed lines 

are the 95% CL interval while the solid lines are the survival curve for each regime.  

 

 

2.4.3 Effect of larval and adult diet quantity on Microsporidia MB intensity and wing size 

in Isofemale lines (IMLs) after vertical transmission 

2.4.3.1a Effect of larval diet quantity on the Microsporidia MB intensity of MB+ IMLs 

MB+ IMLs larvae, which were fed on Tetramin 0.07 mg/larva/d and Tetramin 0.3 mg/larva/d 

concurrently (Figure 12B), showed that the intensity of females was significantly affected by 

the two larval diets with the later having higher intensity compared to the lower diet quantity 

(Kruskal-Wallis, X2 = 6.38, df = 1, P = 0.011). Meanwhile, the males (Figure 12A) showed no 

significant difference in intensity between both diets (Kruskal-Wallis, 

X2= 0.30, df = 1, P = 0.584). The mean Microsporidia MB intensity ± (standard deviation) of 

male mosquitoes for 0.07 mg/larva/d was 191.4 7S/18S and 0.3 mg/larva/d was 197.6 7S/18S 



38 

 

whiles that for females were 230.7 7S/18S and 469.6 7S/18S for 0.07 mg/larva/d and 0.3 

mg/larva/d regimes respectively (Figure 12B). 

 

 

Figure 12: Microsporidia MB intensity under different larval diet quantity of MB+ IML. 

(A) Microsporidia MB intensity of adult males and (B) Microsporidia MB intensity of adult 

females that emerged from larvae reared on Tetramin 0.07 mg/larva/d and 0.3 

mg/larva/d. The dotted lines represent the mean intensity of each diet regime.  

 

2.4.3.1b Effect of Larval diet quantity on the wing length of MB+ and MB- IMLs 

MB+ IMLs larvae fed on Tetramin 0.07 mg/larva/d resulted in significantly smaller wing size 

(given the size and variation parameter) of adult mosquitoes when compared to the 0.3 

mg/larva/d regime (mm) (GLM following family gamma, t = 4.23, df = 1, P < 0.0001, Figure 

13). Males fed on 0.07 mg/larva/d and 0.3 mg/larva/d regimes had relatively smaller wing sizes 

compared to the females on the same treatments (GLM following family gamma, t = 5.93, df 

= 1, P < 0.0001). MB+ IMLs female mosquitoes had significantly longer wing lengths under 

both Tetramin 0.07 mg/larva/d (Kruskal-Wallis, X2 = 4.18, df = 1, P= 0.040) and 0.3 mg/larva/d 

(Kruskal-Wallis, X2= 23.21, df = 1, P < 0.0001) compared to the respective male counterparts.  



39 

 

Furthermore, the wing length of the female MB+ IMLs mosquitoes was significantly larger 

compared to the control MB- IMLs (uninfected IMLs from the same infected mother) when 

fed with Tetramin 0.3 mg/larva/d diet (Kruskal-Wallis, X2= 23, df = 1, P< 0.0001, Figure 13). 

The males under the same diet did not show any difference in size (U(64) = 3.58, P = 0.115 ).  

 

Figure 13: Effects of Microsporidia MB on wing length of MB+ & MB- IMLs under larval 

diet quantity regimes. Microsporidia MB significantly increased the length of female 

wings both under Tetramin 0.07 and 0.3 mg/larva/d respectively. The line and the bars in 

the middle of each treatment indicate the mean with 95% CI respectively.  

 

 

2.4.3.2 Adult diet quantity on the Microsporidia MB intensity of MB+ IMLs 

The analysis conducted here adhered to the methodology outlined in section 2.4.3.1. However, 

in the absence of significant results among the groups (Figure 14A & B), I proceeded to 

incorporate analysis of each diet quantity by sex. There was no significant difference in 

Microsporidia MB intensity between adult males fed on 1% and 6% glucose diet (Figure 14A) 

or females only (Figure 14B) MB+ IMLs mosquitoes (U(62)=434, P = 0.524 and U(85)= 

734, P = 0.153 respectively). Under a low adult quantity (1% concentration) diet, there was no 



40 

 

significant difference between male and female Microsporidia MB intensity (U(39)= 

7.4, P = 0.430) (Figure 14C). However, when adults were fed with the 6% diet, females had 

significantly higher Microsporidia MB intensity compared to their male IMLs counterparts 

(Figure 14D, U(46)= 12.57, P < 0.009). 

 

 

Figure 14: Microsporidia MB intensity under different adult diet quantities of MB+ IMLs 

(A) Effect of low adult diet (1% glucose) on Microsporidia MB intensity in male and 

female IMLs mosquitoes. (B) Effect of high adult diet (6% glucose) on Microsporidia MB 

intensity in male and female IMLs mosquitoes. (C) Effect of low (1% glucose) and high 

(6% glucose) diet on the intensity of Microsporidia MB of male mosquitoes. (D) Effect of 

low (1% glucose) and high (6% glucose) diet on the intensity of Microsporidia MB of 

female mosquitoes. Bars represent the mean with a 95% confidence interval of the 

Microsporidia MB intensities of each diet regime. 

 

 
 



41 

 

 

2.5 DISCUSSION 

This is the first study to investigate how host diet regimes and quantities affect the interaction 

between An. arabiensis and Microsporidia MB.  Understanding the dynamics of this interaction 

is important for predicting the ability of Microsporidia MB to spread naturally in An. arabiensis 

populations. This knowledge is also imperative for establishing optimal methods to rear 

Microsporidia MB-infected An. arabiensis for future vector control implementation strategies 

(Bourtzis et al., 2014; Carvajal-Lago et al., 2021).  

 

Some life history traits were not affected by either Microsporidia MB or diet. The prevalence 

of Microsporidia MB was not significantly affected by diet. However, diet conditions or quality 

could play an essential role in ensuring successful transmission with a significant increase in 

the intensity of the symbiont.  The inference that can be made from the intensity result is that 

in the absence of Tetramin 0.3 mg/larva/d diet regime, GoCat 0.3 mg/larva/d can be used to 

achieve a comparable high Microsporidia MB intensity in the mosquito host. The low-intensity 

results seen under Tetramin 0.07 and Cerelac diets could be attributed to the low protein content 

by quantity (0.07 mg) and composition (14%) ingested by the larvae, respectively.  

 

Microsporidia MB reduced larval mortality in An. arabiensis across four different diet regimes, 

with the greatest impact observed under the Tetramin 0.3 mg/larva/day and GoCat 0.3 

mg/larva/day diets. Regarding larval development time, Tetramin 0.3 mg/larva/day resulted in 

the fastest development, followed by GoCat 0.3 mg/larva/day, Cerelac, and then Tetramin 0.07 

mg/larva/day. MB+ GLs larvae pupated 1.7 days faster than MB- GLs (without the symbiont) 

under the Tetramin 0.3 mg/larva/day diet, though no significant difference was observed under 

Tetramin 0.07 mg/larva/day, highlighting the role of diet quality and quantity in development 



42 

 

(Araújo et al., 2012; Couret et al., 2014). Larval diet conditions impacted the total number of 

adult mosquitoes that emerged from first instar larvae. However, Microsporidia MB did not 

affect the emergence of adults from larvae under any of the four diet regimes, aligning with the 

findings from Herren et. al., (2020). 

 

To conclude, amongst the larval diet regimes tested, Tetramin 0.3 mg/larva/d led to the lowest 

larval mortality, fastest larval development, and highest adult emergence, supporting previous 

research indicating that the diet of a mosquito influences the life history characteristics of the 

mosquito(Carvajal-Lago et al., 2021; Linenberg et al., 2016; Vrzal et al., 2010). I hypothesise 

that Tetramin’s higher protein content, compared to Gocat and Cerelac, was the main dietary 

factor contributing to the observed differences in development, as prior studies have shown 

that a protein content above 50% is essential for achieving larger size and weight in mosquitoes 

(Carvajal-Lago et al., 2021).  These results suggest that both the mosquito host and the 

symbiont are influenced by diet, with the fitness benefits conferred by Microsporidia MB being 

dependent on dietary conditions. Therefore, Tetramin 0.3 mg/larva/day is the most suitable diet 

for mass-rearing Microsporidia MB mosquitoes.  

 

The survival of MB+ GLs adult mosquitoes was also influenced by adult diet concentration. 

MB+ GLs adult mosquitoes survived longer compared to MB- GLs mosquitoes, reaffirming 

the diet-dependent nature of the symbiont’s fitness benefits, a pattern similar to that seen in 

Rickettsia-infected whiteflies (Himler et al., 2011). It is also worth noting that sugar 

concentration in the adult diet did not affect the intensity of Microsporidia MB in the 

mosquitoes but rather