1 Chapter 1 Overview of the use of aromatic plants and their essential oils to treat microbial infections. 1.1 Introduction There can be no doubt that antimicrobials are one of the most important therapeutic agents available today (Clark, 1996). The importance could most likely be attributed to the fact that most modern treatment regimens for ailments are effectively managed by pharmaceuticals. However, the treatment of infectious diseases focuses on a cure rather than symptomatic relief or merely pharmacological management. The impact of improved quality of life and increased life expectancy with chemotherapeutic agents has brought man-kind into an era of antibiotic dependency. Man?s reliance on these antimicrobial agents has led to their global misuse. This has resulted in untreatable infections due to emerging antimicrobial resistance. There are a number of international strategies in place to try and curb this surge of antimicrobial resistance. Infection control, antibody-based therapies and rational use of antibiotics are currently accompanied by the parallel development of new antimicrobial agents. Scientific advances have gone full circle and are presently probing plant derived antimicrobials to seek evidence corroborating the traditional use that has been holistically instilled into the cultural evolution of indigenous people. The importance and need for further research into traditional herbal medicine was first officially recognized by the World Health Organization (WHO) in 1978 (Kim, 2005). Subsequently, the WHO declared the period 2001-2010 as the decade for African Traditional Medicine. Initiatives were set up to prioritise medicinal plant screening with the aim to initiate an African Pharmacopoeia (AACHRD, 2002). In the WHO document for Traditional Medicine 2002-2005, a strategy was suggested to address reliability of methods, safety, quality, access and rational use. In a follow-up presentation in 2005, the global outcomes were presented (WHO, 2005). With the worldwide emphasis on traditional medicines, the basic screening and more in-depth research protocols into natural products cannot be over emphasised. 2 1.2 A historical perspective Historically plants have provided a good source for the treatment against microbial infections. References to the therapeutic application of aromatic plants used for antimicrobial purposes are well documented in ancient history. Hieroglyphic depictions have been found (Figure 1.1 and Figure 1.2) of essential oils used for anointing the sick. Traces of frankincense oil, found in an urn were discovered in Tutankhamen?s tomb. Their use in embalming processes suggests that the preservative nature of the oils were beneficial in warding off putrefaction. In ancient times, it has generally been the essential oils rather than extracts that were considered effective (Rios and Recio, 2005). Records, written on clay tablets dating back to about 2600 BC indicate that amongst others, oils from Cedrus (Cedar) and Commiphora (Myrrh) species were used for ailments ranging from coughs and colds to inflammation (Gurib-Fakim, 2006). The oldest written information from the Egyptians is the Ebers Papyrus. The origin dates back to around 1500 BC. The Ebers Papyrus is a medical ?handbook? documenting therapeutic treatments and pharmaceutical plant preparations. It has been said that the diagnostic precision is very accurate (Heinrich et al., 2004). Accounts of distillation by the Egyptians, Indians and Persians over 2000 years ago have been documented and the first authentic written account of essential oil distillation is ascribed to Villanova ca. 1235-1311 (Burt, 2004). Figure 1.1 Figure 1.2 Figure 1.1 and Figure 1.2 Egyptian hieroglyphic depictions of essential oils used for anointing and medicinal purposes. Figure 1.1 depicts an Egyptian with an incense burner preparing oils for inhalation (http://www.lotuspress.com/lotusbrands/tiferetonline/key.html). Figure 1.2 illustrates the administration of oils for the enhancement of body processes (http://www.ansononline.com/cgi-bin/ez-catalog/cat_display.cgi?0X367200). 3 Biblical citations of medicinal plant use account for at least 30 references with myrrh and frankincense the most noteworthy (Cowan, 1999). The antimicrobial attributes of frankincense (Boswellia spp.) was especially favoured by the Greeks, Romans, Babylonians, Persians and Hebrews where it was used as a treatment for wounds, skin diseases, urinary tract infections and gynaecological disorders. Myrrh (Commiphora spp.) was used in embalming because of its effectiveness in preventing bacterial growth. It was also valued for its effectiveness in the treatment of topical conditions (Dharmananda, 2003; Marshall, 2004). One of the earliest reports of the use of plant material for antimicrobial use was when the European ?iceman? was first discovered. Traces of Piptoporus betulinus, a bracket fungus was found accompanied by the preserved human dating back to 3300 BC. Capasso (1998) postulated that the plant had antimycobacterial as well other microbial inhibitory activities. Later, the Greeks and Romans played a significant role in documenting the use of medicinal plants. The Greek scholar Pedanius Dioscorides (1 BC) described more than 600 medicinal plants which became the prototype of modern pharmacopoeias (Cowan, 1999; Heinrich et al., 2004). Hippocrates, the well-known Greek medical doctor (460-375 BC) to whom the Hippocratic Oath is attributed, fully supported the use of herbal medicines and advocated their use. After the conquest of the Roman Empire, Greek medical texts were translated to Arabic, and then later copied by the Christians. Printed European reports on medicinal plants through the 16th century were still very strongly influenced by the Greek-Roman sources. Herbal documentation became available in various languages and became the driving force of European phytotherapy. During the middle ages aromatic plants were used extensively to combat the plague. Widespread use of perfumed candles, pomanders (plant material adorned the neck of wearers) and burning of oils were believed to offer some immunity to the caregivers of the sick. During this time, plant treatment regimens were greatly contradicted. On the one hand monks were respected for tending to the sick, using herbal medicine (Heinrich et al., 2004). On the other hand the village herbalists were persecuted for witchery. During the renaissance period there was a revival of holistic treatments. It is about this time that the term ?pharmacognosy? was born. The word is derived from the two Greek words 4 ?pharmakon? meaning drug and ?gignosko? meaning to acquire knowledge. The focus of pharmacognosy studies during this time and up to the 20th century were on medicinal substances derived from the oils, waxes, gums and resins of plants (Kinghorn, 2001). It was highly thought of as the mother of all present-day pharmaceutical disciplines (Gurib-Fakim, 2006). With the emergence of antibiotic discovery in 1939, plants as antimicrobials were overlooked in favour of more potent treatment regimens. With the introduction of antibacterial compounds such as penicillin and streptomycin, effective therapy against Staphylococcus and other infections were met (Rice, 2006). These discoveries led to the era of wonder drugs where the ?single, golden bullet? approach to combat diseases were the first line of therapy. 1.3 Present use of antimicrobials and the impact on disease Antimicrobials have been used successfully for over six decades. At least 30 new antimicrobial agents have been introduced into mainstream chemotherapeutic treatment regimens within the first 18 years of the antibiotic era. This has been marked by cycles, comprising of the introduction or development of an antibiotic followed by the subsequent emergence of resistance (Swartz, 2000). Given the current problem with the development of bacterial resistance there is a great surge to discover new potent antibiotics. The present world market for antibiotic drug expenditure on finding new anti-infective agents is expected to increase by 60% from spending levels in 1993. The continued interest in finding newer anti-infectives are exacerbated by emergence of resistance and its impact on the global population encompassing both developed countries and third world countries. Initially antimicrobial resistance was mostly evident in hospital environments where nosocomial infections became prevalent due to the consistent exposure to antimicrobials. More recently the spread of resistance has become community- acquired. The available epidemiological data suggests that levels of antimicrobial resistance have reached unacceptable proportions and trends show further increases (Okeke et al., 2005). Infectious diseases still account for approximately half of all deaths in tropical countries (Iwu et al., 1999). Microbial infections contribute towards the primary morbidity and mortality statistics from hospitals around the world (Rice, 2006). These mortality rates for developed countries are increasing in spite of adequate hygiene, sanitation and education. It has been estimated that as many as 30% of people in industrialized countries still suffer from diarrhoeal-related diseases (Kosek et al., 2003). 5 It is also estimated that approximately 80% of the total global population comprise of developing nations. These developing nations often have poorer sanitation and hygiene levels which contribute to infection susceptibility, Even though the success of antibiotics has benefited mankind, many poorer countries have not had accessibility. The burden of infectious diseases falls heavier on developing countries where it has been estimated that approximately 60% of all deaths are related to communicable diseases. This is in sharp contrast to the 35% mortality from developed countries (Ist?riz and Carbon, 2000). Furthermore, third-world countries are subject to a shortage of funds, adequate medical care and lack the resources for newly developed medicine, which make them dependent on their natural resources. The increase of numerous infectious diseases that have occurred on a global level have further emphasised the need to continue with antimicrobial studies (Cordell and Colvard, 2005). In addition, the recent limitations on the use of antibiotic feeds in farming and the call for the reduction of antimicrobial chemical additives in foods are now regulated more than ever before. Thus the search for natural antimicrobials has become increasingly popular (Devlieghere et al., 2004). 1.4 Ethnomedicinal plants as antimicrobials Phytochemicals have found their way into the arsenal of antimicrobial drugs prescribed by physicians. The sale and use of herbal products worldwide has increased dramatically with Germany as the leading country where 80% of medical practitioners are prescribing herbal remedies (Gilani and Rahman, 2006). Natural products of botanical origin represent almost 40% of all drugs in clinical use. Phillipson (2001) estimated that over 50% of the top 20 drugs used on today?s world market could be linked to natural product research. The modern pharmacopoeias contain at least 25% plant derived drugs (Kim, 2005). The primary benefit of sourcing plant material for medicines is that they are perceived to be safer, less rigorous and more affordable than synthetic alternatives (Iwu et al., 1999). In spite of the successes of conventional antimicrobials sourced from microbial products, botanical substances are still researched for potential biological activity. Borris (1996) hypothesised that plants and micro-organisms are complementary, both systems are natural and have a diverse species base, however, each produce different metabolites and therefore research into both systems are necessary for antimicrobial screening. 6 Rios and Recio (2005) commented on the increase in published articles relating to medicinal plants for antimicrobial therapy. During 1966-1994, there were 115 articles found on Pubmed. In later years (1995-2004) the authors noted that the references more than doubled to 307. Where essential oils were researched, references increased to 323 in 2005 (Rios and Recio, 2005). A recent literature Pubmed search undertaken for ?plants? and ?antimicrobial? yielded 324 citations exclusively for 2006, in comparison with the earlier publications (256 articles in 2003). Similarly using Science Direct, only 44 papers were published in 2002 in comparison with the 84 papers published in 2006. These estimates confirm the exponential rise in antimicrobial ethnobotanical research. The rapid increase in research in this field is possibly due to the increased interest in overcoming antimicrobial resistance to conventional antimicrobials. In addition, improved assessment techniques within pharmacognosy as well as the diversity of naturally occurring metabolites allows for a number of studies of this nature (Clark, 1996). 1.5 Essential oils as antimicrobials Many essential oils extracted from plants are known for their antimicrobial properties and considered the most widely used natural product to date (Nakatsu et al., 2000). The most important antimicrobial application of essential oils are within the fields of dermatology, gastritis, respiratory complaints, wound healing and genital infections (Neuwinger, 2000). Arnald de Villanova, a physician in the 13th century, is believed to be one of the first doctors to have used essential oils therapeutically. By the 17th century essential oils were used extensively by pharmacists where they were applied topically for healing (Hili, 2001). The antiseptic properties of essential oils have been researched since the 1800?s, however, during the course of the 19th and 20th centuries the use of essential oils for fragrance and flavour became a more prominent experimental focus rather than the therapeutic application (Burt, 2004). More recently there has been a renewed interest in essential oil research. The resurgence of aromatherapy and aromachology assisted in launching essential oils into the international markets where they are sold as ?natural antibiotics?. The essential oil of Melaleuca alternifolia (tea tree) is probably the best known example, which is sold for its antimicrobial properties. A number of studies have been conducted on the essential oil of M. alternifolia. Antiviral activities were investigated where the oil significantly decreased a number of local lesions on Nicotiana glutinosa caused by the Tobacco Mosaic virus (Bishop, 1995). Cytotoxicity has also been reported (S?derberg et al., 1996; Hayes et al., 1997; Hammer et al., 2006) but the predominant focus has been on the antimicrobial potential of the oil (Carson et al., 1995; Caelli et al., 2000; Cox et al., 2001; 7 Christoph and Stahl-Biskup, 2001). Other popular antimicrobial commercial oils include Lavendula angustifolia (lavender), Mentha piperita (peppermint) and Rosmarinus officinalis (rosemary) and Thymus vulgaris (thyme). Developing concurrently to this every day use of essential oil products for antimicrobial properties is the scientific exploration which aims at validating the numerous properties ascribed to these natural molecules (Dharmaratne et al., 1999; Iwu et al., 1999). There have been numerous studies on the antimicrobial activities of essential oils in specific plant species. A compendium for the Journal of Essential Oil Research (Lawrence, 1989-2005) documents over ninety antimicrobial publications. These papers include studies on bacteriostatic activities (Clark, 1996; Christoph and Stahl-Biskup, 2001), antifungal activities (Griffin et al., 2000; Dawson-Andoh et al., 2000) and antiviral activities (Bishop, 1995; Chao and Young, 2000). Methodology has been studied extensively due to the problems associated with the volatility and immiscibility of the oil. Diffusion and minimum inhibitory assays (MIC) have pre-dominated many publications and to a lesser extent time-kill methodology (Remmal and Tanaoui-Elaraki, 1993; Hood et al., 2003). The antimicrobial activity of essential oil components has been studied extensively (Hinou et al., 1989; Knobloch et al., 1989; Pauli and Kubeczka, 1997; Pattnaik et al., 1997; Chalchat and Garry, 1997; Nakatsu et al., 2000; Pauli, 2001). Detailed compositional analysis is achieved by gas chromatography and mass spectrometry where the essential oil may yield any number of major constituents. The essential oil constituents that show the highest antimicrobial activity are the phenols followed by the aldehydes, ketones, alcohols, esters and finally the hydrocarbons (Kalemba and Kunicka, 2003). A recent publication (van Zyl et al., 2006) demonstrated the biological activities for 20 essential oil constituents from the seven different structural groups i.e. phenols, aldehydes, ketones, alcohols, esters, terpene hydrocarbons and oxides. Highest antimicrobial activities were found for carvacrol from the phenol group, confirming results from previous studies (Kalemba and Kunicka, 2003). While the study of the antimicrobial properties of essential oil constituents play an important role in determining the overall activity of the plant, it must be taken into account that any of the constituents (which may be more than 70 in any given plant studied) may be responsible for activity. These may be either major or minor constituents or even a combination of constituents that interact synergistically to enhance activity of the plant as a whole. Some in vitro antimicrobial combination studies have been undertaken which support the role of synergism in essential oils, but antimicrobial studies have not been undertaken in great depth (Kang et al., 1992; Lachowicz et al., 1998; Cassella et al., 2002). 8 The objective to include plant-derived antimicrobials in ingestible products to meet the needs of the so called ?green consumer? desired by many Westerners has also become very popular (Burt, 2004). Essential oils have become multifunctional i.e. in the reduction and / or elimination of food-borne pathogens, as a preservative and as flavouring agents in food. The use of spices for antimicrobial prophylaxis has been well documented. In a review on the antimicrobial properties of essential oils in food, Burt (2004), listed 36 studies reporting the efficacious use of oils as preservative in various food stuffs. Essential oils, because of their organoleptic nature and flavouring properties meet the culinary need and with their potential medicinal role may play an active role in present and future food preservation. Such applications could contribute significantly as synergistic agents that not only enhance food flavouring but also work prophylactically to ward off undesirable microbial populations. Due to the known antimicrobial properties of essential oils, an interest has developed in aromatic plants in general, in the continued search for new natural molecules to be used in phytopharmaceuticals (Rates 2001, Cowan 1999). 1.6 The South African perspective Southern Africa boasts a unique and diverse botanical heritage with over 30 000 plant species of which ca. 3 000 species are used therapeutically (van Wyk et al., 1997). Not only is the South African flora rich in diversity but it is also mostly endemic (Mulholland, 2005). In addition to this unique botanical heritage, South Africa has a cultural diversity with traditional healing being integral to each ethnic group. African traditional medicine is the oldest medicinal system, hence culturally often referred to as the Cradle of Mankind (Gurib-Fakim, 2006). The informal verbal record of medicinal use of plant material by the Khoi-San, Nguni, Zulu and the Sotho-speaking ethnic groups has been passed on from generation to generation (Hutchings and van Staden, 1994; van Wyk et al., 1997). There are over 200 000 traditional healers in South Africa who systematically diagnose the cause of disease and thereafter administer plant medicines to holistically improve the health of the patient (van Wyk et al., 1997). Methods such as this are not very different to the western medical system. However, the added benefit of the traditional healer to seek the origin of the ailment makes the use of cultural healing more favourable for many people. There is also the added benefit of availability at low cost, as many ?muthi? medicines are sold informally. It has been estimated that over 20 000 tons of plant material are sold annually on the informal medicinal market (Taylor et al., 2001). 9 Even though conventional drug therapies are the first choice of treatment in first world countries, herbal therapy is a way of life for over 80% of people living in rural South Africa (Shale et al., 1999). Indigenous medicinal plant use is well recorded in the readily available local ethnobotanical literature. Despite the well-documented ethnobotanical literature, very little scientific information (e.g. efficacy, phytochemistry) has become available on indigenous medicinally used plants. It is only recently (1999-2006) that a number of findings have emerged on the chemistry and biological activity of plants used in traditional healing. South African contributions within The Journal Ethnopharmacology for the period 1980-1994 were between 10-20%. However, this increased to approximately 55% in the last five years (Light et al., 2005). This recent emergence in the scientific validation of South African medicinal plants can possibly be attributed to the number of citations in local books confirming the need for such studies (Hutchings and van Staden, 1994; van Wyk et al., 1997). The National Research Foundation has recently (2005) committed 15 million rand to the Indigenous Knowledge Focus area emphasizing the need to make pharmacognosy-related studies a high priority (Mulholland, 2005). More specifically, the antimicrobial properties of South African plants have been sorely neglected. The literature search engines ?Science Direct? and ?Scopus? identified only 4% and 3% (respectively) of citations related to ?plants? and ?antimicrobial? were of South African origin. While South African researchers seem to be lagging behind on a global scale, the antimicrobial investigation of plants compare favourably with researchers from the rest of Africa (Light et al., 2005). An antimicrobial review of the South African literature reveals a broad spectrum of research activity. Van Wyk (2002), includes a number antimicrobial directives in his review on the ethnobotanical research in southern Africa including social relevance, wound healing, and antidiarrhoeal activity. In addition, there have been a number of antimicrobial screening publications (Rabe and van Staden, 1997; Lin et al., 1999; McGaw et al., 2000; Kelmanson et al., 2000; Motsei et al., 2003; Eldeen et al., 2005; Buwa and van Staden, 2006; McGaw and Eloff, 2005). Antimicrobial reports have been undertaken on specific species such as Helichrysum (Meyer and Afolayan, 1995; Meyer and Dilika, 1996; Meyer et al., 1997; Dilika et al., 1997; Afolayan and Meyer, 1997; Mathekga and Meyer, 1998; Lourens et al., 2004; van Vuuren et al., 2006) where research groups have focused their expertise on specific plant groups. There has also been local research activity on medicinal plants used to treat sexually transmitted 10 infections (Tshikalange et al., 2005) and mycobacterial studies (Lall and Meyer, 1999). Other research groups have focused their studies on the microbiological activity of particular botanical families e.g. the Combretaceae (Eloff, 1999; Martini et al., 2004; Eloff et al., 2005). Furthermore, there have been significant contributions toward the methods associated with biological testing of botanical material. Of the numerous methodology publications undertaken, the most prominent has been the minimum inhibitory concentration determination publication by Eloff, 1998a which has been extensively cited. Other microbial related botanical studies have involved the isolation of antimicrobial compounds from South African plant species. This has been successfully undertaken where the need for a multi-disciplinary approach is integral to achieve high quality research. Collaboration between plant chemists, botanists and microbiologists have yielded some valuable contributions (Rabe and van Staden, 2000; Drewes et al., 2005; Drewes et al., 2006). Other studies focusing on specific parts of the plant such as roots, stems or leaf material have also been investigated (Louw et al., 2002; Lewu et al., 2006). Even though South African researchers have made valuable contributions towards ethnopharmacological investigations, the need exists to capitalize on the rich botanical diversity and meet the global need for the compilation of significant monographs. This is necessary in order to meet The World Health Organization?s objective to provide guidelines for the assessment of herbal medicines. The examination of the antimicrobial properties of aromatic plants endemic to South Africa and used in traditional healing forms an integral focus of this thesis. One of the main objectives is to make a significant contribution towards generating microbiological data to validate the traditional use and thus ultimately contribute towards the acceptance of plant-based antimicrobials in westernized healthcare systems. 1.7 Thesis structure The thesis is introduced by a general introduction followed by individualised studies on nine different plant species. In order to avoid methodology repetition reference is made to preceding Chapters. A flow chart (Figure 1.3) presents how each study was systematically followed by initial plant collection and identification. Thereafter chemical analysis of the essential oil for 11 each specific species was undertaken. Antimicrobial screening and viability monitoring followed with further in-depth target assay applications. Findings from each investigation have been presented at a conference, published or have been submitted for review. Abstracts of publications are given in the Appendix. A selection of indigenous oils is comparatively evaluated with popular commercial oils to correlate antimicrobial efficacies. A general methodology Chapter follows which encompasses an overview of all methods followed as well as trouble shooting, literature sourced and many of the challenges encountered when studying the various different plant species. A final conclusion is provided to integrate all Chapters and provide some recommendations and guidelines for further ethnopharmacological antimicrobial studies. Figure 1.3 A schematic outlay of the study protocol for the antimicrobial investigation of aromatic plants. Antimicrobial properties of aromatic medicinal plants Botanical aspects Collection and identification of plant material Structure activity relationships Essential oil constituents Antimicrobial activity Antimicrobial assays (e.g. disc diffusion/MIC/ time-kill) Target interactions (e.g. synergy/ bioautography) Chemical analysis 12 1.7.1 Study objectives The overall intention of this study is to provide a scientific basis for the use of selected traditional aromatic plants used therapeutically for conditions associated with infectious diseases. Within this framework, the following objectives were proposed; 1. To determine the antimicrobial activity of selected indigenous aromatic plants used in traditional healing. 2. To record the essential oil composition of some of the most widely used South African medicinal aromatic plants. 3. To investigate the plant composition with antimicrobial activity relating to structure activity relationships and if possible identify the antimicrobial factor responsible. 4. To study specific pharmacological interactions. 5. To compare the antimicrobial activity (potency) of indigenous aromatic plants to commercially available essential oils with claimed antimicrobial activity. 6. To compare and validate results obtained from various test methodologies (e.g. disc diffusion, MIC and time-kill studies). 7. To provide a scientific rationale for the traditional use of various medicinal aromatic plants to treat infectious diseases. 13 Chapter 2 Myrothamnus flabellifolius Welw., antimicrobial efficacy determined by disc diffusion, minimum inhibitory concentration and time-kill methodology. 2.1 Introduction Myrothamnus flabellifolius Welw. (Myrothamnaceae) is a plant used traditionally by many of the ethnic people of Africa. It has been used medicinally by the Pedi, Zulu, Shona healers and early Rhodesian settlers (Watt and Breyer-Brandwijk, 1962). The use of leaves and stems as flavouring agents has also been reported (van Wyk and Gericke, 2000). Furthermore, the use of M. flabellifolius to treat psychological tribulations is strongly linked to the unusual nature of the plant to revive to its original vegetative state even when completely desiccated. In the rainy season the seemingly dead plants are simply revived from the dormant state. This remarkable feature of the plant is used by the traditional healer to explain the healing action of the plant. The patients are told to take the dry plant home with them, being a symbol of both their psychological and physical state. They are told to place the dried plant in water and as the plant revives to it?s original state, so too will the patient become well again. This property is reflected in the English (resurrection bush) and Zulu (uvukwabafile) vernacular names and is used as a symbol of hope in traditional African treatment against severe depression (Credo Mutwa, personal communication to Prof. B-E van Wyk). Due to the conflicting reports on the composition of the essential oils distilled from Myrothamnus flabellifolius, the oil composition has been re-examined. Some of the major compounds previously identified in the essential oils are pinocarvone and trans-pinocarvone (Chagonda et al., 1999) and camphor, ?-pinene and 1,8-cineole (van Wyk et al., 1997). The antimicrobial properties are investigated here using disc diffusion, minimum inhibitory concentrations (MIC) and time-kill methods. 14 2.2 Botanical description Myrothamnus flabellifolius is a small woody shrub about 0.4 m in height. The small aromatic leaves (Figure 2.1) are fan-shaped and conspicuously toothed on the upper leaf margin. Plants are monoecious bearing inconspicuous flowers. The leaves have a strong aromatic character. The plants have an affinity for rocky areas and are usually wedged into crevices of large boulders as illustrated in Figure 2.2 (van Wyk et al., 1997). Figure 2.1 Aromatic foliage of Figure 2.2 Myrothamnus flabellifolius in habitat. Myrothamnus flabellifolius. 2.3 Distribution Myrothamnus flabellifolius grows on rocky outcrops in the northern regions of South Africa from Gauteng, extending through Mpumalanga to the Northern Province and Botswana. It is also found in central and northern Namibia (Figure 2.3). 2.4 Medicinal uses The leaves and twigs are used in many medicinal preparations. The main traditional uses related to microbial infections as summarized from van Wyk et al. (1997) and Hutchings et al. (1996) include: infusions taken for colds, respiratory ailments and decoctions taken orally to alleviate kidney problems. Externally, the plant may be used to treat abrasions and the dried powdered 15 leaves are used in dressings for burns and wounds, possibly to prevent sepsis. Lotions incorporating M. flabellifolius are prepared and applied to skin abrasions (Watt and Breyer- Brandwijk, 1962). Leaves may be burnt and the smoke inhaled to treat chest pains and asthma. The smoke may also be directed into the vagina to treat infections and pains in the uterus (van Wyk and Gericke, 2000). The local Pedi people of northern South Africa smoke the leaves in pipes to alleviate chest pains. The Karanga communities chew the aromatic leaves for mouth complaints. In central Africa the plant is used as a tonic and as a treatment for breast complaints. Shona healers have used the plant to treat epilepsy, madness and coughs. 2.5 Methods 2.5.1 Chemical aspects Plant collection: The aerial parts were collected in the growing season (March) from a population growing on rocky slopes in Klipriviersberg, Johannesburg, South Africa. A voucher specimen (Table 2.1) is retained in the Department of Pharmacy and Pharmacology, University of Witwatersrand. Due to the large volume of oil required for time-kill studies plant material was collected from several plants in the same population. Figure 2.3 The geographical distribution of M. flabellifolius in South Africa (SANBI). 16 Table 2.1 Plant collection data for M. flabellifolius. Gas chromatography (GC): Due to conflicting reports of the major constituents within M. flabellifolius (Da Cunha, 1974; Gibbs, 1974; Chagonda et al., 1999) and the large amounts of oil required for the antimicrobial testing the oil obtained from each of the plants were analyzed by GC to determine any chemical variation. Analysis was performed on a Shimadzu 17A gas chromatograph using the following parameters; Column: J&W-DB1 (60 m x 0.25 mm id., 0.25 ?m film thickness); Temperatures: injection port 230 ?C, column 60 ?C for 1 min, 5 ?C / min to 180 ?C, 180 ?C for 2 min, (total = 25 min). A small amount (1 ?l) of oil sample was injected with 0.85 ?l hexane (Rochelle) into a stream of an inert gas (helium), where the separated components then emerge from the column at discrete intervals (characteristics for each component) and pass through a flame ionization detector (FID). Chromatograms were recorded from the time of initial elution of compounds. The retention times were used for qualitative identification with peak r areas to provide qualitative information. Plant Voucher Material distilled (g) Essential oil yield (% w/w) M. flabellifolius SVV934 2700.7 0.7 Figure 2.4 Clevenger apparatus used for distil- lation of plant material. Distillation of essential oils: Distillation protocols were undertaken according to Guenther (1948). Fresh plant material was hydrodistilled in a Clevenger type apparatus (Figure 2.4) and the essential oil collected after three hours. A round bottom flask of five litre capacity is packed with fresh leafy plant material. Approximately 500 mL distilled water is added and heated by means of an electronic heating mantel. The essential oils were then condensed in a cooling system where they were collected. The temperature was kept constant by letting the water run through the condenser. Thereafter, the oil was allowed to stand at room temperature to allow for good separation and collected in amber vials for storage at 4 ?C. 17 Gas chromatography combined with mass spectrometry (GC-MS): Oil samples (0.1 ?l in 0.9 ?l hexane) were further quantitatively analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the following conditions: column; HP-Innowax (60 m x 0.25 mm id., 0.25 ?m film thickness) was used with helium as a carrier gas; Temperatures: injection port 250 ?C, column 60 ?C for 10 min, 4 ?C / min to 220 ?C, 220 ?C for 10 min, 1 ?C / min to 240 ?C (total = 80 min). The split ratio was adjusted to 50:1. The MS were taken at 70 eV. The mass range was from m/z 35 to 425. Data was recorded after a 5 min lag phase. Relative percentages of the separated compounds were calculated from total ion chromatogram by the computerized integrator. The MS were compared with those of reference compounds and confirmed with the aid of retention indices from published sources using the in-house Ba?er Library of Essential Oil Constituents in collaboration with Prof Ba?er at the Department of Pharmacognosy, Faculty of Pharmacy, Anadolu University, Eski?ehir, Turkey. 2.5.2 Antimicrobial aspects Culture and media preparation: The NCCLS (2003) guidelines were used to ensure that accurate microbiological assay and transfer techniques were followed. All stock cultures were obtained from the National Health Laboratory Services (NHLS) with the exception of Candida albicans, which was obtained from the South African Bureau of Standards (SABS). Authorization to keep and use cultures for research purposes was obtained from the Department of Health (reference: J1/2/4/16 NO1). Table 2.2 lists the cultures with corresponding reference number (where possible) used in this study. Stock cultures were retained at ?20 ?C, subcultured onto Tryptone Soya (Oxoid) agar, incubated at optimum temperatures and checked for purity. Isolated pure colonies were selected and transferred onto Tryptone Soya (Oxoid) agar and thereafter kept viable by subculturing weekly for stock culture maintenance. All media was prepared according to the instructions provided by the supplier i.e. weighed, dissolved in distilled water and autoclaved (Butterworth) at 121 ?C for 15 min. After sterilization, all media was pre-incubated to confirm sterility before further use. Disc diffusion assays: Disc diffusion assays as described by Janssen et al. (1987), Pauli and Kubeczka (1997) and Hewitt and Vincent (2003) and were performed on the hydrodistilled oils. Base layers of Mueller Hinton (Oxoid) agar were prepared for bacterial studies and depending on the size of the assay plate either 15 mL or 100 mL of pre-sterilized molten agar was poured into a sterile petri dish. Care was taken so that the agar was maintained at a working temperature of 18 50-55 ?C. A temperature higher than 55 ?C may result in either killing or reducing the viable count of the test organism. Temperatures lower than 50 ?C will result in the agar rapidly solidifying making uniformity of the agar surface problematic. Similarly, disc diffusion assays for fungal strains were prepared with the exception that Sabouraud?s Dextrose (Oxoid) agar was used. Spore suspensions yielding an approximate inoculum size of 1 x 108 CFU/mL were thoroughly mixed into the overlaying agar surface of identical volume. With aseptic manipulation, 6 mm discs were saturated with the pure oil (approximately 9 mg) and were placed onto the set agar. Neomycin discs (30 ?g, Oxoid) were used for bacterial controls and Nystatin discs (100 IU, Oxoid) were used for fungal controls. These positive controls were included in all test replicates and repetitions to ensure susceptibility of micro-organisms to conventional antimicrobials. Negative controls i.e. sterile discs were used to ensure no pre-existing antimicrobial activity was evident. All plates were allowed to pre-diffuse at 4 ?C for one hour to allow for the dispersion of essential oil from the disc into the set agar before incubation at 37 ?C for 24 hr, with the exception of the two yeasts Candida albicans and Cryptococcus neoformans which were incubated at 37 ?C for 48 hr and the moulds Aspergillus niger and A. alternata which were incubated at 25 ?C for seven days. Results were recorded at least in triplicate and the mean of the zone of inhibition measured from disc edge to margin of culture growth is documented in Table 2.4. Minimum inhibitory concentrations (MIC): After preliminary screening by disc diffusion assay, the broth MIC micro-dilution bioassay was determined (Carson et al., 1995; Eloff, 1998a; NCCLS, 2003). All bacterial cultures were subcultured from stock agar plates and grown in Tryptone Soya (Oxoid) broth overnight. Yeasts were incubated for a further 24 hr. Microtitre plates were aseptically prepared by the addition of 100 ?l distilled, sterile water into each well. The hydrodistilled M. flabellifolius oils at starting stock concentrations of 128 mg/mL were subsequently transferred into the first rows of a microtitre plate. Essential oils were made up to concentration with acetone (Merck) as the solvent of choice. Negative controls (acetone in microtitre plate without antimicrobial) were included in each assay. Serial dilutions are performed on the M. flabellifolius oils so that concentrations of 32, 16, 8, 4, 2, 1, 0.5 and 0.25 mg/mL were obtained. The overnight cultures were diluted in fresh Tryptone Soya broth at a 1:100 ratio, yielding an approximate inoculum size of 1 x 108 colony forming units (CFU)/mL and 100 ?l was added to all wells. The microtitre plates were sealed with sterile adhesive sealing film (AEC Amersham) to ensure that no loss of essential oil took place. Optimal (37 ?C for 24 hr for bacteria and 48 hr for yeasts) incubation conditions followed. 19 Test organism MIC controls (?g/mL) Pseudomonas aeruginosa (ATCC 9027) 0.25-2.50*2 Escherichia coli (ATCC 25922) 0.004-0.016*1 Salmonella typhimurium (ATCC 14028) 0.63-12.50*2 Proteus vulgaris (clinical strain) 0.25-128*2 Serratia odorifera (ATCC 33132) 0.30-1.25*2 Staphylococcus aureus (ATCC 25923) 0.12-0.50*1 Enterococcus faecalis (ATCC 29212) 0.25-2.00*1 Cryptococcus neoformans (ATCC 90112) 0.625-2.500*2 Candida albicans (ATCC 10231) 1.25-2.50*2 Aspergillus niger (clinical strain) ND*3 Alternaria alternata (clinical strain) ND *3 Positive bactericidal controls i.e. ciprofloxacin (Sigma-Aldrich) at starting stock concentrations of 0.01 mg/mL were included in each assay to confirm the antimicrobial susceptibility. The ciprofloxacin stock solution was prepared aseptically with sterile water. The amphotericin B (Sigma-Aldrich) stock solution for yeast controls was initially diluted with dimethyl sulfoxide (DMSO, Merck) as recommended by the supplier and thereafter diluted with sterile water to a starting concentration of 0.01 mg/mL. The selection of ciprofloxacin and amphotericin B as suitable reference controls is in support of the NCCLS guidelines (2003). A review article (Lahlou, 2004), on the methods in assessing the bio-activity of essential oils also cites a number of publications in support of ciprofloxacin and amphotericin B as standard controls. Conventional antibiotics inhibit microbial growth at certain acceptable quantitative MIC ranges. Any variations to the acceptable MIC values indicate either strain resistance (if the MIC value is higher than that from the range) or unusual strain susceptibility (if the MIC value is lower than the range). This would impact on the ?true? results obtained for the test and therefore monitoring these ranges is important to corroborate the accuracy of results if further clinical studies are to be Table 2.2 Microbial organisms with corresponding reference numbers and acceptable MIC ranges for controls (ciprofloxacin for bacteria and amphotericin B for yeasts). *1 NCCLS, 2003; Andrews, 2004. *2 Acceptable ranges as determined within the Department of Pharmacy and Pharmacology, University of Witwatersrand (controls in this study were run for all test organisms even though only one organism from that indicated in bold is recommended for inclusion in each MIC assay). *3 No MIC?s were performed on moulds due to the extended incubation requirements. 20 undertaken. Documented control ranges are only available for selected pathogens, however, all assays were run with micro-organism controls of the particular pathogen studied. If control guidelines were unavailable, at least six repetitions and replications on controls were done to achieve a standard guideline from which comparison could be made. Table 2.2 gives the acceptable ranges for ciprofloxacin and amphotericin B against pathogens used in the MIC assays. A 0.4 mg/mL p-iodonitrotetrazolium violet (Sigma-Aldrich) solution (INT) was prepared and 40 ?l transferred to all inoculated wells. The use of INT to determine the end point MIC value is based on the principle of detection of dehydrogenase activity where the metabolically active test organism reduces INT to a red-purple colour. The microtitre plates inoculated with bacteria were examined after six hours to determine a colour change in relation to concentration of microbial growth. The yeasts were examined after 24 hr. The lowest concentration having no colour change is defined as the MIC. Minimum inhibitory assays were done in triplicate and the mean tabulated (Table 2.4). Time-kill methodology: Death kinetic assays, as described by Christoph and Stahl-Biskup (2001) were performed on a Gram-positive organism S. aureus (ATCC 25923), a Gram-negative organism P. aeruginosa (ATCC 9027) and a yeast C. albicans (ATCC 10231). Stock cultures were grown in Tryptone Soya (Oxoid) broth and centrifuged (MSE) for 10 min at 5000 rpm. The supernatant was discarded and the pellets resuspended in a 0.9% NaCl (Labchem) solution. Oil concentrations of 0.125, 0.5, 0.75, 1 and 2% (1.8, 5.2, 8.1, 9.5 and 18.1 mg/mL respectively) were incorporated into 50 mL Tryptone Soya broth with 0.5% Tween 80 (Saarchem) and a final inoculum of approximately 1 x 108 CFU/mL. The different concentrations were incubated at 37 ?C with agitation in a shaking water bath (Labotec). At pre-determined time intervals ranging from 0 min - 24 hr, (0, 5, 15, 30, 60, 120, 240 min and 24 hr), aliquots of 1 mL were transferred to 9 mL inactivation broth consisting of 0.1% peptone (Oxoid), 5% lecithin (Merck) and 5% yeast extract (Oxoid). A further four serial dilutions were performed from the inactivation broth into 9 mL of 0.9% NaCl. From each dilution 100 ?l was plated onto Tryptone Soya (Oxoid) agar. The plates were incubated (37 ?C for 24 hr for bacteria and 48 hr for the yeast) and colony- forming units (CFU/mL) counted. Death kinetics were expressed in log10 reduction time-kill plots. 21 Negative controls were included in all time-kill assays having the same broth formulation, inoculum and Tween 80 but without the oil. This control validated the assay conditions so that the test organism?s growth could be monitored when no inhibitor was added. Positive controls, ciprofloxacin for bacterial assays, or amphotericin B for fungal assays at a concentration of 0.063% (0.6 mg/mL) were included in all time-kill assays. This positive control was included to ensure microbial death response to conventional antimicrobials. Each time-kill plot represents the mean of values where the kinetics were determined in duplicate and each figure represents the time period at which most activity could be noted. Various steps in the method are seen in Figure 2.5 and 2.6. Various concentrations of essential oil Inactivation broth with oil sampled at selected time intervals Dilutions made in 0.9% NaCl Each dilution is plated out onto Tryptone Soya agar 100 ?l Figure 2.5 A diagrammatic representation of the steps undertaken for each oil concentration carried out at all time intervals. 22 Figure 2.6 Death kinetic method as detailed clockwise: aseptic addition of M. flabellifolius into culture media; incubation in shaking waterbath; layout of plates for dilution; plating out of successive dilutions. 2.6 Results and discussion 2.6.1 Essential oil chemistry The GC analysis of individual plants within the Klipriviersberg locality indicated consistent chemical composition (data not shown) and consequently the oil obtained from individual plants were pooled and analyzed with GC-MS. Eighty-four compounds were identified in the hydrodistilled M. flabellifolius essential oil representing 86% of the total composition (Table 2.3). 23 RRI* Compound name % 1032 ?-pinene 1.6 1076 camphene 0.1 1118 ?-pinene 0.2 1132 sabinene tr 1174 myrcene 0.2 1203 limonene 6.1 1213 1,8-cineole 0.2 1218 ?-phellandrene tr 1220 cis-anhydrolinalool oxide tr 1224 o-mentha-1(17)5,8-triene 0.2 1280 p-cymene 0.2 1384 ?-pinene oxide tr 1408 1,3,8-p-menthtriene 0.1 1435 ?-campholene aldehyde tr 1452 ?-p-dimethylstyrene 0.2 1466 ?-cubebene tr 1468 trans-1,2-limonene epoxide tr 1482 fenchyl acetate 0.1 1497 ?-copaene 0.8 1528 ?-bourbonene tr 1535 ?-bourbonene 0.2 1536 pinocamphone 0.1 1541 benzaldehyde tr 1549 ?-cubebene 0.1 1553 linalool 1.8 1562 isopinocamphone 0.2 1565 linalyl acetate 0.2 1571 trans-p-menth-2-en-1-ol tr 1586 pinocarvone 11.1 1597 bornyl acetate 1.8 1600 ?-elemene 0.5 1611 terpinen-4-ol 0.4 1616 hotrienol 0.3 1624 cis-dihydrocarvone 0.1 1639 trans-p-mentha-2,8-diene-1-ol 2.7 1648 myrtenal 0.9 1658 sabinyl acetate 0.1 1661 alloaromadendrene 0.1 1661 trans-pinocarvyl acetate tr 1664 trans-pinocarveol 19.6 1678 cis-p-menth-2,8-dien-1-ol 2.3 1687 ?-humulene 0.1 1697 carvotanacetate tr 1704 ?-muurolene tr Table 2.3 The essential oil composition and integration percentage of M. flabellifolius as determined by GC-MS. 24 RRI* Compound name % 1706 ?-terpineol 0.3 1719 borneol tr 1726 germacrene D 1.1 1740 ?-muurolene 0.7 1747 trans-carvyl acetate 0.4 1751 carvone 1.1 1758 cis-piperitol 3.7 1773 ?-cadinene 0.6 1776 ?-cadinene 0.2 1797 p-menthyl acetophenone 0.1 1802 cuminaldehyde 0.1 1804 myrtenol 1.1 1807 perilla aldehyde 0.1 1811 trans-p-menth-1-(7)-8-diene-2-ol 7.4 1845 trans-carveol 1.3 1853 cis-calamenene 0.1 1865 isopiperitenone 0.3 1882 cis-carveol 0.2 1896 cis-p-menth-1-(7)-8-diene-2-ol 7.0 1941 ?-calacorene I 0.1 1945 1,5-epoxy-salvial-4(14)-ene tr 1953 palustrol 0.1 1956 p-isopropyl benzaldehyde 0.1 1984 ?-calacorene II 0.1 2008 caryophyllene oxide 0.1 2037 salvial-4(14)-en-1-one 0.1 2057 ledol 0.3 2069 germacrene D-4?-ol 0.1 2080 cubenol 0.3 2088 1-epi-cubenol 0.4 2104 viridiflorol 0.2 2113 cumin alcohol 0.1 2144 spathulenol 0.8 2187 T-cadinol 0.9 2209 T-muurolol 1.5 2219 ?-cadinol 0.6 2255 ?-cadinol 3.0 2264 intermedeol 0.2 2273 selin-11-en-4-?ol 0.4 2324 caryophylladienol II (=caryophylla-2(12),6(13)-dien-5?-ol) 0.1 TOTAL 86.00 *RRI: Relative retention indices as eluted from a polar column. % calculated from TIC data tr = trace <0.1%. 25 Major compounds: The two major compounds (Figure 2.7) in the essential oil of M. flabellifolius are trans-pinocarveol (19.6%) and pinocarvone (11.1%), representing 30.7% accumulatively. H3C CH3 CH2 O H3C CH3 OH CH2 pinocarvone trans-pinocarveol The major compounds in the essential oil of M. flabellifolius have previously been reported by Da Cunha (1974), indicating carvone and perillic acid as the major constituents. Gibbs (1974) recorded the presence of 1,8-cineole and diosphenol in the essential oil. Chagonda et al. (1999), identified 43 compounds from M. flabellifolius with trans-pinocarveol (28.7-28.8%), pinocarvone (13.4-21.3%), ?-pinene (tr-23.0%) and ?-selinene (5.0-9.9%) as major constituents. The major constituents (Figure 2.7) detected in this study could not corroborate with the results reported by Da Cunha (1974) and Gibbs (1974). The composition however, corresponds to the analysis reported by Chagonda et al. (1999) who also recorded trans-pinocarveol and pinocarvone as major constituents. Although no variation was recorded within the Klipriviersberg area, more populations representing the wide geographical distribution range of this plant may yield information on possible chemotypic variations. 2.6.2 Antimicrobial activity Myrothamnus flabellifolius have been used traditionally to treat various ailments, some associated with microbial infections (Neuwinger, 2000). The method of administration, for example the inhalation of smoke from the burnt leaves indicates the probability that essential oils are responsible for the biological activity. An investigation of the antibacterial and antifungal activity of M. flabellifolius against eleven different pathogens by disc diffusion assay (Table 2.4) Figure 2.7 Chemical structures for major compounds identified in the essential oil of M. flabellifolius. 26 showed inhibition for all micro-organisms except S. typhimurium and A. alternata. While results for bacterial isolates show minimal (1 mm for P. aeruginosa) to moderate sensitivity (3 mm for S. aureus, E. coli and S. odorifera), it was interesting to note that the fungal isolates C. neoformans and A. niger showed marked antimicrobial activity (8 mm and 10 mm respectively) in the essential oil of M. flabellifolius. These two fungal organisms are respiratory pathogens and may enter the host via the respiratory tract. Cryptococcosis is a chronic infection involving mainly the lungs and meninges and A. niger is implicated in bronchopulmonary lung infections (Boyd and Hoerl, 1981). As the smoke from M. flabellifolius is inhaled directly into the lungs, it provides some rational into its possible antimicrobial use in traditional medicine. The notable candidacidal activity reported here (5 mm) could possibly provide scientific support for the traditional use of M. flabellifolius for Candida-related infections (e.g. mouth and vaginal infections). Table 2.4 Zones of inhibition (measured in mm from disc edge to margin of culture growth) and MIC (mg/mL) of M. flabellifolius essential oil. *1 Neomycin and nystatin served as controls for bacteria and fungi respectively. *2 Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. *3 ND = Methods did not incorporate MIC?s for moulds. Test organism Disc diffusion (mm) Disc diffusion Control*1 MIC (mg/mL) MIC controls*2 Staphylococcus aureus (ATCC 25923) 3 7 4 0.30 x 10 -3 Enterococcus faecalis (ATCC 29212) 2 4 8 2.00 x 10 -3 Pseudomonas aeruginosa (ATCC 9027) 1 1 8 0.80 x 10 -3 Escherichia coli (ATCC 25922) 3 5 4 0.02 x 10 -3 Proteus vulgaris (clinical strain) 2 6 4 1.60 x10 -4 Serratia odorifera (ATCC 33132) 3 6 4 0.30 x 10 -3 Salmonella typhimurium (ATCC 14028) 0 4 4 0.80 x 10 -3 Candida albicans (ATCC 10231) 5 7 4 1.30 x 10 -3 Cryptococcus neoformans (ATCC 90112) 8 10 2 2.50 x 10 -3 Aspergillus niger (clinical strain) 10 10 ND *3 ND*3 Alternaria alternata (clinical strain) 0 2 ND *3 ND*3 27 In addition to disc diffusion assays, quantitative MIC assays was also undertaken on the same bacterial and yeast strains (Table 2.4). Congruency was mostly obtained between the two methods. Where disc diffusion assays showed moderate sensitivity (3 mm) similar moderate activities (4 mg/mL) were found in the MIC assay. Similarly, where lower sensitivities were obtained in the disc diffusion assay (1-2 mm), MIC studies indicated less activity (8 mg/mL). Results for C. neoformans indicated one of the highest sensitivities in the disc diffusion methodology (8 mm) and highest sensitivity in the MIC assay (2 mg/mL), however, some variation did occur. Salmonella typhimurium having no activity in the disc diffusion assay showed moderate sensitivity in the MIC assay (4 mg/mL) and conversely C. albicans having high sensitivity (5 mm) in the disc diffusion assay showed only moderate activity (4 mg/mL) in the MIC assay. Such method variations are not unexpected when investigating essential oils as results may be influenced by the choice of growth medium (agar in disc diffusion and broth in MIC), diffusion rate, oil volatility, water insolubility, and the oil complexity (Janssen et al., 1987). Researchers often find the reproducibility of plant-based antimicrobial studies problematic (Janssen et al., 1987; Pauli and Kubeczka, 1997; Njenga et al., 2005). To further substantiate the results obtained for the antimicrobial disc diffusion and MIC assays, time-kill studies were implemented. As this method involved a number of fine dilution parameters and was undertaken for the first time within the Department, tandem assays were undertaken using similar methods under the guidance of Prof. Klepser at the University of Iowa. This international collaboration to standardize and confirm results obtained from both laboratories were in agreement and further comparative analysis is discussed in Viljoen et al. (2002). Three susceptible organisms were selected to demonstrate the rapid onset of antimicrobial activity using time-kill methodology. The bactericidal activity of M. flabellifolius against S. aureus was investigated due to its comparatively high antimicrobial activity seen in the disc diffusion assay (Table 2.4). Staphylococcus aureus is associated with primary infections of the skin, such as boils, carbuncles and impetigo, thus providing some rationale for the traditional use to treat skin abrasions. Even though limited antimicrobial activity was noted for P. aeruginosa in the disc diffusion assay, further time-kill investigation was undertaken based on the hypothesis that P. aeruginosa, could possibly be the causative organism responsible for infections from burn wounds. As M. flabellifolius is traditionally used as a dressing for burns and wounds, possible 28 preventative anti-infective properties may play a role in therapeutic treatment. Lung disease and pseudomonial pneumonia also directly correlates with medicinal preparations inhaled by the ethnic Pedi community and warranted further antimicrobial investigation with P. aeruginosa. Candida albicans was selected for an in-depth fungal time-kill investigation due to its moderate antimicrobial susceptibility, as seen in the MIC assay (4 mg/mL) and susceptible profile noted in the disc diffusion assay (5 mm), but more due to the prevalent pathogenesis. More than 80% of all fungal bloodstream infections are a direct result from the Candida species and since the eighties there has been an explosive rise in the rate of Candidal infections (Klepser et al., 1998). The death kinetics of S. aureus, P. aeruginosa and C. albicans are represented as time-kill curves (log10 reduction values) in Figures 2.8, 2.9 and 2.10 respectively. Positive controls (ciprofloxacin and amphotericin B) presented with an immediate death response and due to presentation scale were not included in the time-kill plots. Negative controls, having no essential oil (indicated in black in Figures 2.8, 2.9 and 2.10) show a continuous growth pattern. For S. aureus (Figure 2.8), the essential oil generally followed a concentration-dependent antibacterial trend. An oil concentration of 0.125% and 0.25% resulted in a decrease in colonies in comparison with the control, however, no cidal action was noted. A bactericidal effect after four hours was observed at 0.75% (8.1 mg/mL) and a more rapid killing rate within one hour for M. flabellifolius oil at concentrations 1% (9.5 mg/mL) and 2% (18.1 mg/mL). The bactericidal effect noted here correlates to a degree with the MIC values (4 mg/mL) where minimal growth was still noted in the time-kill assay at 0.5% (5.2 mg/mL). For P. aeruginosa (Figure 2.9), death kinetics indicate a rapid decrease in the number of surviving cells within 60 min at higher concentrations (1% and 2%) but the CFU?s increased over time. Lower concentrations (0.125%-0.75%) also presented an initial decrease in CFU but not as prominently observed as in higher concentrations. This suggests that the oil produces an initial bacteriostatic effect but regrowth of the organism occurs with time. This regrowth seen in the time-kill assay gives a more detailed account of the essential oil action on P. aeruginosa, not seen in the disc diffusion nor MIC assays where the final limited antimicrobial effect was only observed after 24 hr. Tam et al. (2005), highlighted possible reasons for encountering regrowth during time-kill studies. Previous studies (Mouton et al., 1997) have postulated that regrowth may be attributed to the possibility of delayed growth of sub-populations. These organisms growing at a slower rate allow for resistance mechanisms to be transferred to other sub- populations thus enabling the population to withstand exposure of the antimicrobial test 29 substance. This would explain the initial reduction in colony forming units and later regrowth where microbial populations have been exposed to the essential oil over time. Another postulation explaining regrowth (Tam et al., 2005) is simply by means of adaptation where continuous exposure to the test antimicrobial eradicates sub-populations over time until, the most resistant microbial population becomes increasingly difficult to kill over time. All the oil concentrations were fungicidal for C. albicans (Figure 2.10), with rapid killing rates within the first four hours. The highest oil concentration (2%) was fungicidal within 5 min. The death kinetic comparison between the three organisms studied showed that the killing rate was greater for C. albicans, then S. aureus, and lastly only initial activity for P. aeruginosa. These time-kill studies correlate directly with the disc diffusion and MIC results, which as an initial screening procedure also showed antimicrobial activities highest for C. albicans and lowest for P. aeruginosa of the three pathogens studied. The value of time-kill methodology over other methods is clear as it demonstrates a more descriptive means of explaining the pharmocodynamic interaction of the oil with test organism over time, not seen with methods such as disc diffusion and MIC where results are observed at the end of an observation period. Even though the investigation is in vitro, a more realistic approach is given to what may possibly be occurring in an in vivo situation. 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 1.00E+06 1.00E+07 1.00E+08 1.00E+09 0 1 2 3 4 5 6 Time (hrs) CF U/ m l Control 0.125% 0.5% 0.75% 1% 2% Figure 2.8 Time-kill expressed in Log10 reduction of S. aureus exposed to M. flabellifolius essential oil within the first six hours. 30 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 1.00E+06 1.00E+07 1.00E+08 1.00E+09 1.00E+10 0 5 10 15 20 Time (hrs) CF U/ m l Control 0.125% 0.5% 0.75% 1% 2% Figure 2.9 Time-kill expressed in Log10 reduction of P. aeruginosa exposed to M. flabellifolius essential oil over 24 hr. 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 0 1 2 3 4 5 6 Time (hrs) CF U/ m ll Control 0.125% 0.25% 0.5% 1% 2% Figure 2.10 Time-kill expressed in Log10 reduction units of C. albicans exposed to M. flabellifolius essential oil within the first six hours. While it is not always the major essential oil constituents responsible for antimicrobial activity (Chalchat and Garry, 1997), it is however, interesting to tentatively correlate the observed antimicrobial activity to the essential oil composition for M. flabellifolius. Pinocarvone, a major constituent (11.13%) in M. flabellifolius essential oil is also one of the major constituents of Hyssop oil (Hyssopus officinalis), which is a popular home-remedy with bactericidal properties (Kerrola et al., 1994). Hinou et al. (1989) reported that trans-isomers have higher activities than the corresponding cis-isomers. trans-Pinocarveol and oxygenated bicyclic terpenoid, pinocarvone present in M. flabellifolius are both trans-isomers, possibly contributing to the 31 antimicrobial activities seen here. Kishore et al. (1996), found the essential oil of Chenopodium ambrosioides with trans-pinocarveol as a major constituent (27%) to be fungicidal against several dermatophytes. The high fungicidal activities noted in the disc diffusion assay (5-10 mm) for M. flabellifolius could possibly be attributed to the high concentration of trans-pinocarveol (19.57%). It is worthy of note that trans-pinocarveol, the major essential oil component in M. flabellifolius, is used in pharmaceutical preparations such as OzopulminTM to treat respiratory tract disorders, including asthma. The antimicrobial properties, with speculation on possible structure activity relationships reported here for M. flabellifolius essential oil, may provide some pharmaceutical rationale for the popular traditional use of this plant in African medicinal herbal preparations. 2.7 General conclusions ? The major constituents of the essential oil of Myrothamnus flabellifolius are trans- pinocarveol (19.6%) and pinocarvone (11.1%), representing 30.7% accumulatively. ? The highest antimicrobial activities obtained by means of the disc diffusion assay were noted for the fungal organisms Cryptococcus neoformans (8 mm) and Aspergillus niger (10 mm). ? The highest antimicrobial activity noted in the MIC assay was also for Cryptococcus neoformans (2 mg/mL). ? Death kinetics for the three organisms studied demonstrated that the killing rate was greatest for Candida albicans, then Staphylococcus aureus, and lastly only initial activity for Pseudomonas aeruginosa with regrowth after two hours. ? The ethnobotanical use of Myrothamnus flabellifolius to treat respiratory ailments correlates with the high antimicrobial activities found for Cryptococcus neoformans in the disc diffusion and MIC assay. 32 Chapter 3 Osmitopsis asteriscoides (P.J. Bergius) Less., the antimicrobial efficacy (disc diffusion, minimum inhibitory concentration, time- kill) and role of the major essential oil constituents. 3.1 Introduction Osmitopsis asteriscoides (P.J. Bergius) Less., (Asteraceae) is known locally as ?bels? or ?belskruie?. The plant is known for its abundance of essential oil which imparts a camphorus odour (Watt and Breyer-Brandwijk, 1962). Previous phytochemical work has been undertaken on Osmitopsis (Bohlmann and Zdero, 1974 in van Wyk et al., 1997; Bohlmann et al., 1985) which reported the presence of sesquiterpene lactones. More recently (Scott et al., 2004) described the morphology and phytochemical profile of O. asteriscoides extracts. Microbiological activity was investigated but this was restricted to the disc diffusion assay on aqueous extracts. This Chapter focuses on the microbiological activity and composition of the essential oil, which is the most characteristic feature of this highly aromatic plant. Although the antimicrobial activity of essential oils and their constituents are well-known (Hinou et al., 1989; Yousef and Tawil, 1980; Pattnaik et al., 1997; Hammer et al., 1999) the methods used to assess the in vitro antimicrobial action remains a topic of debate. This study has also documented the variability, reproducibility and accuracy in the results obtained using different methods (disc diffusion, MIC, time-kill and membrane confocal integrity studies). Furthermore the interaction between major compounds was assessed for antimicrobial activity. 3.2 Botanical description The leaves of the plant have numerous surface glands containing essential oil and when brushed the plant emits a strong eucalyptus-camphor smell. The plant is characterised by large thin, leaves and daisy-like white flowers as seen in Figure 3.1 (van Wyk et al., 1997). 33 Figure 3.1 The leaves and flowers of Osmitopsis asteriscoides. 3.3 Distribution Osmitopsis asteriscoides is restricted to the South Western Cape region (Figure 3.2) in South Africa (Bremer, 1972). 3.4 Medicinal uses This Cape-Dutch remedy has been traditionally used to treat various ailments and may be taken orally in the form of a brandy tincture for chest complaints, bronchial congestion and other respiratory related illnesses (Watt and Breyer-Brandwijk, 1962; Scott et al., 2004). Furthermore, O. asteriscoides has been combined with other plant species i.e. Artemisia afra and Eucalyptus globulus to treat respiratory infections (Watt and Breyer-Brandwijk, 1962). The dried plant may also be used externally to treat inflammation, cuts and swelling (van Wyk et al., 1997). The use of O. asteriscoides for intestinal disorders, stomach complaints and fever has also been reported (Watt and Breyer-Brandwijk, 1962). Figure 3.2 The geographical distribution of Osmitopsis asteriscoides in South Africa SANBI). 34 3.5 Methods 3.5.1 Chemical aspects Plant collection and distillation of essential oils: The aerial parts of the plants were collected in the growing season (February) from a natural population near Betty?s Bay, South Western Cape region, South Africa. A voucher specimen (Table 3.1) is retained in the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plant material was distilled as described in Chapter 2.5.1. Table 3.1 Plant collection data for O. asteriscoides. Gas chromatography combined with mass spectrometry (GC-MS): Oil samples were analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. The chiral separation of camphor was performed on a multidimensional gas chromatography / mass spectrometer (MD-GC/MS) system. Two Hewlett Packard GC 6890 systems with MSD and Gerstel multi column switching (MCS) system were used. The cooled injection system (CIS) was kept at 40?C for injection. Helium was used as carrier gas (1 mL/min). The precolumn comprised of an HP-Innowax fused silica capillary column (60 m x 0.25 mm i.d., with 0.25 um film thickness). The GC oven temperature was kept at 60 ?C for 10 min and programmed to 220 ?C at a rate of 4 ?C/min and kept constant at 220 ?C for 10 min, then programmed to 240 ?C at a rate of 1 ?C/min and kept constant at 240 ?C for 40 min. The FID detector temperature was at 250 ?C. The main column comprised: Lipodex E [Octakis (3-O-butyryl-2, 6-di-O-pentyl)-g-cyclodextrin] (70% in OV 1701), (25m x 0.25 mm i.d.). The temperature programme for camphor was 40 ?C for 34 min and programmed to 120 ?C at a rate of 1 ?C/min then kept constant at 120 ?C for 6 min. The MS were taken at 70 eV. Mass range was from m/z 35 to 425. Plant Voucher Material distilled (g) Essential oil yield (% w/w) O. asteriscoides ADCAV174 2520.0 0.7 35 3.5.2 Antimicrobial aspects Culture and media preparation were undertaken according to the NCCLS (2003) guidelines and additional references as described in Chapter 2.5.2. Three different antimicrobial assays (disc diffusion, minimum inhibitory concentrations and time-kill studies) were performed on the hydrodistilled oil to corroborate results. Disc diffusion assays: A broad disc diffusion screening was carried out against many of the same reference pathogens as listed in Table 2.2. Some variation in the selection of pathogens was undertaken. Two additional Gram-positive organisms were added to the study i.e. Bacillus subtilis (ATCC 15244) and Staphylococcus epidermidis (ATCC 2223). Rabe and van Staden (1998) and authors Mangena and Muyima (1999) have shown a better efficacy towards Gram- positives due to membrane structure and this warranted their inclusion in the study. Staphylococcus epidermidis was included in the study due to its pathogenesis being associated with immunocompromised patients, causing severe nosocomial infections such as endocarditis. This opportunistic pathogen resides on the skin as part of the normal biological flora, but in times of stress or reduced immunity it can cause a number of infections (Bannister et al., 2000). The disc diffusion method is described in Chapter 2.5.2. Minimum inhibitory concentrations (MIC): The microplate bioassay minimum inhibitory concentrations were determined using the micro-dilution method (Carson et al., 1995; Eloff, 1998a; NCCLS, 2003) as described in Chapter 2.5.2. Table 3.2 gives the acceptable control ranges for ciprofloxacin and amphotericin B against additional pathogens used in the MIC assays. Where pathogen strain numbers were the same as that used in Chapter 2, refer to Table 2.2. Table 3.2 Acceptable MIC ranges for ciprofloxacin antibiotic control. * Acceptable ranges as determined within the Department of Pharmacy and Pharmacology, University of Witwatersrand (controls in this study were run for all test organisms). Test organism MIC controls (?g/mL)* Escherichia coli (ATCC 11775) 0.04-0.10 Staphylococcus epidermidis (ATCC 2223) 0.30-2.50 Bacillus subtilis (ATCC 6051) 0.09-3.10 36 Minimum inhibitory assays were done in triplicate and the mean tabulated (Table 3.4). A 0.4 mg/mL p-iodonitrotetrazolium violet (Sigma-Aldrich) solution (INT) was prepared and 40 ?l transferred to all inoculated wells. The microtitre plates inoculated with bacteria were examined after six hours to determine a colour change in relation to concentration of microbial growth. The yeasts were examined after 24 hr. Time-kill methodology: Time-kill assays were employed according to methods as described in Chapter 2.5.2 investigating the pathogens S. aureus (Gram-positive), P. aeruginosa (Gram- negative) and a yeast C. albicans. Oil concentrations of 0.5, 1, 1.5 and 2% (5.2, 9.5, 13.8 and 18.1 mg/mL respectively) were selected to determine the death kinetics for O. asteriscoides. As with the M. flabellifolius study (Chapter 2), results were corroborated in a separate laboratory at the University of Iowa (Viljoen et al., 2003). Membrane integrity: Harvesting of bacterial cells, treatment and confocal scanning for the assessment of membrane integrity was undertaken according to standard methods described by Lindsay and von Holy (1999) and Peta et al. (2003). Stock cultures of S. aureus were grown in Tryptone Soya (Oxoid) broth and centrifuged (MSE) for 10 min at 5000 rpm. The supernatants were discarded and the pellets resuspended in 10 mL of a 0.9% NaCl (Labchem) solution with 50 ?l Tween 80 (Saarchem). Osmitopsis asteriscoides oil at 0.5% was added to the buffer. A control having no essential oil was run in tandem. At time intervals comparable to the time-kill assay, 1 mL aliquots were removed and added to 9 mL Neutralizing Buffer (Difco). After centrifugation for 10 min at 5000 rpm, samples were washed with 1 mL of a 0.9% NaCl solution and recentrifuged. The supernatants were discarded and the pellet resuspended in 100 ?l of a 0.9% NaCl solution. Samples (0.5 % O. asteriscoides oil and control) were prepared for examination using the LIVE / DEAD? Baclight Bacterial Viability Kit (Molecular Probes) and thereafter viewed using a confocal scanning electron microscope (Carl Zeiss). The morphology of cells exposed to the oil were compared with those untreated control cells by observing the interaction of the dyes with bacterial cells. Bacteria with intact cell membranes stained fluorescent green and bacteria having damaged or compromised membranes stained orange-red. Experimental procedures were repeated for conformity. Major compound investigation: The compounds 1,8-cineol at 98.0% purity (Lot 1054365, Sigma-Aldrich), (+)-camphor at 98.0% purity (Lot 464-48-3, Sigma-Aldrich) and (-)-camphor at 99.0% purity (Lot 464-48-2, Sigma-Aldrich) were investigated for the role they may play on the 37 anti-candidal activity of the essential oil. Initial MIC determination was undertaken as a screening process to investigate the activity of the individual compounds. Assays were done in triplicate and the mean tabulated (Table 3.5). Individual compounds were prepared at a starting concentration of 128 mg/mL and assayed with the essential oil of O. asteriscoides. A more detailed assessment of the role on the two main chemical components i.e. 1,8-cineole and camphor was undertaken by comparative time-kill studies where C. albicans, a moderately susceptible test organism, was used to demonstrate the efficacy of 1,8-cineole and camphor independently and in combination in concentrations relative to their ratio in the plant over a 24 hr period. Time-kill studies were done in duplicate and the mean plotted in Figure 3.8. 3.6 Results and discussion 3.6.1 Essential oil chemistry Twenty-five compounds were identified in the essential oil of O. asteriscoides representing 96.1% of the total composition (Table 3.3). Table 3.3 Essential oil composition of O. asteriscoides. RRI* Compound name % 1014 tricyclene tr 1032 ?-pinene 3.0 1035 ?-thujene 0.1 1076 camphene 1.8 1118 ?-pinene 0.6 1132 sabinene 1.4 1174 myrcene tr 1188 ?-terpinene tr 1195 dehydro-1,8-cineole 0.3 1203 limonene tr 1213 1,8-cineole 59.9 1255 ?-terpinene 0.3 1280 p-cymene 0.9 1290 terpinolene tr 1391 (Z)-3-hexen-1-ol tr 1450 trans-linalool oxide (furanoid) tr 1474 trans-sabinene hydrate 0.1 1482 longipinene 2.9 1493 ?-ylangene tr 1499 ?-campholenal tr 1522 chrysanthenone tr 1532 camphor, (-) 12.4 38 RRI* Compound name % 1553 linalool 0.3 1556 cis-sabinene hydrate 0.1 1571 trans-p-menth-2-en-1-ol 0.1 1586 pinocarvone 0.1 1611 terpinen-4-ol 2.3 1638 cis-p-menth-2-en-1-ol tr 1648 myrtenal tr 1651 sabinaketone tr 1664 trans-pinocarveol 0.1 1682 ?-terpineol 0.4 1706 ?-terpineol 7.8 1758 cis-piperitol tr 1798 methyl salicylate 0.1 1804 myrtenol tr 1864 p-cymen-8-ol tr 2008 caryophyllene oxide 0.4 2008 p-menta-1,8-dien-10-ol 0.1 2113 cumin alcohol tr 2265 longiverbenone (= vulgarone B) 0.2 TOTAL 96.1 *RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = trace (< 0.1%) Major compounds: The two major compounds (Figure 3.3) in the essential oil of O. asteriscoides are 1,8-cineole (59.9%) and (-)-camphor (12.4%), representing 72.3% accumulatively. O CH 3 CH3CH3 O (-)-camphor 1,8-cineole Figure 3.3 Chemical structures for major compounds identified in the essential oil of O. asteriscoides. 39 3.6.2 Antimicrobial activity Of the twelve test organisms studied in the disc diffusion assay, four (P. aeruginosa, B. subtilis, A. niger and A. alternata) showed no inhibitory effect. The highest activities were obtained for the yeasts (C. neoformans, 3 mm and C. albicans, 2 mm) and S. aureus (3 mm). For the MIC assay highest sensitivities were noted for E. coli and S. epidermidis (4 mg/mL), followed by moderate sensitivity (8 mg/mL) for the three Gram-positive organisms S. aureus, E. faecalis, B. subtilis and the yeasts C. albicans and C. neoformans. Table 3.4 lists the zones of inhibition (mm) and MIC values (mg/mL) for the test organisms studied. The MIC values recorded here do not always correlate with previously published data (Viljoen et al., 2003). Variation may be attributed to the mean been calculated with a further range of reference cultures and more repetitions as the techniques were optimised. The correlation between the two different screening methods was examined and larger zones of inhibition did not always correlate with lower MIC values. Variations occurred for P. aeruginosa where no activity was noted in the disc diffusion assay but in the MIC assay some minor activity (16 mg/mL) was noted. This variation is more clearly seen in the time-kill assay (Figure 3.5) where a steady progression of antimicrobial action is noted over 24 hr. This trend is also evident with the test organism B. subtilis where antimicrobial activity (8 mg/mL) is only noted in the MIC assay. Possible regrowth could have occurred (not visible in the MIC assay and in the disc diffusion assay) due to the presence of resistant spores. Escherichia coli (4 mg/mL, 1.5 mm) and S. epidermidis (4 mg/mL, 1 mm) indicated MIC values lower than what would be expected when observing / correlating with inhibition zones. This variation between methods was also noted by Pattnaik et al. (1996), Chalchat and Garry (1997), Janssen et al. (1987), Pauli and Kubeczka (1997) and Remmal and Tanaoui-Elaraki (1993) and can be attributed mainly to the variation of neat essential oil on disc, disc size, agar composition as well as the volatility of oil in an open air system. The incubation period for the moulds are seven days and the volatility of the essential oils over this extended incubation period may be responsible for the negative inhibition results obtained in the disc diffusion assay. 40 Table 3.4 Disc diffusion and MIC assays of O. asteriscoides essential oil. Test organism Disc diffusion (mm) *1 Disc diffusion controls*2 MIC (mg/mL) MIC controls *4 Pseudomonas aeruginosa ATCC 9027 0 2 16 2.5 x 10 -3 Escherichia coli ATCC 11775 2 10 4 0.08 x10 -4 Salmonella typhimurium ATCC 14028 1 8 32 0.8 x 10 -3 Proteus vulgaris (clinical strain) 1 12 16 1.6 x10 -4 Staphylococcus aureus (ATCC 25923) 3 14 8 0.5 x 10 -3 Staphylococcus epidermidis ATCC 2223 1 2 4 2.5 x 10 -3 Enterococcus faecalis ATCC 29212 1 8 8 2.0 x 10 -3 Bacillus subtilis ATCC 6051 0 10 8 2.5 x 10 -3 Cryptococcus neoformans ATCC 90112 3 16 8 2.5 x 10 -3 Candida albicans ATCC 10231 2 14 8 1.3 x 10 -3 Aspergillus niger (clinical strain) 0 20 ND *3 ND*3 Alternaria alternata (clinical strain) 0 4 ND *3 ND*3 *1 The diameter of the zone of inhibition is expressed in mm from the disc edge. *2 Neomycin and nystatin served as controls for bacteria and fungi respectively. *3 ND = Methods did not incorporate MIC?s for moulds. *4 Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. Results of the time-kill study are presented in Figures 3.4 to 3.6. Positive controls (ciprofloxacin and amphotericin B) presented with an immediate death response and due to presentation scale were not included in the time-kill plots. As with the M. flabellifolius study, results duplicated in an independent laboratory in a collaborative investigation with Prof. Klepser were mainly congruent. Further comparative assessment can be observed in Viljoen et al. (2003). The cidal activity was exhibited for concentrations ranging between 0.5% and 2.0% over 24 hr. For the S. aureus (ATCC 29523) time-kill plot, the essential oil generally followed a concentration- 41 dependent antibacterial activity (Figure 3.4). There is an initial decrease in viable counts for all concentrations within the first 60 min after which the 1.5% (13.8 mg/mL) and 2% (18.1 mg/mL) oil concentrations show increasing bactericidal activity with time where cidal activity is evident after 24 hr. At concentrations of 0.5% (5.2 mg/mL) and 1% (8.1 mg/mL) a bacteriostatic effect was noted after eight hours with viable counts neither increasing nor decreasing after 24 hr. These results correlate with the MIC values (8 mg/mL) obtained for S. aureus. In Figure 3.5 the death kinetics show a decrease in the number of surviving P. aeruginosa cells within 60 min, but an increase in colony forming units over time indicates that, after an initial bacteriostatic effect, regrowth of the organism occurs. The 24 hr time-kill plot (Figure 3.5) for P. aeruginosa shows a more detailed progression of bacteriostatic activity, which can obviously not be recorded by the disc diffusion method. The regrowth effect gives some rationale for the differences of results obtained for the disc diffusion assay where no activity was recorded. Some activity (8 mg/mL), congruent with the time-kill assays was noted for the MIC assay. Figure 3.6 displays the cidal effect of the essential oil on C. albicans (ATCC 10231). The candidacidal activity is rapidly exhibited and viable counts were considerably lowered within 240 minutes for all the concentrations and then increasing cidal action with increasing oil concentration. At 24 hr complete fungicidal activity was exhibited for all oil concentrations. The candidacidal activity reported here could possibly provide scientific support for the traditional use of O. asteriscoides. 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 5 10 15 20 Time (hrs) CF U/ m l Control 0.5% 1% 1.5% 2% Figure 3.4 Death kinetic studies of S. aureus with exposure to the essential oil of O. asteriscoides as seen over a 24 hour period. 42 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 5 10 15 20 Time (hrs) CF U/ m l Control 0.5% 1% 1.5% 2% Figure 3.5 Death kinetic studies of P. aeruginosa with exposure to the essential oil of O. asteriscoides as seen over a 24 hour period. 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 5 10 15 20 Time (hrs) CF U/ m l Control 0.5% 1% 1.5% 2% Figure 3.6 Death kinetic studies of C. albicans with exposure to the essential oil of O. asteriscoides as seen over a 24 hour period. Comparison of the time-kill plots for the three organisms studied showed that the killing rate was the greatest for C. albicans, then S. aureus and then with initial activity followed by subsequent regrowth for P. aeruginosa. For S. aureus, steady killing rates were obtained at 2% essential oil concentration. With P. aeruginosa, regrowth before four hours was noted for all concentrations with growth reduction for the initial 60 min. For C. albicans, a concentration dependent cidal activity was noted with greatest cidal activity at 2% after 12 hr. Membrane integrity: In conjunction with time-kill analysis, a membrane integrity study was undertaken on 0.5% O. asteriscoides essential oil sample. This method correlated very well with the time-kill studies. Figure 3.7 shows a proportion of the confocal image viewed and selected 43 for cell counting. This view of the S. aureus cells when exposed to 0.5% oil of O. asteriscoides indicate viability (green cells) seen for most cells in the control (0-20 hr) and on initial exposure to the oil. There is some cell death as this is acceptable in growing cell populations. Over time, however, the 0.5% O. asteriscoides oil sample show increasing cell death (orange-red). In comparison with the time-kill showing the kinetics for 0.5% O. asteriscoides after five hours (Figure 3.7) results were comparable. The area above the line after five hours depicts viable cells. Below the line shows non-viability, indicative of a bacteriostatic action or cell injury. This corresponds with the data generated in the confocal membrane integrity study (Table 3.5) also having injured populations with live and dead cells in the same proportion. In addition, membrane integrity studies display cell lysis which is some indication of how the essential oil is working to inactivate microbial cells. 0 min 0 min 4 hr 4 hr Control 0.5% O. asteriscoides 20 hr 20 hr Figure 3.7 The confocal scanner laser images (2 ?m) of the control (having no essential oil) and 0.5% O. asteriscoides over 24 hr. 44 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 1 2 3 4 5 Time (hrs) CF U/ m l Control O. asteriscoides Major compound investigation: Further studies on the role of the two major chemical components 1,8-cineole (60%) and (-)-camphor (12%) on the antimicrobial activity of C. albicans was demonstrated with MIC?s and death kinetic studies. Previous MIC studies on the compounds 1,8-cineole and camphor have shown conflicting results when tested against C. albicans. Magiatis et al. (2002), indicated moderate activity when tested against camphor (6.2 mg/mL) and good activity (0.3 mg/mL) when tested against 1,8-cineole. Pattnaik et al. (1997) reported candidal resistance when tested against 1,8-cineole. Setzer et al. (2004) indicated good activity for C. albicans when tested against both 1,8-cineole and camphor (0.6 mg/mL). Griffin et al. (1999), reported moderate candidal activity for 1,8-cineole with camphor being more Cell viability (%) Time Sample Live Dead Injured Control 92.6 7.4 0 0 min 0.5% O. asteriscoides 79.1 20.2 0.7 Control 97.6 2.4 0 30 min 0.5% O. asteriscoides 49.3 34.9 15.8 control 95 4.7 0.3 2 hr, 15 min 0.5% O. asteriscoides 36.4 57.3 6.3 Control 93.9 4.5 1.6 4 hr, 20 min 0.5% O. asteriscoides 20.2 20.2 59.6 Figure 3.8 Death kinetic studies of S. aureus with exposure to the essential oil (0.5%) of O. asteriscoides over five hours. Table 3.5 The confocal cell viability (%) of S. aureus cells when exposed to 0.5% O. asteriscoides over 4 hr and 20 min. 45 active. Contradictory reports may be due to tests been undertaken on different C. albicans reference strains or differentiation in compound purity. From the MIC data seen in Table 3.6, it can be noted that 1,8-cineole (16 mg/mL), (+)-camphor (32 mg/mL) and (-)-camphor (16 mg/mL) are less active having MIC values higher than that of O. asteriscoides (8 mg/mL). When the combination of (-)-camphor with 1,8-cineole was tested against C. albicans, a significantly lower MIC value (0.5 mg/mL) was obtained. This indicates that a combination of these two major compounds may be responsible for the increased activity found in O. asteriscoides. Even though some research has been carried out on essential oil combinations (Geda, 1995; Lachowicz et al., 1998), investigation of the synergistic role of constituents have been neglected. 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 0 0.2 0.4 0.6 0.8 1 Time (hrs) CF U /m l Control 1,8-Cineole (-)-Camphor (+)-Camphor 1,8-Cineole and (-)-Camphor O. asteriscoides Sample MIC (mg/mL) O. asteriscoides 8.0 1,8-cineole 16.0 (+)-camphor 32.0 (-)-camphor 16.0 1:1 (-)-camphor with 1,8-cineole 0.5 nystatin control 1.3 x 10-3 Table 3.6 The MIC of O. asteriscoides essential oil and major constituents against C. albicans. Figure 3.9 Time-kill plot for O. asteriscoides essential oil (1%), 1,8-cineole, (+)- camphor, (-)-camphor, 1,8-cineole and (-)-camphor tested on Candida albicans (ATCC 10231). 46 In order to investigate this further, a time-kill assay (Figure 3.9) was undertaken and results show that both (+)-camphor, (-)-camphor have negligible antimicrobial activity on C. albicans whereas 1,8-cineole indicated microbial reduction after 240 min. As O. asteriscoides only accumulates levorotatory camphor, it was tested in combination with 1,8-cineole and a total reduction of colony forming units was observed within 15 min. This indicates that the positive antimicrobial response was due to possible synergistic interaction between 1,8-cineole and (-)-camphor. From Figure 3.7, it is evident that some minor compounds should also be taken into consideration as 1% essential oil still had a greater killing rate (under 10 min) than 1,8-cineole and (-)-camphor in combination. Camphor is known to be a decongestant and antiseptic (Bruneton, 1995) and it is further known for its topical use as a counter-irritant in fibrositis (Reynolds, 1996). The enantiomeric composition of camphor has been discussed by Demirci et al. (2002) where reference was made to 40 different essential oil samples with (-)-camphor occurring more abundantly than (+)- camphor. Nakatsu et al. (2000), reported inactivity of 1,8-cineole against Gram-positive bacteria, however, with S. aureus, relatively moderate activity (Table 3.4) for O. asteriscoides was noted. Cineole-rich essential oils have been used historically for respiratory infections (Silvestre et al., 1999). Previous studies by Pattnaik et al. (1997), show that 1,8-cineole at 23.2% has an antimicrobial effect against 18 microbial strains, with the exception of P. aeruginosa which showed no inhibitory effect. In antimicrobial studies done on O. asteriscoides having 60% 1,8- cineole, no lasting antimicrobial effect was noted with P. aeruginosa. The antifungal activity of 1,8-cineole, specifically with C. albicans was also noted by Steinmetz et al. (1988) where a decrease of fungal activity was reported after six hours for rosemary essential oil (composition = 51% 1,8-cineole and 11% camphor). Similarly, with O. asteriscoides, there was a 90% reduction in fungal activity after six hours. Jasonia montana, another commonly used folk medicine, having an essential oil composition of 21.79% of camphor and 4.6% 1,8-cineole was reported by Hammerschmidt et al. (1993) to have antifungal activity against C. albicans and C. neoformans. This corroborates the findings summarised in Table 3.4. In a plant screening by Tzakou et al. (2001) of Salvia ringens containing concentrations of 1,8-cineole (46.4-50.7%), comparable to that obtained for Osmitopsis asteriscoides no activity was reported for Staphylococcus aureus and Staphylococcus epidermidis, but very good activity for the Gram-negatives. Results of O. asteriscoides show that the oil is active against S. aureus and S. epidermidis. It was thought to be due to the high concentration of camphor (14.9%). Also of note was the fact that camphor was only present in small quantities (1.6-1.8%) for Salvia ringens (Tzakou et al., 2001). 47 Previous phytochemical work has been undertaken on Osmitopsis (Bohlmann and Zdero, 1974; Bohlmann et al., 1985) who reported the presence of sesquiterpene lactones. These compounds are known for their wide range of biological activity (Bruneton, 1995) and the monoterpenes, contained in the essential oil, especially 1,8-cineole and (-)-camphor, provides a chemical rationale for the established traditional use of O. asteriscoides as a Cape-Dutch remedy, which is still widely used today. 3.7 General conclusions ? The major constituents of the essential oil of Osmitopsis asteriscoides are 1,8-cineole (59.9%) and (-)-camphor (12.4%), representing 72.3% accumulatively. ? The highest antimicrobial activities obtained by means of the disc diffusion assay were noted for the fungal organism Cryptococcus neoformans (3 mm) and Staphylococcus aureus (3 mm). ? The highest antimicrobial activity noted in the MIC assay was against Staphylococcus epidermidis (4 mg/mL) and Escherichia coli (4 mg/mL). ? Death kinetics for the three organisms studied demonstrated that the killing rate was greatest for Candida albicans, then Staphylococcus aureus, and lastly only initial activity for Pseudomonas aeruginosa with regrowth after two hours. ? The ethnobotanical use of Osmitopsis asteriscoides to treat dermal afflictions correlates with the highest antimicrobial activities found for the Staphylococci. ? Method corroboration was achieved between different laboratories (Universities Iowa and Witwatersrand) for the time-kill assay and between methods (MIC / time-kill and time-kill / confocal integrity studies). ? The two major essential oil constituents (1,8-cineole and (-)-camphor) act synergistically to enhance antimicrobial activity. 48 Chapter 4 The antimicrobial activity of Artemisia afra Jacq. ex Willd essential oil and the role of the major constituents, singularly and in combination. 4.1 Introduction The genus Artemisia (Asteraceae) is one of the most well-known medicinal plants and has been used globally to treat infectious diseases (Setzer et al., 2004). Its reputation in medicine dates back to the 15th century when some species e.g. A. absinthium were not only used as an antiseptic but also incorporated into alcoholic beverages and used in folklore celebrations as love charms. The word ?Artemisia? is derived from the Greek goddess Artemis. Of the 180 species in this genus, Artemisia afra Jacq. ex Willd is the only indigenous species (afra means Africa) and is one of the most widely used medicinal plants (Graven et al., 1992). Its popularity may be attributed not only to the wide use as a treatment for mainly respiratory ailments but also its vast abundance in the wild. A recent study on the use of medicinal plants in the Bredasdorp region of South Africa confirmed its popularity when a survey revealed that of the 44 participants interviewed, all reported the medicinal use of A. afra, making it one of the most popular plants used by the local people (Thring and Weitz, 2006). In spite of the international interest this genus has generated both on a commercial market and within indigenous populations, very little quantitative microbiological analysis has been undertaken. Some literature (Graven et al., 1992; Gundidza et al., 1993; Mangena and Muyima, 1999; Huffman et al., 2002; Scott et al., 2004) have established that microbial activity is evident, but no time-kill studies have been undertaken with A. afra. This report serves to validate the antimicrobial activity more quantitatively and investigate further, the role of the major compounds as possible antimicrobial constituents, thus providing a scientific basis for the traditional uses of A. afra. 49 4.2 Botanical description Artemisia is a large and widespread genus in the family of Asteraceae housing approximately 350 species (Tan et al., 1998). Most of the Artemisia species yield essential oils (Pappas and Sturtz, 2001). There is one indigenous (A. afra) and one localized (A. vulgaris) Artemisia species in Southern Africa. Artemisia afra is commonly known as African wormwood, ?umhlonyane? (Xhosa, Zulu), ?lengana? (Sotho, Tswana) and ?wildeals? (Afrikaans) (van Wyk et al., 1997). This plant is a multi-stemmed perennial shrub, growing up to two meters in height. Glaucous feather-like leaves (Figure 4.1) with a pungent aromatic smell, indicative of its rich essential oil, characterize A. afra. The oil is yellow-green. The flowers are yellow, inconspicuous and borne at the ends of branches (von Koenen, 2001). Figure 4.1 The leaves of Artemisia afra. 4.3 Distribution The plant is abundantly distributed in mountainous regions from South Western Cape, extending from the Cederberg Mountains northwards. Distribution generally proceeds along the eastern coast through to the Northern Province (van Wyk et al., 1997). Figure 4.2 illustrates the geographical distribution of A. afra in South Africa. Figure 4.2 The geographical distribution of Artemisia afra in southern Africa (SANBI). 50 4.4 Medicinal uses Members of the genus have been frequently used to treat infections caused by bacteria, fungi and viruses. Watt and Breyer-Brandwijk (1962) reports the inhalation of leaf deconcoctions for blocked nose, headache, or alternatively the tip of a fresh plant is inserted into the nose to treat the same conditions. Fresh twigs may also be inserted into a hollow tooth to relieve toothache. The plant is usually burned and used as an inhalant for ailments of the bronchial passages. This suggests the release of volatile components and hence this study is narrowly focused on the essential oils of this species. The oils of the plant impart a bitter taste and the presence of thujone makes this plant toxic if consumed in sufficient quantities (Hutchings et al., 1996). Artemisia afra has also been used for the treatment of ailments associated with the gastrointestinal tract, fever and topical inflammation (von Koenen, 2001; Scott et al., 2004). In addition, there have been reports where A. afra has been combined with Agrimonia bracteata, O. asteriscoides (Chapter 3), Lippia javanica (Chapter 5), and Tetradenia riparia to treat respiratory disorders (Hutchings et al., 1996). 4.5 Methods 4.5.1 Chemical aspects Plant collection and distillation of essential oils: The aerial parts from a natural population were collected during the growing period (February) from Klipriviersberg, southern region of Gauteng. Due to the larger volumes of essential oil required for the antimicrobial assays, a collective sample from Klipriviersberg was used to obtain the essential oil. A voucher specimen (Table 4.1) is deposited in the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plants were distilled as described in Chapter 2.5.1. Table 4.1 Plant collection data for A. afra. Plant Voucher Material distilled (g) Essential oil yield (% w/w) A. afra SVV173 623.76 0.2 51 Gas chromatography combined with mass spectrometry (GC-MS): Oil samples were analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. 4.5.2 Antimicrobial aspects Culture, media preparation and assays were undertaken according to the NCCLS (2003) guidelines and methods described by Carson et al. (1995), and Eloff (1998a), details of which can be referred to in Chapter 2.5.2. Three different antimicrobial assays (disc diffusion, minimum inhibitory concentrations and time-kill studies) were performed on the hydrodistilled oil. Four additional test micro-organisms were added to the study i.e. Candida tropicalis, Klebsiella pneumoniae, Bacillus cereus and a different strain of S. aureus. Candida spp. are prevalent not only in the urinary tract but are also a cause of lung abscesses in immunocompromised patients. Klebsiella pneumoniae has been associated with septicaemia, cystitis, urinary tract infections, wound infections and pneumonia with pyrogenic respiratory infections being one of the most prevalent causes of lung infections in hospitals (Bannister et al., 2000). These organisms are often hospital acquired complicating antibiotic treatment. The use of A. afra to treat both respiratory infections and disorders of the gut (B. cereus being one of the possible causative organisms responsible for enteric food poisoning) may be attributed to one of these organisms and thus their inclusion in the study was warranted. The acceptable MIC control ranges for these additional test pathogens as determined within the Department of Pharmacy and Pharmacology, University of Witwatersrand is given in Table 4.2. Test organism MIC controls (?g/mL) Klebsiella pneumoniae (NCTC 9633) 0.12-1.30 Staphylococcus aureus (ATCC 12600) 0.12-1.00 Bacillus cereus (ATCC 11778) 0.20-1.00 Candida tropicalis (clinical strain) 1.25-2.50 Table 4.2 Acceptable MIC ranges for controls (ciprofloxacin for bacteria and amphotericin B for the yeast. 52 Time-kill assays were employed according to methods as described in Chapter 2.5.2 investigating two pathogens (K. pneumoniae and C. neoformans). Oil concentrations of 0.063, 0.125, 0.25, 0.5 and 0.75% (0.6, 1.8, 3.0, 5.2 and 8.1 mg/mL respectively) were selected to determine death kinetics of K. pneumoniae, and oil concentrations 0.25, 0.5, 0.75, and 1% (3.0, 5.2, 8.1 and 9.5 mg/mL respectively) were selected to determine death kinetics of C. neoformans. Selection was based on preliminary assays to determine optimal death kinetic response. Results were done in duplicate and the logarithmic mean plotted in Figures 4.4-4.6. Major compound investigation: Individual compounds, obtained from Sigma-Aldrich were 1,8-cineole, at 98.0% purity (Lot 1054365), artemisia ketone at 97.0% purity (Lot 344166/114801), ?-thujone at 96.0% purity (Lot 014071/423103488) and a mixture of ? & ?- thujone composed of 70% ?-thujone and 10% ?-thujone (Lot 414439/112701). Investigation of the role of the four major compounds on the overall microbial activity was investigated by MIC determination and time-kill analysis. The MIC assay was determined for all four major compounds against ten pathogens to determine if the individual compounds contribute towards the overall antimicrobial activity of A. afra. Assays were done in triplicate and the mean documented in Table 4.5. Individual compounds (Sigma-Aldrich) were prepared at starting concentrations 128 mg/mL and assayed with A. afra. After the addition of INT, results were observed after six hours (for bacteria) and 24 hr (for yeasts), as per standard MIC methodology described in Chapter 2.5.2. The results were left for a further 24 hr and re-examined to observe if any change to the original results occurred i.e. any loss of bacteriostatic action or regrowth. To examine whether any of these compounds independently or in combination (1,8-cineole and artemisia ketone; ? & ?-thujone; ?-thujone and 1,8-cineole; ? & ?-thujone and 1,8-cineole; artemisia ketone and ?-thujone; artemisia ketone and ? & ?-thujone; artemisia ketone, ?-thujone and 1,8-cineole as well as artemisia ketone, ? & ?-thujone and 1,8-cineole) were responsible for bactericidal efficacy in A. afra, each compound and various combinations thereof were studied by death kinetic assay at the original concentration to that found in the plant (1,8-cineole at 17.8%, artemisia ketone at 10.1%, ?-thujone at 18.8% and ?-thujone at 12.5%). The number of colony forming units were determined at 0 min, 15 min, 240 min, 8 hr and 24 hr when subjected to K. pneumoniae. Time-kill assays were done in duplicate and the mean CFU counts documented in Table 4.6. 53 4.6 Results and discussion 4.6.1 Essential oil chemistry Forty nine compounds were identified in the essential oil of A. afra representing 99.7% of the total composition (Table 4.3). Previous reports (Graven et al., 1992; van Wyk et al., 1997; Chagonda et al., 1999) have shown that the major components of A. afra essential oils from different localities exhibited chemical variation. For instance, the major component reported from Ethiopian oils was artemisyl acetate (24.4 -32.1%), 1,8-cineole (63.4%) in the Kenyan oil, ? and ?-thujone (52.0%) in the Zimbabwean oil and ?-thujone (52.5-54.2%) in the South African oil. In a more recent study (Viljoen et al., 2006) the aerial parts of sixteen plants from four natural populations within South Africa were hydrodistilled and the essential oil analyzed by GC-MS. Artemisia afra displayed chemical variation in both the major and minor essential oil compounds within and between natural populations. Even though the Klipriviersberg sample was selected for this antimicrobial study one must be cognizant of the chemical variation that exists and take into account the possible variability that might arise should other populations or other individual plants be comparatively investigated. Table 4.3 Essential oil composition of A. afra. RRI* Compound name Area percentage 1032 ?-pinene 0.2 1043 santolinatriene 0.1 1076 camphene 1.0 1118 ?-pinene 0.1 1132 sabinene 1.0 1174 myrcene 0.1 1195 dehydro-1,8-cineole tr 1213 1,8-cineole 17.8 1280 p-cymene 2.2 1358 artemisia ketone 10.1 1403 yomogi alcohol 4.9 1405 santolina alcohol 2.0 1409 santolinyl acetate 0.2 1437 ?-thujone 18.8 1448 artemisyl acetate 1.8 1451 ?-thujone 12.5 1474 trans-sabinene hydrate 0.3 1497 ?-copaene 0.1 1510 artemisia alcohol 5.5 54 RRI* Compound name Area percentage 1532 camphor 8.2 1540 chrysanthenone 0.4 1556 cis-sabinene hydrate 0.3 1571 trans-p-menth-2-en-1-ol 1.0 1586 pinocarvone 0.2 1597 bornyl acetate 0.2 1611 terpinen-4-ol 2.1 1638 cis-p-menth-2-en-1-ol 0.7 1643 dehydrosabina ketone 0.1 1651 sabinaketone tr 1658 sabinyl acetate 0.1 1664 trans-pinocarveol 0.1 1682 ?-terpineol 0.1 1689 trans-piperitol 0.5 1706 ?-terpineol 0.2 1719 borneol 2.7 1742 ?-selinene 0.2 1748 piperitone 0.7 1758 cis-piperitol 0.8 1786 ar-curcumene 0.1 1802 cumin aldehyde 0.1 1864 p-cymen-8-ol 0.1 1889 ascaridol 0.3 1957 epi-cubebol 0.1 2008 caryophyllene oxide 0.3 2113 cumin alcohol 0.1 2126 4-hydroxy-4-methylcyclohex-2-enone 0.1 2144 spathulenol 0.5 2209 T-muurolol 0.5 2264 intermedeol 0.2 TOTAL 99.7 * RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = trace (< 0.1%) Major compounds: The four major compounds (Figure 4.3) in the essential oil of A. afra are 1,8-cineole (17.8%), artemisia ketone (10.1%), ?-thujone at 18.8% and ?-thujone at 12.5% representing 59.2 % accumulatively. 55 CH CH3 CH3 2H3C H3C O O O O artemisia ketone ?-thujone ?-thujone 1,8-cineole Figure 4.3 Chemical structures for major compounds identified in the essential oil of A. afra. 4.6.2 Antimicrobial activity Qualitative disc diffusion and quantitative minimum inhibition concentration (MIC) methods were used to evaluate antimicrobial activity. From the disc diffusion and MIC results presented in Table 4.4, it is clear that some antimicrobial activity is evident for A. afra. This confirms findings by Mangena and Muyima (1999) who reported the activity of African wormwood oil towards B. subtilis, E. coli, K. pneumoniae and S. aureus. According to Graven et al. (1992), the oil has value as a biological agent with greater antimycotic than antibacterial activity. Both author?s results and that reported here, indicate K. pneumoniae as the most sensitive bacterial test organism in the disc diffusion assay. Further correlations were difficult as collection locality, methodology, reference culture strains and interpretation of results differed. While Graven et al. (1992) reported antimicrobial activity against three mould species, in this study no fungal activity was found when assayed by disc diffusion against A. alternata and A. niger. As the assay used to determine antifungal activity differed it may account for the variation. Methodology variation is discussed in greater detail in Chapter 11. Graven et al. (1992) also made no microbial evaluation on the yeast species. In this study, the yeasts displayed some activity with C. tropicalis indicating the highest sensitivity (5 mm) for the disc diffusion assay. Gundidza et al. (1993), using the same methodology as Graven et al. (1992), has shown significant antifungal activity against C. albicans, A. alternata and A. niger. The different localities from which the species were collected can differ in the essential oil composition and this may impact on the microbiological activities being reported. On examination of the literature, it can be noted that A. afra collected from Fort Hare (in the Eastern 56 Cape) had higher ?-thujone (78.7%), ?-thujone (13.1%) and lower 1,8-cineole (8.2%) content than that reported by Graven et al. (1992) and the oil did not contain any camphor (Mangena and Muyima, 1999). An antimicrobial and geographical variation study indicated that the antimicrobial activity within plants from a single population and inter-population variation was evident (Viljoen et al., 2006). Such variation could account for discrepancies observed when comparatively assessing results from various researchers. The MIC values for the essential oil from A. afra ranged between 8-16 mg/mL for most pathogens. The organism S. odorifera showed slightly higher sensitivity (4 mg/mL) and S. typhimurium showed lower sensitivity (32 mg/mL). The yeasts C. neoformans, C. tropicalis and C. albicans showed consistent (8 mg/mL) antifungal activity. Table 4.4 Disc diffusion (mm from edge of disc) and MIC (mg/mL) of A. afra essential oil. Test organism Disc diffusion (mm) *1 Disc diffusion controls*2 MIC (mg/mL) MIC controls *4 Pseudomonas aeruginosa ATCC 9027 0 6 16 0.8 x 10 -3 Klebsiella pneumoniae NCTC 9633 3 6 16 0.8 x 10 -3 Escherichia coli ATCC 11775 2 7 16 0.04 x 10 -3 Salmonella typhimurium ATCC 14028 2 5 32 0.8 x 10 -3 Serratia odorifera ATCC 33132 <1 4 4 0.4 x 10 -3 Staphylococcus aureus ATCC 12600 2 7 8 0.5 x 10 -3 Staphylococcus epidermidis ATCC 2223 1 12 16 0.4 x 10 -3 Enterococcus faecalis ATCC 29212 1 5 8 2.0 x 10 -3 Bacillus cereus ATCC 11778 1 10 16 0.8 x 10 -3 Bacillus subtilis ATCC 6051 3 8 16 0.1 x 10 -3 Cryptococcus neoformans ATCC 90112 3 8 8 2.5 x 10 -3 Candida tropicalis (clinical strain) 5 8 8 2.5 x 10 -3 57 Test organism Disc diffusion (mm) *1 Disc diffusion controls*2 MIC (mg/mL) MIC controls *4 Candida albicans ATCC 10231 1 8 8 1.3 x 10 -3 Aspergillus niger (clinical strain) 0 5 ND *3 ND*3 Alternaria alternata (clinical strain) 0 4 ND *3 ND*3 *1 The zone of inhibition is expressed in mm from the disc edge. *2 Neomycin and nystatin served as controls for bacteria and fungi respectively. *3 ND = Methods did not incorporate MIC?s for moulds. *4 Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. The pathogens C. neoformans and K. pneumoniae indicating moderate antimicrobial sensitivities in the disc diffusion assay were used to demonstrate the killing rate over time in the death kinetic studies. The time-kill plots for K. pneumoniae (Figure 4.4 and 4.5) demonstrate the concentration dependent antimicrobial activity where the bactericidal efficacy is greatest at 0.75% (8.1 mg/mL) within 10 min, followed by 0.5% (5.2 mg/mL) within 30 min, 0.25% (3.0 mg/mL) within 60 min and 0.125% (1.8 mg/mL) within 24 hr. Very little antimicrobial activity was noted at concentration 0.063% (0.6 mg/mL) and the growth curve was very similar to the control having no essential oil. As the time-kill response for K. pneumoniae is mostly within the first two hours, an enlarged section of Figure 4.4 is included to depict the detailed cidal activity occurring within the first hour (Figure 4.5). The positive control ciprofloxacin resulted in an immediate death response and due to presentation scale was not included in the time-kill plot. Time-kill plots for C. neoformans (Figure 4.6) do not follow a concentration dependent cidal pattern, however, all concentrations (0.25% to 1%) have a cidal effect within four hours. Concentrations 0.75% (8.1 mg/mL) and 1% (9.5 mg/mL) show similar death kinetic patterns, having bactericidal activity at 60 min. At a concentration of 0.25% (3.0 mg/mL), after an initial drop in CFU?s over time, there was a levelling off of viability before bactericidal activity was reached at two hours. At a concentration of 0.5% (5.2 mg/mL) the A. afra essential oil followed a steady concentration dependent bactericidal tendency but after one hour the reduction of CFU?s over time decreased at a much slower rate than that of the other higher concentrations. The use of the essential oil of A. afra to treat respiratory ailments is validated by showing this rapid cidal rate over time for K. pneumoniae and C. neoformans, both respiratory pathogens associated with lung infections. 58 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 5 10 15 20 Time (hrs) CF U/ m l Control 0.063% 0.125% 0.25% 0.5% 0.75% Figure 4.4 The death kinetics of K. pneumoniae on exposure to A. afra essential oil at concentrations 0.063%-0.75% over a 24 hr period. 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 0.2 0.4 0.6 0.8 1 Time (hrs) CF U/ m l Control 0.063% 0.125% 0.25% 0.5% 0.75% Figure 4.5 The death kinetic study of K. pneumoniae on exposure to A. afra essential oil showing bactericidal activity within one hour. 1.E-01 1.E+02 1.E+05 1.E+08 1.E+11 1.E+14 1.E+17 0 1 2 3 4 5 6 Time (hrs) CF U/ m l Control 0.25% 0.5% 0.75% 1% Figure 4.6 The death kinetics of C. neoformans on exposure to A. afra essential oil at concentrations 0.063%-0.75% over a six hour period. 59 The GC-MS results (Table 4.3) showed that the main chemical constituents in the essential oil of A. afra are ?-thujone, ?-thujone, artemisia ketone and 1,8-cineole. Previous antimicrobial studies on the constituent thujone have reported moderate to high activities (Dorman and Deans, 2000) and no activity (Hinou et al., 1989) by means of disc diffusion assay. A more quantitative evaluation indicated poor activity by means of MIC determination (van Zyl et al., 2006). Thujone belongs to the aliphatic ketone group of constituents which is known for microbial inhibition rather than cidal efficacy (Pauli, 2001). Thujone is known for its toxicity (Kr?ner et al., 2005; (http://www.sahealthinfo.org/tradionalmeds/monographs/artemisia.htm) and one would expect microbial cell destruction in response to the toxicity. However, thujone has very poor solubility (http://www.erowid.org/chemicals/absinthe_info3.shtml) and possibly this may account for the poor antimicrobial activities noted in this and previous studies. For 10 of the 15 organisms previously studied (Table 4.4), the MIC of ?-thujone, ?-thujone, camphor, 1,8-cineole together with the essential oil from A. afra were recorded after six hours and 24 hr. The antimicrobial activity remained constant after the full 24 hr period for the essential oil of A. afra. The MIC values for almost all of the compounds against most of the pathogens studied increased indicating static rather than cidal activity for the independent compounds (Table 4.5). The regrowth after 24 hr suggests that other compounds are responsible for the cidal action noted for A. afra. A study undertaken by Setzer et al. (2004) where MIC?s were determined for the major compounds (amongst others) in A. douglasiana also indicated that ?-thujone, ?-thujone, artemisia ketone and 1,8-cineole had no significant in vitro activity and it was thus suggested that further investigative procedures be carried out to determine which components are responsible for the activity found in A. douglasiana. To further corroborate disc diffusion and MIC findings (Table 4.4) and take into account the possible synergistic role of the major compounds, time-kill studies were undertaken for various combinations of the major compounds: 1,8-cineole and artemisia ketone; ? & ?-thujone; ?- thujone and 1,8-cineole; ? & ?-thujone and 1,8-cineole; artemisia ketone and ?-thujone; artemisia ketone and ? & ?-thujone; artemisia ketone, ?-thujone and 1,8-cineole as well as artemisia ketone, ? & ?-thujone and 1,8-cineole. Table 4.6, Figure 4.7 and Figure 4.8 illustrate the results (CFU/mL) obtained for the chemical constituents independently and in combination over a 24 hr time period. None of the compounds showed any bactericidal effect when investigated singularly over a 24 hour period with K. pneumoniae, thus corroborating findings 60 from the MIC studies. Of the eight combination of compounds, six (? & ?-thujone; ?-thujone and 1,8-cineole; artemisia ketone, ?-thujone and 1,8-cineole; ? & ?-thujone and 1,8-cineole; artemisia ketone and ? & ?-thujone; artemisia ketone ?- thujone and 1,8-cineole) showed a reduction in colony forming units after 15-240 min but regrowth after 24 hr. Compound combinations of artemisia ketone and 1,8-cineole as well artemisia ketone and ?-thujone show no antimicrobial death kill over the 24 hr test period. Table 4.5 MIC (mg/mL) determination for the major compounds and essential oil of A. afra. * Ciprofloxacin and nystatin served as controls for bacteria and yeasts respectively. Table 4.6 The CFU/agar plate obtained for the various major compound combinations together with A. afra essential oil tested against K. pneumoniae. Chemical standard Time Scale*1 0 min 15 min 240 min 8 hr 24 hr Klebsiella pneumoniae*2 >1000 >1000 >1000 >1000 >1000 0.5% Artemisia afra >1000 208 0 0 0 1,8-cineole >1000 >1000 >1000 >1000 >1000 artemisia ketone >1000 >1000 >1000 >1000 >1000 ?-thujone >1000 >1000 >1000 >1000 >1000 artemisia ketone ?-thujone ? and ? thujone 1,8-cineole A. afra Test organism 6 hr 24 hr 6 hr 24 hr 6 hr 24 hr 6 hr 24 hr 6 hr 24 hr Control* S. aureus ATCC 12600 16 >32 8 >32 16 >32 16 >32 8 8 1.0 x 10 -3 S. epidermidis ATCC 2223 16 16 16 >32 16 >32 16 >32 8 8 0.4 x 10 -3 B cereus ATCC 11778 8 32 8 >32 8 >32 8 >32 16 16 0.8 x 10 -3 B. subtilis ATCC 6051 16 32 16 32 8 32 16 >32 16 16 0.1 x 10 -3 E. coli ATCC 11775 16 >32 16 >32 16 >32 32 >32 16 16 0.04 x 10 -3 K. pneumoniae NCTC 9633 16 >32 8 >32 8 >32 16 >32 16 16 0.8 x 10 -3 S. typhimurium ATCC 14028 8 16 8 16 12 >32 32 >32 32 32 0.8 x 10 -3 S. odorifera ATCC 33132 16 16 16 >32 16 >32 16 >32 4 4 0.4 x 10 -3 C. albicans ATCC 10231 16 16 12 16 15 >32 16 >32 8 8 2.5 x 10 -3 C. neoformans ATCC 90112 32 32 >32 >32 6 >32 >32 >32 8 8 1.25 x 10 -3 61 Chemical standard Time Scale*1 0 min 15 min 240 min 8 hr 24 hr ? and ?-thujone >1000 6 0 697 >1000 artemisia ketone + 1,8-cineole >1000 >1000 >1000 >1000 >1000 ?-thujone + 1,8-cineole >1000 92 23 1003 >1000 ? and ?-thujone + 1,8-cineole >1000 7 0 0 1023 artemisia ketone + ?-thujone >1000 >1000 >1000 >1000 >1000 artemisia ketone + ? and ?-thujone >1000 90 1 19 >1000 artemisia ketone + ?-thujone + 1,8-cineole >1000 120 1 12 >1000 artemisia ketone + ? and ?-thujone + 1,8-cineole >1000 15 0 0 >1000 * 1 Shaded areas indicate CFU >1000. *2 Control test organism without any compound or antimicrobial indicating acceptable growth kinetics. A. afra alpha & beta thujone 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (h) CF U /m L Control A. afra 1,8-cineole artemisia ketone alpha-thujone alpha & beta thujone artemisia ketone + 1,8-cineole alpha thujone + 1,8-cineole alpha & beta thujone + 1,8-cineole artemisia ketone + alpha thujone artemisia ketone + alpha & beta thujone + 1,8-cineole artemisia ketone + alpha thujone + 1,8-cineole artemisia ketone + alpha & beta thujone Figure 4.7 The death kinetic representation for the various major compound combinations together with A. afra (0.5%) essential oil tested against K. pneumoniae. 62 Chemical standard* CFU at time intervals 0 min 15 min 240 min 8 hr 24 hr Klebsiella pneumoniae control 0.5% Artemisia afra 1,8-cineole ? and ?-thujone ?-thujone + 1,8-cineole ? and ?-thujone + 1,8-cineole artemisia ketone + ?-thujone + 1,8-cineole artemisia ketone + ? & ?-thujone + 1,8-cineole Both MIC and time-kill studies on the major compounds ascertain that the antimicrobial activities cannot be attributed to the major compounds or combination thereof. As suggested by Chalchat and Garry (1997), when investigating the correlation between chemical composition Figure 4.8 The CFU?s observed when performing the compound combination time- kill assay tested against K. pneumoniae. * Figure 4.8 represents only a selected portion of the study for illustrative purposes. Note that the study was done in duplicate, with a 5 times dilution made for all combinations and final mean CFU counts represented in Table 4.6 and Figure 4.7. 63 and antimicrobial activity one needs to take into account less abundant components. The essential oil of A. afra has 49 compounds as determined by the GC-MS data. To determine the antimicrobial activity on each individual compound, as well as taking into consideration the possibility of synergy between compounds is not feasible as a certain number of trace compounds would still be unaccounted for. As seen in this study with A. afra essential oils, one or more of the 45 minor compounds are probably associated with the antimicrobial activity seen here. The use of the whole A. afra plant to treat microbial infections is conclusively validated in this study. As noted, investigating single compounds to determine bioactivity does not always yield the complete pharmacological action and in searching for single bioactive compounds scientists may be overlooking other important synergistic interactions between constituents. Many phytomedicines on the market today, such as Ginkgo biloba and Kava kava, are sold as whole extracts and it is believed that the synergistic interactions between the constituents are responsible for their therapeutic efficacy (Williamson, 2001). It is also known that many traditional healers rely not only on a single plant constituent for therapeutic regimens but often combine various plant parts and even different species in the belief that efficacy may be enhanced. In this way the efficacy of A. afra may be biologically more active when all constituents within the plant are accounted for. 4.7 General conclusions ? The major constituents of the essential oil of Artemisia afra analysed in this study are 1,8- cineole at 17.8%, artemisia ketone at 10.1%, ?-thujone at 18.8% and ?-thujone at 12.5% representing 59.2 % accumulatively. ? The highest antimicrobial activities obtained by means of the disc diffusion assay were noted for the fungal organism Candida tropicalis (5 mm). ? The highest antimicrobial activity noted in the MIC assay was against Serratia odorifera (4 mg/mL). 64 ? Rapid death kinetics for the respiratory pathogens Klebsiella pneumoniae and Cryptococcus neoformans verify the ethnobotanical use of Artemisia afra to treat respiratory infections. ? The major compounds of Artemisia afra when accessed independently and in combination had negligible antimicrobial activity, thus suggesting that the major chemical compounds may be acting in a synergistic manner with the minor compounds, or that the minor compounds, independently or in combination, may be responsible for the antimicrobial efficacy noted for Artemisia afra. 65 Chapter 5 Lippia javanica (Burm.f.) Spreng, the antimicrobial activity and an in vitro synergy study when combined with Artemisia afra to treat K. pneumoniae infections. The genus Lippia (Burm.f.) Spreng, a member of the Verbenaceae family is represented by approximately 200 herbs, shrubs and small trees which are often of an aromatic nature (Terblanch? and Kornelius, 1996). Lippia javanica is commonly referred to as ?fever tea?. This name is probably derived from its use as a weak infusion for fever symptoms. In Afrikaans it is referred to as ?koorsbossie?, which literally translated means ?fever bush?. The Zulu name ?Umsuzwane?, refers to the strong odour of the plant (Ngwenya et al., 2003). The aromatic properties of the plant have resulted in its use both medicinally and for culinary purposes as a flavouring and a sweetener (Pascual et al., 2001). A number of species have been used in African tribal customs as a treatment to ward off demons and even in contradictory conditions such as a stimulant and sedative (Pascual et al., 2001). The plant is used traditionally as a hand and body wash after burial procedures (Ngwenya et al., 2003). While a number of authors have reported on the chemistry of the genus (Velasco-Negueruela et al., 1993; Terblanch? and Kornelius, 1996; Chagonda and Makanda, 2000; Braga et al., 2005), antimalarial properties (Valentin et al., 1995) and the pharmacology (Pascual et al., 2001), the species L. javanica has been neglected with regard to antimicrobial studies. Manenzhe et al., (2004), has reported briefly on the antimicrobial activity. Viljoen et al. (2005), highlighted the chemical variation in natural populations in South Africa as well as the antimicrobial efficacy for L. javanica against three pathogens. The linalool chemotype has been investigated in this Chapter to further elaborate on the antimicrobial activities. The combination of L. javanica with A. afra as an antimicrobial is also studied. While literature sites the traditional use of plant combinations (Hutchings et al., 1996; van Wyk and Gericke, 2000) there has been very little scientific evidence to support the traditional use of L. javanica in combination with A. afra. 5.1 Introduction 66 5.2 Botanical description Lippia javanica is an erect woody shrub approximately two metres in height. Its hairy leaves have compound veins and are highly aromatic with a strong lemon smell. Dense rounded heads of small yellowish-white flowers (Figure 5.1) appear during the flowering season (van Wyk et al., 1997). Figure 5.2 The geographical distribution of L. javanica in South Africa (SANBI). 5.3 Distribution The plant is distributed in southern Africa and northwards, mainly in the eastern regions. Plant populations extend from the Eastern Cape through to KwaZulu-Natal, Mpumalanga, Gauteng, North-West Province and Northern Province with a few L. javanica populations present in southern Botswana (Figure 5.2). 5.4 Medicinal uses Both the layman and traditional healers use the plant extensively in traditional medicine to treat minor ailments. The leaves and stems are often used and in some cases the roots as well (van Wyk et al., 1997). Strong leaf infusions are made, which are commonly used as inhalants and Figure 5.1 The aerial leaves and white flower heads of Lippia javanica. 67 also topically for scabies and lice (van Wyk et al., 1997; Gelfand et al., 1985). In Tanzania, the species is used to treat skin infections where fresh bruised leaves are rubbed over the infected skin area (Ngassapa et al., 2003). More commonly, leaf and stem infusions are used as a tea to treat coughs, colds, fever and bronchitis (Watt and Breyer-Brandwijk, 1962; Hutchings et al., 1996). It has also been used for bronchial ailments and influenza (Hutchings et al., 1996). Roots are used as antidotes for suspected food poisoning and bronchitis. The Vhavenda people use leaf infusions as antihelmintics, respiratory and febrile ailments and as a prophylactic against dysentery, diarrhoea and malaria (Mabogo, 1990). Roots are used as antidotes for suspected food poisoning, for bronchitis and sore eyes (Hutchings et al., 1996). Ethnobotanical literature documents its uses for fever and influenza in combination with leaves of A. afra (Hutchings et al., 1996). 5.5 Methods 5.5.1 Chemical aspects Plant collection and distillation of essential oils: The aerial parts of the plants were collected in the growing season (November) from a natural population in Fairland, in the northern regions of Gauteng. A voucher specimen (Table 5.1) is retained in the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plant material was distilled, as described in Chapter 2.5.1. Table 5.1 Plant collection data for L. javanica. Gas chromatography combined with mass spectrometry (GC-MS): Oil samples were quantitatively analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. Plant Voucher Material distilled (g) Essential oil yield (% w/w) L. javanica AMV355 6517.9 0.2 68 5.5.2 Antimicrobial aspects Culture, media preparation and assays were undertaken according to the NCCLS (2003) guidelines and methods described by Carson et al. (1995) and Eloff (1998a) in Chapter 2.5.2. Three different antimicrobial assays (disc diffusion, MIC and time-kill studies) were performed on the hydrodistilled oil. For the time-kill assay oil concentrations of 0.25, 0.5, 0.75 and 1% (3.0, 5.2, 8.1 and 9.5 mg/mL respectively) were selected to determine the death kinetics of K. pneumoniae, C. neoformans and B. cereus. The MIC was determined on a racemic mixture of linalool at 95% purity (Lot 409782, Sigma-Aldrich) at a starting concentration of 128 mg/mL, against all bacterial and yeast isolates from this study. Time-kill assays were employed according to methods as described in Chapter 2.5.2 investigating three pathogens (K. pneumoniae, B. cereus and C. neoformans). Results were recorded in duplicate and the mean plotted in Figures 5.4-5.6. A time-kill assay was used to assess the ethnobotanical use of L. javanica with A. afra in combination therapy against the respiratory pathogen K. pneumoniae. Essential oil at 0.25% concentration obtained from L. javanica and A. afra (Chapter 4.5.1) were run independently with a combination of L. javanica and A. afra (together totalling 0.25%). A control (K. pneumoniae having no essential oil) was included in the study. Aliquots were sampled (as described in Chapter 2.5.2.) at pre-determined time intervals ranging from 0 min - 48 hr, (5 min, 15 min, 240 min, 8 hr, 24 hr and an extended 48 hr). Colony-forming units were counted and death kinetics expressed in Table 5.4 and Figure 5.7. The study was done in duplicate. 5.6 Results and discussion 5.6.1 Essential oil chemistry Thirty-four compounds were identified in the essential oil of L. javanica representing 99.40 % of the total composition (Table 5.2). Table 5.2 Essential oil composition of L. javanica. RRI* Compound name % 1032 ?-pinene 0.2 1076 camphene 0.3 1174 myrcene and ?-phellandrene 2.6 69 RRI* Compound name % 1203 limonene 0.5 1218 ?-phellandrene 0.5 1246 (Z)-?-ocimene 13.0 1266 (E)-?-ocimene 6.2 1280 p-cymene 0.9 1319 dihydrotagetone 0.2 1382 cis-alloocimene 0.2 1444 ipsenone (tentative, Wiley) 0.8 1450 trans-linalool oxide (furanoid) 0.2 1452 1-octen-3-ol 0.1 1478 cis-linalool oxide (furanoid) 0.2 1497 ?-copaene 0.4 1500 cis-tagetone 0.2 1522 trans-tagetone 0.3 1532 camphor 0.2 1553 linalool 65.2 1612 ?-caryophylline 3.6 1661 alloaromadendrene 0.1 1687 ?-humulene 0.2 1704 ?-muurolene 0.1 1719 borneol 0.4 1726 germacrene-d 1.5 1740 ?-muurolene 0.1 1755 bicyclogermacrene 0.1 1758 (E,E)-?-farnesene 0.2 1773 ?-cadinene 0.2 1830 2,6-dimethyl-3(E),5(E),7-octatriene-2-ol 0.1 2001 isocaryophylline oxide 0.1 2008 caryophylline oxide 0.4 2050 (E)-nerolidol 0.1 2071 humulene epoxide II tr TOTAL 99.40 *RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = trace <0.1%. Major compounds: The three major compounds (Figure 5.3) in the essential oil of L. javanica are linalool (65.2%) and (Z)-?-ocimene (13.0%), representing 78.2% accumulatively. 70 OH linalool (Z)-?-ocimene Figure 5.3 Chemical structures for the major compounds identified in the essential oil of L. javanica. Mwangi et al. (1992) and Terblanch? and Kornelius (1996) have cited myrcene, myrcenone, ocimene, (E)-tagetenone, (Z)-tagetenone and cis-tagetone as major compounds in L. javanica. Of these, only (Z)-?-ocimene is present as a major constituent in this sample analysed from the Fairland population. Variation in chemical composition was also noted by Ngassapa et al. (2003), when evaluating two different L. javanica samples from the same locality in Tanzania. Further analysis of the composition variation was undertaken (Viljoen et al., 2005) where sixteen samples (representing five natural populations) were investigated by cluster analysis and five chemotypes were identified: a myrcenone rich type (36?62%); a carvone rich type (61?73%); a piperitenone rich type (32?48%); an ipsenone rich type (42?61%) and a linalool rich type (>65%). The myrcenone and linalool chemotypes have been mentioned in literature (Chagonda and Makanda, 2000; Ngassapa et al., 2003; Terblanch? and Kornelius, 1996) but the carvone, ipsenone and piperitenone chemotypes have not previously been reported for L. javanica. Of the five localities studied, the Fairland population was selected for further antimicrobial investigation. Selection was based on its chemotype (rich in linalool), not identified in any of the other samples, preliminary antimicrobial analysis (Subramoney, 2003) and the added advantage that the vast quantity of plant material required for time-kill analysis was easily accessible. As linalool represents the major component of the essential oil, it warranted further assessment to determine if the antimicrobial efficacy may be attributed to this major constituent. The MIC (Table 5.3) was determined for a racemic mixture of linalool against nine reference bacterial strains and three yeast strains. Pattnaik et al. (1997) reported moderate antibacterial sensitivity for linalool against S. aureus, E. coli and K. pneumoniae. For C. albicans and C. neoformans high sensitivities were found. Resistance was reported for P. aeruginosa. In this study only the bacterial results show partial congruency. Pattnaik et al. (1997) noted that MIC?s from essential oils were in many cases lower than the major constituents independently suggesting that synergy 71 between constituents may be contributing to the enhanced activity. Minimum inhibitory concentration values for L. javanica essential oil indicated higher sensitivities than the linalool constituent for pathogens P. aeruginosa, C. albicans and C. neoformans. In another study (Inouye et al., 2001a) where the vapour was analysed for antimicrobial activity, linalool was assessed with 14 other essential oils. Results for an oil rich in linalool (Coriander, 73.2% linalool) indicted the same MIC value for linalool and Coriander against S. aureus and E. coli. These organisms together with C. tropicalis also gave linalool MIC results equivalent to that found in the L. javanica essential oil signifying that linalool may be responsible for the antimicrobial activity noted against these pathogens. In a more recent study (Alviano et al., 2005), indicated by means of a bio-autographic assay, that the linalool rich essential oil of Croton cajucara, while indicating antimicrobial activity, showed no antibacterial effect against the separated linalool constituent. While the study by Alviano et al. (2005) seems to contrast with previous results obtained by Inouye et al. (2001a), one must take into account that different pathogens were used to assess antimicrobial activity. The study by Alviano et al. (2005) was restricted to oral bacterial biofilms only. As noted earlier, the MIC?s obtained for linalool in this study with L. javanica also indicated no activity with selected pathogens. Thus, the antimicrobial activity may or may not be attributed to the major constituent linalool (65.2%), depending on the pathogen studied. 5.6.2 Antimicrobial activity Antimicrobial disc diffusion and MIC assays (Table 5.3) were performed on the essential oil samples obtained from L. javanica and tested against five Gram-positive bacteria, four Gram- negative bacteria and five fungi. Highest disc diffusion sensitivities were noted for K. pneumoniae (5 mm), E. faecalis (4 mm) and C. neoformans (4 mm). Interestingly, these pathogens are associated with respiratory infections and hence the correlation to the traditional use of L. javanica. None of the oil samples showed any apparent activity in the disc diffusion assay against P. aeruginosa, S. epidermidis and the mould A. niger. The MIC data has indicated moderate activity (8 mg/mL) for most of the bacterial isolates. No significant antimicrobial activity was found in the MIC assay for L. javanica. Ngassapa et al. (2003) indicated similar MIC results on one of the L. javanica species (sample B) collected from Tanzania, even though the chemical composition varied considerably to the chemical composition of the L. javanica sample collected from Fairlands. The authors documented 72 limonene (11.3%), neral (13.7%) and geraniol (21.5%) as major components whereas linalool (1.8%) and ?-ocimene (trans 0.3% and cis 0.3%) were only present as minor components. Previous reports (Knobloch et al., 1989) on ocimene have shown poor to moderate antimicrobial activity and no conclusive correlation could be made with the major compound ocimene and the antimicrobial activity noted in L. javanica. Table 5.3 Disc diffusion (mm from edge of disc) and MIC (mg/mL) of L. javanica essential oil. *1 The diameter of the zone of inhibition is expressed in mm from the disc edge. *2 Neomycin and nystatin served as controls for bacteria and fungi respectively. *3 ND = Methods did not incorporate MIC?s for moulds. *4 Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. Test organism Diameter (mm) *1 DD controls*2 MIC (mg/mL) Linalool MIC controls*4 P. aeruginosa ATCC 9027 0 6 8 16 1.00 x 10 -3 K. pneumoniae NCTC 9633 5 5 16 4 0.13 x 10 -3 E. coli ATCC 11775 1 5 4 4 0.08 x 10 -3 S. odorifera ATCC 33132 1 6 8 4 1.25 x 10 -3 S. aureus ATCC 25923 2 10 16 16 0.50 x 10 -3 S. epidermidis ATCC 2223 0 12 8 4 0.63 x 10 -3 E. faecalis ATCC 29212 4 3 16 2 2.00 x 10 -3 B. cereus ATCC 11778 1 10 8 8 0.63 x 10 -3 B. subtilis ATCC 6051 1 10 16 8 0.09 x 10 -3 C. neoformans ATCC 90112 4 5 4 8 2.50 x 10 -3 C. tropicalis (clinical strain) 3 5 32 32 2.50 x 10 -3 C. albicans ATCC 10231 1.5 7 16 32 1.30 x 10 -3 A. niger (clinical strain) 0 7 ND *3 ND*3 ND*3 A. alternata (clinical strain) 0.5 3 ND *3 ND*3 ND*3 73 There was very little correlation between disc diffusion and MIC data. For instance with P. aeruginosa and S. epidermidis, no antimicrobial activity was found in the disc diffusion assay, yet moderate activity (8 mg/mL) was noted in the MIC assay. Conversely, significant activity was noted for K. pneumoniae (5 mm) in the disc diffusion assay but lower sensitivities (16 mg/mL) noted in the MIC assay. Some correlation did exist, as in the case of S. aureus (2 mm; 16 mg/mL), B. subtilis (1 mm; 16 mg/mL), C. neoformans (4 mm; 4 mg/mL) and C. albicans (1.5 mm; 16 mg/mL). Method variation is further discussed in Chapter 11. As L. javanica is mainly used in African traditional medicine to treat respiratory disorders such as coughs, colds and bronchitis, time-kill assays were performed on the three respiratory pathogens from three different micro-organism groups, K. pneumoniae (Gram-negative), C. neoformans (yeast) and B. cereus (Gram-positive) which are commonly associated with opportunistic infections in immune-compromised patients. The essential oil displays mostly moderate antimicrobial activity against respiratory pathogens in the disc diffusion and MIC methodology. By further examination using death kinetics, a more detailed pharmacodynamic approach may be observed for the oils when exposed to respiratory pathogens. This antimicrobial assay shows the bactericidal effect of the oil at concentrations 0.25, 05, 0.75 and 1%. For K. pneumoniae (Figure 5.4), a total rapid bactericidal efficacy was reached within 30 min for all oil concentrations. The lower oil concentrations (0.25% and 0.5%) indicated an initial gradual decline in CFU within 20 min after which the viable bacterial counts were reduced rapidly. The higher oil concentration (0.75%) also indicated an initial rapid reduction in viability within 8 min and thereafter a gradual decline in viability. The highest oil concentration investigated (1%) indicated an immediate reduction in CFU over time resulting in very few viable colonies still evident after 7 min. Total cidal activity was obtained at 30 min. This correlates with the high disc diffusion activities noted for this pathogen. 74 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 0.5 1 1.5 2 2.5 3 3.5 4 Time (hrs) CF U/ m l control 0.25% 0.50% 0.75% 1% 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 1 2 3 4 5 6 7 8 Time (hrs) CF U/ m l Control 0.25% 0.5% 0.75% 1% Death kinetics for C. neoformans (Figure 5.5) showed a killing rate for concentrations 0.5, 0.75 and 1% within one hour. The lowest concentration of 0.25% (3 mg/mL) took eight hours before a cidal effect was noted and death kinetics correlate with the MIC value (4 mg/mL) obtained for this pathogen. While it initially appears that L. javanica is more effective against K. pneumoniae, one must take into account the growth response time of the organism C. neoformans. The growth response time is longer for yeasts, as seen in the control where growth only increases exponentially after four hours. Cidal efficacy is therefore expected to take longer. The MIC (4 mg/mL) and disc diffusion (4 mm) results indicate one of the highest sensitivities towards C. Figure 5.4 Time-kill plot of L. javanica essential oil showing death kinetics of K. pneumoniae (NCTC 9633) represented over the first four hours of a 24 hr test period. Figure 5.5 Time kill plot of L. javanica essential oil showing death kinetics of C. neoformans (ATCC 90112) represented over the first eight hours of a 24 hr test period. 75 neoformans and further substantiate the use of L. javanica essential oils to treat infections associated with this organism. 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 0 0.5 1 1.5 2 2.5 3 3.5 4 Time (hrs) CF U/ m l control 0.25% 0.5% 0.75% 1% Bacillus cereus (Figure 5.6) showed a reduction in growth within 30 min for all the concentrations studied. Thereafter, growth maintained a generally consistent non-responsive pattern neither increasing nor decreasing CFU over the 24 hr test period. While this initially suggests that the essential oil of L. javanica has no significant effect on B. cereus, one must consider that the test pathogen does contain spores and while initial results show a reduction in CFU possibly suggesting the cidal efficacy against vegetative cells, the later growth could possibly be due to the germination of spores. When comparatively examining the growth response of the essential oil concentrations against the control after 30 min, it can be noted that the CFU remained consistently lower than the control. This indicates that a bacteriostatic effect is occurring. In therapeutic treatment regimens, a patient is usually given more than one antimicrobial treatment dose and one could postulate that if another dose (at the concentration given initially) of essential oil were to be introduced after 30 min (when cidal action begins to cease) it may have resulted in cidal action instead of the static effect noted. The death kinetic plot for B. cereus does correlate with the disc diffusion assay (1 mm) and to a lesser extent the MIC (8 mg/mL). Figure 5.6 Time-kill plot of L. javanica essential oil showing death kinetics of B. cereus (ATCC 11778) represented over four hours of a 24 hr test period. 76 Comparison of the time-kill plots for the three organisms studied showed that the killing rate was the greatest for K. pneumoniae, thereafter C. neoformans and very little reduction of microbial populations for B. cereus. A time-kill study was undertaken with L javanica and A. afra, two plants often used in combination therapy to treat respiratory infections amongst the ethnic populations. The combination time-kill study for L. javanica and A. afra (Figure 5.7) and tabulated results (Table 5.4) are shown for each plant independently and in combination. Artemisia afra when studied independently showed initial microbial destruction within one hour, but regrowth after 24 hr. These results correlated with the death kinetics performed on A. afra found in Figure 4.4 (Chapter 4) where cidal activity was seen within one hour, however, the extended assay given here over 48 hr indicated that the antimicrobial efficacy for A. afra at 0.25% could not be maintained by long term exposure as regrowth occurred after 24 hr. For the L. javanica oil at 0.25%, death kinetics was observed within 40 min but regrowth after four hours. As with A. afra, the regrowth was not noted in previous death kinetic assays on the same oils. The microbial load was increased in this study and this could account for the extra bioburden placed on the oils. When the two plants were combined, a bactericidal effect was maintained for the full 48 hours of testing. This synergistic effect scientifically validates the combined use of L. javanica and A. afra for the treatment of respiratory infections associated with K. pneumoniae and corroborates with the traditional use of these two plants when used in combination. CFU over time* Sample 0 min 15 min 1 hr 8 hr 24 hr 48 hr K. pneumoniae control >1000 >1000 >1000 >1000 >1000 >1000 Artemisia afra (0.25% ) >1000 300 0 0 0 >1000 Lippia javanica (0.25%) >1000 >1000 0 1000 >1000 >1000 A. afra (0.125%) and L. javanica (0.125%) >1000 >1000 0 0 0 0 * Shaded areas indicate CFU >1000. Table 5.4 The CFU on agar plate obtained for 0.25% essential oils against K. pneumoniae of L. javanica and A. afra independently and in combination. 77 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 4 8 12 16 20 24 28 32 36 40 44 48 Time (hrs) CF U/ m l Control A. afra L. javanica A. afra + L. javanica A review of the literature on plant combination therapy revealed that no death kinetic assays have been previously presented for plant combinations. A disc diffusion study was undertaken (Jain and Kar, 1971; Geda, 1995), where oils were mixed and applied to discs in a 1:1 ratio. These studies have shown selective synergy for some oils. In another study (Lachowicz et al., 1998) linalool was combined with methyl charvicol at v/v ratios of 1:0; 0.8:0.2; 0.6:0.4; 0.4:0.6; 0.2:0.8 and 0:1. It was observed that the combined concentrations showed higher efficacies than the constituents independently, when tested against two test micro-organisms. Furthermore, medicinal plants administered by traditional healers often make use of a combination of species to enhance efficacy. A comprehensive study on the ethnobotanical use of plant mixtures is given by Cano and Volpato (2004), where 170 plant species from Cuba were examined for their combined medicinal use. Other pharmacological studies on plant species have also been undertaken. A study on the combined synergistic efficacy between Kava kava and Passiflora was undertaken where there was a significant sedative prolongation with combined use (Capasso and Sorrentino, 2005). Panax ginseng with Ginkgo biloba was shown to have improved cognitive function when combined (Scholey and Kennedy, 2002). More specifically, in spite of the many anecdotal accounts of plants used in combination for the treatment of microbe related infections, very few in vitro antimicrobial combination studies have been undertaken to validate the role of antimicrobial synergism in phytotherapy. Some documented accounts include Portulaca quadrifida with Monadenium lugardiae to treat stomach complaints; Trichilia emetica with Cyathula natalensis for leprosy and Momoridica foetida with Pittosporum viridiflorum for boils (Hutchings et al., 1996). A follow-up literature search yielded only three studies on plants used Figure 5.7 Time-kill plot of L. javanica and A. afra independently and in combination for 0.25% essential oil against K. pneumoniae (NCTC 9633) represented over 48 hr. 78 in combination for antimicrobial therapy. A disc diffusion study was undertaken by Hsieh et al., (2001) where Cinnamomum cassia was combined with Allium tuberosum and the fruit of Cornus officinalis at different ratios. Results indicated combined efficacy at various ratio combinations. A more recent study on the combined effect of Origanum vulgare and Vaccinium macrocarpon was undertaken using the disc diffusion assay, optical density evaluation and colony counts. After treatment over a number of days, efficacy was enhanced in combination (Lin et al., 2005). Another combination study using isolobolograms to plot the combined MIC in relation to the MIC as determined independently was undertaken with Salvia chamelaeagnea and Leonotis leonurus. The ethnobotanical use of S. chamelaeagnea and L. leonurus in combination to treat respiratory infections was validated (Kamatou et al., 2006). While these studies cannot be directly compared to the time-kill efficacy examined here, they do support the hypothesis that synergy between plants of different species exist. 5.7 General conclusions ? The major constituents of the essential oil of Lippia javanica are linalool (65.2%), and (Z)- ?-ocimene (13.0%), representing 78.2% accumulatively. ? The highest antimicrobial activities obtained by means of the disc diffusion assay were noted for Klebsiella pneumoniae (5 mm). ? The highest antimicrobial activity noted in the MIC methodology was against Escherichia coli and Cryptococcus neoformans (4 mg/mL). ? Death kinetic assays for the three organisms studied showed that the killing rate was the greatest for Klebsiella pneumoniae, thereafter Cryptococcus neoformans and very little reduction of microbial populations for Bacillus cereus. ? The ethnobotanical use of Lippia javanica to treat dysentery, diarrhoea and suspected food poisoning is validated by the highest antimicrobial activity noted in the MIC methodology against Escherichia coli. 79 ? The ethnobotanical use of Lippia javanica to treat coughs, colds, bronchial and respiratory ailments is validated by the highest efficacies obtained in the MIC methodology, time-kill assays as well as the use in combination with Artemisia afra. ? The time-kill study for Lippia javanica with Artemisia afra indicates that the oils in combination act synergistically against the test organism Klebsiella pneumoniae. 80 Chapter 6 Helichrysum cymosum (L.) D. Don subsp. cymosum, antimicrobial activity of the essential oil, extract and isolated bioactive compound helihumulone. 6.1 Introduction The genus Helichrysum, belonging to the Asteraceae family consists of approximately 500 species (Bougatso et al., 2004). South Africa harbours in the region of 245 species of Helichrysum and biological studies have been performed on a number of these species. Of the many species that have been investigated (Hutchings and van Staden, 1994; Meyer and Dilika, 1996; Afolayan and Meyer, 1997; Dilika et al., 1997; Mathekga and Meyer, 1998; Lourens et al., 2004; Scott et al., 2004; Yani et al., 2005) Helichrysum cymosum D. Don subsp. cymosum has received very little attention. A non-quantitative antimicrobial screening has been reported for the ethanol extracts of H. cymosum where activities were selectively observed for Gram-positive organisms only (Sindambiwe et al., 1999). More recently (Bougatso et al., 2004; van Vuuren et al., 2006) reported the composition of the essential oil and antimicrobial activity. Up until now, the research focus of this thesis has been on the volatile components of aromatic plants used in traditional healing rites. However, one must take cognizance that non-volatile compounds may be responsible for antimicrobial efficacy and one should not focus on the volatile fraction only when studying aromatic plants. With this in mind, the study of H. cymosum subsp. cymosum was undertaken to additionally investigate the activity of the non-volatile compounds not present in the essential oils. 6.2 Botanical description Members of the genus Helichrysum are usually aromatic, perennial shrubs, having dense leaves with hardy yellow flower heads (Figure 6.1). The genus is characterized by hairy leaves and persistent flower heads, which when dried keep their shape for years (van Wyk et al., 1997). The word Helichrysum is derived from the Greek word ?helios? meaning sun and ?chrysos? meaning gold, referring to the distinctive flowers of the genus. 81 6.3 Distribution Figure 6.2 illustrates the geographical distribution of H. cymosum subsp. cymosum in South Africa. The species is distributed mainly along the south and easterly coastal regions from the Western Cape through to KwaZulu-Natal. 6.4 Medicinal uses The medicinal uses of Helichrysum spp. have been widely reported. Extracts of various species have been used to treat topical infections, respiratory ailments and have been incorporated in dressings used after circumcision rites (Meyer and Dilika, 1996; Mathekga and Meyer, 1998). Administration by inhalation suggests that volatile compounds may play a role in antimicrobial therapy and several studies indicate significant antimicrobial properties for Helichrysum oils (Hutchings and van Staden, 1994; Roussis et al., 2000). The traditional use as an analgesic and anti-inflammatory suggests that it may be used for the treatment of other tropical diseases such as malaria (Hutchings and van Staden, 1994; van Vuuren et al., 2006). Of all the Helichrysum species found in South Africa, H. cymosum subsp. cymosum is among the best known and most commonly used in traditional healing practices (Bougatso et al., 2004). The route of administration varies from inhalation of burned leaves (possibly suggesting active volatile Figure 6.1 Helichrysum cymosum subsp. cymosum in habitat. Figure 6.2 Geographical distribution of H. cymosum subsp. cymosum in South Africa (SANBI). 82 constituents) to oral ingestion of a tea preparation (possibly extracting more polar non-volatile constituents), with extracts applied to topical infections. 6.5 Methods 6.5.1 Chemical aspects Plant collection and distillation of essential oils: Plant material was harvested from the wild, during April near Oudtshoorn, one kilometre east of Robinsons Pass. A voucher specimen (Table 6.1) is deposited in the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plants were hydrodistilled as described in Chapter 2.5.1. Table 6.1 Plant collection data for H. cymosum subsp. cymosum. Gas chromatography combined with mass spectrometry (GC-MS) data: Oil samples were analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. 6.5.1.1 Isolation of antimicrobial compound Extraction: The extracts were prepared by submerging the dried macerated plant material (297 g) in acetone overnight. Plant material was retained at room temperature and thereafter filtered. The extraction solvent was evaporated at 60?C using a rotary evaporator (B?chi) for approximately three hours. Bio-autographic assays: Thin layer bio-autographic assays (Marston and Hostettmann, 1999) were carried out by placing 5 ?l of the acetone extract (50 mg/mL) on silica gel thin layer chromatography (TLC) plates (Alugram ? Sil G/UV254, 0.2 mm). The plates were eluted with toluene 72%, dioxane 20% and acetic acid 8% (Merck). Two TLC plates were prepared under identical conditions, one to be incubated with the test organism S. aureus and Tryptone Soya agar, and a reference plate without test organism and agar, kept at ambient temperature. A base Plant Voucher Material distilled (g) Essential oil yield (% w/w) H. cymosum subsp. cymosum JV790 1440 0.2 83 layer (15 mL) of Tryptone Soya agar was poured into a sterile petri dish. The developed TLC plate was dried and sterilized under ultra violet (UV) light at 254 nm for one hour. Thereafter it was placed onto the agar surface with the silica side facing upwards. An overlay of Tryptone Soya agar (15 mL) containing a bacterial culture S. aureus, ATCC 12600 (approximate inoculum size, 1 x 106 CFU/mL) was placed on top of the TLC plate. The plates were allowed to prediffuse at 4 ?C for one hour prior to incubating at 37 ?C for 24 hr. The resulting inhibition zones were visualized around active compounds by spraying with 0.04% (w/v) INT and compared to the reference TLC plate (Figure 6.3). The acetone extract of H. cymosum subsp. cymosum was subsequently subjected to bioassay-guided fractionation. Bioactivity-guided fractionation and isolation using column chromatography: Column chromatography was used as a purification technique to isolate the antimicrobial compound. Several solvent systems were prepared and run with the crude acetone extract to determine the ideal mobile phase, of which 100% chloroform (Merck) was selected. The extract (4.8 g) was dissolved in chloroform to produce a slurry. The column was prepared by plugging the tap with cotton wool and filling the column with silica gel 60 (0.063-0.2 mm / 70-230 mesh ASTM, Macherey-Nagel). The slurry (extract together with chloroform) was applied to the top of the column. Four consecutive columns, packed with silica gel 60 and eluted with 100% chloroform were prepared. With each column, fractions were collected, and samples run on TLC which was subsequently analyzed by viewing the chromatograms under UV light at 254 nm and 366 nm. Similar fractions were combined and concentrated to dryness with the aid of a rotary evaporator. Combined fractions were subjected to bio-autographic assays against S. aureus (Figure 6.4). The bioactive fractions were further purified on Sephadex LH-20 (Sigma-Aldrich) eluted with methanol to yield 277 mg of an isolated biologically active compound. Figure 6.5 summarizes the bioactivity-guided fractionation procedure. 84 A prominent zone of inhibition was observed for the compound. Solvent system = 100% chloroform. Rf = 0.26 Crude extract Three bioactive fractions from the second column. Figure 6.3 Bio-autographic assay of the crude acetone extract of H. cymosum subsp. cymosum against S. aureus. Figure 6.4 Follow-up isolation of the biologically active compound by assay-guided fractionation using S. aureus as a test pathogen. Solvent system = 72% toluene; 20% dioxane; 8% acetic acid. Rf = 0.69 85 Crude H. cymosum subsp. cymosum acetone extract 5.70 g Silica gel 60 Chloroform 422 fractions fractions 1-15*1 fractions 16-135*1 fractions 136-255*2 fractions 256-422*1 0.59 g 1.21g 3.35 g 0.55 g R f = 0.51 R f = 0.56 R f = 0.60 R f = 0.65 Silica gel 60 Chloroform 321 fractions fractions 1-107*1 fractions 108-143*2 fractions 144-152*2 fractions 153-248*2 fractions 249-321*1 0.20 g 0.16 g 0.19 g 1.50 g 0.67 g R f = 0.71 R f = 0.71 R f = 0.69 R f = 0.69 R f = 0.70 Silica gel 60 Chloroform 58 fractions fractions 1-10 fractions 11-24*2 fractions 25-37*1 fractions 38-53*1 fractions 54-58*2 0.44 g 0.61 g 0.25 g 0.16 g 0.03 g discard R f = 0.64 R f = 0.64 R f = 0.61 R f = 0.70 70 fractions Silica gel 60 Chloroform fractions 1-19 fractions 20-25*2 fractions 26-70*2 220 mg 170 mg 600 mg discard R f = 0.74 R f = 0.75 Sephadex LH-20 86 Methanol 33 fractions fractions 1-10 fractions 11-18*2 fractions 31-32*1 fraction 33*1 219 mg 277 mg 4.9 mg 9.87 mg discard R f = 0.75 R f = 0.71 R f = 0.70 Figure 6.5 Schematic representation of the bioactivity guided fractionation procedure. *1 No antimicrobial activity was noted on further bio-autography investigations *2 Antimicrobial activity was noted for S. aureus on further bio-autography investigation. Identification and elucidation of isolated compounds: The isolated compound was subjected to nuclear magnetic resonance spectrometry (NMR) and MS for identification in collaboration with the Department of Chemistry and Biochemistry at the University of Johannesburg. Spectra were recorded on a Varian Inova 2000 300 MHz spectrometer. All spectra were recorded at 25 ?C in deuterated chloroform and the chemical shifts were recorded in ppm referenced to tetramethylsilane or the solvent shift. Electron ionization MS of the compound was performed by direct inlet at 70 eV on the GCMS-QP2010 gas chromatography-mass spectrometer. The crude extract constituents were analysed by high performance liquid chromatography (HPLC) coupled to MS and using a Waters 2690 HPLC system (Phenomenex Aqua C18 column, 250 mm x 2.1 mm at 40 ?C) and equipped with a 996 photodiode array detector (PDA) and Thermabeam mass selective detector (TMD). The samples were dissolved in methanol at a concentration of 50 mg/mL and 10 ?l injected under the following conditions: mobile phase flow rate; 0.2 mL/min, nebulizer gas flow; 30 L/h, nebulizer temperature; 80 ?C, expansion region; 90 ?C and source temperature; 225 ?C. The initial mobile phase was acetonitrile, 90% water containing 100 mM formic acid and the solvent ratio was changed through a linear gradient to 90% acetonitrile, 10% water with 100 mM formic acid at 40 min. This ratio was maintained for Antimicrobial compound. Intense blue fluorescence at 366 nm, started crystallizing immediately after isolation. 87 10 min after which the solvent ratio was converted to that of the initial starting conditions. Data analysis was performed using the Empower? software program. 6.5.2 Antimicrobial aspects Culture, media preparation and disc diffusion assays were undertaken according to the NCCLS (2003) guidelines and additional references as described in Chapter 2.5.2. Preliminary antimicrobial screening by disc diffusion assays was performed with the essential oil and the acetone extract on ten reference strains: Yersinia enterocolitica, E. faecalis, B. cereus, B. subtilis, S. aureus, P. aeruginosa, E. coli, K. pneumoniae, C. albicans and C. neoformans. The MIC was determined using the micro-dilution bioassay method as in Chapter 2.5.2, where the essential oil, crude acetone extract and the isolated bioactive fraction were examined against the same ten pathogens as investigated in the disc diffusion assay. Two strains were introduced which were not studied in Chapters 2-5. The acceptable control ranges for ciprofloxacin against these additional pathogens used in the MIC assays are given in Table 6.2. The starting concentration for the essential oil was 128 mg/mL. The starting concentration for the extract and bioactive fraction for MIC determination was 64 mg/mL and 5 mg/mL respectively. Table 6.2 Acceptable MIC ranges for ciprofloxacin control. Test organism MIC controls (?g/mL)* Yersinia enterocolitica (ATCC 23715) 0.39-1.00 Klebsiella pneumoniae (ATCC 13883) 0.12-0.80 6.6 Results and discussion 6.6.1 Essential oil chemistry Sixty-five compounds were identified in the H. cymosum subsp. cymosum essential oil representing 85.2% of the total composition (Table 6.3). * Acceptable ranges as determined within the Department of Pharmacy and Pharmacology, University of the Witwatersrand (controls in this study were run for all test organisms). 88 Table 6.3 Essential oil composition of H. cymosum subsp. cymosum. RRI* Compound name Area percentage 1032 ?-pinene 12.4 1072 ?-fenchene 0.1 1076 Camphene 0.4 1118 ?-pinene 3.7 1132 Sabinene 0.2 1174 Myrcene 1.2 1188 ?-terpinene 1.2 1203 Limonene 7.2 1213 1,8-cineole 20.4 1246 (Z)-?-ocimene 9.5 1255 ?-terpinene 1.4 1266 (E)-?-ocimene 3.6 1280 p-cymene 2.4 1290 Terpinolene 0.4 1320 2-heptanol 0.1 1386 1-hexanol tr 1382 cis-allo ocimene 0.4 1391 (Z)-3-hexen-1-ol 0.1 1398 2-nonanone 0.4 1426 2-octanol 0.5 1452 1-octen-3-ol 0.3 1493 ?-ylangene 0.1 1497 ?-copaene 1.20 1521 2-nonanol 0.1 1535 ?-bourbonene 0.6 1541 Benzaldehyde 0.1 1544 ?-gurjunene 0.1 1553 Linalool 0.1 1586 pinocarvone 0.1 1591 fenchylalcohol 0.7 1612 ?-caryophyllene 10.8 1628 aromadendrene 1.5 1631 selina-5,11-diene 0.3 1664 trans-pinocarveol 0.6 1682 ?-terpineol 0.9 1687 ?-humulene 1.3 1700 Limonene-4-ol 0.2 1706 ?-terpineol 2.6 1719 Borneol 1.8 1730 ?-guaiene 1.1 1740 ?-muurolene 0.3 89 RRI* Compound name Area percentage 1773 ?-cadinene 0.6 1804 myrtenol 0.6 1838 2-phenyl ethyl acetate tr 1845 trans-carveol 0.2 1853 cis-calamenene 0.6 1864 p-cymen-8-ol 0.1 1889 benzylacetone 0.3 1896 cis-p-mentha-1(7), 8-dien-2-ol tr 1941 ?-calacorene 0.1 2001 isocaryophyllene oxide 0.1 2008 caryophyllene oxide 1.6 2033 epiglobulol 0.3 2071 humulene epoxide II 0.2 2073 ?-caryophyllene alcohol 0.4 2098 globulol 0.6 2104 viridiflorol 0.8 2144 rosifoliol 0.2 2184 cis-p-menth-3-en-1,2-diol 0.1 2204 clovenol 0.1 2255 ?-cadinol 0.1 2316 caryophylladienol I tr 2324 caryophylladienol II 0.4 2389 caryophyllenol I 0.2 2392 caryophyllenol II 0.1 TOTAL 85.2 *RRI: Relative retention indices as eluted from a polar column. % calculated from TIC data tr = Trace <0.1% Major compounds: The chemical constituents of the essential oil (Table 6.3) analyzed by GC- MS indicate the major compounds of H. cymosum subsp. cymosum (Figure 6.6) as ?-pinene (12.4%), 1,8-cineole (20.4%) and ?-caryophyllene (10.8%) comprising 43.6% of the total oil composition. O H H 1,8-cineole ?-pinene ?-caryophyllene Figure 6.6 Chemical structures for major compounds identified in the essential oil of H. cymosum subsp. cymosum. 90 6.6.2 Elucidation and identification of the isolated compound Based on 1H, 13C, COSY, HMQC and HMBC experiments, the compound was identified as the known helihumulone, a phloroglucinol-derived compound previously isolated from H. cymosum (Bohlmann et al., 1979). The chemical shift data (Table 6.6) as well as the carbon (13C, Figures 6.8) and proton (1H, Figure 6.9) NMR spectra for helihumulone can be found as an appendix to this Chapter. O O OH HO CH3O 1 2 3 4 5 6 7 8 910 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 helihumulone Figure 6.7 The HPLC profile of H. cymosum subsp. cymosum with the major peak identified as helihumulone. The HPLC profile (Figure 6.7) indicates a single major compound. Based on 1H, 13C, COSY, HMQC and HMBC experiments, this compound was identified as the known helihumulone, a phloroglucinol-derived compound previously isolated from H. cymosum subsp. cymosum (Bohlmann et al., 1979). AU 0.00 0.20 0.40 0.60 0.80 1.00 Minutes 5 .00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 91 6.6.3 Antimicrobial activity The preliminary disc diffusion results (Table 6.4), indicated good activities only for the Gram- positive organisms when exposed to the acetone extracts. * Neomycin and nystatin served as controls for bacteria and yeasts respectively. On further bio-autographic investigation of the crude acetone extract, a single prominent zone of inhibition was observed for the compound with a Rf value of 0.26 when tested against S. aureus (Figure 6.3) thus suggesting that a single compound could possibly be responsible for antimicrobial activity in this species. Helihumulone has previously been isolated from H. cymosum (Bohlmann et al., 1979), however, no report was made on any of the biological activities of the compound. Helihumulone was further quantitatively examined for antimicrobial activity by the MIC assay together with the essential oil and acetone extract (Table 6.5). Even though antimicrobial activity was only selectively found in the disc diffusion assay, MIC evaluations were performed for all test pathogens, as false negatives, often associated with disc diffusion assays may have occurred. This method variation is further discussed in Chapter 11. Pathogen Extract Oil Control* Enterococcus faecalis (ATCC 292192) 3.7 0 4.0 Bacillus cereus (ATCC 11778) 7.0 0 6.0 Bacillus subtilis (ATCC 6051) 5.7 0 11.0 Staphylococcus aureus (ATCC 25923) 8.0 0 3.0 Pseudomonas aeruginosa (ATCC 9027) 0 0 3.0 Escherichia coli (ATCC 11775) 0 0 5.0 Yersinia enterocolitica (ATCC 23715) 0 0 5.0 Klebsiella pneumoniae (ATCC 13883) 0 0 6.0 Cryptococcus neoformans (ATCC 90112) 0 0 5.0 Candida albicans (ATCC 10231) 0 0 5.0 Table 6.4 Zones of inhibition (measured in mm from edge of disc) of the acetone extract and essential oil of H. cymosum subsp. cymosum. 92 * Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. The MIC values (mg/mL) ranges from 1 - 8 (essential oil), 0.078 - 0.313 (crude extract) and 0.016 - 0.125 (helihumulone) indicating that the crude extract is at least six times more antimicrobially active than the essential oil, and similarly the isolated compound is more active than the crude extract. The highest sensitivity for the essential oils was against E. faecalis (1 mg/mL). This pathogen is associated with respiratory disorders and could possibly be correlated with the traditional inhalation use of H. cymosum. Previous studies on the antimicrobial activity of ?-pinene, a major compound (12.4%) found in the essential oil of H. cymosum have shown moderate (Knobloch et al., 1989; Dorman and Deans, 2000; Tzakou et al., 2001) to poor activity (Hinou et al., 1989; Moleyar and Narasimham, 1992; Griffin et al., 1999; van Zyl et al., 2006). Thus, it is probable that this compound does not play a major role in the antimicrobial efficacy of the plant. The highest sensitivities for the extracts were against B. cereus and C. neoformans (78 ?g/mL). The highest sensitivity for the bioactive compound helihumulone may be attributed to P. aeruginosa (31 ?g/mL), a pathogen often associated with wound infections. The numerous accounts (Watt and Breyer-Brandwijk, 1962; Hutchings et al., 1996; Meyer and Dilika, 1996; Mathekga and Meyer, 1998) of the Helichrysum genus as a dressing for wounds is validated by the high efficacies noted for P. aeruginosa. With the exception of the activity for S. aureus (125 ?g/mL), helihumulone indicated good activity with values less than 64 ?g/mL. A review publication on anti-staphylococcal compounds (Gibbons, 2004) has suggested that only compounds with a MIC value of less than 64 ?g/mL have clinical antimicrobial significance. Table 6.5 MIC (mg/mL) for H. cymosum subsp. cymosum essential oil, acetone extract and helihumulone against ten pathogens. Pathogen Essential oil Extract Helihumulone Control* Enterococcus faecalis (ATCC 292192) 1.000 0.156 0.031 2.0 x 10-3 Bacillus cereus (ATCC 11778) 8.000 0.078 0.063 1.0 x 10-3 Bacillus subtilis (ATCC 6051) 8.000 0.313 0.063 3.0 x 10-3 Staphylococcus aureus (ATCC 25923) 8.000 0.313 0.125 0.5 x 10-3 Pseudomonas aeruginosa (ATCC 9027) 4.000 0.313 0.016 1.2 x 10-3 Escherichia coli (ATCC 11775) 8.000 0.313 0.063 0.1 x 10-3 Yersinia enterocolitica (ATCC 23715) 8.000 0.313 0.063 1.0 x 10-3 Klebsiella pneumoniae (ATCC 13883) 8.000 0.313 0.063 0.2 x 10-3 Cryptococcus neoformans (ATCC 90112) 4.000 0.078 0.031 1.0 x 10-3 Candida albicans (ATCC 10231) 4.000 0.313 0.063 2.5 x 10-3 93 Considering this statement one can therefore deduce that the antimicrobial activity of helihumulone was significant for nine of the ten pathogens studied. No antimicrobial activity was found in the essential oil study of the Tanzanian H. cymosum species (Bougatso et al., 2004) when investigated against nine test micro-organisms. No starting concentrations were given in the methodology and therefore comparative examination cannot be made, as starting concentrations may have been lower than the MIC?s obtained in this study. Major compounds in the Tanzanian H. cymosum essential oil were trans-caryophyllene (27.02%) and p-cymene (7.55%). ?-Pinene, a major compound found in the essential oil of this study (12.4%) was detected only as a minor compound in the Tanzanian H. cymosum essential oil (0.75%). Also 1,8-cineole (20.4%), a major compound in this study is absent in the Tanzanian H. cymosum essential oil. From these essential oil comparisons it is not surprising that different biological activities were noted. As noted in previous studies (Griffin et al., 2000; Dorman and Deans, 2000), the antimicrobial activity is merely an expression of the chemical composition and activity is dependent on the structural configuration. These results highlight the importance of examining both essential oils and extracts when the bioactivities of aromatic plants are investigated. As was noted, the highest antimicrobial activity can be attributed to helihumulone isolated from the crude extract. Helihumulone is not found in the essential oil, which could possibly indicate why the essential oil showed lower antimicrobial activities than the crude extract. 6.7 General conclusions ? The major constituents of the essential oil of Helichrysum cymosum subsp. cymosum are ?- pinene (12.4%), 1,8-cineole (20.4%) and ?-caryophyllene (10.8%). ? Using bioassay-guided fractionation, the most active compound was isolated with column chromatography and identified as the known phloroglucinol derivative, helihumulone. ? The MIC ranges for the essential oil were 1? 8 mg/mL with the highest activity noted for the respiratory pathogen Enterococcus faecalis, thus validating the traditional use of the plant to treat respiratory ailments. 94 ? The MIC ranges for the crude extract were 0.078 ? 0.313 mg/mL with highest activity noted for Bacillus cereus and Cryptococcus neoformans. ? The MIC ranges for the isolated compound helihumulone were 0.016 ? 0.125 mg/mL with the highest activity noted for Pseudomonas aeruginosa, thus validating the traditional use of the plant for wound dressings. 95 Appendix Helichrysum cymosum subsp. cymosum Table 6.6 Correlation of NMR chemical shift data 13C and 1H for helihumulone with literature values obtained by Bohlmann et al. (1979). Position 13 C NMR data ?C 13 C NMR literature 1H NMR data ?H 1H NMR data literature 1 141.0s*1 141.0s 2,6 128.2 128.5d 7.25m*5 7.27m 3,5 128.2 128.4d 7.25m 7.35m 4 126.1d*2 126.1d 7.25m 7 39.5t*3 39.7t 2.95m 2.95t 8 41.2t 42.1t 3.30m 9 201.2s 201.3s 10 112.9s 113.3s 11 190.2s 190.4s 12 108.0s 108.3s 13 166.9s 166.4 s 14 84.0s 84.7s 15 192.3s 192.5s 16 31.1t 31.3t 2.54m 2.54dd 17 115.0d 115.4d 4.88m 4.81t(br) 18 137.8s 138.0s 19 17.6q*4 17.8q 1.49 1.59s(br) 20 25.7q 25.8q 1.60 1.55s(br) 21 21.1t 21.3t 3.10m 3.19d(br) 22 121.1d 121.3d 5.12m 5.15t(br) 23 132.5s 132.4s 24 17.6q 17.8q 1.69 25 25.6q 25.7q 1.74 1.79s(br) OMe 53.9q 54.0q 3.15 *1 singlet; *2 doublet; *3 triplet; *4 quartet;*5 multiplet 96 Figure 6.8 The NMR carbon spectra (13C) for helihumulone. 97 Figure 6.9 The NMR proton spectra (1H) for helihumulone. 98 Chapter 7 Croton gratissimus Burch. var. subgratissimus, the antimicrobial activity of the essential oil and extracts, with evidence of synergy for plant part combinations. 7.1 Introduction Despite the traditional use of the genus Croton for its medicinal properties (Watt and Breyer- Brandwijk, 1962; van Wyk et al., 1997; van Wyk and Gericke, 2000), scientific studies validating the therapeutic properties of the indigenous species Croton gratissimus Burch. var. subgratissimus (from the family Euphorbiaceae) are lacking. The leaves of this tree are aromatic and often used in herbal preparations and may also be used with the bark and roots (Hutchings et al., 1996; van Wyk et al., 1997; von Koenen, 2001). It has been suggested when investigating aromatic plants that one should not restrict research only to the volatile essential oils (van Vuuren et al., 2006). This was noted in the study on H. cymosum subsp. cymosum (Chapter 6) where the antimicrobial spectrum of activity for extracts was much higher than that for the essential oils. There are numerous ethnobotanical reports (Watt and Breyer-Brandwijk, 1962; Hutchings et al., 1996; van Wyk et al., 1997; von Koenen, 2001; Ngwenya et al., 2003) of plants used in combination therapy i.e. where plant parts such as roots and leaves are combined and used in therapeutic regimens. Croton is one such plant where the bark, roots and leaves are used independently as well as in combination. This study focused on the antimicrobial potential of C. gratissimus var. subgratissimus and the interactive role of the different plant parts when used in combination for antimicrobial therapy in order to validate the traditional use. 7.2 Botanical description This tree can grow up to ten meters in height. Leaves are distinctive (Figure 7.1) as they are dark green on the upper side and have a shiny gray to silver coloration on the lower surface (van Wyk et al., 1997). The under side of the leaf is scattered with dark brown glands. The bark is rough and gray in colour. Cream-coloured flowers are produced in autumn, which yield small fruit capsules (van Wyk et al., 1997). 99 7.3 Distribution Figure 7.2 illustrates the geographical distribution of C. gratissimus var. subgratissimus in southern Africa. The tree occurs naturally over a large area (Northern Cape, North-West Province, Gauteng, northern KwaZulu-Natal, Mpumalanga and Northern Province) and in the northern parts of South Africa (van Wyk et al., 1997). A number of populations have been recorded in central to northern Botswana and Namibia. 7.4 Medicinal uses The Afrikaans name for the plant ?koorsbessie? (?koors? = fever) suggests that these trees are used as a pyrogenic. All parts of the plant have reputed medicinal value. Leaf infusions are used for coughs and placed in steam baths to reduce fever (van Wyk et al., 1997; Hutchings et al., 1996). Eye lotions are made for livestock from cold infusions of the leaf (Watt and Breyer- Brandwijk, 1962). A deconcoction of the roots has been used for chest complaints, coughs, fever and sexually transmitted diseases such as syphilis (von Koenen, 2001). The bark has the highest reported use including the treatment of bleeding gums, abdominal disorders, skin inflammation, earache and chest complaints, suggesting microbial infections may be responsible (Hutchings et al., 1996; von Koenen, 2001). Various combinations of C. gratissimus have been used. The Figure 7.2 The geographical distribution of Croton gratissimus var. subgratissimus in southern Africa. (SANBI). Figure 7.1 The leaves and fruit of Croton gratissimus var. subgratissimus. 100 combination of roots and bark to treat respiratory disorders has been reported (von Koenen, 2001). Accounts of the administration with other species (Hutchings et al., 1996) have been noted i.e. for the treatment of swellings where the bark of C. gratissimus is combined with the root of Amaryllidaceae species and rubbed into incisions. Also noted, was the use of the bark of C. gratissimus together with the bark of Ocotea bullata where powdered forms are blown into the womb to treat uterine disorders. Even though the plant is considered toxic, the very nature of the irritancy action may serve as a purgative or eruptive emetic thus ridding the body of toxins and microbial infection. 7.5 Methods 7.5.1 Chemical aspects Plant collection and distillation of essential oils: The leaf, bark and root samples (Figure 7.3) used in this study was collected from a single tree in the wild on the western slopes below Diep- in-die Berg, in the Tshwane area during May. A voucher specimen (Table 7.1) is deposited in the Department of Pharmacy and Pharmacology, University of Witwatersrand. Leaf material was hydrodistilled, as described in Chapter 2.5.1. Figure 7.3 The leaves, bark and root of C. gratissimus var. subgratissimus. Root Bark Leaves 101 Table 7.1 Plant collection data for C. gratissimus var. subgratissimus. Gas chromatography combined with mass spectrometry (GC-MS) data: Oil samples were analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. Preparation and analysis of the crude extracts: Extracts of the leaves, bark and roots were prepared by submerging the dried macerated plant material in a 1:1 mixture of methanol (Merck) and chloroform (Merck) overnight. Plant material was retained at ambient temperature and thereafter filtered. The solvent was evaporated at 60 ?C using a rotary evaporator for approximately three hours. High performance liquid chromatography (HPLC) analysis: In order to determine if any major chemical variation between subterranean and above ground foliage exists, HPLC analysis was undertaken on the leaf, bark and root extracts of C. gratissimus var. subgratissimus. The HPLC analysis was undertaken on a Waters 26690 HPLC system operating under the same conditions as described in Chapter 6.5.1.1. 7.5.2 Antimicrobial aspects Culture, media preparation and assays were undertaken according to the NCCLS (2003) guidelines and methods described by Carson et al. (1995) and Eloff (1998a) as described in Chapter 2.5.2. Minimum inhibitory concentrations were determined for the hydrodistilled oil, leaf, bark and root extracts independently on ten test organisms (Figure 7.7). Commercial antimicrobials (ciprofloxacin for bacteria and amphotericin B for yeasts) were included as positive controls in all MIC repetitions and replications. Synergy / antagonistic interaction: Once the independent MIC was determined on the hydrodistilled oil and leaf, bark and root extracts, the interaction between plant parts was Plant Voucher Material distilled (g) Essential oil yield (% w/w) C. gratissimus var. subgratissimus PB975 470.8 0.3 102 investigated in two ways. Firstly, stock solutions (64 mg/mL) of each individual plant part extract were prepared and mixed together in the following combinations: ? 1:1 leaf and root ? 1:1 bark and root ? 1:1 leaf and bark ? 1:1:1 leaf, bark and root. An MIC was determined for these combinations to evaluate if there was a possibility that synergistic interactions were at play. The fractional inhibitory concentration (FIC) was calculated for the leaf with root, bark with root and leaf with bark combination. The FIC is expressed as the interaction of two agents where the concentration of each test agent in combination is expressed as a fraction of the concentration that would produce the same effect when used independently (Berenbaum, 1978). The FIC is then calculated for each test sample independently as specified in the following equations; MIC (a) in combination with (b) FIC (*a) = MIC (a) independently MIC (b) in combination with (a) FIC (*b) = MIC (b) independently (*a) and (*b) = in this study represents either leaf, bark or root. The sum of the FIC, known as the FIC index is thus calculated as: ?FIC = FIC (*a) + FIC (*b) The FIC index (Schelz et al., 2006), is determined as the correlation between the two combined test substances and may be classified as either synergistic (? 0.5), additive (>0.5 ?1), indifferent (>1 ? 4) or antagonistic (?4). 103 Antagonism  MIC of Y in combination Additive / MIC of Y independently Synergy MIC of X in combination / MIC of X independently The second part of this study involved combining the roots and leaves of C. gratissimus var. subgratissimus in nine ratios i.e. 90:10%; 80:20%; 70:30%; 60:40%; 50:50%; 40:60%; 30:70%; 20:80% and 10:90%. The MIC was determined for all ratios and extracts independently. The study was undertaken in triplicate and the mean MIC values (mg/mL) were plotted on an isobologram using Graphpad Prism? software, allowing for a graphical representation of the interaction of the various combinations. The isobologram can be interpreted by examining the data points of the ratios where the MIC for each concentration is determined in relation to the independent MIC?s (shown as a straight line), and extrapolating synergy (below the line), antagonism (above the line) and additive in the vicinity closest to or on the line (Figure 7.4). Conventional antimicrobials (not shown on isobologram) were included in all repetitions. Only values within the 0-1.25 range were plotted on the isobologram. 7.6 Results and discussion 7.6.1 Essential oil chemistry Ninety-four compounds were identified in the C. gratissimus var. subgratissimus essential oil representing 93.3% of the total composition (Table 7.2). The toxicity of the plant may possibly be attributed to the high concentrations of the cyclic monoterpene ?-phellandrene (20.7%) which acts as a skin irritant and may cause vomiting and diarrhoea if ingested (Windholz, 1976). The presence of ?-phellandrene is not restricted to the gratissimus species and can be found in other Croton species such as C. stellulifer as a major constituent (Martins et al., 2000) and C. sakamaliensis as a minor constituent (Radulovi? et al., 2006). Figure 7.4 Isobologram depicting antagonism, synergy and an additive effect where the relative ratio of Y is plotted to the relative ratio of X. 104 Table 7.2 Essential oil composition of C. gratissimus var. subgratissimus. RRI* Compound name % 1032 ?-pinene 6.7 1035 ?-thujene 1.7 1076 camphene 1.8 1118 ?-pinene 1.5 1176 ?-phellandrene 20.7 1188 ?-terpinene 0.7 1203 limonene 2.7 1213 1,8-cineole 8.3 1218 ??phellandrene 1.2 1246 (Z)-?-ocimene 2.2 1255 ?-terpinene 1.3 1266 (E)-?-ocimene 4.5 1280 p-cymene 2.8 1290 terpinolene 1.1 1296 methyl heptanoate tr 1386 1-octenyl acetate 0.2 1391 methyl octanoate tr 1402 (Z)-2-hexen-1-ol 0.1 1450 trans-linalool oxide (Furanoid) tr 1453 1-octen-3-ol 2.1 1466 ?-cubebene 0.1 1476 cis-linalool oxide (Furanoid) 0.1 1479 ?-elemene 0.1 1482 (Z)-3-hexenyl-2-methylbutyrate 0.1 1493 ?-ylangene 0.1 1497 ?-copaene 1.6 1532 camphor 5.0 1535 ?-bourbonene 0.2 1544 ?-gurjunene tr 1553 linalool 2.6 1554 methyl nonanoate 0.1 1571 trans-p-menth-2-en-1-ol 0.1 1589 ?-ylangene 0.1 1597 bornyl acetate 0.1 1601 ?-copaene 0.1 1611 terpinen-4-ol 0.9 1612 ?-caryophyllene 2.1 1617 selina-5,11-diene 0.1 1638 cis-p-menth-2-en-1-ol 0.1 1661 alloaromadendrene 0.5 1677 epi-zonarene 0.1 1682 ?-terpineol 0.1 1687 ?-humulene 0.4 1689 ?-gurjunene tr 1706 ?-terpineol 1.7 105 RRI* Compound name % 1719 borneol 0.9 1725 verbenone 0.1 1726 germacrene D 8.6 1730 bicyclosesquiphellandrene 0.1 1739 ?-muurolene 0.3 1744 ?-selinene 0.1 1755 bicyclogermacrene 0.5 1759 ?-cuprenene 0.1 1773 ?-cadinene 0.8 1776 ?-cadinene 0.3 1808 nerol 0.1 1823 p-mentha-1(7),5-dien-2.ol 0.1 1830 2,6-dimethyl-3(E),5(E),7-octatrien-2-ol 0.1 1854 germacrene B 0.2 1864 p-cymen-8-ol tr 1918 epicubebol 0.1 1941 ?-calacorene tr 1942 benzyl cyanide tr 1943 palustrol tr 1957 cubebol 0.2 1984 ?-calacorene tr 2008 caryophyllene oxide 0.2 2017 methyl eugenol 0.4 2037 salvial-4(14)-en-1-one 0.1 2057 ledol 0.2 2064 ?-copaen-11-ol tr 2069 germacrene D 4-ol 0.1 2080 cubenol 0.1 2088 1-epi-cubenol 0.1 2096 elemol 0.3 2101 germacrene D 1,10-epoxide tr 2127 10-epi-??eudesmol tr 2128 salviadienol tr 2146 spathulenol 0.4 2164 6-epi-cubenol tr 2184 eugenol 0.2 2185 T-cadinol 0.2 2209 T-muurolol 0.2 2214 torreyol 0.1 2255 carvacrol tr 2256 ?-eudesmol 0.2 2257 torilenol 0.1 2258 ?-eudesmol 0.6 2264 intermedeol 0.3 2272 alismol tr 2273 selin-11-en-4?-ol tr 2369 (2E,6E)-farnesol 0.7 2389 eudesma-4(15),7-dien-1?-ol 0.1 106 RRI* Compound name % 2622 phytol 0.1 TOTAL 93.3 *RRI: Relative retention indices calculated against n-alkanes % calculated from FID data tr = Trace (< 0.1 %) Major compounds: Major compounds (Figure 7.5) in the essential oil of C. gratissimus var. subgratissimus comprise of ?-phellandrene (20.7%) and to a lesser extent germacrene D (8.6%) and 1,8-cineole (8.3%), representing 37.6% accumulatively. O ?-phellandrene 1,8-cineole germacrene D 7.6.2 The HPLC analysis of the leaf, bark and root extract The yield for the root, leaf and bark extracts were 4.34%, 4.04% and 6.41% w/v respectively. The HPLC chromatograms for the root, bark and leaf extracts indicate some similarity (Figure 7.6) and the retention times, UV absorbance maxima and percentage areas of the major peaks (>4.00%), identified in each extract, are summarized in Table 7.3. Two major compounds (indicated in bold, Table 7.3) were present in all three samples (leaf, bark and root) with two major constituents present in both leaf and bark only. Six major compounds were noted in the root material with one compound showing similarity with the bark extract. Further quantitative and qualitative similarities were noted with the minor compounds where analysis indicates similarities between bark and root for five compounds. Similarities between eight minor compounds were also noted between bark and leaf. The leaf and root had only two common compounds. Only three compounds present at low concentrations (0.15-2.25%) were present in all samples. The leaves contained 14 compounds not detected in the other samples. Similarly, the bark exhibited six compounds and root eight compounds, not detected in other parts of the plant. Figure 7.5 Chemical structures for major compounds identified in the essential oil of C. gratissimus var. subgratissimus. 107 By combining leaf and root extracts, the number of compounds increases from 14 (leaf) or 8 (root), observed independently, to 22 compounds in combination. As the number of compounds is increased when leaf and root are mixed it is likely that there would be an increased possibility of synergistic interactions that may occur to enhance activity. Figure 7.6 The C. gratissimus var. subgratissimus HPLC chromatogram for root (?), bark (?) and leaf (?). Table 7.3 The HPLC quantification of leaf, bark and root compounds of C. gratissimus var. subgratissimus. Content and area% Retention time UV absorbance maxima leaf bark root 1.625 201.7;281.7;376.7 0.65 - 0.23 1.652 204.0;279.3 - 0.94 - 2.667 205.2;376.7;598.5 2.18 1,68 0.43 4.113 205.2;375.5;592.2 6.01 4.88 3.43 leaf bark root 108 Content and area% Retention time UV absorbance maxima leaf bark root 4.833 205.2;376.7 3.54 - - 5.032 256.9;592.2 - 3.90 - 5.272 264.0;588.4 - - 2.88 7.460 202.9;273.4;354.2 1.25 1.32 - 10.164 208.7;273.4;326.8 19.58 13.87 6.51 11.451 208.7;273.4;354.2 2.94 1.42 - 11.925 218.1;260.4;291.2 1.63 1.92 - 14.056 204.0;279.3;376.7 4.64 1.19 - 14.981 224.0;267.5;301.9 - 11.94 7.23 15.153 234.5;324.4 1.23 - - 16.397 255.7;347.1 0.16 - - 17.007 228.6;269.9;348.3 2.40 - - 17.585 231.0;255.7;348.3 1.23 - - 19.659 234.5;280.5;309.0 - - 0.37 19.818 231.0;268.7;337.5 0.69 - - 20.252 229.8;269.9;337.5 1.58 - - 24.710 242.8;279.3;373.1 - 0.39 - 25.114 233.3;291 - - 1.44 26.197 243.9 - - 0.46 28.556 245.1;373.1 0.44 - - 28.847 236.9;266.3;314.9 4.80 4.56 - 29.238 266.3;313.7;434.7 2.78 2.58 - 30.934 245.1;265.2;469.5 - 0.94 4.64 31.169 266.3;312.5;452.5 0.81 2.58 - 31.368 242.8;565.4 - - 0.71 31.555 251.0;434.4;595.8 - 1.58 - 32.359 247.5;373.1 2.95 - 0.60 32.539 248.6;399.5;382.4 - 3.60 3.61 32.604 252.2 9.76 7.43 5.83 32.830 241.6;373.1;434.4 1.05 3.70 - 33.310 249.8;373.1;403.1 1.25 1.42 - 33.583 251.0;373.1;434.4 - 0.57 0.31 33.916 254.5;350.6 1.40 - - 33.935 399.5;565.4 - - 4.88 34.037 249.8;348.3 2.47 - - 34.174 246.3;334.0;373.1 1.27 0.88 2.25 34.421 243.9;373.1;457.5 - 3.00 - 35.491 24. 245.1;373.1;430.8 0.37 0.81 0.15 35.565 243.9;374.3 1.50 - - 36.409 242.8;373.1;428.4 - 0.72 1.03 37.235 266.3;565.4 - - 0.83 37.324 251.0;373.1;399.5 0.48 0.47 - 37.605 249.8;374.3;406.7 - - 0.07 39.119 251.0;374.3;399.5 0.58 - - 39.374 245.1;565.4 - 1.14 1.09 43.374 249.8;273.4;374.3 0.39 1.64 7.65 45.197 268.7;331.6 0.66 - - 109 Content and area% Retention time UV absorbance maxima leaf bark root 48.280 256.9;273.4;374.3 - 1.82 - 49.857 254.5;319.7;376.7 1.27 - - 52.610 255.7;582.4 - - 1.82 TOTAL 33 29 24 7.6.3 Antimicrobial activity As noted in literature (Janssen et al., 1987; Hinou et al., 1989; Yousef and Tawil, 1980; Pattnaik et al., 1997; Hammer et al., 1999; van Vuuren et al., 2006) and in a number of Chapters in this thesis (i.e. O. asteriscoides, Chapter 3; L. javanica, Chapter 5 and H. cymosum subsp. cymosum, Chapter 6) the correlation between disc diffusion and MIC assays are not always congruent. As MIC determination demonstrates a more quantitative assay for assessing susceptibility, it was selected as the screening method to determine microbiological susceptibility towards the hydrodistilled oil, leaf, bark and root extracts of C. gratissimus var. subgratissimus prior to the combination studies. Figure 7.7 represents the antimicrobial activity determined by the MIC assay (mg/mL) of C. gratissimus var. subgratissimus essential oil, and leaf, bark and root extracts. A literature search on Croton species indicated that a number of species had positive antimicrobial activity for their essential oils (Martins et al., 2000; Alviano et al., 2005). In this study, the essential oils showed moderate activity (4-11 mg/mL) against six of the pathogens studied. Four pathogens (S. aureus, S. epidermidis, B. subtilis and E. coli) showed poor susceptibilities (13-32 mg/mL) towards the essential oils. The leaf, bark and root extract, however, showed higher efficacies than the oils against all the pathogens studied. Upon further analysis of the extracts, no general trend could be found for the Gram-positive organisms. All three Gram-negative organisms (E. coli, P. aeruginosa and K. pneumoniae) showed highest efficacy for the leaf extracts with the bark and root having equivalent susceptibility patterns. Both yeasts indicated a consistent MIC value for all three extracts i.e. C. albicans (6 mg/mL) and C. neoformans (2 mg/mL). In a study by Shale et al. (1999), information was provided by the traditional healers of Lesotho on the administration of plant material. Healers indicated that the roots were the most commonly used part of the plant. From the leaf, bark and root efficacies noted in this study, the root was found to have either equivalent activities (for six pathogens) or higher efficacies (for four pathogens) of all pathogens studied. 110 2432 0 2 4 6 8 10 12 14 16 18 S. au re us S. ep ide rm idis E. fae ca lis B. ce re us B. su bti lis E. co li P. ae ru gin os a K. pn eu m on iae C a lbic an s C.n eo for m an s Micro-organism M IC v al u e (m g/ m L) EO leaves bark root The traditional use of the root, leaf and bark extracts (van Wyk et al., 1997; Hutchings et al., 1996; von Koenen, 2001) together with the findings that the extracts indicated higher antimicrobial properties than the oils, led the focus of this study towards examination of the interaction between the plant parts when used in combination therapy. The leaf, bark and root of C. gratissimus var. subgratissimus were mixed in equal aliquots and thereafter a MIC assay carried out on the combinations. Results (Table 7.4) indicate that when a combination of leaf, bark and root is studied, antimicrobial efficacy is mostly enhanced for all pathogens except one (S. epidermidis) where the MIC in combination is lower than the leaf and root independently, but higher than the MIC obtained for the bark extract alone. The antimicrobial activity for all other pathogens indicated a lower than or equal to MIC value in combination, than when the leaf, bark and root extracts were investigated independently. When the leaf and bark extracts were combined activity was enhanced only for E. faecalis and C. albicans. Antagonism (where the FIC = >4) was found against three of the ten pathogens studied. A predominantly indifferent interaction was noted for leaf and bark combination and only one test organism (C. albicans) demonstrated an additive profile. Figure 7.7 The MIC (mg/mL) of C. gratissimus var. subgratissimus essential oil (EO), and leaf, bark and root extracts. 111 Table 7.4 The MIC values (mg/mL) for the leaf, root and bark C. gratissimus var. subgratissimus extracts, singularly and in combination with the FIC*1. Pathogen Leaf Bark Root 1:1:1 Leaf, bark and root 1:1 Leaf and bark 1:1 Bark and root 1:1 Leaf and root Control *2 Staphylococcus aureus ATCC 12600 2.3 0.6 0.4 0.3 2.0 (4.2) 0.4 (1.6) 1.0 (2.9) 1.0 x 10 -3 Enterococcus faecalis ATCC 29212 6.0 6.0 2.6 1.3 4.0 (1.4) 1.3 (0.7) 1.5 (0.8)1 1. 1.5 (0.8) 0.3 x 10 -3 Bacillus cereus ATCC 11778 1.5 0.9 0.8 0.3 1.0 (1.8) 0.3 (0.7) 0.7 (1.3) 0.5 x 10 -3 Staphylococcus epidermidis ATCC 2223 4.0 0.8 4.0 1.0 4.0 (6.0) 3.0 (4.5) 2.0 (1.0) 1.6 x 10-3 Bacillus subtilis ATCC 6051 1.5 0.5 0.3 0.2 2 .0 (5.3) 0.2 (1.0) 0.5 (1.9) 0.3 x 10 -3 Escherichia coli ATCC 11775 2.0 4.0 4.0 2.0 3.0 (2.3) 1.5 (0.8) 4 .0 (3.0) 0.1 x 10 -3 Klebsiella pneumoniae ATCC 13883 4.0 6.0 6.0 2.0 4.0 (1.7) 3.0 (1.0) 2.0 (0.8) 0.2 x 10 -3 Pseudomonas aeruginosa ATCC 9027 1.0 2.0 2.0 0.6 2.0 (3.0) 1.5 (1.5) 0.5 (0.8) 0.3 x 10-3 Candida albicans ATCC 10231 6.0 6.0 5.3 1.0 4.0 (0.7) >8.0 (>8.0) 2.0 (0.8) 1.3 x 10 -3 Cryptococcus neoformans ATCC 90112 2.0 2.0 2.0 0.6 4.0 (2.0) 0.8 (0.8) 0.4 (0.4) 0.6 x 10-3 FIC*1 is shown in brackets for all 1:1 combinations where two plants were studied. *2 Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. For the bark and root combination, a lower MIC value was obtained for pathogens E. faecalis, B. cereus, B. subtilis, E. coli, K. pneumoniae and C. neoformans, than when investigated against the bark or root independently. The FIC for these pathogens indicated an additive effect (FIC = 0.5- 1) when combined in a 1:1 ratio. For S. aureus having a FIC of 1.6 and P. aeruginosa having a FIC of 1.5, an indifferent profile was observed. The greatest antagonism (FIC = ?4) was noted for the pathogen S. epidermidis (FIC 4.5) and C. albicans (FIC >8). Candida albicans also showed the least sensitivity when combined, with a MIC value of >8 mg/mL when compared with the root and bark independently having MIC values of 5.3 mg/mL and 6 mg/mL respectively. The FIC for C. albicans showed the highest antagonism of all the pathogens studied. No synergy was noted for the root and bark when used in a 1:1 combination. When the leaf and roots were combined in a 1:1 ratio, the test pathogens B. cereus, E. faecalis, S. epidermidis, P. aeruginosa, K. pneumoniae, C. albicans and C. neoformans had MIC values lower in combination than when investigated independently. The FIC for these pathogens with 112 the exception of B. cereus and C. neoformans showed an additive profile. Only C. neoformans demonstrated synergism when combined. The pathogens S. aureus, B. subtilis and E. coli had varying sensitivity patterns in combination with the MIC?s, indicating values lower than or equal to one plant part only. While these results give an indication of interaction between the plant parts, the methodology is restricted to only a 1:1 combination. Isobologram interaction studies were thus undertaken which took into account more than one ratio of the test substance. The leaf and bark combination was selected for further isobologram studies to validate their ethnobotanical therapeutic use. Table 7.5 Raw data for the isobologram construction for of B. cereus. *The 1:1 ratio as determined from the raw data is presented as a red square on the isobolograms. The isobolograms for all ten pathogens are given with adjacent MIC plots (for relative comparison) in Figures 7.8-7.27. In order to clarify how the isobologram were constructed, a layout of the raw data for one test organism (B. cereus) is given in Table 7.5. The first column (A.) represents the microtitre well in which the mixture of the two test substances was prepared. Any arbitrary number may be given. The concentration of the two test substances X and Y were mixed in various ratios. This is reflected in column B. For this study, the roots and leaves were mixed so that the concentration in the first well before further dilution of each ratio is given. Column C presents the MIC results in mg/mL as determined relative to the original concentration given in column B. Column D is the ratio of the MIC results (Column C) with reference to the independent MIC results when not mixed. This ratio is then plotted on the A. B. C. D. Concentration in mg/mL MIC values Ratio values Row in plate X (roots) Y (leaves) X (roots) Y (leaves) X (roots) Y (leaves) 1 16.0 0.0 0.80 0.00 1.000 0.000 2 14.4 1.6 0.90 0.10 1.125 0.050 3 12.8 3.2 0.80 0.20 1.000 0.100 4 11.2 4.8 0.70 0.30 0.875 0.150 5 9.6 6.4 0.90 0.60 1.125 0.300 6 8.0* 8.0* 0.25* 0.25* 0.313* 0.125* 7 6.4 9.6 0.20 0.30 0.250 0.150 8 4.8 11.2 0.15 0.35 0.188 0.175 9 3.2 12.8 0.20 0.80 0.250 0.400 10 1.6 14.4 0.20 1.80 0.250 0.900 11 0.0 16.0 0.00 2.00 0.000 1.000 113 isobologram which figuratively presents the results as either synergistic, antagonistic or additive (Figure 7.4). Ratios are plotted within the scale 1.25:1.25. Any outlying values with antagonistic profiles greater than that within the scale are not displayed on the isobologram. The raw MIC data can be noted in Table A1, Appendix A. The isobologram combination studies for the Gram-positive organisms do not show any specific pattern. For B. cereus, synergy was observed in four ratio points on the isobolograms where the leaf material were equivalent and in majority (Figure 7.9). However, the FIC (Table 7.4) for B. cereus shows indifference (1.3) for the 1:1 ratio of leaf and root combination. Observation of the MIC data for B. cereus (Figure 7.8, Table 7.4), indicates that the combined MIC value (0.7 mg/mL) is lower than the MIC value for leaves (1.5 mg/mL) and MIC value for roots (0.8 mg/mL). The Staphylococci having FIC values of 2.9 and 1.0 for the species aureus and epidermidis respectively (Table 7.4) show similar concentration dependent additive to indifferent profiles, with the 1:1 ratio of the isobole being on the line (Figure 7.11 and 7.13). Results obtained for both combination methods are consistent. Antimicrobial activity of B. cereus 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root B. cereus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.8 Figure 7.9 114 Antimicrobial activity of S. aureus 0 0.5 1 1.5 2 2.5 Plant part M IC v a lu e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root S. aureus 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 MIC roots in combination/MIC roots alone M IC le av es in co m bi n at io n/ M IC le av es al on e Figure 7.10 Figure 7.11 Antimicrobial activity of S. epidermidis 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root S. epidermidis 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.12 Figure 7.13 Antimicrobial activity of B. subtilis 0 0.5 1 1.5 2 2.5 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root B. subtilis 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.14 Figure 7.15 115 Antimicrobial activity of E. faecalis 0 1 2 3 4 5 6 7 Plant part M IC va lu e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root E. faecalis 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.16 Figure 7.17 Figure 7.8 - Figure 7.17 The antimicrobial efficacy of the Gram-positive test organisms (B. cereus, S. aureus, S. epidermidis, B. subtilis and E. faecalis respectively) against the leaf, bark and root extract independently and in a 1:1 combination with the adjacent isobologram of leaf and root combination. Bacillus subtilis indicated antagonism for five of the nine ratios with the highest antagonism noted where the roots were at a 90% concentration. Due to the scale, this is not noted on the isobologram. Where the roots were in majority antagonism was observed. As the root concentration decreases and leaf concentration increases, synergy becomes apparent (Figure 7.15). The isobologram, for this pathogen indicates that depending on the concentration of the combination, synergy may be attained, however, should the concentration vary, antagonism would be encountered. This can also be noted in the adjacent bar graph (Figure 7.14) where the combination is synergistic when comparatively examined with the leaf but not with the root. For E. faecalis, all ratios (Figure 7.17) where leaf concentration was higher than root, a mostly additive profile predominated. Where root concentration is higher than leaf concentration and where concentrations are equal a synergistic profile predominated. The 1:1 ratio (Table 7.4) indicates how efficacy is enhanced when comparatively observing the independent profiles. The FIC (0.8) for E. faecalis indicates an additive effect in the 1:1 combination thus corroborating results obtained in the isobologram study. 116 Antimicrobial activity of K. pneumoniae 0 1 2 3 4 5 6 7 Plant part M IC va lu e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root K. pneumoniae 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.18 Figure 7.19 Antimicrobial activity of P. aeruginosa 0 0.5 1 1.5 2 2.5 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root P. aeruginosa 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n/ M IC le av es al on e Figure 7.20 Figure 7.21 Antimicrobial activity of E. coli 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root E. coli 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n /M IC le av es al on e Figure 7.22 Figure 7.23 Figure 7.18 - Figure 7.23 The antimicrobial efficacy of the Gram-negative test organisms (K. pneumoniae, P. aeruginosa and E. coli) against the leaf, bark and root extract independently and in a 1:1 combination with the adjacent isobologram of leaf and root combination. 117 The isobolograms for the Gram-negative test organisms: K. pneumoniae (Figure 7.19), P. aeruginosa (Figure 7.21) and E. coli (Figure 7.23) all indicate a ratio dependent pattern. All ratios fall mainly within the additive profile range. The 1:1 combination of leaf and root for K. pneumoniae and P. aeruginosa indicated an additive effect when the FIC was determined (Table 7.4). Both yeasts, C. albicans and C. neoformans investigated in combination studies showed synergism for most ratios when observing the isobologram data (Figure 7.25 and 7.27). The FIC for C. albicans (0.8) indicates an additive effect whereas for C. neoformans the FIC (0.4) was synergistic. In both 1:1 FIC combination studies and isobologram ratios the leaf / root combination for C. neoformans indicate the greatest synergistic patterns. The efficacy against C. neoformans, a respiratory pathogen, may be correlated with the traditional use of C. gratissimus where the plant is used as a treatment for lung inflammation. The adjacent MIC bar graphs (Figure 7.24 for C. albicans and Figure 7.26 for C. neoformans) confirm synergism, where it can be noted that MIC values show higher efficacies in combination than when investigated independently. Antimicrobial activity of C. albicans >8 0 1 2 3 4 5 6 7 8 9 Plant part M IC v al u e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root C. albicans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n /M IC le av es al on e Figure 7.24 Figure 7.25 118 Antimicrobial activity of C. neoformans 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Plant part M IC va lu e (m g/ m l) leaf bark root 1:1:1 combination 1:1 bark & leaf 1:1 bark & root 1:1 leaf & root C. neoformans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC roots in combination/MIC roots alone M IC le av es in co m bi na tio n /M IC le av es al on e Figure 7.26 Figure 7.27* Figure 7.24 - Figure 7.27 The antimicrobial efficacy of the yeasts (C. albicans and C. neoformans) against the leaf, bark and root extract independently and in a 1:1 combination with the adjacent isobologram of leaf and root combination. *Due to the close proximity of the ratio points on the isobologram, no red square was inserted for the 1:1 ratio as this would impair data visualization. Research into antimicrobial synergy configurations has in the past been very vaguely defined and experimental procedures and conclusions based on only one experimental procedure (Lambert et al, 2003). Synergy as determined by the FIC method varies in interpretation of results and may either be expressed as an FIC of <1 (Berenbaum, 1978) or ? 0.5 (ESCMID, 2000). The FIC synergistic investigation of C. gratissimus var. subgratissimus was standardized according to the most recent literature (Schelz et al., 2006) with true synergy been expressed as a FIC of ? 0.5. Antimicrobial synergy assessment for this study was based on more than one method and as results indicate, methodologies were not always in agreement. In Chapter 11.13, method variation with regard to synergy studies is discussed further. If one compares the FIC methodology with the isobolograms, five of the ten pathogens show congruency, with two additional pathogens (S. aureus and B. subtilis) showing partial congruency. One possible hypothesis for the variability of these results could be due to dissimilar mechanisms of interaction as noted by different methods (Beale and Sutherland, 1983). It is clear, however, that when comparing the different plant part combinations in a 1:1 ratio, the combination of leaf and root demonstrates the greatest efficacy of the 1:1 combinations, having five additive profiles and one synergistic profile for the ten test pathogens studied. The benefit of combining leaf and root plant parts of C. gratissimus var. subgratissimus is also apparent in the isobologram studies where a predominant trend is seen where most ratios show an additive profile, or close to an 119 additive profile, for most pathogens studied. The pathogens showing the highest synergistic profiles (E. faecalis, C. albicans, C. neoformans and selectively for B. cereus) correlate somewhat with diseases for which the plant is traditionally used for. Studies such as this give credibility to the use of plant part mixtures for therapeutic treatment. There are many screening studies that focus on the antimicrobial efficacies of different plant parts. Earlier studies (Vlietinck et al., 1995) have examined antimicrobial efficacies of the fruit, leaf, root, barks, seeds and whole plant of Rwandese medicinal plants. More recent studies (Eldeen et al., 2005) have examined the leaf, root and bark of ten South African trees for their antimicrobial potential. These studies and many others (Caceres et al., 1991; Rabe and van Staden, 1997; Lin et al., 1999; Drewes et al., 2005) have however, investigated the different parts of the plant as separate entities. A recent study by Lewu et al. (2006) comparatively evaluated the antimicrobial efficacy of the roots and leaf material of Pelargonium sidoides. Even though the ethnobotanical use of many plants incorporates mixes of the different plant parts, there has been very little scientific evidence to support the interactive efficacies of the different parts of the plant when combined. Furthermore, studies on Croton species have narrowly focused on the investigation of specific compounds from the species, such as the isolation of diterpines from the leaves of Croton zambesicus (Block et al., 2004), the ethnopharmacological study of the chemical constituents within the roots of Croton cajucara (Maciel et al., 2000) and investigations on the chemical composition of the bark oil of Croton stellulifer (Martins et al., 2000). The outcome of this study was to examine the interaction of plant parts rather than taking a reductionist approach to assess the antimicrobial activity of a plant. 7.7 General conclusions ? The major constituents of the essential oil of Croton gratissimus var. subgratissimus are ?- phellandrene (20.7%), germacrene D (8.6%) and 1,8-cineole (8.3%), representing 37.6% accumulatively. ? The extracts of leaf, bark and root when investigated independently indicated better antimicrobial activity (0.3-6.0 mg/mL) than the hydrodistilled leaf essential oil (4.0-32 mg/mL). 120 ? The MIC and FIC results indicated variable efficacies for the plant combinations with the highest efficacy noted for Cryptococcus neoformans in the leaf and root combination (MIC value 0.4 mg/mL and FIC of 0.4). ? The pathogens showing the highest synergistic profiles in the isobole method of determining synergy were Bacillus cereus, Enterococcus faecalis, Candida albicans and Cryptococcus neoformans which correlate with the ethnobotanical uses of the plant to treat respiratory disorders (Enterococcus faecalis and Cryptococcus neoformans) and abdominal complaints (Bacillus cereus). 121 Chapter 8 A seasonal and geographical study of Heteropyxis natalensis Harv. essential oil and effect of stereochemistry on antimicrobial activity. 8.1 Introduction The family Heteropyxidaceae has only three representatives in southern Africa; Heteropyxis canescens, Heteropyxis dehniae and Heteropyxis natalensis. The most noteworthy is Heteropyxis natalensis Harv., a tree known for its therapeutic properties (Watt and Breyer-Brandwijk, 1962; van Wyk et al., 1997; van Wyk and Gericke, 2000). A literature review on this species identified two seasonal variation studies that had been carried out (Weyerstahl et al., 1992; Chagonda et al., 2000), however, the major essential oil constituents established in these two investigations varied, even though plant sampling was undertaken in the same region. No antimicrobial investigation was undertaken in these two studies. Another study (Gundidza et al., 1993) reported the antimicrobial properties and phytoconstituents of Zimbabwean H. natalensis species, but studies were restricted to disc diffusion methods only. It is known that phytochemicals are often governed by external factors such as soil quality and climate, and biological activity is directly related to the chemical composition of a plant, which may be subject to quantitative and qualitative variation. Seasonal fluctuation of phytoconstituents may or may not impact on the antimicrobial activity. While H. natalensis has been cultivated in Zimbabwe for further commercial development (Chagonda et al., 2000), this study on the seasonal variation of South African H. natalensis species was undertaken to determine if there was any quantitative and / or qualitative variation in essential oil composition as noted in the Zimbabwean oils. The purpose of this study was to collect, hydrodistil and analyze the essential oils obtained from three individual trees of H. natalensis and to characterize the essential oils thereof on a monthly basis. The geographical variation within South Africa was also investigated to determine if oil composition and microbiological activity is dependent on the origin of the plant material. 122 8.2 Botanical description Heteropyxis natalensis is a deciduous tree growing to approximately ten metres in height. It has strongly aromatic foliage, a branched trunk and a pale-green bark, which has a distinctive mottled appearance (Figure 8.1). The leaves are small, simple, narrow, oblong / elliptic and are arranged spirally. They are pale-green on the underside with a shiny dark-green upper. The leaves may appear red-tinged when the tree is young (van Wyk et al., 1997). The leaves have a strong lavender odour, which is emitted in the summer; hence the vernacular name ?lavender tree?. The flowers (Figure 8.2) are yellowish-green, inconspicuous and are in bloom between December and March. They are usually followed by small, oval, dry capsules (fruit), which appear between March and May. The genus name is derived from Greek ?Hetero?- meaning different, and a Latin ?-pyxis? meaning ?the container with the lid?. The latter refers to the fruit (capsule), which is sometimes called the pyxidium (Venter and Venter, 2002). Figure 8.1 Figure 8.2 Heteropyxis natalensis (Figure 8.1) and leaves with small inconspicuous flowers (Figure 8.2). 8.3 Distribution Heteropyxis natalensis grows naturally in the north-eastern part of South Africa (Figure 8.3), on the coastal and inland regions extending from the tree?s place of origin, which is KwaZulu-Natal, through to Mpumalanga, northern Gauteng regions and Northern Province. 123 Figure 8.3 The geographical distribution of H. natalensis in southern Africa (SANBI). 8.4 Medicinal uses Heteropyxis natalensis is used traditionally to treat respiratory disorders, as a decongestant and as an anti-infectant (van Wyk et al., 1997). The leaves are mainly used for medicinal purposes and prepared as a tea (van Wyk and Gericke, 2000). A deconcoction of the roots is prepared and the steam inhaled. Both the Venda and Zulu communities have reported on the medicinal values of H. natalensis (Watt and Breyer-Brandwijk, 1962). 8.5 Methods 8.5.1 Chemical aspects Plant collection and distillation of essential oils: For the seasonal variation study, the aerial parts from three individual trees (A, B and C) were collected from the Johannesburg Botanical Garden (JHB BG) on a monthly basis, for the year 2004. Due to the semi-deciduous nature of the tree no plant samples were obtained from for June to September and only one tree produced sufficient foliage in October. 124 Sample Voucher Material distilled (g) Essential oil yield (% w/w) Average temperature (?C)* A, January SVV897 1 042.5 0.20 B, January SVV898 419.0 0.10 C, January SVV899 257.0 0.10 20.5 A, February SVV922 948.4 0.20 B, February SVV923 400.2 0.10 C, February SVV924 405.6 0.20 19.5 A, March SVV954 510.0 0.20 B, March SVV955 587.8 0.10 C, March SVV956 485.1 0.10 19.0 A, April SVV972 460.6 0.20 B, April SVV973 490.0 0.10 C, April SVV974 460.6 0.10 16.0 A, May SVV993 336.7 0.10 B, May SVV994 158.8 0.10 C, May SVV995 455.2 0.10 13.0 A, October SVV1069 450.0 0.10 17.5 A, November SVV1071 780.0 0.10 B, November SVV1072 455.0 0.04 C, November SVV1073 816.0 0.20 18.5 A, December SVV1074 1 296.0 0.10 B, December SVV1075 811.0 0.10 C, December SVV1076 840.0 0.10 19.5 For the geographical variation study, plants were collected from Cullinan and Verena (Gauteng); Nelspruit (Mpumalanga); Lagalametse and Waterberg (Northern Province) and Balakane (Swaziland). With the ex```ception of the Balakane sample (collected in November), all collections were undertaken in the month of February to avoid any impact on possible seasonal variations that may occur. Voucher specimens for seasonal (Table 8.1) and geographical (Table 8.2) sampling were deposited in the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plants were distilled as described in Chapter 2.5.1. * Average temperature at sampling is calculated as the mean of the maximum and minimum temperature for that month (http://www.weathersa.co.za/FcastProducts/LongRange/ViewSeasonEyeBall.jsp). Table 8.1 Collection data for H. natalensis harvested in the Johannesburg Botanical Garden. 125 Table 8.2 Collection data for H. natalensis sourced from various localities. Locality Voucher Material distilled (g) Essential oil yield (% w/w) Cullinan AMV771 700.0 0.3 Verena PMB703 559.0 0.2 Nelspruit ADC721 421.5 0.2 Lagalametse ADCAV181 450.0 0.1 Waterberg PMB738 499.2 0.1 Balakane ADCAV140 380.0 0.1 Gas chromatography (GC) and GC-MS data: The GC analysis was performed on a Shimadzu 17A gas chromatograph using the parameters as described in Chapter 2.5.1. Oil samples were quantitatively analyzed with GC-MS using the Hewlett-Packard 1800A GCD system (Chapter 2.5.1.). Using quantitative data obtained from the GC-MS analysis, a cluster analysis was performed on the essential oils from three seasonal samples (February, plants A, B and C) and a further six H. natalensis samples from different localities. A dendrogram (Figure 8.8) presents the data using the NTSYS-pc software package (Version 2.0) analysed by a phenetic program (Rohlf, 1998). Correlation of similarity was selected for the basis of analysis and the unweighted pair-group method with arithmetic average (UPGMA) was used for cluster definition. Thin layer chromatography (TLC): TLC plates were developed for the essential oil samples obtained from the seasonal variation study to establish if there was obvious seasonal variation in the essential oil components. Samples were prepared by placing 2 ?l of diluted (1:7 hexane, Merck) H. natalensis essential oil on silica gel TLC plates (Alugram ? Sil G/UV254, 0.2 mm). The plates were developed with ethyl acetate and toluene (Merck) at a ratio of 7:93 and dried at ambient temperature. Plates were observed under ultraviolet light (254 nm) and thereafter sprayed with an anisaldehyde-acetic-sulphuric acid reagent. This was prepared by mixing 0.5 mL anisaldehyde (Fluka) with 10 mL glacial acetic acid (Merck), followed by 85 mL methanol (Merck). The TLC plates were subsequently placed in the oven at 110 ?C for five minutes to enable visualization of the separated compounds (Wagner and Bladt, 1995). 126 8.5.2 Antimicrobial aspects Culture, media preparation and assays were undertaken according to the NCCLS (2003) guidelines (Chapter 2.5.2.). Quantitative MIC assays (Carson et al., 1995; Eloff, 1998a; NCCLS, 2003) were performed on the hydrodistilled oil to determine if: ? The antimicrobial efficacy supports the traditional use. ? Results correlate with that found in literature (Gundidza et al., 1993). ? Seasonal variation exists. ? Geographical variation is evident between distant populations. ? Structure-activity relationships between major constituents and biological activity were evident. As the traditional use of H. natalensis is for respiratory tract infections, eight microbial test organisms were selected for the seasonal variation study based on their respiratory pathogenesis. A clinical Gram-negative respiratory pathogen, Moraxella catarrhalis was included in the study. This opportunistic respiratory pathogen has severe implications in upper and lower respiratory tract infections, bronchitis, pneumonia and may even be associated with sinusitis and otitis media (Laurans and Orfila, 1991; Verduin et al., 2002). The study on the Zimbabwean H. natalensis species undertaken by Gundidza et al. (1993) has shown antimicrobial sensitivity towards this pathogen and was included in this study to determine if similar sensitivity patterns exist for the South African H. natalensis samples. In addition to the H. natalensis samples examined for seasonal variation, a further six H. natalensis samples were selected from different regions to determine any geographical variation within the MIC data. Antimicrobial MIC assays were limited to five pathogens (Table 8.6) due to limited oil acquisition. Pathogen selection was based on the traditional use of H. natalensis. All MIC assays were undertaken at least in triplicate with the seasonal mean and standard deviation presented in Table 8.5. The geographical mean with standard deviation is presented in Table 8.6. The geographical MIC data (Table 8.6) was grouped according to sensitivity, with poor susceptibility classified as 16 - >16 mg/mL, moderate susceptibility with MIC values 8 - <16 mg/mL, moderate to good susceptibility with MIC values 4 - <8 mg/mL and good susceptibility with MIC values <4 mg/mL. A cluster analysis was performed from this data using the same parameters as described with the GS-MS cluster analysis and presented in Figure 8.10. 127 Interaction between limonene and 1,8-cineole: The GC-MS data confirmed that limonene and 1,8-cineole are predominantly present as the two major compounds in H. natalensis. The enantiomeric configuration of limonene with 1,8-cineole was investigated for antimicrobial activity according to the MIC microtitre plate method as described previously. The compounds 1,8-cineole at 98.0% purity (Lot 1054365) was obtained from Sigma-Aldrich. S(-)-Limonene at 99.0% purity (Lot 054076) was obtained from Fluka and (R)-(+)-limonene at 97.0% purity (Lot 301Tl-101) was obtained from Sigma-Aldrich. The compounds were prepared at starting concentrations 128 mg/mL and assayed with the H. natalensis essential oil. Assays were undertaken in triplicate and the mean documented in Table 8.7. Table 8.3 Raw data for the isobologram construction of S. aureus. Concentration in mg/mL MIC values Ratio values X Y X Y X Y (+/- limonene) (1,8-cineole) (+/- limonene) (1,8-cineole) (+/- limonene) (1,8-cineole) 32.0 0.0 4.00 0.00 1.000 0.000 28.8 3.2 7.20 0.80 1.800 0.100 25.6 6.4 4.80 1.20 1.200 0.150 22.4 9.6 4.20 1.80 1.050 0.225 19.2 12.8 3.60 2.40 0.900 0.300 16.0* 16.0* 3.00* 3.00* 0.750* 0.375* 12.8 19.2 2.40 3.60 0.600 0.450 9.6 22.4 2.40 5.60 0.600 0.700 6.4 25.6 1.60 6.40 0.400 0.800 3.2 28.8 0.80 7.20 0.200 0.900 0.0 32.0 0.00 8.00 0.000 1.000 *The 1:1 ratio as determined from the raw data is presented as a square on the isobolograms. The optically active monoterpene, limonene was used to determine if microbes respond differently when exposed to the (+), (-) and / or racemic mixture of limonene in combination with 1,8-cineole. The isobologram ratio method as described in Chapter 7 (Berenbaum, 1978; Williamson, 2001) was used and nine ratios i.e. 90:0%; 80:20%; 70:30%; 60:40%; 50:50%; 40:60%; 30:70%; 20:80% and 10:90% of each enantiomeric form and the racemate were thoroughly mixed with 1,8-cineole. The MIC values (mg/mL) were determined for each concentration independently and for the nine ratios, and raw MIC data can be noted in Table A2, Appendix A. The MIC of all relevant ratios (values within the 0-1.25 range) were plotted as points on an isobologram relative to the MIC of limonene and 1,8-cineole independently (shown as a straight line), allowing for a figurative representation of the interaction of the various combinations. Investigations were undertaken in duplicate on the following reference test 128 organisms: S. aureus ATCC 12600, (Gram-positive); P. aeruginosa ATCC 9027, (Gram- negative) and a yeast C. neoformans ATCC 90112. Conventional antimicrobials (not shown on the isobologram) were included in all repetitions to ensure susceptibility of the test organism. The raw data for the isobologram construction for one test organism (S. aureus) is given in Table 8.3. 8.6 Results and discussion 8.6.1 Essential oil chemistry The GC-MS data identified approximately 173 compounds in the H. natalensis essential oil representing 72.0?97.8% of the total composition (Table 8.4). Major compounds: Two major compounds (limonene and 1,8-cineole) identified by GC were predominantly present in all seasonal samples. Figure 8.7 shows the monthly variation of the different concentrations of the main compounds limonene and 1,8-cineole in relation to the sum of the remainder of the constituents as determined by GC. Three individual trees (samples February A, B and C) were selected for further GC-MS analysis. Major compounds confirmed were 1,8-cineole (23.9-41.2%) and limonene (18.1-25.4%). The H. natalensis (geographical study) samples identified limonene and 1,8-cineole as major constituents in four localities. Linalool was found as a major constituent in the samples from Nelspruit (16.7%) and Lagalametse (11.4%). (Z)-3-Hexenyl nonanoate (16.0%) was only present in the Lagalametse sample. ?-Pinene at 25.2% was identified as the major constituent in the Balakane sample. (E)- Nerolidol was present as a major constituent in the samples JHB BG, sample A (12.5%) and C (16.3%) and only present in negligible quantities in the other geographical samples. (E)- Nerolidol, limonene and ?-pinene were not present in the Lagalametse sample. All major compounds detected in the various H. natalensis samples are given in Figure 8.4. 129 Table 8.4 Essential oil composition as determined by GC-MS for H. natalensis samples (seasonal and geographical variation study). Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane 1032 ?-pinene 1.7 3.2 3.1 3.0 1.9 2.7 tr 3.1 1.9 1035 ?-thujene 0.4 1.2 1.1 0.6 0.6 1.0 - 0.7 tr 1076 camphene - - tr 0.1 - - - - 0.1 1118 ?-pinene 1.1 2.0 1.9 7.6 3.2 6.2 - 2.5 25.2 1132 sabinene - 0.1 0.4 0.1 0.1 0.6 - 0.1 tr 1146 2-methylbutyl acetate - - - - - - 0.1 - - 1174 myrcene 0.5 1.8 2.0 1.6 1.2 3.2 - 1.4 5.1 1183 p-mentha-1,7(8)-diene (=Pseudolimonene) 5.2 7.1 4.6 6.0 3.7 3.0 - 6.5 tr 1188 ?-terpinene - - - 0.1 - - - - - 1203 limonene 18.1 25.4 16.5 22.8 19.2 15.0 - 23.6 1.8 1213 1,8-cineole 33.6 41.2 23.9 23.5 29.4 21.7 0.2 39.2 - 1246 (Z)-?-ocimene - 0.3 0.1 0.4 tr 0.2 - - 0.4 1255 ?-terpinene - 0.2 - 1.1 - - - - tr 1265 (E)-?-ocimene - 0.8 - 1.5 tr 0.7 - - 2.5 1274 2-heptyl acetate - - - 0.1 0.1 0.1 0.2 - 0.2 1275 2-methylbutyl butyrate 0.1 tr 0.3 0.3 0.1 0.1 0.4 tr 1280 p-cymene 3.5 2.5 3.3 3.7 4.3 4.2 0.1 5.8 0.4 1290 terpinolene - 0.1 - 0.2 - - - - 0.1 1303 amyl isovalerate - - - - - - - 0.1 - 1345 4-pentenyl butyrate? 0.4 0.5 1.2 1.2 tr 0.2 0.4 0.3 0.2 1391 (Z)-3-hexenol - - - - - - - 0.4 tr 1395 2-nonanone - - 0.3 0.7 0.3 0.6 2.4 - 1.1 1398 3-methyl-2-butenyl butyrate? 0.3 0.3 1.1 0.6 0.4 0.1 0.4 0.2 0.1 1413 rose furan - - 0.2 - - 0.1 - - 0.1 1429 perillen - - - - tr 0.1 - - tr 1450 trans-linalool oxide (Furanoid) - - - tr tr 0.4 8.8 0.1 tr 1458 cis-1,2-limonene epoxide 0.3 - 0.1 - tr 0.1 - - - 1466 ?-cubebene - - - - - - - - 0.1 1468 trans-1,2-limonene epoxide 0.2 - - - - - - - - 1469 3-methyl butyl hexanoate (=Isoamyl hexanoate) - - - - - 0.1 - - tr 130 Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane 1470 2-nonyl acetate - - - - - - 0.3 - - 1471 (Z)-3-hexenyl butyrate - - 0.2 0.1 - - - 0.1 - 1473 (E)-2-hexenyl butyrate - tr - - 0.2 - - - - 1474 trans-sabinene hydrate - - - - - 0.1 - - - 1476 (Z)-?-ocimene epoxide - - tr - - tr - - tr 1478 cis-linalool oxide (Furanoid) - - tr tr - 0.3 8.0 0.1 tr 1482 fenchyl acetate 0.1 tr tr tr - - - - tr 1483 octyl acetate - - - - - tr - - tr 1497 ?-copaene - - tr 0.1 0.1 - - tr 0.1 1498 (E)-?-ocimene epoxide - - 0.1 - 0.1 0.1 - - 0.1 1506 decanal - - - - - - 0.1 - - 1521 2-nonanol - - - tr - - 0.4 - - 1541 benzaldehyde - tr tr - tr tr - 0.1 - 1553 linalool 0.1 1.2 0.5 8.0 0.2 16.7 11.4 2.1 1.0 1559 8,9-limonene epoxide-I tr - - - - - - - - 1565 linalyl acetate - tr - 0.1 - 0.1 0.1 tr - 1571 trans-p-menth-2-en-1-ol 0.1 tr 0.1 tr 0.1 0.1 - - - 1586 pinocarvone - - tr - - 0.1 - - 0.1 1589 ?-ylangene - - - - - - - - 0.1 1590 3-methyl-2-butenyl hexanoate? - - tr - - - 0.1 - tr 1591 fenchyl alcohol - - - - - tr - - 0.2 1597 bornyl acetate - - - - - - - - 0.2 1600 ?-elemene - - 0.1 - - 0.1 - - 0.1 1601 ?-copaene - - tr - 0.1 - - - 0.2 1602 6-methyl-3,5-heptadien-2-one - - 0.2 - 0.1 - - - - 1605 2-undecanone - - - 0.1 - 0.1 0.8 - - 1611 terpinen-4-ol 1.4 1.6 1.4 0.7 0.8 2.1 - 2.3 0.1 1612 ?-caryophyllene - 0.1 0.5 0.7 0.1 0.1 - tr 1.9 1616 hotrienol - - - - - - 0.3 - - 1625 4,4-dimethyl but-2-enolide - - - - - - 0.1 - - 1628 aromadendrene - - 0.1 - tr - - - - 1630 terpinen-4-yl acetate (=4-Terpinenyl - - - - tr tr - - - 131 Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane acetate) 1638 cis-p-menth-2-en-1-ol - - - tr - 0.1 - - - 1639 trans-p-mentha-2,8-dien-1-ol 0.1 tr 0.1 - 0.2 - - - 1641 cis-?-terpineol - tr tr - tr 0.1 - - - 1648 myrtenal - - - - - 0.1 - - 0.3 1661 alloaromadendrene - - - - - - - - 0.3 1664 nonanol - - - - - - 0.2 - - 1670 trans-pinocarveol 0.1 0.1 tr 0.2 0.1 - - 0.4 1671 acetophenone - tr - - - - - 0.1 - 1678 cis-p-mentha-2,8-dien-1-ol - - 0.1 - 0.1 - - - - 1681 4-methyl-4-vinyl butyrolactone - - - - - - 5.3 - - 1682 ?-terpineol 0.5 0.4 0.2 0.1 0.3 0.2 - 0.4 - 1687 ?-humulene 0.1 0.2 0.8 0.3 0.3 0.4 - 0.1 0.5 1688 selina-4,11-diene(=4,11-Eudesmadiene) - - - - - - - - 0.2 1700 p-mentha-1,8-dien-4-ol (=Limonen-4-ol) - tr tr - 0.1 - - 0.1 - 1704 myrtenyl acetate - - - - - tr - - - 1705 ?-muurolene - - - - - - - - 0.3 1706 ?-terpineol 3.1 3.2 2.3 1.0 3.5 3.0 - 2.8 2.5 1707 ?-selinene - - - - - - - - 0.1 1709 ?-terpinyl acetate 0.1 tr - 0.1 0.1 0.1 - 0.1 - 1718 4,6-guaiadiene (=?-Guaiene) - - - - - - - - tr 1719 borneol - - - - - - - - 0.2 1723 cis-1,2-epoxy-terpin-4-ol 0.2 tr 0.1 - 0.1 0.1 - tr tr 1733 neryl acetate - - - - 0.1 0.1 - - - 1739 ?-muurolene - 0.1 0.2 - 0.3 0.1 - - 1.0 1740 valencene - - 0.1 - 0.1 - - - 0.2 1742 ?-selinene - - - 0.1 - 0.2 - - - 1744 ?-selinene - tr 0.1 0.1 0.1 0.1 - - 0.6 1750 cis-linalool oxide (Pyranoid) - - - - - - 1.9 - - 1751 carvone 0.5 0.1 0.2 - 0.4 tr - 0.2 - 1758 cis-piperitol - tr tr - tr tr - - - 1759 (E,E)-?-farnesene - - - - - - - - tr 132 Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane 1765 geranyl acetate - tr - 0.1 0.2 0.1 0.2 0.1 0.1 1770 trans-linalool oxide (Pyranoid) - - - - - tr 2.0 - - 1773 ?-cadinene 0.1 0.1 0.1 0.1 0.1 0.1 - 0.1 1.2 1776 ?-cadinene - 0.1 0.2 0.1 0.4 0.1 - - 0.6 1797 p-methyl acetophenone - - - - - tr - - - 1798 methyl salicylate - - - tr - tr - - - 1804 myrtenol 0.1 - - - 0.2 0.1 - - 0.3 1807 ?-cadinene - - - tr tr tr - - - 1808 nerol - - - - - 0.1 - - - 1811 trans-p-mentha-1(7),8-dien-2-ol (= trans-2-hydroxy pseudolimonene) - tr - - - - - - - 1826 ?-heptalactone - - - - - - 1.1 - - 1830 2,6-dimethyl-3(E),5(E),7-octatriene-2-ol - tr - tr - 0.1 - - - 1845 trans-carveol 0.5 0.1 0.3 tr 0.5 0.1 - 0.2 - 1853 cis-calamenene 0.1 tr 0.1 - 0.2 0.1 - 0.1 0.6 1857 geraniol - - - 0.1 - 0.3 - - - 1864 p-cymen-8-ol 0.2 tr 0.1 - 0.2 0.2 - 0.1 0.1 1875 trans-2-hydroxy-1,8-cineole 0.1 - - - - - - - - 1882 cis-carveol 0.2 0.1 - - tr - 0.1 - 1883 benzyl butanoate 0.1 0.1 0.3 0.1 0.2 tr 0.2 0.1 - 1896 cis-p-mentha-1(7),8-dien-2-ol (= cis-2-hydroxy pseudolimonene) - tr - - - - - tr - 1901 geranyl butyrate - tr 0.1 0.1 0.2 - tr - 1941 ?-calacorene - tr tr tr - tr - - 0.1 1949 (Z)-3-hexenyl nonanoate - - - - - - 16.0 - - 1969 cis-jasmone - - - 0.1 0.1 tr - tr 0.1 1989 2,6,10-trimethyl-7,10-epoxy-2,11-dodecadien-6-ol (=Nerolidol oxide) 0.4 - - - - - - - - 2001 isocaryophyllene oxide - - 0.2 - 0.5 0.2 - tr 0.1 2008 caryophyllene oxide 1.3 1.2 2.2 4.2 7.2 1.5 4.6 1.3 4.5 2016 2,6,10-trimethyl-7,10-epoxy-2,11-dodecadien-6-ol isomer (=Nerolidol oxide 0.6 - - - - - - - - 133 Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane isomer) 2029 perilla alcohol - tr - - 0.1 0.1 - tr 0.1 2037 salvial-4(14)-en-1-one - - - - 0.1 - - tr 0.2 2045 humulene epoxide-I 0.1 0.1 0.1 0.1 0.1 0.1 - tr 0.1 2050 (E)-nerolidol 12.5 tr 16.3 0.1 0.2 0.8 - tr tr 2051 gleenol - - - - - - - tr tr 2071 humulene epoxide-II 1.2 0.6 1.6 1.0 0.9 0.6 - 0.2 0.7 2079 (E)-?-ionone 0.1 - - - - - - - - 2080 cubenol - tr - - 0.1 tr - tr 0.5 2088 1-epi-cubenol 0.1 0.1 0.1 0.1 0.2 0.1 - 0.1 1.0 2096 elemol - - - - - 0.4 - - - 2098 globulol 0.3 0.2 0.6 - 0.8 - 0.3 - - 2100 heneicosane - - - 0.1 - - - - - 2104 viridiflorol - - tr 0.1 - tr - 0.1 0.8 2106 guaiyl acetate - - - - - - - - 0.2 2123 methyl-4-(4?-methyl-3?-pentenyl)-3- cyclohexenyl ketone? - - - - - 0.1 - - - 2126 3,7-dimethyl-1,7-octadien-3,6-diol - - - - - - 3.4 - - 2144 rosifoliol - - - - 0.1 tr - - 0.3 2146 spathulenol 0.4 0.6 0.7 0.1 0.9 0.5 1.0 0.1 0.8 2148 (Z)-3-hexen-1-yl benzoate - - - - 0.1 - - - - 2153 neointermedeol - tr - 0.1 - 0.1 - tr 0.6 2164 6-epi-cubenol tr tr 0.1 0.1 0.3 0.1 - tr 0.1 2174 cinnamyl acetate tr - - - - - - - - 2184 cis-p-menth-3-en-1,2-diol - 0.1 - - 0.5 - - 0.1 - 2185 T-cadinol 0.3 0.1 0.3 0.3 0.8 0.1 - 0.2 1.4 2187 ?-eudesmol - - - 0.2 - 0.9 - - 5.2 2196 eremoligenol - - - 0.1 - 0.1 - - 1.7 2206 ?-guaiol - - - - - - - - 0.5 2209 T-muurolol 0.1 tr 0.1 tr 0.3 tr - 0.1 0.2 2211 clovenol - tr - 0.1 0.1 - - tr - 2214 torreyol 0.1 tr 0.1 tr 0.2 0.1 - tr 0.6 134 Gauteng Mpumalanga Northern Province Swaziland JHB BG (Feb) RRI* Compound name A B C Cullinan Verena Nelspruit Lagalametse Waterberg Balakane 2247 trans-?-bergamotol - tr 0.1 - - - - - - 2250 ?-eudesmol - - - 0.6 - 1.1 - 7.5 2255 ?-cadinol 0.3 0.1 0.3 - 0.8 - -- 0.2 tr 2257 ?-eudesmol - - - 0.8 - 1.4 - - 3.1 2272 alismol - - - - 0.4 0.2 - 0.1 0.7 2273 selin-11-en-4?-ol - tr 0.2 0.2 0.3 0.1 - - 0.2 2316 caryophylla-2(12),6(13)-dien-5?-ol (=Caryophylladienol I) - - 0.1 - - - - - - 2324 caryophylla-2(12),6(13)-dien-5?-ol (=Caryophylladienol II) - 0.1 - 0.4 0.3 - - 0.1 0.4 2328 (E,E)-10,11-epoxyfarnesyl acetate - - 0.4 - - - - - - 3,7,11-trimethyl-10,11-epoxy-2,6- dodecadien-1-yl acetate ? (=10,11- Epoxyfarnesyl acetate) 1.0 - - - - - - - - 2350 carvone hydrate? (=Aralone) 0.2 - - - - - - - - 2352 (2E,6E)-farnesol - 0.1 0.3 0.3 0.8 - 0.1 - 2375 eudesma-4(15),7-dien-1?-ol - tr 0.1 0.1 0.2 0.1 - 0.1 0.5 2385 10-hydroxy calamenene - - 0.3 - 0.4 0.1 - 0.1 - 2389 caryophylla-2(12),6-dien-5?-ol (=Caryophyllenol I) - 0.1 tr 0.4 0.4 - - 0.1 0.2 2392 caryophylla-2(12),6-dien-5?-ol (=Caryophyllenol II) 0.2 0.1 0.2 0.4 0.5 - 0.2 0.1 0.1 2438 kaur-16-ene - - - - 0.2 - - - - 2441 3,7-dimethyloct-1-en-3,6,7-triol ? - - - - - - 0.5 - - 2501 phytol acetate? - - - - - - - - 1.1 2503 agglomerone ? - - - - - - - 0.4 - 2622 phytol - 0.1 0.4 - 0.6 0.2 - - 2.7 TOTAL 92.3 97.8 93.6 97.6 93.3 95.3 72.0 97.2 89.1 * RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = Trace (< 0.1%), ? = Tentative identification from Wiley, MassFinder, Adams libraries. 135 O OH OH E 1,8-cineole limonene linalool (E)-nerolidol O HH O (Z)-3-hexenyl nonanoate ?-pinene Seasonal variation: The essential oil yield for the seasonal study on H. natalensis varied (0.04- 0.2%) with higher yields in the late summer months. Preliminary TLC comparisons were undertaken to investigate if variation in chemical profiles existed between seasonal samples. Similar profiles were obtained for all samples from the three individual trees studied with a representative profile given in Figure 8.5 and Figure 8.6. Confirmation of chromatograms depicting 1,8-cineole and limonene as major constituents were made by comparison with the commercially obtained standards (not shown in Figure 8.5 and Figure 8.6). The monthly fluctuation of 1,8-cineole and limonene levels in the essential oil showed tree-to-tree variation within a single site as well as variability on a monthly basis (Figure 8.7). The highest tree- to-tree variability for limonene is noted in November for plant B, which contains 4% as opposed to plant A with 26%. Other variations within the individual plant sampling show monthly fluctuations in limonene content, as is the case in May with Plant B having 7% and plant A having 29%. Both these months fall on either side of the winter months and seasonal changes could possibly account for this variation. The total plant to plant variation within a population for limonene was between 4- 29% (depicted in blue, Figure 8.7). Figure 8.4 Chemical structures for major compounds identified in the essential oil of various H. natalensis samples. 136 Jan. Feb. Mar. April May Oct. Nov. Dec. A B C December Figure 8.5 Figure 8.6 0 10 20 30 40 50 60 70 80 90 100 JANUARY FEBRUARY MARCH APRIL MAY OCTOBER NOVEMBER DECEMBER % Co n ce n tr at io n Figure 8.7 The monthly variation of limonene and 1,8-cineole recorded for three individual H. natalensis plants at the same locality. Others 1,8-Cineole Limonene The TLC chromatograms indicating similar profiles for plant A, sampled monthly (Figure 8.5) and (Figure 8.6) representing similar profiles for three individual plants within a population. 137 Fluctuation of 1,8-cineole within plant populations showed the highest variation for January (plant A: 60% and plant B: 79%) and December (plant A: 47% and plant B: 28%). Both months fall within the mid summer season. In general plant to plant variation within a population for 1,8-cineole was between 13-20% (depicted in red, Figure 8.7). The concentration of limonene in plant A has shown an 11% seasonal variation throughout the period of analysis. 1,8-Cineole, indicated higher concentrations in January, February, March, and April (55-65%), with lower concentrations obtained for May, October, November and December, (38-47%). The overall seasonal variation for 1,8-cineole in plant A was 18%. For plant B (Figure 8.7), the concentration of limonene was low in January (11%), increased in February, March, and April (21-23%), and decreased again in May, November and December (4- 15%). 1,8-Cineole had higher concentrations in January and February (65-79%), with concentrations decreasing in March and April (38-44%), increased in May (57%), and again decreased in November and December (28-29%). The seasonal variation for plant B is 19% for limonene and 51% for 1,8-cineole. For plant C (Figure 8.7), the concentrations of limonene were lower in January (7%) and February (18%) than in the later months (21-26%), with a seasonal variation of 19%. 1,8-Cineole was found in higher concentrations in January (78%), which decreased in the later months (46-52%), with a seasonal fluctuation of 32%. The two major compounds (limonene and 1,8-cineole) collectively present in the essential oils of H. natalensis made up between 32% (plant B, November) to 90% (plant B, January) of the total composition. One would assume that three plants, growing in close proximity within a population in cultivated conditions (Johannesburg Botanical Gardens) where irrigation is regular, would yield a relatively standard composition. This was not the case in this study where fluctuations of the sum of the two major constituents show up to a 58% variation. Seasonal variation studies on H. natalensis by Weyerstahl et al. (1992), on the leaves of plants collected in Zimbabwe had a lower oil yield (0.9%) to that found in this study. More interestingly, however, was the difference in composition. The major compounds found in the Zimbabwean oils in 138 the summer months were (E)-??ocimene (29%) and linalool (26%). In the winter months limonene (21%) and 1,8-cineole (40%) were dominant. The GC-MS analysis of the three February samples (A, B and C) from JHB BG indicates that these compounds were found in negligible quantities. (E)- ???cimene was absent in plant A, with minor constituent composition 0.3% and 0.1% in plants B and C respectively. Linalool, although present as major constituents in other geographical regions, were only present as a minor constituent here representing 0.1%, 1.2% and 0.5% for plants A, B and C respectively. This study also noted that limonene and 1,8-cineole, both major constituents, were consistently predominant throughout the winter and summer months. In another seasonal study, also undertaken in Zimbabwe (Chagonda et al., 2000), linalool (26.5%) and 1,8-cineole (37.2%) were found to be the major constituents in winter for both the wild and cultivated plants. In the summer months, linalool (26.5%), limonene (13.3%) and 1,8-cineole (24.7%) were seen as major components. For this study, limonene concentrations for the three separate plants within the month of February varied between 16.5-25.4%. 1,8-Cineole composition varied between 23.9-41.2%. These results correlate more closely with the study undertaken by Chagonda et al. (2000), than when comparatively examined with the study done by Weyerstahl et al. (1992). The average climate for Harare, Zimbabwe (information obtained from an eight year mean, http://www.climate- zone.com/climate/zimbabwe/celsius/harare.htm), indicated very little temperature variation when compared to the seasonal sampling in this study. The overall variation for the Zimbabwean oils was between 2-5 ?C with more variation in the winter months. For the South African oils studied here, a temperature variation of 1-7.5% was noted with the greatest difference found in May, also being a cooler month of the year. Thus, one may deduce that it is unlikely that essential oil variation for H. natalensis is determined by climatic conditions. Geographical variation: The chemical diversity noted between the Zimbabwean H. natalensis essential oils and that noted from this study brought about the postulation that geographical distribution might play a role in composition variability. This geographical variability is not novel and can be noted in studies on A. afra (Viljoen et al., 2006) and Eriocephalus species (Njenga et al., 2005) where the oil composition varied quantitatively and qualitatively within and between natural populations and showed little or no correlation to geographical distribution. A study investigating the geographical variability within South African H. natalensis populations was hence undertaken. The geographical oil yield of H. natalensis from six different localities showed little variation (0.1- 139 0.2%) with only the sample from Cullinan having a somewhat higher yield (0.3%). The oil compositions are given in Table 8.4. The percentage composition of the essential oil from H. natalensis representing seven different localities was studied and a dendrogram (Figure 8.8) was generated using the 173 different chemical compounds identified by GC-MS (Table 8.4). Four main types were identified with two groups closely related with a similarity coefficient of 0.8-1. The H. natalensis sample from Lagalametse is distinctly different from all other samples, where the characteristic major constituents limonene and 1,8-cineole are absent. (Z)-3-Hexenyl nonanoate predominantly present (16.0%) was absent from all other H. natalensis samples. Also uniquely different was the H. natalensis sample from Balakane having ?-pinene (25.2%) as a major constituent. The samples from Gauteng are similarly clustered all having limonene and 1,8-cineole as major constituents. These major compounds were also present in the Nelspruit and Waterberg samples. In addition to these major compounds, linalool was present as a major constituent in the Nelspruit sample, distinguishing it from Gauteng and Waterberg. Figure 8.8 Dendrogram constructed from the essential oil data matrix (Table 8.4) for H. natalensis. JHB A Verena Waterberg JHB C JHB B Cullinan Nelspruit Balakane Lagalametse Coefficient 0.0 0.2 0.52 0.7 1.0 140 8.6.2 Antimicrobial activity Seasonal study: As the chemical variation may impact on the antimicrobial efficacy of the plant, an antimicrobial study (MIC method) was undertaken on the oil samples to determine if antimicrobial variability is present. The acceptable MIC control ranges for all pathogens except M. catarrhalis have been previously determined (Chapters 2, 3 and 4) either by the NCCLS (2003), Andrews (2004), or standards determined within the Department of Pharmacy and Pharmacology, University of Witwatersrand. The MIC control ranges for M. catarrhalis was set at a range of 0.3-0.6 ?g/mL for ciprofloxacin as determined by the mean of ten repetitions. Results for 16 of the 22 oil samples against eight pathogens are represented in Table 8.5. Seasonal samples for January (plant C), February (plant A), March (plant B), May (plant A and B) and November (plant B) were not evaluated by the MIC assay due to insufficient oil sample. With seasonal variation, one would expect the biological activity to correlate with the fluctuating levels of the major constituents. No trend exists where the two major constituents (1,8-cineole and limonene) show similarities with certain seasonal conditions. This is also seen with the microbiological assessment. Most oil samples show similar antimicrobial patterns over the season with slight variations for some samples. These variations were sporadic and restricted to individual plant samples unrelated to climatic conditions. For the Gram-positive test organisms S. aureus, B. cereus and E. faecalis, the average MIC values obtained were 6.4, 3.3 and 8.8 mg/mL respectively with the highest variability noted for S. aureus with a standard deviation of ? 3.3. Similarly, the Gram-negative test organisms showed little variability with average MIC values of 11.1, 4.8, 8.8 and 5.1 mg/mL for E. coli, P. aeruginosa, K. pneumoniae and M. catarrhalis respectively. The greatest variability noted amongst the Gram-negative test organisms was recorded for K. pneumoniae with a standard deviation of ? 2.9. The yeast, C. neoformans, showed the highest sensitivity of all pathogens studied having an average MIC value of 2.1 mg/mL with the least variance (standard deviation of ? 0.3) noted between samples (Table 8.5). The antimicrobial activity for the Zimbabwean H. natalensis species (Gundidza et al., 1993) showed good activities against pathogens E. coli, K. pneumoniae, S. aureus and Moraxella species with lower sensitivities for P. aeruginosa. In this study E. coli showed poor susceptibility and K. pneumoniae moderate activity. Even though the major constituents 1,8-cineole, and limonene, found in the oil studied by Gundidza et al. (1993) 141 correlate with that found in the South African oils, results are not comparable due to the different antimicrobial methods (Gundidza et al., 1993, made use of disc diffusion techniques). Geographical study: Table 8.6 presents the MIC data obtained when investigating the possibility of antimicrobial related geographical variation. The Gauteng samples (Cullinan, Verena and JHB BG) all indicate consistent moderate antibacterial activity (3.0-16.0 mg/mL). Only C. neoformans displayed higher sensitivities (0.8-2.0 mg/mL). Nelspruit, Waterberg and Balakane H. natalensis samples also displayed similar antimicrobial efficacies. This was clearly demonstrated when testing these samples against E. faecalis (Figure 8.9). 1 2 3 4 5 6 7 8 9 10 11 12 Figure 8.9 The MIC plate for E. faecalis showing uniform microbial inhibition for selected H. natalensis samples. Key to Figure 8.9 Columns 1 and 2 = Waterberg*1 Columns 3 and 4 = Verena*1 Columns 5 and 6 = Nelspruit*1 Columns 7 and 8 = JHB BG (sample Feb B)*1 Columns 9 and 10 = Swaziland*1 Column 11 = Positive control (ciprofloxacin)*2 Column 12 = Negative control (having no test substance) showing no inhibition. *1 Starting concentration 128 mg/mL. *2 Stock antibiotic (ciprofloxacin) solution starting concentration 0.01 mg/mL. 142 Table 8.5 The mean MIC (mg/mL) for selected H. natalensis essential oils for January ? December, 2004. * Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively Plant sample Staphylococcus aureus ATCC 12600 Bacillus cereus ATCC 11778 Enterococcus faecalis ATCC 29212 Escherichia coli ATCC 11775 Pseudomonas aeruginosa ATCC 9027 Klebsiella pneumoniae ATCC 13883 Moraxella catarrhalis (clinical strain) Cryptococcus neoformans ATCC 90112 January A 12 5 8 11 6 16 8 3 January B 7 4 8 11 4 8 5 2 February B 5 6 8 11 4 8 11 2 February C 5 2 8 11 4 6 4 2 March A 12 3 8 8 4 8 4 2 March C 3 3 8 11 4 8 4 2 April A 4 3 8 11 4 8 4 2 April B 2 3 8 11 4 8 4 2 April C 5 3 8 13 4 8 4 2 May C 4 3 13 11 4 8 4 2 October A 6 3 8 11 4 8 4 2 November A 4 3 16 11 4 8 4 2 November C 11 2 8 11 4 8 4 2 December A 11 5 12 11 2 16 6 3 December B 5 1 6 13 12 6 5 2 December C 7 3 6 11 8 8 7 2 Annual mean 6.4 ? 3.3 3.3 ? 1.2 8.8 ? 2.6 11.1 ? 1.1 4.8 ? 2.3 8.8 ? 2.9 5.1 ? 2.0 2.1 ? 0.3 Control * 0.30 x 10-3 0.30 x 10-3 1.25 x 10-3 0.10 x 10-4 0.30 x 10-3 0.60 x 10-3 0.50 x 10-3 1.60 x 10-3 143 The antimicrobial geographical variation was most apparent when observing the E. faecalis MIC data obtained from Cullinan (16 mg/mL) and Lagalametse (3 mg/mL) accounting for the large standard deviation of ? 3.8, whereas the geographical variation was least apparent when observing the C. neoformans MIC data with a ? 0.5 standard deviation. Table 8.6 The mean MIC (mg/mL) for H. natalensis essential oils selected from different localities. *1 Mean of plant B and C harvested in February (included as a comparison with other samples from Gauteng). *2Ciprofloxacin and amphotericin B served as controls for bacteria and yeast respectively. The antimicrobial MIC data from the seven different localities (Table 8.6) is presented in a dendrogram (Figure 8.10). With the exception of the Lagalametse sample, all microbiological data was closely clustered approximately around the 0.75-0.85 similarity coefficient. Clustering similarities were evident between both the essential oil data matrix (Figure 8.8) and the microbiological data (Figure 8.10) with the Lagalametse sample existing as a distant outlier in both dendrograms. H. natalensis sample Staphylococcus aureus ATCC 12600 Enterococcus faecalis ATCC 29212 Pseudomonas aeruginosa ATCC 9027 Klebsiella pneumoniae ATCC 13883 Cryptococcus neoformans ATCC 90112 Cullinan 8.0 16.0 4.0 8.0 0.8 Verena 4.0 8.0 3.0 4.0 1.0 JHB BG *1 5.0 8.0 4.0 7.0 2.0 Nelspruit 6.0 8.0 3.0 6.0 1.0 Waterberg 8.0 8.0 4.5 4.0 1.0 Balakane 4.0 8.0 4.0 8.0 0.5 Lagalametse 1.6 3.3 2.0 3.3 0.5 Geographical mean 5.2 ? 2.3 8.5 ? 3.8 3.5 ? 0.9 5.8 ? 2.0 1.0 ? 0.5 Control *2 0.3 x 10-3 1.25 x 10-3 0.3 x 10-3 0.6 x 10-3 1.6 x 10-3 144 Figure 8.10 Dendrogram constructed from the microbial MIC data in Table 8.6 for H. natalensis. With the exception of the Balakane sample, all geographical samples have 1,8-cineole and limonene as major constituents. ?-Pinene, present as a major constituent only in the Balakane sample, has previously shown either poor (Hinou et al., 1989; Kang et al., 1992) or moderate (Dorman and Deans, 2000; van Zyl et al., 2006) antimicrobial activity against the Gram-negatives. The authors mainly concur that better efficacy was noted for the Gram-positive organisms. With the exception of the MIC results shown for C. neoformans, the Balakane sample does not display any distinct antibacterial disparities and therefore it can be deduced that ?-pinene has no major influence on the overall bactericidal activity of the plant, however, the low MIC values (0.5 mg/mL) seen in this sample for C. neoformans could possibly be attributed to alterations in mitochondrial and plasma membrane function. Previous studies by Uribe et al. (1985) have shown that ?-pinene has an adverse effect on yeast membrane functions. (E)-Nerolidol present as a major constituent in both samples A and C from JHB BG have shown poor antimicrobial efficacy in a previous study (van Zyl et al., 2006) and thus it can be assumed to have no antimicrobial impact on the H. natalensis oils, JHB B Cullinan JHB C Verena Balakane Nelspruit Waterberg Lagalametse Coefficient 0.0 0.2 0.7 1.00.5 145 however, previous studies on nerolidol have shown that this aliphatic sesquiterpene has a significant role as a penetration enhancer (Xiong et al., 1996) and further studies (Brehm-Stecher and Johnson, 2003) demonstrated the non-specific bacterial permeability when nerolidol was combined with polymyxin B. Possibly (E)-nerolidol could act synergistically with other H. natalensis essential oil components where the compound enhances penetration of other more active molecules. The H. natalensis sample from Lagalametse displayed the lowest antimicrobial efficacies of all samples and against all pathogens. The chemistry of the essential oil of the Lagalametse sample is very different from the rest of the samples, having (Z)-3-hexenyl nonanoate and linalool as major constituents. A previous study (Dorman and Deans, 2000) has shown that linalool has far greater broad-spectrum antimicrobial activity when comparatively assessed with limonene. The authors presented disc diffusion data where inhibition was greater for linalool against 22 of the 25 test organisms studied. Another study (Carson and Riley, 1995) demonstrated higher efficacies for linalool when comparatively assessed with 1,8-cineole. Higher efficacies were found for linalool in both disc diffusion and MIC methodology. Pattnaik et al. (1997) also demonstrated increased efficacies for linalool for 21 of the 30 pathogens studied. A more recent publication (van Zyl et al., 2006) has shown increased efficacies against linalool when comparatively assessed with both limonene and 1,8-cineole against four pathogens. This could possibly account for the increased efficacies found in the Lagalametse sample, however, the Nelspruit sample also having linalool as a major constituent (16.7%) indicated lower susceptibilities for the pathogens studied. Thus, (Z)-3-hexenyl nonanoate exclusively found in the Lagalametse sample could possibly account for the increased antimicrobial activity noted. (Z)-3-Hexenyl nonanoate has been associated with food additives and possibly included into food perishables due to its antimicrobial efficacy, however, this cannot be confirmed and further investigation is recommended on this novel compound. 8.6.3 The enantiomeric influence on microbiological efficacy It is well-known for several pharmaceutically active compounds that stereochemistry is often related to the specific structure activity relationships of a molecule (Hutt and O? Grady, 1996; van Miert, 2003; Cordato et al., 2003). For example, levofloxacin is the optically active L-isomer of ofloxacin. Not only is levofloxin twice as active as ofloxacin but it also has a greater spectrum of antimicrobial activity, being 8-128 times more potent (Morrissey et al., 1996). This resulted in the postulation that when investigating constituent combinations, one should also take into account the enantiomeric 146 form. With this in mind an investigation was undertaken on the interaction (antagonistic, synergistic, inert or additive) between the two major constituents of H. natalensis, with regard to the specific stereochemistry of the optically active molecule i.e. (+), (-) and racemic mixtures of limonene. As indicated in the GC-MS analysis (Table 8.4), TLC profiles (Figure 8.5 and 8.6) and GC composition (Figure 8.7), the major compounds (limonene and 1,8-cineole) identified in the H. natalensis oils, predominated both the seasonal and geographical samples, and thus served as a model to study their microbiological interaction further. Previous studies have shown pathogen selective antimicrobial activity for limonene (Lis-Balchin, et. al., 1996; Neirotti et al., 1996; Dorman and Deans, 2000; Chalchat et al., 2000). For this study the MIC results determined for (+) and (-)-limonene as well as the racemic mixture show that depending on the chirality different microbiological results may be obtained. Analysis of the results outlined in Table 8.7 show that in general the (-)-limonene is the more active component in comparison to the (+)-limonene with higher sensitivities for five of the eight pathogens studied. A disc diffusion study by Dorman and Deans (2000), on the antimicrobial efficacy of different plant volatile oil components indicate moderate to no inhibition for (+)-limonene, depending on the pathogen studied, with highest sensitivity towards E. coli (11.2 mm diameter zone of inhibition). Even though the disc diffusion results are not comparable, the general antimicrobial trend for (+)-limonene indicating moderate efficacy was also noted in this study, however, the highest sensitivities found in this study were noted for B. cereus and C. neoformans, both having MIC values of 3 mg/mL. These pathogens were not investigated by Dorman and Deans (2000). In another study by Chalchat et al. (2000), antimicrobial efficacy was greater for the (-)-enantiomer. Conversely to that which was noted here, and with Dorman and Deans (2000), other studies (Lis-Balchin et al., 1996; Aggarwal et al., 2002) have indicated higher antimicrobial efficacies for (+)-limonene in the disc diffusion assay. In addition to methodology variation between the literature review and the values reported in this study, further comparative analysis could not be made as Lis-Balchin et al. (1996), did not specify which pathogens were investigated. Aggarwal et al. (2002), indicated sensitivities only for (+)- limonene, with no values given for (-)-limonene. One common finding in all of the above mentioned studies is that antimicrobial variability between isomers is apparent. The greatest variation in this study was noted for S. aureus where there was at least a three-fold difference between the enantiomeric forms of limonene (Table 8.7).xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 147 Table 8.7 The mean MIC (mg/mL) for the major constituents limonene and 1,8-cineole independently, and in combination with FIC (in brackets), determined for 1:1 combinations. * Ciprofloxacin and amphotericin B served as controls for bacteria and yeast respectively. Constituent Staphylococcus aureus ATCC 12600 Bacillus cereus ATCC 11778 Enterococcus faecalis ATCC 29212 Escherichia coli ATCC 11775 Pseudomonas aeruginosa ATCC 9027 Klebsiella pneumoniae ATCC 13883 Moraxella catarrhalis (clinical strain) Cryptococcus neoformans ATCC 90112 (+)-Limonene 13 3 27 11 4 12 8 3 (-)-Limonene 4 3 27 8 4 6 4 2 1,8-Cineole 8 2 23 8 4 8 16 2 1:1 (+) and (-)- Limonene 8 (1.1) 4 (2.6) 8 (0.6) 8 (1.8) 4 (1.0) 8 (2.0) 1 (0.4) 4 (2.3) 1:1 (+)-Limonene and 1,8-Cineole 16 (3.2) 8 (6.6) 16 (1.3) 32 (6.9) 8 (4.0) 16 (3.3) 16 (3.0) 3 (2.5) 1:1 (-)-Limonene and 1,8-Cineole 8 (3.0) 4 (3.3) 8 (0.6) 27 (6.8) 8 (4.0) 16 (4.6) 16 (5.0) 2 (2.0) 1:1 (+/-)-Limonene and 1,8-Cineole 8 (2.0) 4 (1.5) 8 (1.4) 16 (2.5) 8 (4.0) 16 (4.0) 16 (16) 2 (1.5) Control * 0.30 x 10-3 0.30 x 10-3 1.25 x 10-3 0.40 x 10-4 0.30 x 10-3 0.60 x 10-3 0.30 x 10-3 1.60 x 10-3 148 When the racemic mixture was examined, there was no consistent sensitivity pattern. The best antimicrobial activity noted for the racemate in comparison to the two isomeric forms was noted with studies on M. catarrhalis where the racemic form had an MIC value of 1 mg/mL and synergistic FIC value of 0.4. The (+) and (-)-limonene had MIC values of 8 mg/mL and 4 mg/mL respectively. This pattern was also noted for E. faecalis with MIC values for the racemate at 8 mg/mL and additive FIC value of 0.6. Both the (+) and (-)-limonene exhibited MIC values of 27 mg/mL. Comparative evaluation of the compounds studied indicated that efficacy was pathogen specific and none of the compounds showed broad-spectrum efficacy. 1,8-Cineole, known for its biological activity (Magiatis et al., 2002), demonstrated the highest efficacy of 2 mg/mL against B. cereus and C. neoformans. The 1:1 combinations of 1,8-cineole with (+), (-) and (+/-)- limonene mostly show an indifferent interaction indicating FIC?s between 1 and <4 for 13 samples. Plant essential oils however, are complex in nature having a number of constituents and antimicrobial efficacy may depend not only on the chirality of the constituent, but also on the distribution ratio of constituents. The biological activities for a number of individual oil constituents have been studied (Hinou et al., 1989; Chalchat and Garry, 1997; Griffin et al., 1999). Studies have also been conducted on essential oil combinations (Geda 1995; Lachowicz et al., 1998). Nakatsu et al. (2000) demonstrated a few instances where essential oil constituents show synergistic and antagonistic profiles. A study (Viljoen et al., 2003, Chapter 3) showed that the combination of camphor and 1,8-cineole contributed synergistically to the overall antimicrobial activity of O. asteriscoides. A study on the combination of the major constituents in A. afra (Chapter 4, Viljoen et al., 2006) has shown that the major constituents and combination thereof were not responsible for the overall activity. Even though some pharmacognostic research has been undertaken on constituent combinations, very little biological activity has been performed on the enantiomeric forms. With this study on H. natalensis, results (Figure 8.11 - 8.13) indicate how the chirality may impact on the microbiological activities obtained. When S. aureus was exposed to the various isomeric forms of limonene together with 1,8- cineole, some antagonistic activity was observed for each isomer (Figure 8.11). Where (+)- limonene and 1,8-cineole were combined, the antimicrobial activity was dependent on the 149 concentration ratios assessed. Antagonistic profiles were observed for all combinations where (+)-limonene was higher. Where 1,8-cineole and (+)-limonene were present in a 1:1 ratio combination a synergistic pattern was observed, with less significant synergistic profiles for most of the ratios where 1,8-cineole is in a higher concentration than (+)-limonene. The (-)-limonene and 1,8-cineole combination show antagonistic profiles for all ratios. Four antagonistic ratios are not plotted on the isobologram as their values are greater than the range expressed. When a racemic mixture of limonene was combined with 1,8-cineole, ratios were predominantly antagonistic with only one ratio (40% limonene and 60% 1,8-cineole) showing an additive or close to additive profile. The mixture of 1,8-cineole with (+)-limonene at a 1:1 concentration shows the highest synergistic efficacy of all the profiles studied against S. aureus. Interestingly with the single compound study (Table 8.7), (-)-limonene showed the highest efficacies of the different enantiomers, yet it is when (+)-limonene is combined with 1,8-cineole in the isobologram study that better efficacies, i.e. synergy are noted. S. aureus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+)-limonene in combination/MIC (+)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e S. aureus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (-)-limonene in combination/MIC (-)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e S. aureus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+/-)- limonene in combination/MIC (+/-)- limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e Figure 8.11 Isobologram plots for S. aureus (ATCC 12600) when exposed to the combination of (+)-limonene with 1,8-cineole, (-)-limonene with 1,8-cineole and a racemic mixture of limonene with 1,8-cineole. The 1:1 ratio as determined from the raw data is presented as a square. 150 The isobologram for the Gram-negative pathogen P. aeruginosa (Figure 8.12) show mostly additive profiles when both (+) and (-) isomeric forms of limonene were combined with 1,8- cineole. Antagonistic activity is only noted for three ratios where concentrations of (+)-limonene are greater than that of 1,8-cineole i.e. in concentration ratios observed for: 90% (+)-limonene and 10% 1,8-cineole (not noted on isobologram); 80% (+)-limonene and 20% 1,8-cineole; 70% (+)-limonene and 30% 1,8-cineole. Synergy was observed for seven ratios where the racemate of limonene was combined with 1,8-cineole. For the 1:1 MIC study (Table 8.7), an indifferent profile was obtained for all combinations of limonene with 1,8-cineole. P. aeruginosa 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+)-limonene in combination/MIC (+)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e P. aeruginosa 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (-)-limonene in combination/MIC (-)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e P. aeruginosa 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+/-)- limonene in combination/MIC (+/-)- limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e 8.12 Isobologram plots for P. aeruginosa (ATCC 9027) when exposed to the combination of (+)-limonene with 1,8-cineole, (-)-limonene with 1,8-cineole and a racemic mixture of limonene with 1,8-cineole. 151 The isobolograms (Figure 8.13) where (+)-limonene and the racemate were combined with 1,8- cineole, and tested against C. neoformans, displayed a predominantly additive profile. For the (-)-limonene / 1,8-cineole combination an antagonistic profile exists for most ratios where 1,8- cineole is in higher concentration. With ratios where (-)-limonene predominates (90:10 and 80:10) additive profiles are noted. When observing the MIC data (Table 8.7) of the compounds studied against C. neoformans, all 1:1 combinations of limonene with 1,8-cineole show indifference with FIC?s 1.5-2.5. These results correlate with the isobolograms where the 1:1 ratio (depicted within the square) is either on or in the close proximity to the straight line. C. neoformans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+)-limonene in combination/MIC (+)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e C. neoformans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (-)-limonene in combination/MIC (-)-limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e C. neoformans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MIC (+/-)- limonene in combination/MIC (+/-)- limonene alone M IC 1, 8- ci n eo le in co m bi n at io n /M IC 1, 8- ci n eo le al o n e Figure 8.13 Isobologram plots for C. neoformans (ATCC 90112) when exposed to the combination of (+)-limonene with 1,8-cineole, (-)-limonene with 1,8-cineole and a racemic mixture of limonene with 1,8-cineole. 152 From these results obtained with test organisms selected from each microbial group, it is clearly shown that not only may the chirality influence antimicrobial efficacy but also the concentration of the composition may impact on the overall activity. While the independent antimicrobial efficacies of the various enantiomeric forms of limonene highlight the significance of stereochemical configurations for antimicrobial agents, the combination with 1,8-cineole shows that consideration between constituent combinations are just as important. The variation between the (+), (-) and racemic forms of limonene when combined with 1,8-cineole differ from pathogen to pathogen. This was most apparent when observing the racemic mixture of limonene and 1,8- cineole for S. aureus where antagonism was noted. For P. aeruginosa the racemic mixture indicated synergy and for C. neoformans an additive profile was obtained. Variation within the pathogen profile were also apparent with the highest variation noted against P. aeruginosa expressing synergy, antagonism and an additive effect depending on the enantiomeric form studied. Such variation may ultimately impact on the overall activity of the whole plant, particularly when trying to correlate structure-activity relationships and must be considered as researchers could be overlooking relevant influences on antimicrobial activities. This study of H. natalensis has shown both the seasonal (with minimal microbiological variation) and geographical characterization as well as enantiomeric impact on biological activity. 8.7 General conclusions ? The major constituents for seven of the nine Heteropyxis natalensis samples analysed by GC-MS confirm 1,8-cineole (21.7-41.2%) and limonene (15.0-25.4%) as major constituents. Other major constituents were linalool at 16.7% and 11.4% for Nelspruit and Lagalametse samples respectively. (Z)-3-Hexenyl nonanoate (16.0%) was only present in the Lagalametse sample and ?-Pinene (25.2%) was identified as the major constituent in the Balakane sample. (E)-Nerolidol was present in the samples JHB BG, sample A (12.5%) and C (16.3%). ? Moderate antimicrobial activity (3.0-16.0 mg/mL) was found for most pathogens with higher sensitivities for Cryptococcus neoformans, a respiratory pathogen associated with lung infections. As this plant is traditionally used to treat respiratory disorders, efficacies tested with this pathogen may give credibility to the ethnobotanical use of Heteropyxis natalensis. 153 ? The seasonal variation of Heteropyxis natalensis samples selected from three trees in the Johannesburg Botanical Garden indicate similar chemical profiles with fluctuation in the quantity of major constituents (1,8-cineole and limonene). The antimicrobial study indicated little variation between samples. ? The geographical variation of Heteropyxis natalensis indicates similar major constituent profiles for the Gauteng, Nelspruit and Waterberg samples. The Lagalametse sample showed distinct variation both chemically and microbiologically, and antimicrobial efficacy was higher than in all the other samples. ? The microbiological enantiomeric investigation of the major constituents (limonene and 1,8- cineole) indicate varying activity profiles with (+/-)-limonene in combination with 1,8- cineole having the most significant synergistic ratios against Pseudomonas aeruginosa. Staphylococcus aureus also displayed synergistic profiles when investigated with the (+)- limonene / 1,8-cineole combination. 154 Chapter 9 Tarchonanthus camphoratus L. and Plectranthus grandidentatus G?rke, antimicrobial activity and pharmacological interaction of the non-volatile and volatile fractions. 9.1 Introduction Medicinal plants are often administered through inhalation (Caceres et al., 1991; Inouye et al., 2001a). This suggests the volatile constituents may (in part) be responsible for antimicrobial activity. Several papers (Brantner et al., 1994; Chagonda et al., 2000) have focused on the antimicrobial efficacies of either the essential oils or extracts independently. Some papers report on the bioactivity of both volatile and non-volatile plant constituents (Hammer et al., 1999; El-Shazly et al., 2004; van Vuuren et al., 2006), however, when dealing with aromatic plants, there is a need to determine the specific contribution of both the volatile and non- volatile constituents to test the hypothesis that the aromatic constituents, while independently may exhibit poor activity, in combination with non-volatile constituents act synergistically to enhance biological activity. This study aims to verify the need for the co-existence of volatile and non-volatile constituents for enhanced antimicrobial efficacy using two plants Tarchonanthus camphoratus L. (Asteraceae and Plectranthus grandidentatus G?rke (Lamiaceae) as a model for further investigation. Tarchonanthus camphoratus, or the wild camphor bush / tree as it is commonly known, is a strongly scented tree (van Wyk et al., 1997; Watt and Breyer-Brandwijk, 1962). The name ?anthos? from Tarchon- ?anthus? is derived from the Greek word meaning flower, whereas the name ?camphoratus? refers to the strong camphor odour, which the leaves impart (van Wyk et al., 1997). Hutchings and van Staden (1994), mentions T. camphoratus for its potential antimicrobial properties. A literature search on T. camphoratus yielded only four articles. One published more or less at the same time as Hutchings and van Staden?s article, reporting the volatile constituents of T. camphoratus in Kenya (Mwangi et al., 1994). Another later publication (McGaw et al., 2000) evaluated the biological properties of T. camphoratus extracts 155 together with a number of other species and found no antimicrobial activity when tested at a starting concentration of 12.5 mg/mL. A recent publication (Matasyoh et al., 2007) indicated marked activity in the disc diffusion assay and conflicting poor activity in the MIC assay for selected pathogens tested against the essential oil of T. camphoratus. Some taxonomic revision has also been undertaken on the camphoratus complex (Herman, 2002). The large Plectranthus genus (over 350 species), of which 300 species occur in Africa, have a rich diversity of ethnobotanical use. A number of reports on the antimicrobial activity have been documented. In a study by Rabe and van Staden (1998), the crude leaf extracts of a number of Plectranthus species were investigated for their potential antimicrobial activity. Some of the species i.e. P. ambiguus, P. ciliatus, P. ecklonii, P. ernsti, P. fruiticosus, P. hadiensis, P. hilliardiae, P. laxiflorus, P. madagascariensis, P. oertendahli, P. oribiensis, P. petiolaris, P. purpuratus, P. reflexus, P. rubropunctatus, P. strigosus, P. succatus, P. verticllatus and P. zuluensis as investigated by disc diffusion assay, exhibited variable antimicrobial activity. A further study undertaken by Maistry (2003), investigated the essential oil composition and antimicrobial activity of eight species (P. ciliatus, P. fruticosus, P. grandidentatus, P. hadiensis, P. neochilus, P. porphyranthus, P. venteri and P. zuluensis). While Rabe and van Staden (1998) focused on the Plectranthus extracts and Maistry (2003) examined the essential oil fraction, it is clear that both studies indicated activity for the Gram-positive test organisms albeit in the disc diffusion assay. From the numerous literature reports on the Plectranthus species, it was noted that P. grandidentatus has been poorly investigated. In a review article on the Plectranthus species (Abdel-Mogib et al., 2002), seven species were noted for their medicinal properties. Plectranthus grandidentatus however, was not mentioned. In a more recent review (Lukhoba et al., 2006), the species P. grandidentatus was listed within the medicinal ethnomedicinal category. Maistry (2003) did indicate some antimicrobial activity for this species but this was not further explored. Even though antimicrobial assessment was undertaken for P. grandidentatus by Teixeira et al. (1997), the antimicrobial quantitative values were only reported for the isolated compounds. This Chapter focuses not only on the antimicrobial properties of the essential oil and extracts of T. camphoratus and P. grandidentatus, but also on whether pharmacological interactions between the volatile and non-volatile fractions are evident. 156 9.2 Botanical description Tarchonanthus camphoratus is a small shrub-like tree having thin oblong leaves which are dark green on the upper side and pale velvety-gray on the underside (van Wyk et al., 1997). The tree is semi-deciduous growing approximately two to nine metres in height. Small whitish flowers are produced giving rise to woolly fruits (van Wyk et al., 1997), as clearly noted in Figure 9.1. These woolly fruits, produced mostly from March through to June are strongly scented (van Wyk and Gericke, 2000). Plectranthus grandidentatus is classified as a perennial semi-succulent herb. The leaves of P. grandidentatus show size variation with ovate to broadly ovate characteristics. The stems have a trailing, dishevelled appearance growing up to two metres (Codd, 1985). White to purple flowers are borne on terminal lateral shoots (Figure 9.3). 9.3 Distribution Tarchonanthus camphoratus is widely distributed in southern Africa (Figure 9.2), extending from Namibia to Botswana, from the Northern Province, Gauteng, Free State to the Northern Cape (van Wyk et al., 1997; Herman, 2002). The T. camphoratus complex has recently been revised, where a number of different taxa have been reclassified due to differences in Figure 9.1 The leaves of Tarchonanthus camphoratus with woolly fruits. Figure 9.2 The geographical distribution of T. camphoratus in southern Africa (SANBI). 157 morphology and geographical distribution (Herman, 2002). Based on locality and voucher material, T. camphoratus L. was the taxon identified for this study. Plectranthus grandidentatus is naturally occurring within dry, rocky areas or in open woodland. The species is distributed along the eastern region of South Africa (Figure 9.4) from the Soutspansberg mountain range, northwards to Mpumalanga and then through to Swaziland, KwaZulu-Natal Midlands and extending through to the eastern area of the Cape (Codd, 1985). 9.4 Medicinal uses The leaves of T. camphoratus have been used for numerous medicinal applications, from an infusion for stomach ailments and bronchitis to the inhalation for headaches. Also noted was the use of the leaves as a hot poultice for chest complaints. Chewing of the leaves is said to alleviate toothache (Hutchings et al., 1996). The green branches are burnt and the smoke emanating from this is inhaled for sinuses (van Wyk and Gericke, 2000). The use of the plant as a treatment for sexually transmitted diseases has also been noted (Watt and Breyer-Brandwijk, 1962). Africans have used the burned leaf and seed for fumigation and as an anointment in funeral rituals. Interestingly the word ?Tarchos? is derived from the Greek word meaning funeral flower. Figure 9.4 The geographical distribution of P. grandidentatus in South Africa (SANBI). Figure 9.3 The inflorescence of Plectran- thus grandidentatus. 158 The ethno-medical information available on the Plectranthus species indicates its use for coughs, colds and chest complaints, and administered by means of an infusion made from crushed leaves (Hutchings et al., 1996). Root deconcoctions are prepared for respiratory complaints and abdominal disorders (Watt and Breyer-Brandwijk, 1962). Fifteen species have been identified for the treatment of respiratory conditions and twenty-one species have been categorized for digestive complaints encompassing stomach-ache, diarrhoea and oral infections (Lukhoba et al., 2006). In the same review publication, 20 species were recorded as being used for dermal applications where some species are used as a dressing on wounds, burns and skin ulcerations. The medicinal use of this genus encompasses a number of species, however, very little specific medicinal properties could be found for the species P. grandidentatus. In the extensive review of the ethnobotanical literature of the Plectranthus species, where over 200 references are cited, only one reference (Cerqueira et al., 2004) refers to the medicinal properties of P. grandidentatus. 9.5 Methods 9.5.1 Chemical aspects Plant collection and distillation of essential oils: The aerial parts of T. camphoratus and P. grandidentatus within a single population were collected from the Walter Sisulu Botanical Garden during February. Voucher specimens (Table 9.1) were deposited within the Department of Pharmacy and Pharmacology, University of Witwatersrand. The plants were hydrodistilled as described in Chapter 2.5.1, however, distillation time was extended by one hour to allow for the complete removal of all volatile constituents from the plant material. Table 9.1 Plant collection data for T. camphoratus and P. grandidentatus. Gas chromatography combined with mass spectrometry (GC-MS) data: Oil samples were quantitatively analyzed with GC-MS using the Hewlett-Packard 1800A GCD system operating under the same conditions as described in Chapter 2.5.1. Plant Voucher Material distilled (g) Essential oil yield (% w/w) T. camphoratus SVV1100 1243.0 0.10 P. grandidentatus SVV1099 4.6 0.02 159 Extract preparation with soxhlet apparatus: Plant samples were divided into three quantities for separate extraction procedures with the soxhlet apparatus. The soxhlet extraction method was selected on the basis of the method being heat activated, thus being more comparable to distillation than cold extraction methods. The plant material remaining in the Clevenger apparatus after distillation was dried at 40 ?C in an oven (Memmert), weighed (T. camphoratus 8.39 g; P. grandidentatus 5.96 g) and placed inside an extraction thimble (Whatman). This was placed into a soxhlet extractor (Figure 9.5). The extractor was attached to a flask containing 250 mL of a 1:1 methanol and chloroform mixture. The solvent was selected for its ability to extract both polar and non-polar constituents. Heating of the solvent allowed for the extraction of plant material by movement of the solvent through the outer chamber of the soxhlet apparatus, through to the condenser, after which permeation into the extraction thimble holding the plant material occurred. When the solvent in the soxhlet chamber (in contact with plant material) was the same colour as the pure solvent, extraction was complete (approximately three hours). The operating temperature of the soxhlet extraction process was maintained at approximately 100 ?C, thus allowing for both the chloroform (boiling point 61 ?C) and methanol (boiling point 65 ?C) to extract a full range of constituents. Thereafter, excess solvent was removed by using a rotary evaporator. A second portion of plant material (T. camphoratus, 14.80 g; P. grandidentatus, 23.30 g) was similarly extracted but with fresh plant material. The third portion of fresh leaf material was oven-dried at 37 ?C for 24 hr. Thereafter, the plant material (T. camphoratus, 7.90 g; P. grandidentatus, 6.18 g) was placed in a soxhlet apparatus and extracted in the same way as mentioned above. The different preparative techniques of the plant samples allowed for the examination of the non-volatile and volatile constituents. The methods and resultant constituents with a relevant key are as follows:kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk ? EO = Essential oil prepared by distillation (compounds obtained: volatile constituents only). ? NV = Non-volatile constituents prepared by soxhlet extraction of fresh plant material remaining in the still thus having no (or negligible) volatile constituents. ? FC = Volatile and non-volatile constituents from fresh plant material that have been soxhlet extracted (compounds obtained: non-volatile compounds [polar and non-polar], and essential oil constituents). ? DC = Volatile and non-volatile constituents that have been prepared from dried plant material that have been soxhlet extracted (compounds obtained: non-volatile compounds [polar and non-polar], and essential oil constituents). 160 Figure 9.5 The soxhlet apparatus. 9.5.2 Antimicrobial aspects Culture, media preparation and MIC assays were undertaken according to the NCCLS (2003) guidelines, Carson et al. (1995) and Eloff (1998a). Comparative assays were performed on eight test mico-organisms. Acetone was used as the antimicrobial solvent for P. grandidentatus samples, however, due to the insolubility of the T. camphoratus plant extracts in acetone, dimethyl sulfoxide (DMSO, Merck) was used as the solvent for reconstitution. As DMSO exhibits some antimicrobial activity, MIC values equivalent to or greater than that found for the DMSO control were omitted from the data (Table 9.6) and considered not susceptible (NS) at the highest concentration tested (64mg/mL). This was inclusive for P. grandidentatus samples having little or no susceptibility. Although DMSO is not the preferred choice of solvent for antimicrobial studies, it is accepted for extract miscibility (Okeke et al., 2001; van Es et al., 2005; Magee et al., 2007) and is used extensively in laboratory procedures (Santos et al., 2003). Synergy / antagonistic interaction: Once the independent MIC values were determined, the interaction between the essential oil containing component of the plant (EO) and the non- volatile (NV) component was investigated based on the hypothesis that efficacy may be enhanced by having non-volatile constituents combined with the volatile constituents. The essential oil, at a starting concentration of 64 mg/mL, was combined in selected ratios with the non-volatile extract at a starting concentration of 64 mg/mL. Isobolograms were constructed based on the methodology as described in Chapter 7.5.2. Assays were conducted in duplicate or triplicate where there was a difference in the correlation of results, and the mean plotted on an 161 isobologram. Conventional antimicrobials (controls) were run in tandem with assays to assess pathogen sensitivity patterns. These are not shown on isobolograms and were included in all replications. Assay controls having solvent only were undertaken with each pathogen where ratios of acetone (used with essential oil and P. grandidentatus extract reconstitution) and DMSO (used with T. camphoratus extract reconstitution) were assayed to determine the antimicrobial effect of the solvents independently and in various combinations. Raw MIC data for T. camphoratus can be noted in Table A3, Appendix A. Raw MIC data for P. grandidentatus can be noted in Table A4, Appendix A Concentration in mg/mL MIC values Ratio values X Y X Y X Y (essential oil) (non-volatile) (essential oil) (non-volatile ) (essential oil ) (non-volatile ) 16.0 0.0 8.00 0.00 1.000 0.000 14.4 1.6 3.60 0.40 0.450 0.100 12.8 3.2 3.20 0.80 3.200 0.200 11.2 4.8 2.80 1.20 0.350 0.300 9.6 6.4 2.40 1.60 0.300 0.400 8.0* 8.0* 2.00* 2.00* 0.250* 0.500* 6.4 9.6 1.60 2.40 0.200 0.600 4.8 11.2 1.20 2.80 0.150 0.700 3.2 12.8 0.80 3.20 0.100 0.800 1.6 14.4 0.40 3.60 0.050 0.900 0.0 16.0 0.00 4.00 0.000 1.000 *The 1:1 ratio as determined from the raw data is presented as a square on the isobolograms. Concentration in mg/mL MIC values Ratio values X Y X Y X Y (essential oil ) (non-volatile) (essential oil ) (non-volatile ) (essential oil ) (non-volatile ) 16.0 0.0 2.00 0.00 1.000 0.000 14.4 1.6 0.23 0.03 0.113 0.071 12.8 3.2 0.40 0.10 0.200 0.286 11.2 4.8 0.35 0.15 0.175 0.429 9.6 6.4 0.30 0.20 0.150 0.571 8.0* 8.0* 0.25* 0.25* 0.125* 0.714* 6.4 9.6 0.10 0.12 0.050 0.343 4.8 11.2 0.75 0.18 0.375 0.500 3.2 12.8 0.10 0.40 0.050 1.143 1.6 14.4 0.10 0.90 0.050 2.571 0.0 16.0 0.00 0.35 0.000 1.000 *The 1:1 ratio as determined from the raw data is presented as a square on the isobolograms. Table 9.2 Raw data for the isobologram construction for the volatile (EO) constituent interaction with non-volatile constituents of T. camphoratus against E. faecalis. Table 9.3 Raw data for the isobologram construction for volatile (EO) constituent interaction with non-volatile constituents of P. grandidentatus against S. aureus. 162 For P. grandidentatus, the study was limited to three pathogens S. aureus, K. pneumoniae and E. faecalis due to low oil acquisition. Pathogen selection was based on the ethnobotanical use of P. grandidentatus. The raw data for the isobologram construction where the volatile and non- volatile constituents are combined is demonstrated for two test organisms E. faecalis (T. camphoratus) and S. aureus (P. grandidentatus) and given in Table 9.2 and Table 9.3 respectively. 9.6 Results and discussion 9.6.1 Essential oil chemistry Tarchonanthus camphoratus Ninety-eight compounds were identified in the T. camphoratus essential oil, representing 94.3% of the total composition (Table 9.4). Earlier studies on T. camphoratus species harvested from Kenya identified 1,8 cineole (16.5%) and ?-fenchyl alcohol (29.1%) as major constituents with ?-caryophyllene only present in trace amounts (Mwangi et al., 1994). In this study ?-fenchyl alcohol was not present at all and the bicyclic sesquiterpene ?-caryophyllene featured predominantly. A more recent publication by Matasyoh et al. (2007), also examined the essential oil from Kenyan T. camphoratus species where the major compounds 1,8 cineole (14.3%) and fenchol (15.9%) were identified. ?-Caryophyllene was only present as a minor (0.6%) compound. Similarity between both Kenyan species (Mwangi et al., 1994; Matasyoh et al., 2007) and that investigated here were the presence of ?-pinene (3.9-6.9%) and 1,8-cineole (9.3- 16.5%). The two Kenyan studies although harvested from different localities display more similarity. Geographical variation is not uncommon (Viljoen et al., 2005 and 2006) and possibly the variance of ?-caryophyllene between the Kenyan species and this study can be attributed to differences in locality. Furthermore, it is interesting to note that although the vernacular name (camphor bush or camphor wood) suggests the presence of camphor, no evidence of such was present in either the analysed oil in this study, nor the T. camphoratus oil studied by Matasyoh et al. (2007). Camphor was only present in minor quantities (0.38%) in the study undertaken by Mwangi et al. (1994). 163 Table 9.4 Essential oil composition of T. camphoratus. RRI* Compound name % 1032 ?-pinene 3.9 1035 ?-thujene 0.1 1076 camphene 1.7 1118 ?-pinene 1.0 1146 ?-2-carene 4.1 1176 ?-phellandrene 1.0 1188 ?-terpinene 0.1 1203 limonene 1.8 1213 1,8-cineole 9.3 1246 (Z)-?-ocimene 0.1 1255 ?-terpinene 0.1 1266 (E)-?-ocimene tr 1280 p-cymene 2.1 1290 terpinolene 0.1 1327 (Z)-3-hexenyl acetate 0.3 1360 hexanol 0.4 1391 (Z)-3-hexen-1-ol 0.3 1400 nonanal 0.1 1402 (Z)-2-hexen-1-ol 0.1 1406 ?-fenchone tr 1450 trans-linalool oxide (Furanoid) tr 1466 ?-cubebene 0.1 1476 isoledene tr 1493 ?-ylangene 0.4 1497 ?-copaene 1.2 1528 ?-bourbonene tr 1535 ?-bourbonene 0.5 1544 ?-gurjunene 0.1 1553 linalool 0.8 1559 ?-maaliene tr 1571 trans-p-menth-2-en-1-ol tr 1589 ?-ylangene 0.8 1601 ?-copaene 1.2 1611 terpinen-4-ol 0.7 1612 ?-caryophyllene 13.4 1617 6,9-guaiadiene 0.2 1628 aromadendrene 0.2 1645 cadina-3,5-diene tr 1659 ?-gurjunene 0.1 1661 alloaromadendrene 3.4 1677 epi-zonarene 0.3 1682 ?-terpineol tr 1687 ?-humulene 1.1 1705 ?-muurolene 3.2 164 RRI* Compound name % 1706 ?-terpineol 1.0 1708 ledene 0.2 1719 borneol 0.1 1720 4,6-guaiadiene (=?-Guaiene) 0.1 1725 zonarene 1.3 1739 ?-muurolene 1.6 1740 valencene 0.7 1744 ?-selinene 0.1 1755 bicyclogermacrene 0.2 1773 ?-cadinene 5.1 1776 ?-cadinene 2.0 1799 cadina-1,4-diene (=Cubenene) 0.4 1807 ?-cadinene 0.6 1853 cis-calamenene 1.5 1864 p-cymen-8-ol 0.1 1918 ?-Calacorene tr 1941 ?-Calacorene 0.5 1943 palustrol 0.2 2001 isocaryophyllene oxide 0.1 2008 caryophyllene oxide 2.2 2017 maaliol 0.1 2029 perilla alcohol 0.1 2033 epiglobulol 0.1 2037 salvial-4(14)-en-1-one 0.2 2057 ledol 0.5 2058 13-tetradecanolide 0.1 2071 humulene epoxide-II 0.1 2072 cubeban-11-ol? 0.4 2074 caryophylla-2(12),6(13)-dien-5-one 0.1 2080 cubenol 1.1 2088 1-epi-cubenol 1.8 2098 globulol 1.1 2104 viridiflorol 0.5 2144 rosifoliol 0.6 2146 spathulenol 0.7 2153 neointermedeol 0.2 2164 6-epi-cubenol 0.1 2165 1,10-diepicubenol? 0.8 2185 T-cadinol 2.6 2209 T-muurolol 2.0 2214 torreyol 1.0 2247 trans-?-bergamotol 0.3 2255 ?-cadinol 4.3 2264 intermedeol 0.1 2265 4-(4?-Methyl-3?-pentenyl)-3-cyclohexenyl-propyl ketone 0.1 2272 alismol 0.1 2273 selin-11-en-4?-ol 0.2 165 RRI* Compound name % 2316 caryophylla-2(12),6(13)-dien-5?-ol (=Caryophylladienol I) 0.2 2324 caryophylla-2(12),6(13)-dien-5?-ol (=Caryophylladienol II) 0.8 2375 caryophylla-2(12),6-dien-5?-ol (=Caryophyllenol I) 0.4 2384 10-hydroxy calamenene 0.3 2389 eudesma-4(15),7-dien-1?-ol 0.3 2392 caryophylla-2(12),6-dien-5?-ol (=Caryophyllenol II) 0.1 2622 phytol 0.6 TOTAL 94.3 * RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = Trace (< 0.1 %) ? = Tentative identification from Wiley, MassFinder, Adams libraries. Major compounds: The major compounds (Figure 9.6) identified by GC-MS of T. camphoratus are ?-caryophyllene (13.4%) and to a lesser extent 1,8-cineole (9.3%), representing 22.7% accumulatively. H H O ?-caryophyllene 1,8-cineole Figure 9.6 Chemical structures for major compounds identified in the essential oil of T. camphoratus. Plectranthus grandidentatus Sixty-two compounds were identified in the P. grandidentatus essential oil, representing 95.6% of the total composition (Table 9.5). A previous study on the essential oils of numerous Plectranthus species displayed quantitative and qualitative variation in major compound composition. Some Plectranthus species (P. ciliatus ex Ferncliff and P. zuluensis ex hort C. Potgieter) did not have T-cadinol and camphor at all. Other species displayed only T-cadinol as both a minor compound (P. neochilus, 1.3%) and a major compound (P. hadiensis, 26.9%). 166 Plectranthus hadiensis was the only other species shown to include both T-cadinol and camphor as major constituents (Maistry, 2003). Table 9.5 The essential oil composition of P. grandidentatus (Maistry, 2003). RRI* Compound name Percentage 1032 ?-pinene 0.6 1035 ?-thujene 0.4 1076 camphene 0.2 1132 sabinene 0.8 1176 ?-phellandrene 0.7 1188 ?-terpinene 0.1 1203 limonene 9.4 1218 ?-phellandrene 0.5 1246 (Z)-?-ocimene 0.1 1255 ?-terpinene 0.4 1266 (E)-?-ocimene 1.9 1280 ?-cymene 0.6 1290 terpinolene 0.2 1391 (Z)-3-hexene-1-ol tr 1393 3-octanol 0.5 1406 ?-fenchone 4.3 1452 1-octen-3-ol 0.6 1495 bicycloelemene 0.3 1505 + 1497 dihydroedulane 2 + ?-copanene 0.5 1532 camphor 15.2 1553 linalool 0.5 1571 trans-?-menth-2-en-1-ol tr 1594 trans-?-bergamotene 0.1 1600 ?-elemene 0.3 1612+1611 ?-caryophyllene + terpinen-4-ol 7.7 1628 aromadendrene 0.1 1638 cis-?-menth-2-en-1-ol 0.1 1655 9-epi-?-caryophyllene 0.6 1661 alloaromadendrene 0.6 1668 (E)-?-farnesene 0.1 1687 ?-humulene 1.0 1706 ?-terpineol 0.6 1719 borneol 1.5 1726 germacrene D 1.2 1737 (Z)-E-?-farnesene 0.5 1755 bicyclogermacrene 3.3 1764 sesquicineol 9.2 1773 ?-cadinene 0.4 1776 ?-cadinene 1.2 167 RRI* Compound name Percentage 1784 (E)-?-bisabolene 0.5 1810 3,7 guaiadiene 0.1 1845 trans-carveol 0.1 1853 cis-calamenene 0.2 1864 ?-cymene-8-ol tr 2001 isocaryophyllene oxide 1.0 2008 caryophyllene oxide 3.9 2050 (E)-nerolidol tr 2051 gleenol tr 2069 germacrene D-4-ol 0.5 2080 cubenol 5.6 2098 globulol 0.5 2104 viridiflorol 0.3 2144 spathulenol 2.7 2170 ?-bisabulol 0.3 2187 T-cadinol 10.3 2209 T-muurolol 0.2 2232 ?-bisabolol 1.9 2247 trans-?-bergamotol 0.2 2255 ?-cadinol 0.7 2324 caryophyllandienol-2 0.1 2389 caryophyllenol-1 0.2 2392 caryophyllenol-2 0.3 TOTAL 95.6 * RRI: Relative retention indices calculated against n-alkanes % calculated from TIC data tr = Trace <0.1%. Major compounds: The two major compounds (Figure 9.7) in the essential oil of P. grandidentatus are T-cadinol (10.3%) and camphor (15.2%), representing 25.5% accumulatively. OH O T-cadinol camphor Figure 9.7 Chemical structures for major compounds identified in the essential oil of P. grandidentatus. 168 9.6.2 Antimicrobial activity In previous studies with H. cymosum subsp. cymosum (van Vuuren et al., 2006, Chapter 6) and C. gratissimus var. subgratissimus (Chapter 7), it was observed that the non-volatile constituents in a plant showed higher antimicrobial activity than the volatile constituents. Studies with H. natalensis (Chapter 8) have shown that plant essential oil constituents may play a synergistic role. This study on T. camphoratus and P. grandidentatus examined the interactive role of volatile and non-volatile constituents to determine if when combined, constituents would have antimicrobial synergistic properties. The MIC data for the EO, NV, FC and DC for both T. camphoratus and P. grandidentatus is presented in Table 9.6. Tarchonanthus camphoratus The MIC values for the essential oils (EO) ranged between 1.5-16.0 mg/mL depending on the pathogen studied. These values were significantly higher in activity than that found for the Kenyan T. camphoratus oil, exhibiting MIC values within the range 113-450 mg/mL (Matasyoh et al., 2007). Previous studies on ?-caryophyllene have shown poor antimicrobial activity when investigated against 16 test organisms (Kang et al., 1992). This could possibly indicate that it is not the single major compound ?-caryophyllene that is responsible for activity but that a synergistic interaction exists between compounds, or even that one or more of the minor compounds not present in the Kenyan samples could be responsible for the activity noted in the South African species. The non-volatile fraction (NV) devoid of volatile constituents displayed higher antimicrobial efficacies with a narrow MIC range between 2.0-4.0 mg/mL, with the exception of E. faecalis and K. pneumoniae showing no susceptibility at the highest concentration (16.0 mg/mL) tested. 169 Table 9.6 The mean MIC (mg/mL) for the essential oil (EO) and plants that have undergone soxhlet extraction (NV, FC, DC) for T. camphoratus*1 and P. grandidentatus*2. *3Ciprofloxacin and amphotericin B served as controls for bacteria and yeasts respectively. NS = Not susceptible at the highest concentration tested (64mg/mL). Essential oil (EO) Non-volatile (NV) Combined fresh material (FC) Combined dried material (DC) Pathogen T. camp*1 P. grand*2 T. camp*1 P. grand*2 T. camp*1 P. grand*2 T. camp*1 P. grand*2 Control*3 Staphylococcus aureus ATCC 25923 6.0 4.0 2.0 0.5 2.0 8.0 1.5 0.16 0.2 x 10 -3 Bacillus cereus ATCC 11778 3.5 3.0 3.0 0.02 4.0 4.0 2.0 0.63 0.2 x 10 -3 Enterococcus faecalis ATCC 29212 NS 12.0 NS NS NS 6.0 NS 0.06 0.6 x 10 -3 Escherichia coli ATCC 11775 16.0 NS 3.0 NS 3.5 NS 4.7 NS 0.1 x 10 -4 Pseudomonas aeruginosa ATCC 9027 6.0 4.0 4.0 3.3 4.0 NS 3.3 1.5 0.3 x 10 -3 Klebsiella pneumoniae NCTC 9633 NS NS NS 4.0 4.0 8.0 4.0 4.0 0.8 x 10 -3 Candida albicans ATCC 10231 6.0 6.0 2.0 3.0 0.6 8.0 0.3 6.0 2.5 x 10 -3 Cryptococcus neoformans ATCC 90112 1.5 3.0 2.0 3.0 3.0 4.0 1.5 0.13 2.5 x 10 -3 170 Of the two combined preparations, the dried plant material showed the most favourable synergistic profile with MIC values between 0.3-4.7 mg/mL, depending on the pathogen studied. Comparative investigation of two pathogens (E. coli and K. pneumoniae) investigated by both McGaw et al. (2000) and the combined fractions in this study indicate moderate antimicrobial activity. Even though each pathogen showed different sensitivity patterns, it is clear that when volatile and non-volatile constituents are combined efficacy is enhanced. This is seen to a lesser extent for the fresh plant material (two pathogens) and more notable for the dried plant material (five pathogens). Although ethnobotanical reports (Hutchings et al., 1996; van Wyk et al., 1997; van Wyk and Gericke, 2000) make more reference to the use of the dried plant material, the traditional use of the plant is not restricted to the dried material and depending on the ethnobotanical use, fresh plant material is also administered. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 S. aureus B. cereus E. faecalis E. coli P. aeruginosa K. pneumoniae C. albicans C. neoformans Micro-organism M IC v a lu e (m g/ m l) Volatile Non-volatile Dried combined Figure 9.8 The comparative MIC (mg/L) for the volatile constituents (EO), non-volatile constituents (NV) and dried combined (DC) constituents for T. camphoratus. 171 Figure 9.8 shows a condensed version of the MIC results (EO, NV and DC) obtained from Table 9.6, allowing for easier comparative evaluation. The FC plant material and controls have been excluded allowing for the more active combined plant material (DC) to be comparatively assessed with the essential oils and non-volatiles. Results show that for all pathogens with the exception of C. neoformans and K. pneumoniae where efficacies are equivalent, the non-volatile extracts devoid of any essential oils show higher antimicrobial efficacy than the volatile constituents. For B. cereus, K. pneumoniae and C. neoformans, both volatile and non-volatile constituents were more or less equal in efficacy. It is interesting to note that two of these organisms (K. pneumoniae and C. neoformans) are predominantly associated with respiratory infections. Tarchonanthus camphoratus is administered by inhalation thus suggesting that the volatile constituents may be the active constituents. The synergistic roles of the volatile and non-volatile constituents in various ratios are depicted in isobolograms (Figures 9.9-9.11). The control isobolograms for all pathogens showed that when acetone was in higher concentration, no antimicrobial activity was obtained. When DMSO was in greater concentration, antimicrobial activity was minimal. When plotted on an isobologram no comparative ratios were noted thus indicating that the solvents had no marked effect on the results obtained for the test plant material. The isobolograms for the Gram-positive organisms indicate synergy for most ratios. Both B. cereus and E. faecalis exhibited antagonism for one ratio where 80% non-volatile constituents are combined with 20% volatile constituents. These ratios are not seen on the isobolograms as the points fall out of the 1.25:1.25 scale (Figure 9.9). This predominantly synergistic pattern suggests that the presence of the essential oils may be necessary for increased efficacy as observed in the ratios where higher essential oil (volatile) content results in synergy. The Gram- positive test organisms S. aureus and B. cereus show similar synergy patterns in comparison with the MIC study (Table 9.6). No comparison could be made for E. faecalis due to lack of susceptibility in the MIC assay. 172 S. aureus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e B. cereus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e E. faecalis 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e Figure 9.9 The antimicrobial efficacy of the Gram-positive test organisms (S. aureus, B. cereus and E. faecalis respectively) against varying concentrations of the non-volatile (NV) and volatile (EO) constituents of T. camphoratus. Isobologram studies for the Gram-negative pathogens (Figure 9.10) mostly indicate synergistic interactions. It was only with the isobologram for E. coli that antagonism was seen for five ratios (three of which are not seen on isobologram, as the value falls out of the 1.25:1.25 scale). When cross-examining with MIC data, it can be noted that the combined plant material (fresh and dried) showed no increased efficacy with E. coli and this was the only pathogen studied which showed this pattern. For isobologram studies with P. aeruginosa and K. pneumoniae, results were congruent with the MIC values. As noted with E. faecalis, similar results were obtained for P. aeruginosa where ratios having a higher non-volatile component showed an additive profile 173 and in ratios having a higher volatile concentration synergy was observed. This supports the need for the presence of volatile constituents in order for synergy to occur. Synergy was observed for all ratios, except one, when studied against the test organism K. pneumoniae. E. coli 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e P. aeruginosa 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e K. pneumoniae 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e Figure 9.10 The antimicrobial efficacy of the Gram-negative test organisms (E. coli, P. aeruginosa and K. pneumoniae respectively) against varying concentrations of the non-volatile (NV) and volatile (EO) constituents of T. camphoratus. The isobolograms for the yeasts (Figures 9.11) show synergistic profiles for all ratios studied. This again confirms the need for volatile constituents to enhance activity. Correlation with the 174 MIC study is true for C. albicans where both the fresh and dried combined plant material had higher efficacies than when studied independently. C. albicans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e C. neoformans 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e Figure 9.11 The antimicrobial efficacy of the yeasts (C. albicans and C. neoformans respectively) against varying concentrations of the non-volatile (NV) and volatile (EO) constituents of T. camphoratus. The overall comparative assessment of the volatile, non-volatile with the combined volatile / non-volatile extracts show that for the eight pathogens studied in the MIC assay, five showed increased efficacy for the dried combined material, thus supporting the theory that volatile constituents increase efficacy. This was further validated by the isobologram studies where a predominant synergistic profile for all pathogens was presented. All 1:1 ratios on the isobolograms depicted as a square show synergism for all pathogens including B. cereus. Due to the close proximity of the ratio points on the B. cereus isobologram, no square was inserted for the 1:1 ratio as this would impair data visualization. Plectranthus grandidentatus A number of studies have been undertaken on the chemistry of the Plectranthus genus (Batista et al., 1994; Teixeira et al., 1997; Dellar et al., 1996; Gaspar-Marques et al., 2005) and in an excellent review (Abdel-Mogib et al., 2002) where the isolation and identification of antimicrobial compounds amongst others were initiated. From these studies it is clear that the main phytochemical constituents found in the extracts of the Plectranthus genus are the 175 diterpenoids. Approximately 140 diterpenoids have been identified in the leaf extracts of Plectranthus species of which vinylogous quinones, acylhydroquinones, dimeric abietanoids and royleanones were isolated from P. grandidentatus (Abdel-Mogib et al., 2002). Other compounds isolated include grandidone A and B (Uchida et al., 1981) and 7-epigrandidone D (Teixeira et al., 1997). From bio-autographic studies undertaken by Teixeira et al. (1997), it was confirmed that the royleanones were the compounds responsible for antimicrobial activity. Previous antimicrobial studies on Plectranthus species have shown variable antimicrobial activity (Rabe and van Staden, 1998). Of the four test organisms studied, the greatest antimicrobial activity was found for S. aureus against 26 Plectranthus species. The other test organisms, S. epidermidis, B. subtilis and M. lutea indicated mostly poor to no activity in the disc diffusion assay. The antimicrobial activity for P. grandidentatus was not investigated, possibly due to previous work undertaken on the species (Teixeira et al., 1997). The authors (Rabe and van Staden, 1998) however, did recommend that further antimicrobial investigation be carried out on the essential oil of the species. Table 9.6 presents the MIC data obtained for P. grandidentatus essential oils and extracts prepared with and without inclusion of the volatile constituents. The essential oil MIC data for P. grandidentatus (Table 9.6) indicate variable sensitivities, depending on the pathogen studied. Two of the three Gram-positive test organisms indicated moderate to good activity (S. aureus, 4 mg/mL; B. cereus, 3 mg/mL). Only E. faecalis indicated poor sensitivity (12 mg/mL). Of the three Gram-negative organisms investigated, only P. aeruginosa showed sensitivity towards the essential oil of P. grandidentatus with an MIC value of 4 mg/mL. These results were not congruent with the disc diffusion results as seen with Maistry?s (2003) report where antimicrobial efficacy was only noted for B. cereus and C. neoformans. For the MIC results obtained in this study these two pathogens showed the highest sensitivity patterns. As seen in previous studies and discussed in further detail in Chapter 11.9, disc diffusion methodology is not always comparable to MIC assays. Maistry (2003) did perform a MIC assay with B. cereus (4 mg/mL) where results were more or less congruent with those obtained here (3 mg/mL). Antimicrobial correlation with the essential oil chemistry suggests that possibly the major compound, camphor, may in part be responsible for the antimicrobial efficacy as camphor is known to possess some antimicrobial properties (Griffin et al., 1999; Magiatis et al., 2002). The bactericidal effect of T-cadinol, also present as a major constituent in the P. grandidentatus oil has been previously investigated against S. aureus and E. coli where results indicated high 176 susceptibility for S. aureus (24 ?g/mL) and no susceptibility for E. coli (Claeson et al., 1992). The MIC values for S. aureus (4 mg/mL) and E. coli (not susceptible at the highest concentration tested, 64mg/mL) noted in this study correlate with the results obtained by Claeson et al. (1992) and suggest that T-cadinol may in part be responsible for the antimicrobial activity noted in the P. grandidentatus oil. The MIC results for P. grandidentatus, obtained when the volatile constituents were removed from the extract, indicate mostly increased activity. Only E. faecalis and E. coli demonstrated no susceptibility in the non-volatile extract. These pathogens also displayed poor to no antimicrobial activity when exposed to the oils. When fresh plant material was extracted containing both volatile and non-volatile constituents increased efficacy was only noted for E. faecalis. As in the case with most microbial species studied against T. camphoratus, the dried combined plant material indicated higher antimicrobial activities with four pathogens (S. aureus, E. faecalis, P. aeruginosa and C. neoformans), displaying increased efficacies when volatiles and non-volatiles were combined. Thus only the dried combined plant sample containing volatile and non-volatile constituents were selected to comparatively investigate the essential oil (volatile) and non-volatile extract (Figure 9.12). When observing the ethnobotanical use of the species, no correlation could be found for the preferred use of dried material over fresh material and a review of the Plectranthus genus by Lukhoba et al. (2006), when reviewing 62 species, make no reference to the use of fresh or dried plant material. The isobologram ratio study for S. aureus show predominantly synergistic profiles for seven of the nine ratios studied (Figure 9.13). The two ratios expressing antagonism is where the extract is at highest concentrations, (80:20%) and the ratio 90:10%. The point is not seen on isobologram as the ratio is greater than the scale expressed. It is interesting to note than when observing the MIC data (Table 9.6) the essential oil (4 mg/mL) has far less efficacy than the non-volatiles devoid of oil (0.5 mg/mL), yet when combined, efficacy is enhanced both in the isobologram study and with the MIC study (DC). Enterococcus faecalis indicated the highest synergistic effect when observing both the fresh combined material and dried combined material combination (Table 9.6). This enhanced 177 synergistic pattern was corroborated in the isobologram study where synergy was noted for all ratios investigated (Figure 9.13). >32 >16 >32 0.02 >16 0.06 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 S. aureus B. cereus E. faecalis E. coli P. aeruginosa K. pneumoniae C. albicans C. neoformans Micro-organism M IC v al u e (m g/ m l) Volatile Non-volatile Dried combined >32 Figure 9.12 The comparative MIC (mg/L) for the volatile constituents (EO), non-volatile constituents (NV) and dried combined (DC) constituents for P. grandidentatus. This pattern was also noted for the isobologram ratio study for K. pneumoniae (Figure 9.13) and MIC data (Table 9.6), however, to a lesser extent. The dried combined plant material having volatile and non-volatile constituents displayed a MIC value of 4 mg/mL while the essential oil exhibited no activity. All ratios within the isobologram study for K. pneumoniae displayed a synergistic pattern. All 1:1 ratios on the isobolograms (depicted as squares), show synergism. Due to the close proximity of the ratio points for the K. pneumoniae and E. faecalis isobologram no square was inserted as this would impair data visualization. 178 S. aureus 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e K . pneumoniae 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e E. faecalis 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 MICEO in combination/MICEO alone M IC N V in co m bi n at io n /M IC N V al o n e Figure 9.13 The antimicrobial efficacy of the test organisms S. aureus, K. pneumoniae and E. faecalis respectively against varying concentrations of the non-volatile (NV) and volatile (EO) constituents of P. grandidentatus. From this work it is shown that there may be considerable differences in microbiological activity between volatiles obtained by distillation and non-volatiles obtain through extraction. The greatest variation seen in the study on T. camphoratus is where E. coli indicated an MIC value of 16.0 mg/mL for the essential oil and the non-volatile plant extract had a MIC value of 3.0 mg/mL, thus exhibiting more than a 5-fold difference. Similarly for P. grandidentatus where B. cereus has a MIC value of 3.0 mg/mL for the essential oil and the non-volatile plant extract had a MIC value of 0.02 mg/mL. When plant material comprised of both volatile and non-volatile constituents, efficacy was enhanced for most pathogens. Increased efficacies were predominantly 179 found for the dried combined plant material over the fresh material. Eloff (1998b), affirmed that the choice of dried over fresh plant material for biological studies were favoured by a number of researchers. In a similar study (Lalli, 2006), undertaken on three Pelargonium species, the comparative TLC and HPLC data were included and indicated no decomposition of non-volatile compounds during distillation. Confirmation was also made using TLC to ensure that the plant material remaining in the distillation apparatus (NV) contained no essential oil constituents. Even though the ratio combination of the volatile / non-volatile mixtures used in this study does not truly reflect the natural ratios in the whole plant, it is clear that synergistic mechanisms exist, which necessitates the inclusion of volatile compounds in the antimicrobial assessment. One also needs to consider the dosage and administration of herbal medicines when observing targeted activity. The various different administration methods (infusions, inhalations, tinctures, deconcoctions etc.) are usually carefully selected by the healer for the greatest efficacy, where preparation often serves to neutralize toxins (van Wyk et al., 1997). It has also been demonstrated (Bonati, 1980) that the tincture of Valeriana is more potent than the preparation in water. The administration of an infused oil, where the oil acts as a penetrative enhancer for active principles of the extract is also widely accepted (Shealy, 1998) and conceivably the combination of volatile / non-volatile constituents improve solubility and bioactivity of the active principles. The actual mode of action where the essential oil and non-volatiles interact to enhance activity is complex and may be attributed to a number of different mechanisms. It is known that essential oils disrupt the bacterial cell wall (Burt, 2004) and a number of studies have confirmed that aromatic compounds exert their antimicrobial efficacy on the cytoplasmic membrane by altering its structure and function (Cox et al., 2000; Lambert et al., 2001; Holley and Patel, 2005). This was clearly observed with confocal membrane integrity studies undertaken on O. asteriscoides (Chapter 3.6.2), however, the antimicrobial effects of essential oils may be dependent on a number of other factors such as pathogen specificity, oil concentration, level of activity and relevant function of the microbial membrane such as pH gradient, ATPase activity, electrochemical potential, ion leakage, changes in composition of fatty acid and intracellular absorption of material (Nychas et al., 2002). Such complexities require further investigation, however, one could postulate that the essential oils provide a different mode of action to the non- 180 volatiles and thus act synergistically to enhance activity. This synergistic interaction could possibly enhance activities in a similar manner to other combined antimicrobials such as amoxicillin and clavulanic acid. Here the mechanism of action is known, where clavulanic acid while independently exerts little antimicrobial activity, in combination with amoxicillin, increases the spectrum of activity by inactivating ?-lactamases, blocking active enzyme sites and thus allowing for a greater efficacy (Gibbon, 2005). Similarly essential oils could enhance entry into the cell and allow for selected non-volatile components to exert bactericidal activity. 9.7 General conclusions ? The major constituents of the essential oil of Tarchonanthus camphoratus are ?- caryophyllene (13.4%) and to a lesser extent 1,8-cineole (9.3%), representing 22.7% accumulatively. The two major compounds in the essential oil of Plectranthus grandidentatus are T-cadinol (10.3%) and camphor (15.2%), representing 25.5% accumulatively. ? The non-volatile constituents devoid of any essential oils show higher antimicrobial efficacy than the volatile constituents. The only exceptions were Cryptococcus neoformans and Enterococcus faecalis when exposed to Tarchonanthus camphoratus and Plectranthus grandidentatus respectively. ? When volatile constituents were combined with non-volatile constituents (fresh and dried plant material), MIC values most often indicated better efficacy for the dried combined (volatile and non-volatile) plant material. ? Isobologram representation of the combination of various ratios of essential oil and non- volatile extracts devoid of essential oils, present a predominant synergistic profile for all pathogens for Tarchonanthus camphoratus and Plectranthus grandidentatus. ? The correlation between MIC data and isobolograms were mainly congruent where the greatest synergy for Tarchonanthus camphoratus was noted for Bacillus cereus, Klebsiella pneumoniae Candida albicans and Cryptococcus neoformans. The greatest synergy for Plectranthus grandidentatus was observed with Enterococcus faecalis. These efficacies 181 correlate with the traditional use of the plants to treat stomach ailments and respiratory conditions. 182 Chapter 10 A comparative investigation of the antimicrobial properties of indigenous South African aromatic plants with popular commercially available essential oils. 10.1 Introduction Popular commercial oils, such as lavender (Lavendula angustifolia), thyme (Thymus vulgaris), tea tree (Melaleuca alternifolia), peppermint (Mentha piperita) and rosemary (Rosmarinus officinalis) have been used extensively to treat bacterial and fungal infections associated with the urinary tract, wounds, skin disorders, digestion, as well as the respiratory tract (Shealy, 1998). Their popular use for antimicrobially related disorders validated the choice to comparatively investigate these commercial essential oils with five indigenous South African essential oils used in ethnic healing rites. While literature (Carson and Riley, 1995; Mangena and Muyima, 1999; Tassou et al., 2000 Angioni et al., 2004) documents and supports the antimicrobial action of these commercial oils, their popularity has been complimented by excellent marketing strategies. Even though the majority of the rural South African population consult traditional practitioners, for urban dwellers there is still a trend to rather use globally recognized herbal therapies, of which popular essential oils is considered the most widely used of all natural therapies (Nakatsu et al., 2000). Based on traditional use, consultation with literature and results obtained from research, five indigenous aromatic plants were selected from this thesis for comparative study: M. flabellifolius (Chapter 2), O. asteriscoides (Chapter 3), A. afra (Chapter 4), L. javanica (Chapter 5) and H. natalensis (Chapter 8). While antimicrobial activities have been assessed for these indigenous oils, this study aims to directly compare, under identical laboratory conditions, the antimicrobial activity against the five commercial oils mentioned above. The objective was to demonstrate the potential to actively promote South African indigenous aromatic plants and their essential oils for antimicrobial therapy. 183 10.2 Botanical description and distribution An abridged botanical description with distribution map for indigenous oils can be found in the relevant corresponding Chapters. For the commercial plant oils: Lavendula angustifolia, commonly known as lavender is a member of the Lamiaceae family. It is native to the Mediterranean but is widely cultivated throughout Europe. This bushy evergreen shrub bears grey leaves turning greener with age. Most commonly blue to mauve-coloured flowers (Figure 10.1) are borne in the summer months (Lis-Balchin, 2002). Thymus vulgaris or thyme as it is commonly known, is also from the Lamiaceae family. It is a small hardy shrub perennial to the Mediterranean. Leaves are dark grey to green in colour- bearing small white or violet flowers (Figure 10.2) which bloom from spring to summer (van Wyk and Wink, 2004). Melaleuca alternifolia or tea tree, is a member of the Myrtaceae family, indigenous to Australia. The evergreen tree grows in swampy, low-lying areas having needle-like leaves (Figure 10.3). Stalkless, yellow or purple flowers are borne in the flowering season (Shealy, 1998). Rosmarinus officinalis (rosemary), from the Lamiaceae family is native to the Mediterranean, but cultivated world-wide for its culinary, perfumery and medicinal uses. It is a woody perennial shrub with evergreen spiky leaves. The flowers are mainly purple in colour but may vary between pale pink to blue (Figure 10.4). This highly aromatic herb has many traditional uses and was often planted in cemeteries where the odour emitted was thought to stimulate memories of deceased loved ones (Shealy, 1998). Mentha piperita, a perennial mint plant from the Lamiaceae family, originates from Europe and is well-known as a flavouring and therapeutic agent. The use of the plant dates back to 300 BC where hieroglyphics show the use of mint as an ingredient in incense. Historical anecdotes have shown that the Greeks and Romans have used mint at their feasts and the Arabs have prepared mint in beverages. Of the many species of mint available, Mentha piperita is regarded as therapeutically more superior due to the high menthol content. The copious amount of menthol produced is a characteristic feature of the plant, giving it the distinctive peppermint odour. The 184 leaves are short with finely-toothed margins (Figure 10.5) producing spiked pink to mauve flowers in late summer (Wootton, 2005). Figure 10.1 Lavendula angustifolia Figure 10.2 Thymus vulgaris (thyme) (lavender) Figure 10.3 Melaleuca alternifolia Figure 10.4 Rosmarinus officinalis (tea tree)*1 (rosemary) Figure 10.5 Mentha piperita (peppermint) *2 *1 http://yourskin101.com/natural-skin-care/tea-tree-oil.aspx; *2 http://sophy.u-3mrs.fr/photo-id/Men 185 10.3 Medicinal uses A description of the medicinal uses for indigenous oils can be found in the relevant corresponding Chapters. For the commercial plant oils: Lavendula angustifolia: The essential oil from this species is one of the most popular and appreciated in phytotherapy (Lis-Balchin, 2002). The name of the plant is derived from the Latin word ?lavare? meaning ?to wash? referring to the medicinal use for the washing of wounds. The medicinal properties range widely from the treatment of coughs and colds by steam inhalation to applying topically for the treatment of burns and acne (Shealy, 1998). Thymus vulgaris: The essential oil from this species was traditionally used as a medicine for treating coughs and respiratory disorders associated with the upper respiratory tract. It is known as an antiseptic to treat wounds and urinary tract infections (Shealy, 1998). There have been a number of reports validating the antimicrobial efficacy of thyme (Delespaul et al., 2000; Valero and Salmer?n, 2003) and in a publication by Lai and Roy (2004), where the antimicrobial activity is reviewed by a number of authors against bacterial isolates such as S. aureus, B. subtilis and E. coli, in conjunction with other organisms. Due to its known antimicrobial efficacy it is often used in combination with other oils as a preservative. Melaleuca alternifolia: The Australian aborigines have originally used tea tree oil as an antiseptic and the leaves were used as a poultice for skin infections. It is a popular choice for the effective treatment of respiratory complaints where a few drops of the oil are placed into hot water and the vapours inhaled. Diluted tea tree oil is used as a mouthwash to protect against dental caries. The demand for tea tree oil as a disinfectant increased when it was used by Australian soldiers participating in World War I (http://www.vitacost.com/science/hn/Herb/Tea_ Tree.htm). The use of tea tree oil to treat topical fungal infections (with particular reference to Candida albicans) is well-known. Several in vitro studies have revealed that the oil inhibits many strains of bacteria, yeasts and fungi (Carson and Riley, 1995; Shealy, 1998; Caelli et al., 2000; Cox et al., 2001). Rosmarinus officinalis: The oil is known for its antiseptic and disinfectant properties (Shealy, 1998). The use of rosemary dates back to the ancient Egyptians where it was used in cleansing rituals. In former times this aromatic herb was rubbed on uncooked meat to delay putrefaction. During outbreaks of the plague, rosemary was a common ingredient used in posies carried to 186 ward off infection. More recently rosemary is used in the fumigation of French hospitals. Of the many virtues bestowed on rosemary, the therapeutic value has been accredited to the treatment of bronchitis, sinusitis, as an expectorant, as a mucolytic and antiseptic. The therapeutic application by inhalation has also been used by both aromatherapists and traditional Chinese medicinal practitioners. Furthermore, it has been used as a scalp treatment for the eradication of dandruff (Holmes, 1999; Wootton, 2005). Mentha piperita: The medicinal use of the oil has been used to treat sinusitis by inhalation therapy. It has been used for the treatment of skin disorders but is mostly known for digestive disorders, hence the traditional consumption of peppermints after a large meal (Shealy, 1998). In addition to the numerous reports on the antimicrobial properties of mint, as cited by Logan and Beaulne (2002), the use of enteric coated peppermint oil was found to be beneficial in the treatment of overgrowth of intestinal flora, often associated with irritable bowel syndrome. 10.4 Methods 10.4.1 Chemical aspects Essential oils: The aerial parts of the five indigenous plants were collected from various localities within South Africa and hydrodistilled for three hours in a Clevenger apparatus according to methods described in Chapter 2.5.1. Plant samples were retained for identification and voucher specimens housed within the Department of Pharmacy and Pharmacology at the University of the Witwatersrand. The corresponding voucher numbers are given in Table 10.1. The five commercial essential oils with their technical data were obtained from Robertet (France) where distillation and analysis was carried out according to ISO compliance. Gas chromatography (GC) and gas chromatography combined with mass spectrometry (GC-MS): Even though the GC and GC-MS analysis of the indigenous oils have been previously undertaken, all ten essential oils and the standards were confirmed on a Shimadzu 17A gas chromatograph using the following parameters: Column: J&W-DB-1 (30 m x 0.25 mm, 0.25 ?m film thickness), Temperatures: injection port 230 ?C, column 60 ?C for 1 min., 5 ?C / min to 180 ?C, 180 ?C for 2 min., (total = 25 min.). Quantification was achieved by correlation of relative retention times with authentic standards. The major compounds and their relative percentages are presented in Table 10.1 and Figure 10.6. 187 10.4.2 Antimicrobial aspects The culture preparation was undertaken according to the NCCLS (2003) guidelines (Chapter 2.5.2) and the microplate MIC bioassay method was used (Carson et al., 1995; Eloff, 1998a; NCCLS, 2003), as in previous Chapters. Tests were performed in triplicate and identical laboratory conditions were maintained to ensure that inoculum, batch media, preparative techniques and incubation times were standardized. Where MIC data indicated coinciding values between commercial and indigenous oils, the MIC was further refined to narrow the increments. This was achieved by altering the dilution factor by preparing consecutive dilutions of increments 0.75, 0.72 and 0.65 (depending on the predetermined MIC). The MIC assay is performed on the principle of doubling dilutions with a ratio of 0.5 for each dilution. By manipulating the dilution factor i.e. changing the ratio, a narrower MIC value can be obtained. This would enable one to obtain a MIC reading somewhere between the two doubling dilutions. This can be demonstrated by observing the Myrothamnus flabellifolius oil sample, initially having an MIC value of 8 mg/mL when using the MIC doubling dilution principle. By changing the dilution ratio to 0.72, a MIC reading of either 16.59; 11.94; 8.60 (actual value obtained); 6.19 or 4.46 could be obtained, thus reducing the doubling dilution error normally encountered in MIC methodologies. Time-kill studies: To comparatively demonstrate the time-kill efficacy, commercial and indigenous oils were exposed to S. aureus, K. pneumoniae and C. albicans (Figures 10.7-10.9) under identical laboratory conditions and the cidal efficacy plotted over time against the logarithm of viable colonies. The assays were carried out in duplicate according to methodology described in Chapter 2.5.2. Only one oil concentration (0.5%; 5.2 mg/mL) for all ten test samples was comparatively evaluated. 10.5 Results and discussion 10.5.1 Essential oil chemistry The major compounds in the commercial oils, identified by gas chromatography are presented in Figure 10.6. This ensured that commercial oils were of a high quality and gave some indication of which indigenous chemotype was included in this study. The major compounds for the indigenous oils have been diagrammatically presented in previous Chapters. A summary of data for all oils comparatively evaluated can be observed in Table 10.1. 188 Lavendula angustifolia O - Ac OH linalyl acetate linalool Thymus vulgaris OH p-cymene thymol Melaleuca alternifolia OH terpinen-4-ol ?-terpinene Mentha piperita O OH menthone menthol 189 Rosmarinus officinalis O O camphor 1,8-cineole ?-pinene Table 10.1 A summary of the major compounds and therapeutic use for the five indigenous and five commercial essential oils. Species, family and voucher number Common name Major compounds and relative % Therapeutic use Lavendula angustifolia (Lamiaceae) Commercial sample Lavender linalyl acetate (31.3%), linalool (30.8%) Acne, skin abrasions and wounds, antiseptic, respiratory tract infections. Thymus vulgaris (Lamiaceae) Commercial sample Thyme thymol (47.2%), p-cymene (22.1%) Wounds, antiseptic, urinary tract infections, colds, coughs and flu. Melaleuca alternifolia (Myrtaceae) Commercial sample Tea Tree Terpinen-4-ol (38.6%), ?-terpinene (21.6%) Acne, skin abrasions and wounds, antiseptic, respiratory tract infections, fever, fungal infections. Mentha piperita (Lamiaceae) Commercial sample Peppermint menthone (18.2%), menthol (42.9 %) Mainly for digestive disorders, respiratory ailments, skin lesions and acne. Rosmarinus officinalis (Lamiaceae) Commercial sample Rosemary camphor (12.4%), 1,8-cineole (41.4%), ?-pinene (13.3%) Antiseptic and disinfectant. Lippia javanica (Verbenaceae) Voucher number: AMV331 Fever Tea ?-ocimene (19%), linalool (65%) Respiratory tract infections, fever, digestive disorders. Artemisia afra (Asteraceae) Voucher number: AdcAV173 African Wormwood 1,8-cineole (17.8%), camphor (8.2%), ?-thujone (18.7%), ?-thujone (12.5%), artemisia ketone (10.1%) Respiratory tract infections, fever, digestive disorders. Figure 10.6 Chemical structures for the major compounds identified in the commercial oils of L. angustifolia, T. vulgaris, M. alternifolia, M. piperita and R. officinalis. 190 Species, family and voucher number Common name Major compounds and relative % Therapeutic use Heteropyxis natalensis (Heteropyxidaceae) Voucher number: AMV771 Lavender Tree 1,8-cineole (31.4%), limonene (17.9%), ?-terpineol (7.3%) Respiratory tract infections. Osmitopsis asteriscoides (Asteraceae) Voucher number: ADCAV174 ?Bels? 1,8-cineole (56%), (-)-camphor (14.8%) Respiratory tract infections, inflammation, skin lesions, digestive disorders, fever. Myrothamnus flabellifolius (Myrothamnaceae) Voucher number: AV169 Resurrection Bush pinocarvone (11.1 %) trans-pinocarveol (19.6%) limonene (6.1%) trans-p-mentha-1(7), 8- dien 2-ol (7.4%) cis-p-mentha-1(7), 8-dien 2-ol (6.9%) Respiratory tract infections, skin abrasions and wounds, urinary tract infections, mouth ulcers. 10.5.2 Antimicrobial activity Initial MIC data indicated that indigenous essential oils compare favourably with the commercial oils as noted for S. aureus where M. flabellifolius and O. asteriscoides (4 mg/mL) showed higher sensitivity than the commercial oils: L. angustifolia (lavender), M. alternifolia (tea tree), M. piperita (peppermint) and R. officinalis (rosemary) at 8 mg/mL. However, in many instances there was very little variance between MIC values for indigenous and commercial oils. This was noticed for B. cereus where O. asteriscoides, H. natalensis, T. vulgaris (thyme), M. alternifolia (tea tree), M. piperita (peppermint) and R. officinalis (rosemary) oils where the same MIC value (4 mg/mL) was recorded (not shown in Table 10.2). A narrower MIC was hence determined. The final refined MIC values are depicted in Table 10.2. Minimum inhibitory values against all ten pathogens generally ranged between 4-16 mg/mL for most oils. A few exceptions are seen e.g. with the indigenous oils M. flabellifolius indicating the highest sensitivity towards B. cereus (2 mg/mL) and S. epidermidis (3.2 mg/mL). This was also noted for O. asteriscoides (1 mg/mL) against C. neoformans. Similarly, T. vulgaris (thyme) oil showed the highest sensitivity towards S. aureus (1.3 mg/mL) and E. coli (0.5 mg/mL). Previous comparative studies (Hammer et al., 1999) also support high antimicrobial activity for T. vulgaris (thyme) where the lowest MIC reported was against E. coli (0.03%), in comparison with oils from 20 other genera. 191 Table 10.2 The MIC (mg/mL) of the five indigenous and five commercial oils. Pathogen Lippia javanica Artemisia afra Heteropyxis natalensis Osmitopsis asteriscoides Myrothamnus flabellifolius Lavendula angustifolia (Lavender) Thymus vulgaris (Thyme) Melaleuca alternifolia (Tea Tree) Mentha piperita (Peppermint) Rosmarinus officinalis (Rosemary) Control* Staphylococcus aureus ATCC 12600 16.0 4.5 8.6 4.0 4.0 8.6 1.3 8.6 11.9 6.2 1.0 x 10 -3 Staphylococcus epidermidis ATCC 2223 7.6 16.0 5.7 6.2 3.2 6.2 4.7 6.2 6.2 10.1 2.5 x 10 -3 Bacillus cereus ATCC 11778 16.0 11.9 4.5 3.7 2.0 11.9 8.8 2.4 3.7 3.7 1.0 x 10 -3 Bacillus subtilis ATCC 6051 32.0 32.0 32.0 6.2 8.6 32.0 11.9 6.2 4.5 32.0 2.5 x 10 -3 Escherichia coli ATCC 11775 1.6 8.6 4.5 3.7 5.7 6.2 0.5 3.7 5.7 4.5 0.1 x 10 -3 Pseudomonas aeruginosa ATCC 9027 16.0 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 6.2 2.5 x 10 -3 Enterococcus faecalis ATCC 29212 13.5 5.7 18.0 7.0 6.2 18.0 4.0 8.6 6.2 24.0 2.0 x 10 -4 Klebsiella pneumoniae NCTC 9633 7.6 10.1 10.1 11.9 5.7 16.0 4.0 4.0 6.2 16.0 1.3 x 10 -3 Candida albicans ATCC 10231 8.6 4.5 4.5 5.7 5.7 5.7 2.4 3.7 2.4 5.7 1.3 x 10 -3 Cryptococcus neoformans ATCC 90112 2.4 8.0 5.7 1.0 2.0 4.0 4.0 4.0 4.0 1.6 2.5 x 10 -3 *The bacterial control was ciprofloxacin and fungal control was amphotericin B. 192 Results of these studies encompassing all five of the commercial oils were mostly congruent with previous findings. Carson and Riley (1995), confirmed results obtained for tea tree oil where it was also observed that E. coli was more susceptible than S. aureus. In another study, where eleven oils were comparatively evaluated for antibacterial activity, thyme oil showed a much higher inhibitory action than peppermint oil against fifteen of the eighteen pathogens studied (?zkan et al., 2003). Similarly in this study thyme showed higher inhibitory action in eight out of ten pathogens investigated. While MIC data for some of these oils have been reported, studies are not always directly comparable due to different methods, microbial sensitivities etc., hence the data represented here can be accurately compared due to the standardization of test conditions. Thymus vulgaris (thyme) oil showed the highest overall antimicrobial activity in the MIC assay followed by M. flabellifolius and then O. asteriscoides and M. alternifolia (tea tree) oil. Time-kill data for S. aureus (Figure 10.7) indicate that of the ten oils studied, three of the four having bactericidal activity are from indigenous South African plants. Indigenous oils O. asteriscoides indicated bactericidal efficacy at 24 hr and both M. flabellifolius and A. afra displayed a cidal effect at four hours. Thymus vulgaris (thyme) was the only commercial oil showing bactericidal efficacy (at eight hours). These results showed some correlation with the MIC data where the same four oils had the lowest MIC values for S. aureus. The remaining six oils (L. angustifolia, R. officinalis, M. alternifolia, M. piperita, H. natalensis and L. javanica) show very similar bacteriostatic patterns, averaging approximately 50% reduction in CFU / mL after 24 hr. 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U/ m l Control Heteropyxis Artemisia Osmitopsis Lippia Myrothamnus Tea tree Lavender Rosemary Thyme Peppermint Figure 10.7 A comparative death kinetic study of S. aureus, exposed to 0.5% essential oil from five commercial and five indigenous plants over a 24 hr period. 193 Death kinetics for K. pneumoniae (Figure 10.8) indicated less sensitivity. Although a rapid reduction of CFU was observed within one hour for O. asteriscoides, A. afra and M. piperita (peppermint) oil, regrowth of K. pneumoniae was evident within one hour for A. afra and M. piperita (peppermint) oil and more gradually (after four hours) with O. asteriscoides. Bactericidal efficacy was noted for T. vulgaris (thyme) oil, M. alternifolia (tea tree) oil, as well as M. flabellifolius oil, having a time-kill effect within one hour. Again these results confirm similar susceptibility patterns as seen in the MIC data where the same three oils have the lowest MIC values for K. pneumoniae. 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 1.E+09 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U/ m l Control Lippia Artemisia Heteropyxis Osmitopsis Myrothamnus Lavender Thyme Tea tree Peppermint Rosemary Figure 10.8 A comparative death kinetic study of K. pneumoniae, exposed to 0.5% essential oil from five commercial and five indigenous plants over a 24 hr period. Most of the oils showed fungicidal efficacy for C. albicans (Figure 10.9) with the exception of L. angustifolia (lavender) and H. natalensis having limited static activity. Even though an approximate 25% reduction of CFU was noted for these two oils, no cidal activity was observed after the 24 hr test period. The oils of O. asteriscoides, A. afra and T. vulgaris (thyme) indicated time-kill death rates within four hours. The oil M. flabellifolius exhibited the most rapid (within one hour) bactericidal activity against C. albicans. Surprisingly M. alternifolia (tea tree) oil, marketed for it?s anti-candidal activity did not indicate cidal efficacy in the first four hours but only after 24 hr, with oil samples R. officinalis (rosemary), M. piperita (peppermint) and L. javanica. This reduced time-kill effect was also noted in a comparative study (Hammer et al., 1999) in which the same five commercial oils, amongst others, were studied. Results were 194 mainly congruent with M. alternifolia (tea tree) oil being less effective than T. vulgaris (thyme) and L. angustifolia (lavender) oils for C. albicans. Of the four oils demonstrating cidal efficacy for C. albicans within four hours, three (M. flabellifolius, O. asteriscoides and A. afra) are attributed to the indigenous oils. 1.E-01 1.E+01 1.E+03 1.E+05 1.E+07 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U/ m l Control Lippia Artemisia Heteropyxis Osmitopsis Myrothamnus Lavender Thyme Tea tree Peppermint Rosemary In light of these findings we can confirm that the indigenous oils compared most favourably with commercial oils and as indicated in the death kinetic assay M. flabellifolius, an indigenous oil had the most rapid cidal effect for all three pathogens S. aureus, K. pneumoniae and C. albicans. Earlier studies (Chapter 2, Viljoen et al., 2002) state that the major compound (trans- pinocarveol, 19.6%) in M. flabellifolius is also used in pharmaceutical preparations such as OzopulminTM, thus further substantiating the plant?s potential for medicinal use. Comparative death kinetics (Lattaoui and Tantaoui-Elaraki, 1994) for the oils of three Thymus species (T. broussonettii, T. zygis and T. satureioides) had time-kill efficacies within three and a half hours against S. aureus (0.4% oil concentration) and C. albicans (0.2% oil concentration). While T. vulgaris (thyme) oil used in this study showed the highest microbial efficacies of all five commercial oils studied at 0.5% concentrations, the death kinetic rate was four hours and eight hours for C. albicans and S. aureus respectively. From both these studies, it can be confirmed that some Thymus species possess strong antimicrobial properties. In a time-kill study by May et al. (2000), the rapid bactericidal effect of M. alternifolia (tea tree) oil within 30 min was also Figure 10.9 A comparative death kinetic study of C. albicans, exposed to 0.5% essential oil from five commercial and five indigenous plants over a 24 hr period. 195 noted, confirming the time-kill data presented in this study. A time-kill study of M. alternifolia oil (Christoph and Stahl-Biskup, 2001) indicated a very similar cidal trend against S. aureus to that seen in this study, however, the authors demonstrated that combinations with other oils, such as manuka, showed excellent bactericidal efficacy, thus one should not exclude synergistic oil combinations in microbial therapy. In summary these results validate the use of South African aromatic plants for their anti-infective properties and emphasize the need to promote the natural botanical resources in South Africa. Many urban South African people are growing European herbs for medicinal purposes but are reluctant or ignorant of the use of African plants for therapeutic use (Derwent, 2004). Thus, by publicising the efficacy of indigenous oils in comparison with those that are internationally well marketed (van Vuuren and Viljoen, 2006), a bridge may possibly be formed between western herbal practices and that of indigenous sources that have been known for centuries within the ethnic people. We need to take into account however, that while these in vitro antimicrobial results support the local traditional use of indigenous aromatic plants, harvesting limitations need to be considered to make further commercial development a viable proposition. Also, of note is that not all in vitro antimicrobial results correlate with their therapeutic use and other factors such as chemophysical, pharmacological or physiological aspects may contribute to further enhance the efficacy of these essential oils. 10.6 General conclusions ? Comparative MIC results for popular commercial oils (Lavendula angustifolia, Thymus vulgaris, Melaleuca alternifolia, Mentha piperita and Rosmarinus officinalis) against the essential oils of five indigenous plants (Myrothamnus flabellifolius, Osmitopsis asteriscoides, Heteropyxis natalensis, Artemisia afra and Lippia javanica) indicated little variance for many samples with MIC values mainly ranging between 4-16 mg/mL. ? Time-kill data for Staphylococcus aureus indicate that of the ten oils studied, three of the four having bactericidal activity are from indigenous South African plants i.e. Osmitopsis asteriscoides, Myrothamnus flabellifolius and Artemisia afra with Thymus vulgaris as the only commercial oil showing bactericidal efficacy. 196 ? Death kinetics for Klebsiella pneumoniae indicated less sensitivity, with bactericidal efficacy noted for Thymus vulgaris, Melaleuca alternifolia and Myrothamnus flabellifolius oil, having a time-kill effect within one hour. ? Eight of the ten oils investigated demonstrated fungicidal efficacy for Candida albicans in the time-kill assay. ? Of all oils studied, Myrothamnus flabellifolius showed the most rapid cidal effect against all three pathogens tested. ? Results indicate that the antimicrobial activity of South African indigenous essential oils compare favourably with the commercial oils investigated in this study. 197 Chapter 11 A review of the methods used to study the antimicrobial activity of aromatic plants. 11.1 Introduction Many papers on the antimicrobial efficacy of plant essential oils and extracts have been published, however, there still remains a number of disparities in the techniques, where there is a need to explore standardization (Rios and Recio, 2005). This is evident even though numerous publications have been dedicated to the methods involved in bio-activity studies of extracts and oils (Remmal and Tanaoui-Elaraki, 1993; Brantner et al., 1994; Carson et al., 1995; Pauli and Kubeczka, 1997; Janssen et al., 1987; Eloff, 1998a; Marston and Hostettmann, 1999; Lahlou, 2004; Rios and Recio, 2005). As microbiological methods incorporate living principles (test micro-organisms), predictability of the outcome is not always clear and subject to many environmental influences that may impact on a response. Hence, the problems encountered when trying to standardize techniques. Standardization of protocols for antimicrobial testing of plants has limitations since the study objectives and experimental design amongst researchers often differ (Holley and Patel, 2005). The methodologies involved in the empirical approach to discover biological activities should be carefully considered and approached so that results yield accurate and reproducible outcomes (Marston and Hostettmann, 1999). It is of the utmost importance that parameters such as plant collection, validation of laboratory equipment, chemical analysis and various intricacies of antimicrobial investigations are carefully defined. This Chapter serves to document the standardization of techniques based on literature reviews and practical suggestions when performing microbiological assays. Microbiological assays on natural products are techniques where the potency of a test substance (plant extract, essential oil or compound) may be determined against the growth of a micro- organism. 198 The classification of such studies can be four-fold. There should be an initial screening process. Ideally this should be relatively cost effective, quick and need not be quantitative (Marston and Hostettmann, 1999). In these microbiological studies, disc diffusion assays meet these requirements. Secondly, screening should follow which confirms qualitative biological activity (Marston and Hostettmann, 1999). This has been achieved by determining the quantitative concentration that will inhibit microbial growth using the minimum inhibitory concentration (MIC) method. Both disc diffusion and MIC techniques may be carried out with the aid of automated equipment, however, all assays undertaken in this thesis have been carried out manually. Thirdly, monitoring assays should be in place for studies, such as isolation of antimicrobial constituents from whole plants, as in the case of bioassay-guided fractionation procedures. Secondary testing should be in place after monitoring studies to confirm activity, specifically for studies involving the isolation of active compounds. Hostettmann (1999) recommended that biological assays be used in conjunction with chemical screening to allow for a multidisciplinary approach that may yield important information for future drug development. Monitoring assays also include time-kill studies where viability is assessed over a selected time period. Lastly, a more outcome based approach may be undertaken to study the activity in more detail, as with synergy studies where a specific outcome is sought to test a hypothesis (Marston and Hostettmann, 1999). 11.2 Plant collection and essential oil isolation For all plant collections voucher specimens were established and species correctly identified in order to botanically authenticate the plant material. For many follow-up studies plant material was re-collected, always from the same locality to avoid any chemotypic variability. There are several methods that can be used to isolate the essential oils i.e. cold press extraction, extraction of a single oil with another, steam or hydrodistillation, simultaneous distillation coupled with solvent extraction, supercritical fluid extraction, microwave extraction, solid phase extraction and fluorocarbon extraction (Nakatsu et al., 2000). For these studies isolation was standardized by hydrodistillation as described in Chapter 2.2. This is the most common and best suited method for the isolation of volatile constituents in plant-based studies (Guenther, 1948). As recommended, distilled oils were stored at 4 ?C, in airtight amber vials to prevent alteration of composition (Burt, 2004). Traceability of oil samples to voucher numbers and oil yields were documented in an established reference book. 199 11.3 Equipment All equipment i.e. from laminar flow benches, balances to micropipettes were calibrated according to standard operating procedures. This is imperative for good laboratory technique to ensure accuracy and reproducibility of results. Documentation supporting this is housed within the Department of Pharmacy and Pharmacology, University of Witwatersrand. 11.4 Chemical evaluation of essential oils A review paper (Lahlou, 2004) on techniques used in the investigation of the bioactivity of essential oils lists the analytical requirements that should be adhered to when performing investigations on essential oils. Parameters such as equipment specifications, column type and dimension, carrier gas-flow rate, temperature of column and detector should be specified. Specifications for these studies have been documented in Chapter 2.5.1. 11.5 Effect of solvents, media selection and additions A number of different solvents may be used for initial sample dilution preparation. The choice of an adequate biological solvent is crucial and examples include methanol, acetone, dimethyl sulfoxide (DMSO), and ethanol, amongst others. As many of these solvents are antimicrobial in nature they should be avoided if possible. Unfortunately studies on certain extracts show poor solubility in acetone as was the case with the T. camphoratus study (Chapter 9). Therefore a solvent such as DMSO which has a broad polarity range was used to facilitate dissolution of extract. For such studies, controls should be run of the solvent independently to determine the antimicrobial end points of the solvent. This was undertaken in the T. camphoratus study and DMSO showed inhibition between 4-8 mg/mL depending on the pathogen studied. Any antimicrobial activity within this range obtained with the T. camphoratus extracts were excluded, as activity could not be categorically attributed to the extract and may be as a result of the solvent. When investigating bio-activity, acetone should be used wherever possible as the solvent of choice, due to its inert properties and miscibility with water (Eloff, 1998b). The use of acetone for antimicrobial susceptibility testing is widely accepted as the solvent of choice and supported by over 444 references when performing a database literature search. 200 11.6 Reference standards All assays must include conventional antimicrobial controls that are appraised under identical laboratory conditions to that of the assay. These confirm that the micro-organism is susceptible to the selected antimicrobial. In the literature review by Lahlou (2004), a number of references are cited where ciprofloxacin is recommended as one the ideal antibacterial controls and similarly amphotericin B is recommended for control testing against fungal organisms. Antimicrobial controls were assessed with every test procedure and evaluated against an acceptable standard (Chapter 2.5.2). 11.7 Pathogen selection Many plants have reported use for the treatment of cuts, wounds, sores, ear infections, respiratory disorders, gastrointestinal upsets etc. indicating a possible bacterial or fungal infectious agent. In this current study the pathogen selection was based on the traditional use of the plant as determined from the ethnobotanical literature. Studies on the medicinal use of the plants were restricted to infectious test organisms. Even though a number of other pathogens exist, which may offer pertinent information for specific diseases, for this study a standard set of pathogens relevant to the ethnobotanical uses of the plant have been used. This would also allow for standardization of growth requirements. The culturing of many other pathogens involves either the addition of blood i.e. the Streptococcus species, provision of growth supplementation as in the case of Mycobacterium species and may even require enrichment of the incubation atmosphere with the addition of 5% CO2 as in the case for Lactobacillus species. An overall pathogenesis of all the test organisms studied and their relative importance as infectious agents are given here so that pathogen selection based on the ethnobotanical use is clarified. 11.7.1 The Gram-positive test organisms Staphylococcus aureus: Even though an opportunistic pathogen residing in the upper respiratory tract, S. aureus is one of the most virulent bacterial pathogens known to man-kind. The organism is responsible for superficial skin diseases (pustules, boils, carbuncles, abscesses, pimples and impetigo), systemic infections such as food poisoning and respiratory diseases causing pneumonia. This organism is commonly associated with bloodstream infections responsible for endocarditis, soft tissue infection and osteomyelitis (Rubin et al., 1999). A review of nosocomial 201 cardiac infections (Giamarellou, 2002) cites S. aureus as one of the major causative pathogens of bacteraemia responsible for 52-57% infections. The rate of increase of such infections has increased exponentially since 1980, escalating by 122% to 283%, depending on the hospital studied. With the ever increasing use of conventional antimicrobials such as methicillin (Hanberger et al., 2001) and vancomycin (Gibbons, 2004), resistance has emerged requiring a more aggressive search for newer anti-staphylococcal etiological agents. The emergence of methicillin resistant S. aureus (MRSA) strains have become a global problem where at certain epidemiology surveillance centres it is now referred to epidemic methicillin resistant S. aureus (EMRSA). Acquisition of MRSA carries increased healthcare costs and seriously limits effective antimicrobial therapy (Bannister et al., 2000). With the unsuccessful attempts to control multiresistant strains of S. aureus with conventional antimicrobials, it is no wonder that the search for newer antimicrobial sources from natural products has become increasing popular. Staphylococcus epidermidis: Of the many infections transmitted, those entering via the skin usually occur due to natural immune mechanisms becoming suppressed or through abrasions to the skin barrier. The most common infections are primary and associated with localized inflammation and invasion of the pathogen, however, because of the skin?s contact with the environment, it is usually the first port of entry where infections are spread systematically throughout the body. Staphylococcus epidermidis is commonly isolated from the skin but also resides in the upper respiratory tract, lower intestine, urethra and vagina. Although commensal, recent medical invasive therapies (such as the insertion of intravascular cannulae) have allowed this organism to enter the blood stream of immunocompromised patients, resulting in the organism becoming pathogenic (Bannister et al., 2000). In a study on nosocomial infections (Giamarellou, 2002), S. epidermidis was featured together with S. aureus as the most common infectious agent isolated from ICU patients. It is well-known that S. aureus is resistant to a number of conventional antimicrobials and it is only a matter of time before we see the same thing occurring with S. epidermidis. Bacillus cereus: Not only is this organism associated predominantly with gastrointestinal infections but there have also been reports of ocular infections, infections from wounds, pneumonia and infections in the urinary tract (Kotiranta et al., 2000). An outbreak of a respiratory tract infection associated with a B. cereus strain has been documented by Gray et al. (1999), where the authors stated that such outbreaks are not uncommon. A review of the organism B. cereus classified infections into six groups i.e. local (including burns), bacteremia, 202 systemic (including meningitis), respiratory, endocarditis and gastrointestinal (Drobniewski, 1993). This spore bearing organism has been associated with an increasing number of food- borne diseases and given the ever present rise in such outbreaks, especially in rural areas where poor hygiene and a lack of adequate education is present, it would be ideal to use endemic plants as a prophylaxis to prevent such infections occurring. Enterococcus faecalis: This pathogen is a regular inhabitant of the intestinal tract and often used as an indicator of faecal pollution. The species has been implicated in urinary tract infections, abdominal lesions and endocarditis (Boyd and Hoerl, 1981). In recent years the organism has emerged as a significant antibiotic resistant nosocomial pathogen, developing resistance to both aminoglycosides and vancomycin (Giamarellou, 2002). Vancomycin-resistant E. faecalis (VRE) has acquired specific resistant genes enabling the organism to withstand vigorous treatment protocols (Walsh and Amyes, 2004). Recently, the added concern that VRE strains could exchange genetic material with MRSA has resulted in pharmaceutical companies increasing their efforts to identify and produce compounds that would be active against these multi-resistant strains (Rice, 2006). Bannister et al. (2000), reports that over 25% of nosocomial infections resulting from surgery involving intravascular cannulae were attributed to E. faecalis species. Bacillus subtilis: Even though rarely pathogenic, many ethnobotanical antimicrobial publications include this organism in their test organism data set, possibly due to the fact that this spore bearing Gram-positive organism may yield common inhibitory trends with other Gram- positive pathogens. Kalemba and Kunicka (2003) specified that this organism is one of the most common test strains when investigating antimicrobial efficacies against essential oils. There have been occasional reports of food poisoning associated with the toxins produced by B. subtilis (Bannister et al., 2000), but this is rare. 11.7.2 The Gram-negative test organisms Klebsiella pneumoniae: Even though K. pneumoniae forms part of the commensal flora of the intestine, the organism becomes an infectious agent when inhabiting the upper respiratory tract causing pneumonia. The pathogen has been identified as one of the major causes of septicaemia in paediatric wards (Boyd and Hoerl, 1981). Fatality is as high as 90% in untreated cases and the severity was highlighted with the recent outbreak of K. pneumoniae infections in South African neonatal hospital wards where the deaths of nine babies were reported (Green and Rondganger, 203 2004). Klebsiella associated meningitis, endophthalmitis, cholangitis, endocarditis and urinary tract infections, even though uncommon have also been noted (Bannister et al., 2000). The emerging antimicrobial resistance to ampicillin and amoxicillin shows the need to develop more resilient means of preventing Klebsiella nosocomial infections. Pseudomonas aeruginosa: This organism can invade virtually any living tissue and is frequently responsible for a wide spectrum of severe infections (Hanberger et al., 2001). The growth requirements for P. aeruginosa are minimal and thus often found in inappropriately stored water. This opportunistic pathogen is responsible for a vast number of nosocomial infections in particular that of burn patients, pneumonia, dermatitis, open wounds and otitis externa where a high degree of antibiotic resistance has been reported. The organism has been associated with nosocomial cardiac infections (Giamarellou, 2002) as well as infections of the lungs with a death mortality of over 80%. Pseudomonas aeruginosa can cause keretoconjunctivitis where infections have derived from non-sterile ophthalmic cleaning fluids (Bannister et al., 2000). Multi-drug treatment regimens are often necessary to eradicate this organism. Pseudomonas species together with MRSA have been identified as the most prominent organisms that are rapidly becoming resistant to all existing antibiotic agents (Rice, 2003). Moraxella catarrhalis: One of the most common bacterial agents responsible for acute sinusitis. This opportunistic pathogen may also be responsible for bronchitis and pneumonia in the lower respiratory tract (Laurans and Orfila, 1991). The incidences of M. catarrhalis infections have resulted in this organism being a well-established pathogen, presenting itself in a number of respiratory infections as well as septicaemia and meningitis. The organism resides in the upper respiratory tract and has become the most common bacterial species isolated from patients presented with laryngitis (Verduin et al., 2002). Because this organism often co-invades with other infectious agents future treatment developments should include targeting poly-microbial infections. Escherichia coli: While E. coli is a consistent resident of the small intestine, some strains are pathogenic, causing intestinal and urinary tract infections as well as neonatal meningitis. Most infections are usually food-borne or spread from person to person via the faecal-oral route (Bannister et al., 2000). Their role in human pathogenesis has only recently been appreciated, due to the distinguishing difficulties between typing pathogenic and commensal strains. Antimicrobial resistance towards E. coli has already been reported worldwide. Enterotoxigenic E. coli is presently recognized as the primary cause of diarrhoea in infants and adults, 204 particularly associated with traveller?s diarrhoea. A recent publication by Qadri et al. (2005), strongly recommends that serious efforts be required for management of E. coli infections. In addition, the widespread misuse of veterinary antimicrobials has resulted in the introduction of antibiotic resistant E. coli strains into the endogenous microflora of animals. Resistant genes are then transferred to human strains where resistance proliferates (von Baum and Marre, 2005). Yersinia enterocolitica: Transmission occurs through contaminated food, milk and water where Y. enterocolitica will cause gastroenteritis accompanied by severe abdominal pain (Bannister et al., 2000). Another means of transmission is the result of animal-to-human contact. The organism is highly acid resistant, contributing to its survival and virulence in the stomach (Koornhof et al., 1999). Infections arising from this organism may also cause enlarged lymph nodes, subcutaneous nodules and septicaemia, with fatality rates higher than 50% (Boyd and Hoerl, 1981). Salmonella typhimurium: Infections associated with Salmonella are a global problem and it has been estimated that over 16 million cases are reported yearly (House et al., 2001). Consummation of infected poultry is one of the most common sources of salmonellosis where up to 60% of all incidences occur. Salmonella food poisoning and colitis usually result from exposure and occasionally there have been reports of Salmonella-related infections in the bones, joints and soft tissue, indicating blood-borne infections (Bannister et al., 2000). Disease is spread by person to person contact via the faecal-oral route. Many countries have experienced an unprecedented rise in salmonellosis and in South Africa, nosocomial death mortality is on the rise due to outbreaks of S. typhimurium (personal communication: personnel from the Reference Centre for Enteric Disease, NHLS). Proteus vulgaris: This commensal gastrointestinal organism is often associated with bladder infections where there is inappropriate urine release resulting in chronically stagnant bladder reservoirs (Bannister et al., 2000). The most frequently isolated human pathogen, P. vulgaris, is often acquired in hospitals where transfer has been associated with inadequate hygienic protocols in the manipulation of catheters (Boyd and Hoerl, 1981). Serratia odorifera: This micro-organism is free-living and capable of causing sepsis and urinary tract infections. Since the emergence of antibiotic resistance Serratia spp. infections have become increasingly prevalent. The organism requires very little growth stimulant and has been 205 identified in hand-washing solutions from hospitals. The main route of infection is person to person contact with immunocompromised hospital patients (Boyd and Hoerl, 1981), however, the natural reservoir for S. odorifera is unknown (Lee et al., 2006). A fatal outbreak of S. odorifera occurred in neonatal wards in South Africa in the early nineties where contaminated parenteral intravenous solutions resulted in a number of infant deaths (Frean et al., 1994). 11.7.3 The fungal test organisms Cryptococcus neoformans: This opportunistic pathogen can fatally infect those with compromised immune systems. Airborne spores enter the lungs where infection becomes asymptomatic until entry into the upper respiratory tract where pathogenesis is presented as Cryptococcus meningitis. Buchanan and Murphy (1998) cited a number of references supporting the emerging pathogenicity of C. neoformans resulting in 100% fatality for cryptococcal meningoencephalitis. Recently, Cryptococcus infections have become a problem in HIV patients (Bannister et al., 2000). In immunocompromised individuals, the infection may disseminate systemically causing infection of the central nervous system. Candida albicans: This yeast is part of the normal microbial flora of the skin, mucosa and bowel. A fine balance with other flora is achieved in healthy individuals, however, during immunocompromised conditions or when antibiotic treatment is administered the balance is affected allowing for the growth proliferation of this organism. Moist skin exacerbates candidiasis resulting in superficial skin infections and inflammation at the site of infection. Cutaneous infection is commonly associated with nappy rash and genital infections (Bannister et al., 2000). Most candidal infections are effectively treated with antifungal agents such as nitroimidazoles, nevertheless the frequent use of potent antibiotics has often resulted in the over colonization of C. albicans. Over colonization often leads to mucosal infections of the mouth and throat, intestinal overgrowth and invasive persistent systemic infections. Candida related nosocomial infections have recently emerged and in a review by Giamarellou (2002), 73% of fungal infections resulting from prosthetic valve endocarditis were attributed to Candida spp. Candida tropicalis: The number of reported infections associated with Candida species other than albicans has increased in incidence in the last two decades. In a report on Candida epidemiology surveillance, C. tropicalis was found in 16% of clinical Candida associated cases (Weinberger et al., 2005). In another study on candidemia patients, C. tropicalis was the third 206 most isolated Candida species isolated from ICU patients resident in Brazilian tertiary care hospitals (Colombo et al., 1999). The pathogeneses is similar to that of C. albicans and has been associated with nosocomial cardiac infections (Giamarellou, 2002). Aspergillus niger: Aspergillosis infections usually reside in the external ear and lungs. Because of the characteristic fungal nature of spore formation and preference for damp conditions, this organism often resides in buildings and ventilation systems which may manifest in clinical respiratory infections associated with sick building syndrome. Lung infections may become difficult to treat, especially when colonizing immunosuppressed patients. If no response to the recommended treatment by amphotericin B occurs, surgery may become necessary to excise the infected lung segment (Bannister et al., 2000). Colonization of the organism in the cerumen of the ear may eventually lead to loss of hearing (Boyd and Hoerl, 1981). Alternaria alternata: Although considered non-pathogenic, allergic responses are common when this organism has come into contact with the respiratory system. Airborne fungal spores enter the upper respiratory tract by inhalation where they may contribute towards allergic fungal sinusitis. Alternaria alternata spores are larger than other fungal spores and thus are more likely to be deposited within the mucus of the upper airway. A study by Shin et al. (2004), demonstrated a five fold increase in allergic immune response when exposed to A. alternata. The spores can also be a contributing factor in sick building syndrome. 11.8 Inoculum For cultures, the strain number should always be specified (Paull and Marks, 1987; Janssen et al., 1987). In many papers, this has been neglected (Izzo et al., 1995; Rabe and van Staden, 1997; Natarajan et al., 2005). Wherever possible reference strains should ideally be used, as clinical strains, isolated from patients may have been exposed to conventional antimicrobials, thus having the potential to have devised a means of increasing resistance. The use of microbial culture reference standards has been standardized internationally. This allows for direct assay comparison whereas clinical strains are not universally available and could yield results that may differ from reference strains. The importance of standardizing the inoculum size is discussed in a number of publications (Denyer and Hugo, 1991; Pauli and Kubeczka, 1997; Lambert, 2000). The general findings were that with the increase of inoculum size the MIC decreases. The recommended inoculum size for 207 assays testing antimicrobial agents should be standardized by visual inspection with a 0.5 McFarland suspension where the density is approximately equivalent to 1 x 104-8 CFU/mL or alternatively adjusted photometrically (Klepser et al., 1998; ESCMID, 2000; NCCLS 2003; Andrews, 2004). As the latter is based on turbidometric techniques, one must take into account that photometric readings cannot always predict viability of the test organism and even though readings may be standardized, if assays are not undertaken promptly, changes in broth density will occur due to exponential growth of the test organism. In order to verify the standardization of inoculum when investigating biological activities of plant preparations, a selection of essential oils and commercial antibiotics were assessed with various different inoculum ranges. Eleven concentration ranges (0.1 ? 30 mL) were prepared where a 24 hr growth of test organism (S. aureus, K. pneumoniae and C. albicans) was diluted with 100 mL Tryptone Soya broth. The MIC was determined for the two essential oil samples O. asteriscoides and H. natalensis (Table 11.1), as well as ciprofloxacin and amphotericin B (Table 11.2) where various inoculum sizes were introduced. Both, Table 11.1 and Table 11.2, summarise the MIC data obtained for the 11 concentration ranges as determined in triplicate and thus present the data as a range. It was interesting to note that the essential oils showed little variation in MIC value with increasing inoculum size and only slight variation was noted for the H. natalensis essential oil when investigating bacterial strains, however, with the yeast C. albicans, some variation was noted (Table 11.1). The MIC ranges for the conventional antimicrobials ciprofloxacin and amphotericin B indicated a substantial variation with increasing inoculum and thus extremely dependent on inoculum size (Table 11.2). This variation, although not predominantly noted with the essential oil samples, validates the need to standardize results and to keep in line with the literature recommendations. All assays undertaken in this thesis were thus adjusted according to the 0.5 McFarland standard. MIC range (mg/mL) Plant sample S. aureus K. pneumoniae C. albicans H. natalensis 8-16 8 8-32 O. asteriscoides 4 8 4-16 Table 11.1 The MIC (mg/mL) for oil samples when tested with a varying inoculum density ranging from 0.1-30 mL:100 mL (24 hr culture: broth ratio). 208 * Not tested, as amphotericin B is an inappropriate antimicrobial for the test organism. 11.9 Disc diffusion assay The principle of disc diffusion assay is based on the outward diffusion of a growth inhibiting test substance into an agar base, which has been suitably inoculated with a pure culture of test organism. The plates are allowed to pre-diffuse at 4 ?C for one hour allowing for dispersion of the test substance from the disc into the surrounding agar. Thereafter optimum incubation conditions follow which is predetermined depending on the selected test organism. Results are determined by measuring the clear zone of inhibition contrasting with the remainder of the agar having microbial growth (Hewitt and Vincent, 2003). For more detail of the method refer to Chapter 2.5.2. A number of critical factors are involved in this assay and may affect the outcome. These are the following: Measurement of the zone diameter: The simplest method of measurement is achieved with hand-held vernier callipers, where measurement is made to the nearest mm. The use of a light- box aids with visualization of zone size determination, however, in numerous publications much variation exists in obtaining an accurate measurement. For some publications (Hinou et al., 1989; Brantner and Grein, 1994; Kim et al., 1995; Mangena and Muyima, 1999), measurement of the inhibition zone includes the entire diameter including the disc (Figure 11.1). When no inhibition is noted, the result is presented as the size of the disc (6 mm). This is a relatively large area of ?no-activity? and may become very difficult when comparing samples having slight activity i.e. 7/8 mm, as the actual inhibition is less than the size of the disc. In addition, one sees a negative result having a positive value. Other authors (Eksteen et al., 2001; Lourens et al., 2004; Njenga et al., 2005) make reference to only the distance from the disc edge to the area of MIC range (?g/mL) Control S. aureus K. pneumoniae C. albicans Ciprofloxacin 0.3-1.9 0.3-2.5 * Amphotericin B * * 2.5-18.9 Table 11.2 The MIC (?g/mL) for controls when tested with a varying inoculum density ranging from 0.1-30 mL:100 mL (24 hr culture: broth ratio). 209 growth (Figure 11.2). This is a more accurate means of measurement, as the disc is not taken into account. It should be assumed that the active antimicrobial agents have diffused from the disc during the pre-diffusion period and thus should not be considered as an inhibitory area. Some publications, however, have not mentioned how the zone size has been determined (Akinpelu and Olorunmola, 2000; Dorman and Deans, 2000; Ezeifeka et al., 2004; Natarajan et al., 2005) and such discrepancies can greatly influence comparative literature evaluations. All disc diffusion assays undertaken in this thesis are standardized to measurements from disc edge to the area of microbial growth. Choice of agar: The selection criteria for a growth medium for bioassays should be based on the nutrient requirements of the test organism. Pauli and Kubeczka (1997), in their overview of microbiological testing methods list 15 different media that have been used in the assessment of essential oils. A recent study (Moon et al., 2006) compared the efficacy of results achieved with two different media, IsoSensitest agar and Nutrient agar. No generalization concerning the superiority of one test medium over the other could be established. Mueller-Hinton media, however, is the recommended choice for bacterial assays (NCCLS, 2003). Both bacterial and fungal cultures were assayed in this thesis and the growth requirements for fungi differ from that of bacteria. As Mueller-Hinton media is not suitable for fungal culturing, Sabourauds Dextrose agar was selected. In later studies, to keep variables consistent, Tryptone Soya agar was used for both bacteria and fungi. In addition, this versatile medium with high nutrient value was selected due to its ability to produce a higher density of inoculum. In this way a more accurate definition between presence and absence of growth could be observed. Figure 11.1 Figure 11.2 Agar with seeded test organism Agar with seeded test organism Figure 11.1 and 11.2 Agar with seeded test organism where the diameter of the zone of inhibition is measured (Figure 11.1) and where the radius of the zone of inhibition is measured (Figure 11.2). 210 Disc size and sample reservoir: The volume and concentration of test sample placed on the disc is important and should be consistent for all repetitions and between all test organisms (Hewitt and Vincent, 2003; Janssen et al., 1987). Standardization of disc size to 6 mm has been undertaken in all disc diffusion studies and is recommended to ensure even absorption of test sample. Diffusion: Even though disc diffusion methodology is a quick simple means of screening for antimicrobial activity, problems may arise associated with the hydrophobic nature of the samples assayed. Depending on the nature of the particular sample i.e. extracts and essential oils may differ and diffusion rates may result in inconsistent results. Essential oils do not easily diffuse through the agar and even with the pre-diffusion time allocation of one hour (Hewitt and Vincent, 2003), false negatives may still be encountered and the possibility of activity could be underestimated. Another prominent factor to be considered is the volatility of the oil samples. Depending on the incubation time a proportion of the oil is inclined to be lost due to evaporation and this too may impact on false negative results. This is particularly true for fungal test organisms due to their long incubation periods of 48 hr - 7 days (Janssen et al., 1987). It has been suggested (Southwell et al., 1993) that the hydrocarbon components are particularly prone to evaporation and thus one tends to evaluate antimicrobial activity against only a proportion of the essential oil compounds found in the whole essential oil sample. Such false negative results may become problematic, particularly when comparatively evaluating methodologies. This was encountered in a number of instances in this thesis and is addressed further in Section 11.13 (Method variation). 11.10 Minimum inhibitory concentration (MIC) The MIC measurement to determine antimicrobial activity is a quantitative method based on the principle of contact of a test organism to a series of dilutions of test substance. The assay determines the lowest concentration of antimicrobial agent that inhibits the growth of the test organism (Bannister et al., 2000). Assays involving MIC methodology are widely used and an accepted criterion for measuring the susceptibility of organisms to inhibitors (Lambert and Pearson, 2000). Although current methods are straightforward, to obtain useful comparative information from tests is a challenge as factors such as inoculum size, incubation temperature, reading of results, growth medium and assay technique can greatly affect the results (Christofilogiannis, 2001). Three main references were used to carry out the MIC assays: 211 ? The NCCLS (2003), used for obtaining an adequate accepted inoculum range. ? Eloff (1998a), suitable because the method has been refined for plant-based studies. ? Carson et al. (1995), suitable as the method takes into account essential oil volatility. Due to the volatility of oil samples all microtitre plates were sealed with sterile adhesive sealing film. This ensured no loss of volatile actives. Inouye et al. (2001b) demonstrated the increased effect observed when sealing agar plates during agar dilution MIC assays. With this principle in mind, preliminary observations were undertaken on the MIC micro-well methodology where sterile sealing tape was placed on microtitre plates and comparatively evaluated with unsealed plates. Table 11.3 confirms the increased efficacy when the microtitre plate was sealed. With all three plant samples (L. javanica, T. camphoratus and C. gratissimus var. subgratissimus), where the essential oils and extracts were investigated, the MIC values were lowered with sealing. The results for ciprofloxacin (non-volatile) did not indicate any variation between sealed and unsealed areas. Selection of growth medium: As with disc diffusion assays, Tryptone Soya medium was selected for most of the MIC studies as it is considered the most suitable medium for the growth of all test organisms. Studies on medium comparison were undertaken where the MIC?s were performed in both Mueller-Hinton and Tryptone Soya broth under the same test conditions using the same test samples. Results indicated no difference in the outcome. Thus, to standardize procedures, it was decided to perform all MIC assays with Tryptone Soya medium. Table 11.3 The mean MIC (mg/mL) for plant samples exposed to K. pneumoniae in microtitre plates that have been sealed in comparison to the uncovered microtitre plates. Plant sample Sample type Uncovered (mg/mL) Sealed (mg/mL) Essential oil 16 12 L. javanica Extract >32 6 Essential oil >32 12 T. camphoratus Extract 12 4 Essential oil >32 4 C. gratissimus var. sub- gratissimus Extract >32 8 Ciprofloxacin control 0.2 x 10-3 0.2 x 10-3 212 Reading of results: The use of tetrazolium salts as an indicator for biological activity has been successfully employed for many years (Begue and Kline, 1972). Its application for end point determination of MIC readings in natural product research has been widely accepted (Hamburger and Cordell, 1987; Brantner et al., 1994; Eloff, 1998a). p-Iodonitrotetrazolium violet (INT) forms a red-purple dye on reduction i.e. in the presence of biological growth and thus when added to all wells within a microtitre plate, binds to living cells making end point MIC visualization more clear, however, in spite of the numerous references to microtitre MIC methods incorporating INT, no recommended time between the adding of indicator and reading of results could be clarified in literature. This factor is crucial as results may differ completely when read at different times. Practical experimentation demonstrated that if results were read within the first three hours of adding INT a much lower MIC value was obtained, however, this is not ideal, as the indicator salts have not had a full colour development and for this reason the reading of all MIC values was standardized at six hours after adding INT for bacterial test organisms. Continuous repetitions and replications confirmed that this time allocation resulted in a more reliable reading with full development of colour and minimum change of MIC. Ciprofloxacin did not show any variation in the MIC readings when observed periodically within the six hour colour development. For yeast cultures, an INT exposure time was increased to 24 hours as colour change developed at a much slower pace, possibly due to the delayed growth of yeasts. This was evident for both plant samples and the conventional antifungal control amphotericin B. The reporting of MIC results for pharmacognosy-based studies differs somewhat from conventional MIC reports with known MIC breakpoints. To elaborate, one needs to understand how the clinician interprets data obtained from MIC testing. For each conventional antibiotic there is an acceptable breakpoint range which has been predetermined by susceptibility criteria and quality control (NCCLS, 2003). Thus MIC determinations on known antibiotics can be effectively compared to the recommended categories. Results are then categorized as either susceptible (an appropriate antimicrobial to treat infection), intermediate (where response may be lower) or resistant where the MIC values depict possibility of resistance arising. With pharmacognosy studies however, there are no comparative breakpoints as the antimicrobial activities are reported for test substances that have not been previously examined and used for therapeutic purposes. The activities found for plant-based assays are much higher then conventional antimicrobials. Activities of 8 mg/mL are considered acceptable (Fabry et al., 1998) and many literature citations refer to natural products having activities over 1 mg/mL 213 (Gibbons, 2004). This is in contrast with commercial antimicrobials obtaining therapeutic susceptibilities in the lower ?g range and should not be used as a direct comparison. The control antimicrobials are thus only included as a reference to monitor the response of the test organism to a known antimicrobial. Much debate has arisen over the use of statistical methods, particularly standard deviation to report MIC results. Standard deviation is defined as the square root of variance and is used to compare the variance between samples within a data set (Roscoe, 1995). Such criteria are only useful and an acceptable means of measuring activity when a large number of samples are included in the data set and the variance between samples is narrow. With MIC reporting, the methodology incorporates doubling dilutions and therefore the variance, if reported as standard deviation becomes so high that even one variation i.e. results 16 to 32 mg/mL, are represented as a very high probable variance. The reporting of an MIC value is hence not a true absolute point and the true value is somewhere between the lowest test concentration that inhibits the test organism and the next lowest test concentration (Opalchenova and Obreshkova, 2003). For this reason many studies using MIC methodology do not report standard deviation. The use of standard deviation only becomes a valuable statistical tool when comparably evaluated to acceptable MIC limits given by the NCCLS (2003) quality control. Here the degree of variance can be reported, as direct comparisons can be made to the acceptable breakpoint limits. Even though the statistical method of using standard deviation was not used to analyze data, the experimental procedures should always be repeated to confirm data, as well as replicated at different times to ensure reproducibility, as was undertaken with the current studies. The use of a dispersing agent: Due to the oil volatility and problems associated with miscibility of the oil in an aqueous based medium, the use of Tween has been suggested to facilitate an emulsive mix of test substance. In Remmal and Tanaoui-Elaraki (1993) and Delespaul et al. (2000), the use of Tween resulted in increased MIC values. Inouye et al., (2001b) reviewed a number of contradictory publications either supporting or rejecting the use of a dispersing agent in antimicrobial assessments. Practical application of the use of Tween (Saarchem) at 0.1-0.5% showed no difference in MIC values. To avoid interference of a dispersing agent on the biological system, the use of an emulsifier was excluded in the assays. 214 11.11 Bio-directed fractionation (bio-autographic assays) The principle of the method is based on the isolation of the test substance (whole plant and / or active compound) by basic thin layer chromatography conditions (Chapter 6.5.1.1). The test compounds are then allowed to diffuse into a suitable agar medium containing a test micro- organism and incubated at optimum incubation temperatures. Antimicrobial compounds are thereafter detected by the lack of microbial growth in the vicinity of the test compound. Test compounds to be isolated are preferentially obtained from non-volatile constituents rather than essential oils, as the volatile constituents are often lost during incubation, leading to poor detection of inhibition. A number of variations of this method have been used to determine antimicrobial activity by target-directed isolation of antimicrobial compounds. The method most commonly used is where the selected test culture is mixed with suitable growth medium and sprayed onto a TLC plate. This was not selected for this study as aseptic conditions are difficult to maintain and with spraying, an aerosol of culture is created thus potentially contributing to contamination if multiple assays are being performed. This direct method also has the disadvantage of media drying out which would ultimately result in inconsistent inhibition zones. Similar inconsistencies would arise if the agar-culture spray mix is not of adequate uniform thickness. The method best selected, having the least amount of variables, was the over layer method (Chapter 6.5.1.1). This method is applicable to a wide range of micro-organism, produces well defined zones of inhibition and is not susceptible to possible contamination (Marston and Hostettmann, 1999; Hostettmann, 1999). 11.12 Time-kill analysis Time-kill studies provide descriptive information on the relationship between bactericidal activity and the concentration of test substance (Tam et al., 2005). Even though the methodology is labour intensive and requires a number of steps where variables may be introduced, the method gives valuable information of the cidal action over time. A number of considerations need to be taken into account when analyzing time-kill data. The type of organism studied may impact on the results observed. This was particularly noted for studies on P. aeruginosa where regrowth after initial reduction in CFU/mL was encountered. Previous studies have postulated that the regrowth phenomenon be attributed to preferential killing of sub-populations, however, this could not be substantiated in further studies as indicated by Tam et al. (2005) who attributed 215 regrowth more to the adaption of the test micro-organism to the environment containing test antimicrobial. Tam et al. (2005) stated that the constant exposure to a test antimicrobial may result in a more resistant mixture of bacterial populations which become increasingly difficult to kill over time. This substantiates results where there is an immediate reduction in CFU/mL over time followed by regrowth. This was noted in time-kill essential oil studies with M. flabellifolius (Chapter 2) and to a lesser extent O. asteriscoides (Chapter 3) when investigated with P. aeruginosa. This organism is known to produce multiple emerging resistant strains (Livermore, 2002) and time-kill studies have shown that even with conventional antimicrobials such as meropenem, regrowth is observed (Tam et al., 2005). The regrowth effect may also be attributed to a concentration dependent dose, as noted in a death kinetic study on C. gratissimus var. subgratissimus. As limited essential oil was obtained, time-kill assays were employed for only one pathogen (K. pneumoniae). Selection was based on the moderate activity obtained in the MIC assay and the traditional use of C. gratissimus to treat coughs, colds and inflammation of the lungs. The essential oil, when exposed to K. pneumoniae, indicated initial bacteriostatic activity (Figure 11.3, not shown in Chapter 7) for all three concentrations (0.5%, 0.75% and 1%). The lowest concentration studied (0.5%) showed a 99.9% reduction in CFU within the first four hours, however, after eight hours regrowth occurred exponentially over the 24 hr test period. Both the 0.75% and 1% oil concentration indicated an immediate drop in CFU and within one hour bactericidal action was obtained and maintained for the full 24 hr test period. These results indicate that essential oil concentrations for C. gratissimus var. subgratissimus should be greater than 0.5% in order to achieve prolonged efficacy. 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U /m l Control 0.5% 0.75% 1% Figure 11.3 Death kinetic studies of K. pneumoniae exposed to the essential oil of C. gratissimus var. subgratissimus over 24 hr. 216 One must also take into consideration that in vivo pharmacodynamics of conventional antimicrobial drugs relies on repeated doses to achieve complete cidal activity. Should this principle be applied to time-kill analysis, one could extrapolate that a second dose provided with an extended time-kill may result in a cidal outcome. 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U/ m l control 0.13% 0.25% 0.5% 0.75% 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 1.E+10 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hrs) CF U/ m l control 0.25% 0.5% 0.75% 1% The presence of spores may also affect the results. Bacterial spores are the most resistant of all cell forms to bacterial agents and only with vigorous destructive methods are spores inactivated. This was noted in the study on P. grandidentatus. The MIC data for B. cereus demonstrated Figure 11.5 Death kinetic studies of K. pneumoniae exposed to the essential oil of H. natalensis over 24 hr. Figure 11.4 Death kinetic studies of B. cereus exposed to the essential oil of P. grandidentatus over 24 hr. over 24 hr. 217 relatively high sensitivity (3 mg/mL) for the essential oil. Thus, a theoretical decision was made to perform a time-kill study to assess the pharmacodynamic effect over time. It came to light upon examination of the results (Figure 11.4) that the poor reduction of CFU/mL over time was due to the fact that this organism possesses spores which re-germinated in spite of the possibility of the vegetative cells been inactivated. Previous research on disinfectant evaluation with B. cereus has shown increased resistance of this species due to the presence of spores (Lindsay and von Holy, 1999; Lindsay et al., 2002). Even though time-kill studies are a valuable means of determining the inter-relationship between test essential oil and selected test pathogen, there are certain circumstances where studies may not yield informative data, as was the case with a study on a collective sample of H. natalensis essential oil where death kinetics performed on K. pneumoniae indicated little reduction in CFU/mL over time (Figure 11.5). These results do correlate with the moderate MIC data (seasonal mean of 8.8 mg/mL), but the time-kill assay gives a more detailed account of the pathogens immediate and prolonged response with exposure to the test oil. 11.13 Method variation Although the antimicrobial activity of essential oils and their constituents are well-known (Yousef and Tawil, 1980; Hinou et al. 1989; Pattnaik et al. 1997; Hammer et al., 1999), the methods used to assess such activity may result in varied reporting of results. In a number of studies, observed from both publications and determined through experimental procedures, variability between results was demonstrated. The variability and problems associated with the disc diffusion assay have been well documented (Janssen et al., 1987; Pauli and Kubeczka, 1997; Njenga et al., 2005; van Vuuren et al., 2006). Variation between disc diffusion results and MIC data was observed to a certain degree for all studies incorporating both methods. This was noted with the study on A. afra (Chapter 4) where lower disc diffusion efficacies were observed for three of the 13 pathogens when a comparative evaluation was made with the MIC methodology. The L. javanica study (Chapter 5) also incorporating both disc diffusion and MIC methodology indicated better efficacies with the MIC assay for five of the 12 pathogens studied. For O. asteriscoides (Chapter 3), results for four of the ten pathogens were incongruent. The greatest variation was noted for B. subtilis where no activity was noted in the disc diffusion assay and moderate activity (8 mg/mL) detected in the MIC assay. The variation of results noted using different methodologies was greatest with the H. 218 cymosum subsp. cymosum study (Chapter 6). All ten pathogens showed no activity for the essential oils in the disc diffusion assay and only Gram-positive sensitivity for the extracts, however, when the same samples were tested with the MIC method, all pathogens showed activity for the essential oils and the crude extracts were at least six times more active than the oil. For a limited number of pathogens it was found that the disc diffusion methodology gave higher efficacies than the MIC methodology. This was evident for A. afra (Chapter 4) where B. subtilis and C. tropicalis had disc diffusion readings of 3 mm and 5 mm respectively. Less sensitive MIC values were obtained for B. subtilis (16 mg/mL with poor activity) and C. tropicalis (8 mg/mL with moderate activity). This was also noted for L. javanica where E. faecalis and C. tropicalis were more sensitive having disc diffusion values of 4 mm and 3 mm respectively and poor MIC values of 16 mg/mL and 32 mg/mL respectively. This was also noted in a study on Eriocephalus species (Njenga et al., 2005) where the disc diffusion results yielded better activities than MIC values. It is often difficult to compare results obtained from MIC assays with literature presenting disc diffusion results only. This was noted in a number of studies such as results obtained for A. afra (Chapter 4) compared to the investigation by Graven et al. (1992). This was also noted in the H. natalensis (Chapter 8) study where correlating data with the non-quantitative results by Gundidza et al. (1993). When comparing results with literature, the latter trend seems to support the use of quantitative MIC methodology for pharmacognosy studies rather than earlier disc diffusion antimicrobial assessments. This is observed by the number of earlier publications using disc diffusion methods (Hinou et al., 1989; Graven et al., 1992; Gundidza et al., 1993; Brantner and Grein, 1994; Kim et al., 1995; Mangena and Muyima, 1999) and later publications utilizing MIC methods (Negi et al., 2005; Gaspar-Marques et al., 2006). The use of the more quantifiable MIC assay techniques are presently the preferred method of antimicrobial assessment (Kalemba and Kunicka, 2003; Burt, 2004). 11.14 Validation of various methodologies As microbial cell viability is the key in determining antimicrobial activity, any reliable method of determining rate of growth or inhibition of growth can be used to assess antimicrobial activity. Diffusion assays, MIC?s, UV methods and time-kill analysis are all such methods used to 219 confirm antimicrobial activity. It has been noted that different methods (Tassou et al., 2000), may not always be comparable, however, the ideal is to use reliable and reproducible techniques that can yield consistent information. It has been shown (Chapters 3, 6 and 8) that results from literature do not always correlate with experimental conditions. Standardization of techniques are thus valuable in order to provide a common basis for comparison. This was achieved for time-kill assays, undertaken in two different laboratories to confirm reliability and consistency of results. Collaboration was undertaken with researchers at the University of Iowa whom have been involved with a number of studies reviewing the methodology of time-kill studies (Klepser et al., 1997; Klepser et al., 1998; Ernst et al., 1999). Time-kill assays were conducted using similar methods, same oil sample and same culture reference standards to determine the reliability and degree of variability. The results obtained for both comparative time-kill studies against M. flabellifolius (Viljoen et al., 2002) and O. asteriscoides (Viljoen et al., 2003) confirmed congruency. It was also shown how time-kill analysis and membrane integrity studies undertaken on O. asteriscoides correlated very well. Congruency between disc diffusion, MIC and time-kill methods was examined for all relevant Chapters and the greatest correlation was noted in the study with M. flabellifolius (Chapter 2) where five pathogens showed consistent results between disc diffusion and MIC methods. Furthermore death kinetics of S. aureus and P. aeruginosa were congruent with both disc diffusion and MIC methods. 11.15 Interactive combination studies The investigation of medicinal plants with a view to ascertain a single chemical entity for antimicrobial activity is becoming more and more improbable and research is now moving towards the investigation of a combination of substances to achieve efficacy. For plants to rely on a single compound in their biochemical warfare with pathogens would be equivalent to relying on the ?single golden bullet? approach and thus as researchers investigating the activity of single compounds, we would be ignoring the evolutionary approach that plants may have developed various metabolic mechanisms for the production of structurally and functionally diverse compounds to overcome emerging resistance. To ensure future success in natural product research we must incorporate interactive phytochemical studies with existing practices in the hope that developments may be used as a foundation and driving force for novel 220 chemotherapeutic agents. The problem arises with the standardization of methods and even in the classification of interactive procedures (Williamson, 2001; Lambert et al., 2003). The effect of combination studies may result in outcomes that are synergistic, antagonistic or additive. Synergism is evident where the combined action of test substances exceeds the effects of the individual components. Antagonism occurs where there is a reduced effect in combination. An additive effect occurs when the effect of the combination is equal to the sum of the components in combination (ESCMID, 2000). There are a number of different methodologies that can be utilized to express synergy or antagonism and different researchers have their preferred method of assessment. The time-kill method, has been praised as one of the best methodologies to study synergy (Beale and Sutherland, 1983) even though the methodology had been shunned by Berenbaum (1980). Further validation of the death kinetic method for synergistic studies was given by Acar (2000), where the method was praised from a clinical perspective. Time-kill studies have shown to be an effective procedure in demonstrating synergistic interactions between constituents of O. asteriscoides (Chapter 3, Viljoen et al., 2003). In this study the combined effect of the two major constituents, camphor and 1,8-cineole, indicated enhanced efficacy for C. albicans when in combination. Similar studies were conducted on A. afra (Chapter 4, Viljoen et al., 2006) where the major constituents, independently and in various combinations, were investigated to determine the role in the overall activity of the plant. Results indicate that all 11 samples (independently and in combination) when investigated at the concentration relative to that found in the whole plant, do not play a role in the overall activity of A. afra. In another study, time-kill methodology was used to validate the traditional use of A. afra when combined with L. javanica showing how the therapeutic properties may be enhanced when used in combination (Chapter 5). An algebraic method to determine synergy by means of the fractional inhibitory concentration index (FIC) has been defined in Chapter 7.5.2 where the roots, stems and leaves of C. gratissimus var. subgratissimus were studied independently and in combination. Even though the definition and equation for determining the FIC is a widely accepted means of measuring interaction, the interpretation has evolved. Table 11.4 demonstrates the refinement of measuring the FIC index, with the most recent reference being the most concise FIC interpretation. It was therefore decided that the most comprehensive and up to date definition (Schelz et al., 2006) would be used to classify synergy or antagonistic interactions. 221 * Values not given by author The isobole method of determining interaction is presently the favoured method of assessment (Williamson, 2001) and although complicated, gives a more accurate assessment of each agent when studied in various combinations. The method involves the combination of two samples at various ratios (Table 11.5), and determining the MIC value. The MIC for each sample in the concentration ratio is determined independently (Table 11.6). The MIC result for each sample in the combination is comparatively assessed against the MIC obtained for the independent sample and expressed as a dose ratio (Table 11.6). These ratios are then presented on an isobole graph. The adjoining line of the two axes (Figure 11.6) indicates the individual doses. It can be shown that one cannot always rely on a 1:1 combination and that depending on the ratio of the combination, differences in activity may be noted. This is demonstrated in a study where the volatile and non-volatile constituents of T. camphoratus were investigated with E. coli. Figure 11.6 in relation to Table 11.5 and Table 11.6 indicate where synergistic ratios were found (ratios 3-6, depicted in red). Antagonism was noted for ratio 8 (depicted in green) and an additive effect is noted for ratio 2 (depicted in brown). The novel isobologram representation of synergy thus presents three different interactions for the volatile and non-volatile constituents of T. camphoratus studied with E. coli, depending on the relative ratio in which the test compounds were combined. Interaction Reference Synergy Additive Indifference Antagonism Kerry et al., 1975 ?0.7 * * ?1.3 Berenbaum, 1978 <1.0 1.0 * >1.0 Krogstad and Moellering, 1986 ?0.5 * 1.0 ?2.0 ESCMID, 2000 ?0.5 >0.5-1.0 >1.0-<2.0 ?2.0 Mackay et al., 2000; Odds, 2003 ?0.5 * >0.5-?4.0 >4.0 Schelz et al., 2006 ?0.5 >0.5-1.0 >1.0-?4.0 >4.0 Table 11.4 The classification of the FIC index in accordance with the corresponding authors. 222 Additive Synergistic Antagonistic Ratio (%) Samples 1 2 3 4 5 6 7 8 9 10 11 Volatile 100 90 80 70 60 50 40 30 20 10 0 Non-volatile 0 10 20 30 40 50 60 70 80 90 100 Concentration in mg/mL MIC values Ratio values Row in plate X (volatile) Y ( non-volatile) X ( volatile) Y ( non-volatile) X (volatile) Y ( non-volatile) 1 16.0 0 8.00 0.00 1 0 2 14.4 1.6 7.20 0.80 0.900 0.200 3 12.8 3.2 3.20 0.80 0.400 0.200 4 11.2 4.8 2.80 1.20 0.350 0.300 5 9.6 6.4 2.40 1.60 0.300 0.400 6 8.0 8.0 2.00 2.00 0.250 0.500 7 6.4 9.6 3.20 4.80 0.400 1.200 8 4.8 11.2 2.40 5.60 0.300 1.400 9 3.2 12.8 1.60 6.40 0.200 1.600 10 1.6 14.4 0.80 7.20 0.100 1.800 11 0 16 0.00 4.00 0.000 1.000 E. coli 0.00 0.25 0.50 0.75 1.00 1.25 0.00 0.25 0.50 0.75 1.00 1.25 volatile n o n - vo la til e Table 11.5 The relative ratios of the volatile and non-volatile constituents of T. camphoratus. Table 11.6 The raw data for the volatile and non-volatile constituents of T. camphoratus. Figure 11.6 The isobole of the non-volatile constituents and volatile constituents combined in different ratios demonstrating the different profiles (antagonism, synergy and additive) dependent on the ratio studied (Table 11.6). 223 As seen with the comparison of disc diffusion and MIC methodologies, variation may also be noted when looking at different methods to assess synergistic and antagonistic interactions. This was highlighted by Odds (2003), wherein it was stated that reproducibility errors using MIC methodologies are common. Mackay et al., (2000) found good correlation of time-kill methods with calculation of FIC indices if methods were performed simultaneously. Up until recently the checkerboard MIC methodology and the calculation of FIC indices were the only methods adopted for synergy investigations and many studies (Ryan et al., 1981; Lambert et al., 2003) have focused on the correlation between the two. More recently e-tests have been incorporated in antimicrobial synergistic studies, however, these are only applicable to commercially available antimicrobials and have little relevance to plant-based studies. The use of isobolograms for antimicrobial combinations has been poorly explored in spite of its wide acceptance in other interactive pharmacological studies, such as its application in drug interactions, antiparasitic studies, antiviral combinations, toxicology interactions and numerous in vivo investigations. A literature search yielded very few (Codd and Deasy, 1998; Cedergreen et al., 2006; Kamatou et al., 2006) interactive antimicrobial studies using isobole methods. Recently, a mathematical 3-dimensional isobologram approach for the examination of antimicrobial agents was presented by Boucher and Tam (2006) validating isobologram construction as an effective tool for the examination of antimicrobial combinations. In the editorial by Odds (2003), researchers were encouraged to seek more sophisticated approaches to the measurement of synergy, to overcome limitations imposed by chequerboard methods. Thus interactive approaches should be further explored and validated with different assays to confirm interactions. With such validations in place, the justification and development of antimicrobial combinations could lead to patentable entities making research in the field of phytosynergy not only viable but also commercially beneficial (Lambert et al., 2003). 11.16 Summary The standardization of antimicrobial assay techniques based on literature reviews and practical suggestions may be summed up briefly as follows: ? There should be suitable identification and documentation for all plant collections (Janssen et al., 1987; Cos et al., 2006). 224 ? Analytical GC-MS data for all oil constituents should be included in microbiological analysis to assess structure-activity relationships (Lahlou, 2004). ? Ideally, all microbial strains should belong to a standard collection i.e. ATCC (Cos et al., 2006). ? More than one antimicrobial method should be undertaken in order to validate activity (Cos et al., 2006). ? Selection of the test micro-organism should be guided by the ethnobotanical use of the plant (Cos et al., 2006). ? Standardization of inoculum should be undertaken for all assays (Pauli and Kubeczka, 1997; Cos et al., 2006). ? Inclusion of appropriate controls in each test replicate i.e. conventional antimicrobials and test blank should be incorporated with all assays (Cos et al., 2006). ? The occurrence of pharmacological interactions should be explored (Burt, 2004; Verpoorte et al., 2005; Cos et al., 2006). 225 Chapter 12 General discussion, conclusion and further recommendations. 12.1 General discussion Antimicrobial pharmacological activities of aromatic plants consider many different parameters, depending on the focus of the study. The studies presented herewith have demonstrated a number of factors that should be considered when performing antimicrobial plant-based studies. Moreover, essential oil studies are mplex due to the heterogeneous nature of the oil. Components not only exhibit distinct molecular formula, have specific physiochemical properties but also interact synergistically or antagonistically within the mixture (Lahlou, 2004). This thesis has thus followed a structure-activity related approach into the antimicrobial study of some of the most widely used aromatic medicinal plants used for the treatment of infectious diseases. 12.2 Thesis summary Nine indigenous medicinal aromatic plants: M. flabellifolius, O. asteriscoides, A. afra, L. javanica, H. cymosum subsp. cymosum, C. gratissimus var. subgratissimus, H. natalensis, T. camphoratus and P. grandidentatus have been selected to investigate their essential oil composition and biological activity, establishing scientific evidence for their local use in herbal preparations. Figure 12.1 outlines the thesis in brief indicating plants studied, methods used and outcomes. A more extensive discussion follows based on the objectives listed in Chapter 1. 226 Helichrysum cymosum Chapter 6 Figure 12.1 Outline of thesis: The antimicrobial activity and essential oil composition of medicinal aromatic plants used in African traditional healing. Plectranthus grandidentatus and Tarchonanthus camphoratus Chapter 9 Methods: ? Disc diffusion ? MIC ? Time-kill Methods: ? Disc diffusion ? MIC ? Time-kill Methods: ? Disc diffusion ? MIC ? Bioassay guided fractionation Methods: ? MIC/FIC ? Isobologram synergy studies Methods: ? MIC ? Isobologram synergy studies Outcomes: ? Antimicrobial evaluation. ? Method standardization. Outcomes: ? Antimicrobial evaluation. ? Efficacy of major compounds. Outcomes: ? Antimicrobial evaluation. ? Method comparability. ? Synergy between major constituents. Outcomes: ? Antimicrobial evaluation. ? Synergistic combination with A. afra. Outcomes: ? Antimicrobial evaluation. ? Isolated compound. Outcomes: ? Antimicrobial evaluation. ? Efficacy of plant part combinations. Outcomes: ? Seasonal and geographical variation. ? Enantiomer selectivity. Myrothamnus flabellifolius Chapter 2 Osmitopsis asteriscoides Chapter 3 Artemisia afra Chapter 4 Lippia javanica Chapter 5 Croton gratissimus Chapter 7 Heteropyxis natalensis Chapter 8 Methods: ? MIC ? Volatile vs. non- volatile activity ? Isobologram synergy studies Outcomes: ? Antimicrobial evaluation. ? Role of volatile constituents. Indigenous/commercial comparison Chapter 10 Methods: ? MIC ? Time-kill Outcomes: ? Antimicrobial evaluation. ? Comparable activities. 227 Objective 1: To determine the antimicrobial activity of selected indigenous aromatic plants used in traditional healing. In a review of over 500 publications detailing the antimicrobial activity of plants, no set criteria could be found that classifies an essential oil as having good, moderate or poor activity, and many researchers base the assessment on their own particular data attained. It is clear from this study that the essential oils exhibit much higher MIC values (poorer activity) than the extracts. Antimicrobial activity with MIC values 4 mg/mL or lower were considered noteworthy for these studies. Based on this criterion highest activity was attained for M. flabellifolius oil where seven of the nine pathogens studied exhibited MIC values of 4 mg/mL or lower. Similarly good activity was found for four of the ten pathogens studied with H. cymosum subsp. cymosum oil and four of the eight pathogens studied with P. grandidentatus oil. With this study on aromatic plants, it was noted that it was not always the volatile (essential oils) constituents that were responsible for the highest antimicrobial activity, but also the non-volatile (extracts) constituents, as documented with the studies on H. cymosum subsp. cymosum, C. gratissimus var. subgratissimus, T. camphoratus and P. grandidentatus. Extracts having activities where MIC values are below 8 mg/mL (Fabry et al., 1998) are considered to possess some antimicrobial activity, and natural products having activities where MIC values are below 1 mg/mL are considered noteworthy (Gibbons, 2004; Rios and Recio, 2005). Antimicrobial activities for extracts with MIC values 1 mg/mL or lower were attained for all pathogens when investigated against the extract of H. cymosum subsp. cymosum. Other extracts exhibiting high activities include C. gratissimus var. subgratissimus (bark) with four of the ten pathogens having MIC values lower than 1 mg/mL and P. grandidentatus (dried plant material), also with four pathogens having MIC values lower than 1 mg/mL. Compounds having antimicrobial activities 64-100 ?g are accepted as having clinical relevance for antimicrobial activity (Gibbons, 2004; Rios and Recio, 2005). The isolated compound helihumulone from H. cymosum subsp. cymosum had MIC values below 64 ?g against nine of the ten pathogens tested. The highest antimicrobial activities obtained using MIC methodology, noted for all plants within this study are given in Table 12.1. 228 Table 12.1 The highest antimicrobial activities found for all plants in this study. Plant Component of plant Test organism showing highest antimicrobial activity Antimicrobially related traditional use*1 MIC value (mg/mL) M. flabellifolius Essential oil Cryptococcus neoformans ATCC 90112 Colds, respiratory ailments, chest pains and coughs 2.00 Escherichia coli ATCC 1175 Stomach complaints and intestinal disorders O. asteriscoides Essential oil Staphylococcus epidermidis ATCC 2223 Cuts and swellings 4.00 A. afra Essential oil Serratia odorifera ATCC 33132 Ailments associated with the gastrointestinal tract 4.00 Cryptococcus neoformans ATCC 90112 Coughs, colds, bronchial and respiratory ailments L. javanica Essential oil Escherichia coli ATCC 11775 Dysentery, diarrhoea and suspected food poisoning 4.00 Bacillus cereus ATCC 11778 Topical infections Leaf extract Cryptococcus neoformans ATCC 90112 Respiratory ailments 0.08 H. cymosum subsp. cymosum Helihumulone Pseudomonas aeruginosa ATCC 9027 Topical infections 0.02 Bacillus cereus ATCC 11778 Abdominal disorders and skin inflammation 1:1:1 Leaf, bark and root Staphylococcus aureus ATCC 12600 Skin inflammation C. gratissimus var. subgratissimus 1:1 Bark and root Bacillus cereus ATCC 11778 Abdominal disorders and skin inflammation 0.30*2 H. natalensis Essential oil (Lagalametse and Balakane samples) Cryptococcus neoformans ATCC 90112 Respiratory disorders and as a decongestant 0.50 T. camphoratus Dried leaf extract Candida albicans ATCC 10231 Stomach ailments 0.30 P. grandidentatus Dried leaf extract Enterococcus faecalis ATCC 29212 Respiratory complaints 0.06 *1 Adapted from reference literature within relevant Chapters. *2Bacillus subtilis is excluded from data set due to the organism?s rare pathogenic status. 229 Objective 2: To record the essential oil composition of some of the most widely used South African medicinal aromatic plants. The major constituents for each plant essential oil are summarized in Table 12.2. The most predominant constituent present in five of the nine plants studied was 1,8-cineole. This monoterpene compound has previously attained pharmacological status as an inhalant and expectorant with application for respiratory complaints (Windholz, 1976; Silvestre et al., 1999). Thus, it is not surprising that it is associated with the aromatic plants studied here. The antimicrobial property of 1,8-cineole has also been widely investigated (Table 12.3) with varied reports attained by different authors. Variation was noted between strains having the same reference number as well as strains having different reference numbers. Such variations emphasise the need to standardize reference cultures, as well as corroborate findings between different laboratories. Mostly the authors are in general agreement that some antimicrobial efficacy may be attributed to 1,8 cineole. One must not forget however, that synergistic interactions between constituents may occur, such as that noted in the O. asteriscoides study where 1,8-cineole in combination with (-)-camphor act synergistically to enhance the antimicrobial activity. Objective 3: To investigate the plant composition with antimicrobial activity relating to structure activity relationships and if possible identify the antimicrobial factor responsible. Results observed from the microbiological activity for the various oils have shown that the correlation between chemical structure and biological activity are integrated and the essential oil chemistry has provided insights into the antimicrobial activity. Studies on the structure activity related antimicrobial activity was demonstrated with the interaction of the major oil constituents of O. asteriscoides, where the combination of camphor and 1,8-cineole were shown to enhance antimicrobial efficacy. Investigation of the four major compounds most abundant in the essential oil of A. afra (artemisia ketone, 1,8-cineole, ? and ?-thujone) indicated minimal antimicrobial activity against K. pneumoniae after 24 hours, when investigated independently and in various combinations. It is probable that the mono- and sesquiterpenes could exert antimicrobial properties by working in a synergistic manner or that minor compounds greatly contribute to the antimicrobial activity. Studies on the essential oil, extract and isolated biological compound helihumulone from H. cymosum subsp. cymosum demonstrated higher activities for the extract. 230 Helihumulone, not found in the volatile essential oil, could possibly indicate why the essential oil showed lower antimicrobial activities than the crude extract. Enantiomer studies, as undertaken with H. natalensis indicated how the variation of results obtained between the (+), (-) and racemic forms of limonene when combined with 1,8-cineole as well as their combination in different ratios clearly impact on the bio-activity. The geographical investigation of the H. natalensis oils revealed that the anomalous chemotype from Lagalametse having (Z)-3-hexenyl nonanoate as a major constituent could possibly account for the increased antimicrobial activity found in this sample. In a recent publication by Holley and Patel (2005), mention was made of the lack of information available on essential oil constituent interactions. The work presented here sheds some light on these possible interactions. Table 12.2 Summary of the major oil constituents obtained from all plants in this study. Plant essential oil Major components Component % % within total essential oil trans-pinocarveol 19.6 M. flabellifolius pinocarvone 11.1 30.7 1,8-cineole 59.9 O. asteriscoides (-)-camphor 12.4 72.3 1,8-cineole 17.8 artemisia ketone 10.1 ?-thujone 18.8 A. afra ?-thujone 12.5 59.2 linalool 65.2 L. javanica (Z)-?-ocimene 13.0 78.2 1,8-cineole 20.4 ?-caryophyllene 10.8 H. cymosum subsp. cymosum ?-pinene 12.4 43.6 1,8-cineole 8.3 germacrene D 8.6 C. gratissimus var. subgratissimus ?-phellandrene 20.7 37.6 limonene 15.1-25.4 linalool (Nelspruit and Lagalametse samples) 11.4-16.7 (Z)-3-hexenyl nonanoate (Lagalametse sample) 16.0 H. natalensis ?-pinene (Balakane sample) 25.2 Variable* 1,8-cineole 9.3 T. camphoratus ?-caryophyllene 13.4 22.7 T-cadinol 10.3 P. grandidentatus camphor 15.2 25.5 *Composition % within major constituents displayed seasonal and geographical variation. 231 Table 12.3 The antimicrobial properties of 1,8-cineole as determined by corresponding authors. ND = not determined. Test micro-organism with corresponding reference strain number Publication S. aureus S. epidermidis B. cereus E. coli P. aeruginosa K. pneumoniae C. albicans Carson and Riley (1995) 0.5 (NCTC 6571) ND ND 0.25 (NCTC 8359) ND ND 1.0 (ATCC 10231) Griffin et al. (1999) 18.1 (NCTC 8325) ND ND 9.1 (AG100) 18.1 (NCTC 6749) ND 9.1 (KEMH 5) Tzakou et al. (2001) 9.5 (ATCC 25923) 9.5 (ATCC 12228) ND 2.0 (ATCC 25922) 2.8 (ATCC 227853) 3.0 (ATCC 13883) 0.3 (not given) Magiatis et al. (2002) 9.5 (ATCC 25923) 9.5 (ATCC 12228 ND 2.4 (ATCC 25922) 2.8 (ATCC 227853) 3.2 (ATCC 13883) 0.3 (ATCC 10231) Setzer et al. (2004) 1.3 (ATCC 29213) ND 0.3 (ATCC 14579) 0.6 (ATCC 25922) 0.6 (ATCC 27853) ND 0.6 (ATCC 10231) Cha et al. (2005) 12.8 (ATCC 21293) 0.8 (ATCC 12228) ND 3.2 (ATCC 25922) ND ND ND van Vuuren (Chapter 8) 8.0 (ATCC 12600) ND 2.0 (ATCC 11778) 8.0 (ATCC 11775) 4.0 (ATCC 9027) 8.0 (ATCC 13883) ND 232 Objective 4: To study specific pharmacological interactions. The topic of synergy has been highlighted recently by many authors. In the synergy review by Williamson (2001), the need for further interactive investigations was emphasized. Rios and Recio (2005) recommended that a high priority should be given to further studies involving plant combinations and Gilani and Rahman (2006) amplified that there is a scarcity of ethnopharmacognosists performing a much needed study into medicinal plant combinations. Holley and Patel (2005) stated that very little information was available on constituent interaction. The emphasis on synergistic studies was also highlighted by Prof. H Wagner (Editor: Phytomedicine) at the Annual Conference of the Society for Medicinal Plant Research, Florence (21-25 August, 2005) when he stated ?Progress in the field of synergy research could give phytotherapy new legitimacy and the possibility to treat diseases which up to now were reserved for chemotherapy only?. This study thus identified possible interaction between various constituents of the essential oils as was noted with the interaction of the major oil constituents of O. asteriscoides where camphor and 1,8-cineole, in combination act synergistically to enhance activity. Conversely, the major volatile constituents of A. afra in various combinations had no significant role on the antimicrobial activity of the plant. The interaction between different plant species was studied using time-kill methodology, where the synergistic interaction was demonstrated for the combined use of L. javanica with A. afra for the treatment of respiratory infections associated with K. pneumoniae. Results show that in combination, efficacy is enhanced, which supports with the traditional use of the plant. Antimicrobial studies on C. gratissimus var. subgratissimus show synergistic, additive, antagonistic or non-interactive action between plant parts, depending on the specific ratio in which the plant parts are combined. Higher synergistic sensitivities were noted for the root / leaf combinations. Other interactive studies have shown that the volatile fractions within T. camphoratus and P. grandidentatus serve to enhance overall activity of the plant. Objective 5: To compare the antimicrobial activity (potency) of indigenous aromatic plants to commercially available essential oils with claimed antimicrobial activity. Chapter 10 deals exclusively with this objective. By means of comparatively determining the MIC of popular commercial oils (Lavendula angustifolia, Thymus vulgaris, Melaleuca 233 alternifolia, Mentha piperita and Rosmarinus officinalis) against the essential oils of five indigenous plants (M. flabellifolius, O. asteriscoides, H. natalensis, A. afra and L. javanica), efficacy was determined against eight bacterial reference strains and two yeast reference strains. The laboratory conditions and inoculum were standardized to ensure all ten essential oils (commercial and indigenous) were evaluated under identical conditions. Results indicated that South African indigenous essential oils compare favourably with the commercial oils studied herein. Of all oils studied, M. flabellifolius showed the most rapid cidal effect against all three pathogens tested with the time-kill method. Objective 6: To compare and validate results obtained from various test methodologies (e.g. disc diffusion, MIC and time-kill studies). Many variables are involved in plant-based studies and microbiological assays are only as reliable and sensitive as the test system employed. It was therefore necessary to incorporate more than one microbiological method to validate the results attained. The methods used for each study is summarised in Figure 12.1. Detailed methodology development and critical validation assessment was reviewed in Chapter 11. Objective 7: To provide a scientific rationale for the traditional use of various medicinal aromatic plants to treat infectious diseases. Incorporated in Table 12.1 is a summary of the traditional use of the plants studied with reference to the test organisms having the most significant activity. Other significant findings were noted for plant combinations, as was the case for the combination of L. javanica and A. afra, where the time-kill study indicated increased antimicrobial efficacy for the respiratory pathogen K. pneumoniae. These findings support the traditional use of these plants when used together in traditional therapeutic practices. The benefit of combining roots and leaves of C. gratissimus var. subgratissimus as used traditionally was also apparent in the isobologram studies, where extracts showing the strongest synergistic profile were noted with studies on B. cereus, C. albicans and C. neoformans, correlating with diseases that the plant is traditionally used to treat. 234 12.3 Future Trends Antimicrobial studies on plants have evolved from multidisciplinary investigations where researchers have used the antimicrobial data to complement the phytochemical study, to focused studies pertaining to specific information about the plants? antimicrobial activity. This reflects the importance of such studies and conti nued investigations have emerged as a result of many of these positive preliminary screening procedures. A number of recommended strategies for future antimicrobiological analysis in pharmacognosy should thus be considered. Toxicity: Although it is considered that traditional plant medicines, having been used for centuries, are generally safe, it can not be taken for granted that these treatments do not have any potentially toxic, mutagenic and / or carcinogenic properties and thus it has been recommended that pharmacological studies should always be accompanied by toxicology screening (Taylor et al., 2001; Fennel et al., 2004; Cos et al., 2006). The toxicology of the oils studied have all, with the exception of P. grandidentatus (still in progress), been assessed by Dr. Robyn van Zyl at the Pharmacology Division of the Department of Pharmacy and Pharmacology, University of Witwatersrand. Toxicology data (Table 12.4) is arranged in ascending order from the least toxic (O. asteriscoides) to the most toxic (C. gratissimus var. subgratissimus) plant species. Toxicity comparisons were made with quinine, considered as an acceptable control, presenting low toxicity. It comes as no surprise that C. gratissimus var. subgratissimus demonstrates the highest toxicity profile as this species is known for its toxicity (Watt and Breyer-Brandwijk, 1962; Hutchings et al., 1996). The presence of copious amounts ?-phellandrene (20.7%) in the oil may also contribute towards the overall toxicity, as this compound is known for its skin irritancy effect and ability to cause vomiting and diarrhoea if ingested (Windholz, 1976). A further recommendation to investigate the toxicity profiles for the phenolic extracts of C. gratissimus var. subgratissimus (bark, leaves and roots), T. camphoratus and P. grandidentatus is advised. 235 Essential oils Mean toxicity profile IC50 (%) Osmitopsis asteriscoides 0.00908 ? 0.00140 Lippia javanica 0.00730 ? 0.00065 Artemisia afra 0.00522 ? 0.00088 Heteropyxis natalensis 0.00246 ? 0.00032 Myrothamnus flabellifolius 0.00226 ? 0.00010 Tarchonanthus camphoratus 0.00214 ? 0.00112 Helichrysum cymosum subsp. cymosum * 0.00175 ? 0.00310 Croton gratissimus var. subgratissimus 0.00157 ? 0.00016 Quinine control 0.01360 ? 0.00041 *(van Vuuren et al., 2006) Mode of action studies: While investigations presented here demonstrate the microbial inhibition on a selection of indigenous medicinal plant essential oils, the mechanism of action requires further insight. Mode of action studies are complex and destruction of microbial cells could occur on many different levels. There is no doubt that essential oils compromise membrane integrity, as noted in studies on O. asteriscoides (Chapter 3) and with studies by Cox et al. (2000), Griffin et al. (2001), Lambert et al. (2001) and Fitzgerald et al. (2004), however, there are a number of other mechanisms that need to be elucidated. Leakage of potassium ions and nucleic acids from bacterial cells (Hada et al., 2003; Inoue et al., 2004), impairment of a variety of enzyme systems, electrochemical potential, measurement of microbial respiration (Cox et al., 2000), ATPase activity (Burt, 2004), cellular metabolic processes (Holley and Patel, 2005), changes in composition of fatty acid and intracellular absorption of material (personal communication Dr. Panos Skandamis) and even the role of constituents as penetration enhancers require further investigation. One also needs to take into consideration that given the number of essential oil constituents within one plant, there may be constituents independently exhibiting different modes of action that when combined, act synergistically. Table 12.4 The toxicity profiles for the eight essential oils studied (van Zyl, publication in preparation). 236 Dosage schedules and administration for indigenous medicines: Research design often overlooks these factors, which may be in part due to the lack of information provided to researchers on the details of the traditional use. But moreover, laboratory techniques often exclude aqueous extracts, which is the most widely used traditional administration, in favour of working with apolar extracts in order to maximize activities. Technically, aqueous extracts are more difficult to work with and especially with microbiological work, contaminate rapidly, however, these studies are important as bioactive compounds may be overlooked. In addition, positive results from laboratory tests may indicate activities greater than that found in the traditional preparative methods, as indicated in a study by J?ger (2003). With the rapid rate of plant species extinction, there is an urgent need to document the medicinal potential of plants in an effort to preserve the knowledge and natural resources. It is a future challenge to translate the applied knowledge gained from intricate assays and make it meaningful to the ethnic people. Future research design should also look at regular or consistent administration of antimicrobials to determine if better efficacy would be achieved with sub chronic administrations in comparison with single acute administrations (Etkin, 2001; Etkin and Elisabetsky, 2005). As in scheduled antimicrobial therapy, consistent dosages are required to produce a cidal effect. Such studies could be carried out with plant test extracts or oils in extended time-kill methods, where the viability can be monitored over time with regular dosages. Pharmacological interactions: The identification of a single active chemical entity that is responsible for the antimicrobial activity of a plant is becoming more and more improbable. Research is now focussing on the investigation of a combination of compounds to achieve a greater efficacy. To ensure future success in natural product research we must incorporate interactive phytochemical studies with existing practices in the hope that developments involving combinations may be used as a foundation and driving force in the search for novel chemotherapeutic agents. Combinations with conventional antimicrobials: Studies supporting the possible use of medicinal plants together with known antibiotics may serve to increase potency and reduce the dosage of potent drugs. In so doing, the toxicity levels may be reduced. With natural product research gaining credibility and the escalated use of herbal remedies by the public sector, less burden and reliance may be placed on conventional antimicrobial drugs. In so doing we can 237 preserve these important allopathic antimicrobials and delay the emergence of resistance. In a recent publication by Rios and Recio (2005), this proposal is given high priority. Metabolomics: The metabolomic study of plants is a rapidly growing field, where the metabolites of plant species are quantitatively and qualitatively analyzed in order to obtain more information on the overall mechanism of action. Information regarding metabolites can be correlated with biological activities to determine if a specific fingerprint exists. Such studies encompass all aspects of plant metabolites and are a major challenge as this moves away from the single targeted bioactivity approach of drug development (Verpoorte et al., 2005). Pathogen specific studies: More attention needs to be directed towards the ethnobotanical use of the plant and the rational antimicrobial screening that follows. Plants used for skin afflictions should be tested against micro-organisms such as Epidermophyton floccosum, Microsporum canis and Trichophyton mentagrophytes, in addition to the more commonly tested organisms such as S. aureus and S. epidermidis. Similarly, plants used for oral complaints should be tested against Streptococcus mutans, S. sobrinus and Porphyromonas gingivalis. With studies on plants used for the treatment of sexually transmitted diseases, choice of bacterial and fungal test strains are limited. While Neisseria gonorrhoeae remains the most commonly studied organism, other organisms responsible for sexually transmitted infections (STI) include Treponema pallidum (Syphilis) and Haemophilus ducreyi (Chancroid). Studies on T. pallidum prove difficult as the micro-organism cannot be cultured in common bacteriological growth media. Studies on H. ducreyi are no longer feasible as outbreaks no longer occur (personal communication Prof. Lewis at the National Centre for Communicable Diseases), however, in spite of the difficulties associated with culturing of sexually transmitted organisms, there is a need to investigate the use of medicinal plants for the treatment of STI?s, even if studies are only restricted to N. gonorrhoeae. Whilst it is understandable that all these organisms mentioned here are fastidious and require more intensive cultural techniques, it is a necessary progression for further antimicrobial pharmacognosy studies. We can no longer assume that broad spectrum activity is adequate for plants used for specific afflictions. In vivo and whole system studies: Even though a number of successful studies have identified specific plant species as having antimicrobial activity in an in vitro model, it is necessary to 238 subject these plants to animal and human studies to determine their efficacy in whole organism systems. Traditional medicines are esteemed not only for their therapeutic value but also from a holistic administration approach, in which the plant is given to treat the patient on various levels. Thus, when studying traditional medicines one must not forget that there may be other physiological effects on the body that act beyond the symptomatic treatment of the disease. In addition, studies on the efficacy of plants against resident beneficial bacteria such as Lactobacillus acidophilus and Bifidobacterium bifidum could yield information that may make them more appropriate than the presently administered allopathic antimicrobials, which inevitably destroy the beneficial commensal organisms together with the invading pathogens. In line with this, studies on biofilms should be initiated as these mixed populations play an active role in the microbiological ecology of the human body. Ethnopharmacological studies have shown that many plant essential oils are broad spectrum in their activity, unlike conventional chemotherapeutic agents which often target specific bacterial modes of action. This multi- faceted approach may be effective in treating polymicrobial infections and may also prove to be effective in inhibiting undesirable biofilm formation. Formulation studies: The next logical step in the investigation of the antimicrobial efficacy of plant extracts and essential oils would be to establish suitable formulations that retain the efficacy demonstrated in the in vitro screening procedures. Formulations such as tinctures, teas, ointments, capsulation or tableting incorporating plant material should be investigated to determine feasibility and bioavailability. 12.4 Conclusion The antimicrobial study of some of the most significant indigenous aromatic plants presented here focuses not only on antimicrobial screening but also on addressing many of the concerns and recommendations mentioned in literature. By addressing the complexity of standardizing various methods, as recomme nded by Burt (2004) as well as Rios and Recio (2005), many of the variables encountered when performing antimicrobial assays were examined. Furthermore the interaction between constituents, plant parts, as well as different species, emphasises that the reductionist approach used in examining bio-activity may often be short-sighted (Rios and Recio, 2005; Holley and Patel, 2005; Verpoorte et al., 2005; Cos et al., 2006). 239 It is interesting to note that amongst all the modern technologically advanced medical treatments available today, it is still the simplest bacterium that may result in humankind?s demise. One can only hope that with continued research we can devise mechanisms that may delay or prevent this from happening. We need to plan for future therapeutic treatments that will withstand the growing population needs and emergence of ?superbugs? such as MRSA and VRE. In support of this, the study of the antimicrobial activities of medicinal plants has clearly become a progressive trend. It has been predicted (Mulholland, 2005) that multidisciplinary studies within the ethnopharmacology sector will escalate within the next decade. With advances in laboratory techniques, renewed interest in the field and the scientific validation of the traditional use, the possibility now exists to bring traditional medicine to such a level of recognition that it becomes accepted alternate regimen to western healthcare systems. The anticipated future of antimicrobial phytotherapy should not attempt to replace conventional antimicrobials but rather focus on the integration into conventional medicinal practices where the general practitioner may recommend either phytomedicine or allopathic medicine or even both, depending on the diagnosis (Tyler, 1999). 240 References AACHRD Enhancing Research into Traditional Medicine in the African Region. A Working Document Prepared for the 21st Session of the African Advisory Committee for Health Research and Development, Port Louis, Mauritius 2002. Abdel-Mogib M., Albar H.A., Batterjee S.M. Chemistry of the genus Plectranthus. Molecules 7, 271-301, 2002. Acar J.F. Antibiotic synergy and antagonism. Medical Clinics of North America 84 (6), 1391- 1407, 2000. Afolayan A.J., Meyer J.J.M. The antimicrobial activity of 3,5,7-trihydroxyflavone isolated from the shoots of Helichrysum aureonitens. Journal of Ethnopharmacology 57, 177-181, 1997. Aggarwal K.K., Khanuja S.P.S., Ahmad A., Kumar T.R.S., Gupta V.K., Kumar S. Antimicrobial activity profiles of the two enantiomers of limonene and carvone isolated from the oils of Mentha spicata and Anethus sowa. Flavour and Fragrance Journal 17, 59-63, 2002. Akinpelu A.D., Olorunmola F.O. Antimicrobial activity of Bridelia ferruginea fruit. Fitoterapia 71, 75-76, 2000. Alviano W.S., Menon?a-Filho R.R., Alviano D.S., Bizzo H.R., Souto-Padr?n T., Rodrigues M.L., Bolognese A.M., Alviano C.S., Souza M.M.G. Antimicrobial activity of Croton cajucara Benth linalool-rich essential oil on artificial biofilms and planktonic microorganisms. Oral Microbiology Immunology 20, 101-105, 2005. Andrews J.M. Determination of minimum inhibitory concentrations. Journal of Antimicrobial Chemotherapy 48, 18-30, 2004. Angioni A., Barra A., Cereti E., Barile D., Coisson J., Arlorio M., Dessi S., Coroneo V., Cabras P. Chemical composition, plant genetic differences, Antimicrobial and antifungal activity investigation of the essential oil of Rosmarinus officinalis. Journal of Agricultural and Food Chemistry 52, 3530-3535, 2004. Bannister B.A., Begg N.T., Gillespie S.H. Infectious Diseases. 2nd Edition, Livingstone, London, UK, ISBN0632053194, 2000. Batista O., Duarte A., Nascimento J., Sim?es M.F. Structure and antimicrobial activity of diterpenes from the roots of Plectranthus hereroensis. Journal of Natural Products 57 (6), 858-861, 1994. Beale A.S., Sutherland R. Measurement of Combined Antibiotic Action in Antibiotics: Assessment of Antimicrobial Activity and Resistance. Russell A.D., Quesnel L.B., (Editors), Academic Press (Inc), London, UK, ISBN0126044805, 1983. Begue W.J., Kline R.M. The use of tetrazolium salts in bioautographic procedures. Journal of Chromatography 64, 182-184, 1972. 241 Berenbaum M.C. A method for testing for synergy with any number of agents. The Journal of Infectious Diseases 137 (2), 122-130, 1978. Berenbaum M.C. Correlations between methods for measurement of synergy. The Journal of Infectious Diseases 142 (3), 476-478, 1980. Bishop C.D Antiviral activity of the essential oil of Melaleuca alternifolia (Maiden & Betche) cheel (tea tree) against tobacco mosaic virus. Journal of Essential Oil Research 7 (6), 641-644, 1995. Block S., Baccelli C., Tinant B., van Meervelt L., Rozenberg R., Habib Jiwan J-L., Llabr?s G., De Pauw-Gillet M.-C., Quetin-Leclercq J. Diterpenes from the leaves of Croton zambesicus. Phytochemistry 65 (8), 1165-1171, 2004. Bohlmann F., Zdero C. Neue sesquiterpenlactone aus Osmitopsis asteriscoides (L.). Cass Chemische Berichte 107, 1409-1415, 1974. Bohlmann F., Ziesche J., Mahanta P. Neue chalkon-derivate und humulon-?hnliche verbindungen aus Helichrysum-arten. Phytochemistry 18, 1033-1036, 1979. Bohlmann F., Zdero C., Jakupovic J., Rourke J. cis-Guajanolide aus Osmitopsis asteriscoides. Liebigs Annalen der Chemie, ISBN2342B2351, 1985. Bonati A. Medicinal plants and industry. Journal of Ethnopharmacology 2, 167-171, 1980. Borris R.P. Natural products research: perspectives from a major pharmaceutical company. Journal of Ethnopharmacology 51, 29-38, 1996. Boucher A.N., Tam V.H. Mathematical formulation of additivity for antimicrobial agents. Diagnostic Microbiology and Infectious Diseases 55 (4), 319-325, 2006. Bougatso C., Ngassapa O., Runyoro D.K.B., Chinou I.B. Chemical composition and in vitro antimicrobial activity of the essential oils of two Helichrysum species from Tanzania. Verlag der Zeitschrift f?r Naturforschung 59, 368-372, 2004. Boyd R.F., Hoerl B.G. Basic Medical Microbiology. 2nd Edition, Livingstone, London, UK, ISBN0316104337, 1981. Braga M.E.M., Ehlert P.A.D., Ming L.C., Meireles M.A.A. Supercritical fluid extraction from Lippia alba: global yields, kinetic data and extract chemical composition. Journal of Supercritical Fluids 34, 149-156, 2005. Brantner A., Grein E. Antibacterial activity of plant extracts used externally in traditional medicine. Journal of Ethnopharmacology 44, 35-40, 1994. Brantner A., Pfeiffer K.P., Brantner H. Applicability of diffusion methods required by the pharmacopoeias for testing antibacterial activity of natural compounds. Pharmazie 49, 512-516, 1994. Brehm-Stecher B.F., Johnson E.A. Sensitization of Staphylococcus aureus and Escherichia coli to antibiotics by the sequiterpenoids nerolidol, farnesol, bisabolol and apritone. Antimicrobial Agents and Chemotherapy 47 (10), 3357-3360, 2003. 242 Bremer K. The genus Osmitopsis (Compositae). Botaniska Notiser 125, 9-45, 1972. Bruneton J. Pharmacognosy, Phytochemistry, Medicinal Plants. Springer, Hampshire, ISBN1898298130, 1995. Buchanan K.L., Murphy J.W. What makes Cryptococcus neoformans a pathogen? Emerging Infectious Diseases. http://www.cdc.gov/ncidod/eid/vol4no1/buchanan.htm 1998. Burt S. Essential oils: their antimicrobial properties and potential application in foods - a review. International Journal of Food Microbiology 94 (3), 223-253, 2004. Buwa L.V., van Staden J. Antibacterial and antifungal activity of traditional medicinal plants used against venereal diseases in South Africa. Journal of Ethnopharmacology 103, 139- 142, 2006. Caceres A., Alvarez A.V., Ovando A.E., Samayoa B.E. Plants used in Guatemala for the treatment of respiratory diseases. Screening of 68 plants against Gram-positive bacteria. Journal of Ethnopharmacology 31, 193-208, 1991. Caelli M., Porteous J., Carson C.F., Heller R., Riley T.V. Tea tree oil as an alternative topical decolonization agent for methicillin-resistant Staphylococcus aureus. Journal of Hospital Infection 46, 236-237, 2000. Cano J.H., Volpato G. Herbal mixtures in the traditional medicine of Eastern Cuba. Journal of Ethnopharmacology 90, 293-316, 2004. Capasso A., Sorrentino L. Pharmacological studies on the sedative and hypnotic effect of Kava- kava and Passiflora extracts combination. Phytomedicine 12, 39-45, 2005. Capasso L. 5300 Years ago, the Ice Man used natural laxatives and antibiotics. The Lancet 352, 1864, 1998. Carson C.F., Hammer K.A., Riley T.V. Broth micro-dilution method for determining the susceptibility of Escherichia coli and Staphylococcus aureus to the essential oil of Melaleuca alternifolia (tea tree oil). Microbios 82, 181-185, 1995. Carson C.F., Riley T.V. Antimicrobial activity of the major components of the essential oil of Melaleuca alternifolia. Journal of Applied Bacteriology 78, 264-269, 1995. Cassella S., Cassella J.P., Smith I. Synergistic antifungal activity of tea tree (Melaleuca alternifolia) and lavender (Lavandula angustifolia) essential oils against dermatophyte infection. The International Journal of Aromatherapy 12 (1), 2-14, 2002. Cedergreen N., Kamper A., Streibig J.C. Is prochloraz a potent synergist across aquatic species? A study on bacteria, daphnia, algae and higher plants. Aquatic Toxicology 78, 243-252, 2006. Cerqueira F., Cordeiro-Da-Silva A., Gaspar-Marques C., Sim?es F., Pinto M.M.M., Nascimento M.S.J. Effect of abietane diterpenes from Plectranthus grandidentatus on T- and B- lymphocyte proliferation. Bioorganic Medicinal Chemistry 12, 217-223, 2004. 243 Cha J-D., Jeong M-R., Jeong S., Moon S-E, Kim J-Y., Kil B-S, Song Y-H. Chemical composition and antimicrobial activity of the essential oils of Artemisia scoparia and A. capillaries. Planta Medica 71, 186-190, 2005. Chagonda L.S., Makanda C., Chalchat J.C. Essential oils of four wild and semi-wild plants from Zimbabwe: Colospermum mopane (Kirk ex Benth.) Kirk ex Leonard, Helichrysum splendidum (Thunb.) Less, Myrothamnus flabellifolius (Welw.) and Tagetes minuta L. Journal of Essential Oil Research 11, 573-578, 1999. Chagonda L.S., Makanda C.D. Essential oils of wild and cultivated Lippia javanica (Spreng) and L. oatesii (Rolfe) from Zimbabwe. Journal of Essential Oil Research 12, 1-6, 2000. Chagonda L.S., Makanda C.D., Chalchat J.C. Essential oils of cultivated Heteropyxis natalensis (Harv.) and cultivated Heteromorpha trifoliata (Wendl.) Eckl. & Zey. from Zimbabwe. Journal of Essential Oil Research 12, 317-321, 2000. Chalchat J.C., Garry R. Correlation between chemical composition and antimicrobial activity. VI. Activity of some African essential oils. Journal of Essential Oil Research 9, 67-75, 1997. Chalchat J.C., Chiron F., Garry R.Ph., Lacoste J. Photochemical Hydoperoxidation of terpenes: Antimicrobial activity of ?-pinene, ?-pinene and limonene hydroperoxides. Journal of Essential Oil Research 12, 125-134, 2000. Chao S.C., Young D.G. Screening for inhibitory activity of essential oils on selected bacteria, fungi and viruses. Journal of Essential Oil Research 12, 639-649, 2000. Christofilogiannis P. Current inoculation methods in MIC determination. Aquaculture 196, 297- 302, 2001. Christoph F., Stahl-Biskup E. Death kinetics of Staphylococcus aureus, exposed to commercial tea tree oils. Journal of Essential Oil Research 13, 98-102, 2001. Claeson P., R?dstr?m P., Sk?ld O., Nilsson ?., H?glund S. Bactericidal effect of the sesquiterpene T-cadinol on Staphylococcus aureus. Phytotherapy Research 6, 94-98, 1992. Clark A.M. Natural products as a resource for new drugs. Pharmaceutical Research 13, 1133- 1141, 1996. Codd J.E., Deasy P.B. Synergistic antifungal interaction between miconazole nitrate and chlorhexidine acetate. International Journal of Pharmaceutics 173, 3-11, 1998. Codd L.E. Lamiaceae, Flora of Southern Africa. Botanical Research Institute, Pretoria, South Africa, ISBN0621082686, 1985. Colombo L., Nucci M., Salomao R., Branchini M.L.M., Richtmann R., Derossi A., Wey S.B. High rate of non-albicans Candidemia in Brazilian tertiary care hospitals. Diagnostic Microbiology and Infectious Diseases 34, 281-286, 1999. Cordato D.J., Mather L.E., Herkes G.K. Stereochemistry in clinical medicine: a neurological perspective. Journal of Clinical Neuroscience 10 (6), 649-654, 2003. 244 Cordell G.A., Colvard M.D. Some thoughts on the future of ethnopharmacology. Journal of Ethnopharmacology 100, 5-14, 2005. Cos P., Vlietinck A.J., Vanden Berghe D., Maes L. Anti-infective potential of natural products: How to develop a stronger in vitro 'proof-of-concept'. Journal of Ethnopharmacology 106, 290-302, 2006. Cowan M.M. Plant products as antimicrobial agents. Clinical Microbiology Reviews 12 (4), 564- 582, 1999. Cox S.D., Mann C.M., Markham J.L., Bell H.C., Gustafson J.E., Warmington J.R., Wyllie S.G. The mode of antimicrobial action of the essential oil of Melaleuca alternifolia (tea tree oil). Journal of Applied Microbiology 88, 170-175, 2000. Cox S.D., Mann C.M., Markham J.L., Gustafson J.E., Warmington J.R., Wyllie S.G. Determining the antimicrobial actions of tea tree oil. Molecules 6, 87-91, 2001. Da Cunha P. Identification and determination of principle constituents of essential oil of Myrothamnus flabellifolius collected in Angola. Boletim Da Fuculdade De Farmacia. Univ. Coimbra Ed Cient 34, 49-61, 1974. Dawson-Andoh B.E., Lovell R., Kamdem D.P. Inhibitory and compatibility effects of essential oils on sapstain and biological control of fungi. Journal of Essential Oil Research 12, 509-515, 2000. Delespaul Q., de Billerbeck G.V., Roques C.G., Georges M. The antifungal activity of essential oils as determined by different screening methods. Journal of Essential Oil Research 12, 256-266, 2000. Dellar J.E., Cole M.D., Waterman P.G. Antimicrobial abietiane diterpenoids from Plectranthus elegans. Phytochemistry 41, 735-738, 1996. Demirci B., Tabanca N., Ba?er K.H.C. Enantiomeric distribution of some monoterpenes in the essential oils of some Salvia species. Flavour and Fragrance Journal 17, 54-58, 2002. Denyer S.P., Hugo W.B. (Editors) Mechanisms of Action of Chemical Biocides. Blackwell Scientific, London, UK, ISBN0632029285, 1991. Derwent S. Tradition and medicine. Sawubona October, 118-122, 2004. Devlieghere F., Vermeiren L., Debevere J. New preservation technologies: Possibilities and limitations. International Dairy Journal 14, 273-285, 2004. Dharmananda S. Myrrh and frankincense, spiritual significance. Internet Journal 2, 1-6, 2003. Dharmaratne H.R.W., Wijesinghe W.M.N.M., Thevanasem V. Antimicrobial activity of xanthones from Calophyllum species, against methicillin-resistant Staphylococcus aureus (MRSA). Journal of Ethnopharmacology 66, 339-342, 1999. Dilika F., Afolayan A.J., Meyer J.J.M. Comparative antimicrobial activity of two Helichrysum species used in male circumcision in South Africa. South African Journal of Botany 63, 158-159, 1997. 245 Dorman H.J.D., Deans S.G. Antimicrobial agents from plants: antibacterial activity of plant volatile oils. Journal of Applied Microbiology 88, 308-316, 2000. Drewes S.E., Khan F., van Vuuren S.F., Viljoen A.M. Simple 1,4-benzoquinones with antibacterial activity from stems and leaves of Gunnera perpensa. Phytochemistry 66, 1812-1816, 2005. Drewes S.E., Madau K.E., van Vuuren S.F., Viljoen A.M. Antimicrobial monomeric and dimeric diterpenes from the leaves of Helichrysum tenax var tenax. Phytochemistry 67, 716-722, 2006. Drobniewski F.A. Bacillus cereus and related species. Clinical Microbiology Reviews 6 (4), 324-338, 1993. Eksteen D., Pretorius J.C., Nieuwoudt T.D., Zietsman P.C. Mycelial growth inhibition of plant pathogenic fungi by extracts of South African plant species. Annuals of Applied Biology 139, 243-249, 2001. El-Shazly A.M., Hafez S.S., Wink M. Comparative study of the essential oils and extracts of Achillea frantissima (Forssk.) Sch. Bip. and Achillea santolina L. (Asteraceae) from Egypt. Pharmazie 59, 227-230, 2004. Eldeen I.M.S., Elgorashi E.E., van Staden J. Antibacterial, anti-inflammatory, anti-cholinesterase and mutagenic effects of extracts obtained from some trees used in South African traditional medicine. Journal of Ethnopharmacology 102, 457-464, 2005. Eloff J.N. A sensitive and quick microplate method to determine the minimal inhibitory concentration of plant extracts for bacteria. Planta Medica 64, 711-713, 1998a. Eloff J.N. Which extractant should be used for the screening and isolation of antimicrobial components from plants? Journal of Ethnopharmacology 60, 1-8, 1998b. Eloff J.N. The antibacterial activity of 27 southern African members of the Combretaceae. The South African Journal of Science 95, 148-152, 1999. Eloff J.N., Famakin J.O., Katerere D.R.P. Combretum woodii (Combretaceae) leaf extracts have high activity against Gram-negative and Gram-positive bacteria. African Journal of Biotechnology 4 (10), 1161-1166, 2005. Ernst E.J., Klepser M.E., Ernst M.E., Messer S.A., Pfaller M.A. In vitro pharmacodynamic properties of MK-0991 determined by time-kill methods. Diagnostic Microbiology and Infectious Diseases 33, 75-80, 1999. ESCMID Terminology relating to methods for the determination of susceptibility of bacteria to antimicrobial agents. The European Society of Clinical Microbiology and Infectious Diseases Eucast Definitive document E. def 1.2., CMI 6 (9), 503-515, 2000. Etkin N.L. Perspectives in ethnopharmacology: forging a closer link between bioscience and traditional empirical knowledge. Journal of Ethnopharmacology 76, 177-182, 2001. Etkin N.L., Elisabetsky E. Seeking a transdisciplinary and culturally germane sciences: The future of ethnopharmacology. Journal of Ethnopharmacology 100, 23-26, 2005. 246 Ezeifeka G.O., Orji M.U., Mbata T.I., Patrick A.O. Antimicrobial activities of Cajanus cajan, Garcinia kola and Xylopia aethiopica on pathogenic microorganisms. Biotechnology 3 (1), 41-43, 2004. Fabry W., Okemo P.O., Ansorg R. Antibacterial activity of East African medicinal plants. Journal of Ethnopharmacology 60, 79-84, 1998. Fennel C.W., Lindsey K.L., McGaw L.J., Sparg S.G., Stafford G.I., Elgorashi E.E., Grace O.M., van Staden J. Assessing African medicinal plants for efficacy and safety: pharmacological screening and toxicology. Journal of Ethnopharmacology 94, 205-217, 2004. Fitzgerald D.J., Stratford M., Gasson M.J., Ueckert J., Bos A., Narbad A. Mode of antimicrobial action of vanillin against Escherichia coli, Lactobacillus plantarum and Listeria innocua. Journal of Applied Microbiology 97, 104-113, 2004. Frean J.A., Arntzen L., Rosekilly I., Isaacson M. Investigation of contaminated parenteral nutrition fluids associated with an outbreak of Serratia odorifera septicaemia. Journal of Hospital Infection 27, 263-273, 1994. Gaspar-Marques C., Sim?es M.F., Rodriguez B. A trihomoabietane diterpenoid from Plectranthus grandidentatus and an unusual addition of acetone to the ortho-quinone system of cryptotanshinone. Journal of Natural Products 68, 1408-1411, 2005. Gaspar-Marques C., Rijo P., Sim?es M.F., Duarte M.A., Rodriguez B. Abietanes from Plectranthus grandidentatus and P. hereroensis against methicillin- and vancomycin- resistant bacteria. Phytomedicine 13 (4), 267-271, 2006. Geda A.K. Antibacterial activity of essential oils and their combinations. Fat Science Technology 12, 458-460, 1995. Gelfand M., Mavi, S., Drummond, R.B., Ndemera, S.B. The Traditional Medical Practitioner in Zimbabwe. Mambo Press, Gweru, Zimbabwe, ISBN0123351855, 1985. Giamarellou H. Nosocomial cardiac infections. Journal of Hospital Infection 50, 91-105, 2002. Gibbs R.D. Chemotaxonomy of Flowering Plants. Vol.3. McGill-Queens University Press, USA, ISBN0773500987, 1974. Gibbon C.J. (Editor) South African Medicines Formulary. The Health and Medical Publishing Group of the South African Medical Association, South Africa, ISBN1875098399, 2005. Gibbons S. Anti-staphylococcal plant natural products. Natural Products Report 21, 263-277, 2004. Gilani A.H., Rahman A. Trends in ethnopharmacology. Journal of Ethnopharmacology 100, 43- 49, 2006. Graven E.H., Deans S.G., Svoboda K.P., Mavi S., Gundidza M.G. Antimicrobial and antioxidative properties of the volatile (essential) oil of Artemisia afra Jacq. Flavour and Fragrance Journal 7, 121-123, 1992. 247 Gray J., George R.H., Durbin G.M., Ewer A.K., Hocking M.D., Morgan M.E.I. An outbreak of Bacillus cereus respiratory tract infections on a neonatal unit due to contaminated ventilator circuits. Journal of Hospital Infection 41, 19-22, 1999. Green J., Rondganger L. Polluted drips kill six infants. Gauteng hospitals warned of danger. The Star, 10 March, 2004. Griffin S.G., Wyllie S.G., Markham J.L., Leach D.N. The role of structure and molecular properties of terpenoids in determining their antimicrobial activity. Flavour and Fragrance Journal 14, 322-332, 1999. Griffin S.G., Markham J.L., Leach D.N. An agar dilution method for the determination of the minimum inhibitory concentration of essential oils. Journal of Essential Oil Research 12 (2), 249-255, 2000. Griffin S.G., Wyllie S.G., Markham J.L., Role of the outer membrane of Escherichia coli AG100 and Pseudomonas aeruginosa NCTC 6749 and resistance / susceptibility to monoterpenes of similar chemical structure. Journal of Essential Oil Research 13, 380- 386, 2001. Guenther E. The Essential Oils. Krieger, Florida, USA, ISBN0882750739, 1948. Gundidza M., Deans S.G., Kennedy A.I., Mavi S., Waterman P.G., Gray A.I. The essential oil from Heteropyxis natalensis Harv: Its antimicrobial activities and Phytoconstituents. Journal Science Food Agriculture 63, 361-364, 1993. Gurib-Fakim A. Medicinal plants: Traditions of yesterday and drugs of tomorrow. Molecular Aspects of Medicine 27, 1-93, 2006. Hada T., Inoue Y., Shiraishi A., Hamashima H. Leakage of K+ ions from Staphylococcus aureus in response to tea tree oil. Journal of Microbiological Methods 53, 309-312, 2003. Hamburger M.O., Cordell G.A. A direct bioautographic TLC assay for compounds possessing antibacterial activity. Journal of Natural Products 50 (1), 19-22, 1987. Hammer K.A., Carson C.F., Riley T.V. Antimicrobial activity of essential oils and other plant extracts. Journal of Applied Microbiology 86, 985-990, 1999. Hammer K.A., Carson C.F., Riley T.V., Nielsen J.B. A review of the toxicity of Melaleuca alternifolia (tea tree) oil. Food and Chemical Toxicology 44 (5), 616-625, 2006. Hammerschmidt F.J., Clark A.M., Soliman F.M., El-Kashoury E.A., El-Kawy M.M.A., El- Fishawy A. Chemical composition and antimicrobial activity of essential oils of Jasonia candicans and J. montana. Planta Medica 59, 68-70, 1993. Hanberger H., Diekema D., Fluit A., Jones R., Struelens M., Spencer R., Wolff M. Surveillance of antibiotic resistance in European ICUs. Journal of Hospital Infection 48, 161?176, 2001. Hayes A.J., Leach D.N., Markham J.L., Markovic B. In vitro cytotoxicity of Australian tea tree oil using human cell lines. Journal of Essential Oil Research 9 (5), 1097-1106, 1997. 248 Heinrich M., Barnes J., Gibbons S., Williamson E.M. Fundamentals of Pharmacognosy and Phytotherapy. Churchill Livingstone, London, UK, ISBN0443071322, 2004. Herman P.P.J. Revision of the Tarchonanthus camphoratus complex (Asteraceae- Tarchonantheae) in southern Africa. Bothalia 32 (1), 21-28, 2002. Hewitt W., Vincent S. Theory and Application of Microbiological Assay. Academic Press (Inc), London, UK, ISBN0849318246, 2003. Hili P. The Antimicrobial Properties of Essential Oils. Winter Press, Surrey, UK, ISBN1874581835, 2001. Hinou J.B., Harvala C.E., Hinou E.B. Antimicrobial activity screening of 32 common constituents of essential oils. Pharmazie 44, 302-304, 1989. Holley R.A., Patel D. Improvement in shelf-life and safety of perishable foods by plant essential oils and smoke antimicrobials. Food Microbiology 22 (4), 273-292, 2005. Holmes P. Rosemary oil. The International Journal of Aromatherapy 9 (2), 62-66, 1999. Hood J.R., Wilkinson J.M., Cavanagh H.M.A. Evaluation of common antibacterial screening methods utilized in essential oil research. Journal of Essential Oil Research 15, 428-433, 2003. Hostettmann K. Strategy for the biological and chemical evaluation of plant extracts. Pure Applied Chemistry 70 (11), 1-9, 1999. House D., Bishop A., Parry C., Dougan G., Wain J. Typhiod fever: pathogenesis and disease. Current Opinion in Infectious Diseases 14, 573-578, 2001. Hsieh P-C., Mau J-L., Huang S-H. Antimicrobial effect of various combinations of plant extracts. Food Microbiology 18, 35-43, 2001. http://sophy.u-3mrs.fr/photo-id/Men (Accessed 29/08/04) http://www.ansononline.com/cgi-bin/ez-catalog/cat_display.cgi?0X367200 (Accessed 06/05/06) http://www.climate-zone.com/climate/zimbabwe/celsius/harare.htm (Accessed 09/07/05) http://www.erowid.org/chemicals/absinthe_info3.shtml (Accessed 07/03/05) http://www.lotuspress.com/lotusbrands/tiferetonline/key.html (Accessed 06/05/06) http://www.sahealthinfo.org/tradionalmeds/monographs/artemisia.htm (Accessed 07/03/05) http://www.vitacost.com/science/hn/Herb/Tea_Tree.htm (Accessed 20/10/05) http://www.weathersa.co.za/FcastProducts/LongRange/ViewSeasonEyeBall.jsp (Accessed 07/28/2005) http://yourskin101.com/natural-skin-care/tea-tree-oil.aspx (Accessed 09/11/05) 249 Huffman A., Klepser M.E., Ernst E.J., Keele D., Roling E., Viljoen A.M. Susceptibility of opportunistic bacteria and yeasts to the volatile oils of Artemisia afra, Lippia javanica, Lippia scaberrima, Myrothamnus flabellifolius and Osmitopsis asteriscoides. Journal of Infectious Disease Pharmacotherapy 5 (4), 43-50, 2002. Hutchings A., van Staden J. Plants used for stress-related ailments in traditional Zulu, Xhosa and Sotho medicine. Part 1: Plants used for headaches. Journal of Ethnopharmacology 43, 89- 124, 1994. Hutchings A., Scott A.H., Lewis G., Cunningham A.B. Zulu Medicinal Plants - an Inventory. University of Natal Press, Pietermaritzburg, South Africa, ISBN0869809237, 1996. Hutt A.J., O'Grady J. Drug chirality: a consideration of the significance of the stereochemistry of antimicrobial agents. Journal of Antimicrobial Chemotherapy 37, 7-32, 1996. Inouye S., Takizawa T., Yamaguchi H. Antibacterial activity of essential oils and their major constituents against respiratory tract pathogens by gaseous contact. Journal of Antimicrobial Chemotherapy 47, 565-573, 2001a. Inouye S., Takizawa T., Uchida K., Yamaguchi H. Effect of sealing and Tween 80 on the antifungal susceptibility testing of essential oils. Microbiology and Immunology 45 (3), 201-208, 2001b. Inoue Y., Hada T., Hirose K., Hamashima H., Shimada J. The antibacterial effects of terpene alcohols on Staphylococcus aureus and their mode of action. Federation of European Microbiological Societies Microbiology Letters 237, 325-331, 2004. Ist?riz R.E., Carbon C. Antibiotic use in developing countries. Infection Control Hospital Epidemiology 21 (6), 394-397, 2000. Iwu M.M., Duncan A.R., Okunji C.O. New Antimicrobials of Plant Origin. Perspectives on New Crops and New Uses. ASHS Press, 457-462, 1999. Izzo A.A., Di Carlo G., De Fusco R., Mascolo N., Borrelli F., Capasso F. Biological screening of Italian medicinal plants for antibacterial activity. Phytotherapy Research 9, 281-286, 1995. J?ger A.K. Evaluation of antibacterial activity of traditionally prepared South African remedies for infections. South African Journal of Botany 69 (4), 595-598, 2003. Jain S.R., Kar A. The antibacterial activity of some essential oils and their combinations. Planta Medica 20 (2), 118-123, 1971. Janssen A.M., Scheffer J.J.C., Svendsen A.B. Antimicrobial activity of essential oils: A 1976- 1986 literature review. Aspects of the test methods. Planta Medica 53, 395-398, 1987. Kalemba D., Kunicka A. Antibacterial and antifungal properties of essential oils. Current Medicinal Chemistry 10, 813-829, 2003. Kamatou G.P.P., Viljoen A.M., van Vuuren S.F. In vitro evidence of antimicrobial synergy between Salvia chamelaeagnea and Leonotis leonurus. South African Journal of Botany 72, 634-636, 2006. 250 Kang R., Helms R., Stout M.J., Jaber H., Chen Z., Nakatzu T. Antimicrobial activity of the volatile constituents of Perilla frutescens and its synergistic effects with polygodial. Journal of Agricultural and Food Chemistry 40, 2328-2330, 1992. Kelmanson J.E., J?ger A.K., van Staden J. Zulu medicinal plants with antibacterial activity. Journal of Ethnopharmacology 69, 241-246, 2000. Kerrola K., Galambosi B., Kallio H. Volatile components and odour intensity of four phenotypes of hyssop (Hyssopus officinalis L.). Journal of Agricultural and Food Chemistry 42, 776- 781, 1994. Kerry D.W., Hamilton-Miller J.M.T., Brumfitt W. Trimethoprim and rifampicin: in vitro activities separately and in combination. Journal of Antimicrobial Chemotherapy 1, 417- 427, 1975. Kim J., Marshall M.R., Wei C. Antibacterial activity of some essential oil components against five foodborne pathogens. Journal of Agricultural and Food Chemistry 43, 2839-2845, 1995. Kim H-S. Do not put too much value on conventional medicines. Journal of Ethnopharmacology 100, 37-39, 2005. Kinghorn A.D. Pharmacognosy in the 21st century. Journal of Pharmacy and Pharmacology 53, 135-148, 2001. Kishore N., Chansouria J.P.N., Dubey N.K. Antidermatophytic action of the essential oil of Chenopodium ambrosioides and an ointment prepared from it. Phytotherapy Research 10, 453-455, 1996. Klepser M.E., Ernst M.E., Pfaller M.A. Growth medium effect on the antifungal activity of LY 303366. Diagnostic Microbiology and Infectious Diseases 29, 227- 231, 1997. Klepser M.E., Lewis R.E, Pfaller M.A. Therapy of Candida Infections: Susceptibility testing, resistance, and therapeutic options. The Annals of Pharmacotherapy 32, 1353-1361, 1998. Knobloch K., Pauli A., Iberl B., Weigand H., Weis N. Antibacterial and antifungal properties of essential oil components. Journal of Essential Oil Research 1, 119-128, 1989. Koornhof H. J., Smego R. A., Nicol M. Yersinoisis II: The pathogenesis of Yersinia infections, European Journal of Clinical Microbiology and Infectious Diseases 18, 87-112, 1999. Kosek M., Bern C., Guerrant R.L. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bulletin of the World Health Organisation 81 (3), 197-204, 2003. Kotiranta A., Lounatmaa K., Haapasalo M. Epidemiology and pathogenesis of Bacillus cereus infections. Microbes and Infection 12 (2), 189-198, 2000. Krogstad D.J., Moellering R.C. Antimicrobial Combinations in Antibiotics in Laboratory Medicine. Lorian V., (Editor), 2nd Edition, Williams and Wilkins, UK, ISBN0781749832, 1986. 251 Kr?ner L.U., Padosch C.A., Lachenmeier D.W., Madea B. Investigations on the medico-legal relevance of spirits containing thujone with special regard to toxicological analytical aspects. Blutalkohol 42 (4), 263-271, 2005. Lachowicz K.J., Jones G.P., Briggs D.R., Bienvenu F.E., Wan J., Wilcock A., Coventry M.J. The synergistic preservative effects of the essential oils of sweet basil (Ocimum basilicum L.) against acid-tolerant food microflora. Letters in Applied Microbiology 26, 209-214, 1998. Lahlou M. Methods to study the phytochemistry and bioactivity of essential oils. Phytotherapy Research 18, 435-448, 2004. Lai P.K., Roy J. Antimicrobial and chemopreventive properties of herbs and spices. Current Medicinal Chemistry 11, 1451-1460, 2004. Lall N., Meyer J.J.M. In vitro inhibition of drug-resistant and drug-sensitive strains of Mycobacterium tuberculosis by ethnobotanically selected South African plants. Journal of Ethnopharmacology 66, 347-354, 1999. Lalli J.Y.Y. In vitro pharmacological properties and composition of leaf essential oils and extracts of selected indigenous Pelargonium (Geraniaceae) species. Masters Thesis, Faculty of Health Sciences, University of Witwatersrand, 2006. Lambert R.J.W. Susceptibility testing: inoculum size dependency of inhibition using the Colworth MIC technique. Journal of Applied Microbiology 89, 275-279, 2000. Lambert R.J.W., Pearson J. Susceptibility testing: accurate and reproducible minimum inhibitory concentration (MIC) and non-inhibitory concentration (NIC) values. Journal of Applied Microbiology 88, 784-790, 2000. Lambert R.J.W., Skandamis P.N., Coote P.J., Nychas G-J.E. A study of the minimum inhibitory concentration and mode of action of oregano essential oil, thymol and carvacrol. Journal of Applied Microbiology 91, 453-462, 2001. Lambert R.J.W., Johnston M.D., Hanlon G.W., Denyer S.P. Theory of antimicrobial combinations: biocide mixtures - synergy or addition? Journal of Applied Microbiology 94, 747-759, 2003. Lattaoui N., Tantaoui-Elarki A. Comparative kinetics of microbial destruction by the essential oils of Thymus broussonettii T. zygis and T. satureioides. Journal of Essential Oil Research 6, 165-171, 1994. Laurans C., Orfila J. Moraxella (branhamella) catarrhalis and rti: in vitro activity of cefuroxime. M?decine et Maladies Infectieuses 21 (3), 34-40, 1991. Lawrence B. (Editor) The Antimicrobial / Biological Activity of Essential Oils. Journal of Essential Oil Research Allured, USA, ISBN9781932633139, 1989-2005. Lee J., Carey J., Perlman D.C. Pneumonia and bacteremia due to Serratia odorifera. Journal of Infection 53 (3), 212-214, 2006. 252 Lewu F.B., Grierson D.S., Afolayan A.J. The leaves of Pelargonium sidoides may substitute for its roots in the treatment of bacterial infections. Biological Conservation 128, 582-584, 2006. Light M.E., Sparg S.G., Stafford G.I., van Staden J. Riding the wave: South Africa's contribution to ethnopharmacological research over the last 25 years. Journal of Ethnopharmacology 100, 127-130, 2005. Lin J., Opoku A.R., Geheeb-Keller M., Hutchings A.D., Terblanche S.E., J?ger A.K., van Staden J. Preliminary screening of some traditional Zulu medicinal plants for anti-inflammatory and anti-microbial activities. Journal of Ethnopharmacology 68, 267-274, 1999. Lin Y.T., Labbe R.G., Shetty K. Inhibition of Vibrio parahaemolyticus in seafood systems using oregano and cranberry phytochemical synergies and lactic acid. Innovative Food Science and Emerging Technologies 6, 453-458, 2005. Lindsay D., von Holy A. Different responses of planktonic and attached Bacillus subtilis and Pseudomonas fluorescens to sanitizer treatment. Journal of Food Protection 62, 68-379, 1999. Lindsay D., Brozel V.S., Mostert J.F., von Holy A. Differential efficacy of a chlorine dioxide- containing sanitizer against single species and binary biofilms of a dairy-associated Bacillus cereus and a Pseudomonas fluorescens isolate. Journal of Applied Microbiology 92, 352-361, 2002. Lis-Balchin M., Ochocka R.J., Deans S., Asztemborkas M., Hart S. Bioactivity of the enantiomers of limonene. Medical Science 24, 309-310, 1996. Lis-Balchin. Lavender. Taylor and Francis, USA, ISBN04152848664, 2002. Livermore D.M. Does the resistance epidemiology demand new antibiotics? International Journal of Infectious Diseases 6 (2), S2-S3, 2002. Logan A.C., Beaulne T.M. The treatment of small intestinal bacterial overgrowth with enteric- coated peppermint oil: A case report. Alternative Medicine Review 7 (5), 410-417, 2002. Lourens A.C.U., Reddy D., Ba?er K.H.C., Viljoen A.M., van Vuuren S.F. In vitro biological activity and essential oil composition of four indigenous South African Helichrysum species. Journal of Ethnopharmacology 95, 253-258, 2004. Louw C.A.M., Regnier T.J.C., Korsten L. Medicinal bulbous plants of South Africa and their traditional relevance in the control of infectious diseases. Journal of Ethnopharmacology 82, 147-154, 2002. Lukhoba C.W., Simmonds M.S.J., Paton A.J. Plectranthus: A review of ethnobotanical uses. Journal of Ethnopharmacology 103 (1), 1-24, 2006. Mabogo D.E.N. The Ethnobotany of the Vhavenda. Masters Thesis, University of Pretoria, 1990. Maciel M.A.M., Pinto A.C., Arruda A.C., Pamplona S.G.S.R., Vanderlinde F.A., Lapa A.J., Echevarria A., Grynberg N.F., C?lus I.M.C., Farias R.A.F., Costa A.M.L., Rao V.S.N. Ethnopharmacology, phytochemistry and pharmacology: a successful combination in the study of Croton cajucara. Journal of Ethnopharmacology 70 (1), 41-55, 2000. 253 Mackay M.L., Milne K., Gould I.M. Comparison of methods for assessing synergic antibiotic interactions. International Journal of Antimicrobial Agents 15, 125-129, 2000. Magee A.R., van Wyk B-E., van Vuuren S.F. Ethnobotany and antimicrobial activity of sieketroos (Arctopus species). South African Journal of Botany 73 (1), 159-162, 2007. Magiatis P., Skaltsounis A.L., Chinou I., Haroutounian A.S. Chemical composition and in vitro antimicrobial activity of the essential oils of three Greek Achillea species. Verlag der Zeitschrift f?r Naturforschung 57c, 287-290, 2002. Maistry K. The antimicrobial properties and chemical composition of indigenous Plectranthus (Lamiaceae) species. Masters Thesis, Faculty of Health Sciences, University of Witwatersrand, 2003. Manenzhe N.J., Potgieter N., van Ree T. Composition and antimicrobial activities of volatile components of Lippia javanica. Phytochemistry 65, 2333-2336, 2004. Mangena T., Muyima N.Y.O. Comparative evaluation of the oils of Artemisia afra, Pteronia incana and Rosmarinus officinalis on selected bacteria and yeast strains. Letters in Applied Microbiology 28, 291-296, 1999. Marshall S. Myrrh: magi, medicine and mortality. The Pharmaceutical Journal 273, 919-920, 2004. Marston A., Hostettmann K. Biological and Chemical Evaluation of Plant Extracts and Subsequent Isolation Strategy in Bioassay Methods in Natural Product Research and Drug Development. Bohlin L., Bruhn J.G., (Editors), Kluwer Academic Publishers, USA, ISBN079235480X, 1999. Martini N.D., Katerere D.R.P., Eloff J.N. Biological activity of seven antibacterial flavonoids isolated from Combretum erythrophyllum (Combretaceae). Journal of Ethnopharmacology 93, 207-212, 2004. Martins A.P., Salgueiro L.R., Gon?alves M.J., Vila R., Tomi F., Adzet T., Proen?a da Cunha A., Ca?igueral S., Casanova J. Antimicrobial activity and chemical composition of the bark oil of Croton stellulifer, an endemic species from S. Tom? e Principe. Planta Medica 66, 647-650, 2000. Matasyoh J.C., Kiplimo J.J., Karubiu N.M., Hailstorks T.P. Chemical composition and antimicrobial activity of essential oil of Tarchonanthus camphoratus. Food Chemistry 101 (3), 1183-1187, 2007. Mathekga D.M., Meyer J.J.M. Antibacterial activity of South African Helichrysum species. South African Journal of Botany 64, 293-295, 1998. May J., Chan C.H., King A., Williams L., French G.L. Time-kill studies of tea tree oils on clinical isolates. Journal of Antimicrobial Chemotherapy 45, 639-643, 2000. McGaw L.J., Eloff J.N. Screening of 16 poisonous plants for antibacterial, antihelmintic and cytotoxic activity in vitro. South African Journal of Botany 71, 302-306, 2005. McGaw L.J., Jager A.K., van Staden J. Antibacterial, antihelmintic and anti-amoebic activity in South African medicinal plants. Journal of Ethnopharmacology 72, 247-263, 2000. 254 Meyer J.J.M., Afolayan A.J. Antibacterial activity of Helichrysum aureonitens (Asteraceae). Journal of Ethnopharmacology 47, 109-111, 1995. Meyer J.J.M., Dilika F. Antibacterial activity of Helichrysum pedunculatum used in circumcision rites. Journal of Ethnopharmacology 53, 51-54, 1996. Meyer J.J.M., Afolayan A.J., Taylor M.B., Erasmus D. Antiviral activity of galangin isolated from the aerial parts of Helichrysum aureonitens. Journal of Ethnopharmacology 56, 165- 169, 1997. Moleyar V., Narasimham P. Antibacterial activity of essential oil components. International Journal of Food Microbiology 16, 337-342, 1992. Moon T., Cavanagh H.M., Wilkinson J.M. Which is better for disc diffusion assays of essential oils - IsoSensitest or Nutrient agar? Journal of Essential Oil Research 18, 578-580, 2006. Morrissey I., Hoshino K., Sato K., Yoshida A., Hayakawa I., Bures M.G., Shen L.L. Mechanism of differential activities of ofloxacin enantiomers. Antimicrobial Agents and Chemotherapy 40 (8), 1775-1784, 1996. Motsei M.L., Lindsey K.L., van Staden J., J?ger A.K. Screening of traditionally used South African plants for antifungal activity against Candida albicans. Journal of Ethnopharmacology 86, 235-241, 2003. Mouton J.W., Vinks A.A.T.M.M., Punt N.C. Pharmacokinetic-Pharmacodynamic modelling of activity of ceftazidime during continuous and intermittent infusion. Antimicrobial Agents and Chemotherapy 41 (4), 733-738, 1997. Mulholland D.A. The future of ethnopharmacology: A southern African perspective. Journal of Ethnopharmacology 100, 124-126, 2005. Mwangi J.W., Addae-mensah I., Muriuki G., Munavu R., Lwande W., Hassanali A. Essential oils of Lippia species in Kenya. IV. Maize weevil (Sitophilus zeamais) repellancy and larvicidal activity. International Journal of Pharmacognosy 30, 9-16, 1992. Mwangi J.W., Achola K.J., Lwande W., Hassanali A., Laurant R. Volatile constituents of essential oil of Tarchonanthus camphoratus L. Journal of Essential Oil Research 6, 183- 185, 1994. Nakatsu T., Lupo A.T., Chinn J.W., Kang R.K.L. Biological activity of essential oils and their constituents. Studies in Natural Products Chemistry: Atta-ur-Rahman 21, 571-631, 2000. Natarajan D., Britto S.J., Srinivasan K., Nagamurugan N., Mohanasundari C., Perumal G. Antibacterial activity of Euphorbia fusiformis ? A rare medicinal herb. Journal of Ethnopharmacology 102 (1), 123-126, 2005. NCCLS. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria. 6th Edition, USA, ISBN1562384864, 2003. Negi P.S., Chauhan A.S., Sadia G.A., Rahinishree Y.S., Ramteke R.S. Antioxidant and antibacterial activities of various seabuckthorn (Hippophae rhamnoides L.) seed extracts. Food Chemistry 92, 119-124, 2005. 255 Neirotti E., Moscatelli M., Tiscornia S. Antimicrobial activity of the limonene. Arquivos de Biologia e Tecnologia 39 (2), 233-237, 1996. Neuwinger H.D. African Traditional Medicine: A Dictionary of Plant Use and Applications. Medpharm, Stultgart, Germany, ISBN3887630866, 2000. Ngassapa O., Runyoro D.K.B., Harvala E., Chinou I.B. Composition and antimicrobial activity of essential oils of two populations of Tanzanian Lippia javanica (Burm.f.) Spreng. (Verbenaceae). Flavour and Fragrance Journal 18, 221-224, 2003. Ngwenya M.A., Koopman A., Williams R. Zulu Botanical Knowledge: An Introduction. National Botanical Institute, South Africa, ISBN19196844409, 2003. Njenga E.W., van Vuuren S.F., Viljoen A.M. Antimicrobial activity of Eriocephalus L. species. South African Journal of Botany 71(1), 81-87, 2005. Nychas G-J.E., Tassou C., Skandamis P.N. Making the Most of Herbs, Spices and Their Active Components in Natural Antimicrobials for the Minimal Processing of Foods. Roller S., (Editor), Woodhead Publishing, Ltd., Cambridge, UK, ISBN 1855736691, 2002. Odds F.C. Synergy, antagonism, and what the chequerboard puts between them. Journal of Antimicrobial Chemotherapy 52, 1-3, 2003. Okeke M.I., Iroegbu C.U., Eze E.N., Okoli A.S., Esimone C.O. Evaluation of extracts of the root of Landolphia owerrience for antibacterial activity. Journal of Ethnopharmacology 78 (2- 3), 119-127, 2001. Okeke I.N., Laxminarayan R., Bhutta Z.A., Duse A.G., Jenkins P., O'Brien T.F., Pablos-Mendez A., Klugman K.P. Antimicrobial resistance in developing countries. Part I: recent trends and current status. The Lancet 5 (8), 481-493, 2005. Opalchenova G., Obreshkova D. Comparative studies on the activity of Basil - an essential oil from Ocimum basilicum L. - against multidrug resistant clinical isolates of the genera Staphylococcus, Enterococcus and Pseudomonas by using different test methods. Journal of Microbiological Methods 54, 105-110, 2003. ?zkan G., Sa?dic O., ?zcan M. Note: Inhibition of pathogenic bacteria by essential oils at different concentrations. Food Science and Technology International 9, 85-88, 2003. Pappas R.S., Sturtz G. Unusual alkynes found in the essential oil of Artemisia dracunculus L. var. dracunculus from the Pacific northwest. Journal of Essential Oil Research 13 (3), 187-188, 2001. Pascual M.E., Slowing K., Carretero E., S?nchez M.D., Villar A. Lippia: traditional uses, chemistry and pharmacology: a review. Journal of Ethnopharmacology 76, 201-214, 2001. Pattnaik S., Subramanyam V.R., Kole C.R. Antibacterial and antifungal activity of ten essential oils in vitro. Microbios 86, 237-246, 1996. Pattnaik S., Subramanyam V.R., Bapaji M., Kole C.R. Antibacterial and antifungal activity of aromatic constituents of essential oils. Microbios 89, 39-46, 1997. 256 Paull A., Marks J. A new method for the determination of bactericidal antibiotic synergy. Journal of Antimicrobial Chemotherapy 20, 831-838, 1987 Pauli A., Kubeczka K.H. Evaluation of Inhibitory Data of Essential Oil Constituents Obtained with Different Microbiological Testing Methods in Essential Oils: Basic and Applied Research. Franz C., Mathe A., Buchbauer G., (Editors), Allured, USA, ISBN0931710596, 1997. Pauli A. Antimicrobial properties of essential oil constituents. The International Journal of Aromatherapy 11, 126-133, 2001. Peta M.E., Lindsay D., Br?zel V.S., von Holy A. Susceptibility of food spoilage Bacillus species to chlorine dioxide and other sanitizers. South African Journal of Science 99, 375-380, 2003. Phillipson J.D. Phytochemistry and medicinal plants. Phytochemistry 56 (3), 237-243, 2001. Qadri F., Svennerholm A-M., Faruque A.S.G., Sack R.B. Enterotoxigenic Escherichia coli in developing countries: epidemiology, microbiology, clinical features, treatment and prevention. Clinical Microbiology Reviews 18 (3), 465-483, 2005. Rabe T., van Staden J. Antibacterial activity of South African plants used for medicinal purposes. Journal of Ethnopharmacology 56, 81-87, 1997. Rabe T., van Staden J. Screening of Plectranthus species for antibacterial activity. South African Journal of Botany 64 (1), 6, 1998. Rabe T., van Staden J. Isolation of an antibacterial sesquiterpenoid from Warburgia salutaris. Journal of Ethnopharmacology 73, 171-174, 2000. Radulovi? N., Mananjarasoa E., Harinantenaina L., Yoshinori A. Essential oil composition of four Croton species from Madagascar and their chemotaxonomy. Biochemical Systematics and Ecology 34, 648-653, 2006. Rates S.M.K. Plants as source of drugs. Toxicon 39, 603-613, 2001. Remmal A., Tantaoui-Elaraki A. Improved method for the determination of antimicrobial activity of essential oils in agar medium. Journal of Essential Oil Research 5, 179-184, 1993. Reynolds J.E.F. (Editor) Martindale. The Extra Pharmacopoeia. The Royal Pharmaceutical Society of Great Britain. 31st Edition, London, UK, ISBN0853693420, 1996. Rice L.B. Do we really need new anti-infective drugs? Current Opinion in Pharmacology 3, 459- 463, 2003. Rice L.B. Unmet medical needs in antibacterial therapy. Biochemical Pharmacology 71, 991- 995, 2006. Rios J.L., Recio M.C. Medicinal plants and antimicrobial activity. Journal of Ethnopharmacology 100, 80-84, 2005. 257 Rohlf J.F. NTSYS-pc software package. Numerical taxonomy and multivariate analysis system, Version 2.0 User guide. Allied Biostatics (Inc), New York 1998. Roscoe J.T. Fundamental Research. Statistics for the Behavioural Sciences. 2nd Edition, Holt, Rinehart and Winston (Inc), New York, USA, ISBN0030919347, 1995. Roussis V., Tsoukatou M., Petrikis P.V., Chinou I., Skoula M., Harborne J.B. Volatile constituents of four Helichrysum species growing in Greece. Biochemical Systematics and Ecology 28, 163-175, 2000. Rubin R.J., Harrington C.A., Poon A., Dietrich K., Greene J.A., Moiduddin A. The economic impact of Staphylococcus aureus infection in New York City hospitals. Emerging Infectious Diseases 5 (1), 1999. Ryan R.W., Kwasnik I., Tilton R.C. Methodological variation in antibiotic synergy tests against enterococci. Journal of Clinical Microbiology 13 (1), 73-75, 1981. Santos N.C., Figueira-Coelho J., Martins-Silva J., Saldanha C. Multidisciplinary utilization of dimethyl sulfoxide: pharmacological, cellular and molecular aspects. Biochemical Pharmacology 65, 1035-1041, 2003. Schelz Z., Molnar J., Hohmann J. Antimicrobial and antiplasmid activities of essential oils. Fitoterapia 77, 279-285, 2006. Scholey A.B., Kennedy D.O. Acute, dose-dependent cognitive effects of Ginkgo biloba, Panax ginseng and their combination in healthy young volunteers: differential interactions with cognitive demand. Human Psychopharmacology 17 (1), 35-44, 2002. Scott G., Springfield E.P., Coldrey N. A pharmacognostical study of 26 South African plant species used as traditional medicines. Pharmaceutical Biology 43 (3), 186-213, 2004. Setzer W.N., Volger B., Schmidt J.M., Leahy J.G., Rives R. Antimicrobial activity of Artemisia douglasiana leaf essential oil. Fitoterapia 75, 192-200, 2004. Shale T.L., Strik W.A., van Staden J. Screening of medicinal plants used in Lesotho for anti- bacterial and anti-inflammatory activity. Journal of Ethnopharmacology 67, 347-354, 1999. Shealy C.N. The Illustrated Encyclopaedia of Healing Remedies. Element books, Australia, ISBN1862041873HB, 1998. Shin S-H., Ponikau J.U., Sheeis D.A., Congdon D., Frigas E., Homburger H.A., Swanson M.C., Gleich G.J., Kita H. Chronic rhinosinusitis: An enhanced immune response to ubiquitous airborne fungi. Journal of Allergy and Clinical Immunology 114 (6), 1369-1375, 2004. Silvestre A.J.D., Cavaleiro J.A.S., Feio S.S., Roseiro J.C., Delmond B., Filliatre C. Synthesis of some new benzylic ethers from 1,8-cineole with antimicrobial activity. Monatshefte fur Chemie 130, 589-595, 1999. Sindambiwe J.B., Calomme M., Cos P., Tott? J., Pieters L., Vlietinck A., Vanden Berghe D. Screening of seven selected Rwandan medicinal plants for antimicrobial and antiviral activities. Journal of Ethnopharmacology 65, 71-77, 1999. 258 Southwell I., Hayes A., Markham J., Leach D. The search for optimally bioactive Australian tea tree oil. Acta Horticulture 334, 256-265, 1993. S?derberg T.A., Johansson A., Gref R. Toxic effects of some conifer resin acids and tea tree oil on human epithelial and fibroblast cells. Toxicology 107, 99-109, 1996. Steinmetz M.D., Moulin-Traffort J., R'egli P. Transmission and scanning electron study of the action of Sage and Rosemary essential oils and eucalyptol on Candida albicans. Mycoses 31, 40-51, 1988. Subramoney S. The composition, geographical variation and antimicrobial activity of Lippia javanica (Verbenaceae) leaf essential oils. Masters Thesis, Faculty of Health Sciences, University of Witwatersrand, 2003. Swartz N.M. Impact of antimicrobial agents and chemotherapy from 1972 to 1998. Antimicrobial Agents and Chemotherapy 44 (8), 2009-2016, 2000. Tam V.H., Schilling A.N., Nikolaou M. Modelling time-kill studies to discern the pharmacodynamics of meropenem. Journal of Antimicrobial Chemotherapy 55, 699-706, 2005. Tan X.R., Zheng W.F., Tang H.Q. Biologically active substances from the genus Artemisia. Planta Medica 64, 295-302, 1998. Tassou C., Koutsoumanis K., Nychas G-J.E. Inhibition of Salmonella enteritidis and Staphylococcus aureus in nutrient broth by mint essential oil. Food Research International 33, 273-280, 2000. Taylor J.L.S., Rabe T., McGaw L.J., J?ger A.K., van Staden J. Towards the scientific validation of traditional medicinal plants. Plant Growth Regulation 34, 23-37, 2001. Teixeira A.P., Batista O., Sim?es M.F., Nascimento J., Duarte A., De La Torre M.C., Rodr?guez B. Abietane diterpenoids from Plectranthus grandidentatus. Phytochemistry 44 (2), 325- 327, 1997. Terblanch? F.C., Kornelius G. Essential oil constituents of the genus Lippia (Verbenaceae)- A literature review. Journal of Essential Oil Research 8, 471-485, 1996. Thring T.S.A., Weitz F.M. Medicinal plant use in the Bredasdorp / Elim region of the southern Overberg in the Western Cape Province of South Africa. Journal of Ethnopharmacology 103, 261-275, 2006. Tshikalange T.E., Meyer J.J.M., Hussein A.A. Antimicrobial activity, toxicity and the isolation of a bioactive compound from plants used to treat sexually transmitted diseases. Journal of Ethnopharmacology 96, 515-519, 2005. Tyler V.E. Phytomedicines: back to the future. Journal of Natural Products 62, 1589-1592, 1999. Tzakou O., Pitarokili D., Chinou I.O., Harvala C. Composition and antimicrobial activity of the essential oil of Salvia ringens. Planta Medica 67, 81-83, 2001. 259 Uchida V-M., Miyase T., Yoshizaki F., Bieri J.H., R?edi P., Eugster C.H. Isolation of 14- hydroxytaxodione from Plectranthus grandidentatus G?RKE and of seven new dimeric diterpenoids from P. grandidentatus, P. myrianthus and Coleus carnosus. Structures of grandidone A, 7-Epi-B, C, D and 7-Epi-D. Helvetica Chimica Acta 64 (7), 2227-2251, 1981. Uribe S., Ramirez J., Pe?a A. Effects of ?-pinene on yeast membrane functions. Journal of Bacteriology 161, 1195-1200, 1985. Valentin A., P?lissier Y., Benoit F., Marion C., Kone D., Mallie M., Bastide J.-M, Bessi?re J. Composition and antimalarial activity in vitro of volatile components of Lippia multiflora. Phytochemistry 40 (5), 1439-1442, 1995. Valero M., Salmer?n M.C. Antibacterial activity of 11 essentail oils against Bacillus cereus in tyndallized carrot broth. International Journal of Food Microbiology 85, 73-81, 2003. van Es T., Staskun B., van Vuuren S.F. Sulphur-substituted Pyrrolo [3,4-b] quinolines: synthesis, chemistry and antimicrobial activity. South African Journal of Chemistry 58, 74-81, 2005. van Miert A.S.J.P.A.M. Stereoselectivity in drug development. A clinical perspective? The Veterinary Journal 165 (1), 3-4, 2003. van Vuuren S.F., Viljoen A.M., van Zyl R.L., van Heerden F.R., Ba?er K.H.C. The antimicrobial, antimalarial and toxicity profiles of helihumulone, leaf essential oil and extracts of Helichrysum cymosum (L.) subsp. cymosum. South African Journal of Botany 72, 287-290, 2006. van Vuuren S.F., Viljoen A.M. A comparative investigation of the antimicrobial properties of indigenous South African aromatic plants with popular commercially available essential oils. Journal of Essential Oil Research 18, 66-71, 2006. van Wyk B-E., van Oudtshoorn B., Gericke N. Medicinal Plants of South Africa. Briza, South Africa, ISBN1875093095, 1997. van Wyk B-E., Gericke N. Peoples Plants. A Guide to Useful Plants of Southern Africa. 1st Edition, Briza, South Africa, ISBN1875093192, 2000. van Wyk B-E. A review of ethnobotanical research in southern Africa. South African Journal of Botany 68, 1-13, 2002. van Wyk B-E, Wink M. Medicinal Plants of the World. Briza, South Africa, ISBN1875093443, 2004. van Zyl R.L., Seatholo S., van Vuuren S.F., Viljoen A.M. The biological activities of 20 nature identical essential oil constituents. Journal of Essential Oil Research 18, 129-133, 2006. Velasco-Negueruela A., P?rez-Alonso M.J., Guzm?n C.A., Zygadlo J.A., Ariza-Espinar L., Sanz J., Garc?a-Vallejo M.C. Volatile constituents of four Lippia species from C? rdoba (Argentina). Journal of Essential Oil Research 5, 513-524, 1993. Venter F., Venter J-A. Making The Most of Indigenous Trees. Briza, South Africa, ISBN1875093338, 2002. 260 Verduin C.M., Hol C., Fleer A., van Dijk H., van Belkum A. Moraxella catarrhalis: from emerging to established pathogen. Clinical Microbiology Reviews 15 (1), 125-144, 2002. Verpoorte R., Choi Y.H., Kim H.K. Ethnopharmacology and systems biology: A perfect holistic match. Journal of Ethnopharmacology 100, 53-56, 2005. Viljoen A.M., Klepser M.E., Ernst E.J., Keele D., Rolling E., van Vuuren S.F., Demirci B., Ba?er K.H.C., van Wyk B-E. The composition and antimicrobial activity of the essential oil of the resurrection plant, Myrothamnus flabellifolius. South African Journal of Botany 68, 100-105, 2002. Viljoen A.M., van Vuuren S.F., Ernst E., Klepser M., Demirci B., Ba?er K.H.C., van Wyk B-E. Osmitopsis asteriscoides (Asteraceae) - the antimicrobial activity and essential oil composition of a Cape-Dutch remedy. Journal of Ethnopharmacology 88, 137-143, 2003. Viljoen A.M., Subramoney S., van Vuuren S.F., Ba?er K.H.C., Demicri B. The composition, geographical variation and antimicrobial activity of Lippia javanica (Verbenaceae) leaf essential oils. Journal of Ethnopharmacology 96, 271-277, 2005. Viljoen A.M., van Vuuren S.F., Gwebu L., Demirci B., Ba?er K.H.C. The geographical variation and antimicrobial activity of African Wormwood (Artemisia afra Jacq.) essential oil. Journal of Essential Oil Research 18, 19-25, 2006. Vlietinck A.J., van Hoof L., Tott? J., Lasure A., Vanden Berghe D., Rwangabo P.C., Mvukiyumwami J. Screening of hundred Rwandese medicinal plants for antimicrobial and antiviral properties. Journal of Ethnopharmacology 46, 31-47, 1995. von Baum H., Marre R. Antimicrobial resistance of Escherichia coli and therapeutic implications. International Journal of Medical Microbiology 295 (6), 503-511, 2005. von Koenen E. Medicinal Poisonous and Edible plants in Namibia. Klaus Hess, Namibia, ISBN9991674748, 2001. Wagner H., Bladt S. Plant Drug Analysis. A Thin Layer Chromatography Atlas. 2nd Edition, Springer, Germany, ISBN3540586768, 1995. Walsh F.M., Amyes S.G.B. Microbiology and drug resistance mechanisms of fully resistant pathogens. Current Opinion in Microbiology 7 (5), 439-444, 2004. Watt J.M., Breyer-Brandwijk M.G. The Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd Edition, Livingstone, London, UK, 1962. Weinberger M., Leibovici L., Perez S., Samra Z., Ostfeld I., Levi I., Bash E., Turner D., Goldschmied-Reouven A., Regev-Yochay G., Pitlik S.D., Keller.N. Characteristics of candidaemia with Candida albicans compared with non-albicans Candida species and predictors of mortality. Journal of Hospital Infection 61 (2), 146-154, 2005. Weyerstahl P., Christiansen C., Gundidza M., Mavi S. Constituents of the essential oil of Heteropyxis natalensis. Journal of Essential Oil Research 4, 439-445, 1992. WHO Strategy for Traditional Medicine 2002-2005. 2002. 261 WHO Global and Regional Perspectives on Development of Traditional Medicine. Regional Consultation on Development of Traditional Medicine in the South East Asia Region, Pyongyang, DPR Korea Document 6, 2005. Williamson E.M. Synergy and other interactions in phytomedicines. Phytomedicine 8 (5), 401- 409, 2001. Windholz M. (Editor) An Encyclopaedia of Chemicals and Drugs. 9th Edition, Merck and Co. (Inc), ISBN91910263, 1976. Wootton S. (Editor) Aromatherapy and Natural Health. GE Fabbri Ltd, London, UK, ISBN14636328, 2005. Xiong G.L., Quan D., Maibach H.I. Effects of penetration enhancers on in vitro percutaneous absorption of low molecular weight heparin through human skin. Journal of Controlled Release 42, 289-296, 1996. Yani V.V., Oyedeji O.A., Grierson D.S., Afolayan A.J. Chemical analysis and antimicrobial activity of essential oil extracted from Helichrysum aureonitens. South African Journal of Botany 71(2), 239-241, 2005. Yousef R., Tawil G. Antimicrobial activity of volatile oils. Pharmazie 35, 698-701, 1980. 262 Appendix A Raw MIC data (mg/mL) for isobolograms Table A1 Raw MIC data (mg/mL) for isobolograms (Chapter 7) where C. gratissimus roots and leaves were combined. Combined MIC (roots and leaves) in combination (mg/mL) Ratio B. cereus S. aureus S. epidermidis B. subtilis E. faecalis K. pneumoniae P. aeruginosa E. coli C. albicans C. neoformans 100:0 0.8 2.0 2.0 3.0 4.0 2.0 2.0 4.0 8.0 0.5 90:10 1.0 4.0 2.0 8.0 2.0 4.0 4.0 16.0 8.0 0.5 80:20 1.0 2.0 2.0 4.0 2.0 2.0 2.0 8.0 4.0 0.5 70:30 1.0 2.0 2.0 4.0 2.0 2.0 2.0 4.0 3.0 0.6 60:40 1.5 2.0 2.0 4.0 4.0 2.0 2.0 4.0 4.0 0.5 50:50 0.5 2.0 2.0 4.0 1.4 2.0 2.0 4.0 3.0 0.5 40:60 0.5 2.0 2.0 2.0 4.0 2.0 2.0 4.0 3.2 0.3 30:70 0.5 2.0 2.0 2.0 4.0 2.0 2.0 4.0 3.0 0.5 20:80 1.0 2.0 2.0 2.0 4.0 2.0 2.0 4.0 3.0 0.5 10:90 2.0 2.0 2.0 2.0 4.0 2.0 2.0 4.0 8.0 1.0 0:100 2.0 2.0 2.0 3.0 4.0 2.0 2.0 4.0 3.0 1.0 263 Combined MIC of limonene in combination with 1,8-cineol (mg/mL) S. aureus P. aeruginosa C. neoformans Ratio (+)-limonene (-)-limonene (+/-)-limonene (+)-limonene (-)-limonene (+/-)-limonene (+)-limonene (-)-limonene (+/-)-limonene 100:0 16.0 4.0 4.0 4.0 8.0 4.0 2.0 2.0 2.0 90:10 16.0 16.0 8.0 6.0 8.0 8.0 4.0 2.0 4.0 80:20 16.0 9.6 6.0 6.0 8.0 8.0 4.0 2.0 2.0 70:30 16.0 10.4 6.0 6.0 8.0 4.0 2.0 2.0 2.0 60:40 7.0 17.6 6.0 4.0 8.0 4.0 2.0 2.0 2.0 50:50 8.0 8.0 6.0 4.0 8.0 4.0 2.0 2.0 2.0 40:60 8.0 8.0 6.0 4.0 8.0 4.0 2.0 2.0 2.0 30:70 8.0 8.0 8.0 4.0 8.0 4.0 2.0 2.0 2.0 20:80 8.0 8.0 8.0 4.0 8.0 4.0 2.0 2.0 2.0 10:90 8.0 8.0 8.0 4.0 8.0 4.0 2.0 2.0 3.0 0:100 8.0 8.0 8.0 4.0 8.0 6.0 2.0 1.5 2.0 Table A2 Raw MIC data (mg/mL) for isobolograms (Chapter 8) where (+), (-) and the racemic form of limonene was combined with 1,8- cineol. 264 Table A3 Raw MIC data (mg/mL) for isobolograms (Chapter 9) where the volatile and non-volatile constituents for T. camphoratus were combined. Combined MIC (volatile and non-volatile) in combination (mg/mL) for P. grandidentatus Ratio S. aureus E. faecalis K. pneumoniae 100:0 2.0 8.0 4.0 90:10 0.3 1.0 2.0 80:20 0.5 1.0 2.0 70:30 0.5 1.0 1.0 60:40 0.5 1.0 1.0 50:50 0.5 2.0 1.0 40:60 0.2 4.0 1.0 30:70 1.0 1.0 1.0 20:80 0.5 3.0 1.0 10:90 1.0 4.0 1.0 0:100 0.4 32.0 8.0 Combined MIC (volatile and non-volatile) in combination (mg/mL) for T. camphoratus Ratio B. cereus S. aureus E. faecalis K. pneumoniae P. aeruginosa E. coli C. albicans C. neoformans 100:0 3.0 4.0 8.0 8.0 4.0 8.0 6.0 3.0 90:10 1.0 2.0 4.0 2.0 8.0 8.0 2.0 1.5 80:20 1.0 1.5 4.0 2.0 3.0 4.0 2.0 1.5 70:30 1.0 1.5 4.0 4.0 3.0 4.0 2.0 2.0 60:40 0.7 1.5 4.0 4.0 3.0 4.0 2.0 2.0 50:50 0.5 1.5 4.0 4.0 3.0 4.0 2.0 2.0 40:60 0.5 1.0 4.0 4.0 4.0 8.0 2.0 2.0 30:70 0.5 1.0 4.0 4.0 4.0 8.0 2.0 1.0 20:80 0.5 1.0 4.0 4.0 4.0 8.0 2.0 1.0 10:90 0.5 1.0 4.0 4.0 4.0 8.0 3.0 1.0 0:100 2.0 2.0 4.0 8.0 4.0 4.0 4.0 3.0 Table A4 Raw MIC data (mg/mL) for isobolograms (Chapter 9) where the volatile and non-volatile constituents for P. grandidentatus were combined. 265 Appendix B Publications arising from this study 266 267 268 269 270 271 272