INDIGENOUS SALVIA SPECIES ? AN INVESTIGATION OF THEIR PHARMACOLOGICAL ACTIVITIES AND PHYTOCHEMISTRY Guy Paulin Poungou? Kamatou A thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, in fulfillment of the requirements for the degree of Doctor of Philosophy. Johannesburg, 2006 ii Declaration I, Guy Paulin Poungou? Kamatou, declare that this thesis is my own work except where acknowledged. It is being submitted for the Degree of Doctor of Philosophy at the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination at any other University. ????????????????????????????????????.. (Signature of candidate) ???????????..day of ??????????????????....?.2006 iii This thesis is dedicated to my Lord Jesus Christ and the family Poungou? for their love and support. iv Publications Arising from this Study Kamatou, G.P.P., Viljoen, A.M., Gono-Bwalya, A.B., van Zyl, R.L., van Vuuren , S.F., Lourens, A.C.U., Ba?er, K.H.C., Demirci, B., Lindsey, K. L., van Staden, J., Steenkamp, P. 2005. The in vitro pharmacological activities and a chemical investigation of three South African Salvia species. Journal of Ethnopharmacology 102, 382?390 (Abstract in Appendix A). Kamatou, G.P.P., Viljoen, A.M., van Zyl, R.L., van Vuuren, S.F., Figueiredo, A.C., Barroso, J.G., Pedro, L.G., Tilney, P.M. 2006. Chemical composition, leaf trichome types and biological activities of essential oils of four related Salvia species indigenous to South Africa. Journal of Essential Oil Research 18, 72?79 (Abstract in Appendix A). Kamatou, G.P.P, Viljoen, A.M. van Vuuren, S.F., van Zyl, R.L. 2006. In vitro evidence of antimicrobial synergy between Salvia chamelaeagnea and Leonotis leonurus. South African Journal of Botany 72, 634?637. (Abstract in Appendix A) Kamatou, G.P.P., Viljoen, A.M., van Zyl, R.L., van Vuuren, S.F., Figueiredo, A.C., Barroso, J.G., Pedro, L.G. 2006. Investigation of biological activities of essential oil of Salvia muirii L. Bol. Journal of Essential Oil Research 18, 48?51 (Abstract in Appendix A). Kamatou, G.P.P., Viljoen, A.M., Figueiredo, A.C., Tilney, P.M., van Zyl, R.L., van Vuuren, S.F., Pedro, L.G., Barroso, J.G. 2006. Glandular trichomes, essential oil composition and biological activities of Salvia albicaulis Benth. and S. dolomitica Codd , two species from the Cape region of South Africa. South African Journal of Botany , in press (Abstract in Appendix A). Kamatou, G.P.P., van Vuuren, S.F., van Heerden, F.R., Seaman, T., Viljoen, A.M. 2006. Antibacterial and antimycobacterial activities of South African indigenous Salvia species and isolated compounds from S. chamelaeagnea (Abstract in Appendix A, Submitted to Pharmaceutical Biology). v Conferences/presentations Kamatou, G.P.P., Viljoen, A.M., Gono-Bwalya, A.B., van Zyl, R.L., van Vuuren , S.F., Lourens, A.C.U., Ba?er, K.H.C. 2004. In vitro biological properties of Salvia stenophylla, S. repens and S. runcinata. Poster presentation at the 30th Annual Congress of the South African Association of Botanists, 18th-22nd 2004, University of KwaZulu-Natal, Durban South Africa (Abstract in Appendix B1). Kamatou, G.P.P., Viljoen, A.M., van Zyl, R.L., van Vuuren, S.F., Davids, D.H., van Heerden, F.R. and Seaman, T. 2005. Biological activities of solvent extracts and essential oils of four related Salvia species. Podium presentation, 5th Annual Meeting of the Indigenous Plant Use Forum (IPUF), Rhodes University, 27-30 June 2005, Grahamstown, South Africa (Abstract in Appendix B2). Kamatou, G.P.P. Viljoen, A.M., Tilney, P.M., van Vuuren, S.F. and van Zyl R.L. 2005. Trichome morphology and the role of leaf essential oils in the pharmacological activity recorded for indigenous Salvia species. Podium presentation at the University of Johannesburg, Post Graduate Symposium, Johannesburg, 2 November 2005 (Abstract in Appendix B3). Kamatou, G.P.P., Viljoen, A.M., van Zyl, R.L., van Vuuren, S.F., Davids, D.H., Seaman, T. Figueiredo, C. 2006. Biological activities, essential oils composition and leaf trichome morphology of selected Salvia species indigenous to southern Africa. Podium presentation at the 32th Annual Congress of the South African Association of Botanists, 16-19 January 2006, Nelson Mandela Metropolitan University, Port Elizabeth, South Africa (Abstract, South African Journal of Botany 2006. 72, 323 (Abstract in Appendix B4). Kamatou, G.P.P, van Zyl, R.L., Viljoen, A.M., Davids, H., van Heerden, F.R. 2006. The antimalarial activity and cytotoxic effects of solvent extracts of South African Salvia species and isolated compounds from S. radula. 4th International Conference on Pharmaceutical and Pharmacological Sciences (ICPPS), 20-23 September 2006, Gauteng, South Africa (Abstract in Appendix B5) vi Abstract The genus Salvia belongs to the family Lamiaceae and encompasses 900 species worldwide of which 26 are found in southern Africa and many of them are used in local traditional medicine. However, the phytochemistry and pharmacological activities of the South African species have not been extensively investigated. The leaf trichome morphology that may be used to distinguish species was investigated with the scanning electron and light microscopy. Both glandular (capitate or peltate) and non-glandular trichomes were identified in all species. The essential oils were isolated by hydro-distillation and analysed by GC and GC-MS methods. The oil yield was relatively low and ranged from 0.004 (S. radula) to 0.50% (S. muirii) (w/w). Major components identified include ?-pinene, 1,8-cineole, linalool, limonene, myrcene, ?-caryophyllene, spathulenol, ?-caryophyllene oxide, viridiflorol, ?-3- carene and ?-bisabolol. High performance liquid chromatography analysis was used to identify phenolic compounds in 17 solvent extracts. Betulafolientriol oxide was detected in all species. Rosmarinic acid was only absent in S. verbenaca, while S. garipensis and S. radula were the only species which lacked oleanolic acid/ursolic acid. Various in vitro biological activities were investigated. Nearly all the solvent extracts displayed anti-oxidant activity with IC50 values ranging from 1.61 to 74.50 ?g/ml using the DPPH? radical, while the IC50 values ranged from 11.88 to 69.26 ?g/ml with the ABTS?+ radical. The solvent extract of S. schlechteri was three times more active than vitamin C. Total phenolic content based on gallic acid equivalents (GAE) revealed the presence of total soluble phenolics in the extract at 45 to 211 mg of GAE dry sample. Almost all the essential oils exhibited promising anti-inflammatory activity (5-lipoxygenase assay) with IC50 values ranging from 22.81 to 77.32 ?g/ml. The antimalarial activity was determined using [3H]-hypoxanthine method on the Plasmodium falciparum (FCR-3) strain. The IC50 values of the essential oils ranged from 1.20 to 13.50 ?g/ml and were low compared to the solvent extracts (IC50 values ranging from 3.91 to 26.01 ?g/ml). Betulafolientriol oxide and salvigenin isolated from S. radula inhibited the growth of malaria parasites with IC50 values of 4.95 and 24.60 ?g/ml, respectively. With the exception of S. radula, all the solvent extracts displayed moderate to good activity against Staphylococcus aureus, vii Bacillus cereus, Klebsiella pneumoniae, Escherichia coli and Mycobacterium tuberculosis with the MIC values ranging from 0.03 to 8.00 mg/ml. Four compounds, namely carnosol, 7-O-methylepirosmanol, oleanolic acid and its isomer ursolic acid were isolated from S. chamelaeagnea as the active principles against S. aureus. The solvent extracts of Salvia species were tested for in vitro anticancer activity against human breast adenocarcinoma (MCF-7), colon adenocarcinoma (HT-29) and glioblastoma (SF-268) using the sulforhodamine B assay. The extracts inhibited cell proliferation of all three cell lines to varying degrees, with the IC50 values ranging between 9.69 and 43.65 ?g/ml and 8.72 and 59.12 ?g/ml against the MCF-7 and SF-268 cell lines, respectively. The IC50 values against the HT-29 cell line ranged from 17.05 to 57.00 ?g/ml. The in vitro toxicity profile of 28 samples (17 solvent extracts and 11 essential oils) was evaluated on human kidney epithelial cells using the 3-(4,5-dimethylthiazol-2-yl)-2,5 dimethyl tetrazolium bromide method. The samples displayed some degree of toxicity with IC50 values ranging from 1.79 to 22.9 ?g/ml for the essential oils and from 12.12 to 53.34 ?g/ml for the solvent extracts. The essential oil composition of S. africana-caerulea, S. africana-lutea and S. lanceolata, collected at the same locality throughout the 2004/2005 growing season, was compared in terms of essential oil yields, chemical composition and biological activities. Mostly quantitative, rather than qualitative variation was observed. Major seasonal fluctuations of certain essential oil compounds were observed in all three species. Variations in biological activities of the solvent extracts over seasons were noted. The biological activities of the solvent extracts of three Salvia species (Salvia africana- caerulea, S. africana-lutea and S. lanceolata) were evaluated in the presence and absence of essential oils. The solvent extract of S. africana-caerulea without essential oil exhibited the best activity against Gram-positive bacteria (MIC value: 0.1 mg/ml), while the solvent extract containing essential oil of S. africana-lutea was the most active against Gram- negative bacteria. The toxicity profile of all three species was significantly higher (P < 0.05) with the solvent extracts containing essential oils. The in vitro biological activities add scientific support to the use of Salvia species in traditional medicine. viii Acknowledgements This study would never have been completed without support from a number of people. I am highly indebted to my supervisors, Prof. A.M. Viljoen and Dr. R.L. van Zyl for their endless guidance, commitment and constructive criticism throughout the duration of this study. I also thank them for the financial support I received and acknowledge the contribution of members of the Research Committee for their advice. This work, being multidisciplinary, relied on input from several individuals with specific expertise. I collaborated with a number of laboratories and I would like to say thank you for the facilities offered to me as well as to the staff members who generously rendered support. The analysis of the essential oils was carried out at the Universidade de Lisboa (Portugal) under the supervision of Prof. A.C. Figueiredo, Prof. J.G. Barroso and Prof. L.G. Pedro and I would like to thank them for welcoming me to Lisbon and for the opportunity offered to me to visit and use their facilities. My sincere appreciation also goes to Dr. P.M. Tilney (University of Johannesburg) and Mr. C. van Merwe (University of Pretoria) for their assistance and generosity during the leaf trichome morphology assessment. I also acknowledge the contribution of Mrs. S.F. van Vuuren (University of the Witwatersrand) and Miss. T. Seaman (University of Cape Town) for their assistance throughout the antibacterial and antimycobacterial investigation and constructive criticism. Thank you to Dr. H. Davids (University of the Witwatersrand) and Mrs. A.C. Lourens (University of KwaZulu-Natal) who supervised the anticancer part of this study. My gratitude is extended to Prof. F.R. van Heerden (University of KwaZulu-Natal) who helped me to structurally characterise the isolated compounds and to Dr. P. Steenkamp (CSIR, Pretoria) who assisted me with the HPLC analysis. ix Dr. A. Summerfield and the personnel of the South African National Biodiversity Institute (Pretoria) are acknowledged for the collection of plant materials and identification, respectively. I would like to express my appreciation to Prof. B.E. van Wyk and Prof. M. Scholes for the advice and encouragement throughout the duration of this study. My thanks are extended to Mrs. P. Modipa, Mrs. R. Poultney, Mr. D. McCallum for their constructive criticisms and to Dr. J. Lotter and Mr. A. Magee for their technical assistance. I am grateful to M M. Ngounou, G. Toko, J. Mouatcho; my classmates especially Miss. M. Paraskeva and Miss. M.-J. Huang with whom I shared great times. I finally thank the technical and administrative staff of the Department of Pharmacy and Pharmacology of the University of the Witwatersrand. To all those who have contributed to the completion of this thesis and whose names do not appear here, I would like to express my sincere gratitude. I am thankful for financial support in the form of bursaries received from the National Research Foundation (Indigenous Knowledge Systems) and the Faculty Research Committee Grant (University of the Witwatersrand). x Awards Best poster presentation (3rd place) at the at the 30th Annual Congress of the South African Association of Botanists, University of KwaZulu-Natal, Durban, South Africa. Best podium presentation (2nd place) at the 5th Annual Meeting of the Indigenous Plant Use Forum. Grahamstown, 27-30 June 2005. xi Table of Contents DECLARATION????????????????????????????...ii DEDICATION????????????????????????????......iii PUBLICATION ARSING FROM THIS STUDY????????????.....??...iv CONFERENCES/PRESENTATIONS...............????????????????...v ABSTRACT??????????????????????????.??...?..vi ACKNOWLEDGEMENTS????????????????????????viii AWARDS???????????????????????????????..x TABLE OF CONTENTS......................................................................................................xi LIST OF FIGURES???????..???????????????????xviii LIST OF TABLES???????????????????????????xxii LIST OF ACRONYMS AND SYMBOLS.......................................................................xxiv Chapter 1: General Introduction 1.1 Introduction???????????????????????????....1 1.2 The role of secondary metabolites from plants?????????????......2 1.3 Natural compounds and chemotherapeutic agents????????????.....4 1.4 Approaches to plant-based research??????????????????...5 1.5 Future of the natural compounds research???????????????.....5 1.6 The genus Salvia: description, history and traditional uses in South Africa??.....6 1.6.1 Morphological description??????????????????.....6 1.6.2 History??????????????????????????....7 1.6.3 Traditional uses of Salvia species??????????...?????...8 1.7 Chemical constituents of Salvia species????????????????....8 1.7.1 Volatile compounds?????????????????????...8 1.7.2 Non-volatile compounds???????????????????....9 1.8 Biological activities of Salvia species?????????????????..10 1.9 Rationale and motivation of the study????????????????......10 1.10 Aim of the study?????????????????????????...12 1.11 Objectives of the study??????????????????????.....12 1.12 Outline of the thesis???????????????????????......13 xii Chapter 2: Plant Collection, Isolation of Volatiles and Extraction of Non-Volatiles 2.1 Species selection?????????????????????????...16 2.2 Collection of plant materials????????????????????....16 2.3 Preparation of samples??????????????????????.....16 2.3.1 Preparation of the essential oils (volatile fractions)????????.....16 2.3.2 Preparation of the solvent extracts (non-volatile fractions)?????.....17 Chapter 3: Leaf Trichome Types Abstract....................................................................................................................19 3.1 Introduction???????????????????????????..20 3.2 Materials and methods?????..????????????.?????...21 3.2.1 Plant material???????????????????????....21 3.2.2 Preparation of solutions and stain?............................................................21 3.2.3 Morphological and anatomical studies of leaf trichomes??????.....22 3.3 Results?????????????????????????????..23 3.3.1 Scanning electron microscopy observations??.......??????.........23 3.3.2 Light microscopy observations.......??????????????.....30 3.4 Discussion???????????????????????????....40 3.5 Conclusions??????????????????????????......44 Chapter4: Essential Oil Composition Abstract....................................................................................................................45 4.1 Introduction ?????????????????????????.........46 4.2 Materials and methods??????????????????????......47 4.2.1 Gas chromatography (GC) ?????????????????......47 4.2.2 Gas chromatography coupled to mass spectrometry (GC-MS) ???..48 4.3 Data analysis and chemotaxonomic relationship among species ??????....48 4.4 Results????????????????????????????......49 4.4.1 Percentage yield??????????????????????...49 4.4.2 Essential oil composition??????????????????......49 4.4.3 Cluster analysis??????????????????????.....57 4.5 Discussion???????????????????????????....58 4.6 Conclusions???????????????????????????..60 xiii Chapter 5: HPLC-UV-MS Analysis of the Solvent Extracts Abstract....................................................................................................................62 5.1 Introduction?????????????????????????..........63 5.2 Materials and methods?????????????????????.........64 5.2.1 Chemical and reagents and standards?????????????.......64 5.2.2 Preparation of samples.................................................................................64 5.2.3 High performance liquid chromatography...................................................64 5.3 Data analysis.............................................................................................................65 5.4 Results......................................................................................................................65 5.4.1 Identification of compounds.........................................................................65 5.4.2 Detection of other flavonoids in the solvent extracts...................................67 5.5 Discussion................................................................................................................72 5.6 Conclusions..............................................................................................................73 Chapter 6: Anti-Oxidant Activity and Total Phenolic Content Abstract....................................................................................................................74 6.1 Introduction??????????????????????????.....75 6.2 Materials and methods??????????????????????.....76 6.2.1 Reagents, chemicals and standards ??????????????.....76 6.2.2 Evaluation of the anti-oxidant activity?????????????.....76 6.2.3 Determination of total phenolic content?????????????...80 6.3 Data analysis??????????????????????????.....81 6.4 Results?????????????????????????????..82 6.4.1 Qualitative evaluation of the anti-oxidant activity of the solvent extracts.........................................................................................................82 6.4.2 The quantitative determination of the anti-oxidant activity?????.....83 6.4.3 Total phenolic content and relationship between the total phenolic content and anti-oxidant activity????????????????..85 6.5 Discussion???????????????????????????....87 6.6 Conclusions???????????????????????????..90 Chapter 7: Anti-inflammatory activity-Inhibition of the 5- Lipoxygenase Enzyme Abstract....................................................................................................................91 7.1 Introduction???????????????????????????..92 xiv 7.2 Materials and methods??????????????????????.....93 7.2.1 Chemicals, reagents and standard ???????????????...93 7.2.2 Preparation of solutions, enzymes and buffers??????????....94 7.2.3 Preparation of test samples??????????????????..94 7.2.4 The 5-lipoxygenase assay??????????????????....94 7.3 Data analysis??????????????????????????....95 7.4 Results?????????????????????????????..95 7.5 Discussion???????????????????????????....97 7.6 Conclusions???????????????????????????..99 Chapter 8: The Antimalarial Activity of the Essential Oils and Solvent Extracts and Isolated Compounds from Salvia radula Abstract..................................................................................................................100 8.1 Introduction??????????????????????????....101 8.2 Materials and methods??????????????????????...103 8.2.1 Chemicals, reagents and standards??????????????.....103 8.2.2 Preparation of different solutions???????????????...103 8.2.3 Plasmodium falciparum and in vitro culture??????????......105 8.2.4 Preparation of test samples?????????????????.....105 8.2.5 Assay of antimalarial activity????????????????.....106 8.2.6 Isolation and structural characterization procedures from S. radula??..107 8.3 Data analysis??????????????????????????...109 8.4 Results????????????????????????????....110 8.4.1 The antimalarial activity of the essential oils solvent extracts???????...110 8.4.2 Identification and characterization of the isolated compounds???????..113 8.5 Discussion???????????????????????????..116 8.6 Conclusions??????????????????????????....119 Chapter 9: Antibacterial and Antimycobacterial Activities of Salvia Species and Isolated Compounds from S. chamelaeagnea Abstract..................................................................................................................120 9.1 Introduction??????????????????????????....121 9.2 Materials and methods??????????????????????....122 9.2.1 Reagents, chemicals and drugs????????????????...122 9.2.2 Source and maintenance of micro-organisms?????????........123 xv 9.2.3 Sample preparation????????????????????....124 9.2.4 Determination of the antibacterial and antimycobacterial activities??.124 9.2.5 Isolation of antibacterial active compounds from S. chamelaeagnea?...127 9.3 Results???????????????????????????.......130 9.3.1 Antibacterial activity????????????????????..130 9.3.2 Antimycobacterial activity?????????????????.....134 9.3.3 Elucidation and identification of the isolated compound from S. chamelaeagnea?????????????????????...136 9.4 Discussion??????????????????????????......141 9.5 Conclusions??????????????????????????....147 Chapter 10: In Vitro Anticancer Activity of the Solvent Extracts Abstract..................................................................................................................148 10.1 Introduction?????????????????????????........149 10.2 Materials and methods??????????????????????...151 10.2.1 Chemicals and drugs ???????????????????.....151 10.2.2 Preparation of media and cell maintenance ??????????.......151 10.2.3 The sulforhodamine B assay?????????????????..153 10.3 Data analysis??????????????????????????..155 10.4 Results????????????????????????????....155 10.5 Discussion??????????????????????????......159 10.6 Conclusions??????????????????????????....161 Chapter 11: In Vitro Evaluation of Toxic Effects of the Essential Oils and Solvent Extracts Abstract..................................................................................................................163 11.1 Introduction??????????????????????????....164 11.2 Materials and methods??????????????????????....165 11.2.1 Chemicals, reagents and reference compound ?????????......165 11.2.2 Preparation of various media and solutions???????????...165 11.2.3 Cell maintenance???????????????????.??..166 11.2.4 Cell suspension and sample preparations????????????..166 11.3 Data analysis??????????????????????????..168 11.4 Results????????????????????????????....168 xvi 11.5 Discussion???????????????????????????..171 11.6 Conclusions??????????????????????????....173 Chapter 12: Seasonal Variation in Essential Oil Composition and Biological Activities of Three Salvia Species Abstract..................................................................................................................174 12.1 Introduction??????????????????????????....175 12.2 Materials and methods??????????????????????...176 12.2.1 Plant material??????????????????????.....176 12.2.2 Isolation of the essential oils and preparation of the solvent extracts ?...176 12.2.3 Evaluation of the biological activities?????????????.....176 12.3 Data analysis??????????????????????????...177 12.4 Results????????????????????????????....178 12.4.1 Variation in yield and essential oil composition?????????....178 12.4.2 TLC finger print of the solvent extracts over seasons?????............180 12.4.3 Seasonal variation in biological activities????????????..181 12.5 Discussion??????????????????????????......185 12.6 Conclusions??????????????????????????....188 Chapter 13: The Influence of Essential Oils on the Biological Activities of Solvent Extracts Abstract..................................................................................................................190 13.1 Introduction?????????????????????????........191 13.2 Materials and methods?????????????........................................191 13.2.1 Plant material???????....???????????..................191 13.2.2 Isolation of the essential oils and preparation of the solvent extracts??191 13.2.3 Comparison of the TLC plates of the solvent extract with and without essential oils...............................................................................................192 13.2.4 Evaluation of the biological activities?????????????....192 13.3 Data analysis??????????????????????????...193 13.4. Results???????????????????????????..?..193 13.4.1 Comparison of the TLC plates of the solvent extracts with and without the essential oils...............................................................................................193 13.4.2 The antibacterial activity??????????????????....194 13.4.3 The anticancer activity???????????????????...195 xvii 13.4.4 The antimalarial activity?????????????????.........197 13.4.5 The anti-oxidant activity??????????..????????..197 13.4.6 The toxicity profile?????????????????????.198 13.5 Discussion???????????????????????????..199 13.6 Conclusions??????????????????????????....200 Chapter 14: General Conclusions and Recommendations 14.1 General conclusions ???????????????????????..202 14.1.1 Leaf trichome types and essential oil composition....................................202 14.1.2 Identification of phenolic compounds in the solvent extracts....................203 14.1.3 Biological activities and isolated compounds form two Salvia species.....203 14.1.4 Seasonal variation on essential oil composition and biological activities and the influence of the essential oil in the biological activities of the solvent extracts.................................................................................205 14.2 Recommendations.........................................?????????????.....205 References.........................................................................................................................206 Appendix A: Abstracts of papers published/submitted from this thesis..........................229 Appendix B: Abstracts of conferences/presentations.......................................................236 Appendix C: Monographs of Salvia species investigated................................................244 Appendix D: 1H and 13C NMR spectra of compounds isolated from Salvia radula........295 Appendix E: 1H and 13C spectra of compounds isolated from Salvia chamelaeagnea...297 Appendix F: Seasonal variation in essential oil composition of Salvia africana-caerulea, S. africana-lutea and S. lanceolata........................................................................300 xviii List of Figures Figure 2.1 Clevenger apparatus used for the isolation of the essential oils ?.......??....18 Figure 3.1 Leaf area selected for the study of trichome morphology..???????...22 Figure 3.2 SEM micrographs of Salvia africana-caerulea and S. africana-lutea???...24 Figure 3.3 SEM micrographs of Salvia albicaulis and S. aurita?????????.....25 Figure 3.4 SEM micrographs of Salvia chamelaeagnea and S. disermas?????.......26 Figure 3.5 SEM micrographs of Salvia dolomitica and S. garipensis?????...??..27 Figure 3.6 SEM micrographs of Salvia lanceolata and S. muirii??????????28 Figure 3.7 SEM micrographs of Salvia repens and S. schlechteri?????????..29 Figure 3.8 LM micrographs showing transverse sections of Salvia africana-caerulea and S. africana-lutea???????????????????????...30 Figure 3.9 LM micrographs showing transverse sections of Salvia albicaulis and S. aurita???????????????????????????.31 Figure 3.10 LM micrographs showing transverse sections of Salvia chamelaeagnea and S. disermas??????????????????????...???..32 Figure 3.11 LM micrographs of transverse sections of Salvia dolomitica and S. garipensis?????????????????????????...33 Figure 3.12 LM micrographs of transverse sections of Salvia lanceolata and S. muirii....34 Figure 3.13 LM micrographs of transverse sections of Salvia repens and S. schlechteri...35 Figure 3.14 Various types of glandular and non-glandular trichomes found in Salvia species?????????????????????????...36 Figure 4.1 Gas chromatography and gas chromatography coupled to mass spectrometry apparatus.????????????????????????????..47 Figure 4.2 Percentage essential oil yields obtained for eleven indigenous Salvia species?.......................................................................................................49 Figure 4.3 Variation in relative amounts of common and major components of the essential oils of eleven indigenous Salvia species ?????????...54 Figure 4.4 Mass spectra of the unidentified compounds present in Salvia dolomitica oil?........................................................................................................55 Figure 4.5 Cluster analysis of eleven Salvia species constructed using the essential oil composition data.????????????????????????......57 Figure 5.1 UV spectra of standards and isolated compounds from S. radula and S. chamelaeagnea????????????????????..???66 xix Figure 5.2 HPLC chromatograms of South African Salvia species....................................70 Figure 6.1 Mechanism of action of the ABTS?+ radical ???.??????????78 Figure 6.2 Mechanism of action of the DPPH? radical.????...???????..?.79 Figure 6.3 Microtitre plate showing the decolourisation of the DPPH solution in the presence of anti-oxidant compounds ????????????????.......80 Figure 6.4 Free radical scavenging activity of five Salvia species against the DPPH????????????????????????????.....81 Figure 6.5 Calibration curve of gallic acid used to estimate the total phenolic content in solvent extracts, plotted with a 95% confidence interval.??????????82 Figure 6.6 Qualitative determination of the anti-oxidant activity of the solvent extracts.........................................................................................................83 Figure 6.7 Comparison between the anti-oxidant activity of the solvent extracts of Salvia species using the ABTS and the DPPH assays ??????????85 Figure 6.8 Regression analysis of the anti-oxidant activity plotted against the total phenolic content using the DPPH and ABTS methods ????????...86 Figure 7.1 Inflammation process pathways ??????????????????92 Figure 7.2 Plot of the absorbance determined over 10 minutes of essential oil of Salvia runcinata at various concentrations ??????????????????95 Figure 8.1 Life cycle of the malaria parasite?????????????????.102 Figure 8.2 A thick blood smear showing the ring form of P. falciparum in erythrocytes?....................................................................................................105 Figure 8.3 Isolation procedures of the two compounds from S. radula??????...108 Figure 8.4 Antimalarial properties of the solvent extract of Salvia radula, S. runcinata and the essential oils of S. africana-caerulea and S. runcinata???????110 Figure 8.5 Comparison between the antimalarial activity of the essential oils and the solvent extracts of eleven Salvia species..........................................................112 Figure 8.6 Relationship between the antimalarial activity of the solvent extract and essential oil of the same plant..............................................................113 Figure 8.7 Structures of the isolated compounds from S. radula?????????..116 Figure 9.1 BACTEC 460 apparatus with racks containing inoculated vial and the 12B vials containing 14 C carbon substrate...........................................................123 Figure 9.2 An example of the determination of the MIC value using a 96-well microtitre plate??????????????.............................????.125 Figure 9.3 Bio-autographic assay of Salvia chamelaeagnea against xx Staphylococcus aureus??????????????????.............?.129 Figure 9.4 Bio-assay guided fractionation for compounds isolated from S. chamelaeagnea ?..................................................................................??...130 Figure 9.5 Isobolograms constructed from combination study involving Salvia chamelaeagnea and Leonotis leonurus on four bacterial pathogens.????...133 Figure 9.6 Growth index of three Salvia species tested at two different concentrations against Mycobacterium tuberculosis?????????...?...136 Figure 9.7 Structures of the isolated compounds with antibacterial activity from Salvia chamelaeagnea???????????????????????141 Figure 10.1 Microtitre plate showing how extracts were plated out in the anticancer assay and colour change of the SRB dye as a result of cell death ???????......154 Figure 10.2 Percentage cell viability of cancer cells tested at various concentrations for three Salvia species??????????????????????...156 Figure 11.1 Variation between the toxicity profile of the essential oil and solvent extracts of eleven Salvia species???????????????????170 Figure 12.1 Variation in essential oil yields of three Salvia species across seasons??.178 Figure 12.2 Variation of major components of Salvia africana-caerulea essential oil over a one year period?????????????????????...179 Figure 12.3 Variation of major components of Salvia africana-lutea essential oil over a one year period.????????????????????..?179 Figure 12.4 Variation of major components of Salvia lanceolata essential oil over a one year period?????????????????????...180 Figure 12.5 TLC chromatograms of the solvent extracts three Salvia species collected at various seasons of the year????????????????..181 Figure 12.6 Antibacterial activity of the solvent extracts of three Salvia species obtained across four seasons of the year???????????????????...182 Figure 12.7 Antimalarial activity of three Salvia species obtained at different seasons of the year????????????????????????..183 Figure 12.8 Anti-oxidant activity of three Salvia species obtained at different seasons of the year???????????????????.????.....184 Figure 12.9 Toxicity profile of the solvent extracts of three South African Salvia species across seasons???????????......................................??..184 Figure 12.10 Toxicity profile of the essential oils of three South African Salvia species across seasons????.........................................................................................185 xxi Figure 12.11 Variation in ?-caryophyllene and ?-caryophyllene oxide contents in S. lanceolata over seasons??????????????..........................186 Figure 13.1 TLC chromatograms of the solvent extracts of three Salvia species obtained in presence and in absence of the essential oils.......................................194 Figure 13.2 Antibacterial activity of the solvent extracts with and without essential oils of three Salvia species?????????????????...195 Figure 13.3 Anticancer activity of the solvent extracts of three Salvia species evaluated in the presence and in the absence of the essential oils??????????..196 Figure 13.4 Variation in the antimalarial activity of the solvent extracts with and without their respective essential oils of three Salvia species????????197 Figure 13.5 Anti-oxidant activity of the solvent extracts with and without their respective essential oils of three Salvia species??????????.............198 Figure 13.6 Toxicity profile of the solvent extracts with and without essential oils???...............................................................................................199 xxii List of Tables Table 1.1 Structural diversity and number of secondary plant metabolites??????..3 Table 1.2 Natural compounds or naturally derived drugs from plants that are currently used clinically ??????????????????????....4 Table 1.3 Medicinal uses of South African Salvia species ????????????..9 Table 1.4 Some examples of biological activities of Salvia species ????????..11 Table 2.1 Species studied, localities of collections, accession and voucher numbers.........17 Table 3.1 Distribution of trichomes on leaves of Salvia species??????????.38 Table 3.2 Various types of glandular trichomes found in Salvia species???????39 Table 4.1 Percentage of the components identified in the essential oils of 11 South African Salvia species.????????????????????...50 Table 5.1 Chromatographic data obtained from various standards according to their retention times..........................................................................................................67 Table 5.2 Area (%) and identification of standards and isolated compounds in the solvent extracts of 17 indigenous Salvia species......................................................68 Table 5.3 Compounds exhibiting a typical flavonoid spectrum in the solvent extracts of 17 Salvia species..................................................................................................69 Table 6.1 The anti-oxidant activity of solvent extracts determined by the ABTS and DPPH assays and the total phenol content ????????................?...84 Table 7.1 Anti-inflammatory activity of the essential oils and solvent extracts...???..96 Table 8.1 In vitro antimalarial activity of the solvent extracts and essential oils of indigenous Salvia species and isolated compounds??????????...111 Table 8.2 NMR data of compound 1 isolated from Salvia radula??............................114 Table 8.3 NMR data of compound 2 isolated from Salvia radula....................................115 Table 9.1 The MIC values of the solvent extracts and essential oils of Salvia species against Gram-positive and Gram-negative bacteria???????????...131 Table 9.2 The sum of the fractional inhibitory concentration values of Salvia chamelaeagnea and Leonotis leonurus using various combinations ratios??.134 Table 9.3 The effect of plant extracts on the growth of Mycobacterium tuberculosis using the radiometric method??????????????...135 Table 9.4 NMR data of compound 1 isolated from Salvia chamelaeagnea......................137 Table 9.5 NMR data of compound 2 isolated from Salvia chamelaeagnea......................138 Table 9.6 NMR data of the mixture of two triterpenoids: oleanolic and xxiii ursolic acid isolated from Salvia chamelaeagnea..................................................139 Table 10.1 The anticancer activity of the solvent extracts of Salvia species and reference compound obtained against three human cancer cell lines ????...157 Table 11.1 The toxicity profile of the essential oils, solvent extracts and reference compound ???.?????????????????????.............169 xxiv List of Acronyms and Symbols %: Percent ? C: Degree Celsius 5?-FU: 5?-Fluorouracil Abs: Absorbance ABTS: 2,2?-azino-bis-(3- ethylbenzothiazoline-6-sulfonic acid ABTS?+: 2,2?-azino-bis- (3- ethylbenzothiazoline-6-sulfonic acid radical AIDS: Acquired immuno deficiency syndrome amu: Atomic mass units ANOVA: One-way analysis of variance ATCC: American type culture collection br: Broadened CCM: Complete culture medium CD3OD: Methanol CDCL3: Chloroform CFU: Colony forming unit CGT: Capitate glandular trichome cm2: square c entimetre CO2: Carbon dioxide COSY: Correlation spectroscopy CPM: Counts per minute CUT: Cuticle d: Doublet dd: Double of doublets DEPT: Distortionless enhanced polarisation transfert DMEM: Dulbecco?s Modified Eagle?s DMSO: Dimethyl sulfoxide DNA: Deoxyribonucleic acid DPPH: 2,2-diphenyl-1-picrylhydrazyl DPPH?: 2,2-diphenyl-1-picrylhydrazyl radical EDTA: Ethylene diamine tetra-acetic acid EI: Electron impact Eq.: Equation eV: Electron volt FAA: Formaldehyde, alcohol and acetic acid FBS: Fetal bovine serum FCR: Falciparum chloroquine resistant FCS: Fetal calf serum FIC: Fractional inhibitory concentration g: Gram GAE: Gallic acid equivalent GC: Gas chromatography GC-MS: Gas chromatography coupled to mass spectrometry GI: Growth index GMA: Glycol methacrylate h: Hour H: proton HEPES: [N-(2-hydroxyethyl) piperazine-N'-(2-ethanesulfonic acid)] HMBC: Heteronuclear multiple-bond correlation HMQC: Heteronuclear multiple quantum HPLC: High performance liquid chromatography HT-29: Hypertriploid Hz: Hertz xxv i.d.: Inner diameter IC50: Inhibitory concentration (concentration that reduces the effect by 50%) INT: p-Iodonitrotetrazolium J : Coupling constant kgf/cm2: Kilogram force per square centimetre kV: acceleration voltage l: Litre LM: Light microscopy m/z: Mass to charge ratio M: Metre M: Molar m: Multiplet M+: Molecular ion max: Maximum MCF-7: Michigan cancer foundation mCi: Millicurie mg: Milligram MHz: Megahertz MIC: minimum inhibitory concentration min: Minute ml: Millilitre mm: Millimeter mM: Millimolar mm2: Square millimetre mmol: Milimole MS: Mass spectrometry MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5 dimethyl tetrazolium bromide n: Number of repetitions N2: Nitrogen nc: Not calculated NCI: National cancer institute nd: Not determined NGT: Non-glandular trichome NHLS: National health laboratory services nm: Nanometer NMR: Nuclear magnetic resonance NOESY: Nuclear overhauser effect NTCC: National type culture collection O2: Oxygen PBS: Phosphate buffer saline PDA: Photodiode array PGT: Peltate glandular trichome pH: Potential hydrogen ppm: Parts per million PRBC: Parasitized red blood cells q: Quartet r: Pearson?s correlation r2 : correlation coefficient RBC: Red blood cell Rf: Retention factor RI: Retention indices rpm: Revolution per minute RPMI 1640: Roswell park memorial institute Rt: Retention time s.d.: Standard deviation of the mean s.e.: Standard error of the estimate s: Second s: Singlet Scorr: Correlation coefficient SEM: Scanning electron microscopy SRB: Sulforhodamine B t: Triplet xxvi TB: Tuberculosis TCA: Trichloroacetic acid td: triple of doublets TLC: Thin layer chromatography TMD: Thermabeam mass detector TMS: Tetramethylsilane TSA: Tryptone soya agar TSB: Tryptone soya broth U: Enzyme unit u: Units UI-A: Unidentified A UI-B: Unidentified B UI-C: Unidentified C UI-D: Unidentified D UV: Ultra violet UV-VIS: Ultraviolet-visible v/v: Volume per volume w/v: Weight per volume w/w: Weight per weight WHO: World Health Organization ?c: Carbon shift ?GI: Difference in the growth index ?H: Proton shift ?: Wavelength ?A: Microampere ?g: Microgram ?l: Microlitre ?m: Micrometre ?M: Micromolar 1 Chapter 1: General Introduction 1.1 Introduction Traditional medicine is widely practiced in many Third World countries. The WHO estimated that about 80% of populations in developing countries rely on traditional medicine for their primary health care needs (J?ger and van Stade n, 2000; Matu and van Staden, 2003). In rural areas, reliance on traditional medicine is high and can be attributed to both economic and cultural factors. From an economic point of view, the high cost of imported, conventional drugs and/or inaccessibility to western health care facilities implies that the traditional mode of health care is the only form that is affordable to them (Matu and van Staden, 2003). Even when western health care facilities are available, they usually exist side by side with traditional medicine (Sindiga, 1994). The use and preparation of medicinal plants is conveyed verbally from one generation to the next, a process that involves the risk of losing essential details (Atindehou et al., 2004). Furthermore, plant species that form the basis of traditional medicine are disappearing at an alarming rate in many countries of the world. It is therefore imperative to study medicinal plants and to determine and evaluate their potential biological activities and efficacy. Secondary metabolites from a vast array of plants are used in numerous fields such as phytotherapy, perfumes, cosmetics, aromatherapy and spices. Aromatherapy has attracted the attention of many scientists and the screening of plants to study their biological activities has increased (Lahlou, 2004). For thousands of years, medicine and natural products have been closely linked with traditional medicines and chemical studies of these predominantly plant-derived traditional medicines were the basis of most early medicine such as aspirin and quinine (Butler, 2004). More recently, essential oils and various other extracts have stimulated a growing interest in natural product research and many have been screened for their potential use as alternative remedies for the treatment of various ailments (Buchbauer, 2000). Species of the genus Salvia (Lamiaceae) are well known throughout the world for their medicinal properties which have made it an attractive choice for many researchers. Although some Salvia species have been scientifically studied in many parts of the world and are reported to have various biological activities including antibacterial (Ulubelen et 2 al., 2001), anti-oxidant (Couladis et al., 2003), anti-inflammatory (Perry et al., 2003, anticancer (Liu et al., 2000) and anticholinesterase (Perry et al., 2003), not much is known about the South African species. Therefore, there is little scientific basis to validate the claims by traditional medicine practitioners about the effectiveness of local indigenous Salvia species. 1.2 The role of secondary metabolites in plants Plants are essential for all life on Earth since they provide oxygen. In addition, they provide non-nutritional compounds or secondary metabolites which are not necessarily involved in cell metabolism (e.g. essential oils). These natural compounds are formed from diverse precursors such as acetyl-CoA or amino acids, through enzymatic reactions (Harbone, 1993). Plants are sessile organisms and can be attacked easily by snails, insects or vertebrate herbivores. Some authors have suggested that natural compounds might have evolved to provide protection to the plant against attack by herbivores and/or to challenge hostile environments. Experimental evidence supports the view that many alkaloids, cyanogenics, glycosides, glucosinolates, terpenoids and tannins are allochemicals (Wink, 2003). However, plants often need animals to assist in pollination or seed dispersal and in these cases, secondary metabolites serve to attract the animals to the plant. Some secondary metabolites concomitantly carry out physiological functions. For instance, peptides (protease inhibitors) can serve as mobile and toxic nitrogen transport (Harbone, 1993). The observed multiple functions are typical and do not contradict the main role of many secondary metabolites as chemical defense and signal compounds (Wink, 2003). Secondary metabolites are present in all higher plants and display immense structural diversity (Table 1.1). As a rule, a single group of secondary metabolites dominate within a given taxon (Wink, 2003) although few major compounds are often accompanied by several derivatives and minor compounds. Typical secondary metabolites of the Lamiaceae include various terpenoids, especially monoterpenes, sesquiterpenes, as well as various phenolic acids (Wink, 2003). The secondary metabolites are generally concentrated in one particular region of a plant such as the leaves, roots, bark, fruits or glandular hairs. In cases where they occur in various organs of the same plant, they frequently have different chemical profiles (Ara?jo et al., 2003). 3 Table 1.1 Structural diversity and number of secondary plant metabolites (Wink, 2003). Natural products Number Alkaloidsa 12,000 Alkamidesa 150 Aminesa 100 Carbohydratesb > 200 Cyanogenic glycosidesa 60 Diterpenesb 2,500 Flavonoidsb 4,000 Glucosinolatesa 100 Monoterpenesb 2,500 Non-protein amino acidsa 700 Phenylpropanoids, coumarins, lignansb 2,000 Polyacetylenes, fatty acids, waxesb 1,000 Polyketidesb 750 Sesquiterpenesb 5,000 Tetraterpenesb 500 Triterpenes, saponins, steroidsb 5,000 a with nitrogen b without nitrogen The essential oils of plants are important not only because they possess biologically active compounds, but their chemical compositions have been used to resolve some taxonomic problems and discrepancies, especially in the Lamiaceae (Wink, 2003). It is thought that trichomes are structures responsible for the secretion of essential oils. These structures are also important in plant taxonomy, as they have been used to determine the degree of similarities among many species of the Lamiaceae. Considering the difficulties often encountered in distinguishing Salvia species, the study of their chemical composition and the morphology of the leaf trichome types can be an important tool in clarifying this issue. 4 1.3 Natural compounds and chemotherapeutic agents Despite competition from synthetic drugs, natural compounds still occupy a commanding position among drugs that are now used against infective organisms and diseases. Table 1.2 lists a few isolated compounds from plants which are currently in clinical use. Drugs derived from natural compounds are well represented in the top 35 worldwide ethical drug sales between 2000 and 2002 with 40% being derived from natural compounds in the year 2000 (Butler, 2004). Table 1.2 Natural compounds or naturally derived drugs from plants that are currently used clinically (Iwu, 1994; Butler, 2004). Compounds / drugs Plant Disease Artemisinin Atermisia annua Malaria Aspirin Salix species Inflammation, pain Physostigmine Physostigma venonosum Glaucoma Quinine Cinchona pubescens Malaria Reserpine Rauvolfia serpentia Hypertension Taxol Taxus brevifolia Ovarian and breast cancer Vinblastine and vincristine Catharanthus roseus Lung cancer Natural compounds often have complex chemical structures making them difficult, if not impossible, to synthesize. Medicinal plants which form the backbone of traditional medicine have in the last few decades been subjected to intense pharmacological investigations. This has been catalysed by the acknowledgement of the value of medicinal plants as a potential source for new compounds with therapeutic value and as valuable sources for drug discoveries and development (Matu and van Staden, 2003). The common practice of phytochemists to isolate, characterize and publish phytocompounds without exploring the bioactivity is no longer attractive to academic journals and funding companies (Fatope et al., 2000). Therefore, it is important to isolate and identify the active compounds which show good biological activity from plants. 5 1.4 Approach to plant-based research In the approach of searching for new drugs, no decision is taken by the researcher about the specific plant or organisms to be examined. Based, for example, on the impressive quantity and quality of information derived by indigenous cultures that have used these medicinal plants, plants are ethnobotanically selected with the aim of searching for compounds with desirable biological activities. Drug discovery is often refered as a ?number-game?. There are an estimated 250,000 to plant species that inhabit our planet and only a very small percentage of these have been evaluated for potential bioactivities (Clark, 1996). Based on this approach, it would take many years of a systematic and concerted effort to evaluate the Earth?s plants for a specific biological activity. Furthermore, it is a difficult task in terms of financial demands. Another problem that might arise from this approach is that extracts might contain more than one substance, which can obfuscate the assay. Results obtained from the early 1980?s by random screening have been disappointing. Valeriote et al. (1990) estimated that only 1% of randomly selected agents tested for anticancer activity demonstrated sufficient solid tumour selectivity to proceed to in vivo experiments. Corbett et al. (1990) screened more than 15,000 agents against various cancer cell lines and found that only about 5% were selective in vitro. To date, this approach constitutes the only effective way to obtain new and unique classes of agents with novel mechanisms of action (Corbett et al., 1990). It is estimated that approximately 77% of the medicinal agents worldwide were obtained from ethnobotanical surveys (Corbett et al., 1990). 1.5 Future of natural products research Although plants have, up to now, supplied man with a large number of excellent drugs and templates for further developments, some people have expressed concern feeling that research in this field is languishing. In the 1980?s, there was a lack of emphasis on plant drug research in some developed countries because the available programmes did not successfully yield any new marketable medicines. After testing approximately 40,000 plants, the National Cancer Institute (USA) was unable to identify a single new agent to be used in the treatment of human cancer (Tyler et al., 1988). During the same period in Germany, drugs made from plants and even new plant constituents were continually 6 introduced into the market. The discovery of new drugs is a question of hard work and perseverance, as was shown by the National Cancer Institute when in the late 1980?s it developed an exciting new drug (Taxol? ) because of its continued efforts (Clark, 1996). Since the end of the twentieth century, there has been a resurgence of interest in the investigation of natural materials as sources of potential new chemotherapeutic agents (Cragg et al., 1997, Cragg and Newman, 2005). Research activity is stronger now than perhaps at any other time and this renewed interest can be attributed to several factors: (i) unmet therapeutic needs, (ii) remarkable diversity of both chemical structure and biological activity of naturally occurring secondary metabolites, (iii) utility of novel bioactive natural products, (iv) development of novel and sensitive techniques to detect biologically active natural products, and (v) improved techniques to isolate, purify and structurally characterize these active constituents. By coupling the technological capabilities with the creativity and innovation of nature, the future of natural product based research is more promising than ever before. Natural products are still a potential source of new drugs, especially in the anticancer and antihypertensive therapeutic areas. The natural products discovered to date have played a vital role in improving the human condition and this role will continue as long as there are unexplored sources of novel natural products (Clark, 1996). 1.6 The genus Salvia: description, history and traditional uses in South Africa 1.6.1 Morphological description Salvia species are easily recognized by their square stems and opposite, simple pairs of leaves that are usually velvety or hairy. Leaves are often rugose, entire, toothed and lobed. Flowers are clustered in racemes, spike-like racemes, spikes and panicles are usually large and brightly coloured, depending on species. Flowers and stems are key diagnostic characteristics for identification of the genus (Hedge, 1974; Codd, 1985). There are four 7 stamens, but only two bear anthers. Seeds, stem cuttings and divisions are the common mode of propagation. Salvia grows optimally in full sun and needs well-drained soil. In waterlogged soil, the roots may rot. 1.6.2 History Salvia species (sage) belong to the mint family Lamiaceae (formerly Labiatae). The genus name Salvia L. is derived from the Latin salvare meaning ?to heal or to be safe and unharmed? referring to the medicinal propertie s of some of the species (Blumenthal et al., 2000). It has been used for centuries, especially by the Chinese as a herb for longetivity or in Roman ceremonies as a sacred herb. This name was translated to sauge (sage) in French and sawge in Old English. Sage encompasses about 900 species, widespread throughout the world (Codd, 1985; Paton, 1991; Gali-Muhtasib et al., 2000) and includes several ornamental, culinary and medicinal species. Because they readily cross-pollinate, innumerable hybrids, both natural and manmade, are also found. The genus has a sub- cosmopolitan distribution, but is largely absent in the North and most of the low-lying tropical areas of the world such as the Amazon basin and central and west Africa (Paton, 1991). Most of South African species are found in the Cape region (Codd, 1985). Although Mexico has the highest number of species (about 250), the center of origin of the genus has been reported to probably be Afghanistan and Soviet Central Asia where there are a larger range of primitive morphological types (Paton, 1991). In Africa, the greatest number of species is in the north-west and in the southern parts (Hedge, 1974). The genus is absent from most of western and central tropical Africa (J?ger and van Staden, 2000). The distribution of the genus extends all over northern Africa from west to east and southwards to the east African highlands. Although the species in the north-west are either common or affiliated to others species, most southern Africa species are endemic and only display a certain degree of similarity with other African species (Paton, 1991). Southern africa is home to more than 24,000 higher plant taxa (Arnold and De Wet, 1993) and a large proportion of them are endemic in character (Mulholland and Drewes, 2004). In this part of the world, as in many Third World countries, traditional medicine still plays an important role with Salvia species being widely used alone or in combinations with other plants. About 26 species have been identified in South Africa (Codd, 1985; Paton, 8 1991) but species such as S. coccinea, S. officinalis, S. reflexa, S. sclare, S. verbenaca and S. tillifolia have been introduced and are thus not indigenous to the region. 1.6.3 Traditional uses of Salvia species S. officinalis is native to the Mediterranean rim, especially around the Adriatic Sea and cultivated in South Africa and has one of the longest histories of use as a medicinal plant. Salvia has been used throughout the world and the famous proverb ?why should he die who has sage in his garden? was used to describe the medicinal properties of the plant (Gali- Muhtasib et al., 2000). A few centuries ago, it was nearly always at the top of the list of household remedies for the relief of many diseases. Herbalists in Jordan, Syria and Lebanon consider Salvia as a ?panacea? i.e. a univ ersal drug (Gali-Muhtasib et al., 2000). The therapeutic value of Salvia was also recognized by the king of France, Louis XIV, who trusted Salvia more than his doctor (Gali-Muhtasib et al., 2000). The traditional uses of Salvia in South Africa are summarized in Table 1.3. Despite the fact that an important local ethnobotanical bibliographic description of the most commonly used plants in the treatment of various diseases exits (Watt and Breyer-Brandwijk, 1962; van Wyk et al., 2000), there are few experimental studies which validate the therapeutic properties of these plants locally. 1.7 Chemical constituents of Salvia species 1.7.1 Volatile compounds (essential oils) Limited phytochemical work has been carried out on indigenous Salvia species. The essential oil of S. stenophylla collected on the ?h ighveld? of the Free State revealed the presence of monoterpenes. The oxygenated monoterpenoids constituted 5% of the oil and the sesquiterpene hydrocarbons 35.5%. S. stenophylla was found to contain very low amounts of 1,8-cineole and camphor, and lacked ?- and ?-thujene (Jequier et al., 1980). Brunke and Hammerschmidt (1985) also analyzed the essential oil of S. stenophylla and reported the predominance of ?-bisabolol (29.8%). For the exotic Salvia species, the essential oil composition varies between species and 1,8- cineole, ?-thujene, ?-thujene, camphor, (E)-caryophyllene, linalool and linalyl acetate were 9 found to be the most predominant compounds in the majority of species investigated (Topcu et al., 1997; Gali-Muhtasib et al., 2000; Kalemba and Kunicka, 2003; Pitarokili et al., 2003; Pavela, 2004). Azcan et al. (2004) investigated the fatty acid seed oil composition of twelve Salvia species growing in Turkey and the main fatty acids found were oleic, linoleic, linolenic acids and palmitic acids. Table 1.3 Medicinal uses of South African Salvia species (Watt and Breyer-Brandwijk, 1962). Species Part used Mode of uses Ailments S. africana-caerulea Twig Infusion mixed with Epson salt and lemon Colic, diarrhoea, indigestion, abdominal trouble S. africana-lutea Whole plant Infusion, decoction Colds, cough S. chamelaeagnea Leaves Decoction, infusion Colds, cough S. triangularis Whole plant Decoction with Helichrysum latifolium and Commelina africana Liver sickness, barrenness S. repens Roots, leaves, whole plant Decoction, added to bath Sores on the body, stomach problems, diarrhoea S. runcinata Whole plant, leaves, stems and roots Burned, decoction Disinfect dwelling sickness, urticaria S. scabra Leaf Decoction Purgative Salvia species Roots, whole plants Decoction Biliousness, cold, febrile attacks Salvia species Whole plant Infusion in combination with Leonotis leonurus Tuberculosis, cough, influenza, bacterial infections, cold S. stenophylla Whole plant Infusion, decoction Disinfect dwelling sickness 1.7.2 Non-volatile compounds (phenolic compounds) Salvia species are a rich source of polyphenolics and flavonoids. The polar phenolic acids constitute the major part of the Salvia decoction. Caffeic acid plays a central role in the biochemistry of Lamiaceae and occurs predominantly in the dimer form as rosmarinic acid (Gerhardt and Schroeter, 1983). In Salvia species, caffeic acid is the building block of a variety of the plant metabolites from the simple monomers to multiple condensation products that give rise to a variety of oligomers (Lu and Foo, 2002). Phenolic acids such as rosmarinic, caffeic and salvimanolic acids have been isolated (Lu and Foo, 2002). 10 The phenolic glycosides are not very common in Salvia. The rosmarinic acid 3'-glucoside and its methyl ester, as well as the cis- and trans-p-coumaric acid 4-O-(2-O-apyiosyl) glucosides, are the only examples of the glycosylated phenolic acids (Lu and Foo, 2002). In contrast, flavonoids are widely distributed in Salvia species (Ulubelen and Tuzlaci, 1990) and they are mostly present as flavones, flavonols and their glycosides (Lu and Foo, 2002). The majority of O-flavonoids are the flavones, apigenin and luteolin. The 6- hydroxyflavones are the flavonoids that characterise the species of Salvia and they include a variety of 6-hydroxylated apigenin and luteolin derivatives with cirsimaritin. Flavonols are mostly those of kaempferol and quercetin. Flavonone O-glycosides are apparently common in Salvia and most of them are flavone 7-glucosides represented by cosmosiin and cinaroside (Lu and Foo, 2002). A solvent extract of S. stenophylla collected in the Eastern Cape was investigated for flavonoids. Apigenin-7-methyl ether, luteolin and 6-hydroxyluteolin-6,7-dimethyl ether were identified (Wollenweber et al., 1992). A large study on the occurrence of alkaloids in plants included a number of the southern African Salvia species, with S. chamelaeagnea, S. namaensis and S. runcinata testing positive for alkaloids, while in S. africana-caerulea, S. africana-lutea, S. coccinea, S. dolomitica, S. reflexa, S. tillifolia and S. triangularis, no alkaloids were detected (Raffauf, 1996). 1.8 Biological activities of Salvia species A literature search indicated that the genus Salvia has been a popular topic in phytochemical and ethnobotanical research. The solvent extracts, essential oils and compounds isolated from Salvia species revealed that they displayed a broad range of pharmacological properties, both in vitro and in vivo. Table 1.4 lists some examples of biological properties of Salvia including two indigenous species. 1.9 Rationale and motivation of the study Salvia was chosen as the basis of this study because of its long history of traditional use in South Africa and the rest of the world. Despite many reports on the medicinal properties of Salvia, few studies have been carried out on indigenous species (Table 1.4). 11 Table 1.4 Some examples of biological activities of Salvia species. Species Plant parts used Bioactive fraction/compound Activity Methodology used/ organisms used S. aethiopis Roots Aethiopinone Anti-inflammatory 5-Lipoxygenase a, b S. africana-lutea Leaves Solvent extract Anti-analgesic Acetic acid writhing in mice c S. fruticosa Whole plant Rosmarinic acid Anti-oxidant Free radical assay d S. hypargeia Roots Crude extracts, taxodione Antibacterial Mycobacterium tuberculosis e S. lavandulaefolia Aerial parts Essential oils Anticholinesterase Microtitre plate assay f S. miltiorrhiza Roots Tanshionones Anticancer Human tumour cells g, h S. officinalis Leaves Ursolic acid Anti-inflammatory Oedema induced in mice i S. paleastina Leaves Cirsimaritin Antibacterial Gram-positive bacteria j S. prionitis Leaves 3-Keto-4-hydroxysa- prothorquinone Anticancer Stomach cancer cell lines k S. repens Whole plant Dichloromethane:methanol extract Antimalarial Parasite lactate dehydrogenase l S. viridis Roots 1-Oxoferrunol Antibacterial S. aureus, B. cereus, B. subtilis m a Hernandez-Perez et al., 1995; b Benrezzouk et al., 2001; c Amabeoku et al., 2001; d Exarchou et al., 2002; e Ulubelen et al., 1988; f Savelev et al., 2003; g Liu et al., 2000; h Ryu et al., 1997; i Baricevic et al., 2001; j Miski et al., 1983; k Chen et al., 2002; l Clarkson et al., 2004; m Ulubelen et al., 2000. 12 From a literature survey on biological activities of the South African Salvia species, only two publications reported the pharmacological properties in S. africana-lutea and S. repens (Amabeoku et al., 2001; Clarkson et al., 2004) in addition to the composition of S. stenophylla essential oil (Jequier et al., 1980; Brunke and Hammerschmidt, 1985). The reported traditional medicinal uses and the lack of phytochemical and pharmacological studies on indigenous Salvia species prompted this investigation into the phytochemistry and the biological activities of various Salvia species from South Africa. 1.10 Aim of the study The aim of this study was to screen seventeen South African Salvia species in order to evaluate their biological activity, to isolate and identify compounds responsible for the selected biological activities and to assess a chemotaxonomic relationships amongst South African representatives of the genus Salvia. 1.11 Objectives of the study The objectives of this study were to: (i) examine the morphology of leaf trichome types of Salvia species and to assess the taxonomic value of these epidermal structures, (ii) investigate the chemical composition of the essential oils and to assess variation in their chemical composition across the seasons using gas chromatography (GC) and gas chromatography coupled to mass spectrometry (GC-MS) analysis and to identify various groups that may be used in the flavourant and cosmetic industries, (iii) identify the presence of phenolic compounds previously identified in exotic Salvia in the solvent extract of indigenous species using high performance liquid chromatography, (iv) investigate the anti-oxidant activity of the crude extracts and essential oils of indigenous Salvia species using two radicals, (v) evaluate the ability of crude extracts and essential oils to inhibit the 5- lipoxygenase enzyme which is one of the enzymes involved in the inflammatory process, 13 (vi) investigate the effect of the essential oils and solvent extracts on Plasmodium falciparum, the protozoan responsible for malaria by the [3H]- hypoxanthine radiometric assay and isolate the active compound(s) from the most active solvent plant, (vii) evaluate the inhibitory effects of the essential oils and solvent extracts on the growth of selected bacteria and the common tuberculosis pathogen as well as to isolate the active compound(s) from the plant showing the best antibacterial activity, (viii) evaluate the inhibitory effects of the solvent extracts against the growth of breast, neuronal and colon cancer cell lines using a protein detection colorimetric method, (ix) determine the toxicity profile of the essential oils and solvent extracts against a transformed human kidney epithelium cell line using the tetrazolium cell proliferative method, (x) study the seasonal variation in biological activities of selected Salvia species, and (xi) evaluate the influence of the essential oils on the biological activities of solvent extracts. 1.12 Outline of the thesis The rest of the thesis consists of 13 Chapters. Reprints of publications, abstracts for conferences and the monographs are included in the Appendices. Apart from Chapters 2 and 14 which describe the localities and the general conclusions, respectively, each Chapter consists of an abstract, an introduction, materials and methods, results, discussion and conclusion. Chapter 2: The extraction procedures for volatile (essential oils) and non-volatile (solvent extracts) fractions are described. The localities where the plants were collected are also listed. Chapter 3: The variety of trichomes found in various plant organs have been a valuable tool for comparative systematic studies of Angiosperms. In this Chapter, the morphology of the leaf trichome types is examined. 14 Chapter 4: The essential oils have a broad range of applications, including the cosmetic industry, food industry and in aromatherapy. GC and GC-MS were used to analyse the chemical composition of the essential oils isolated by hydrodistillation. The relationship among different species could thus be more clearly established. Chapter 5: Salvia is known to possess various phenolic compounds which contribute to biological activities. Some of the most common compounds in Salvia extracts as well as those isolated from two Salvia species are identified in the investigated species. Chapter 6: There is increasing evidence to suggest that many degenerative diseases such as inflammation, brain dysfunction, immune system suppression and cancer could be the result of cellular damage caused by free radicals. The free radical scavenging activity (anti-oxidant activity) of the solvent extracts and essential oils is evaluated and the phenolic content determined. Chapter 7: The 5-lipoxygenase enzyme is one of the key enzymes involved in the inflammation process. The potential of the solvent extracts and essential oils to inhibit this enzyme is evaluated. Chapter 8: The increasing prevalence of malaria has been attributed to a number of factors, one of them being the emergence and spread of drug resistant parasites. In this Chapter, the essential oils and solvent extracts were tested for their potential to inhibit the growth of Plasmodium falciparum and two compounds from the most active solvent extract were isolated, characterised and tested. Chapter 9: Despite the progress in understanding the life cycle and control of many pathogens, nearly all diseases, affecting millions of people in developing countries, are still caused by micro-organisms. The essential oils and solvent extracts were investigated in order to evaluate their potential to inhibit the growth of common bacterial pathogens. Furthermore, the active compounds were isolated and characterised from the most promising plant. 15 Chapter 10: A major problem with present day cancer chemotherapy is the serious deficiency of active drugs for the curative therapy and their adverse effects on the patient. The anticancer activity of solvent extracts is investigated on three different cell lines and their potential selectivity noted. Chapter 11: There is a common notion among people who frequently use Salvia species to treat different ailments that ?if it does not benefit, it would not harm? and many people are not aware of any side effects or toxic symptoms that may be associated with excessive use of Salvia species. In this Chapter, the toxicity of essential oils and the solvent extracts is explored. Chapter 12: The composition of the essential oils and solvent extracts may vary depending on environmental conditions and season which may, inevitably, affect the biological activities. Seasonal variation in essential oil composition and biological activities of the solvent extracts across seasons are evaluated for three species. Chapter 13: Aromatic plants such as Salvia species possess both the volatile and non- volatile fractions. The contribution of the volatile fractions on the biological activities of the non-volatile fraction is discussed. Chapter 14: General conclusions of the study and recommendations are presented. The reference documents employed for this research are cited subsequent to Chapter 14. 16 Chapter 2: Plant Collection, Isolation of Volatiles and Extraction of Non-Volatiles 2.1 Species selection Species investigated in this study were selected according to their reported traditional uses in South Africa. 2.2 Collection of plant materials Codd (1985) described 26 Salvia species in southern Africa, 17 of 22 species occurring naturally in southern Africa were included in this study. The aerial parts of the selected species were harvested at different localities in South Africa between December 2003 and December 2005. Possible variation between populations was not taken into account and only one representative per species was considered. The taxonomic identification of all species collected was confirmed at the herbarium at the South African National Biodiversity Institute (Pretoria) and voucher specimens were deposited in the Department of Pharmacy and Pharmacology, University of the Witwatersrand, Johannesburg, South Africa. Table 2.1 lists the species investigated and a concise description and photographs of each species are shown in the monographs in Appendix C. The traditional uses of each plant are also included in Appendix C. 2.3 Preparation of samples 2.3.1 Preparation of the essential oils (volatile fractions) The essential oils of the fresh aerial parts were isolated by hydrodistillation using a Clevenger type apparatus for three hours at 100 ?C (Figure 2.1). The oil yield was recorded before being sealed in glass vials with silica septa, protected from light and kept refrigerated until used. 17 Table 2.1 Species studied, localities of collection, accession and voucher numbers. Species Locality Date of collection Accession number (KBG1) Voucher number S. africana-caerulea SWC2 01-12-2003 0554/82 AV 875 S. africana-lutea SWC2 01-12-2003 1185/71 AV 873 S. albicaulis KBG1 06-01-2004 0050/79 AV 894 S. aurita KBG1 01-12-2003 0244/01 AV 1066 S. chamelaeagnea KBG1 06-01-2004 0725/93 AV 848 S. disermas KBG1 31-03-2005 0574/74 AV 1194 S. dolomitica ex Manning 26-09-2003 0038/79 AV 838 S. garipensis KBG1 27-03-2004 0689/91 AV 1193 S. lanceolata SWC2 01-12-2003 0919/96 AV 877 S. muirii KBG1 27-03-2004 189/82 AV 874 S. namaensis* Swartberg 11-07-2002 n.g3 AV 497 S. radula Road to Derby 13-12-2003 n.g3 AV 880 S. repens KBG1 06-01-2004 0609/95 AV 615 S. runcinata Klerkskraal Dam 13-12-2003 n.g3 AV 679 S. schlechteri KBG1 27-03-2004 0173/02 AV 1068 S. stenophylla East of Clarens 06-01-2004 n.g3 AV 893 S. verbenaca De Rust 02-12-2003 n.g3 AV 631 1 KBG: Kirstenbosch Botanical Garden 2SWC: South Western Cape 3 n.g: Accession number not given because the plants were not collected at the KBG * Plant material used was collected in 2002 2.3.2 Preparation of the solvent extracts (non-volatile fractions) Aerial parts of each plant were air dried at room temperature for seven days and pulverised into a fine powder, using a coffee grinder. The pulverised aerial parts were then extracted 18 with methanol:chloroform (1:1, v/v) in a water bath for three hours at 36-40 ?C and filtered using Whatman? No 1 filter paper (pore size: 20-25 ?m). The filtrate was evaporated to dryness under reduced pressure and kept refrigerated until used. It is important to note that various solvent systems (acetone, methanol, dichloromethane, chloroform) were assayed before the methanol:chloroform mixture was finally selected because this mixture extracted compounds over a range of polarities. Figure 2.1 Clevenger apparatus used for the isolation of the essential oils. 19 Chapter 3: Leaf Trichome Types Abstract Trichomes are outgrowths of the epidermis found on various plant organs. Leaf trichome morphology is described here for the first time for 12 species of Salvia collected at the Kirstenbosch Botanical Garden. Leaf surfaces were examined by scanning electron microscopy while transverse sections of leaves were studied by light microscopy. All species investigated contain both glandular and non-glandular trichomes. The non- glandular trichomes are multicellular, unbranched and consist mainly of elongated cells, while the glandular trichomes are either capitate or peltate. The capitate type, present in all species, has a unicellular or multicellular head, one to four short or long stalk cells and one epidermal basal cell. The peltate type, also found in all species, is relatively large with multicellular head cells (up to sixteen), one or two stalk cell(s) and one epidermal basal cell. The non-glandular hairs are abundant in S. africana-lutea, S. lanceolata and S. muirii, while many glandular trichomes are observed in S. africana-caerulea, S. chamelaeagnea, S. disermas, S. garipensis and S. repens. Few glandular trichomes were found in S. albicaulis, S. dolomitica, S. lanceolata and S. muirii. The non-glandular trichomes as well as the types of glandular trichomes observed are taxonomically informative to distinguishing between groups of species. 20 3.1 Introduction Trichomes are hair-like outgrowths derived from specialised epidermal cells on leaf, root or stem surfaces. Scientific interest in plant trichomes is based on their taxonomic importance and on the economic benefit of some trichome-generated products (Davis and Heywood, 1973; Theobald et al., 1979). The great diversity of trichomes and the composition of exudates produced by them are used in plant taxonomy to distinguish between closely related species in some families (Spring, 2000). There is a great range in coverage for both glandular and non-glandular trichomes, and various proportions of types between species, and between the abaxial and adaxial sides of organs such as leaves. Hairs may be present on one side or both sides equally or unequally, sometimes along the margins and/or veins or restricted to intercostal regions in clefts (Werker, 2000). When non-glandular trichomes form a dense indumentum, they may serve as a mechanical barrier against various external factors such as excessive water loss, pathogens and herbivores (Werker, 2000). Although the role of glandular trichomes is not clearly elucidated, they are considered as storage sites of secondary metabolites. Glandular trichomes which secrete lipophilic substances may provide chemical or physiochemical protection against various types of herbivores and pathogens by entrapping, deterring, poisoning them or attract them for pollination or seed dispersal (Wagner, 1991). Glandular trichomes vary in the chemical composition of substances they secrete, mode of production, location and function (Hanlidou et al., 1991; Ascens?o et al., 1995). Secretions include sugars, salts, lipids, essential oils, resins and proteins. Suggestions have been made to classify glandular trichomes according to the nature of their secretion products (Fahn, 2001). However, there are some limitations to this type of classification since some glandular hairs may secrete more than one type of substance (Werker, 2000). The study of trichomes is a valuable tool in the resolution of many taxonomic problems in the Lamiaceae and may be used to determine the relationship between species (Davis and Heywood, 1973; Theobald et al., 1979). Although the indumentum is characteristic in many members of the Lamiaceae, little is known on the structural diversity of the 21 trichomes of South African Salvia species which might facilitate accurate identification. Salvia species are frequently used in traditional medicine and their scarcity makes them highly susceptible to adulteration and substitution. Thus, the wrong plant could easily be used in many herbal preparations with potentially dangerous consequences. The objectives of this chapter were to: (i) investigate the structure of trichomes using scanning electron microscopy (SEM) and light microscopy (LM) in order to determine the leaf trichome types in twelve South African species of Salvia, and (ii) evaluate the structural characteristics of leaf trichomes for their potential in identifying species. 3.2 Materials and methods 3.2.1 Plant material Fresh leaves of all 12 species, namely; S. africana-caerulea, S. africana-lutea, S. albicaulis, S. aurita, S. chamelaeagnea, S. disermas, S. dolomitica, S. garipensis, S. lanceolata, S. muirii, S. repens and S. schlechteri, were collected at the Kirstenbosch Botanical Garden (South Africa) in August 2005. For each species, some leaves were air- dried and others directly preserved in a formaldehyde, alcohol and acetic acid solution (FAA) basically according to Johansen (1940), but the proportions were modified as follows; distilled water:95% ethanol:formaldehyde:glacial acetic acid (28:17:6:2). Only one collection per species was considered. 3.2.2 Preparation of solutions and stain 3.2.2.1 Preparation of glycol methacrylate The glycol methacrylate (GMA) monomer mixture was prepared using 188 ml of glycol methacrylate, 12 ml of polyethylene glycol 200 and 1 g of benzoyl peroxide. The mixture was stirred slowly for one hour and the solution filtered with Whatman? No 1 filter paper. The filtrate was placed in a desiccator and the air removed with a vacuum pump before being stored in the dark at 4 ?C. 22 3.2.2.2 Preparation of toluidine blue The buffer used to prepare the 0.05% (w/v) toluidine blue was made with 0.25 g benzoic acid and 0.3 g sodium benzoate dissolved by heating in 200 ml of distilled water. The pH was adjusted to 4.4 with NaOH and the solution was filtered through Whatman? No 1 filter paper and stored in the dark at room temperature. 3.2.3 Morphological and anatomical studies of leaf trichomes 3.2.3.1 Morphology of leaf trichomes: scanning electron microscopy Leaves preserved in FAA were cut with a razor blade and rinsed with 50% ethanol five times for 30 minutes. The leaf was then dehydrated successively once in 70%, once in 90% and three times in 100% ethanol for 30 minutes followed by critical point drying in liquid carbon dioxide. For dried material, small pieces were obtained. Each leaf portion was then mounted on an aluminium stub by securing it with double-sided adhesive tape, coated with gold and examined. A standard location on each leaf (Figure 3.1) was chosen so that similar positions were examined. Each sample was scanned at a low magnification to reveal the general surface morphology and at higher magnification (where necessary) to show details. Both the abaxial and adaxial surfaces were studied. Samples were examined with a JEOL-5600 scanning electron microscope operating at an acceleration voltage of 6 kV. The trichome types were scored as absent (?) or present (+) and the relative abundance of glandular and non-glandular trichomes noted. Figure 3.1 Leaf area selected for the study of trichome morphology. 3.2.3.2 Microscopic structure of trichomes: light microscopy Small portions (? 3 mm2) of leaves preserved in FAA were cut with a razor blade and prepared according to the method of Feder and O?Brien (1968). The material was 23 dehydrated by passing it through a graded alcohol series (2 x 50% ethanol, 2 x 100% ethanol; 2 x 100% propanol and 2 x 100% butanol). Specimens were left in each alcohol concentration for six hours to ensure that alcohol penetration was complete after which the material was infiltrated with the GMA monomer mixture. Each piece was then placed in a gelatin capsule containing GMA and incubated at 60 ?C for 24 hours to polymerise prior to orienting and sectioning. Several transverse sections (3-5 ?m thick) were cut using a Porter Bl?m ultramicrotome and placed on slid es with a drop of distilled water. Slides were placed on a slide warmer for the sections to adhere and then stained with undiluted Schiff?s reagent and toluidine blue (as desc ribed in Section 3.2.2.2) (Feder and O?Brien, 1968). The slides were finally mounted with Entallen? and observations were made on a light microscope (Leitz Wetzlar) equipped with a digital camera (KY-F1030, JVC) connected to a computer with the Acquis Digital Imaging syncroscopy version 4.0.0.7. The relative abundance of trichomes was determined by counting the number of trichomes under the light microscope. At least 10 fileds were counted to obtain an average number. 3.3 Results 3.3.1 Scanning electron microscopy observations Representative portions of leaf surfaces are shown in Figures 3.2 to 3.7. Two glandular trichome types, namely peltate and capitate could be observed and are present in all the investigated species. There is no difference in trichome type, between adaxial and abaxial surfaces in all taxa. Although the abundance of trichomes differs among the species with either glandular or non-glandular trichomes predominating. In order words, where the glandular trichomes are less dense, the non-glandular trichomes form a dense covering that completely obscures the epidermal surface and this was particularly evident in S. africana- lutea (Figure 3.2C), S. dolomitica (Figure 3.5A), S. lanceolata (Figure 3.6A) and S. muirii (Figure 3.6D). The non-glandular hairs are multicellular, unbranched and consist of long cells and are also present in all the species (Figures 3.2 to 3.7). While most non-glandular trichomes bear one basal cell, S. aurita (Figure 3.3D), S. chamelaeagnea (Figure 3.4A) and S. repens (Figure 3.7A) bore multiple cells at the base. Micropapillae were visible on the non-glandular trichomes of some species such as S. chamelaeagnea (Figure 3.4C), S. lanceolata (Figure 3.6C) and S. muirii (Figure 3.6E). In some species such as S. africana- lutea (Figure 3.2D), S. albicaulis (Figure 3.3B), S. aurita (Figure 3.3F) and S. 24 chamelaeagnea (Figure 3.4B), the rupture of the cuticle revealed that there are up to sixteen secretory cells present in the peltate trichomes, whereas the capitate trichomes are composed of up to four head cells and this can clearly be seen in S. garipensis (Figure 3.5F) and S. muirii (Figure 3.6E). Figure 3.2 SEM micrographs of Salvia africana-caerulea (A and B) and S. africana-lutea (C, D and E). (A): Stalked capitate trichomes, sessile peltate trichomes and non-glandular trichomes on the lower leaf surface, bar = 100 ?m; (B): sessile peltate trichomes on lower leaf surface covered by turgid and intact cuticle and non-glandular multicellular trichomes unbranched, bar = 10 ?m; (C): leaf surfaces showing a glandular trichome (arrow) and dense covering of non-glandular trichomes, bar = 100 ?m; (D): details of the lower leaf surface showing sessile glandular trichome and non-glandular trichomes, bar = 100 ?m; (E): high density of non-glandular trichomes on the lower leaf surface, bar = 100 ?m. 25 Figure 3.3 SEM micrographs of Salvia albicaulis (A, B and C) and S. aurita (D, E and F). (A): lower leaf surface showing a capitate (short arrow) and peltate (long arrow) trichomes, bar = 100 ?m; (B): sessile peltate trichome on lower surface with a ruptured cuticle revealing individual secretory cells, bar = 10 ?m; (C): non-glandular unbranched trichome on the upper leaf surface, bar = 10 ?m; (D): peltate and capitate trichomes, the arrow indicates the base of a non-glandular trichome with multiple cells, bar = 100 ?m; (E): peltate trichome on lower surface with intact cuticle, bar = 100 ?m; (F): a multicellular peltate trichome with a ruptured cuticle, bar = 100 ?m. 26 Figure 3.4 SEM micrographs of Salvia chamelaeagnea (A, B and C) and S. disermas (D, E and F) . (A): lower leaf surface showing glandular trichomes; the black arrow indicates a stoma and the white arrow the base of a non-glandular trichome with many cells, bar = 100 ?m; (B): sessile peltate trichomes on the lower surface with a ruptured cuticle revealing individual secretory cells and neigbbouring stomata; the arrow indicates a capitate trichome of two cells, bar = 100 ?m; (C): non-glandular trichome, the white arrow indicates a capitate trichome composed of four cells, bar = 100 ?m; (D): upper surface showing peltate (long arrow) and capitate (short arrows) trichomes, bar = 100 ?m; (E): capitate trichomes with long and short multicellular stalk cells, rupture of the cuticle visible, bar = 100 ?m; (F): capitate, non-glandular (arrow) and peltate trichomes with intact cuticles, bar = 100 ?m. 27 Figure 3.5 SEM micrographs of Salvia dolomitica (A, B and C) and S. garipensis (D, E and F). (A): upper leaf surface showing a peltate trichome (long arrow), a capitate trichome (short arrow) and a dense covering of non-glandular trichomes, bar = 100 ?m; (B): detail of upper surface of sessile peltate and non-glandular trichomes, bar = 10 ?m; (C): a capitate trichome with a short stalk, bar = 10 ?m; (D): lower surface showing relatively long multicellular non-glandular trichomes (mainly on the vein), peltate and capitate trichomes with short and long multicellular stalk cells, bar = 100 ?m; (E): numerous peltate and capitate trichomes, bar = 10 ?m; (F): capitate trichomes, arrow indicates the four cells revealed by the ruptured cuticle, bar = 100 ?m. 28 Figure 3.6 SEM micrographs of Salvia lanceolata (A, B and C) and S. muirii (D, E and F). (A): upper leaf surface showing capitate (short arrow) and peltate (long arrow) trichomes and relatively dense point shaped multicellular non-glandular trichomes, bar = 100 ?m; (B): sessile peltate trichome covered by turgid and intact cuticle, bar = 10 ?m; (C): peltate trichome and non-glandular trichomes with micropapillae (black arrow), bar = 10 ?m; (D): upper surface with glandular and non-glandular trichomes, bar = 100 ?m; (E): multicellular capitate trichome and multicellular non-glandular trichomes with micropapillae (black arrow), bar = 10 ?m; (F): multicellular peltate trichomes with an intact cuticle, bar = 100 ?m. 29 Figure 3.7 SEM micrographs showing the types of trichomes found on the adaxial leaf surface of Salvia repens (A, B and C) and S. schlechteri (D, E and F). (A): capitate (short white arrow) and peltate (long white arrow) trichomes, the black arrow indicates a multicellular non-glandular trichome with many cells at the base, bar = 100 ?m; (B): capitate trichome with stalk cells and peltate trichome covered by an intact cuticle, bar = 10 ?m; (C): a capitate trichome with two stalk cells, bar = 10 ?m; (D): glandular and very sparse multicellular non-glandular trichomes mainly at the vein, stomata visible (black arrows), bar = 100 ?m; (E): capitate trichome with two secretory cells imbedded in a cavity, bar = 10 ?m; (F): peltate trichome with intact cuticle imbedded in a cavity, bar = 10 ?m. 30 3.3.2 Light microscopy observations The presence of glandular and non-glandular trichomes was confirmed using light microscopy (LM). The general leaf structure in transverse section of each taxon is shown in Figures 3.8 to 3.13, while the various types of trichomes found in all species are illustrated in Figure 3.14. Micro-ornamentation was not evident. No variation in trichome types was found on leaves at different stages of development (mature and immature leaves). The trichome types of the species investigated are summarised in Table 3.1. Figure 3.8 LM micrographs showing transverse sections of Salvia africana-caerulea (A and B) and S. africana-lutea (C and D). (A): capitate trichomes of mature leaf, bar = 0.2 mm; (B): mature leaf revealing the presence of glandular and non-glandular trichomes, bar = 0.1 mm; (C): young leaf showing non-glandular trichomes, bar = 0.2 mm; (D): young leaf showing glandular trichomes, bar = 0.1 mm. CGT: capitate trichome with head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non- glandular trichome; PGT: peltate trichome (multicellular head). 31 Figure 3.9 LM micrographs showing transverse sections of Salvia albicaulis (A and B) and S. aurita C and D). (A): mature leaf with peltate trichomes, bar = 0.2 mm; (B): mature leaf with glandular and non-glandular trichomes, bar = 0.1 mm; (C): mature leaf showing capitate and non-glandular trichomes, bar = 0.2 mm; (D): mature leaf revealing peltate trichomes and the cuticle, bar = 0.1 mm. CGT: capitate trichome with head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non-glandular trichome; PGT: peltate trichome (multicellular head). 32 Figure 3.10 LM micrographs showing transverse sections of Salvia chamelaeagnea (A and B) and S. disermas (C and D). (A): mature leaf showing the general morphology, bar = 0.2 mm; (B) : young leaf with peltate trichomes, bar = 0.1 mm; (C): mature leaf revealing non-glandular trichomes, bar = 0.2 mm and (D): young leaf with glandular trichomes, bar = 0.1 mm. CGT: capitate trichome w ith head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non-glandular trichome; PGT: peltate trichome (multicellular head). 33 Figure 3.11 LM micrographs of transverse sections of Salvia dolomitica (A and B) and S. garipensis (C and D). (A): mature leaf showing a peltate trichome, bar 0.2 mm; (B) : young leaf showing non-glandular and glandular trichomes, bar = 0.1 mm; (C): mature leaf revealing the capitate and non-glandular trichomes, bar = 0.2 mm; (D): mature leaf revealing a peltate trichome, bar = 0.1 mm. CGT: capitate trichome with head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non-glandular trichome; PGT: peltate trichome (multicellular head). 34 Figure 3.12 LM micrographs of transverse sections of Salvia lanceolata (A and B) and S. muirii (C and D). (A): mature leaf showing an abundance of non-glandular trichomes, bar = 0.2 mm; (B) : young leaf showing glandular trichomes, bar = 0.1 mm; (C): mature leaf revealing non-glandular trichomes, bar = 0.2 mm; (D): mature leaf showing various glandular trichomes, bar = 0.1 mm. CGT: capitate trichome with head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non-glandular trichome; PGT: peltate trichome (multicellular head). 35 Figure 3.13 LM micrographs of transverse sections of Salvia repens (A and B) and S. schlechteri (C and D). (A): mature leaf showing non-glandular trichomes, bar 0.2 mm; (B) : capitate and peltate trichomes on a mature leaf, bar = 0.1 mm; (C): mature leaf revealing capitate trichomes, bar = 0.2 mm; (D): mature leaf showing glandular trichomes, bar = 0.1 mm. CGT: capitate trichome with head composed of one or more than one cell; CUT: cuticle of a peltate trichome; NGT: non-glandular trichome; PGT: peltate trichome (multicellular head). 36 Figure 3.14 Various types of glandular and non-glandular trichomes found in Salvia species. (A): type I capitate trichomes of S. africana-caerulea, with unicellular head and short stalks, bar = 10 ?m; (B): capitate trichome type II of S. africana-lutea, with head composed of more than one cell and short stalk, bar = 10 ?m; (C) : capitate type III trichome of S. garipensis, with unicellular head and long stalks, bar = 10 ?m; (D): capitate trichome type IV of S. garipensis, with head composed of more than one cell and long stalk, bar = 10 ?m; (E): peltate trichome of S. africana-caerulea with up to sixteen secretory head cells and short stalk, bar = 10 ?m. 37 Figure 3.14 Contd. Various types of glandular and non-glandular trichomes found in Salvia species: (F): non-glandular trichome with round top (S. aurita), bar = 10 ?m; (G): short non-glandular trichome composed of two cells (S. dolomitica), bar = 10 ?m; (H): sharply-pointed non-glandular trichome (S. africana-caerulea) composed of three cells, bar = 10 ?m. 38 Table 3.1 Distribution of trichomes on the leaves of Salvia species: absence (?) and presence (+) based on trichome type observations in 3.14. Species Accession number (KBG A B C D E F G H S. africana-caerulea 0554/82 + ? ? ? + ? ? + S. africana-lutea 1185/71 + + ? ? + ? ? + S. albicaulis 0050/79 + + ? ? + ? ? + S. aurita 0244/01 + + ? ? + + ? ? S. chamelaeagnea 0725/93 ? + ? ? + ? ? + S. disermas 0574/74 + + + ? + ? ? + S. dolomitica 0038/79 + ? ? ? + ? + ? S. garipensis 0689/91 + + + + + ? ? + S. lanceolata 0919/96 + + ? ? + ? ? + S. muirii 189/82 ? + ? ? + ? + ? S. repens 0609/95 + ? + ? + + ? ? S. schlechteri 0173/02 ? + ? ? + + ? ? (A): Capitate trichomes: single-celled head, two or three short stalk cells, (B): capitate trichome: two-celled head, two or three short stalk cells, (C): capitate trichomes: single-celled head, two or three long stalk cells, (D): capitate trichomes: two celled-head, two or three long stalk cells, (E): peltate trichomes: four to sixteen cells in the head, one or two stalk cells, (F): non-glandular trichome with round top, (G): short non-glandular trichome composed of two cells, (H): sharply-pointed non-glandular trichome composed of three cells. The capitate trichome type varies in the number of secretory cells (up to four cells) with one to four stalk cell(s) and one basal epidermal cell. To accommodate differences in the number of cells comprising the head and the length of the stalk cells, capitate trichomes were classified into four types. Capitate trichome type I refers to a structure composed of a unicellular head, a short stalk cell and one basal epidermal cell. The head of the capitate trichome type II is composed of more than one cell with short stalk cell(s) and one basal epidermal cell. Type III capitate trichome is composed of a unicellular head, a long stalk cell(s) and one basal epidermal cell and type IV capitate trichome comprises a head with more than one cell, a long stalk cell(s) and one basal epidermal cell (Figure 3.14A-D). In general, the head of a capitate trichome has a slightly greater diameter than that of the stalk cells. The types of capitate trichomes found in different species are recorded in Table 3.2. 39 Table 3.2 Various types of capitate trichomes observed in Salvia species. Trichome type Characteristics Salvia species I Unicellular head, short stalk, one basal epidermal cell S. africana-caerulea, S. africana- lutea, S. albicaulis, S. aurita, S. dolomitica, S. disermas, S. garipensis, S. lanceolata, S. repens II Head with more than one cell, short stalk, one basal epidermal cell S. africana-lutea, S. albicaulis, S. aurita, S. chamelaeagnea, S. disermas, S. lanceolata, S. garipensis, S. muirii,S. schelchteri III Unicellular head, long stalk, one basal epidermal cell S. disermas, S. garipensis, S. repens IV Head with more than one cell, long stalk, one basal epidermal cell S. garipensis Peltate glandular trichomes consist of a round broad multicellular head with up to sixteen cells, a short stalk cell(s) with cutinized outer walls and a basal epidermal cell (Figure 3.14E). In the species examined here, the secretory cells of the head are arranged in one (Figure 3.3B) or two (Figure 3.3F) concentric circles. The number of cells in the central circle is usually four, while in the periphery the cell number varies according to the species [e.g. 10 peripheric cell in S. aurita (Figures 3.3F) and 12 peripheric cells in S. chamelaeagnea (Figure 3.4B)]. The stalk cell(s) of the peltate trichomes are often sunken to various degrees into grooves in the epidermis (Figures 3.4A, 3.6F and 3.7F). The non-glandular trichomes are uniseriate, unbranched and multicellular and may vary in shape and number of cells (up to seven) among species. However, a stalk composed of 4 or 5 cells is the most common type and was found in the majority of species (Table 3.1). For most species, the base of the non-glandular trichomes is composed of one cell, however, multiple cells at the base of the non-glandular trichomes could be seen in S. albicaulis, S. aurita and S. repens. The various groups of non-glandular trichomes are illustrated in Figure 3.14F-H. 40 The distribution (presence/absence) of the various trichome types found in each investigated species is listed in Table 3.1. Based on SEM and LM observations, the species studied can be arranged into six groups (Table 3.2). The relative abundance of glandular trichomes follows the increasing order: S. disermas > S. garipensis > S. africana-caerulea > S. chamelaeagnea > S. aurita > S. schlechteri > S. repens > S. africana-lutea, while S. dolomitica, S. albicaulis, S. muirii and S. lanceolata have a lower concentration of glandular trichomes. Although the trichome types found vary between species, the use of trichomes alone could not be used to delimit one species from another. Other characters should also be considered. 3.4 Discussion In the Lamiaceae (Labiatae), many species are densely covered with both glandular and non-glandular hairs (Metcalfe and Chalk, 1950; Serrato-Valenti et al., 1997; Bisio et al., 1999). Two types of glandular trichomes generally occur together in a single leaf, viz. capitate and peltate, the latter being the most characteristic of the family (Metcalfe and Chalk, 1950). The capitate trichomes of Salvia species investigated varied and were designated type I, II, III or IV (Figure 3.14A-D). An SEM study of the Salvia stenophylla complex indigenous to southern Africa, similarly revealed the presence of non-glandular and glandular trichomes with the latter being capitate or peltate (Gono-Bwalya, 2003). Investigation of the various organs of S. glutinosa L. revealed that glandular and non- glandular trichomes were present (Kaya et al., 2003). In this study, the SEM and LM observations revealed the presence of both these trichome types on the adaxial and abaxial surfaces of each species (Figures 3.2 to 3.7 and 3.14A-E). Similar glandular trichome types were found in most species studied. However, taking into account the variation that exists in the nature of the stalk (long or short) and in the number of secretory cells in the head, the species investigated could be divided into four distinct groups (Table 3.2). The taxonomic significance of the morphology of glandular hairs in members of the Lamiaceae is not yet determined and it is still being debated in the literature (Falciani et al., 1995; Fahn, 2001). Generally in this family, capitate trichomes are unicellular or 41 multicelullar with short or long stalk cells and one basal epidermal cell. The peltate trichomes are composed of one to sixteen cells and a short stalk cell(s) and one basal epidermal cell (Metcalfe and Chalk, 1950; Hallahan, 2000). The cells of the heads of peltate trichomes are usually arranged in two concentric circles and the number of cells in the central circle is usually four. Werker et al. (1985) investigated the glandular hairs of eight species of the Lamiaceae and found that the central circle was composed of four cells, while the number of cells on the periphery varied from 6 to 14 cells. In the current study, the peltate trichomes are composed of one or two circles, the central circle being composed of four cells, while the number of cells in the periphery varied from 6 to 12 cells (Figures 3.3F and 3.4B) which concurs with previous studies of the Lamiaceae. There is a discrepancy in the literature regarding the type of glandular trichomes which are widespread in the Lamiaceae, in particular, the genus Salvia. Some authors regard the peltate trichome type as the most common in Salvia species (Metcalfe and Chalk, 1950; Werker et al., 1985) while others consider, a peltate trichome with a single-celled stalk and the capitate trichome with a unicellular head to be the most common (Hanlidou et al., 1991). Metcalfe and Chalk (1950) also state that heads with one or two cells are present in capitate trichomes of Salvia. However, Werker et al. (1985) pointed out that the bicellular head was rather rare. Bourett et al. (1994) also observed small capitate glandular trichomes with two secretory cells, which were similar to those found in the current study (Figure 3.14B). In this study, unicellular, bicellular or even four-celled heads (with short or long stalk cells) were present in all 12 species (Table 3.1). On the basis of external morphology, the glandular trichomes of the species studied are similar to those previously described in the Lamiaceae, namely S. glutinosa L. and S. africana-lutea (Serrato-Valenti et al., 1997; Kaya et al., 2003). Non-glandular trichomes in Lamiaceae may be unicellular or multicelullar with up to seven cells in Salvia species (Werker et al., 1985). This is consistent with the observations recorded here, with up to six cells being observed in some species (Figure 3.7A). Trichome types reported in this study have been identified in exotic species such as S. glutinosa and S. officinalis. It is important to note that trichome morphology can vary between the plant organs (leaf, stem or calyx) studied in addition to the stage of development. However, in this 42 investigation, no variation could be observed between the mature and the immature plant material regarding the trichome types. It was shown in S. glutinosa that the presence or absence of various trichome types, as well as their density varied for different plant organs (Kaya et al., 2003). Changwe (2002) also observed that glandular trichomes, like non- glandular ones on the leaves, stems and calyces, are unevenly distributed in Thorncroftia species. An investigation of other organs in the investigated species may well reveal similar variation, but was beyond the scope of this study. Studies have shown that the short-stalked glands with up to sixteen or more secretory cells in the head secrete pleasantly scented oils such as those of Salvia and other aromatic plants e.g. Rosmarinus officinalis and Thymus vulgaris (Metcalfe and Chalk, 1950). Prasium majus (Lamiaceae) which is not aromatic only possesses capitate trichomes (Werker et al., 1985). It is possible to isolate an individual gland from the leaf surface and analyse its contents for chemical composition and presence of specific oils or enzymes (Svoboda and Svoboda, 2000). This study was however not extensive enough to determine whether the glandular hairs produce the fragrance of sage would be interesting to determine. Glandular trichomes contain essential oils as well as other secretions that can cover the surrounding surface of the leaves. Some trichomes have specifically evolved to act as defense mechanisms. Trichomes that occur in large numbers may protect a plant by making its tissue less edible or inhospitable to feeding and breeding insects (Harbone, 1993). Observations made in this study suggest that the secreted material in Salvia species can be released by breaking the cuticle or via pores in the cuticular structures (Figure 3.4B and 3.14A) (Serrato-Valenti et al., 1997; Bisio et al., 1999). While the first mode is common to many aromatic species, the second has only been described for capitate trichomes in the members of the Lamiaceae (Figure 3.14A) (Werker et al., 1985). The outer surface of the non-glandular trichomes sometimes exhibits micro-ornamentation, which may show great diversity causing it to appear micropapillate (Werker, 2000). Micro-ornamentations were not evident in this study. However, micropapillae were only visible under high magnification in the SEM study on the non-glandular trichomes of some species including S. chamelaeagnea, S. lanceolata and S. muirii (Figure 3.4A, 3.6A and 3.6E, respectively). An expanded cuticle of the peltate trichomes at maturity has been reported in other studies in the Lamiaceae (Hallahan, 2000). This cuticle could be seen in 43 all species investigated (e.g. Figure 3.3 B, 3.3F and 3.4B). It is probably through the rupture of the cuticle that secondary metabolites such as essential oils contained in the heads are released. Non-glandular trichomes are also frequent in the Lamiaceae, and may be uniseriate, tufted or branched (Metcalfe and Chalk, 1950; Werker et al., 1985; Kaya et al., 2003). In this study, they were found to be uniseriate and unbranched (Figure 3.14F-H) and the major difference was the number of cells at the base rather than the number of cells forming the stalk. The basal cells of the non-glandular hairs can be used to distinguish S. aurita, S. chamelaeagnea and S. repens from other species. While glandular hairs act as a chemical defense against herbivores or insects, non-glandular hairs constitute a physical defense of the plant (Werker et al., 1985). The distribution of non-glandular trichomes is variable. For some species (S. africana-lutea, S. dolomitica and S. muirii), they might form a dense cover; while in S. aurita, S. chamelaeagnea and S. schlechteri, they are sparse (Figure 3.2C, 3.3D, 3.4A, 3.5A, 3.6D and 3.7D). The reliability of trichomes as taxonomic characters varies from one group of plants to another and may be influenced by many factors such as the plant parts studied and the stage of development of the plant. Trichome characteristics are regarded as being under strong genetic control and less little affected by the environment (Mueller, 1966; Cutler and Brandham, 1977). But, Stace (1965) and Dilcher (1974) suggest that the characteristics might vary according to environmental factors such as soil type and climate. Furthermore, some attributes of the leaf may have arisen in response to evolutionary pressures. The question has often been posed whether the density of the glandular hairs is fixed during the life span of the leaf or increases with age. Some authors have suggested that the number is established early during the leaf differentiation (Ascens?o and Pais, 1987), while others have concluded that it increases throughout all stages of leaf development (Maffei et al., 1989). Taking into consideration these observations, the density of trichomes should be recorded carefully. This is also true with the number of cells in capitate or peltate trichomes, since the number of cells present at a particular stage may increase during a later stage of development (Ascens?o and Pais, 1987). Substances such as polysaccharides, essential oils, salts, sugars and proteins are secreted by glandular trichomes in species of the Lamiaceae (Werker et al., 1985; Karousou et al., 44 1992; Ascens?o et al., 1995). The study of trichomes demonstrated that the species investigated could be divided into groups, but distinction between species was not possible. The composition of the essential oils (Chapter 4) in addition to leaf trichome types observed in Salvia species may facilitate the identification of Salvia species. 3.5 Conclusions ? Two types of glandular trichomes (peltate and capitate) were observed to be present in all 12 species. ? The peltate trichome type is present in all species but varies in the number of head cells (four to sixteen). ? Similarly, the capitate trichomes vary in the number of head cells and the length of the stalk cells. ? As with other members of the Lamiaceae, species of the genus Salvia showed some degree of diversity in leaf trichome types, but not sufficient to distinguish between the investigated species. ? Salvia aurita, S. chamelaeagnea and S. repens have multiple cells at the base of the non-glandular trichomes, unlike the rest of species only one basal cell. ? Based on the type of capitate trichomes, the species investigated could be divided into four distinct groups. 45 Chapter 4: Essential Oil Composition Abstract Eleven of the seventeen species under investigation are pungently aromatic which prompted a study of their essential oils. The essential oils were isolated by hydrodistillation and analyzed by GC and GC-MS. The oil yield was relatively low and ranged between 0.004 (S. radula) to 0.50% (S. muirii) (w/w). Ninety-three components were identified representing 72.9 to 97.5% of the oils. Quantitative and qualitative variations in oil composition were observed among species. The essential oils were dominated by monoterpene hydrocarbons in S. muirii (53.0%), S. stenophylla (41.5%), S. africana-lutea (35.6%) and S. repens (32.6%). Oxygen-containing monoterpenes dominated in S. dolomitica (71.8%) and S. chamelaeagnea (42.8%), while oxygen- containing sesquiterpenes were dominant in the remaining samples. Major components identified include ?-pinene, 1,8-cineole, linalool, limonene, myrcene, ?-caryophyllene, spathulenol, ?-caryophyllene oxide, viridiflorol, ?-3-carene and ?-bisabolol. Quantitative and qualitative heterogeneity in the oils was supported by the essential oil cluster analysis. Only S. africana-caerulea and S. lanceolata showed a very high correlation (Scorr > 0.9). 46 4.1 Introduction Essential oils are odorous, volatile products of an aromatic plant?s secondary metabolism normally formed in special cells, or groups of cells, such as trichomes (Metcalfe and Chalk, 1950; Werker et al., 1985). They are generally concentrated in one particular region of a plant such as leaves, stems, fruits, barks or roots and when they occur in different organs in the same plant, they frequently have different composition profiles (Ara?jo et al., 2003). Despite the strong competition from synthetic products, essential oils currently still occupy an important place in the perfume, fragrance and therapeutic (mainly aromatherapy) industries (Lis-Balchin et al., 1998). Many aromatic plants are also used in traditional medicine for various purposes (Cimanga et al., 2002). The total value of the essential oil industry worldwide was estimated at about 10 billion US dollars with production from the developing countries representing about 55% of world production (IDRC, 2006). Aromatherapy is the therapeutic use of fragrances, or at least volatiles, to cure, mitigate or prevent disease, infection and indisposition by means of inhalation (Buchbauer, 2000). Essential oils from aromatic and medicinal plants have been known to possess biological activity since time immemorial. There are many reports describing the pharmacological properties of essential oils (Baratta et al., 1998; De Feo et al., 2003). This has recently attracted the attention of many scientists and encouraged them to screen plants for the pharmacological properties of their oils. With the growing interest in the use of essential oils in perfume and pharmaceutical industries, a systematic examination of plants for these properties has become increasingly important. Studies on the chemical composition of essential oils have proven to be a valuable tool in the resolution of some taxonomic problems and can help in determining the relationship between different species in the genus. The study of essential oils is important not only from a taxonomic view point, but also from the pharmacological aspect, as plants with similar chemical compositions may theoretically have very few differences in their biological properties. The objectives of this study were to: (i) determine the chemical composition of 11 essential oils using GC and GC- MS analysis, and 47 (ii) determine any possible chemotaxonomic relationship between the investigated species. 4.2 Materials and methods The essential oils were isolated by hydrodistillation as explained in Section 2.3.1 (Figure 2.1) and were analyzed using gas chromatography (Figure 4.1a) and gas chromatography coupled to mass spectrometry (Figure 4.1B) (Pedro et al., 2001). 4.2.1 Gas chromatography GC analyses were performed using a Perkin Elmer 8700 gas chromatograph equipped with two FIDs, a data handling system and a vaporizing injector port into which two columns of different polarities were installed: a DB-1 fused-silica column (30 m x 0.25 mm i.d., film thickness 0.25 ?m; J & W Scientific Inc., Rancho Cordova, CA, USA) and a DB-17 HT fused-silica column (30 m x 0.25 mm i.d., film thickness 0.15 ?m; J & W Scientific Inc.). Figure 4.1 Gas chromatography (A) and gas chromatography coupled to mass spectrometry (B) apparatus. The oven temperature was programmed to run from 45-175 ?C, at 3 ?C/min, then to increase in temperature by 15 ?C/min up to 30 0 ?C, where it was maintained for 10 min; injector and detector temperatures, 280 ?C and 290 ?C, respectively; carrier gas, hydrogen, adjusted to a linear velocity of 30 cm/s. The samples were injected using the 48 split sampling technique, ratio 1:50. The percentage composition of each oil sample was computed from the GC peak areas using the normalization method; the data were calculated as mean values of two injections from each oil sample without using response factors. 4.2.2 Gas chromatography coupled to mass spectrometry The gas chromatography coupled to mass spectrometry (GC-MS) unit (Figure 4.1B) consisted of a Perkin Elmer Autosystem XL gas chromatograph, equipped with DB-1 fused-silica column (30 m x 0.25 mm i.d., film thickness 0.25 ?m; J & W Scientific, Inc.), and interfaced with Perkin-Elmer Turbomass mass spectrometer (software version 4.1). Oven temperature was as above; transfer line temperature, 280 ?C; ion trap temperature, 220 ?C; carrier gas, helium, adju sted to a linear velocity of 30 cm/s; split ratio, 1:40; ionization energy, 70 eV; ionization current, 60 ?A; scan range, 40-300 u; scan time, 1 s. 4.3 Data analysis and chemotaxonomic relationship among species The identity of the components was assigned by comparison of their retention indices relative to C8-C17 n-alkanes and GC-MS spectra with corresponding data of components of reference oils, laboratory-synthesized components and commercially available standards from a home-made library from the Departamento de Biologia Vegetal, Universidade de Lisboa (Portugal). The percentage composition of the essential oil samples was used to determine the relationship between the different samples of Salvia species by cluster analysis using the NTSYS software (Rohlf, 1992). Correlation was selected as a measure of similarity and the unweighted pair-group method with arithmetic average was used for cluster definition. The degree of correlation was evaluated according to Pestana and Gageiro (2000) where a very high correlation ranged between 0.90 and 1.00, high between 0.70 and 0.89, moderate between 0.40 and 0.69, low between 0.20 and 0.30 and very low if less than 0.20. 49 4.4 Results 4.4.1 Percentage yield The essential oil yields among the 11 species ranged from 0.004 to 0.50% (w/w) wet biomass (Figure 4.2). The highest yield was obtained for S. muirii, while S. radula gave the lowest yield. S. afr ica na -ca eru lea S. afr ica na -lu tea S. alb ica uli s S. ch ame lae ag ne a S. do lom itic a S. lan ce ola ta S. mu iri i S. ra du la S. rep en s S. ru nc ina ta S. ste no ph yll a 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.12 0.15 0.06 0.39 0.34 0.10 0.50 0.004 0.13 0.20 0.25 P er ce nt ag e oi l y ie ld Figure 4.2 Percentage essential oil yields (w/w, wet biomass) obtained for eleven indigenous Salvia species. 4.4.2 Essential oil composition The essential oil compositions determined by GC and GC-MS revealed 93 identified components present in the 11 species representing from 72.9% to 97.5% of the total oil composition. The components identified, together with their respective percentage composition in each species, are shown in Table 4.1, in order of their elution in a DB-1 column. The essential oils with high levels of monoterpene hydrocarbons include; S. muirii (53.0%), S. stenophylla (41.5%), S. africana-lutea (35.6%) and S. repens (32.6%) (Table 4.1). Oxygen-containing monoterpenes dominated in S. dolomitica (71.8%) and S. chamelaeagnea (41.8%), while the oils of the remaining species are dominated by oxygen- containing sesquiterpenes and include S. africana-caerulea, S. albicaulis, S. lanceolata, S. radula and S. runcinata (Table 4.1). 50 Table 4.1 Percentage of the components identified in the essential oils of 11 South African Salvia species. Components RI Sac Sal Sab Sch Sdo Sla Smu Sra Sre Sru Sst Tricyclene 921 - tr - - tr - 0.1 - - - 0.1 ?-Thujene 924 0.1 0.6 tr 0.2 - 0.5 0.3 - 4.7 - 0.1 Benzaldehyde 927 - - - - - - - - - tr tr ?-Pinene 930 4.1 6.0 1.9 10.4 tr 2.6 22.3 - tr tr 1.3 Camphene 938 0.1 tr tr 0.2 - tr 7.6 - 3.3 tr 2.0 Sabinene 958 tr 0.9 tr 0.1 0.1 1.4 0.1 - 2.4 tr 0.4 6-Methylhept-5-en-2-one 960 - - - - - - - - - - tr 1-Octen-3-ol 961 tr 0.9 tr 0.1 0.1 1.4 tr 0.2 tr tr - ?-Pinene 963 1.3 0.2 tr 1.9 tr 1.0 6.7 0.2 tr 0.1 0.2 2-Octanone 967 - - - - - - - 0.8 - - - 3-Octanol 974 - - - - - - - - - 0.1 - Myrcene 975 0.2 11.5 tr 1.6 0.6 0.6 1.9 - 1.0 0.1 3.9 ?-2-Carene 983 - - tr - - - - - 0.9 - - ?-Phellandrene 995 - - - - - - - - - - 0.1 ?-3-Carene 1000 - - tr - - - - - 1.4 - 22.6 o-Ocimene 1000 - - - - - - - - - - 0.3 p-Cymene 1003 1.5 7.6 2.5 0.9 0.2 1.1 1.6 0.2 9.5 tr 0.6 1,8-Cineole 1005 0.5 1.9 9.4 40.5 tr 1.8 23.2 0.2 12.8 tr - ?-Phellandrene 1005 - - - - - - - - - tr 3.7 Limonene 1009 1.5 1.6 9.4 9.7 0.6 1.8 11.6 - 9.4 0.5 4.8 cis-?-Ocimene 1017 tr 5.4 - 0.3 0.6 tr 0.5 - - 0.1 - trans-?-Ocimene 1027 tr 1.8 - 0.4 - tr 0.1 - - - 0.2 ?-Terpinene 1035 - - - - tr - 0.1 - - - 0.2 trans-Sabinene hydrate 1037 tr tr - 0.1 - tr - - - - 0.5 cis-Linalool oxide 1045 - - - - 0.2 - - - - - - n-Octanol 1045 - - - - - - - - - 0.1 - trans-Linalool oxide 1059 - - - - 0.2 - - - - - - Terpinolene 1064 - - - - 0.2 - 0.1 - - - 1.0 cis-Sabinene hydrate 1066 0.1 tr - 0.1 - tr - - - - - Linalool 1074 0.3 0.8 tr 0.6 16.6 1.3 - - tr 0.2 0.4 trans-p-2-Menthen-1- ol 1074 0.6 0.2 tr 0.3 - 0.2 - - - - - ?-Campholenal 1088 - - tr - - - - - - - - Camphor 1095 - - 2.3 - - - 8.3 0.3 5.3 0.9 3.2 neo-iso-Thujanol 1101 - - tr - - - - - - - - 51 Components RI Sac Sal Sab Sch Sdo Sla Smu Sra Sre Sru Sst trans-Pinocarveol 1106 - - - - - - 0.2 - - - - cis-p-2-Menthen-1-ol 1110 - - - - - - - - - - tr allo-Ocimene 1110 - - - - 0.8 - - - - - - trans-Verbenol 1114 - - - - - - 0.1 - - - - Pinocarvone 1121 - - - - - - 0.1 - - - - Nerol oxide 1127 - - - - 0.1 - - - - - - ?-Terpineol 1134 tr 0.2 - 0.2 - 0.5 - - - - - Borneol 1134 - - tr - - - 0.6 - 1.0 1.2 2.2 Terpinen-4-ol 1148 0.5 0.6 - 0.4 0.8 1.0 0.9 7.1 - 0.7 0.8 ?-Terpineol 1159 0.7 0.8 2.7 0.5 6.2 1.3 0.7 0.7 tr 0.6 0.8 Myrtenol 1168 - - 1.5 - - - 0.2 - tr 0.8 3.5 trans-Carveol 1189 tr tr tr tr - 0.1 - - tr - - Cuminaldehyde 1200 - - tr - - - - - tr - - Carvone 1206 0.3 tr tr 0.1 - tr - - - - - Citronellol 1210 - - - - 0.4 - - - - - - Geraniol 1236 - - - - 19.6 - - - - - - Linalyl acetate 1245 - - - - 19.6 - - - - - - Cumin alcohol 1260 - - - - - - - - tr - - Bornyl acetate 1265 - - tr - - - 1.8 - 1.3 0.1 - Neryl acetate 1275 - - - - 4.0 - - - - - - cis-Jasmone 1372 tr tr - tr - tr 0.1 0.4 - 0.1 - Geranyl acetate 1370 - - - - 2.3 - - - - - - ?-Gurjunene 1400 tr 1.1 - 0.6 tr 0.2 - - 1.6 - - Isocaryophyllene 1406 - - - - - - tr 8.6 - 0.2 - ?-Caryophyllene 1414 1.9 3.4 2.1 1.7 0.3 5.7 2.2 tr 8.0 10.5 2.1 ?-Gurjunene* 1426 0.7 0.3 - 0.2 - 0.1 - - 1.4 - - Aromadendrene 1428 0.9 2.7 2.0 2.1 - 0.8 0.4 tr - tr - ?-trans-Bergamotene 1434 - - - - - - 0.3 tr - 0.7 1.1 Eudesmadiene* 1435 0.8 tr - 0.4 - tr - - - - tr ?-Humulene 1447 1.2 1.9 tr 1.0 0.2 4.7 0.2 1.6 2.8 2.3 - trans-?-Farnesene 1455 - - - - 0.2 - 0.1 - - 0.4 0.5 allo-Aromadendrene 1456 tr 1.3 tr 1.0 - 2.4 - - 0.9 - - Geranyl-2-butylether 1466 - - - - 1.8 - - - - - - ar-Curcumene 1475 - - - - - - - - - 0.1 - ?-Selinene 1476 - - - - - - tr - - 0.1 - Viridiflorene 1487 0.2 1.3 - 0.7 0.8 0.2 - - 1.8 - - UI-C 1492 - - - - 1.4 - - - - - - UI-A 1494 - - - - 3.5 - - - - - - ?-Bisabolene 1495 - - - - - - 0.1 - - 0.6 3.0 52 Components RI Sac Sal Sab Sch Sdo Sla Smu Sra Sre Sru Sst ?-Cadinene 1500 1.5 4.7 - 0.6 - 2.5 - - 0.8 - - trans-Calamenene 1505 0.3 0.9 - tr - tr - - - - - ?-Sesquiphellandrene 1508 - - - - - - - - - 0.2 - trans-Nerolidol 1549 1.4 3.5 2.1 1.6 - 2.2 1.8 - - 0.7 - ?-Caryophyllene alcohol 1550 1.4 2.0 - 1.6 - 2.2 - - - - - UI-D 1550 - - - - 2.8 - - - - - - Spathulenol 1551 29.1 2.0 2.0 tr 1.5 18.3 - 7.7 2.6 tr - ?-Caryophyllene oxide 1561 14.6 1.9 5.6 0.6 - 14.3 0.6 22.6 4.9 1.3 - Globulol 1566 tr 0.8 2.2 0.5 - 0.6 - - - - - Viridiflorol 1569 0.0 3.3 24.5 9.3 - 0.9 1.9 1.3 5.3 tr - UI-B 1569 - - - - 2.8 - - - - - - Humulene epoxide* 1579 - - - - - - 0.4 12.6 - 0.8 - Ledol 1580 6.5 0.0 6.6 1.4 - 5.2 - - 4.6 - - epi-Cubenol 1600 0.0 1.8 - 0.4 - 0.6 - - - - - T- Cadinol 1616 3.0 1.9 2.0 2.3 - 1.9 - - 1.1 - tr ?-Cadinol 1618 - - - - 0.6 - - 1.8 1.1 tr - ?-Eudesmol 1620 0.7 1.7 tr 0.1 - 1.3 - - - - - Cubenol 1624 - - - - - - - 1.4 - tr - ?-Cadinol 1626 - - tr - - - - - 3.1 - - ?-Eudesmol 1634 2.0 tr - 1.1 - 0.4 - - - - - ?-Bisabolol 1656 - - - - - - 0.2 3.8 - 65.5 26.1 epi-?-Bisabolol 1658 - - 2.0 - - - - - 1.4 - - (Z)- Lanceol 1692 - - - - - - - - - 0.2 1.3 ?-Bisabolol oxide A* 1702 - - - - - - 0.1 1.4 - 0.8 1.0 % Identification 78.0 77.5 80.8 95.8 78.6 81.1 97.5 72.9 94.4 90.0 88.0 Monoterpenes hydrocarbons 8.8 35.6 13.8 25.7 2.3 9.0 53.0 0.4 32.6 0.8 41.5 Oxygen containing monoterpenes 3.0 4.5 15.9 42.8 71.8 6.2 36.2 8.7 20.4 4.6 11.4 Sesquiterpene hydrocarbons 7.5 17.6 4.1 8.3 1.5 16.6 3.3 10.2 17.3 15.1 6.7 Oxygen containing sesquiterpenes 58.7 18.9 47.0 18.9 2.1 47.9 5.0 52.6 24.1 69.3 28.4 Others 0.0 0.9 0.0 0.1 0.1 1.4 0.0 1.0 0.0 0.2 0.0 Monoterpenes 11.8 40.1 29.7 68.5 74.1 15.2 89.2 9.1 53.0 5.4 52.9 Sesquiterpenes 66.2 36.5 51.1 27.2 3.6 64.5 8.3 62.8 41.4 84.4 35.1 RI: Relative retention indices calculated against n-alkanes. Sac: Salvia africana-caerulea; Sal: S. africana- lutea; Sab: S. albicaulis; Sch: S. chamelaeagnea; Sdo: S. dolomitica; Smu: S. muirii; Sla: S. lanceolata; Sra: S. radula; Sre: S. repens; Sru: S. runcinata; Sst: S. stenophylla. UI-A: unidentified compound A; UI-B: unidentified compound B; UI-C: unidentified compound C; UI-D: unidentified compound D; - : not present, tr: trace (< 0.05 %); *: dete rmined by mass spectra only. 53 The components of essential oils identified in high amounts (> 10%) include ?-pinene, 1,8- cineole, linalool, limonene, myrcene, ?-caryophyllene, spathulenol, ?-caryophyllene oxide, viridiflorol, ?-3-carene and ?-bisabolol. The components identified in all 11 species include ?-pinene, p-cymene, ?-terpineol and ?-caryophyllene. The essential oil composition of each species is described briefly below. Salvia africana-caerulea : Forty-three components representing 78.0% of the total oil were identified. The oil was dominated by oxygen-containing sesquiterpenes (58.7%). The percentage of monoterpenes, both monoterpene hydrocarbons and oxygen-containing monoterpenes, represent only 11.8% of identified oil components (Table 4.1). Major components of this oil include spathulenol (29.1%), ?-caryophyllene oxide (14.6%) followed by ledol (6.5%) and ?-pinene (4.1%). These four compounds represent up to 70% of the total oil (Figure 4.3). Salvia africana-lutea : Forty-three components representing 77.5% of the total oil composition were characterized. Monoterpene hydrocarbons (35.6%) and sesquiterpenes (36.5%), both oxygen-containing sesquiterpenes and sesquiterpene hydrocarbons, are the major identified groups of components with myrcene (11.5%), p-cymene (7.6%), ?-pinene (6.0%) and cis-?-ocimene (5.4%) being the major components. They constitute about 39% of the total oil identified in this species (Figure 4.3). Salvia albicaulis : Thirty-eight components were identified in S. albicaulis essential oil and account for 80.8% of the total oil. This oil was characterized by the predominance of oxygen-containing sesquiterpenes (47.0%) compared to monoterpenes (29.7%) (Table 4.1). Viridiflorol (24.5%), 1,8-cineole (9.4%) and limonene (9.4%) were the most prominent components followed by ledol (6.6%) and ?-caryophyllene oxide (5.6%). All these compounds represent 69% of the total oil (Figure 4.3). Salvia chamelaeagnea: Forty-three components were identified in this species representing 95.8% of the total oil composition. Oxygen-containing monoterpenes and oxygen-containing sesquiterpenes represent 42.8% and 18.9% of the identified groups of components, respectively. This oil is characterized by high levels of 1,8-cineole (40.5%) followed by ?-pinene (10.4%), limonene (9.7%) and viridiflorol (9.3%) (Table 4.1). These four compounds constitute 73% of the total oil of S. chamelaeagnea (Figure 4.3). 54 S. afr ica na ca eru lea S. afr ica na -lu tea S. alb ica uli s S. ch am ela ea gn ea S. do lom itic a S. lan ce ola ta S. mu iri i S. ra du la S. rep en s S. ru nc ina ta S. ste no ph yll a 0 10 20 30 40 50 60 70 80 90 100 ? -P i n e n e 1,8-Cineole Terpineol-4-ol Isocaryophyllene Limonene Ledol Myrcene Camphene Linalyl acetate Geraniol Viridiflorol ? -Pinene p-Cymene cis-? -Ocimene Camphor ?-3-Carene Linalool ?-Caryophyllene Spathulenol ?-Caryophyllene oxide ?-Bisabolol Others P er ce nt ag e oi l c om po si ti on Figure 4.3 Variation in relative amounts of common and major components of the essential oils of eleven indigenous Salvia species: Sac: Salvia africana-caerulea; Sal: S. africana-lutea; Sab: S. albicaulis; Sch: S. chamelaeagnea; Sdo: S. dolomitica; Smu: S. muirii; Sla: S. lanceolata; Sra: S. radula; Sre: S. repens; Sru: S. runcinata; Sst: S. stenophylla. Salvia dolomitica: Thirty-six components representing 78.6% of the total oil were identified in S. dolomitica. The oil of this species is characterized by a high level of oxygen-containing monoterpenes (71.8 %), while the sesquiterpenes represent only 3.6% of the identified components. The major components identified include geraniol (19.6%), linalyl acetate (19.6%), linalool (16.6%) and ?-terpineol (6.2%) (Table 4.1). They represent 79% of the compounds identified in S. dolomitica oil (Figure 4.3). In S. dolomitica, four compounds amounting 11% of the total oil were detected at RIs 1494, 1569, 1492 and 1550 but remained unidentified (UI) and were assigned as UA-A, UI-B, UI-C and UI-D, respectively. The mass spectra of these four compounds previously detected in species belonging to the family Umbelliferae are shown in Figure 4.4. Salvia lanceolata: Forty-three components accounting for 81.1% of the total oil were identified. The oil had a predominance of oxygen-containing sesquiterpenes (47.9%), 55 while the monoterpenes represented only 15.2% of the identified components. Major components were spathulenol (18.3%), ?-caryophyllene oxide (14.3%) followed by ?- caryophyllene (5.7%) and ledol (5.2%) (Table 4.1). The four compounds constitute 54% of the total oil of this species (Figure 4.3). 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 6 9 5 ( 3 4 . 0 7 6 ) Sca n EI+ 8 . 2 2 e 441 8169 5543 53 44 67 65 63 79 77 70 15083 93 91 84 95 10997 105 98 140123 121 110 137 124 149 145 151 177 159 163 205178 191 206 220 UI A, RI 1494 X 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 6 9 5 ( 3 4 . 0 7 6 ) Sca n EI+ 8 . 2 2 e 441 8169 5543 53 44 67 65 63 79 77 70 15083 93 91 84 95 10997 105 98 140123 121 110 137 124 149 145 151 177 159 163 205178 191 206 220 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 8 9 5 ( 3 7 . 7 4 3 ) Sca n EI+ 3 . 2 1 e 49741 8179 695543 53 44 51 67 65 59 77 71 93 91 82 107 105 98 177159 153109 140135 121 117 131 149 160 205178 202192 187 220206 UI B, RI 1569 X 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 8 9 5 ( 3 7 . 7 4 3 ) Sca n EI+ 3 . 2 1 e 49741 8179 695543 53 44 51 67 65 59 77 71 93 91 82 107 105 98 177159 153109 140135 121 117 131 149 160 205178 202192 187 220206 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 6 5 0 ( 3 3 . 2 5 1 ) Sca n EI+ 4 . 3 6 e 441 6943 55 53 44 51 67 65 58 15081 79 77 71 9383 91 84 95 109 107 105 98 123121 110 140 137 124 149 151 177 159 163 205 UI C, RI 1492 X 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 6 5 0 ( 3 3 . 2 5 1 ) Sca n EI+ 4 . 3 6 e 441 6943 55 53 44 51 67 65 58 15081 79 77 71 9383 91 84 95 109 107 105 98 123121 110 140 137 124 149 151 177 159 163 205 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 7 9 0 ( 3 5 . 8 1 8 ) Sca n EI+ 2 . 8 5 e 441 81 7943 55 44 53 51 6967 65 58 77 71 107 93 91 83 95 97 98 108 150122109 120 111 135 123 131 137 139 220 177 151 159 163 205 178 UI D, RI 1550 X 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 0 0 0 1 0 0 % Sd 3 1 0 4 1 7 9 0 ( 3 5 . 8 1 8 ) Sca n EI+ 2 . 8 5 e 441 81 7943 55 44 53 51 6967 65 58 77 71 107 93 91 83 95 97 98 108 150122109 120 111 135 123 131 137 139 220 177 151 159 163 205 178 Figure 4.4 Mass spectra of the unidentified compounds [RIs 1494 (UI-A), 1569 (UI-B), 1492 (UI-C) and 1550 (UI-D)] present in Salvia dolomitica oil. 56 Salvia muirii: Thirty-nine components representing 97.5% of the oil were identified and this oil was dominated by the monoterpene hydrocarbons (53.0%) and oxygen-containing monoterpenes (36.2%) as the major classes of components, while sesquiterpenes represented only 8.3% of the identified components (Table 4.1). The major components of this oil include 1,8-cineole (23.2%), ?-pinene (22.3%) followed by limonene (11.6%), camphor (8.6%), camphene (7.6%) and ?-pinene (6.7%). They represent 82% of the identified compounds in S. muirii oil (Figure 4.3). Salvia radula: As shown in Table 4.1, GC and GC-MS analysis of the oil resulted in the identification of only 22 components representing 72.9% of the oil. The major class of components was the oxygen-containing sesquiterpenes (52.6%), while sesquiterpene hydrocarbons and monoterpenes represented only 10.2% and 9.1% of the identified components, respectively (Table 4.1). ?-Caryophyllene oxide (22.6%) and humulene epoxide (12.6%) were the most prominent components followed by isocaryophyllene (8.6%), spathulenol (7.7%) and terpineol-4-ol (7.1%). They represent 80% of the total oil (Figure 4.3). Salvia repens : Analysis of the essential oil resulted in the characterization of thirty-six components representing 94.4% of the oil. Monoterpene hydrocarbons (32.6%) and oxygen-containing sesquiterpenes (24.1%) were the major structural groups of compounds in this oil. The oil has a predominance of 1,8-cineole (12.8%), p-cymene (9.5%) and limonene (9.4%), followed by ?-caryophyllene (8.0 %), camphor (5.3%) and viridiflorol (5.3%) (Table 4.1). They amount to 53% of the total oil (Figure 4.3). Salvia runcinata: Forty-two components representing 90.0% of the total oil were identified and the major structural class of compound was oxygen-containing sesquiterpenes (69.3%), while monoterpenes represented only 5.4% of the identified components (Table 4.1). This oil was dominated by ?-bisabolol (65.5%) and ?- caryophyllene (10.5%). These two compounds represent 84% of S. runcinata oil (Figure 4.3). Salvia stenophylla : The results of the GC and GC-MS showed that 36 components representing 88.0 % of the oil could be identified. This oil was dominated by monoterpene hydrocarbons (41.5%) and oxygen-containing sesquiterpenes (28.4%) with ?-bisabolol 57 (26.1%) and ?-3-carene (22.6%) being the most prominent components followed by limonene (4.8%) (Table 4.3). These compounds contribute to 61% of the total oil composition (Figure 4.3). 4.4.3 Cluster analysis The cluster analysis was constructed using the 93 essential oil components identified in the 11 Salvia species and UI-A, UI-B, UI-C and UI-D and interpreted according to Pestana and Gageiro (2000). Two major clusters (A and B) could be observed (Figure 4.5). One cluster consists of S. dolomitica alone, while the remaining species constitute another cluster. A very high correlation was observed between the essential oil of S. africana- caerulea and S. lanceolata (Scorr ? 0.92) (Figure 4.5). This strong correlation is based on the abundance of spathulenol and ?-caryophyllene oxide present in the two plants (Table 4.1). Figure 4.5 Cluster analysis of eleven Salvia species constructed using the essential oil composition data. 58 A high correlation was obtained between S. chamelaeagnea and S. muirii (Scorr ? 0.77) (Figure 4.5) due to the presence of ?-pinene, 1,8-cineole and limonene in relatively high amounts. A high correlation was also observed between S. runcinata and S. stenophylla (Scorr ? 0.71) with both containing high amounts of ?-bisabolol (Table 4.1). Moderate correlation was observed between S. albicaulis, S. chamelaeagnea, S. muirii and S. repens and also between S. radula, S. africana-caerulea and S. lanceolata (Figure 4.5). S. africana-lutea and S. dolomitica have a low correlation with the nine other species as indicated by the low correlation coefficient (Scorr < 0.30). 4.5 Discussion This investigation showed variability in yield of essential oils of the species studied with S. muirii having the highest yield and S. radula the lowest. The yield obtained in this study was relatively low compared to those reported in literature for other Salvia species. Bellomaria et al. (1992) reported the yield of S. pomifera, S. willeana and S. fruticosa to be 1.3, 2.8 and 5.3%, respectively. Pitarevic et al. (1984) found that the yield of S. officinalis collected at various seasons throughout the year varied from 1.8 to 3.1%. The difference obtained in yield may be explained by the fact that in the current study, the yield was expressed based on wet biomass, while in other studies, the dry biomass was used. Furthermore, difference in yield varies with soil types, plant part used, climate, season and stage of plant development (Putievsky et al., 1986; Bellomaria et al., 1992). The relatively high yield obtained in S. muirii and S. chamelaeagnea may be attributed to high density of glandular trichomes as it has been reported that these trichomes are responsible for the secretion of the essential oils (Werker et al., 1985; Karousou et al., 1992). The essential oils are qualitatively and quantitatively heterogeneous (Figure 4.4). Constituents of essential oils such as ?-pinene, 1,8-cineole, linalool, limonene, myrcene, ?- caryophyllene, spathulenol, ?-caryophyllene oxide, viridiflorol, ?-3-carene and ?-bisabolol were identified in greater amounts (Table 4.1). Components identified in all the 11 species include ?-pinene, myrcene, p-cymene, ?-terpineol and ?-caryophyllene. Many components identified in the current study have been previously identified in non-indigenous Salvia species and these include 1,8-cineole, myrcene, ?-caryophyllene, borneol, camphene, limonene and spathulenol (Pitarevic et al., 1984; Bellomaria et al., 1992; Ahmed et al., 59 1994; Carta et al., 1996; Foray et al., 1999; Kaya et al., 2003). However, in this study, high levels of geraniol and linalyl acetate were detected in S. dolomitica (Table 4.1) in this study, which have rarely been identified in exotic Salvia species (Bellomaria et al., 1992; Farhat et al., 2001). It is important to note that ?-thujene and ?-thujene, which are major components of the essential oil of S. offinalis (Pitarevic et al., 1984), were detected in relatively low amounts (Table 4.1). Many factors may explain variation in essential oil composition between different species. In addition to the genetic make up of each species and factors listed above which affect the essential oil yield, the mode of harvest and isolation procedure may also influence the essential oil composition (Pitarevic et al., 1984; Bellomaria et al., 1992). Ahmed et al. (1994) investigated the essential oil of S. hispanica L. collected from three geographical areas and found major variations in their essential oil composition. Brunke and Hammerschmidt (1985) analysed the essential oil of S. stenophylla and found that ?-carene (20.0%) and ?-bisabolol (29.8%) were the dominant components. In the current study, the two components represent 22.6 and 26.1%, respectively (Table 4.1). Jequier et al. (1980) found that oxygen-containing monoterpenes represent 5% of the oil and the sesquiterpene hydrocarbons 35% in S. stenophylla. The amount of oxygen-containing monoterpenes and sesquiterpene hydrocarbons is this investigation was 11.4% and 6.7%, respectively (Table 4.1). There are many components that were identified in one plant but were absent in others (Table 4.1 and Figure 4.4) and this confirms the heterogeneity of the oils investigated. Based on Cabo et al. (1987) classification, the essential oils of Salvia species can be classified into three groups: Group I is composed of species with a predominance of ?- thujene, ?-thujene and 1,8-cineole; group II is composed of essential oils rich in linalool and linalyl acetate; and group III comprises species rich in 1,8-cineole and camphor. The oils examined in this study do not fall into any of the groups as described by Cabo et al. (1987). Apparently, these findings display new groups (Table 4.1). S. runcinata may be a potential natural source ?-bisabolol since it constitutes more than 65% of the total oil. Components of the essential oil present in only one species and absent in others with an amount greater than 1% include geraniol (19.6%), linalyl acetate (19.6%), neryl acetate (4.0%), geranyl-2-butylether (1.8%), cubenol (1.4%) and unidentified components (UI-A, UI-B, UI-C and UI-D) with 1.4%, 2.8%, 1.4% and 2.8%, respectivel (Table 4.1). Eight components, including the unidentified (UI) UI-A, UI-B, UI-C and UI-D, representing 55.5% of the total oil, were found only in S. dolomitica and were absent in other species. 60 The presence of these components can explain the very low correlation coefficient found between the essential oil of S. dolomitica and the rest of the species (Figure 4.4). The unidentified components of the oils namely UI-A, UI-B, UI-C and UI-D were previously detected in species of the family Umbelliferae (Figueiredo, pers comm), but have not yet been characterised. Their presence in S. dolomitica differentiates this species from others indigenous Salvia species to date. The oil composition of S. africana-caerulea resembles that of S. lanceolata (Table 4.1) and this is supported by the cluster analysis (Figure 4.5). Codd (1985) placed S. africana-caerulea, S. africana-lutea, S. chamelaeagnea and S. lanceolata within the same group based on morphological characters and SEM observations (Chapter 3). The essential oil of S. chamelaeagnea in this study resembles that of S. muirii as shown by the high correlation coefficient found between the two species (Figure 4.5). On the other hand, low similarity was found in essential oil composition of S. africana-lutea and other species within the complex. In the current study a very high correlation was found between S. africana-caerulea and S. lanceolata and provides chemotaxonomic support for the affinity as described by Codd (1985). Similarly, S. repens, S. runcinata and S. stenophylla were also placed in the same group and a high correlation was found between the essential oil of S. runcinata and S. stenophylla (Figure 4.5). The essential oil composition data has provided additional information in order to better understand the the chemistry of the investigated species which will in turn help to understand the biological activities of the essential oils studied in the next Chapters. 4.6 Conclusions ? The essential oil yield varied between species, with S. muirii giving the highest yield and S. radula the lowest yield. ? Quantitative and qualitative variations were observed between the essential oils of the investigated species. ? S. runcinata is the only species where one compound (?-bisabolol) represents more than 65% of the total oil. ? A very high correlation was found between the essential oil of S. africana-caerulea and S. lanceolata due to their high amount of spathulenol and ?-caryophyllene oxide. 61 ? The correlation between the essential oil of S. dolomitica and the rest of the species was very low due to the presence of geraniol, linalyl acetate and four other unidentified components. ? The essential oil data provides chemotaxonomical confirmation previously suggested relationships between taxa based on morphological evidence alone. 62 Chapter 5: HPLC-UV-MS Analysis of the Solvent Extracts Abstract Salvia species are rich in phenolic compounds which are partly responsible for the various biological activities. High performance liquid chromatography was used to detect caffeic acid, carnosic acid, kaempferol, oleanolic acid, rosmarinic acid and ursolic acid in 17 solvent extracts. Compounds isolated from S. chamelaeagnea (7-O-methylepirosmanol and carnosol) and S. radula (betulafolientriol oxide and salvigenin) were also included to verify their presence in other Salvia species. Betulafolientriol oxide was detected in all the investigated species. Rosmarinic acid, carnosic acid, carnosol and oleanolic acid/ursolic acid were detected in relatively high levels. The level of rosmarinic acid (percentage area) ranged from 3.4 (S. radula) to 52.8% (S. muirii) and was absent in S. verbenaca. S. garipensis and S. radula were the only species which lacked oleanolic acid and ursolic acid. Kaempferol, a compound commonly found in exotic Salvia species was not detected in any of the species studied. The solvent extracts also contained other flavonoids (e.g. flavones). 63 5.1 Introduction Salvia is an important genus widely cultivated and used in flavouring and is a popular ingredient folk medicines. They are used in traditional medicine in many parts of the world for the treatment of various diseases (Liu et al., 1992). The genus has attracted great interest, and has been the subject of numerous phytochemical studies (Lu and Foo, 2002). It is a rich source of polyphenols, with an excess of 160 polyphenols having been identified, of which many are unique to the genus (Lu and Foo, 2002). Many Salvia species have received interest due to their ability to accumulate high levels of active compounds, notably flavonoids (Tanaka et al., 1989; Ulubelen et al., 2001). These classes of compounds are known to exhibit a range of biological activities including; anticancer, antibacterial, anti-oxidant and anti-inflammatory properties (Ulubelen et al., 2001, Cuvelier et al., 1994; Lu and Foo, 2002; Kobayashi et al., 2002). Many Lamiaceae extracts are of commercial interest to the food industry as a source of natural anti-oxidants (Thorsen and Hildebrandt, 2003). The quality of anti-oxidant and the price of commercially available species, such as Rosmarinus officinalis extracts, are highly correlated with the total phenolic compounds, for example the levels of carnosic acid, rosmarinic acid and caffeic acid (Thorsen and Hildebrandt, 2003). A large number of polyphenolic compounds are constructed from the caffeic acid building block via a variety of condensation reactions (Lu and Foo, 2002). As Salvia species are also rich in phenolic compounds, accurate identification is of great importance for commercial development. The identification of these compounds will also help to understand the pharmacological properties of species studied. High performance liquid chromatography (HPLC) of phenolic and triterpenoid compounds is a widely used methodology designed to separate, quantify, identify and analyse components in a chemical mixture (Areias et al., 2001). The objectives of this study were to: (i) determine the presence of some phenolic and triterpenoid compounds, previously identified in exotic Salvia species (caffeic acid, carnosic acid, kaempferol, oleanolic acid, rosmarinic acid and ursolic acid) in the solvent extracts of indigenous Salvia species, 64 (ii) confirm the presence of compounds isolated from S. chamelaeagnea (7-O- methylepirosmanol and carnosol) and S. radula (betulafolientriol oxide and salvigenin) in the other species, and (iii) determine among these compounds those which are the most common in indigenous Salvia species. 5.2 Materials and methods 5.2.1 Chemicals, reagents and standards Analytical grade acetonitrile, methanol and chloroform were purchased from Rochelle Chemicals, while standards of caffeic acid, carnosic acid, kaempferol, oleanolic acid, rosmarinic acid and ursolic acid were obtained from Sigma? . Carnosol and 7-O- methylepirosmanol were isolated from S. chamelaeagnea (Chapter 9), while betulafolientriol oxide and salvigenin [5-hydroxy-6,7-dimethoxy-2-(4-methoxyphenyl)- 4H-1-benzopyran-4-one] were isolated from S. radula (Chapter 8). 5.2.2 Preparation of samples The methanol:chloroform (1:1) extracts were prepared as described in Section 2.3.2. The 17 samples and standards for the analysis were prepared at a concentration of 50 mg/ml in methanol. They were filtered using Whatman? No 1 filter paper, stored in vials at 4 ?C until analysed. 5.2.3 High performance liquid chromatography The detection and identification of compounds were performed using a Waters 2695 HPLC system equipped with both a 2996 photodiode array (PDA) detector and a Thermabeam electron impact (EI) mass selective detector (TMD). The TMD detector was operated in EI mode with the ionizer at 70 eV (fixed) with a gain of 10 and scanning mass range of 50- 550 amu. Analysis was performed on a Phenomenex Aqua C18 column (250 x 2.1 mm, 5 ?m) thermostated at 40 ?C. The flow rate of the HPLC was 0.2 ml/min and the gas flow through the nebulizer 30 l/h. The nebulizer temperature was 80 ?C, the expansion region 90 ?C and the source temperature 225 ?C. The TMD was operated in positive ion mode 65 with no flow splitting, thereby utilizing the full HPLC eluent at a flow rate of 0.2 ml/min. The mobile phase started with 10% acetonitrile, 90% water containing 10 mM formic acid. The solvent ratio was changed through a linear gradient to 90% acetonitrile, 10% water (with 10 mM formic acid) at 40 minutes. This ratio was maintained for 10 minutes where after the solvent ratio was changed back to the initial conditions. Standards of caffeic acid, carnosic acid, kaempferol, oleanolic acid, rosmarinic acid and ursolic acid (10 ?l, 50 mg/ml) and isolated compounds (50 mg/ml) from S. chamelaeagnea (carnosol and 7-O- methyl-epi-rosmanol) and S. radula (betulafolientriol oxide and salvigenin) were injected separately and co-injected with samples (10 ?l) in order to identify and confirm their presence in the extracts. 5.3 Data analysis Comparison of the MS spectra of standards and isolated compounds after fragmentation of the parent ion together with retention time (Rt) and ultra-violet (UV) spectra were used to confirm the presence/absence of these compounds in the solvent extracts using the Empower? software. Flavonoids were tentatively assigned by comparing the UV absorption spectra of different flavonoids (Markham, 1982). The area (%) was calculated from the integrated peak area. 5.4 Results 5.4.1 Identification of compounds The typical UV spectra (Figure 5.1) and the chromatographic data (Table 5.1) obtained for the standards and isolated compounds are shown below. The HPLC analysis revealed the presence of various compounds in Salvia species (Table 5.2). Betulafolientriol oxide, a compound isolated from S. radula was the only compound identified in all 17 species at relatively low levels (Table 5.2). Rosmarinic acid, carnosic acid, carnosol and ursolic acid were present in the majority of the extracts in relatively high levels. The percentage area of rosmarinic acid ranged from 3.41 (S. radula) to 52.82% (S. muirii) (Table 5.2). S. albicaulis (35.23%), S. runcinata (38.40%) and S. muirii (52.82%) were particularly rich in rosmarinic acid (Table 5.2) and S. verbenaca was the only species 66 in which rosmarinic acid was not detected. Carnosol and/or carnosic acid was/were abundant in S. aurita, S. chamelaeagnea, S. namaensis and S. stenophylla (area > 20%). Salvigenin (23.76%) and oleanolic acid/ursolic acid (19.85%) were abundant in S. disermas (Table 5.2). AU 0 . 0 0 0 . 5 0 1 . 0 0 1 . 5 0 2 . 0 0 2 . 5 0 3 . 0 0 n m 2 0 0 . 0 0 2 2 0 . 0 0 2 4 0 . 0 0 2 6 0 . 0 0 2 8 0 . 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 3 2 8 . 0 Rosmarinic acid AU 0 . 4 0 0 . 6 0 0 . 8 0 1 . 0 0 1 . 2 0 1 . 4 0 1 . 6 0 1 . 8 0 2 . 0 0 n m 2 0 0 . 0 0 2 2 0 . 0 0 2 4 0 . 0 0 2 6 0 . 0 0 2 8 0 . 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 2 6 5 . 2 3 6 5 . 9 Kaempherol AU - 0 . 0 8 - 0 . 0 6 - 0 . 0 4 - 0 . 0 2 0 . 0 0 0 . 0 2 0 . 0 4 n m 2 0 0 . 0 0 2 2 0 . 0 0 2 4 0. 0 0 2 6 0 . 0 0 2 8 0 . 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 2 4 0 . 4 2 9 0 . 0 3 7 3 . 1 39 8 . 3 7-O-Methylepi rosmanol AU - 0 . 1 0 - 0 . 0 5 0 . 0 0 0 . 0 5 0 . 1 0 n m 2 0 0 . 0 0 2 2 0. 0 0 2 4 0 . 0 0 2 6 0 . 0 0 2 8 0. 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 2 7 7 . 0 3 2 9 . 2 Salvigenin AU 0 . 0 0 0 . 2 0 0 . 4 0 0 . 6 0 0 . 8 0 1 . 0 0 1 . 2 0 1 . 4 0 1 . 6 0 1 . 8 0 2 . 0 0 2 . 2 0 n m 2 0 0 . 0 0 2 2 0 . 0 0 2 4 0 . 0 0 2 6 0 . 0 0 2 8 0 . 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 2 1 6 . 9 2 3 9 . 2 3 2 5 . 6 Caffeic acid H O H O O H O O H O H H O O H O H O O O O O H O H OO H H O O H H O O O O C H 3 O O OO H O O AU 0 . 0 0 0 . 0 5 0 . 1 0 0 . 1 5 0 . 2 0 0 . 2 5 0 . 3 0 0 . 3 5 0 . 4 0 n m 2 0 0 . 0 0 2 2 0 . 0 0 2 4 0 . 0 0 2 6 0 . 0 0 2 8 0 . 0 0 3 0 0 . 0 0 3 2 0 . 0 0 3 4 0 . 0 0 3 6 0 . 0 0 3 8 0 . 0 0 4 0 0. 0 0 2 8 4 . 1 3 2 5 . 6 3 7 3 . 1 Carnosol O H H O O O Oleanolic acidH O C O O H Ursolic acid H O C O O H Betulafolientriol oxide H O O H O O H AU 0.00 0.10 0.20 0.30 0.40 0.50 nm 200.00 220.00 240.00 260.00 280.00 300.00 320.00 340.00 360.00 380.00 400.0 0 232.2 285. 3 H O O H H H O O C Carnosic acid Figure 5.1 UV spectra of standards and isolated compounds from S. radula and S. chamelaeagnea. The UV spectra of ursolic acid, oleanolic acid and betulafolientriol oxide are not reported because they are not good chromophores. 67 Table 5.1 Chromatographic data obtained for the various standards according to their retention times. Number Standards/Isolated compounds Rt (min) Absorption maxima (?max ) in nm 1 Caffeic acid 15.41 217, 239, 292, 328 2 Rosmarinic acid 23.25 204, 328 3 Kaempferol 30.38 266, 365 4 7-O-Methylepirosmanol 34.39 240, 290, 373, 378 5 Salvigenin 34.60 216, 277, 330 6 Carnosol 34.85 284, 323, 373 7 Carnosic acid 37.03 232, 285 8 Oleanolic acid 41.11 - 9 Ursolic acid 41.09 - 10 Betulafolientriol oxide 43.28 - -: poor UV absorption The only other species in which 7-O-methylepirosmanol was detected was S. namaensis. Kaempferol reported in exotic Salvia species was absent in all the 17 Salvia species investigated (Table 5.2). 5.4.2 Detection of other flavonoids in the solvent extracts The results of the HPLC-UV analyses of the 17 species examined showed several compounds exhibiting UV spectra typical of flavonoids (Markham, 1982). The current investigation focused mostly on flavones. Approximately 70% of the plants studied contained flavones based on UV spectra only (Table 5.3). In S. disermas, various flavones were recorded, while a few were noted in S. africana-caerulea, S. aurita, S. chamelaeagnea, S. dolomitica, S. garipensis, S. lanceolata, S. namaensis, S. radula, S. repens, S. runcinata, S. schlechteri and S. stenophylla (Table 5.3). In five species namely; S. africana-lutea, S. albicaulis, S. lanceolata, S. muirii and S. verbenaca, no flavones other than those mentioned in Table 5.2 were detected (Table 5.3). The flavone at a retention time of 20.76 min was found in S. chamelaeagnea, S. dolomitica and S. garipensis (Table 5.3). The HPLC chromatograms of all the investigated species are presented in Figures 5.2. 68 Table 5.2 Area (%) and identification of standards and isolated compounds in the solvent extracts of 17 indigenous Salvia species. Species Caffeic acid (1) Rosmarinic acid (2) Kaemp- ferol(3) 7- O -Methyl- epirosmanol (4) Salvigenin (5) Carnosol (6) Carnosic acid (7) Oleanolic/ Ursolic acid (8 or 9) Betulafolien-triol oxide (10) S. africana-caerulea 5.29 22.90 - - - - 11.55 3.21 0.92 S. africana-lutea - 23.82 - - - - 14.22 4.49 0.75 S. albicaulis 0.58 35.23 - - - 1.10 - 10.23 1.26 S. aurita 0.84 24.44 - - - 27.61 6.37 0.10 0.68 S. chamelaeagnea 0.40 26.36 - 0.87 - 22.48 27.25 0.49 0.52 S. disermas 0.25 3.62 - - 23.76 - - 19.85 2.06 S. dolomitica 0.75 14.05 - - - - - 0.10 4.53 S. garipensis - 29.22 - - 7.38 - - - 2.91 S. lanceolata 3.29 10.67 - - - 2.05 - 14.28 3.83 S. muirii 0.87 52.82 - - 0.10 4.83 11.80 0.05 1.41 S. namaensis 0.92 15.60 - 1.19 - 30.97 20.25 0.10 1.00 S. radula - 3.41 - - 8.52 - - - 1.94 S. repens 0.47 23.66 - - - 13.40 13.01 0.10 2.56 S. runcinata 0.34 38.40 - - 3.55 - 3.36 0.20 1.05 S. schlechteri 0.33 26.58 - - - 10.88 10.38 0.10 0.80 S. stenophylla - 22.17 - - - 23.60 25.17 0.10 1.80 S. verbenaca - - - - 1.80 - - 2.19 2.76 - : not detected 69 Table 5.3 Compounds exhibiting a typical flavonoid spectrum in the solvent extracts of 17 Salvia species. Retention time UV?max (nm) Sac Sal Sab Sau Sch Sdi Sdo Sga Sla Smu Sna Sra Sre Sru Ssc Sst Sve 13.57 238, 328 + + + 14.98 242, 327 + 15.41 236, 327 + + 16.54 234, 327 + + 17.56 242, 329 + + 19.60 253, 348 + + 20.61 252, 341 + + 20.73 251, 269, 339 + 20.76 256, 345 + + + 20.86 269, 335 + 22.70 250, 341 + + 23.63 252, 347 + 27.83 269, 334 + + 28.55 265, 341 + + 30.33 288, 329 + 30.64 252, 341 + + 31.48 269, 329 + + 33.12 269, 337 + 41.72 272, 324 + + 42.48 272, 325 + 50.46 272, 326 + Sac: S. africana-caerulea, Sal: S. africana-lutea, Aab: S. albicaulis, Sau: S. aurita, Sch: S. chamelaeagnea, Sdi: S. disermas, Sdo: S. dolomitica, Sga: S. garipensis, Sla: S. lanceolata, Smu: S. muirii, Sna: S. namaensis, Sra: S. radula, Sre: S. repens, Sru: S. runcinata, Ssc: S. schlechteri, Sst: S. stenophylla, Sve: S. verbenaca. 70 AU 0. 00 2. 00 4. 00 AU 0. 00 2. 00 AU 0. 00 0. 20 0. 40 0. 60 AU 0. 00 1. 00 2. 00 AU 0. 00 1. 00 2. 00 AU 0. 00 0. 20 AU 0. 00 0. 10 0. 20 AU 0. 00 0. 20 AU 0. 00 0. 10 Mi nut es 0. 00 5. 00 10. 0 0 1 5. 00 20. 00 25. 00 30. 00 35. 00 40. 0 0 45. 00 50. 00 55. 00 60. 0 0 S. africana-caerulea S. africana-lutea S. albicaulis S. aurita S. chamelaeagnea S. disermas S. dolomitica S. garipensis S. lanceolata 2 2 2 2 2 2 2 2 2 6 6 6 7 7 7 7 8/9 8/ 9 8/ 9 8/ 9 8/ 9 Figure 5.2 HPLC chromatograms of South African Salvia species. Rosmarinic acid (2), carnosol (6), carnosic acid (7), oleanolic acid (8) and ursolic acid (9). 71 AU 0. 00 1. 00 2. 00 AU 0. 00 1. 00 2. 00 3. 00 AU 0. 00 0. 20 AU 0. 00 0. 50 1. 00 AU 0. 00 0. 50 1. 00 AU 0. 00 1. 00 AU 0. 00 1. 00 2. 00 AU 0. 00 0. 50 1. 00 Minut e s 0. 00 5 .00 1 0. 00 15. 00 20. 00 2 5. 00 30. 00 35. 00 40. 00 45. 00 50. 00 55. 00 60. 0 0 S. muirii S. namaensis S. radula S. repens S. runcinata S. schlechteri S. stenophylla S. verbenaca 2 2 2 2 2 2 2 6 6 6 6 6 7 7 7 7 7 7 Figure 5.2 Contd. HPLC chromatograms of South African Salvia species. Rosmarinic acid (2); carnosol (6) and carnosic acid (7). 72 5.5 Discussion Phenolic compounds are widely distributed in Salvia species (Adzet et al., 1987; Ulubelen and Tuzlaci, 1990) and the flavonoids are mostly present as flavones, flavonols and their glycosides (Lu and Foo, 2002). Flavones were identified in 12 species (Table 5.3), while in five species namely; S. africana-lutea, S. albicaulis, S. lanceolata, S. muirii and S. verbenaca, no flavones were detected (Table 5.3). The majority of flavonoids found in exotic Salvia species are flavones of the apigenin and luteolin type and their corresponding hydroxylated derivatives, while the flavonols are mostly those of kaempferol and quercetin (Lu and Foo, 2002). However, kaempferol was not detected in any of the investigated species (Table 5.2). The majority of the phenolic acids in Salvia species are exclusively those of caffeic acid derivatives. Caffeic acid plays a central role in the biochemistry of Lamiaceae and occurs predominantly in the dimer form as rosmarinic acid (Gerhardt and Schroeter, 1983). In many Salvia species, caffeic acid is the building block of a variety of plant metabolites, ranging from the simple monomers to multiple condensation products that give rise to a variety of oligomers (Lu and Foo, 2002). The trimers and tetramers are also interesting from a therapeutic point of view as they have demonstrated various biological activities such as anti-oxidant, antimicrobial and anticancer activities (Lu and Foo, 2002). The monomers that are frequently present in Salvia species are represented by caffeic acid (Qian and Li, 1992) and other monomeric derivatives including ferulic acid (Cuvelier et al., 1996) and isoferulic acid (Ai and Li, 1992). The caffeic acid was present in 70% of the solvent extracts investigated (Table 5.2). Caffeic acid was also identified in a dimer form (rosmarinic acid) in 94% of the species investigated. Before the chemical structure of rosmarinic acid was elucidated, rosmarinic acid and similar compounds have been known as ?Labiate ngerbstoffe?, a type of tannin known from species of the Lamiaceae (Petersen and Simmonds, 2003). Rosmarinic acid occurs throughout the Boraginaceae whereas within the Lamiaceae it is restricted to the sub- family Nepetoideae (Janics?k et al., 1999). Rosmarinic acid is the most abundant caffeic dimer in Salvia species (Ai and Li, 1988; Cuvelier et al., 1994; Lu and Foo, 1999) and has been reported to be the major phenolic compound of Salvia samples (Cuvelier et al., 1996). 73 This was confirmed by the results obtained in this study. Rosmarinic acid was detected in nerly all the species srudied and S. verbenaca was the only species lacking this compound (Table 5.2). S. albicaulis (35.23%), S. runcinata (38.40%) and S. muirii (52.82%) contained high levels of rosmarinic acid (Table 5.2). Phenolic and triterpenoid compounds reported here have certainly contributed to various biological activities of the solvent extracts reported in the next Chapters. Several studies have reported on the detection and quantification of rosmarinic acid in various Salvia preparations (Janics?k and M?th?, 1997, Yaun et al., 1998; Janics?k et al., 1999; Areias et al., 2000). The variation in levels of caffeic acid and rosmarinic acid in the same plant has been shown. Janics?k et al. (1999) demonstrated that concentrations of caffeic acid were always much lower than those of rosmarinic acid. In all the species studied where the two compounds were detected together, the same pattern was observed (Table 5.2). All the compounds detected in southern African species investigated have previously been identified in exotic Salvia species (Cuvelier et al., 1996; Baricevic et al., 2001; Liu, 2005). As phenolic and triterpenoid compounds are known to exhibit a range of biological activities, this prompted an investigation on biological activities of the solvent extracts in the subsequent Chapters. 5.6 Conclusions ? Salvia species are a rich source of polyphenolic flavonoids and phenolic acids, notably rosmarinic acid, carnosic acid, caffeic acid, carnosol and ursolic acid. ? Betulafolientriol oxide, a compound isolated from S. radula was detected in low amount in all the species studied and if found in species at various locations, it may be used as taxonomic marker for indigenous species. ? High level of rosmarinic acid was recorded in the majority of the species studied particularly in S. muirii and S. runcinata while it was absent in S. verbenaca. ? Salvia disermas has high levels of salvigenin and oleanolic/ursolic acid. ? Kaempferol, a compound commonly detected in exotic Salvia species was absent in all the species studied. 74 Chapter 6: Anti-Oxidant Activity and Total Phenolic Content Abstract The anti-oxidant activity of 17 solvent extracts and 11 essential oils of Salvia species were evaluated. Anti-oxidant activity was measured using the 2,2?-azino-bis-(3- ethylbenzothiazoline-6-sulfonic acid) (ABTS) and the 2,2,-diphenyl-1-picrylhydrazyl (DPPH) scavenging assays and compared to the results with the controls obtained Trolox? and vitamin C, respectively. Nearly all the solvent extracts displayed anti-oxidant activity with IC50 values ranging from 1.61 to 74.50 ?g/ml using the DPPH? radical, while the IC50 values ranged from 11.88 to 69.26 ?g/ml with the ABTS?+ radical. The solvent extract of S. schlechteri, with an IC50 value of 1.61 ?g/ml, was three times more active than the reference compound, vitamin C (IC50 value: 4.72 ?g/ml). All the essential oils displayed poor activity with all the IC50 values being greater than 100 ?g/ml using both the DPPH? and ABTS?+ radicals. Total phenolic content based on gallic acid equivalents (GAE) confirmed the presence of total soluble phenolics in the various extracts from 45 to 211 mg of GAE dry sample and showed strong association (r2 = 0.90) with anti-oxidant activity (DPPH assay). 75 6.1 Introduction Recent developments in biomedical science emphasise the involvement of free radicals in many diseases. There is increasing evidence to suggest that many degenerative diseases such as brain dysfunction, cancer, heart disease and immune system decline could be the result of cellular damage caused by free radicals and that anti-oxidants may play an important role in disease prevention (Aruoma, 1998). Anti-oxidants are compounds that inhibit or delay the oxidation of other molecules by inhibiting the initiation or propagation of oxidizing chain reactions. The anti-oxidant activity of natural substances is based on their ability to donate hydrogen atoms to free radicals. Many phenolic compounds exhibit a wide range of biological effects including antibacterial and anti-inflammatory activity (Cook and Samman, 1996). Studies have also shown that phenolic compounds are potent scavengers of free radicals and as such, are potentially useful in the prevention of a number of diseases (Zainol et al., 2003). Plants are a potential source of natural anti-oxidants. Plants have played a significant role in the development of new drugs and in many developing countries attention has been paid to exploring natural substances as substitutes for synthetic compounds. In recent years, phytochemicals have appeared in the market and are generally regarded as safe (Wanasundara and Shahidi, 1998). The commonly used anti-oxidants, butylated hydroxyanisol and butylated hydroxytolune are synthetic chemicals and the possible toxicity of these anti-oxidants has resulted in their reduced usage (Ito et al., 1985). Due to health concerns, natural anti-oxidants have been extensively employed in recent years (Yen et al., 2003). The reported high anti-oxidant activity and use of Salvia in local traditional medicine prompted more extensive investigation in order to provide a better understanding of medicinal value of indigenous species. A literature survey revealed no reference to previous work on the anti-oxidant activity of Salvia species found in South Africa. Most studies have focused on S. officinalis (garden sage) because of its well-known and widely documented anti-oxidant properties (Tepe et al., 2006). The objectives of this study were to: (i) evaluate the anti-oxidant activity of South African Salvia species using the 76 ABTS?+ and DPPH? scavenging assays, (ii) determine the total phenolic content present in each species, and (iii) correlate the anti-oxidant activity to the total phenolic content. 6.2 Materials and methods 6.2.1 Reagents, chemicals and standards Dimethyl sulfoxide (DMSO) and HPLC grade methanol were purchased from Rochelle Chemicals. The 2,2-diphenyl-1-picrylhydrazyl (DPPH) and potassium persulfate (K2S2O8) were purchased from Fluka, while the 2,2?- azino-bis (3-ethylbenzothiazoline-6-sufonic acid) (ABTS), 3,4,5-trihydroxybenzoic acid commercially known as gallic acid and the standard 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox? ) were obtained from Sigma? . Sodium hydroxide (NaOH) and vitamin C were obtained from Saarchem. Potassium dihydrogen phosphate (KH2PO4), Folin-Ciocalteu phenol reagent, HPLC grade ethanol and anhydrous sodium carbonate were purchased from Merck. Double distilled Millipore ? (0.22 ?m) water was used throughout. 6.2.2 Evaluation of the anti-oxidant activity The anti-oxidant activity of volatile (essential oils) and non-volatile (solvent extracts) samples were evaluated using both the ABTS and the DPPH assays. 6.2.2.1 The ABTS method The anti-oxidant activity was determined using the method described by Villa?o et al. (2004) with slight modifications. (i) Preparation of the ABTS?+ radical and stock solution of test samples The ABTS solution was prepared at a concentration of 7 mM and the potassium persulfate solution at a concentration of 140 mM with double distilled water. The radical cation ABTS?+ was chemically generated with K2S2O8 and the ABTS solution. Hence 88 ?l of the K2S2O8 solution was added to 5 ml of ABTS solution, mixed well, covered with foil and kept at room temperature for 12-16 hours before used in the experiment. The ABTS?+ radical (stable for 2-3 days) was diluted with cold ethanol and maintained on ice 77 throughout the experiment. The experiment was performed only if the absorbance of the radical was 0.70 ? 0.02 at 734 nm. Stock soluti ons of test samples were prepared at starting concentrations of 10 mg/ml in DMSO (extracts and essential oils). (ii) The ABTS assay From the stock solution of solvent extracts and essential oils, seven two-fold dilutions were prepared with DMSO. To a sample volume of 50 ?l of each concentration, in a cuvette, a volume of 1 ml of the ABTS?+ radical was added and kept at 30 ?C in a water bath for four minutes before the absorbance was recorded at 734 nm. The final concentrations ranged from 0.05 to 238.10 ?g/ml. Each sample was tested in duplicate. The principle of the reaction is that the anti-oxidant compounds react with the ABTS?+ radical. During the reaction, the solution becomes clear or decreases in colour intensity, thus decreasing the absorbance. The mechanism of reaction of the ABTS?+ radical is shown below (Figure 6.1). The percentage of decolourisation was calculated from (Eq. 6.1) using a programme written on Microsoft Excel? . The colour interference of the test sample was not taken into consideration because there was very little variation. % Decolourisation = 100 x (1 - Abs test sample) (Eq 6.1) Abs control Where: Abs = absorbance at 734 nm 6.2.2.2 The DPPH method (i) Preparation of the DPPH? radical and stock solution of test samples The DPPH? radical (96 ?M) was prepared in HPLC grade methanol. The radical, protected from light and stored at 4 ?C (stable for a pproximately a week), has an absorbance of 0.38 ? 0.03 at 550 nm. The stock solution of the test sample (essential oils and solvent extracts) was prepared at a concentration of 10 mg/ml in DMSO. (ii) The DPPH assay The DPPH? radical is a stable radical that can readily undergo reduction by an anti-oxidant (Lu and Foo, 2001). Because of the ease and convenience of this reaction, it now has widespread use in the free radical-scavenging activity assessment (Sanchez-Moreno et al., 1998; Lu and Foo, 2001; Mambro et al., 2003). A DPPH assay (Mambro et al., 2003; Lourens et al., 2004) was also employed to investigate the anti-oxidant activity of different plant extracts and essential oils. 78 - 3 3 - ABTS S OS N E t NN S N E t O S SOS N E t NN S N E t O S S OS N E t NN S N E t O S S OS N E t NN S N E t O S 3 3 S OS N E t NN S N E t O S - - . ABTS Radical S OS N E t NN S N E t O S Potassium persulfate ?e- (peaks at 645, 734, and 815 nm) Figure 6.1 Mechanism of action of the ABTS?+ radical (Dorman et al., 2000). From the stock solution of the test sample, a 1:20 dilution was made with DMSO followed by five consecutive two-fold dilutions. For each concentration, 50 ?l of the test sample was plated out in triplicate in a 96 well-plate with the appropriate control for the test sample (in order to compensate for the colour interference of the test sample). For the DPPH and methanol controls, 50 ?l of DMSO was added into the wells. A volume of 200 ?l of DPPH solution was added to the test samples (including the DPPH control wells with only DMSO), while 200 ?l of HPLC grade methanol was added to the methanol control wells including the methanol control wells with only DMSO). The final concentrations of the test samples in the wells ranged from 3.13 to 100 ?g/ml. However, due to their high activity, 12 dilutions were made for the reference compound as well as for S. schlechteri. The plate was then shaken at 960 rpm for 2 minutes in a microtitre plate reader (Labsystems Multiskan RC connected to a computer equipped with Genesis? version 3.03 software) and incubated in the dark, at room temperature, for 30 minutes. The principle of the method is that anti-oxidant compounds donate electron(s) to DPPH? (Figure 6.2 and 6.3) resulting in decolourisation that is stoichiometric with respect to the electrons captured by DPPH (Hristea et al., 2002). 79 The absorbance was read after incubation using a spectrophotometer (Labsystems Multiskan RC) equipped with Genesis? version 3.03 software at a single wavelength of 550 nm and the percentage of decolourisation was determined using a simple programme written in Microsoft Excel? (Eq. 6.2). % Decolourisation = 100 x (Abs control - (Avg test Abs + Avg Abs methanol) (Eq. 6.2) Abs control Where: Abs = absorbance Avg = average Abs control = Average absorbance DPPH ? Average absorbance methanol Avg test Abs = Average absorbance obtaine d in the wells containing DPPH and test sample Avg Abs methanol = Average absorban ce obtained in the wells containing methanol and test sample (no DPPH) Figure 6.2 Mechanism of reaction of the DPPH? radical (Brand-Williams et al., 1995): ((FlOH: flavonoid which can donate H2; FlO: flavonoid which has donated H2 to the DPPH? radical). DPPH (purple colour) N N N O 2 NO 2 O 2 N . + FlOH O 2 N NO 2 NO 2 NN H + FlO. DPPH (reduced to yellow due to the presence of anti-oxidant compounds) 80 Figure 6.3 Microtitre plate showing the decolourisation of the DPPH solution in the presence of anti-oxidant compounds. HC: high concentrations of test compound with control; LC: low concentration of test compound with control, first and last row are used for DPPH and methanol controls. 6.2.2.3 Qualitative determination of the anti-oxidant activity of the solvent extracts The anti-oxidant activity of the solvent extracts was also evaluated using a TLC method. A 3 ?l sample (5 mg/ml in methanol) was spotted on two separate TLC plates (Alugram? Sil G/UV254, 0.200 mm) and eluted with ethyl acetate:toluene:acetic acid (10:5:0.01). One of the TLC plates was then sprayed with 4% DPPH? radical and another sprayed with the ABTS?+ radical (prepared with 7 mM ABTS solution and 140 mM K2S2O8 in a ratio 3:1). 6.2.3 Determination of total phenolic content 6.2.3.1 Preparation of sodium carbonate and Folin-Ciocalteu reagent Anhydrous sodium carbonate was prepared at a concentration of 75 g/l. The solution was stirred and heated until the sodium carbonate was completely dissolved before the solution was filtered using a Whatman? No 1 paper filter. The Folin-Ciocalteu stock solution was prepared at a concentration of 0.1% (v/v) with double distilled water. 6.2.3.2 Preparation of standard and test samples Gallic acid was used as the standard to quantify the phenolic content in the test samples. The stock solution and subsequent dilutions (concentration ranged from 10 to 250 mg/ml) HC LC Decolourisation of sample extract due to anti-oxidant compounds LC Sample plated out in triplicate (test sample and control) DPPH+DMSO Methanol+DMSO HC 81 were prepared with DMSO. The test samples (4 mg) were dissolved in 4 ml of DMSO and filtered through Whatman? No 1 filter paper. 6.2.3.3 Method for determining total phenolic content The total phenolic content of the solvent extracts was determined by the method using Folin-Ciocalteu and gallic acid as standards (Slinkard and Singleton, 1977; Chandler and Dodds, 1983). A 2 ml volume test sample was mixed with 10 ml of Folin-Ciocalteu reagent followed by the addition of 8 ml of anhydrous sodium carbonate. After incubation at room temperature in the dark for 2 hours, the absorbance of the reaction mixture was read twice at 765 nm against a sample containing only DMSO, using a double beam spectrophotometer (Hitachi). 6.3 Data analysis For both the ABTS and DPPH assays, the IC50 values (concentration at which 50% of decolourisation is obtained) were determined from the log sigmoid-dose response curves generated by the Enzfitter? version 1.05 software (Figure 6.4). Gallic acid was used as a standard to produce the calibration curve for total phenolic content (Figure 6.5). -2 -1 0 1 2 20 40 60 80 100 S. schlechteri S. chamelaeagnea S. radula S. africana-caerulea S. lanceolata log [ Salvia extract] ?g/ml % D ec ol ou ri sa ti on Figure 6.4 Free radical scavenging activity of five Salvia species against the DPPH?. 82 0 50 100 150 200 250 300 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Y = 0.0119 X + 0.0868 r2 = 0.990 Concentration of Gallic acid (mg/ml ) A bs or ba nc e at 7 34 n m Figure 6.5 Calibration curve of gallic acid used to estimate the total phenolic content in solvent extracts, plotted with a 95% confidence interval. To determine the total phenolic content, the mean of two readings was recorded and the quantification expressed in mg/g of gallic acid equivalent dry sample (mg/g of GAE). The correlation between the anti-oxidant activity and total phenolic content was determined using the linear regression equation. Trolox? and vitamin C were used as standard anti- oxidant compounds against the ABTS and DPPH methods, respectively. All results of the anti-oxidant activity were obtained from three independent experiments and given as mean ? standard deviation (s.d.). E xperimental data were analyzed using the Student t-tests. with Statistica version 5.0 and Prism? version 3.0 software. Throughout the analysis, P < 0.05 was considered significant. 6.4 Results 6.4.1 Qualitative evaluation of the anti-oxi dant activity of the solvent extracts The qualitative analysis of the solvent extracts showed that similar compounds are responsible for the anti-oxidant activity when the ABTS?+ or the DPPH? radicals were used (Figure 6.6). Anti-oxidant compounds are indicated by the white spots on the green background (ABTS?+ radical) or purple background (DPPH? radical). The two TLC plates also showed that the anti-oxidant compounds can be polar or less polar. However, the majority of the white spots have relatively low Rf values implying that the anti-oxidant compounds were mostly polar. 83 Sac Sal Sab Sau Sch Sdi Sdo Sga Sla Tro Vit Smu Sna Sra Sre Sru Ssc Sst Sve Figure 6.6 Qualitative determination of the anti-oxidant activity of the solvent extracts: (A): ABTS?+ radical and (B) DPPH? radical: Sac: S. africana-caerulea, Sal: S. africana- lutea, Aab: S. albicaulis, Sau: S. aurita, Sch: S. chamelaeagnea, Sdi: S. disermas, Sdo: S. dolomitica, Sga: S. garipensis, Sla: S. lanceolata, Tro: Trolox? , Vit: Vitamin C, Smu: S. muirii, Sna: S. namaensis, Sra: S. radula, Sre: S. repens, Sru: S. runcinata, Ssc: S. schlechteri, Sst: S. stenophylla, Sve: S. verbenaca. 6.4.2 The quantitative determination of the anti-oxidant activity Several solvent extracts displayed potent activity against both the ABTS?+ and the DPPH? radicals (Table 6.1), while all the essential oils were poorly active and were excluded from Table 6.1. The IC50 values ranged from 11.88 ? 1.52 to 69.26 ? 3.22 ?g/ml and from 1.61 ? 0.03 to 74.15 ? 2.85 ?g/ml against the ABTS and DPPH assays, respectively. A B 84 Table 6.1 The anti-oxidant activity of solvent extracts determined by the ABTS and DPPH assays (mean ? s.d., n = 3) and the tota l phenol content. ABTS assay DPPH assay Species/Controls IC50 values in ?g/ml Total phenol content (mg/g of GAE) S. africana-caerulea 39.72 ? 1.76 33.40 ? 3.73 115.06 S. africana-lutea 30.35 ? 2.97 47.58 ? 2.61 67.83 S. albicaulis 24.22 ? 1.28 19.85 ? 1.01 100.27 S. aurita 22.94 ? 2.57 16.59 ? 0.22 119.09 S. chamelaeagnea* 14.61 ? 2.25 12.75 ? 1.04 211.78 S. disermas 33.05 ? 2.86 55.07 ? 4.01 69.01 S. dolomitica 49.94 ? 1.17 > 100 53.04 S. garipensis 41.66 ? 2.71 74.15 ? 2.85 45.56 S. lanceolata 25.63 ? 0.66 68.09 ? 3.69 54.22 S. muirii* 11.88 ? 1.52 11.06 ? 0.52 186.15 S. namaensis 16.57 ? 2.56 10.63 ? 0.75 190.5 S. radula 69.26 ? 3.22 > 100 55.73 S. repens 18.24 ? 1.87 15.47 ? 1.75 178.17 S. runcinata* 19.42 ? 2.78 19.34 ? 0.61 149.34 S. schlechteri 17.51 ? 2.05 1.61 ? 0.03 209.18 S. stenophylla 20.84 ? 0.86 14.92 ? 0.93 161.28 S. verbenaca 37.47 ? 1.77 45.83 ? 2.78 73.29 Trolox? 2.43 ? 0.07 2.51 ? 0.41 ? Vitamin C ? 4.72 ? 0.14 ? * denotes no significant difference between the ABTS and the DPPH methods ? not determined The solvent extract of S. schlechteri displayed the most favorable activity against the DPPH? radical, which was statistically more active than the reference compound, vitamin C (P < 0.05), while the extract of S. muirii was the most active against the ABTS?+ radical. The extracts of S. radula and S. dolomitica, which were poorly active (IC50 values > 100 ?g/ml) using the DPPH? radical, were moderately active against the ABTS?+ radical (IC50 values: 69.26 ? 3.22 and 49.94 ? 1.17 ?g/ml, respectively) (Figure 6.7; Table 6.1). Although the two methods measure the percentage decolourisation of two radicals, there is 85 a statistically significant difference in IC50 values generated by the ABTS and the DPPH methods (Figure 6.7) (Table 6.1). The ABTS method in general gave lower IC50 values compared to the DPPH method (Table 6.1). However, the Student t-tests did not show significant differences between the IC50 values of S. chamelaeagnea, S. muirii and S. runcinata (P > 0.05). None of the 11 essential oils tested displayed promising anti-oxidant activity with either the ABTS or DPPH methods. All the IC50 values obtained were greater than 100 ?g/ml. S. afr ica na -ca eru lea S. afr ica na -lu tea S. alb ica uli s S. au rit a S. ch am ela ea gn ea S. dis erm as S. do lom itic a S. ga rip en sis S. lan ce ola ta Tr olo x Vi tam in c S. mu iri i s. na ma en sis S. ra du la S. rep en s S. ru nc ina ta S. sch lec hte ri S. ste no ph yll a S. ve rb en ac a 0 10 20 30 40 50 60 70 80 90 100 ABTS assay DPPH assay > 100 > 100 A nt i- ox ida nt a ct iv it y (I C 5 0 va lu es in ? g /m l) Figure 6.7 Comparison between the anti-oxidant activity of the solvent extracts of Salvia species using the ABTS and the DPPH assays: Sac: S. africana-caerulea, Sal: S. africana- lutea, Aab: S. albicaulis, Sau: S. aurita, Sch: S. chamelaeagnea, Sdi: S. disermas, Sdo: S. dolomitica, Sga: S. garipensis, Sla: S. lanceolata, Tro: Trolox? , Vit: Vitamin C, Smu: S. muirii, Sna: S. namaensis, Sra: S. radula, Sre: S. repens, Sru: S. runcinata, Ssc: S. schlechteri, Sst: S. stenophylla, Sve: S. verbenaca. 6.4.3 Total phenolic content and its relationship with the anti-oxidant activity The anti-oxidant activity exhibited by the solvent extracts depends on the type of compounds present. A calibration curve used to estimate the total phenolic content is displayed in Figure 6.5. The phenolic values estimated in the organic extracts of the various species ranged from 45.56 to 211.78 mg/g of GAE dry sample (Table 6.1). Species such as S. chamelaeagnea, S. muirii, S. repens, S. runcinata, S. schlechteri and S. 86 stenophylla with high phenolic content (> 150 mg of GAE) had the most promising anti- oxidant activity (Table 6.1). After inverse transformation of IC50 values obtained (to accommodate the anti-oxidant capacity from low to high values on the X-ax is), there was a stronger correlation between the total phenolic content and the anti-oxidant activity for the DPPH assay compared to the ABTS assay (r2 = 0.90 and 0.76, respectively) (Figure 6.8). This implies that the phenolic compounds are partly, if not totally, responsible for the anti-oxidant activity observed in the species studied. 0.00 0.02 0.04 0.06 0.08 0.10 0 4 0 8 0 1 2 0 1 6 0 2 0 0 2 4 0 r2 = 0.90 DPPH assay 1/IC50 (?g/ml ) T ot al p he no li c co nt en t (m g/ g of G A E ) 0.00 0.02 0.04 0.06 0.08 0.10 0 40 80 120 160 200 240 r2 = 0.76 ABTS assay 1/IC50 (?g/ml ) T ot al p he no li c co nt en t (m g/ g of G A E ) Figure 6.8 Regression analysis of the anti-oxidant activity plotted against the total phenolic content using the DPPH and ABTS methods. 87 6.5 Discussion The majority of solvent extracts exhibited potent anti-oxidant activity against the ABTS?+ and the DPPH? radicals (IC50 values < 75 ?g/ml) with the exception of S. dolomitica and S. radula which showed weak activity in the DPPH assay (Table 6.1). The solvent extract of S. schlechteri was more active than vitamin C. In contrast, all the essential oils were poorly active (IC50 values > 100 ?g/ml) in both assays. The results obtained in the present study are in agreement with previous investigations on Salvia species. S. tomentosa (IC50 value: 18.7 ?g/ml), S. hypageia (IC50 value: 34.6 ?g/ml), S. sclarea (IC50 value: 23.4 ?g/ml) and S. caepitosa (IC50 value: 41.3 ?g/ml) have been reported to display anti-oxidant activity against the DPPH? radical (Tepe et al., 2005, Tepe et al., 2006). The anti-oxidant activity of S. officinalis has also been reported (Cuvelier et al., 1994). However, as far as the literature survey has ascertained, there are no published reports on the anti-oxidant activity of southern African Salvia species. Salvia dolomitica and S. radula demonstrated weak anti-oxidant activity against the DPPH? radical and moderate activity against the ABTS?+ radical. The differences obtained between the two methods indicated that the DPPH assay recorded lower anti-oxidant activity as compared to the ABTS assay. A similar incongruence has been reported previously between the two methods (Kim et al., 2002) and these differences may be attributed to absorbance interference at 550 nm by other compounds in the DPPH assay. In addition, the ABTS method is very sensitive, requiring a shorter reaction time (four minutes) compared to 30 minutes for the DPPH method. It can be used in both organic and aqueous solvent systems (Lee et al., 2003). Therefore, the ABTS assay may be preferable to the DPPH method for evaluating the total anti-oxidant activity of a given sample (Lee et al., 2003). The total phenolic content (Table 6.1) found in the species was high and varied from 45.56 to 211.78 mg/g of GAE dry sample and a strong correlation using the DPPH method was obtained between the anti-oxidant activity and the total phenolic content (Figure 6.6). It is well known that Salvia species contain various phenolic compounds including flavonoids, rosmarinic acid, caffeic acid, carnosol and carnosic acid (Lu and Foo, 2002; Tepe et al., 2006). The anti-oxidant activity of sage has been attributed to the phenolic compounds (Cuvelier et al., 1994, Cuvelier et al., 1996; Hohmann et al., 1999). Tepe et al. (2005) demonstrated that S. tomentosa, which exhibited good anti-oxidant activity, also possessed 88 a high phenolic content (200 ?g/mg). Several reports have conclusively shown a close relationship between total phenolic content and anti-oxidant activity (Velioglu et al., 1998; Deighton et al., 2000). In the present study it was found that S. chamelaeagnea and S. schlechteri contained a substantial amount of phenolics, that are at least partly, if not entirely, responsible for the anti-oxidant activity (Table 6.1). As can be observed from the present data, anti-oxidant activity does not necessarily correlate with high amounts of phenolics as seen by the moderate correlation between the anti-oxidant activity using the ABTS method and the total phenolic content (Figure 6.8) (K?hk?nen et al., 1999). The primary anti-oxidant activity of rosemary and sage is related to the presence of carnosic acid (Wenkert et al., 1964). It was also reported that carnosic acid and carnosol are responsible for 90% of anti-oxidant activity of rosemary (Cuvelier et al., 1996). Caffeic acid, rosmarinic acid, ursolic acid, oleanolic acid, carnosic acid and carnosol are common in many plants and all are strong radical scavengers. These compounds and derivatives have been shown to display anti-oxidant activity comparable to the reference compound vitamin C. Carnosol (isolated from S. chamelaeagnea, Chapter 9), standards of caffeic acid and rosmarinic acid also exhibited good anti-oxidant activity using the DPPH assay with the IC50 values 6.10 ? 0.6; 2.55 ? 0.17, 2.48 ? 0.20 ?g/ml, respectively. The HPLC data (Chapter 5) revealed the presence of rosmarinic acid, caffeic acid carnosol and derivatives (carnosic acid) in the majority of the extracts. For example, S. aurita, S. chamelaeagnea, S. muirii, S. repens, S. runcinata and S. schlechteri with the presence of caffeic acid, carnosol, rosmarinic acid and/or carnosic acid exhibited good anti-oxidant activity, while those lacking these compounds were poorly active (e.g. S. dolomitica, S. radula and S. verbenaca). These compounds might therefore account for the anti-oxidant activity obtained in most investigated species. Although the chemical composition and structures of active extract components are important factors governing the efficacy of natural anti-oxidants, the anti-oxidant activity of an extract cannot be explained based only on their phenolic content without further characterization (Heinonen et al., 1998). Studies have indicated that the anti-oxidant activity of phenolic acids is related to their structure. It has been reported that phenolic compounds with ortho- and para-dihydroxylation or a hydroxy and a methoxy group are more effective than simple phenolics (Frankel et al., 1995). 89 It has been reported that polar extracts exhibited stronger activity than non-polar extracts (Duan et al., 1998; Tepe et al., 2005; Tepe et al., 2006). For instance, if an electron donating group, especially a hydroxyl group, is located on the o- or p- positions of the phenolic compound, it makes the compound polar and therefore anti-oxidant activity is increased (Duan et al., 1998). The strong anti-oxidant activity of some Salvia species could be attributed to these groups that enable phenols to donate hydrogen atoms easily, thereby disrupting the chain-reaction of auto-oxidation. Weng and Wang (2000) showed that the anti-oxidant activity of S. plebeia was concentrated in the acidic, weakly acidic and phenolic fractions of the extract, while the activity of neutral fractions was negligible. DPPH? and ABTS?+ radical scavenging activity of the essential oils was very low (IC50 values > 100 ?g/ml) and this was obviously related to their chemical composition. In addition, essential oils have a low phenolic content (Cuvelier et al., 1994, Cuvelier et al., 1996). Essential oils are generally non-polar and many studies have shown that the anti- oxidant activity is associated with polar fractions (Tepe et al., 2005; Tepe et al., 2006). Other members of the Lamiaceae also showed poor activity against the DPPH? radical. For instance, ?ahin et al. (2006) demonstrated that Origanum vulgare spp. vulgare was poorly active against the DPPH? radical (IC50 value > 8900 ?g/ml). In contrast, Tepe et al. (2004) demonstrated that S. multicaulis and S. crypthantha essential oils were more active than the reference anti-oxidant compounds such as vitamin C and butylated hydroxy toluene (IC50 values of 2.4 and 3.9 ?g/ml, respectively). The GC-MS data (Chapter 4) indicated that the essential oils were mostly devoid of constituents generally displaying anti-oxidant activity such as eugenol, thymol, carvacrol and ?-terpinene (Baratta et al., 1998; Ruberto and Baratta, 2000, van Zyl et al., 2006). However, these individual components of the oils were generally tested at only one concentration (1000 ?g/ml). The essential oils found in the current study may also reveal some degree of activity at 1000 ?g/ml, but this value is above the cut-off used to evaluate the anti-oxidant activity of samples in this study. It was found that sesquiterpene hydrocarbons exerted a low, if any, anti-oxidant effect, while monoterpene hydrocarbons showed a significant protective effect with several variants due to the functional group (Ruberto and Baratta, 2000; Dorman et al., 2000). The quantitative difference in their amount, as well as the antagonism in the mixture of essential oils may also explain why the activity was low. Furthermore, there are various methods for anti- oxidant activity determination of a given sample such as linoleic acid peroxidation, ferric thiocynate and thiobarbituric acid methods and perhaps the essential oils may display 90 greater activities using another method. The studies reported here provide initial results on the anti-oxidant properties of South African indigenous Salvia species. In order to investigate whether good anti-oxidant extracts can also exhibit good anti-inflammatory properties, the anti-inflammatory activity of Salvia species is discussed in the next Chapter. 6.6 Conclusions ? Indigenous Salvia species are a good source of anti-oxidants as they are rich in phenolic compounds. ? The DPPH? radical was less sensitive than the ABTS?+ radical. ? Similar compounds are responsible for the anti-oxidant activity of the solvent extracts against the DPPH? and ABTS?+ radicals; however the degree of activity varies significantly for some species ? S. schlechteri and S. muirii were the most active solvent extracts against the DPPH? and ABTS?+ radicals, respectively. ? The activity of the solvent extracts of S. schlechteri was 3 times greater than that of the reference compound, vitamin C. ? Although a correlation between the anti-oxidant activity using the DPPH method and phenolic content has been confirmed, however the anti-oxidant activity is not always highly correlated to the total phenolic content. ? Essential oils have a poor free radical scavenging capacity against the DPPH? and the ABTS?+ radicals. 91 Chapter 7: Anti-inflammatory activity-Inhibition of the 5- Lipoxygenase Enzyme Abstract 5-Lipoxygenase is one of the key enzymes involved in the inflammation process. The anti- inflammatory activity of essential oils and solvent extracts of Salvia species was evaluated using the 5-lipoxygenase assay. Essential oils exhibited better anti-inflammatory activity when compared to the solvent extracts. Almost all the essential oils exhibited promising anti-inflammatory activity with IC50 values ranging between 22.81 and 77.32 ?g/ml. The essential oil of S. runcinata exhibited the most favourable activity, while S. africana-lutea displayed the lowest activity. With the exception of S. radula, solvent extracts displayed poor ability to inhibit the enzyme with all IC50 values being greater than 100 ?g/ml. The components of the oils such as ?-caryophyllene, trans-nerolidol, ?-bisabolol, ?-pinene, ?- limonene, linalool and its ester linalyl acetate, may be responsible for the anti- inflammatory activity of the investigated essential oils. 92 7.1 Introduction Inflammation is the normal physiological response of the body that occurs when it is invaded by infective agents, responding to stimulation or when it is exposed to physical, chemical or traumatic changes (Baylac and Racine, 2003). Inflammatory reactions may be initiated by a variety of mediators including eicosanoids derived from the cyclo-oxygenase and 5-lipoxygenase pathways (Figure 7.1). The 5-lipoxygenase enzyme is the initiator of the conversion of the membrane phospholipid-derived fatty acid, arachidonic acid, into a number of compounds such as leukotrienes, which cause redness, pain and irritation. Figure 7.1 Inflammation process pathways (Davis et al., 2004). Leukotrienes form a group of highly potent molecules that mediate inflammatory reactions contributing to various diseases such as asthma, cystic fibrosis, urticaria and central nervous system disorders. This important class of messengers is pro-inflammatory, resulting from the metabolism of arachidonic acid and which begins with oxidation by the 93 5-lipoxygenase enzyme. The biological effects of leukotrienes can be prevented by suppressing leukotriene production via inhibition of the lipoxygenase pathway (Figure 7.1). In recent years, a large body of evidence has accumulated, demonstrating that free radicals are important components of inflammation (Kehrer, 1993) and anti-oxidant compounds have been found to display anti-inflammatory activity (Sala et al., 2003). Compounds such as carnosic acid and ursolic acid which have potent anti-oxidant activity, have also been shown to possess anti-inflammatory activity (Baricevic et al., 2001). As listed in Table 1.4 and 1.5, plants belonging to the genus Salvia have been used for various ailments and are reported to have anti-inflammatory properties. Studies have shown that the solvent extracts or isolated compounds of Salvia also displayed anti- inflammatory activity both in vitro and in vivo (Baricevic et al., 2001). Essential oils are low molecular weight compounds which can rapidly diffuse through the skin barrier and reach the underlying tissue. They are used in many fields, such as food products, pharmaceuticals, cosmetics, perfumes and household items. Furthermore, a number of essential oils are used in aromatherapy for their therapeutic benefits (Pauli, 2001; Baylac and Racine, 2003). Previous studies carried out on Salvia stenophylla reported the anti-inflammatory properties of its essential oil (Baylac and Racine, 2003). Nevertheless, South African species have not yet been fully investigated. Together with the traditional uses of sage to treat body wounds (Watt and Breyer-Brandwijk, 1962), the potent anti-oxidant capacity and the reported anti-inflammatory activity of Salvia prompted this investigation on the anti-inflammatory activity of indigenous species. The objective of this study was to evaluate the effectiveness of the essential oils and solvent extracts to inhibit the in vitro activity of the 5-lipoxygenase enzyme. 7.2 Materials and methods 7.2.1 Chemicals, reagents and standard 5-Lipoxygenase and nordihydroguaiaretic acid (NDGA) were obtained from Cayman Chemicals Company; DMSO, potassium phosphate and hydroxyl sodium from Saarchem, while Tween? 20 and linoleic acid were purchased from Fluka. Double distilled Millipore ? (0.22 ?m) water was used throughout this experiment. 94 7.2.2 Preparation of solutions, enzymes and buffers The potassium phosphate buffer (pH 6.3) was prepared by mixing 50 ml of 0.1 M potassium dihydrogen phosphate and 9.7 ml of 0.1 M sodium hydroxyde. The solution of linoleic acid was prepared at a concentration of 0.168% in ethanol (w/v). The enzyme was aliquoted in eppendorfs (100 U ? 12 ?l) and kept at -80 ?C until used. 7.2.3 Preparation of test samples The stock solution of the test samples (essential oils and solvent extracts) was prepared at a concentration of 10 mg/ml in the DMSO and Tween? 20 mixture and, if necessary, further dilutions were made with the same mixture. The solution of DMSO and Tween? 20 mixture was prepared by weighing 1.5 g of Tween? 20 and made up to 55 g with DMSO. 7.2.4 The 5-lipoxygenase assay The 5-lipoxygenase assay was used as an indication of the anti-inflammatory activity. An aliquot of 50 ?l of the stock solution (10 mg/ml) of each test sample was placed in a 3 ml cuvette, followed by 2.95 ml of pre-warmed potassium phosphate buffer and 48 ?l of linoleic acid solution. The mixture was covered with a cap and gently shaken six times before the cuvette was placed back in the UV-VIS spectrophotometer in order to measure the absorbance reference at 234 nm. Thereafter, 12 ?l of the ice-cold buffer (potassium phosphate) was mixed with 100 U of the thawed enzyme (for 10 seconds to avoid its degradation) in order to activate the enzyme. The mixture was then transferred to the cuvette and the content of the cuvette was shaken and placed back into the spectrophotometer (Specord Analytikjena-40 connected to Winaspect? software), before the absorbance was recorded at 234 nm. It is important to note that, prior to testing samples; two samples were prepared as mentioned above but only with DMSO and Tween? 20 mixture to serve as controls (no enzyme inhibition). 5-Lipoxygenase is known to catalyze oxidation of unsaturated fatty acids containing 1-4 diene (Alitonou et al., 2006) and the modification of linoleic acid (diene 1-4 into diene 1- 3) can be detected at 234 nm. 95 7.3 Data analysis An increase in absorbance at 234 nm arising from modification of the linoleic acid (diene 1-4 into 1-3) was measured for 10 minutes at room temperature. The initial reaction determined from the slope of the linear portion (over 60 seconds) of the curve (Figure 7.2) and the percentage activity of the enzyme was calculated by comparison to the controls. The slope of the control represents 100% activity of the enzyme. Figure 7.2 Plot of the absorbance, determined over 10 minutes, of essential oil of Salvia runcinata at various concentrations. For samples exhibiting activity at 100 ?g/ml, three 2-fold serial dilutions were prepared and each test sample was run once at each concentration. The concentration inhibiting 50% enzyme activity (IC50) was used to evaluate the test sample and the IC50 value was obtained by plotting the percentage activity against the log concentration using the Enzfitter? version 1.05 software. NDGA was used as the reference anti-inflammatory compound. 7.4 Results Results obtained for inhibition of the 5-lipoxygenase enzyme by the solvent extracts and essential oils are presented in Table 7.1. Essential oils displayed favourable activity when compared to the solvent extracts. Almost all the essential oils inhibited the 5-lipoxygenase enzyme with the IC50 values ranging from 22.81 ? 0.01 to 77.32 ? 6.27 ?g/ml in comparison to the reference compound NDGA (IC50 value: 4.95 ? 0.07 ?g/ml). 96 Table 7.1 Anti-inflammatory activity (IC50 ? s.e) of the essential oils and solvent extracts. Essential oils Solvent extracts Species / Control IC50 values in ?g/ml S. africana-caerulea 73.42 ? 5.27 > 100 S. africana-lutea 77.32 ? 6.27 > 100 S. albicaulis 39.22 ? 3.93 > 100 S. aurita nd > 100 S. chamelaeagnea 48.58 ? 0.07 > 100 S. disermas nd > 100 S. dolomitica 64.58 ? 6.47 > 100 S. garipensis nd > 100 S. lanceolata 43.38 ? 3.52 > 100 S. muirii > 100 > 100 S. namaensis nd > 100 S. radula > 100 78.78 ? 5.89 S. repens 28.04 ? 4.50* > 100 S. runcinata 22.81 ? 0.01 > 100 S. schlechteri nd > 100 S. stenophylla 49.17 ? 4.87 > 100 S. verbenaca nd > 100 NDGA 4.95 ? 0.07 nd = not determined, essential oil not available * Essential oil was obtained from Lady Grey whic h is different from where extract was collected NDGA: Nordihydroguaiaretic acid s.e.: standard error of the estimate The essential oil of S. runcinata exhibited the most potent activity (IC50 value: 22.81 ? 0.01 ?g/ml), while S. africana-lutea displayed the lowest activity (IC50 value: 77.32 ? 6.27 ?g/ml). The essential oils of S. muirii and S. radula displayed poor activity at a starting concentration of 100 ?g/ml. Nearly all the solvent extracts exhibited poor anti-inflammatory activity except S. radula which displayed moderate activity (IC50 value: 78.78 ? 5.89 ?g/ml). The IC 50 values of the solvent extracts were greater than 100 ?g/ml against the 5-lipoxygenase enzyme. It was 97 interesting to note that in contrast to poor activity of the essential oil of S. radula, the solvent extract was fractionally more potent, whilst S. muirii was found to possess no anti- inflammatory properties in either essential oil or solvent extract (Table 7.1). 7.5 Discussion Baylac and Racine (2003) found that the essential oil of S. stenophylla has the potential to inhibit the in vitro 5-lipoxygenase enzyme (10 < IC50 value < 30 in ?g/ml). In this study, S. stenophylla also displayed anti-inflammatory activity (IC50 value: 49 ?g/ml). The difference found between the two values reported, although the same method was used, may be attributed to variation in chemical composition of the two essential oils. Many factors such as harvesting time, geographical location, climate and stage of development of the plant have been reported to affect the chemical composition of the essential oil (Putievsky et al., 1986). The biological activity of essential oils or plant extracts is mainly associated with their chemical composition. The essential oil data given in Chapter 4 can be used to explain the results obtained in this study. Essential oils are complex mixtures comprising of a multitude of compounds and each constituent contributes to the beneficial or adverse effect of the oils (Lahlou, 2004). It is therefore difficult to determine exactly which compound(s) is/are responsible for the biological activity when crude essential oils are used. Aromatherapy reports on the anti-inflammatory activity of terpenes and sesquiterpenes, and data suggests that the presence of these compounds may account for the activity of the essential oils as determined by the 5-lipoxygenase in vitro enzyme. Sesquiterpenoids and their derivatives are credited with various biological action including anti-asthmatic, antibacterial, antifungal, anti-inflammatory and antineoplastic activities (Farnworth and Bingel, 1977). Structure-activity relationships to define broad categories of activities for the different classes of molecules found naturally in essential oils have been investigated. Terpenic hydrocarbons (?-pinene, 31 < IC50 value < 50 ?g/ml), sesquiterpenic hydrocarbons (?-caryophyllene, 10 < IC50 value < 30 ?g/ml), sesquiterpenic alcohols (trans-nerolidol, 10 < IC50 value < 30 ?g/ml) and ?-bisabolol, (10 < IC50 value < 30 ?g/ml) exhibited a potential ability to inhibit the 5-lipoxygenase enzyme (Baylac and Racine 2003). Peana et al. (2002 ) demonstrated that linalool and its ester linalyl acetate (Table 98 4.1) exhibited anti-inflammatory activity. These two compounds were found in S. dolomitica (16.6 and 19.6%, respectively) and may have partly contributed to the anti- inflammatory activity. It was reported that monoterpenes such as 1,8-cineole, ?-pinene and limonene and sesquiterpenes such as ?-caryophyllene also inhibit the in vitro 5- lipoxygenase assay (Juergens et al., 1998; Baylac and Racine, 2003). These compounds were present in the majority of species in varying concentrations (1,8-cineole: trace amounts to 40.5 %; ?-pinene : trace amounts to 22.3 % and limonene: 0.5 to 11.6 % )(Table 4.1). All these components identified in the essential oils may have contributed to the anti-inflammatory activity of the essential oils. Furthermore, ?-bisabolol, which is widely reported to have a skin soothing action, strongly inhibited the 5-lipoxygenase enzyme in vitro (Baylac and Racine, 2003). It is therefore expected that essential oils containing high levels of this sesquiterpenic alcohol like that of S. runcinata (65%) and S. stenophylla (26%), can be expected to inhibit 5-lipoxygenase enzyme as well. Moretti et al. (1997) evaluated the anti-inflammatory activity (induced by carrageenin in rats) of essential oils of S. sclarea and found that the oil exhibited a significant anti-inflammatory effect. The effect of the oil was correlated to the presence of linalool, ?-terpineol and linalyl acetate. They also found that these constituents produce less anti-inflammatory action when administered separately than the oil in toto and are less active than the oxygenated fractions obtained by flash chromatography of the oil. This clearly indicated that the action of the essential oil might be determined by the synergistic interactions of its constituents. Effective anti-oxidant treatment has also been found to be effective against inflammation (Sala et al., 2003). For instance, phenolic acids such as carnosic and ursolic acids, which are good anti-oxidants, were reported to be good anti-inflammatory compounds (Liu, 1995; Baricevic et al., 2001). However, ursolic acid exhibited poor ability to inhibit the key enzyme of leukotriene biosynthesis, namely 5-lipoxygenase (Safayhi et al., 1997). Only the solvent extract of S. radula exhibited some degree of activity, while other species were inactive at the starting concentration of 100 ?g/ml (Table 7.1). In addition, the anti- inflammatory activity of the solvent extracts or isolated compounds of Salvia species is widely reported. For example, Baricevic et al., (2001) investigated the topical anti- inflammatory activity of S. officinalis leaves and found that the solvent extracts inhibited ear oedema in mice in a dose-dependent manner, while aethiopinone, isolated from S. aethiopis roots, exhibited anti-inflammatory activity (Hern?ndez-P?rez et al., 1995). 99 Solvent extracts of Salvia species are rich in flavonoids which possess potent anti-oxidant activity. Although many pentacyclic triterpenes (ursolic acid) also bind to the 5- lipoxygenase enzyme (Safayhi et al., 1995), the presence of an 11-keto-group and a hydrophilic function of the pentacyclic ring system are crucial for potent inhibition of 5- lipoxygenase and ursolic acid turned out to be non-inhibitory (Sailer et al., 1996). This compound was found in some of the solvent extracts. This suggests that the components of the extract responsible for the anti-oxidant activity are not necessarily those responsible for the anti-inflammatory activity. It is important to note that the ?quasi absence? of 5- lipoxygenase inhibitory activity in this study does not imply that the solvent extracts do not possess any anti-inflammatory activity. The activity may be observed at higher concentrations. Furthermore, the biological response of any substance may be influenced by various factors, including variation in assay methods employed (Houghton and Raman, 1998). As shown in Figure 7.1, inflammation is a very complex cascade of events involving not only the 5-lipoxygenase pathway, but also the 12- and 15-lipoxygenases, as well as cyclo-oxygenase-1, cyclo-oxygenase-2, cytochrome P450 and epoxygenase pathways. Further investigation involving other inflammatory routes is required in order to determine whether the solvent extracts do exert anti-inflammatory activity in other steps of the complex anti-inflammatory cascade. 7.6 Conclusions ? The essential oils exhibited the ability to inhibit the in vitro 5-lipoxygenase enzyme. ? The solvent extracts were poorly active and only S. radula inhibited the in vitro 5- lipoxygenase enzyme below a concentration of 100 ?g/ml. ? The uses of Salvia species by inhalation support the traditional uses of these plants and imply that essential oils may be responsible for the activity. Furthermore, the topical use of Salvia species may be validated. ? None of the solvent extracts or essential oils exhibited anti-inflammatory activity comparable to nordihydroguaiaretic acid. ? The possibility that the solvent extracts may inhibit other enzymes or receptors involved in the inflammatory pathway should not be disregarded and this warrants further investigation. 100 Chapter 8: The Antimalarial Activity of the Essential Oils and Solvent Extracts and Isolated Compounds from S. radula Abstract The resistance of Plasmodium falciparum to currently used drugs has become a serious problem and efforts are being directed at obtaining new drugs with different structural features. One option favoured is the search for new plant-derived antimalarial drugs. Eleven essential oils and 17 solvent extracts of indigenous Salvia species used in traditional medicine against various ailments, including malaria, were subjected to pharmacological testing in order to evaluate their potential to inhibit the growth of Plasmodium falciparum FCR-3 strain. The investigation was conducted using the [3H]- hypoxanthine radiometric method. The outcome showed that both the essential oils and the solvent extracts displayed antimalarial activity. The IC50 values of the essential oils ranged from 1.20 to 13.50 ?g/ml and were low compared to the solvent extracts (IC50 values ranging from 3.91 to 26.01 ?g/ml). The essential oil of S. runcinata was the most active killing 50% of parasites at a concentration of 1.20 ?g/ml. Among the plants screened, the solvent extract of S. radula displayed the most favorable activity, while S. lanceolata was the least active. The extract of S. radula was selected to determine the compound(s) which may be responsible for the observed antimalarial activity since it displayed the best activity. Two compounds were isolated and characterised as betulafolientriol oxide and salvigenin. The two compounds displayed comparable or lower antimalarial activity (IC50 values: 4.95 and 24.60 ?g/ml, respectively) than the crude solvent extract. The in vitro antimalarial activity supports the continued use of Salvia species in folk medicine. 101 8.1 Introduction Not considering the future implications of the AIDS pandemic, malaria is currently one of the most important human diseases in developing countries and is still unconquered. It is the world?s leading cause of fatalities among the infectious diseases (Wanyoike et al., 2004). Most of the lethal cases are caused by Plasmodium falciparum, the most virulent of the four Plasmodium species that infect humans and which is distributed in tropical Africa, Asia and Latin America (Karou et al., 2003). Despite the extensive control efforts, the incidence of the disease is not decreasing, especially in developing countries where malaria remains a major public health problem (Karou et al., 2003). Some 3.2 billion people live in areas at risk of malaria transmission (WHO, 2005a). An estimated 350-500 million clinical malaria episodes occur annually. Plasmodium falciparum induced malaria causes more than 1 million deaths each year (WHO, 2005a). The spread of parasites resistant to agents such as chloroquine and quinine, and insecticide- resistant mosquitoes, has led to major difficulties in the treatment and control of this disease and has hence triggered the search for new drugs. One possible approach to the identification of new antimalarial drug candidates is to search for compounds or plants that reportedly cure malaria. For many centuries, plants have formed the basis of sophisticated traditional medicine systems and more recently, natural products have yielded lead compounds, especially against infectious diseases (Schwikkard and van Heerden, 2002). In South Africa and in most Third World countries, the use of medicinal plants is widespread and there is real hope of finding an alternative to synthetic drugs. Plants are known to be the main source of drug therapy in traditional medicine (Prozesky et al., 2001) and the use of traditional medicinal plant preparations is transmitted verbally from one generation to the next, a process than runs the risk of losing essential information or misinformation. The recognition and validation of traditional medical practices and the search for plant-derived drugs could lead to a new strategy in malaria control. It is therefore necessary to obtain more scientific information concerning the efficacy and safety of the remedies in use, as many people already use and depend on herbal medicines for the treatment of malaria (Prozesky et al., 2001). Many studies have been carried out on the antimalarial activity of various plant species with some encouraging results (Prozesky et al., 2001; van Zyl and Viljoen, 2002). Furthermore, two of the most potent antimalarial 102 drugs have originated from plants: quinine from Cinchona trees and artemisinin from Artemisia annua (Gessler et al., 1994; Foley and Tiller, 1997; Olliaro et al., 2001). Nevertheless, only 20% of plants which are reported to have bioactivity have been subjected to bioassay screening (Houghton, 2001). Malaria can occur despite taking antimalarial drugs and symptoms of infection usually occur within 9 to 14 days after exposure. The general symptoms of malaria include headache, nausea, fever, cough, sweating, vomiting and flu-like symptoms. These symptoms may differ depending on the type of Plasmodium that is responsible for the infection (WHO, 1998). The life cycle of a malaria parasite is presented in Figure 8.1. Figure 8.1 Life cycle of the malaria parasite (National Center of Infectious Disease, 2006). The genus Salvia has been used in traditional medicine for the treatment of a variety of diseases, including malaria (Watt and Breyer-Brandwijk, 1962). However, no exhaustive studies on the in vitro antimalarial activity of these plants have been carried out on indigenous species thus far. In a continuation to verify the efficacy of traditional medicines against malaria, South African Salvia species were investigated. 103 The objectives of this study were to: (i) study the essential oils and solvent extracts of the Salvia species for their potential to inhibit the in vitro growth of the malaria parasite, P. falciparum, (ii) isolate, identify, characterize and determine the antimalarial activity of isolated compound(s) from the most promising species, and (iii) generate reliable scientific data to validate the traditional use of these plants in the treatment of malaria. 8.2 Materials and methods 8.2.1 Chemicals, reagents and standards Roswell Park Memorial Institute media (RPMI 1640) was purchased from GibcoTM; quinine sulphate and the reagent anisaldehyde were obtained from Fluka. Gentamicin sulphate, hypoxantine (6-hydroxypurine), NaHCO3 and the standard chloroquine diphospate were purchased from Sigma? , while KH2PO4, Na2HPO4.2H2O and glucose were obtained from Merck. [N-(2-hydroxyethyl) piperazine-N'-(2-e thanesulfonic acid)] (HEPES) was obtained from Highveld Biological, while DMSO and D-sorbitol were obtained from Saarchem. Sterile water was used throughout the experiments. The [3H]- hypoxanthine isotope was purchased from Amersham. All other solvents and/or reagents were of the best available grade. Silica gel was obtained from Macherey-Nagel. 8.2.2 Preparation of different solutions 8.2.2.1 Incomplete hypoxanthine negative medium and the incomplete hypoxanthine culture medium The incomplete hypoxanthine negative medium was prepared by dissolving RPMI-1640 (10.4 g), HEPES buffer (5.9 g) and glucose (4 g) in one liter of distilled Millipore? water which was filtered through a Sterivex-GS 0.22 ?m filter unit to sterilize the solution. The incomplete hypoxanthine culture medium was prepared in the same way as the hypoxanthine negative medium except that hypoxanthine (44 mg) and gentamicin (100 ?l; 50 mg/ml) were included. Both solutions were stored at 4 ?C until required. 104 8.2.2.2 Sodium bicarbonate and phosphate buffer saline The 5% (w/v) sodium bicarbonate was prepared by dissolving 50 g NaHCO3 into one liter of distilled Millipore ? water and filtered. The phosphate buffer saline (PBS) (pH 7.4) was prepared with NaCl (8 g), KCl (0.3 g), Na2HPO4.2H2O (0.73 g) and KH2PO4 (0.2 g) in one liter of distilled Millipore? water which was then autoclaved for 20 min at 120 ?C and at 1.5 kgf/cm 2 pressure. These solutions were stored at 4 ?C until required. 8.2.2.3 Complete culture medium The complete culture medium (CCM) was prepared by supplementing the incomplete hypoxanthine negative medium (85.8 ml) with 10% (v/v) human plasma (10 ml) and 4.2 ml of 5% NaHCO3. The CCM was prepared when required and discarded when it had turned dark pink which indicated that it was too alkaline for use. 8.2.2.4 Preparation of human plasma and blood The plasma pooled from healthy donors (South African Blood Transfusion Services) was inactivated at 56 ?C in a water bath for two hours, centrifuged at 3000 rpm for 20 minutes, aliquoted into sterile 50 ml centrifuge tubes and stored at -20 ?C until required. The whole blood collected from healthy donors (South African Blood Transfusion Service) was preserved in citrate phosphate dextrose adenosine-1 (CPDA-1) to prevent coagulation and, when required, was washed thrice with PBS at 2000 rpm for five minutes each. The buffy coat was removed with each wash and the remaining erythrocytes stored in hypoxanthine negative medium at 4 ?C until required. 8.2.2.5 D-sorbitol and Giesmsa buffer The D-sorbitol was made at a concentration of 5% (w/v) with distilled Millipore? water, filtered and stored at 4 ?C. The Giesmsa buffer was prepared by dissolving 3.5 g of KH2PO4 and 14.5 g of Na2HPO4.12H2O in a final volume of one liter of distilled Millipore ? water and autoclaved for 20 min at 120 ?C and at 1.5 kgf/cm 2 pressure. 105 8.2.3 Plasmodium falciparum and in vitro culture maintenance The chloroquine-resistant Gambian FCR-3 strain was cultured continuously according to the procedure described by Trager and Jensen (1976) taking into account the modifications as described by van Zyl and Viljoen (2002). The parasite culture was suspended in complete culture medium and maintained daily. The culture consisted of at least a 2-6% parasitaemia in a 5% haematocrit of parasitised and uninfected erythrocytes in CCM (Section 8.2.2.3) which was gassed with 2.06% O2, 5.00% CO2, 92.94% N2 and incubated at 37 ?C. Fresh uninfected human erythrocytes (less than 3 weeks old) were added when the parasites were predominantly in the trophozoite/schizoite stages. In order to synchronise the parasites for experimental purposes, 5% (w/v) D-sorbitol was added to a predominantly ring-staged (Figure 8.2) the culture every second day (10-15 minutes) and then washed with PBS three times at 2000 rpm five minutes each. Daily visualisation of parasites was done by preparing a thin smear and the observation was made under oil immersion objective with a light microscope (Nikon). The percentage parasitaemia was then calculated using at least 10 fields counted over the length of the smear (Figure 8.2). Figure 8.2 A thick blood smear showing the ring form of P. falciparum in erythrocytes. 8.2.4 Preparation of test samples Stock solutions of the solvent extracts (10 mg/ml) were prepared with DMSO, aliquoted and stored at -20 ?C until requi red. On the day of experimentation, seven dilutions of each test sample were prepared from a stock solution made with an incomplete hypoxanthine negative medium. The stock solution (10 mg/ml) and dilutions of the essential oils were 106 prepared on the day of the experiment with DMSO. A final concentration of 1% DMSO in the well had no effect on parasite growth. 8.2.5 Assay of antimalarial activity The antimalarial activity was determined using the tritiated hypoxanthine radiometric assay over a single growth cycle (48 hours) (Desjardins et al., 1979; van Zyl and Viljoen, 2002). Before running the experiment, a thin blood film stained with diluted Giesmsa for 10-20 minutes was microscopically examined (Figure 8.2) and the percentage parasitaemia calculated (Eq. 8.1). At least 10 fields were counted over the length of the smear to obtain an average percentage parasitaemia. The percentage parasitaemia normally ranged from 2 to 6%. To prepare a suspension to be used for the in vitro experiment, the parasite suspension, mainly in the ring stage, was adjusted to 0.5% parasitaemia and 1% haematocrit using an incomplete hypoxanthine negative medium supplemented with 10% human plasma and 5% NaHCO3. % Parasitaemia = Number of infected cells x 100 (Eq. 8.1) Total number of infected + uninfected cells For each dilution, 25 ?l of solvent extract and isolated compounds, or 2 ?l of essential oil, was plated out in triplicate in a 96-well culture plate (NunclonTM). In the control wells, 25 ?l of incomplete hypoxanthine negative medium was plated out. The hypoxanthine negative medium (23 ?l) was added to the wells containing essential oils to ensure a total volume of 25 ?l in each well. Subsequently, the complete hypoxanthine negative medium containing 1% red blood cells (200 ?l) was plated out in four wells and used as a control. While 200 ?l of the parasitised red blood cells were added to the remaining 92 wells (each plate consists of four test samples and each test sample had seven dilutions). The plate was placed in a humidified candle jar, relatively free of oxygen and incubated for 24 hours at 37 ?C. The essential oils were always assaye d in separate microtitre plates and incubated in different candle jars due to their volatile properties. The [3H]-hypoxanthine isotope (25 ?l) was then added to each well and the plate was incubated at 37 ?C for an additional 24 hours. The final concentration of the solvent extracts in the wells ranged from 1 to 100 ?g/ml, while the essential oil concentrations ranged from 0.05 to 50 ?g/ml. Further dilutions were prepared where necessary. The 107 [3H]-hypoxanthine was prepared by mixing 10 ?l of [3H]-hypoxanthine (5 mCi, 28.0 Ci/mmol) with 2.7 ml of incomplete hypoxanthine negative medium. At the end of the second incubation period, parasite DNA was harvested onto Wallac? printed glass fiber filter mats with a 96-well semi-automated Titertek? cell harvester. The glass-fiber filter mats were left to dry overnight at room temperature and transferred to a sample bag containing 10 ml of Wallac Betaplate? liquid scintillation fluid and heat- sealed. The in vitro uptake of [3H]-hypoxanthine by malaria parasites was used as an indicator of parasite growth. Hypoxanthine is required by parasites for DNA synthesis and is only supplied by the human host, as it is not produced de novo. The amount of radioactivity in counts per min (CPM) associated with the [3H]-hypoxanthine incorporated into the parasite DNA trapped in the mat was determined with a ?-scintillation counter (Wallac? ). At least three independent experiments were carried out for each sample, with chloroquine diphosphate and quinine sulphate used as reference antimalarial drugs. 8.2.6 Isolation and structural characterization procedures from S. radula S. radula was selected to isolate the compounds which may be responsible for the antimalarial activity because it showed the best activity compared to the other tested species. Dried aerial parts (723.5 g) collected on the road to Derby (North West) were powdered and exhaustively extracted with methanol:chloroform (1:1, v/v). The supernatant was then concentrated at 70 ?C with a rotary evaporator under vacuum. 8.2.6.1 Isolation by column chromatography The residue obtained (22.4 g) was mixed with a small amount of silica 60 (0.063-0.2 mm/70-230 mesh ASTM), dried at 60 ?C in the oven, reduced to fine powder, added to the top of the column (4 cm x 60 cm) and eluted with hexane:dichloromethane (9:1), dichloromethane:methanol (6:1) and finally with 100% methanol, respectively. Five fractions (A-E) were collected and each fraction was tested for antimalarial activity. Fraction C (12.5 g) displayed the best activity and was selected for further analysis. It was re-chromatographed in a smaller column (2.5 cm x 45 cm) with a solvent system of toluene:ethyl acetate (10:3). This solvent system was chosen as it displayed good separation on a TLC plate. A total number of 222 fractions in 3 ml test tubes were 108 collected and spotted onto a TLC plate. The separated spots on TLC were identified under UV lights (254 nm and 366 nm) and also sprayed with 0.5% anisaldehyde sulfuric acid reagent in glacial acetic acid:methanol:sulfuric acid (10:85:5). Fractions with similar Rf values were combined and concentrated to afford 7 major fractions. Fractions 3 and 5 were selected based on the results of their biological activity. Fraction 3 was eluted with toluene:ethyl acetate (10:3) and 189 sub-fractions (in 3 ml test tubes) were collected and combined into two major fractions resulting in the isolation of compound 1 (29.4 mg) with hexane:ethyl acetate:toluene (1:1:1) as eluent. Compound 2 (24.1 mg) was obtained after repeating column chromatography three times with toluene:ethyl acetate (10:2) from fraction 5. The isolation procedure of the two compounds is summarized in Figure 8.3. Crude extract (22.4 g) Silica gel 60 Hexane:Dichloromethane (9:1) Dichloromethane:Methanol (6:1) Methanol (100%) Fraction C (12.5 g) Silica gel 60 Toluene:Ethyl acetate (10:3) Fraction 3 (2.89 g) Fraction 5 (1.89 g) Silica gel 60 Toluene:Ethyl acetate (10:3) Silica gel 60 Silica gel 60 Hexane:Ethyl acetate:Toluene (1:1:1) Toluene:Ethyl acetate (10:2) Subfraction 83-95 Subfraction 123-142 Compound 1 (29.4 mg) Compound 2 (24.1 mg) Figure 8.3 Isolation procedures of the two compounds from S. radula. (NB: Fractions 1, 2, 4 and 6 had low activity and were not considered for further analysis). 109 The isolated compounds were structurally characterized using nuclear magnetic resonance (NMR) and mass spectrometry (MS) and data were compared to those found in the literature. 8.2.6.2 Structural characterization (i) Nuclear magnetic resonance spectroscopy 1H, 13C, HMBC and COSY nuclear magnetic resonance (NMR) data of compounds 1 and 2 were recorded on a Varian Inova 2000 300 MHZ spectrometer. All spectra were recorded at 25 ?C in deuterated chloroform and the chem ical shifts were recorded in ppm referenced to tetramethylsilane (TMS). (ii) Mass spectrometry Electron ionization MS of the isolated compounds were performed by direct inlet at 70 eV on the GC-MS QP 2010 gas chromatography-mass spectrometry. 8.3 Data analysis The percentage inhibition of radio-activity incorporation or growth (relative to the positive control values of P. falciparum without test sample) was calculated based upon the mean radio-active contents (CPM) of the experimental wells using a simple programme written in Microsoft Excel? (Eq. 8.2). % parasite growth = CPM (test compound sample) ? CPM ( mean of RBC control) x 100 (Eq. 8.2) CPM (mean of PRBC control) ? CPM (mean of RBC control) Where: CPM: Counts per minute RBC: Red blood cells PRBC: Parasitised red blood cells The log sigmoid dose-response curve was then constructed using the Enzfitter? version 1.05 and Prism? version 3.0 software and the IC50 values (the concentration required to kill 50% parasite) calculated as a measure of the antimalarial activity of the solvent extracts, isolated compounds and essential oils (Figure 8.4). The Student t-tests and one-way analysis of variance (ANOVA) were used to compare data. A linear regression was used in order to determine any possible relationship between the activity of the solvent extract and the essential oil of the same plant. For all statistical analyses, P < 0.05 was considered 110 significant. The IC50 values are given as the mean ? s.d. of three independent experiments. All the statistical analyses were performed using the Statistica? version 5.0 software. 8.4 Results 8.4.1 The antimalarial activity of the essential oils and solvent extracts The results of the in vitro tests of the essential oils and solvent extracts against chloroquine-resistant P. falciparum (FCR-3) are presented in Table 8.1. The essential oils and extracts of the evaluated species were found to exhibit antimalarial activity at IC50 values of less than 30 ?g/ml. The activity of essential oils (IC50 values) ranged from 1.20 ? 0.08 to 13.50 ? 2.20 ?g/ml. Among the plan ts tested, the essential oil from S. runcinata was significantly the most active (IC50 value: 1.20 ? 0.08 ?g/ml) (P < 0.05). -7.5 -5.0 -2.5 0.0 2.5 20 40 60 80 100 120 S. radula extract S. runcinata extract S. runcinata essential oil S. africana-caerulea essential oil log [ Salvia ] ( ?g/ml ) % P ar as it e gr ow th Figure 8.4 Antimalarial properties of the solvent extract of Salvia radula, S. runcinata and the essential oils of S. africana-caerulea and S. runcinata. An ANOVA showed that the essential oils of five species namely S. africana-caerulea, S. africana-lutea, S. albicaulis, S. dolomitica, S. muirii and S. stenophylla exhibited similar antimalarial activity (P > 0.05). The antimal arial activity of the solvent extracts (IC50 values) ranged from 3.91 ? 0.52 to 26.01 ? 2.95 ?g/ml. The extract of S. radula was the most active (IC50 value: 3.91 ? 0.52 ?g/ml) (P < 0.05). 111 Table 8.1 In vitro antimalarial activity of the solvent extracts and essential oils of indigenous Salvia species and isolated compounds (mean ? s.d., n = 3). Essential oils Solvent extracts Species/Isolated/ reference compounds IC50 values in ?g/ml S. africana-caerulea* 4.76 ? 0.48 22.68 ? 2.80 S. africana-lutea* 5.45 ? 0.35 15.86 ? 5.04 S. albicaulis* 6.41 ? 1.97 15.83 ? 1.94 S. aurita nd 8.92 ? 2.63 S. chamelaeagnea** 8.63 ? 0.96 8.71 ? 1.96 S. disermas nd 24.17 ? 4.10 S. dolomitica * 4.81 ? 0.74 7.62 ? 1.44 S. garipensis nd 13.95 ? 3.76 S. lanceolata* 7.83 ? 0.98 26.01 ? 2.95 S. muirii* 5.93 ? 1.03 11.87 ? 2.13 S. namaensis nd 25.38 ? 2.11 S. radula 13.50 ? 2.20 3.91 ? 0.52 S. repens* 1.65 ? 0.26*** 8.25 ? 2.09 S. runcinata* 1.20 ? 0.08 16.61 ? 3.33 S. schlechteri nd 17.51 ? 2.05 S. stenophylla* 4.13 ? 0.63 6.50 ? 1.37 S. verbenaca nd 23.97 ? 1.10 Compound 1 Compound 2 Chloroquine diphosphate Quinine sulphate 4.95 ? 2.00 a 24.60 ? 1.38 a 0.06 ? .01 0.14 ? 0.03 *Denotes activity of the essential oil is gr eater than activity of the solvent extract * *Denotes no significant difference between the activity of the solvent extract and essential oil (P >0.05) *** Essential oil obtained from Lady Grey (not from the same location as the solvent extract) a: mean ? s.e. ( n = 1) due to the small quantity of the isolated compound nd: Not determined, essential not available One-way analysis of variance showed that there was no difference in the antimalarial activity of the solvent extracts of S. aurita, S. chamelaeagnea, S. dolomitica, S. repens and S. stenophylla (P > 0.05). The activity of S. africana-lutea, S. albicaulis, S. garipensis, S. 112 muirii, S. runcinata and S. schlechteri were also not different (P > 0.05). Finally, S. africana-caerulea, S. disermas, S. lanceolata, S. namaensis and S. verbenaca also exhibited the same activity statistically (P > 0.05). Comparison of the activity of the volatile (essential oils) and non-volatile fractions showed significant differences with the volatile fractions generally exhibiting higher activity (P < 0.05) (Figure 8.5). It was interesting to note that S. radula, with the lowest activity among the essential oils (IC50 value: 13.50 ? 2.20), displayed th e best activity of the solvent extracts (Figure 8.1) (Table 8.1). In addition, there was no statistical difference between the activity of the solvent extract and essential oil of S. chamelaeagnea (P > 0.05). The rest of the species showed significant differences between the activity of the essential oil and solvent extract, with the activity of the essential oil being greater than that of the solvent extracts (P < 0.05). S. afr ica na -ca eru lea S. afr ica na -lu tea S. alb ica uli s S. ch am ela ea gn ea S. do lom itic a S. lan ce ola ta S. mu iri i S. ra du la S. rep en s S. ru nc ina ta S. ste no ph yll a 0 5 10 15 20 25 30 Essential oils Solvent extracts A nt im al ar ia l a ct iv it y (I C 5 0 va lu es in ? g /m l) Figure 8.5 Comparison between the antimalarial activity of the essential oils and the solvent extracts of eleven Salvia species (mean ? s.d., n = 3). The activity of the essential oils and solvent extracts was low compared to the reference antimalarial drugs such as chloroquine and quinine (P < 0.05) (Table 8.1). The essential oil of S. runcinata (IC50 value: 1.20 ?g/ml) was 20 and 9 times less active than the reference antimalarial drugs chloroquine and quinine, respectively. Similarly, the activity of S. radula (IC50 value: 3.91 ?g/ml) was 65 and 28 times less active than chloroquine and quinine, respectively. 113 The very poor correlation (r2 = 0.04) (Figure 8.6) between th e antimalarial activity of the essential oils and solvent extracts implies that the compounds responsible for the activity in the essential oils were most probably different to those found in the solvent extracts. Y = -0.245X + 0.945 r2 = 0.04 0.00 0.25 0.50 0.75 1.00 1.25 1.50 0.00 0.25 0.50 0.75 1.00 1.25 1.50 log IC50 of solvent activity (mg/ml ) lo g IC 5 0 of e ss en ti al oi ls a ct iv it y (? g /m l) Figure 8.6 Relationship between the antimalarial activity of the solvent extract and essential oil of the same plant. 8.4.2 Identification and characterization of the isolated compounds The 1H-NMR and the 13C-NMR of the isolated compounds are displayed in Table 8.2 and Table 8.3 and were compared to the data found in the literature (Voirin, 1983; Harraz et al., 1995; Rouf et al., 2001). The 1H and 13C spectra of the two isolated compounds are in Appendix D. Compound 1 was isolated as a white anhydrous powder. From the mass spectrum, compound 1 has a M+ at m/z 476 which corresponds to a molecular formula of C30H52O4. This compound, a triterpene, was identified as betulafolientriol oxide. 114 Table 8.2 NMR data (in CDCl3) of compound 1 isolated from Salvia radula. Position ?C ?H 1 33.6 2 25.4 3 76.2 3.39, br, s 4 37.6 5 49.6 6 18.2 7 34.6 8 40.0 9 50.3 10 37.3 11 31.2 12 71.0 3.51, td, J = 10.6, 4.7 Hz 13 49.4 14 52.1 15 32.6 16 28.6 17 47.9 18 15.4 0.98, s 19 16.1 0.86, s 20 86.5 21 27.6 1.26, s 22 31.2 23 25.0 24 85.4 3.84, dd, J = 9.0, 6.7 Hz 25 70.1 26 26.1 1.09, s 27 27.9 1.27, s 28 28.3 0.93, s 29 22.0 0.83, s 30 18.2 0.91, s OH 5.58, s; 3.89, s CDCl3: chloroform, s: singlet, d: doublet, dd: double of doublets, td: triple of doublets, t: triplet, br: broadened, ?H: proton shift, ?c: carbon shift: J: coupling constant, Hz: hertz. 115 Compound 2 was isolated as a yellow anhydrous powder, and was determined to be a flavonoid known as hydroxy-6,7-dimethoxy-2-(4-methoxyphenyl)-4H-1-benzopyran-4-one (salvigenin) with a molecular formula of C18H16O6. The UV and chromatographic data of betulafolientriol oxide and salvigenin are shown in Figure 5.1 and Table 5.1, while the structures of the two compounds are presented in Figure 8.7. Table 8.3 NMR data (in CDCl3) of compound 2 isolated from Salvia radula. Position ?C ?H Correlations observed in HMBC 2 162.6 3 104.1 6.59, s C-1?,4 4 182.7 4a 106.1 5/8a 153.2 6 132.6 7 158.7 8 90.5 6.55, s, H-8 C-4,6 8a/5 153.1 1? 123.6 2?,6? 128.0 7.85, d, J = 9.0 Hz C-2?,4?,6? 3?,5? 114.5 7.02, d, J = 9.0 Hz C-1?,2,3?,5? 4? 164.0 6-OCH3 60.8 3.93, s C-6 7-OCH3 56.3 3.97, s C-7 4?-OCH 3 55.5 3.89, s C-4? 5-OH 12.77, s C-4a, 5, 6 CDCl3: chloroform, s: singlet, d: doublet, ?H: proton shift, ?c: carbon shift: J: coupling constant, Hz: hertz, HMBC: heteronuclear multiple bond correlation. The two isolated compounds were tested for antimalarial activity. Betulafolientriol oxide exhibited the best activity compared to the flavanoid (salvigenin) with 5 ?g/ml killing 50% of parasites (Table 8.1). Although statistical comparison was not possible (because the 116 isolated compounds were tested only once due to the small amount obtained through the isolation procedure), salvigenin was less active, while betulafolientriol oxide exhibited comparable activity to the crude extract of solvent S. radula. the two compounds were less active than some essential oils and the reference antimalarial drugs (Table 8.1). H O OH O OH O 8a 4a 4 3 2 8 7 6 5 O 1' 6' 5' 4' 3' 2' O CH3 O CH3 OH O CH3 (1) (2) Betulafolientriol oxide 5-Hydroxy-6,7-dimethoxy-2-(4-methoxyphenyl)- (C30H52O4) 4H-1-benzopyran-4-one (C18H16O6) (Salvigenin) Figure 8.7 Structures of the isolated compounds from Salvia radula. 8.5 Discussion Malaria is a worldwide problem and the increasing spread of P. falciparum strains that are resistant to standard treatment has initiated numerous studies aimed at identifying new antimalarials. Traditional medicine seems to be a very important source of active molecules (Bodeker, 2000). Indeed, at this moment, compounds from plant origin, such as quinine or its alkaloids and, more recently, artemisinin and its derivatives, are the most efficient drugs available to combat malaria (Price et al., 1998). The essential oils and organic extracts from Salvia species in this study exhibited antimalarial properties against the P. falciparum FCR-3 strain with all IC50 values lower than 30 ?g/ml. These results confirm that species belonging to the genus Salvia contain metabolites with considerable antimalarial potential against P. falciparum. Essential oils were significantly more active than the solvent extracts (P < 0.05). Essential oil constituents have low molecular weights and can easily diffuse through the cell 117 membrane. This may partly explain why they are more active than the solvent extracts. The direct application of an essential oil in the treatment of malaria has not yet been considered. However, the presence of potential antimalarial compounds in the essential oils, as well as solvent extracts, might open possibilities for further investigation in the field of antimalarial compounds (Valenti et al., 1995). The biological activity of the solvent extracts and essential oils depends on the type of compounds present in each fraction. For instance, the antimalarial activity of essential oils may be attributed to their high sesquiterpene content. Sesquiterpenoids and their derivatives are credited with numerous biological properties, including antimalarial activity. Recently, nerolidol (an acyclic oxygenated sesquiterpene) from the leaves of Virola surinamensis (Rol.) Warb. was identified as an active component of an essential oil that displayed antimalarial activity (Lopes et al., 1999). Furthermore, parasites treated with nerolidol showed decreased ability to synthesize co-enzyme Q in all intra-erythrocytic stages (Boyom et al., 2003). van Zyl et al. (2006) evaluated the antimalarial activity of twenty individual commercial essential oil constituents (alcohols, aldehydes, esters, ketones, phenols and terpene hydrocarbons). Nerolidol, linalyl acetate, pulegone and ?- pinene were amongst the most active constituents (IC50 values < 0.3 ?g/ml). ?-Pinene is present in the majority of essential oils (traces-22.3%), while linalyl acetate was only detected in S. dolomitica (19.6%) (Table 4.1). The antimalarial activity of many essential oils from other plant families is well reported in literature. The essential oils of Tetradenia riparia (Lamiaceae) obtained from the leaves and stems exhibited moderate antimalarial activity against chloroquine-sensitive D10 and the chloroquine-resistant strain FAC-8 (IC50 values: 25.5 and 50.3 ?g/ml, respectively). Major components of the oils were ?-terpineol, fenchone, ?-fenchyl alcohol and perillyl alcohol and these oil components were absent in the current study and may explain the difference in the antimalarial activity (Campbell et al., 1997). Valenti et al. (1995) investigated the antimalarial activity of Lippia multiflora (Verbenaceae) and found that the essential oils also displayed activity. The major components of the oil, nerolidol (45.2%) and linalool (20.2%), were thought to be responsible for the activity. Nyiligira (2004) investigated the antimalarial activity of Vitex species against the chloroquine-resistant FCR-3 and found that the oils were also active (IC50 values < 9.3 ?g/ml). Although the essential oils studied comprise a multitude of compounds, it can be speculated that the presence of the compounds listed above and identified in indigenous Salvia essential oils may have contributed to the activity observed. 118 It is important to note that not all essential oils exhibit antimalarial properties. For instance, Milhau et al. (1997) studied the activity of the essential oil of S. officinalis and found that the activity was very poor (IC50 value > 1000 ?g/ml). This poor activity may, however, be attributed to the fact that a different parasite strain was used and the chemical composition of the oil was also different. Clarkson et al. (2004) demonstrated that S. repens displayed antimalarial activity against P. falciparum (IC50 value: 10.8 ?g/ml). In this study, the IC50 value of S. repens was 8.25 ? 2.09 ?g/ml which is similar to that obtained by Clarkson et al. (2004). Phenolic compounds are credited with various biological activities including anticancer, antibacterial, anti-oxidant activities. Oliveira et al. (2004) demonstrated that the antimalarial activity of Bidens pilosa (Asteraceae) was mainly correlated to its flavonoid content. In Chapter 5, the HPLC analysis revealed the presence of flavonoids in the solvent extracts of the investigated species. This class of compound may have also contributed to the antimalarial activity of the solvent extracts. Alkaloids also exhibited antimalarial activity against P. falciparum (Sanon et al., 2003). Alkaloids isolated from Sida acuta (Malvaceae) were also tested for antimalarial activity and results pointed out that they were 18 times more active than the solvent extract (Karou et al., 2003). Although no research was done in this study to confirm the presence of alkaloids in species investigated, alkaloids have previously been identified in some indigenous Salvia species, such as S. chamelaeagnea, S. namaensis and S. runcinata (Raffauf, 1996). Salvia species are traditionally used to treat fever, cough and diarrhoea, all symptoms of malaria infection. This study showed the in vitro efficacy of Salvia species against P. falciparum, the causative organism. The isolated compounds from S. radula, and the presence of betulafolientriol oxide (IC50 value: 5 ?g/ml) in all the investigated species (HPLC analysis, Chapter 5), may have contributed to the overall activity of the solvent extracts. Betulafolientriol oxide may act alone or in combination with other compounds present to determine the overall activity of the solvent extract. 119 It has also been demonstrated that methoxylated aromatic compounds, such as compound 2, display antimalarial activity (Kraft et al., 2002). In this study, salvigenin (compound 2) was not as active the crude extract or reference antimalarial drugs (Table 8.1). Phytochemical studies of plants used to treat the symptoms of malaria in traditional medicine of different countries revealed the presence of methoxylated aromatic compounds (Reddy et al., 2003). Furthermore, the methoxylated flavonoids have been found to enhance the in vitro activity of artemisinin (Liu et al., 1989; Bilia et al., 2002). The activity of the isolated compounds was low compared to that of the crude extract (Table 8.1). Williamson (2001) demonstrated that the compound known as ginkgolides, isolated from Ginko biloba, was less effective than the crude extract. The activity of the crude extract may be the result of a multitude of compounds present in the solvent extracts acting in a synergistic manner rather than the effect of only one compound. 8.6 Conclusions ? Salvia species exhibit antimalarial properties against P. falciparum FCR-3. ? Generally, the activity of the volatile fraction was statistically greater than that of the non-volatile fractions. ? Betulafolientriol oxide and hydroxy-6,7-dimethoxy-2-(4-methoxyphenyl)-4H-1- benzopyran-4-one isolated from S. radula also displayed antimalarial activity, but the activity of the two compounds was low compared to the antimalarial activity of the crude extract. ? The essential oils and solvent extracts were less active than chloroquine diphosphate and quinine sulfate. ? This preliminary study provides a scientific basis for the traditional use of Salvia species in treating malaria, although more thorough in vivo evaluation is required. 120 Chapter 9: Antibacterial and Antimycobacterial Activities of Salvia Species and Isolated Compounds from S. chamelaeagnea Abstract Salvia species are commonly used in traditional medicine to treat various microbial infections. Essential oils and solvent extracts of South African Salvia species were investigated for their antibacterial and antimycobacterial activities using the micro-dilution and BACTECTM 460 radiometric methods, respectively. The micro-organisms included two Gram-positive (Staphylococcus aureus and Bacillus cereus), two Gram-negative (Escherichia coli and Klebsiella pneumoniae) bacterial strains, as well as the organism responsible for tuberculosis, Mycobacterium tuberculosis. With the exception of S. radula, all the solvent extracts displayed moderate to good activity against all the Gram-positive and Gram-negative bacterial strains with the MIC values ranging from 0.03 to 8.00 mg/ml. In general, S. chamelaeagnea exhibited the most favorable activity. The solvent extracts also exhibited promising activity against M. tuberculosis with the MIC values ranging between 0.10 and 0.50 mg/ml with S. dolomitica, S. radula and S. verbenaca displaying the best activity (MIC value: 0.10 mg/ml). The essential oils showed moderate activity against B. cereus and K. pneumoniae, while negligeable activity was observed against E. coli and S. aureus (MIC value > 12 mg/ml). The in vitro antibacterial activity of S. chamelaeagnea and Leonotis leonurus combined at various ratios showed in general synergistic interaction against the Gram-positive bacterial strains, while antagonistic and/or additive interactions were observed for the Gram-negative organisms. The antibacterial bio-assay guided fractionation of S. chamelaeagnea resulted in the isolation, identification and characterization of four compounds, namely carnosol, 7-O-methylepirosmanol, oleanolic acid and its isomer ursolic acid as the active principles against S. aureus. The in vitro antibacterial and antimycobacterial activities add scientific support to the use of Salvia species in traditional medicine against tuberculosis and a range of other bacterial infections. 121 9.1 Introduction Micro-organisms are involved in the pathogenesis of many diseases and cause deterioration of a variety of products. Despite the progress in understanding the life cycle and control of many pathogens, nearly all the diseases affecting millions of people in developing countries are still caused by micro-organisms. This is mainly due to factors such as inadequate sanitation, poor hygiene and overcrowded living conditions (Wagstaff, 2002). Tuberculosis (TB), a mycobacterial infection, remains a serious health problem in many regions of the world. It is estimated that a third of the world?s populat ion is infected with Mycobacterium tuberculosis and approximately 1.7 million people died of TB in 2003 (WHO, 2005b). In 2004, TB killed half a million people in Africa, mostly young men and women in their most productive years. Africa is the only continent where the TB rates are increasing. In just 15 years, overall rates have doubled, tripled in HIV endemic areas, and quadrupled in countries worst affected by HIV and TB (WHO, 2005b). The current threat to the success of TB treatment lies in the emergence of strains resistantance to the best antituberculosis drugs, namely isoniazid and rifampicin (Mativandlela et al., 2006) and the extreme drug-resistant TB strain that emerged in South Africa has drastically compromised the treatment programmes to eradicate TB. Common symptoms associated with TB include night sweats, fatigue, chest pains, loss of appetite resulting in weight loss (WHO, 2006a). Infections associated with bacterial pathogens are among some of the conditions treated using traditional remedies in South Africa (Watt and Breyer-Brandwijk, 1962). In the past, the development of resistance by pathogens to many of the commonly used antimicrobial agents provided sufficient impetus for further attempts to search for new antimicrobial agents to combat infection (Grange and Davey, 1990). Sage is a potential source for new antimicrobial agents since it was used prior to the discovery of antimicrobial agents as a common component of herbal mixtures to treat the symptoms of tuberculosis and microbial infections (Watt and Breyer-Brandwijk, 1962). The use of synergistic combinations is often needed in the treatment of serious infection and to reduce the risk of developing resistant strains. In traditional medicine, healers mostly rely on a combination of plants to treat diseases (Iwu, 1994). However, no in vitro 122 studies have as yet been reported on the effect of combining Salvia species with other plants against bacteria. A survey of the literature revealed that Salvia species were usually used by Europeans in the Western Cape of South Africa with Leonotis leonurus (L.) R.Br. (Lamiaceae) to treat influenza, chest inflammation, pulmonary tuberculosis, skin diseases, colds and coughs (Watt and Breyer-Brandwijk, 1962). This information prompted an investigation into the in vitro antibacterial activity of S. chamelaeagnea and L. leonurus in combination. The in vitro antibacterial and antimycobacterial screening provided the preliminary information for the selection of those active crude extracts of plants that were subjected to further chemical analysis. The objectives of this study were to: (i) investigate the antibacterial activity of the essential oils and solvent extracts of indigenous Salvia species, (ii) evaluate the potential of the solvent extracts in inhibiting the growth of the tuberculosis pathogen, (iii) evaluate the in vitro antibacterial activity of S. chamelaeagnea and L. leonurus in combination, and (iv) isolate, identify and characterise the compounds exhibiting antibacterial activity from one of the active species. 9.2 Materials and methods 9.2.1 Reagents, chemicals and drugs Tryptone soya agar (TSA), Tryptone soya broth (TSB) and ciprofloxacin were purchased from CA Milsch, while p-iodonitrotetrazolium chloride (INT), ursolic acid, oleanolic acid and rifampicin were from Sigma? . The BACTECTM 12B media was purchased from Becton Dickinson. All solvents used were analytical grade from Rochelle Chemicals. The blood agar was obtained from the National Heath Laboratory Services (NHLS, Johannesburg). 123 9.2.2 Source and maintenance of micro-organisms Four strains of bacteria were investigated and included Bacillus cereus (ATCC 11778), Staphylococcus aureus (ATCC 25923), Klebsiella pneumoniae (NTCC 9633) and Escherichia coli (ATCC 8739). The reference stock cultures were obtained from the NHLS (Johannesburg) and were maintained in the microbiology laboratory of the Department of Pharmacy and Pharmacology, University of the Witwatersrand, Johannesburg, South Africa. The maintenance was carried out using TSA prepared by dissolving 30 g in 750 ml of sterile water. The agar was autoclaved for 15 minutes at 121 ?C before use in order to sterilise the medium. Mycobacterium tuberculosis H37Ra ATCC 25177 was used for the antimycobacterial experiments. The stock culture in glycerol was stored at -70 ?C until experimentation. The stock was used to inoculate a 12B vial (Figure 9.1) and incubated at 37 ?C. The 12B medium is an enriched Middlebrook 7H9 broth base containing 14C-labeled palmitic acid for radiometric detection of mycobacterial growth. Figure 9.1 (A): BACTECTM 460 apparatus with racks containing inoculated vials; (B): the 12B vials containing the 14C carbon substrate. If micro-organisms are viable in the test sample, they utilize the 14C-labeled substrate (palmitic acid) present in the medium and release 14CO2 into the head space above the medium. When tested on the BACTECTM 460 instrument, the gas is aspirated from the vial and the 14CO2 radio-activity is measured and displayed as a growth index (GI) ranging from 0 to 999 (Middlebrook et al., 1977). The daily increase in the GI is directly proportional to the rate and amount of growth in the medium. 124 9.2.3 Sample preparation Stock solutions of both plant extract (64 mg/ml) and essential oil (128 mg/ml) for the were prepared with acetone for the antibacterial study. For the combination experiment, the concentration of each plant was first carried out at 64 mg/ml and various ratios using the stock solution prepared. The stock solution used in the antimycobacterial activity testing was prepared at a concentration of 42 mg/ml with 50% methanol. The animycobacterial activity of the essential oils was not performed because the quantity of oil was not sufficient. 9.2.4 Determination of the antibacterial and antimycobacterial activities 9.2.4.1 Antibacterial activity: micro-dilution method The in vitro antibacterial activity was evaluated by a micro-dilution assay (Eloff, 1998). Dilutions were prepared in a sterile, flat-bottom, 96-well microtitre plate. In a 96-well plate, 100 ?l of sterile water was plated out in all the wells. In the first row, 100 ?l (stock solution) of plant extract/essential oil was added to the well and 100 ?l transferred to the next row after thorough mixing with the water in the 96-well microtitre plate using a multichannel pipette. This dilution series is prepared from the top to the bottom of the plate until the last row where 100 ?l from each well was discarded. The bacterial suspension (100 ?l), yielding an inoculum size of approximately 1 x 106 CFU/ml was added to all the wells. The final concentration in the wells ranged from 16.00 to 0.13 mg/ml (extract) and from 32.00 to 0.50 mg/ml (essential oil). Further serial dilutions were made where necessary. The plate was then sealed and incubated at 37 ?C overnight. After incubation, 50 ?l of 0.04% (w/v) INT was added directly into each well and the plate left at room temperature for six hours before the readings were taken. The method is based on the principle that the colourless tetrazolium compound (INT) acts as an electron acceptor and is reduced to a coloured product by biologically active organisms (Eloff, 1998). Sterile water was used as a negative control, while ciprofloxacin was used as the positive antibacterial control. The final concentration (3%) of acetone in the well did not adversely affect the bacterial growth. The minimum inhibitory concentration (MIC) value was determined as the lowest concentration of sample required to inhibit the growth of test organisms. In the microtitre plate, the MIC can be seen as the lowest concentration where the clear well is observed 125 (Figure 9.2). In some cases, the colour of the extract prevents a good visualisation of the bacterial growth/inhibition. Thus, the agar plate was prepared in order to double check whether there was effectively no bacterial growth. Figure 9.2 An example of the determination of the MIC value using a 96-well microtitre plate. The first clear well from the bottom indicates the MIC value, i.e. the concentration at which no bacterial growth is obtained and the purple well indicates bacterial growth. 9.2.4.2 Combination study of S. chamelaeagnea and L. leonurus To study the combined antibacterial activity of S. chamelaeagnea and L. leonurus (methanol:chloroform, 1:1 v/v), ten combinations at various ratios were prepared utilising extracts with an initial concentration of 64 mg/ml in acetone. The following ratios of S. chamelaeagnea and L. leonurus were considered, respectively: 10:0; 9:1; 8:2; 7:3; 6:4; 5:5; 3:7; 2:8; 1:9 and 0:10. A volume of 100 ?l of different combinations was introduced into a 96-well microtitre plate containing 100 ?l of water and the experiment was then carried out as described in Section 9.2.4.1. In order to analyse the outcome of the combination study, the sum of fractional inhibitory concentration of S. chamelaeagnea and L. leonurus (sum FICSL) was mathematically determined (Eq. 9.1) (Berenbaum, 1978). All ratios were presented graphically by construction of isobolograms as described by Canfield et al. (1995). The data points for each pathogen were obtained for various ratios by plotting the MICS in combination divided by the MICS alone on the X axis versus the MIC L in combination divided by the 0.25 mg/ml 0.13 mg/ml Negative control Decreasing concentration 16.0 mg/ml 32.0 mg/ml Decreasing concentration Essential oil Extract 8.0 mg/ml 1.0 mg/ml 126 MICL alone (on the Y axis). Based on the sum FICSL values, values < 1 indicate synergistic interactions, values > 1 show antagonistic interactions, while values ? 1 represent additive interactions. On the basis of the isobolograms, data points above the straight line have antagonistic interactions, those below have synergistic interactions and those in the vicinity of the line have additive interactions. sum FICSL = MIC S in combination + MICL in combination (Eq 9.1) MICS alone MICL alone Where: FICSL = Sum of the fractional i nhibitory concentration of S. chamelaeagnea and L. leonurus. MICS = Minimum inhibitory concentration of S. chamelaeagnea. MICL = Minimum inhibitory concentration of L. leonurus. For each analysis, samples were performed in duplicate and at least three experiments were carried out per test sample. 9.2.4.3 Bio-assay on M. tuberculosis : BACTECTM method The susceptibility testing of the solvent extracts against M. tuberculosis was conducted using the radiometric BACTECTM technique as described by Siddiqi et al. (1981), Roberts et al. (1983) and Scarparo et al. (2002). This method, despite the relatively high cost of the 12B culture medium, was preferred to the agar method for plant extract susceptibility testing because the radiometric method allows greater cell-to-drug interaction in the liquid medium as described by Lall and Meyer (1999). The 12B vials were inoculated with M. tuberculosis from the stock solution, incubated at 37 ?C and changes in the GI were recorded daily until a value of 999 was obtained. The mycobacterial suspension was adjusted to approximately match that of a McFarland No. 1 turbidity standard (Lall and Meyer, 1999). Thereafter, a 12B vial was inoculated with 100 ?l of the mycobacterial suspension and incubated at 37 ?C. When the growth in this vial reached 400-500, the 7H9 broth culture was used undiluted to inoculate a set of 12B vials (100 ?l per vial), yielding approximately 1 x 104-1 x 105 CFU/ml (Heifets et al., 1985). Before the extracts were added to the vials, each sample was filtered through a 0.22 ?m Whatman? filter unit to prevent any contamination. The final extract concentration of 1, 0.5 and 0.1 mg/ml were initially prepared in 50% methanol. Vials were incubated at 37 ?C 127 and visually monitored every day for turbidity as an indicator of contamination and any suspected contaminants were confirmed by culturing on 2% blood agar plates (NHLS, Greenpoint) made with Mueller Hinton (Oxoid) and 2% blood. Two plant extract-free vials were used as controls. One vial was inoculated with the organism in the same way as the vial containing plant extracts and the other was inoculated with a 1:100 dilution of the inoculum obtained by adding 0.10 ml of inoculum to 9.90 ml of 12B medium. A methanol control at 2.38%, equivalent to the maximum solvent concentration in the 12B test samples, had no negative impact on the normal growth of M. tuberculosis. The radiometric susceptibility method indicates that the concentration of the extract that results in a daily increase and final ?GI reading lower than in the control (1:100) can be considered the concentration at which more than 99% of the bacterial population is inhibited. Rifampicin, a standard antituberculosis drug, was used as positive control at a final concentration of 2.00 ?g/ml. The difference in the GI values for the last two days is designated as ?GI. The GI readings of the plant extracts were compared with the control vial containing 1:100 dilution of the inoculum. Once the ?GI of the control reached 30, the results were interpreted by comparing the increase with that in the test vial. The following formula was used to interpret the results: ?GI control > ?GI drug = susceptible; ?GI control < ?GI drug = resistant and the MIC va lue was defined as the lowest concentration of drug inhibiting more than 99% of the M. tuberculosis population. The final MIC value was confirmed by repeat testing. 9.2.5 Isolation of antibacterial active compounds from S. chamelaeagnea S. chamelaeagnea which displayed the most favourable antibacterial activity was selected to isolate the active principles. Combinations of various methods were used in isolation, identification and characterization procedures. 9.2.5.1 Extraction Dried aerial parts (423.4 g) collected at the Kirstenbosch Botanical Garden were powdered and extracted with methanol:chloroform (1:1, v/v) and the residue was evaporated under reduced pressure to yield 19.6 g. 128 9.2.5.2 Column chromatography The residue was mixed with a small quantity of Silica gel 60 (0.063-0.200 mm/70-230 mesh ASTM, Macherey-Nagel), dried at 60 ?C in an oven, reduced to a fine powder and added to the top of the column. The column (4 cm x 60 cm) was eluted successively with hexane:dichloromethane (9:1), then dichloromethane:methanol (6:1) followed by 100% methanol. The collection of these crude fractions were based on colour change. The fractions were collected in 250 ml conical flasks and similar spots on TLC were combined to yield five main fractions (A-E). 9.2.5.3 Thin layer chromatography and bio-autographic assay A direct antibacterial bio-assay (Dilika and Meyer, 1996) on a TLC plate was employed using the agar-overlay method with S. aureus as the test organism. S. aureus was selected because it is amongst the most difficult bacterial strains to combat with classic antibacterial agents (Tom?s-Barber?n et al., 1990). Each of the five fractions (Section 9.2.5.2) were spotted on a TLC plate (Alugram? Sil G/UV254, 0.200 mm) developed with toluene:dioxan:acetic acid glacial (95:25:5) as eluent. The dried TLC plate was then sterilised under UV light (254 nm) for one hour before being placed in a petri dish containing the solidified agar. The UV light was used to sterilise the dried extract instead of autoclaving in order to avoid denaturing the active constituents (Okoli and Iroegbu, 2004). The plate was then overlayed with S. aureus culture (1 x 106 CFU/ml) in TSA and left to solidify. After one hour at 4 ?C (to allow the compounds on the TLC to diffuse), the plate was removed and incubated at 37 ?C fo r 24 hours. At the end of the incubation period, the TLC plate was sprayed with 0.04% (w/v) aqueous solution of INT to visualise the inhibition zone. The inhibition zone was indicated by the clear zone on the purple background (Figure 9.3). Only fraction D (Figure 9.3) was active and thus selected for further analysis. 9.2.5.4 Bioactivity-guided fractionation and isolation Fraction D was evaporated under vacuum to yield 6.21 g. This fraction was rechromatographed on a small column (2.5 cm x 45 cm) using hexane:ethyl acetate: acetic acid (12:3:0.02) as eluent. A total of 322 sub-fractions in 3 ml test tubes were collected. The fractions were spotted on the TLC plate and developed in toluene:dioxan:acetic acid glacial (95:25:5) and similar fractions were combined and subjected to a bio-autographic 129 assay on TLC to detect fraction(s) containing the active compound(s). The eluted fractions were monitored on a TLC plate developed with the same solvent system. Figure 9.3 Bio-autographic assay of Salvia chamelaeagnea against Staphylococcus aureus: Inhibition zones are shown by clear spots on the purple background. Pooled fractions were concentrated with a rotary evaporator and rechromatographed using the same column with hexane:ethyl acetate:acetic acid (12:2:0.01) as the solvent system. The column chromatography of the active fraction was finally carried out with toluene:ethyl acetate (10:3) and 293 sub-fractions were obtained. Throughout the isolation procedure, crystals formed were further purified by recrystalisation from acetone. The bioactivity-guided methodology is summarized in Figure 9.4. 9.2.5.5 Identification and elucidation of isolated compounds The isolated compounds were structurally characterised using NMR and MS and data compared to those found in the literature [1 and 2 (Urones et al., 1998; Ahmed et al., 2006), 3 and 4 (Mahato and Kundu, 1994)]. (i) Nuclear magnetic resonance (NMR) 1H, 13C, HSQC, HMBC, DEPT, NOESY and COSY NMR data of compounds 1, 2, 3 and 4 were recorded on a Varian Inova 2000 300 MHZ spectrometer. All spectra were recorded at 25 ?C in deuterated chloroform and the chem ical shifts were recorded in ppm referenced to TMS. 130 (ii) Mass spectrometry Electron ionization MS of the isolated compounds was performed by direct inlet at 70 eV on the GC-MS QP 2010 GC coupled to MS. Crude extract (19.6 g) Silica gel 60 Hexane:Dichloromethane (9:1) Dichloromethane:Methanol (6:1) Methanol (100%) Fraction D (6.21 g) Silica gel 60 Hexane:Ethyl acetate:Acetic acid (12:3:0.02) Fractions 18-95 (0.63 g) Silica gel 60 Fractions 18-45 (0.43 g) Silica gel Compound 1 (120 mg) Hexane:Ethyl acetate:Acetic acid (12:2:0.01) (Crystals) Flash silica Toluene:Ethyl acetate (10:3) Sub-fraction 18-25 Sub-fraction 47-54 Compound 2 (12 mg) Compound 3 and 4 (9 mg) Figure 9.4 Bio-assay guided fractionation for compounds isolated from S. chamelaeagnea. 9.3 Results 9.3.1 Antibacterial activity The solvent extracts and essential oils affected the test organisms to different degrees and the results of the antibacterial activity showed that the solvent extracts exhibited better antibacterial activity than the essential oils (Table 9.1). 131 Table 9.1 The MIC values (mg/ml) of the solvent extracts and essential oils of Salvia species against Gram-positive and Gram-negative bacteria (n = 3). nd: not determined, essrential oil insufficient or not available; EO: essential oil; Ext: solvent extract. E. coli (ATCC 8739) K. pneumoniae (NCTC 9633) B. cereus (ATCC 1177) S. aureus (ATCC 25923) Species/Isolated compound/Controls Ext EO Ext EO Ext EO Ext EO S. africana-caerulea 4.00 16.00 4.67 8.00 6.00 0.75 3.00 4.0 S. africana-lutea 3.00 > 32.00 3.00 8.00 0.75 2.33 0.75 8.00 S. albicaulis 4.00 > 32.00 4.67 8.00 1.00 2.33 1.00 8.00 S. aurita 4.00 nd 4.000 nd 0.25 nd 0.05 nd S. chamelaeagnea 1.00 > 32.00 3.0 12.00 0.03 2.00 0.06 > 32.00 S. disermas 4.00 nd 2.00 nd 3.00 nd 0.75 nd S. dolomitica 3.33 > 32.00 2.00 8.00 0.75 9.33 0.03 12.00 S. garipensis 1.88 nd 2.81 nd 0.75 nd 4.00 nd S. lanceolata 4.00 32.00 2.00 16.00 3.00 1.00 2.00 8.00 S. muirii 2.00 > 32.00 3.00 6.00 0.25 9.33 0.36 > 32.00 S. radula > 16.00 nd 4.00 nd 1.00 nd 0.06 nd S. repens 2.00 nd 3.00 nd 0.03 nd 0.25 nd S. runcinata 2.00 > 32.00 4.00 > 32.00 0.03 2.00 0.25 > 32.00 S. schlechteri 4.00 nd 3.00 nd 0.75 nd 0.25 nd S. stenophylla 4.00 > 32.00 2.00 8.00 0.03 1.50 0.06 > 32.00 S. verbenaca 8.00 nd 2.00 nd 2.00 nd 3.0 nd Carnosol 2.00 2.00 0.02 0.03 Oleonolic acid 3.41 1.71 1.28 3.75 Ursolic acid 2.50 3.75 7.32 x 10-3 0.01 7-O-Methyl-epi-rosmanol nd nd 0.01 0.02 Ciprofloxacin 4.00 x 10-5 1.60 x 10-4 4.10 x 10-5 3.1 x 10-4 132 Nearly all the solvent extracts evaluated for antibacterial activity were active against the strains used. While essential oils exhibited moderate activity against B. cereus and K. pneumoniae and were almost completely inactive against E. coli and, to a certain degree, S. aureus (Table 9.1). Among the plants screened, only the solvent extract of S. radula failed to inhibit the growth of E. coli, although displaying good activity against other pathogens. The best activity of the solvent extracts against B. cereus was obtained with S. chamelaeagnea, S. radula, S. runcinata and S. stenophylla (MIC value: 0.03 mg/ml), while S. africana- caerulea displayed the lowest activity (MIC value: 6.00 mg/ml). The best activity against S. aureus was obtained with S. dolomitica (MIC value: 0.03 mg/ml). Five species, namely S. disermas, S. dolomitica, S. lanceolata, S. stenophylla and S. verbenaca were the most active against K. pneumoniae (MIC value: 2.00 mg/ml). Finally, the best activity against the Gram-negative E. coli was obtained with the solvent extract of S. chamelaeagnea (MIC value: 1.00 mg/ml) and S. muirii (MIC value: 1.88 mg/ml). It was interesting to observe that the essential oil of S. africana-caerulea was more active than its solvent extract against B. cereus. The isobolograms of S. chamelaeagnea and L. leonurus at various ratios against the four pathogens are shown in Figure 9.5. Synergistic interactions were obtained against Gram- positive bacteria for nearly all ratios. Only the ratio 2:8 of S. chamelaeagnea and L. leonurus showed antagonistic interactions against S. aureus (data point corresponding to 2.50 and 1.24, cannot be seen on the graph, but on Table 9.2 as 3.74). Mostly antagonistic or additive interactions were observed with Gram-negative bacteria (Figure 9.5). However, ratios 5:5, 3:7 and 2:8 showed synergistic interactions against E. coli, while ratios 7:3 and 5:5 were also synergistic against K. pneumoniae. The sum of fractional inhibitory concentration values of S. chamelaeagnea and L. leonurus (sum FICSL) are displayed in Table 9.2. It can be seen that the response of the combination between the two plants varied with ratios and the specific test organism. Similarly, only ratio 2:8 (FICSL = 3.74) showed antagonistic interactions against the Gram-positive bacteria, while all the remaining ratios were synergistic (FICSL < 1). 133 E. coli 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 MICS in combination / MICS alone M IC L in c om bi na ti on / M IC L al on e K. pneumoniae 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 MICS in combination / MICS alone M IC L in c om bi na ti on / M IC L al on e S. aureus 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 MICS in combination / MICS alone M IC L in c om bi na ti on / M IC L al on e B. cereus 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 MICS in combination / MICS alone M IC L in c om bi na ti on / M IC L al on e Figure 9.5 Isobolograms constructed from combination study involving Salvia chamelaeagnea and Leonotis leonurus on four bacterial pathogens. The combination of S. chamelaeagnea and L. leonurus at various ratios showed synergistic interactions at only three ratios (5:5, 3:7 and 2:8) against E. coli, while against K. pneumoniae, the synergistic interactions were obtained at only two ratios (7:3 and 5:5) (sum FICSL < 1, Table 9.2). This study revealed that Gram-positive bacteria were more sensitive compared to Gram- negative bacteria. E. coli was in general the most resistant strain especially against the essential oils with all MIC values obtained greater than 32 mg/ml. 134 Table 9.2 The sum of the fractional inhibitory concentration (sum FICSL) values of Salvia chamelaeagnea and Leonotis leonurus using various combinations ratios. Pathogens Ratios E. coli (ATCC 8739) K. pneumoniae (NCTC 9633) B. cereus (ATCC 11778) S. aureus (ATCC 11778) 10:0 1.00 1.00 1.00 1.00 9:1 1.39 1.00 0.55 0.99 8:2 1.10 1.00 0.50 0.87 7:3 1.15 0.67 0.45 0.78 6:4 1.20 1.00 0.53 0.68 5:5 0.83 0.67 0.35 0.89 3:7 0.81 1.00 0.66 0.81 2:8 0.93 1.00 0.78 3.74 1:9 1.45 1.00 0.75 0.84 0:10 1.00 1.00 1.00 1.00 Values < 1 denote synergism interactions; values > 1 denote antagonism interactions and values ? 1 denote additive interactions. 9.3.2 Antimycobacterial activity Mycobacterium tuberculosis was sensitive to all the solvent extracts at a concentration of 1.00 mg/ml (Table 9.3). Three species, namely S. dolomitica, S. radula and S. verbenaca inhibited mycobacterial growth at a concentration of 0.10 mg/ml. Therefore, further dilutions yielded a MIC value of 0.10 mg/ ml for all three extracts (Table 9.3). Since the three species exhibited the same activity, the GI was plotted at 0.10 and 0.05 mg/ml and the results are depicted in Figure 9.6. It is clear that S. verbenaca was the most active extract because its GI readings at 0.10 and 0.05 mg/ml were low compared to that of S. radula and S. dolomitica (Figure 9.6) when tested at the same concentration. It was noted that these three species also exhibited good antibacterial activity against S. aureus and B. cereus (Table 9.1). 135 Table 9.3 The effect of plant extracts on the growth of Mycobacterium tuberculosis using the radiometric method. ?GI values (mg/ml) Plant species 1.00 0.50 0.10 MIC values (mg/ml) S. africana-caerulea 0 (S) -3 (S) 73 (R) 0.50 S. africana-lutea 0 (S) 0 (S) 50 (R) 0.50 S. albicaulis 0 (S) 0 (S) 41 (R) 0.50 S. aurita 0 (S) 4 (S) 33 (R) 0.50 S. chamelaeagnea (S) 1 (S) 37 (R) 0.50 S. disermas 0 (S) 18 (S) 520 (R) 0.50 S. dolomitica 2 (S) -1 (S) 11 (S) 0.10 S. lanceolata 0 (S) -1 (S) 40 (R) 0.50 S. muirii 0 (S) 7 (S) 33 (R) 0.50 S. radula -1 (S) 4 (S) 10 (S) 0.10 S. repens 0 (S) 2 (S) 48 (R) 0.50 S. runcinata 0 (S) 4 (S) 40 (R) 0.50 S. schlechteri 0 (S) 19 (S) 38 (R) 0.50 S. stenophylla 0 (S) 8 (S) 38 (R) 0.50 S. verbenaca -2 (S) 3 (S) 5 (S) 0.10 Rifampicin 0 (S) 0 (S) 0 (S) 0.002 S, sensitive; R, resistant, ?GI values of the control vials were 29, 29 and 28 at 1, 0.5 and 0.1 mg/ml, respectively. For the sample to be active against M. tuberculosis, ?GI must be less or equal to ?GI of the control vial. A rifampicin control vial (2 ?g/ml) was included in each experiment. 136 4 6 8 0 200 400 600 800 1000 S. dolomitica (50 ?g/ml) S. dolomitica (1 0 0 ?g/ml) S. radula (50 ?g/ml) S. radula (100 ?g/ml) S. verbenaca (50 ?g/ml) S. verbenaca (100 ?g/ml) Rifampicin ( 2 ?g/ml) Culture control Days B A T E C TM g ro w th in de x re ad in gs Figure 9.6 Growth index of three Salvia species tested at two different concentrations against Mycobacterium tuberculosis. 9.3.3 Elucidation and identification of the isolated compound from S. chamelaeagnea The 1H-NMR and 13C-NMR data of the four isolated compounds are presented in Table 9.4 to Table 9.6. The 1H, 13C spectra are shown in Appendix E. Based on the 1H-NMR , 13C- NMR, DEPT, COSY spectra and comparison with the data obtained in the literature, compound 1 was identified as 2H-9,4a-(epoxymethano) phenanthren-12- one,1,3,4,9,10,10a-hexahydro-5,6-dihydroxy-1,1-dimethyl-7-(1-methylmethylethyl)- 4aR,9S,10aS)-(9CI) (carnosol) and established to have a molecular formula of C20H26O4. Compound 2 was a diterpene identified as the 7-O-methylepirosmanol and established to have a molecular formula of C21H28O5. Compound 3 and 4 (triperpenoid ) were obtained as co-eluting mixture and identified as [(3?)-3-hydroxyurs-12-en-28-oic acid], commonly known as ursolic acid and the chemical formula is (C30H48O3), while compound 4, which is an isomer of compound 3, was characterized as [(3?)-3-hydroxyolean-12-en-28-oic acid] commonly known as oleanolic acid with the chemical formula (C30H48O3). The UV and chromatographic data of the isolated compounds are shown in Figure 5.1 and Table 5.1, while the absolute stereochemistries of the isolated compounds are presented in Figure 9.7. 137 Table 9.4 NMR data (in CD3OD) of compound 1 isolated from Salvia chamelaeagnea. Position ?C ?H Correlations observed in HMBC NOESY 1 29.0 2.82-2.88, dt C-2, 10, 20 2 18.9 1.64, t 3 41.0 1.58-1.62, m 4 34.3 5 45.9 1.74, m C-9, 10, 18, 19, 20 6 29.7 2.25, m C-4, 7, 8, C-5 7 78.6 5.5, s 8 132.2 9 121.8 10 48.6 11 143.6 4.90, s 12 143.1 4.90, s 13 134.9 14 111.4 6.74, s C-7, 9, 12, 15 H-7 15 26.8 3.30, m C-12, 13, 14, 16, 17 16 22.1 1.22-1.28, m C-15, 17 17 22.1 1.22-1.28, m C-15, 16 18 31.0 0.80, s C-3, 4, 5, 18 19 18.9 0.80, s C-3, 4, 5, 19 20 178.2 CD3OD: methanol: HMBC: heteronuclear multiple bond correlation, NOESY: nuclear overhauser effect spectroscopy, singlet, dt: double of triplets, t: triplet, m: multiplet, ?C: carbon shift, ?H: proton shift. 138 Table 9.5 NMR data (in CDCl3) of compound 2 isolated from Salvia chamelaeagnea. Carbon ?C ?H 1 26.9 2 18.8 3 37.7 4 31.7 5 55.5 6 74.2 4.35, d, J = 2.8 Hz 7 78.0 4.85, d, J = 2.8 Hz 8 125.7 9 124.8 10 47.8 11 143.9 12 141.2 13 135.3 14 119.7 6.84, s 15 27.1 16 22.0 1.11, d 17 22.5 1.13, d 18 31.5 0.91, 18 19 21.8 0.87, 19 20 179.4 OMe 56.0 3.50, s CDCl3: chloroform: ?C: carbon shift, ?H: proton shift, singlet, d: doublet, J: coupling constant, Hz: hertz. 139 Table 9.6 NMR data (in CDCl3) of the mixture of two triterpenoids: oleanolic and ursolic acid isolated from Salvia chamelaeagnea. ?C Mixture ?C Oleanolic acid (Mahato and Kundu, 1994) ?C Ursolic acid (Mahato and Kundu, 1994) 180.7 (s) 181.0 (s, C-28) 180.5 (s), 180.4 (C-28) 143.7 (s) 143.4 (s, C-13) 138.0 (s) 148.1 (s, C-13) 125.3 (d) 125.5 (d, C-12) 122.1 (d) 122.1 (d, C-12) 78.7 (d) (2C) 78.7 (d, C-3) 78.9 (d, C-3) 55.1 (d) (2C) 55.2 (d, C-5) 55.1 (d, C-5) 52.7 (d) 52.7 (d, C-18) 47.7 (s) 47.7 (s, C-17) 47.5 (d) 47.6 (d, C-9) 47.4 (d) 47.5 (d, C-9) 46.2 (s) 46.6 (s, C-17) 45.8 (t) 45.8 (t, C-19) 41.9 (s) 42.0 (s, C-14) 41.6 (s) 41.6 (s, C-14) 41.0 (d) 41.3 (d, C-18) 39.3 (s) 39.3 (d, C-8) 39.4 (d, C-8) 39.0 (s) 38.7 (s, C-4) 38.9 (d) 39.0 (d, C-19) 38.7 (d) 38.6 (d, C-20) 38.52 (t) 38.8 (t, C-1) 38.49 (s) 38.6 (s, C-4) 38.3 (t) 38.5 (t, C-1) 36.8 (s) 37.0 (s, C-10) 36.7 (s) 37.0 (s, C-10) 36.6 (t) 36.7 (t, C-22) 33.7 (t) 33.8 (t, C-21) 32.8 (q) 33.1 (q, C-29) 140 ?C Mixture ?C Oleanolic acid (Mahato and Kundu, 1994) ?C Ursolic acid (Mahato and Kundu, 1994) 32.8 (t) 33.0 (t, C-7) 32.6 (t) 32.8 (t, C-22) 32.4 (t) 32.6 (t, C-7) 30.5 (t) 30.6 (C-20) 30.6 (t, C-21) 27.84 (q) 28.1 (q, C-23) 28.2 (q, C-23) 27.80 (q) 28.0 (t, C-15) 27.5 (t) 27.9 (t, C-2) 26.7 (t) 27.7 (t, C-15) 26.6 (t) 27.4 (t, C-2) 25.6 (q) 26.0 (q, C-27) 24.0 (t) 24.2 (t, C-16) 23.3 (q) 23.6 (q, C-30) 23.2 (t) 23.4 (t, C-16) 23.1 (t) 23.3 (t, C-11) 22.9 (t) 23.1 (t, C-11) 20.9 (q) 21.1 (q, C-30) 18.5 (t) (2C) 18.3 (t, C-6) 18.2 (t, C-6) 16.8 (q) 16.9 (q, C-6, 27) 16.7 (q) 16.8 (q, C-26) 16.6 (q) 16.8 (q, C-29) 15.40 (q) 15.7 (q, C-25) 15.35 (q) 15.6 (q, C-24) 15.2 (q) 15.4 (q, C-24) 15.1 (q) 15.3 (q, C-25) ?H 5.21 (1H, br. s, H-12), 5.18 (1H, br. s, H-12), 3.14 (2H, t, J = 7.7 Hz, H-3). CDCL3: chloroform, s: singlet, d: doublet, t: triplet, q: quartet, ?H: proton shift, J: coupling constant, Hz: hertz. 141 O 7 6 5 4 3 2 1 10 20 8 9 11 12 13 14 CH 318 CH 3 19 O OH O H 15 CH 316 CH 317 H O 6 5 4 3 2 1 10 20 7 8 9 11 12 13 14 O O OH OH 15 CH3 16 CH3 17 CH3 18 CH3 19 CH3 (1) (2) 2H-9,4a-(epoxymethano)phenanthren-12-one, 1,3,4,9,10,10a-hexahydro-5,6-dihydroxy-1,1- 7-O-Methylepirosmanol dimethyl-7-(1-methylethyl)-4aR,9S,10aS)-(9CI) (C21H28O5) (C20H26O4) (Carnosol) 13 14 8 18 10 17 9 5 12 4 28 19 15 7 11 161 6 22 3 O 20 2 21 CH 327 O H CH 3 23 CH 324 OH CH 3 29 H H H CH 330 CH 326 CH 325 13 14 8 18 10 17 9 5 12 4 28 19 15 7 11 161 6 22 3 O 20 2 21 CH3 27 OH CH3 23 CH3 24 OH H H H CH3 30 CH3 26 CH3 25 CH3 29 (3) (4) (3?)-3-hydroxyurs-12-en-28-oic acid (3?)-3-hydroxyolean-12-en-28-oic acid (C30H48O3) (Ursolic acid) (C30H48O3) (Oleanolic acid ) Figure 9.7 Structures of the isolated compounds with antibacterial activity from Salvia chamelaeagnea. 9.4 Discussion The solvent extract displayed good to moderate antibacterial activity as shown by the low MIC values (Table 9.1). According to Fabry et al. (1998), the solvent extracts of plants with MIC values around 8.00 mg/ml may have some antimicrobial activity. The values obtained in this study were less than 8.00 mg/ml against all the pathogens tested and the 142 species studied can therefore be considered to display appreciable activity. More precisely, species such S. dolomitica, S. radula and S. verbenaca displayed good antibacterial and antimycobacterial activity with MIC values less than 3 mg/ml against B. cereus and S. aureus and MIC values of 0.10 mg/ml against M. tuberculosis, respectively (Table 9.1 and 9.3). The current results are consistent with the pattern for the in vitro activity emerging from other studies. The antibacterial activity of exotic Salvia species is reported in the literature (Lee et al., 1999; Ulubelen et al., 2001; Tepe et al., 2004) although the MIC values, the methodology used and pathogens differed. Variations reported between antimicrobial data are frequent and this can be explained by the difficulty to cultivate micro-organisms under the same test conditions, due to species-specific requirements (e.g. incubation temperature, growth medium and atmosphere). A further explanation for the differences in antibacterial activity data is the different sensitivity of strains, the mode and choice of solvent extraction and variation in the composition of the extracts. Indigenous Salvia species also exhibited good antimycobacterial activity with MIC values less than 1 mg/ml. The ethanol extract of S. aethiopis and compounds isolated from S. multicaulis (eg. multiorthoquinone) have been reported to display promising activity against M. tuberculosis (MIC values: 50 and 2 ?g/ml, respectively) (Cantrell et al., 2001; Tosun et al., 2004). Seaman (2005) found that a S. africana-lutea extract exhibited promising activity against M. aurum and M. tuberculosis (MIC values: 2 and 1 mg/ml, respectively) along with moderate activity against M. smegmatis (MIC value: 8 mg/ml). Salvia species are rich in phenolic compounds and the latter are known to display a broad range of biological activities such as antibacterial, anti-oxidant, anticancer and anti- inflammatory activities (Lu and Foo, 2001; Ulubelen et al., 2001; Yamazaki et al., 2004). The species investigated revealed high phenolic content (Chapter 5) may contribute to the observed activity. The essential oils showed moderate antibacterial activity, especially against B. cereus, while little activity was obtained against S. aureus (Table 9.1). The essential oils of many Lamiaceae have been known since antiquity to display various biological activities, notably their antimicrobial potency (Barrata et al., 1998). The study of biological and pharmacological activities of essential oils in aromatherapy is highly correlated to the chemical composition (Lawrence, 2000). The results relating to the hydrocarbon and the 143 oxygenated fractions of S. officinalis essential oil showed that the hydrocarbon fractions did not inhibit the growth of micro-organisms, while the oxygenated fractions exhibited a significant dose response effect (Carta et al., 1996). Delaquis et al. (2002) found that alcohols and aldehydes were particularly active against Gram-positive bacteria. Purified compounds derived from essential oils, such as carvacrol, borneol, geraniol, eugenol, linalool, camphor, 1,8-cineole and thymol, inhibit a wide variety of micro-organisms, while ?-pinene (monoterpene hydrocarbon) had slight activity against a range of micro- organisms (Hinou et al., 1989; Zakarya et al., 1993; Carta et al., 1996; Tzakou et al., 2001). The low potency (MIC values > 32 mg/ml) observed for some of the essential oils may possibly be due to the high concentration of hydrocarbon components (15-56%), while the moderate activity (MIC < 8 mg/ml) may be correlated to the oxygenated components present in the oils (22-74%) (Table 4.1). Compounds such as p-cymene, ?- pinene and limonene have been previously been shown to exhibit low antibacterial activity (Chalchat et al., 1997). The relationship between composition and biological activity suggests that the compounds present in the greatest proportions (e.g. linalool, linalyl acetate, ?-bisabolol, limonene, 1,8-cineole, ?-pinene) are not necessarily responsible for the greatest share of the total antibacterial activity. Thus, the involvement of the less abundant constituents (e.g. ?-caryophyllene alcohol, terpinen-4-ol) should be considered. Essential oils could possibly be rendered more useful if the relative concentrations of antimicrobial components were adjusted to levels that consistently provide the required potency and spectrum of inhibition (Delaquis et al., 2002), but the safety levels also need to be considered. Several factors may contribute to the disparity in observations with reported activity of essential oils. Variability in qualitative and quantitative estimates of activity has been ascribed to differences in analytical techniques (Mann and Markham, 1998). The volatility and poor solubility of most essential oils is problematic, particularly with methods that rely on dilution of the test substance in an aqueous medium (Delaquis et al., 2002). Furthermore, the essential oils are volatile and one has to take into account a loss of the active substances, especially when the incubation period is long (Pauli, 2001). They are complex mixtures comprising many individual compounds and each of these constituents may contribute to the beneficial or adverse effects of these oils (Lahlou, 2004). It was interesting to note in the current study that the activity of the essential oil of S. africana-caerulea was higher than that of the solvent extract (Table 9.1). In many studies, 144 the antibacterial activity of solvent extracts is generally high compared to that of essential oils (Lourens et al., 2004; Njenga et al., 2005, van Vuuren et al., 2006). Eriocephalus punctulatus and E. ericoides essential oils were also found to exhibit better activity in comparison to the acetone extract (Njenga et al., 2005). Lourens et al. (2004) also observed that Helichrysum excisum essential oil was more active than the methanol and acetone extract. The Gram-positive bacteria were found to be more sensitive than the Gram-negative bacteria (Table 9.1) which is in agreement with other studies (Delaquis et al., 2002; Viljoen et al., 2003; Njenga et al., 2005). Reasons for the resistance of many Gram- negative bacteria will likely remain speculative until the mode of action is better understood. Current hypotheses favour adverse effects on the integrity of the bacterial cell membrane (Delaquis et al., 2002). Studies have reported that the sensitivity of bacteria was related to the morphological structure and chemical composition of bacteria membranes. Gram-negative bacteria in general have an outer membrane composed mainly of polysaccharides which is rather impermeable, thereby preventing inhibitors from passing through (Nikaido and Vaara, 1985). Given the differences in the cell envelope of the Gram-negative bacteria, it is plausible that access through the cell membrane is more restricted in Gram-negative bacteria. Infections caused by S. aureus are among the most difficult to treat with standard antibacterial agents (Tom?s-Barber?n et al., 1990). The growth of S. aureus was inhibited by nearly all the extracts (Table 9.1), especially by S. aurita, S. chamelaeagnea, S. dolomitica, S. radula and S. stenophylla (MIC values ? 0.06 mg/ml). Thus, these plants could yield compounds that can improve the treatment of infections caused by S. aureus in particular. It was reported that many species of Salvia are used in decoctions or infusions in South Africa by Sotho people against possible bacterial infections such as throat inflammation, colds, coughs and to treat pulmonary diseases and tuberculosis (Watt and Breyer-Brandwijk, 1962). The in vitro antibacterial and antimycobacterial activity obtained in this study strongly support the traditional medicinal uses of these plants. As a rule, a single group of compounds dominates within a given taxon with major compounds often accompanied by several derivatives and minor components (Wink, 2003). The biological activity of any plant is primarily a function of the type of 145 compounds present. Those isolated from S. chamelaeagnea as well as some standards (rosmarinic acid, carnosic acid and caffeic acid) known to display antimicrobial activity were also present in most of the investigated Salvia species (Table 5.2) and have probably contributed to the observed activity. More precisely, some of the isolated compounds contain phenolic hydroxyl groups which are reported to be associated with antibacterial activity (Tom?s-Barber?n et al., 1990). The MIC values of the isolated compounds indicated promising activity against S. aureus and to a certain degree, against B. cereus. The solvent extracts are crude preparations and one should expect the activity of the isolated molecules to be higher than the crude extracts. This was not observed and may partly be attributed to the lability of the isolated compounds. Furthermore, it is not possible to quantify the antibacterial activity from bio-autographic assay where the inhibition spot is by nature a positive or negative response and does not reveal detail about the degree of activity of a specific compound. The mechanism of many phytomedicines is still unknown and there are several instances of total herb extract showing a better effect than an equivalent dose of an isolated compound. The active isolated compounds have previously been detected in many species of the Lamiaceae. Carnosol has been reported to display antibacterial activity against various micro-organisms including S. aureus, Bacillus subtillus, E. coli and Candida albicans (Collins and Charles, 1987; Dimayuga et al., 1991). It is a common compound in Salvia species and was first isolated from S. carnosa (Dougl.) (White and Jenkins, 1942). Since then, this compound has been detected in Rosmarinus officinalis and many other species belonging to the Lamiaceae family (Wu et al., 1982). A mixture of ursolic acid and its isomer, oleanolic acid, was previously isolated from Swertia arisanensis (Gentianaceae) (Lin et al., 1987) and the two compounds have been found in more than 120 plant species including the Lamiaceae family. Medicinal plants containing ursolic acid have been used in folk medicine before it was known which constituents were responsible for their therapeutic effectiveness (Liu, 1995). Contemporary scientific research that led to the isolation and identification of ursolic acid revealed and confirmed that several pharmacological properties such as antitumour, hepatoprotective, anti-inflammatory, anti-ulcer, antimicrobial, antihyperlipidemic and antiviral activity can be attributed to this compound (Liu, 1995). Numerous plants from the Lamiaceae family containing ursolic acid and its derivatives, exhibited antimicrobial 146 activity and these therapeutic effects have been confirmed by laboratory testing (Sattar et al., 1995). Ursolic acid was also identified as one of the active principles in Rosmarinus officinalis that inhibited the growth of some food-borne pathogens (Collins and Charles, 1987). The isolated compounds have long been found in Salvia and other plants around the world. However, this is the first investigation reporting their presence in S. chamelaeagnea. A major characteristic of traditional medicine that has remained constant is the use of plant mixtures to cure various diseases (Iwu, 1994). The mixtures of the active compounds in producing desired biological effects are common practice in health care and clinical treatment (Liu, 2005). Most of these phytomedicines are on the drug market as whole plants extracts. Synergistic interactions between the compounds of individual or concoctions of various plants are a vital part of the therapeutic efficacy of the traditional practitioners. The combination study involving S. chamelaeagnea and L. leonurus displayed synergistic or antagonistic interactions depending on the test organism (Table 9.2; Figure 9.5). Boik (2001) conducted a large number of combination studies using various natural substances and the results strongly suggest that when used in combination, natural substances can produce synergistic interactions in vitro. Phenolic compounds such as flavonoids may have a role to play in increasing the biological activity of other compounds by synergistic interactions (Williamson, 2001). The synergistic interactions obtained with the Gram-positive bacteria with nearly all ratios (Table 9.2; Figure 9.5) indicated that the combination of natural products might be an alternative approach in the development of new anti-infective drugs. However, one should note that antagonistic actions could also occur, as can be seen with the Gram-negative bacteria. Research on chemical mixtures has been difficult but it is important to fully understand the use of natural occurring compounds in the treatment of various diseases (Liu, 2005). The antimicrobial activity of plants used alone or in combinations may involve distinct active principles with distinctive mechanisms of action (Okoli and Iroegbu, 2004). Although the investigations into the biological activities of chemical mixtures are difficult, it is important to fully understand the mechanisms by which these naturally occurring compounds are effective and interact in the treatment of many diseases (Liu, 2005). 147 9.5 Conclusions ? The solvent extracts and the essential oils displayed varying degrees of activity against bacteria, with the solvent extracts being more active than essential oils. ? Salvia dolomitica, S. radula and S. verbenaca exhibited the best activity against M. tuberculosis, while S. chamelaeagnea, S. muirii, S. runcinata and S. stenophylla displayed the most promising activity against the two Gram-positive and two Gram-negative bacteria. ? The synergistic interaction between S. chamelaeagnea and L. leonurus against S. aureus and B. cereus, indiates that the concurrent use of these two plants may broaden the therapeautic indications for Salvia species. ? Bioassay-guided fractionation led to the isolation of four compounds with antibacterial activity from S. chamelaeagnea. ? Carnosol, oleaonolic acid, ursolic acid and 7-O-methylepirosmanol inhibited the growth of micro-organisms to varying degrees. ? As carnosol, ursolic acid and oleanolic acid are known to exhibit good anti-oxidant activity, these compounds may prove a useful substitute for some of the phenolic anti-oxidants in current use, whilst offering an improved inhibitory capacity to microbial growth. 148 Chapter 10: In Vitro Anticancer Activity of the Solvent Extracts Abstract Salvia species are used in traditional medicine to treat various diseases including cancer. As part of an ongoing effort to verify the traditional uses of South African medicinal aromatic plants, 17 Salvia species were tested for in vitro anticancer activity. The experiment was conducted on three human cell lines, including the breast adenocarcinoma (MCF-7), the colon adenocarcinoma (HT-29) and the glioblastoma (SF-268) cell lines, using the sulforhodamine B assay. The results showed that the extracts inhibited cell proliferation of all three cell lines to varying degrees. The concentration required to inhibit 50% of cell growth (IC50 values) ranged between 9.69 and 43.65 ?g/ml and between 8.72 and 59.12 ?g/ml against the MCF-7 and SF-268 cell lines, respectively, with S. radula and S. africana-caerulea being the most active. IC50 values against the HT-29 cell line ranged from 17.05 to 57.00 ?g/ml, with the extract from S. lanceolata being the most active. Cell line specificity was observed for S. dolomitica and S. garipensis, while S. lanceolata, S. muirii, S. namaensis, S. repens, S. runcinata and S. verbenaca showed some degree of cell- type selectivity. In the search for new anticancer agents, plants that display selective activity against specific cell lines should be further investigated. 149 10.1 Introduction Cancer is caused by the uncontrolled growth and spread of cells that may affect almost any tissue of the body (WHO, 2006b). The ability of ionising radiation to provoke the development of cancer has long been recognized, and the direct absorption of energy by DNA, resulting in damage to this molecule, may be part of the basis for radiation-induced carcinogenesis. There is extensive, indirect evidence supporting the role of free radicals in the promotion of cancer, as many free radical-generating compounds are tumour promotors (Kehrer, 1993). Cancer is a worldwide public health problem, with more than 11 million people being diagnosed every year. It is estimated that by 2020, there will be 16 million new cases annually. It is reported that cancer causes 7 million deaths each year and results in 12.5% of deaths worldwide (WHO, 2006b). Cancer produces no signs or symptoms that exclusively indicates the presence of the disease. Complaints such as haemorrhoids, anaemia, bronchitis, sinusitis, weight loss and fever may be associated with cancer. Some common symptoms of cancer include a persistent cough or blood-tinged saliva, a change in bowel movements, blood in the stool or urine, unexplained anaemia, breast lump or discharge, change in urination (frequent urination, a small amount of urine and/or weak urine flow), night sweats and unexpected weight loss (WHO, 2006b). A major problem with the available cancer chemotherapy is the serious deficiency of active drugs for curative therapy and the resistance developed by many tumours to treatment with standard anticancer agents (Cragg and Newman, 2005) since the majority of cancer chemotherapeutic agents severely affect the ?host?s? normal cells (Mascarenhas, 1994). Plants have a long history of use in the treatment of cancer. In his review, Hartwell (1982) lists more than 3,000 plant species that have reportedly been used in the treatment of cancer, but in many instances the ?cancer? is undefined or reference is made to conditions such as ?hard swelli ng?, abscesses, calluses, corns, to name a few (Cragg and Newman, 2005). Such symptoms would generally apply to skin or visible conditions, but many of the claims for efficacy should be viewed with some skepticism because cancer, as a specific disease entity, is likely to be poorly defined in terms of folklore and traditional medicine (Cragg and Newman, 2005; Steenkamp and Gouws, 2006). Despite these observations, plants have played an important role as a source of effective anticancer 150 agents and it is significant that over 60% of currently used anticancer agents are derived in one way or another from natural sources including plants, marine organisms and micro- organisms (Cragg et al., 1997; Valeriote et al., 2002). The first agents to advance into clinical use were the so-called vinca alkaloids, vinblastine and vincristine, isolated from the Madagascar periwinkle, Catharanthus roseus (Apocynaceae). Others compounds, such as paclitaxel (Taxol? ), topotecan and irinotecan have been isolated. A number of promising new agents from plants including flavopiridol and combretastin are in clinical development based on selective activity against cancer-related molecular targets (Cragg and Newman, 2005). In South Africa, reports on plants used for the treatment of cancer are rare, and can be ascribed to the fact that cancer involves a complex set of signs and symptoms (Steenkamp and Gouws, 2006). With this in mind, it is recommended that when investigating plants for anticancer potential, ethnopharmacological properties such as immune and skin disorders and inflammatory diseases be taken into account when selecting plants used to treat cancer, since these could reflect disease states bearing relevance to cancer or a symptom thereof (Cordell et al., 1991; Popoca et al., 1998). Plants have been a prime source of highly effective conventional drugs for the treatment of many forms of cancer and, while the isolated compounds from the plants may not often serve as the drugs, they may provide leads for the development of potential novel agents (Cragg and Newman 2005). Salvia species have been used against various infectious and inflammatory diseases. Furthermore, they have been used in traditional medicines in China, South Africa and many other countries to treat symptoms associated with cancer (Watt and Breyer-Brandwijk, 1962; Ulubelen et al., 1999; Shoemaker et al., 2005). There has been substantial progress in the development of anticancer drugs, but a number of common cancers appear to be resistant to the available drugs (e.g. Tamoxifen and AndriamycinTM) (Salmon, 1984; Shoemaker et al., 2005). Furthermore, compounds were selected primarily on their gross cytotoxicity rather than on their selective ability to inhibit the growth of tumour cells (Valeriote et al., 2002). Another reason for the search for new anticancer drugs is the fact that current drugs have serious adverse effects. The objectives of this study were to: (i) evaluate the in vitro anticancer activity of the solvent extracts (Section 3.2.2) of Salvia species against three human cell lines using the sulforhodamine B assay, 151 (ii) assess any possible cell line specificity, and (iii) support the use of indigenous Salvia species used in traditional medicine as anticancer agents. 10.2 Materials and methods 10.2.1 Chemicals and drugs RPMI-1640 medium, Dulbecco?s Modified Ea gle?s Medium (DMEM), 0.25% trypsin- 0.1% ethylene diamine tetra acetic acid (EDTA), fetal bovine serum (FBS), sodium pyruvate and 10 mg/ml penicillin G-sodium/streptomycin sulphate (100 x) were from Highveld Biological; sulforhodamine B (SRB), gentamicin sulfate and trypan blue were obtained from Sigma? . Trichloro-acetic acid (TCA) and DMSO were purchased from Saarchem. Tris (hydroxymethyl aminomethane) and the reference drug, 5?-fluorouracil, were purchased from Merck. 10.2.2 Preparation of media and cell maintenance 10.2.2.1 Preparation of different media and solutions The routine culture medium used to maintain the SF-268 and MCF-7 cells was prepared using 470 ml of RPMI-1640, 25 ml of FBS and 5 ml of L-glutamine, while the experimental medium was prepared in the same manner with the addition of 500 ?l of 0.1% gentamicin sulphate to prevent any contamination. The DMEM used to maintain the HT-29 cell line was prepared with 13.53 g DMEM and 3.7 g NaHCO3 in one liter of distilled Millipore? water (further sterilised). Penicillin G-sodium/streptomycin sulphate (1 ml), sodium pyruvate (2 ml) and FBS (25 ml) were added to DMEM (470 ml) to constitute the routine culture medium. The SRB solution was prepared at a concentration of 0.4% (w/v) in 1% acetic acid and the TCA solution made at a concentration of 50% (w/v) with distilled Millipore? water. Trypan blue was prepared at a concentration of 2 mg/ml in PBS (Section 8.2.2.2) and the solution of Tris (hydroxymethyl) aminomethane was prepared at a concentration of 10 mM (pH 10.5) with distilled Millipore ? water. All solutions were stored at 4 ?C until required. 152 10.2.2.2 Source and maintenance of cell lines Cells lines representing the most common human cancers (WHO, 2006b) were obtained from the National Cancer Institute (NCI). The cell lines include the breast adenocarcinoma (MCF-7), the glioblastoma (SF-268) and the colon adenocarcinoma (HT-29). They were chosen for the following reasons: (i) adaptability to growth in RPMI-1640 or DMEM plus FBS with glutamine, (ii) suitability for use in microculture, and (iii) mass doubling time that allows for harvesting of approximately 1.5 x 105 cells/ml every week. The three cell lines were cultured in 200 ml flasks (NunclonTM) with the complete culture medium (RPMI-1640 or DMEM). The cells were maintained at 37 ?C in a 5% CO 2 humidified incubator. The culture medium (20 ml) was replaced three times weekly and the cells were trypsinised when nearly confluent. 10.2.2.3 Preparation of cell suspension All the cells were detached from the culture flasks by addition of 1 ml of 0.25% trypsin- 0.1% EDTA. Trypsin was then inactivated by the addition of 10 ml of experimental medium (RPMI-1640 or DMEM). Cells were separated into a single cell suspension by a gentle pipetting action and 1 ml of the cell suspension was returned to the culture flask containing 20 ml of culture medium. The remaining cell suspension was centrifuged at 1000 rpm (Sorvall? T6000D) for 3 minutes and the supernatant discarded. Experimental medium (20 ml) was added to the cells which were then repeatedly pipetted to allow cells to disperse in to the medium. An aliquot of the trypsinised cell suspension was stained with 2 mg/ml trypan blue (1:1 ratio) in order to obtain a more quantitative analysis of the cell suspension. Trypan blue is a water-soluble dye, which is insoluble in the cell membrane lipids. It will thus only cross cell membranes of dead/non-viable cells. On visualisation with the light microscope (Nikon), viable cells appear transparent, while non- viable cells are stained dark blue. The haemocytometer was used to determine the total number of viable and non-viable cells by counting cells in the 25 squares (each square is subdivided into 16 smaller squares of 0.1 mm2) at the top and bottom of the haemocytometer and the average number of cells per unit volume (millimeter, ml) of medium calculated. The cell suspension was adjusted with experimental medium to 153 approximately 1.5 x 105 cells/ml. The experiment was carried out only when there was at least 95% cell viability. 10.2.2.4 Preparation of test samples and standard The stock solution (10 mg/ml) of each test sample (solvent extract and standard) was prepared with DMSO and a further 1:50 dilution was made with appropriate experimental medium. The final DMSO concentration in the well of 1% had no effect on the growth of the cells. Stock samples were not filtered because DMSO has a sterilizing effect. Stock solutions were stored at -20 ?C to provide consistency between tests. The essential oils were not tested against cancer cell lines due to the small amount of the oils available. 10.2.3 The sulforhodamine B assay The colourimetric SRB microculture assay was performed according to Monks et al. (1991) and Wu et al. (1993). A volume of 100 ?l cell suspension containing 150,000 cells/ml was seeded into a 96-well microtitre plate (final count 15,000 cells/well), except the blank and the test sample background colour (only sample plus experimental medium) wells where 100 ?l of appropriate experimental medium was plated (Figure 10.1, rows A and H). The effect of each concentration of the solvent extract on the cell line was performed in triplicate wells. For instance, sample 1 corresponds to wells B3-D3, while well A3 was used as its corresponding background colour control which contained test sample and experiemental medium (no cells) (Figure 10.1). Wells, which contained cells with experimental medium in the absence of the test samples (Figure 10.1, B1-G1; B2-G2), were used as drug-free controls, while blank wells (A1, A2, H1 and H2) contained experimental medium only. The plate was incubated for 24 hours at 37 ?C, 5% CO 2 and 100% relative humidity to allow for cell attachment prior to the addition of the test samples. Thereafter, 100 ?l of the different sample dilutions were added to the appropriate microtitre wells, resulting in the required final sample concentration. The plate was incubated for a further 48 hours. At the time of sample addition, a separate reference plate for cell growth at time zero (the time at which drugs were added) was also prepared as described above. 154 For each dilution, a control was also plated out (in order to take into consideration the absorbance resulting from the colour of extract with experimental medium alone: rows A and H (Figure 10.1). The final concentration in the wells ranged from 0.34 to 100 ?g/ml for 5?-fluorouracil (5?-FU) and from 6.125 to 100 ?g/ml for the solvent extracts. At the end of the 48 hour incubation period, cells were fixed in situ by layering 50 ?l of ice-cold TCA (50%, w/v) onto the medium in each well and incubating the plate for one hour at 4 ?C. The supernatant was then discar ded and the plate was rinsed at least five times with tap water to remove the TCA, growth medium and low-molecular weight metabolites, before air drying the plate at room temperature. Figure 10.1 Microtitre plate showing how extracts were plated out in the anticancer assay and colour change of the SRB dye as a result of cell death. Each sample had five dilutions plated out in triplicate, plus one control well for measuring the colour of the sample in the presence of the experimental medium. The arrow indicates the direction of increasing concentration of the extract: L: low and H: high. SRB is a water-soluble dye that binds to basic amino acids of cellular proteins that are only synthesised by viable cells (Voigt, 2005). The colour intensity in each well thus corresponds to the number of viable cells and inhibitory effect of the test compounds 155 (Figure 10.1). TCA-fixed cells were stained with 100 ?l of SRB solution (0.4%, w/v). After 10-20 minutes at room temperature, the excess SRB solution was discarded and the cells washed thoroughly with 1% acetic acid to remove unbound dye. The plate was then air-dried at room temperature for approximately 2 hours until no moisture was visible. Bound dye was solubilised with 200 ?l of Tris (hydroxymethyl) aminomethane (10 mM). The SRB solution has a maximum absorbance at 492 nm. The absorbance was read against a Tris (hydroxymethyl) aminomethane blank on an automated spectrophotometer plate reader (Labsystems iEMS reader MF) connected to the Ascent? version 2.4 software, at a single wavelength after the plate was shaken at 960 rpm for 3 minutes. 10.3 Data analysis The percentage cell viability (in reference to control growth) to controls of the various extracts was calculated using the Microsoft Excel? programme (Eq. 10.1). The IC50 values (concentration at which 50% of cells were killed) were calculated from the log sigmoid- dose response curve using Enzfitter? version 1.05 software. The IC50 values were determined only for those extracts that displayed at least 80% inhibition when tested at 100 ?g/ml. At least three independent runs were performed for each sample. One-way analysis of variance (ANOVA) and Student t-tests were used to compare data using Statistica version 5.0 software at a 95% confidence limit. Percentage cell viability was calculated as follows: % Cell viability = Abs test sample ? Ab s background sample ? Mean Abs blank x 100 (Eq. 10.1) Mean Abs control ? Mean Abs blank Where: Abs: absorbance at 492 nm 10.4 Results The plant extracts exhibited anticancer activity against the HT-29, MCF-7 and SF-268 cell lines with the inhibitory effect increased as the concentration of the solvent extract increased (Figure 10.2). This pattern was observed for all solvent extracts against all the cell lines tested. 156 6.25 12.5 25 50 100 0 20 40 60 80 100 S. africana-caerulea (SF-268) S. lanceolata (HT-29) S. radula (MCF-7) Concentration (?g/ml ) % C el l v ia bi li ty Figure 10.2 Percentage cell viability of cancer cells tested at various concentrations for three Salvia species (mean ? s.d., n = 3): S. africana-caerulea (SF-268), S. lanceolata (HT-29) and S. radula (MCF-7). The inhibition of tumour cell proliferation by the solvent extracts of the 17 indigenous Salvia species investigated are displayed in Table 10.1. All the investigated solvent extracts were significantly less potent than 5?-FU (P < 0.05). The best activity against the HT-29 cell line was obtained with the extracts of S. lanceolata and S. stenophylla (IC50 value: 17.05 ? 3.50 ?g/ml) and S. africana-lutea (IC50 value: 20.00 ? 3.41 ?g/ml), S. aurita (IC50 value: 24.58 ?6.41 ?g/ml) and S. namaensis (IC50 value: 24.39 ? 3.42 ?g/ml) which showed no significant difference in their activity (P > 0.05). Five species, viz S. africana-caerulea, S. albicaulis, S. aurita, S. chamelaeagnea, S. disermas and S. namaensis, did not differ significantly from one another in the inhibitory effects (P > 0.05). S. radula exhibited the most favourable activity (IC50 value: 9.69 ? 0.92 ?g/ml) (P < 0.05), while S. africana-lutea and S. runcinata were amongst the least active against the MCF-7 cells (Table 10.1). ANOVA analysis showed no significant difference in the activity exhibited by S. aurita, S. chamelaeagnea, S. schlechteri and S. verbenaca against the MCF-7 cells (P > 0.05). Similarly, the activ ity against the MCF-7 cell line did not differ for the following species: S. africana-lutea, S. albicaulis, S. disermas, S. dolomitica, S. garipensis, S. muirii, S. namaensis, S. runcinata and S. verbenaca (P > 0.05). On average, all of the 17 species showed an IC50 value of less than 44 ?g/ml for the MCF-7 cells, divided as follows: four species, viz. S. aurita, S. chamelaeagnea, S. radula and S. 157 schlechteri with IC50 values less than 20 ?g/ml; four species, viz S. africana-caerulea, S. lanceolata, S. repens and S. stenophylla had IC50 values ranging from 20 to 30 ?g/ml, while the IC50 values of the remaining species were greater than 30 ?g/ml. Two species, S. dolomitica and S. garipensis, also showed a degree of cell selectivity, as they were not active against the HT-29 (IC50 value > 100 ?g/ml) and SF-268 (IC 50 value > 100 ?g/ml), but active against the MCF-7 cell line (IC50 value: 39 ? 4 ?g/ml). None of the extracts were as active as 5?-FU (P < 0.05). Table 10.1 The anticancer activity (IC50 values) of the solvent extracts of Salvia species and reference compound obtained against three human cancer cell lines. HT-29 MCF-7 SF-268 Species / Reference IC50 values in ?g/ml S. africana-caerulea 27.10 ? 1.08 23.36 ? 4.20 8.72 ? 1.52 S. africana-lutea 20.00 ? 3.41 43.65 ? 8.38 nc S. albicaulis 25.33 ? 1.11 35.25 ? 5.66 27.50 ? 0.89 S. aurita 24.58 ? 6.41 17.28 ? 1.71 44.87 ? 2.48 S. chamelaeagnea 29.53 ? 5.14 18.12 ? 1.67 34.98 ? 4.07 S. disermas 26.87 ? 0.57 38.56 ? 1.28 59.12 ? 2.75 S. dolomitica nc 37.05 ? 2.10 nc S. garipensis nc 39.44 ? 3.90 nc S. lanceolata 17.05 ? 3.50 26.15 ? 2.15 nc S. muirii 55.63 ? 1.30 39.07 ? 2.86 nc S. namaensis 24.39 ? 3.42 36.36 ? 3.07 nc S. radula 32.10 ? 2.93 9.69 ? 0.92 27.55 ? 4.52 S. repens 43.62 ? 5.17 23.36 ? 2.16 nc S. runcinata 55.37 ? 2.10 43.42 ? 5.46 nc S. schlechteri 57.00 ? 11.67 18.37 ? 0.47 54.40 ? 4.42 S. stenophylla 17.41 ? 2.65 23.74 ? 1.96 43.86 ? 3.43 S. verbenaca 50.04 ? 5.39 31.50 ?13.70 nc 5?-FU 7.00 ? 2.02 1.11 ? 0.31 nc nc: IC50 values not calculated because the percentage inhibition at 100 ?g/ml was less than 80% 158 In contrast, S. africana-caerulea, with an IC50 value of 8.72 ? 1.52 ?g/ml, was the most active extract against the SF-268 cells (P < 0.05). The inhibition by S. africana-lutea, S. dolomitica, S. garipensis, S. lanceolata, S. muirii, S. namaensis, S. repens, S. runcinata and S. verbenaca tested at a concentration of 100 ?g/ml exhibited less than 80% inhibition of the cell growth. The IC50 values were thus not calculated for these species (Table 10.1). The IC50 values of nearly all the species were above 30 ?g/ml. No significant difference in the activity of S. albicaulis and S. radula against the SF-268 cells (P > 0.05) was observed. 5?-Fluorouracil was not active againt against the SF-268 cell line, while S. africana- caerulea, S. albicaulis and S. radula exhibited some degree of activity (IC50 value < 30 ?g/ml) (Table 10.1). Treatments with Salvia extracts resulted in growth inhibition in the majority of cases. For the three cell lines examined, MCF-7 cells were the most sensitive, as they wewre inhibited by all the solvent extracts, while the SF-268 cell line was the least sensitive. About 53% (9/17) of samples exhibited less than 80% inhibition at 100 ?g/ml (Table 10.1). In the screening of the anticancer activity of plants, species which need to be considered for further analysis are those that show activity against certain cell lines and are not toxic to others. In this study, cell line selectivity was observed. The solvent extracts of S. dolomitica, S. garipensis, S. lanceolata, S. muirii, S. namaensis, S. repens, S. runcinata and S. verbenaca showed some degree of activity against MCF-7, but were, however, not active against the SF-268 and/or HT-29 cell lines (Table 10.1). Ursolic acid and oleanolic acid were isolated from S. chamelaeagnea as an inseparable mixture (Chapter 9). Ursolic acid (standard) exhibited low activity (IC50 value > 100 ?g/ml), while its isomer, oleanolic acid (standard), displayed moderate activity against the MCF-7 cells (IC50 value > 72.77 ? 5.23 ?g/ml). Carnosol, also isolated from S. chamelaeagnea, inhibited HT-29 cells with an IC50 value of 11.79 ? 0.94 ?g/ml. This activity was, however lower than that of 5?-FU (IC 50 value: 7.00 ? 2.02 ?g/ml). Salvigenin, a compound isolated from S. radula (Chapter 8), exhibited moderate activity against the MCF-7 cells (IC50 value: 67.78 ? 3.78?g/ml). 159 10.5 Discussion Some phytochemicals such as flavonoids and alkaloids, present in medicinal plants, are reported to possess substantial anti-oxidant, antibacterial, anti-inflammatory and anticarcinogenic activities. These phytochemicals have been explored extensively for their potential in the treatment of cancer. Well-known compounds like paclitaxel, a diterpenoid from Taxus brevifolia, and vincristine, an alkaloid from Catharanthus roseus, are used in chemotherapy today (Wall and Wani, 1996). Many Salvia species have been used in traditional medicine in South Africa, in Chinese herbal medicine, as well as in many other countries for the treatment of cancer (Watt and Breyer-Brandwijk, 1962; Ryu et al., 1997). The in vitro anticancer activity of indigenous South African species has, however, not yet been fully investigated. The organic extracts of indigenous Salvia species were shown to have inhibitory effects on cancer cells in an apparent dose-dependent manner (Figure 10.2). In this study, there may be compounds in Salvia extracts that may induce a cytotoxic action against various cancer cell lines and initiate cell death. The American National Cancer Institute guidelines set the limit of activity for crude extracts at a 50% inhibition (IC50 value) of proliferation of less than 30 ?g/ml (Suffiness and Pezzuto, 1990). The extracts exhibited various responses depending on the cell line used. The percentage of plant extracts with IC50 values less than 30 ?g/ml was 17%, 47% and 53% against the SF-268, MCF-7 and HT-29 cells, respectively. S. africana-caerulea was the only species to exhibit anticancer activity against all three cell lines with an IC50 value of less than 30 ?g/ml (Table 10.1). The extract of S. hypargeia has been investigated by Ulubelen et al. (1999) against a panel of cell lines, including the human colon cells (COL-2), breast cancer cells (BC-1) and lung cancer cells (LU-1). Results indicated that the plant extracts possess anticancer activity (IC50 values < 20 ?g/ml), which is lower than values obtained in this study, when tested against the breast cancer cell line. Other members of the family Lamiaceae have also been shown to inhibit the in vitro or in vivo growth of cancer cells. Ye et al. (2002) found that Scutellaria baicalensis (Lamiaceae) inhibited the growth of the most common human cancers, including breast cancer (MCF-7), colon cancer (HCT-15) and hepato-cellular carcinoma (HepG-2), with IC50 values of 1.1, 0.9 and 1.5 mg/ml, respectively, when tested in vivo in rats. Like most members of the Lamiaceae family, Salvia species are aromatic in 160 nature and are rich in essential oils (Chapter 4). Foray et al. (1999) investigated the in vitro anticancer activity of essential oils of S. lavandulifolia, S. officinalis and S. sclarea against various cancer cell lines and recorded IC50 values ranged from 0.6 x 10-4 ?g/ml to 4.4 x 10-1 ?g/ml. This higher cytotoxicity of the essential oil is due to the fact that several essential oil compounds have highly hydrophobic character. Their accumulation in the cellular membrane perturbs the membrane, resulting in increased permeability to protons and ions and eventually cell-death (Sikkema et al., 1995). Compounds with anticancer properties have been isolated from many Salvia species and include cryptotanshinone, methylenetanshiquinone and tanshindiol A from S. miltiorrhiza (Ryu et al., 1997). These compounds were tested against the non-small cell lung (A549) and central nervous system cell lines. The IC50 values obtained less than 0.7 ?g/ml, but were still less potent than the reference compound used in the study (doxorubicin, IC50 value < 0.1 ?g/ml). Some of the plants investigated in this study have been studied from a chemical point of view. Phytochemical research revealed that the principal secondary metabolites of Salvia species are mainly found to be flavonoids which are responsible for the antibacterial, anti- oxidant, anti-inflammatory, antimalarial and anticancer activities (Ulubelen et al., 1999; Kobayashi et al., 2002). Terpenoids have been isolated from many Salvia species, including this study (Chapter 9) and this class of compounds probably contributes to the anticancer activity exhibited by the extracts. Ursolic acid and oleanolic acid isolated from S. chamelaeagnea are shown to have marked anti-tumour effects, with the activity of ursolic acid being greater than that of oleanolic acid against the human colon carcinoma cell line LCT-15 (Li et al., 2002). S. chamelaeagnea was not, however, the most active extract. In this study, it was found that ursolic acid (standard) exhibited low activity (IC50 value > 100 ?g/ml), while its isomer, oleanol ic acid, displayed moderate activity (IC50 value: 72.77 ? 5.23 ?g/ml) against the MCF-7 cell line. This is in agreement with the study conducted by Cragg and Newman (2005), which showed that ursolic and oleanolic acids are associated with weak anti-inflammatory and antitumour activities. Carnosol, also isolated from S. chamelaeagnea (Chapter 9), inhibited the HT-29 colon adenocarcinoma cells with an IC50 value of 11.79 ? 0.97 ?g/ml. All th e isolated compounds tested in this study were significantly less active (P < 0.05) than 5?-FU. Oleanolic acid, ursolic acid acid 161 and carnosol are good anti-oxidant (Chapter 5, Cuvelier et al., 1994) and may also offer an improved inhibitory capacity to cancer cells. Flavonoids are biosynthesised by plants and have strong anti-oxidant activity for scavenging free radicals which are involved in cell damage and tumour promotion. Epidemiological studies suggest that flavonoids, such as the isoflavone genistein, play an important role in the prevention of carcinogenesis (Barnes, 1995). Indigenous Salvia species also possess flavonoids (Chapter 5) and one flavonoid has been isolated from S. radula as part of this research (Chapter 8). The presence of flavonoids may also contribute to the anticancer activity. The flavonoid isolated from S. radula (salvigenin) was tested against the MCF-7 cells and it exhibited moderate activity, although not significantly comparable to the activity of 5?-FU (IC 50 value: 67.78 ? 3.78 ?g/ml) . Testing could not be done against the HT-29 and SF-268 cells due to the small amount of flavonoid isolated. This is the first in vitro investigation to determine the anticancer activity on indigenous Salvia species. The investigated species have shown their ability to inhibit the proliferation of three human cancer cells. Furthermore, the selectivity of some of the solvent extracts indicated that Salvia species could be considered in the search for novel anticancer lead compounds. 10.6 Conclusions ? Salvia species inhibited the growth of three different cancer cell lines in a dependent manner. ? The best activity against the HT-29 cells was obtained with extracts obtained from S. africana-lutea, S. aurita, S. lanceolata, S. namaensis and S. stenophylla. ? The best activity against the MCF-7 and SF-268 was obtained with extracts from S. radula and S. africana-caerulea, respectively. ? Solvent extracts of S. africana-lutea, S. dolomitica, S. garipensis, S. lanceolata, S. muirii, S. namaensis, S. repens, S. runcinata and S. verbenaca were active against the MCF-7 cell line, but not active against the SF-268 and/or the HT-29 cell line. ? None of the solvent extracts were as effective as 5?-fluorourac il against MCF-7 and HT-29. The 5?-fluorouracil di d not inhibit SF-268 cells, but S. africana-caerulea, S. albicaulis and S. radula inhibited it with an IC50 value of less than 30 ?g/ml 162 (NCI cut off) and other species were more active against the cell line that is resistant to the effects of 5?-fluorouracil. ? The MCF-7 cell line was the most sensitive cell line, with 80% of cell growth being inhibited by a concentration of 100 ?g/ml of the solvent extracts. ? The SF-268 cell line was the least sensitive cell line as it was inhibited by only 47% of the extracts when tested at 100 ?g/ml. ? This study intends to further investigate Salvia species in the search for anticancer compounds and could help to explain their use in traditional medicine as chemotherapeutic or anticancer agents. 163 Chapter 11: In Vitro Evaluation of Toxic Effects of the Essential Oils and Solvent Extracts Abstract The in vitro toxicity profile of 28 samples (17 solvent extracts and 11 essential oils) was evaluated on human kidney epithelial cells using the 3-(4,5-dimethylthiazol-2-yl)-2,5 dimethyl tetrazolium bromide (MTT) colourimetric method. The samples displayed some degree of toxicity with IC50 values ranging from 1.79 ? 0.43 to 22.9 ? 5.28 ?g/ml for the essential oils and from 12.12 ? 2.02 to 53.34 ? 3.90 ?g/ml for the solvent extracts. The toxicity profile of the essential oils was significantly higher compared to the solvent extracts (P < 0.05). The essential oil of S. runcinata (IC50 value: 1.79 ? 0.43 ?g/ml) and the solvent extract of S. stenophylla (IC50 value: 12.12 ? 2.02 ?g/ml) were the most toxic, while the S. disermas extract (IC50 value: 53.34 ? 3.90 ?g/ml) and S. radula essential oil (IC50 value: 22.90 ? 5.28 ?g/ml) were least toxic. The essential oils and solvent extracts were significantly more toxic than 5?-fluorouracil (P < 0.05). 164 11.1 Introduction In recent years, there has been a worldwide trend towards the use of natural products as opposed to synthetic ones for medicinal purposes. Research and identification of active constituents are important in the search for new biologically active compounds in the pharmaceutical field. Many species of the Lamiaceae have potential therapeutic activities, which include anti-oxidant, anti-inflammatory, anticholinesterase, antibacterial and antimalarial activities (Cuvelier et al., 1996; Ulubelen et al., 2001; Hayes and Markovic, 2002; Perry et al., 2003; Benoit-Vical et al., 2003). Salvia species are also used extensively in perfumery and aromatherapy (Baylac and Racine, 2003; Kaya et al., 2003). The common notion among people who use Salvia to treat various conditions is that ? if it does not benefit, it would not harm? (Gali-Muhtasib et al., 2000) and many people are not aware that excessive intake of sage extracts can be toxic and can lead to convulsions and permanent brain damage (Arnol, 1988). The increased worldwide use of plants in aromatherapy and pharmaceutical fields has raised a number of concerns in relation to their adverse effects, which warrants further investigation (Woolf, 1999; Thompson and Wilkinson, 2000). Poisoning by traditional remedies, especially by plants is well documented, particularly in South Africa (Joubert, 1990). But only a few plants used in traditional medicine contain pharmaco-active chemicals which are toxic. However, these make a significant contribution to morbidity and mortality (Steenkamp, 2000). In order to verify whether the biological activities (especially the anticancer activity) exhibited by plants are not correlated to their general cytotoxicity, the evaluation of their toxic effects against a normal cell line is necessary. Plant material contains a large number of detectable and potentially active compounds, but only a few are responsible for the major pharmacological or adverse effects. Such compounds may be expressed in the plant at different levels depending on the location, soil in which it is growing and the season (Steenkamp, 2000). In addition, the toxic principle(s) may be located in only one part of the plant (e.g. leaves, seeds, roots or bark). Toxic compounds of plants are divided into various groups according to their chemical properties (e.g. alkaloids and glycosides) with some Salvia species having tested positive for alkaloids (Raffauf, 1996). Several studies on biological activities of Salvia species have been published (Cuvelier et al., 1994; Ulubelen et al., 2001; Baricevic et al., 2001; Tepe et al., 2006), but only a few have been conducted on the toxicity of these plant 165 species and may therefore place limitations on their use in conventional medicine. In phytomedicine as well as in clinical medicine, the toxicity profile of a given substance must always be evaluated, although plants found to be toxic in vitro may not necessarily be toxic in vivo because of the complexity of the human body. The lack of information on the toxic profile of Salvia species and the pharmacological activities exhibited in previous Chapters prompted this investigation in order to determine whether they are safe for use as unconventional medicines. The objectives of this chapter were to: (i) evaluate the toxicity profile of the essential oils and the solvent extracts using the MTT assay, and (ii) verify the safety of Salvia species in traditional medicine. 11.2 Materials and methods 11.2.1 Chemicals, reagents and reference compound Trypsin containing EDTA, fetal calf serum (FCS) and Ham F10 were obtained from Highveld Biological, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) was purchased from USB? . NaHCO3, gentamicin sulfate, trypan blue and 5?-fluorouracil were obtained from Sigma? . 11.2.2 Preparation of various media and solutions 11.2.2.1 Culture medium and experimental medium The Ham F10 solution consisted of 9.38 g of Ham F10 medium and 1.18 g of NaHCO3 in a total volume of 1 liter of Millipore? water and the solution was then filtered through a Sterrivex-GS 0.22 ?m filter unit. The culture medium consisted of Ham F10 solution, gentamicin sulfate (10 mg/ml in PBS) and FCS in a proportion of 96:1:5, respectively. The experimental medium was prepared in the same way as the culture medium, except that gentamicin sulfate was not included. 166 11.2.2.2 Preparation of 3-[4,5-dimethylthi azol-2-yl]-2,5-diphenyltetrazolium bromide solution The MTT solution was prepared at a concentration of 50 mg/ml with the PBS (Section 8.2.2.2), filtered and stored at 4 ?C until used. 11.2.3 Cell maintenance Human kidney epithelial cells were maintained continuously in a 500 ml flask (NunclonTM) with 50 ml of culture medium. The medium was changed thrice weekly or when it turned yellow. The cells were maintained at 37 ?C in a 5% CO 2 humidified incubator and were trypsinised when nearly confluent. 11.2.4 Cell suspension and sample preparations 11.2.4.1 Single cell suspension The cells were detached from the flask using 4 ml of trypsin (0.25%) containing EDTA (0.1%) after which 6 ml of the experimental medium was added to neutralize the effects of the trysin. The solution was gently agitated with a pipette until all the cells were dispersed into a single cell suspension. Subsequently, 1 ml of the cell suspension was returned to the culture flask with 49 ml culture medium. The remaining cell suspension was centrifuged at 1500 rpm for 5 minutes (Sorvall? T6000D), the supernatant-trypsin was discarded and 20 ml experimental medium added to the cells which were again agitated with a pipette until a single cell suspension was obtained. The cell density was determined using the trypan blue exclusion assay (Section 10.2.2.3), with 20 ?l of cell suspension being stained with an equal volume trypan blue. The cells were counted using the haemocytometer and adjusted with the experimental medium to prepare a working solution of approximately 0.25 million cells/ml. 11.2.4.2 Preparation of samples The stock solution of the solvent extracts and the 5?-FU were prepared with DMSO at a concentration of 10 mg/ml and stored at -20 ?C. The dilutions were prepared with experimental medium on the day of the experiment and the concentrations of the dilutions used were dependent on the results obtained from the initial broad screening. The toxicity profile of the isolated compounds from S. radula (salvigenin and betulafolientriol oxide) 167 and S. chamelaeagnea (carnosol); as well as the commercial standards of ursolic and oleanolic acid were also tested. These samples were prepared in the same as the solvent extracts. The stock solution of each essential oil was prepared at a concentration of 10 mg/ml with DMSO and the dilutions were also prepared with DMSO on the day of the experiment. The final concentration of DMSO (1%) in the wells had no effect on the normal growth of cells. Seven dilutions were considered for each sample. The final concentration in the wells ranged from 1 to 100 ?g/ml for the solvent extracts and 1 to 50 ?g/ml for the essential oils. Concentrations of the reference compound varied from 1 to 500 ?g/ml. Because essential oils are volatiles, they were always tested on separate microtitre plates and placed in different incubators to the solvent extracts. 11.2.4.3 The 3-[4,5-dimethylthiazol-2-yl] -2,5-diphenyltetrazolium bromide assay The MTT colourimetric assay was performed on human kidney epithelial cells according to the method of Mosmann (1983) taking into consideration modifications by van Zyl and Viljoen (2002). For this assay, a volume of 180 ?l of the cell suspension was plated out into a 96-well plate (Nunclon?), except for two wells which serv ed as the blank controls to which 180 ?l of experimental medium were added. In order to test the effect of the solvent on the cells, two wells containing 1% DMSO were also included. After six hours of incubation (to allow for cell attachment), 20 ?l for each dilution for the solvent extracts and reference compound or 2 ?l of the essential oils, were added in triplicate to a 96-well culture plate. The experimental medium (18 ?l) was added to the wells containing essential oils to ensure a total volume of 20 ?l in each well. In the drug-free control wells (those only containing 180 ?l of cells), 20 ?l of experimental medium was added. The plate was then incubated in a humidified environment at 37 ?C and 5% CO 2 for 44 hours. The validity of the MTT assay depends on the assumption that only viable cells reduce tetrazolium salts to coloured formazan which can be quantified spectrophotometrically (Mosmann, 1983). After 44 hours of incubation, 40 ?l of the MTT solution (50 mg/ml) was added to each well and the plate incubated for a further four hours. Thereafter, 200 ?l of the contents of each well was removed and discarded. Subsequently, 150 ?l of DMSO 168 was added to each well to stop the reaction and dissolve the formazan crystals. The plate was then shaken at 1020 rpm for four minutes and the absorbance was read on an automated spectrophotometer plate reader (Labsystems iEMS reader MF) connected to the Ascent? version 2.4 software at the test wavelength of 540 nm and the reference wavelength of 690 nm. 11.3 Data analysis The percentage cell viability was determined using a simple programme written in Microsoft Excel? by comparing the absorbance between the test samples and the controls (Eq. 11.1). % Cell viability = Abs of test sample ? (Mean Abs of control ? Mean Abs of blank) x 100 (Eq. 11.1) Mean Abs of controls ? Mean Abs of blank Where: Abs = absorbance Abs of test sample = Abs 540 of test sample ? Abs690 of test sample Abs of control = Abs 540 control ? Abs690 control Abs of blank = Abs 540 of blank ? Abs690 of blank The IC50 values (concentration at which 50% of cell viability is obtained) were determined using Enzfitter? version 1.05 software. At least three independent runs were performed for each sample and results are given as mean ? s.d. An ANOVA and Student t-tests were used to analyse results. A correlation between the anticancer activity and toxicity profile was assessed using Pearson?s correlation coe fficient. All statistical analyses were performed using Statistica? version 5.0 software at 95% confidence limits. The toxicity of each sample was compared to that of the 5?-FU (ra tio of toxicity of the 5?-FU to toxicity of the test sample). For the toxicity assay, the lower the IC50 values, the more toxic the sample. 11.4 Results The toxicity profile of the essential oils, solvent extracts and the reference compound are depicted in Table 11.1. The essential oils and solvent extracts showed diverse levels of toxicity. The IC50 values ranged from 1.79 ? 0.43 to 22.90 ? 5.28 ?g/ml for the essential oils. Of all the plants screened, the toxicity profile of the essential oil of S. africana- 169 caerulea, S. runcinata and S. stenophylla were similar (P > 0.05) and were also the most toxic (P < 0.05). Table 11.1 The toxicity profile of the essential oils, solvent extracts and reference compound. Species/Reference compound Essential oils (IC50 values in ?g/ml) Solvent extract (IC50 values in ?g /ml) Safety Index SI1 SI2 S. africana-caerulea 1.87 ? 0.52 14.38 ? 3.39 72.75 9.46 S. africana-lutea 7.24 ? 0.67 25.01 ? 1.94 18.79 5.44 S. albicaulis 2.75 ? 0.72 37.29 ? 0.58 49.47 3.64 S. aurita nd 28.31 ? 4.94 nd 4.80 S. chamelaeagnea 6.00 ? 0.74 24.76 ? 7.60 22.67 5.49 S. disermas nd 53.34 ? 3.90 nd 2.55 S. dolomitica 7.74 ? 0.27 40.26 ? 9.07 17.57 3.37 S. garipensis nd 42.44 ? 3.03 nd 3.20 S. lanceolata 3.66 ? 0.92 26.71 ? 6.41 37.17 5.09 S. muirii 3.09 ? 0.53 37.00 ? 5.41 44.03 3.68 S. namaensis nd 21.91 ? 2.89 nd 6.21 S. radula* 22.90 ? 5.28 20.12 ? 4.02 5.94 6.76 S. repens 6.66 ? 0.90 23.24 ? 4.72 20.42 8.85 S. runcinata 1.79 ? 0.43 22.00 ? 3.75 76.01 6.18 S. schlechteri nd 28.16 ? 4.69 nd 4.83 S. stenophylla 1.98 ? 0.47 12.12 ? 2.02 68.71 11.22 S. verbenaca nd 20.85 ? 2.59 nd 6.53 5?-FU 136.06 ? 16.60 nd: not determined, essentialoil not available * no significant difference between the oil and the solvent extract (P > 0.05) SI1: toxicity ratio of 5?-FU over essential oil sample SI2: toxicity ratio of the 5?-F U over solvent extract sample The essential oil of S. radula displayed the lowest toxicity profile (P < 0.05) (Table 11.1). An ANOVA showed no variation in the toxicity profile of S. albicaulis, S. lanceolata and 170 S. muirii (P > 0.05). Similarly, S. africana-lutea, S. dolomitica, S. repens and S. chamelaeagnea exhibited the same toxicity profile (P > 0.05). The toxicity of the solvent extracts ranged from 12.12 ? 2.02 to 53.34 ? 3.90 ?g/ml. The Student t-tests showed that the solvent extracts of S. africana-caerulea and S. stenophylla were similar (P > 0.05) and were the most toxic, while S. disermas was the least toxic relative to the other extracts (P < 0.05). An ANOVA showed that the solvent extracts of ten species namely S. africana-lutea, S. aurita, S. chamelaeagnea, S. lanceolata, S. namaensis, S. radula, S. repens, S. runcinata, S. schlechteri and S. verbenaca was not statistically different from each other (P > 0.05). Nearly all the essential oils were significantly more toxic than their corresponding solvent extracts (P < 0.05) (Figure 11.1). However, S. radula showed no statistical difference between the toxicity profile of its oil and solvent extract even though the toxicity of the extract was slightly higher than that of the essential oil (Figure 11.1; Table 11.1). S. afr ica na -ca eru lea S. afr ica na -lu tea S. alb ica uli s S. ch am ela ea gn ea S. do lom itic a S. lan ce ola ta S. mu iri i S. ra du la S. rep en s S. ru nc ina ta S. ste no ph yll a 0 10 20 30 40 50 Essentials oils Solvent extracts T ox ic it y pr of il e (I C 5 0 va lu es in ?g /m l) Figure 11.1 Variation between the toxicity profile of the essential oil and solvent extracts of eleven Salvia species (mean ? s.d., n = 3). The toxicity of the essential oils and solvent extracts was compared to that of 5?-FU (safety index; SI1 and SI2, respectively) (Table 11.1). The standard 5?-FU was significantly less toxic compared to the essential oils and solvent extracts (P < 0.05). The solvent extracts were 3 to 11 times more toxic than the reference compound, whilst the essential oils were 6 171 to 76 times more toxic (Table 11.1). More precisely, the essential oil of S. runcinata, the most toxic essential oil (IC50 value: 1.79 ? 0.43 ?g/ml) was 76 times more toxic than 5?-FU and the oil of S. radula (IC50 value: 22.90 ? 0.43 ?g/ml) seven times more toxic than 5?- FU. A poor correlation was observed between the anticancer activity and the toxicity profile of the solvent extracts (r = 0.37 and 0.07 against the MCF-7 and HT-29 cell lines, respectively). It may be assumed that compounds responsible for the inhibition or death of cancer cells are not necessarily those involved in the toxicity of the plants. This can also be seen by the IC50 value difference of 5?-FU which is about 20 times more effective in inhibiting the HT-29 cancer cells. The isolated compounds, namely betulafolientriol oxide and salvigenin from S. radula (Chapter 8), carnosol from S. chamelaeagnea (Chapter 9) and standards of ursolic and oleanolic acids, exhibited poor toxicity against the human kidney epithelial cells with all IC50 values greater than 100 ?g/ml. 11.5 Discussion Plants used in traditional medicine are generally non-toxic at the administered dosage. However, it is known that all plants are toxic at certain concentrations (Liu, 2005). The essential oils and the solvent extracts displayed some degree of toxicity, with the toxicity of the essential oils being greater than that of the solvent extracts (Table 11.1). This higher toxicity of the essential oil is due to the fact that several essential oil compounds have highly hydrophobic in nature. Their accumulation in the cellular membrane disrupts the membrane, resulting in increased permeability and eventually cell death (Sikkema et al., 1995). Salvia species are not known to be toxic (Gali-Muhtasib et al., 2000). However, most of the tests done on the toxicity of these species used in vivo models (rats) rather than in vitro models, as is the case in the present investigation. The solvent extracts were generally less toxic than the essential oils in this study. In traditional medicine, aqueous extracts are used and a small amount of volatile fractions (containing generally non-polar compounds) would probably be extracted in hot water extracts. However, if cold water is used, essential oils will not be extracted. Furthermore, the percentage of essential oil 172 present in the plant is very low (Figure 4.2) and this will probably have little influence on the decoction of plant extracts as prepard traditionally. There are various types of toxic compounds in plants and these include tannins, glycosides and alkaloids. The identification of tannins in Salvia species has not yet been reported in the literature, but alkaloids and glycosides are generally found in the genus (Lu and Foo, 2002). Many alkaloids are poisonous to both animals and humans. Some of the species investigated, such as S. chamelaeagnea, S. namaensis and S. runcinata, tested positive for alkaloids (Raffauf, 1996). The presence of these compounds was not confirmed, however if present, they may contribute to the toxicity of the solvent extracts. The essential oils displayed greater toxicity than the solvent extracts (Table 11.1) and the toxicity of the essential oils is well reported in the literature (van Zyl and Viljoen, 2003; Nyiligira, 2004; van Vuuren et al., 2006). Foray et al. (1999) investigated the in vitro cytotoxicity of three essential oils of Salvia lavandulifolia, S. officinalis and S. sclarea against various cell lines. All three essential oils were toxic with IC50 values ranging from 6.3 x 10-5 to 4.4 x 10-1 ?g/ml. The toxicity of the essential oils found in this study is higher than that found in other oils such as Helichrysum picardi (IC50 value: 30 ?g/ml) (De La Puerta et al., 1993). However, these results are comparable to, and occasionally lower than essential oils from S. lavandulifolia, S. officinalis and S. sclarea (IC50 value < 0.44 ?g/ml), cinnamon (IC50 value: 3.75 ?g/ml) and clove (IC50 value: 7.31 ?g/ml) (De La Puerta et al., 1993). It is difficult to deduce the constituent(s) of an individual oil that may contribute to its toxicity. In a study of 20 essential oil constituents, van Zyl et al. (2006) demonstrated that limonene, nerolidol, p-cymene, ?-thujene, ?-thujene, ?-terpinene and ?-bisabolol displayed some degree of toxicity (IC50 value ? 10 ?g/ml) in comparison to chloroquine diphosphate and quinine sulphate (two antimalarial drugs) and the 5?-FU tested in this study. Nerolidol was also found to be toxic to kidney cells (van Zyl and Viljoen, 2003). In addition, camphor is known to be highly toxic if used in prolonged treatment or when ingested in small amounts. A survey of literature conducted by Burkhard et al. (1999) showed that the essential oils of 11 plants, including Salvia were powerful anticonvulsants due to their terpenic and ketone contents. The toxicity associated with the Salvia essential oils may be due to the presence of the above compounds, even if present in relatively small amounts. The essential oil of S. runcinata was most toxic probably due to the high amount of ?-bisabolol (65%, Table 4.1). Some compounds are present in relatively high amounts: 173 ?-thujene (4.7% in S. repens); camphor (8.3% in S. muirii) and p-cymene (9.5% in S. repens). It is also possible that other compounds present even in small amount may act in a synergistic or additive manner to increase the toxicity of the essential oils (van Zyl, pers comm). Although the anticancer activity and the toxicity profile measures cell viability or inhibition, a poor correlation between the anticancer activity and the toxicity of the extracts was observed. It was, however, interesting to note that with S. lanceolata, S. radula and S. stenophylla, which displayed anticancer activity, also exhibited some degree of toxicity. In these cases, it could be speculated that the same compound(s) may be responsible for the anticancer activity and the ?gener al? toxicity profile of the solvent extracts. The isolated compounds from S. radula (betulafolientriol oxide and salvigenin, Chapter 8) and S. chamelaeagnea (carnosol, Chapter 9), standards of ursolic acid and oleanolic acid demonstrated poor toxicity towards the human kidney epithelial cells with IC50 values greater than 100 ?g/ml. Thus, the activity of these compounds cannot be correlated to the overall cytotoxic profile of the Salvia extracts. 11.6 Conclusions ? The essential oils and solvent extracts displayed some degree of toxicity (compared to the anticancer drug, 5?-fl uorouracil) with the toxicity of the essential oils being generally greater than that of the solvent extracts. ? The compounds responsible for the anticancer activity are not necessarily those responsible for the toxicity profile of the solvent extracts of Salvia. ? Salvia species are used in traditional medicine as acqueous extracts, and water will extract negligeable amount of essential oils which showed some degree of toxicity. 174 Chapter 12: Seasonal Variation in Essential Oil Composition and Biological Activities of Three Salvia Species Abstract Aromatic plants contain both volatile and non-volatile fractions and the chemical composition of these two fractions may be influenced by seasonal changes. The essential oil composition of S. africana-caerulea, S. africana-lutea and S. lanceolata, collected at the same locality throughout the 2004/2005 growing season, was compared in terms of essential oil yields, chemical composition and the biological activities. Mostly quantitative, rather than qualitative variation was observed in the essential oil composition of each species. Major fluctuations in S. africana-caerulea oil composition include limonene (1.5-32.6%) and viridiflorol (1.6-22.8%). Levels of ?-pinene (0.5-11.9%), myrcene (2.2-11.5%) and ?-eudesmol (0.1-12.9%) fluctuated seasonally in S. africana- lutea oil. In S. lanceolata, considerable changes were noted for ?-caryophyllene (trace- 12.9%), ?-caryophyllene oxide (1.4-21.2%) and ledol (trace-11.7%). S. lanceolata harvested in winter was more active in winter against Gram-positive bacteria. S. africana- caerulea exhibited the most favourable antimalarial activity when harvested in winter (IC50 value: 11.61 ?g/ml) which contrasts with the lowest antimalarial activity of S. lanceolata obtained at the same period (IC50 value: 42.60 ?g/ml). The anti-oxidant activity of the solvent extracts also displayed variation over seasons with winter collection giving the most favourable anti-oxidant activity with S. africana-lutea (IC50 value: 10.05 ?g/ml). All the solvent extracts prepared from samples collected in winter exhibited the lowest toxicity profile (29.89 < IC50 value < 60.29 ?g/ml), while the three essential oils obtained from autumn collection were more toxic (0.03 < IC50 value < 0.4 ?g/ml). 175 12.1 Introduction Biological activities are correlated to the presence of chemical compounds (particularly secondary metabolites). The presence of these compounds may assist in predicting some traditional uses of medicinal plants (Rasoanaivo and Ratsimamanga-Urverg, 1993). Many plants concentrate certain secondary metabolites in specific organs resulting in the variation in biological activities between different parts of the same plant (e.g. barks, roots, leaves, stems) (Houghton, 1999). The season and even the number of hours plants receive sunlight per day, may influence the phytochemistry of the plant because some compounds may be accumulated at a particular period to respond to environmental changes (Koenen, 2001). A collection of plant material made at different times of the year may contain other possibly novel compounds with certain bio-activities (Eloff, 1999). This suggests that the biological activities should be evaluated at different times in order to determine the period of the year a plant exhibits its best activity. A quick and effective way to assess variations in the chemical composition of plant extracts is TLC (Wagner and Bladt, 1996). While TLC can also be used to evaluate the variation in chemical composition of the essential oils, GC and GC-MS are preferred because they are not just qualitative (as with the TLC method), but also quantitative. Chemical data obtained by TLC, GC and GC-MS are important in addition to morphological features in taxonomic studies as they re-enforce the relationships of plants at a different level (McGaw et al., 2002). The objectives of this study were to: (i) determine the monthly variation in the essential oil composition of three randomly selected species, (ii) create finger-prints using TLC for the three species, which could generate information on the chemical constituents of the solvent extract of each plant over the seasons, (iii) evaluate the antibacterial, antimalarial and anti-oxidant activity of the solvent extracts over the seasons, investigate the toxicity profile of the essential oils and solvent extracts over seasons and determine if there is a correlation between the toxicity of the essential oils and their major compounds, and 176 (iv) recommend the best time of the year for harvesting of plants in terms of essential oil yields, biological activities and toxicity profile. 12.2. Materials and methods 12.2.1 Plant material Three randomly selected species of Salvia, namely Salvia africana-caerulea, S. africana- lutea and S. lanceolata, all collected from the South Western Cape (Table 2.1), were investigated. The aerial parts of each species were harvested at the end of each month for 12 consecutive months (November 2004 to October 2005) at the same location. 12.2.2 Isolation of the essential oils and preparation of the solvent extracts The essential oils were isolated by hydrodistillation and the solvent extracts prepared with methanol:chloroform (1:1) as previously described (Section 2.3.1 and Section 2.3.2, respectively). 12.2.2.1 Analysis of the essential oils The essential oils were analyzed by GC and GC-MS (Section 4.2 and Section 4.3, respectively). 12.2.2.2 TLC fingerprint To evaluate the possible impact of seasonal change on the phytochemistry, 5 ?l (2 mg/ml) of the methanol:chloroform extract was separated by TLC (Alugram? Sil G/UV254) and developed using ethyl acetate:toluene:acetic acid (10:5:0.01). The separated spots on TLC were examined under UV light (254 and 366 nm) and also sprayed with 0.5% anisaldehyde sulfuric acid reagent in glacial acetic acid, methanol, concentrated sulfuric acid (10:85:5) and heated at 105 ?C for five minutes. 12.2.3 Evaluation of the biological activities The biological activities of the solvent extracts evaluated include the antibacterial, antimalarial and anti-oxidant activities. The toxicity profile was also assessed. These 177 activities were determined at the end of each season of the year as follows: November (spring), February (summer), May (autumn) and August (winter). The results of the biological activities corresponding to the spring samples are those reported in previous Chapters. 12.2.3.1 The antibacterial activity The MIC values of the plant extracts over seasons was determined against Gram-positive B. cereus and S. aureus and Gram-negative E. coli and K. pneumoniae test organisms using the micro-dilution method (Section 9.2.4.1). 12.2.3.2 The antimalarial activity The antimalarial activity was assessed using the [3H]-hypoxanthine radiometric method (Section 8.2.5). 12.2.3.3 The anti-oxidant activity The effect of seasonal change on the anti-oxidant activity of the three species was evaluated using the DPPH method as previously described (Section 6.2.2.2). 12.2.3.4 The toxicity profile The effect of seasonal variations on the toxicity profile of the solvent extracts and essential oils was performed on human kidney epithelial cells using the MTT proliferation colourimetric method (Section 11.2.4.3). 12.3. Data analysis The identity of the components of each oil was assigned by comparison of their retention indices, relative to C8-C17 n-alkanes and GC-MS spectra with corresponding data of components of reference oils, laboratory-synthesized components and commercially available standards from a home-made library. Comparison of biological activities between seasons was determined using ANOVA. The IC50 values are given as a mean ? s.d. of at least three replicate experiments. Pearson?s correlation co efficient was used to determine any correlation between the toxicity profile and major components of each essential oil. 178 12.4 Results 12.4.1 Variation in yield and essential oil composition The essential oil yield of each species varied with the respective seasons. The highest yield was obtained from plants collected in late winter for S. africana-caerulea (August), spring for S. africana-lutea and S. lanceolata (October and September, respectively) (Figure 12.1). All three species demonstrated the lowest essential oil yield in early or mid- winter (July for S. africana-caerulea and S. africana-lutea and June for S. lanceolata). 0 0.05 0.1 0.15 0.2 0.25 No ve mb er De ce mb er J a nu ar y Fe br ua ry M ar ch Ap ril M ay Jun e Ju ly Au gu st Se pt em be r Oc tob er % O il yi el d S. africana-caerulea S. africana-lutea S. lanceolata Figure 12.1 Variation in essential oil yields of three Salvia species over a one year period. The components of essential oil found in each species at different seasons are those reported in Table 4.1. However, quantitative variation in the components of individual oil was noted (Appendix F). Only components of the oils greater than 4% of the total oil at least in one month are reported here. Major fluctuations in components of S. africana- caerulea includes ?-pinene (trace-8.2%), limonene (1.50-32.6%), terpinen-4-ol (0.50- 8.3%), trans-calamenene (0.2-8.9%) and viridiflorol (1.60-22.8%) (Figure 12.2). In the November sample of S. africana-caerulea, spathulenol and ?-caryophyllene oxide were present in high amounts but were totally absent during the rest of the year. The components reported represent up to 60% of the total oil in January and only 20% of the total oil in December (Figure 12.2). The identified compounds of S. africana-lutea over a one year period are displayed in Appendix F (Table F-1). 179 Jan ua ry Fe bru ary M arc h Ap ril Ma y Ju ne Ju ly Au gu st Se pte mb er Oc tob er No ve mb er De ce mb er 0 10 20 30 40 50 60 70 80 90 100 Limonene Viridiflorol ?-Pinene Ledol Terpinen-4-ol Myrcene trans Calamenene Spathulenol ?-Phellandrene ?-Caryophyllene oxide OthersR el at iv e % o il c om po si ti on Figure 12.2 Variation of major components of Salvia africana-caerulea essential oil over a one year period. The major variations in S. africana-lutea essential oil includes ?-pinene (0.5-11.9%), myrcene (2.2-11.5%), o-cymene (0.4-10.3%), spathulenol (1.9-10.9%) and ?-eudesmol (0.1-12.90) (Figure 12.3). These five components of S. africana-lutea oil represent up to 45% of the total oil in May and just below 20% in November (Figure 12.3). The identified compounds of S. africana-lutea over a one year period are displayed in Appendix F (Table F2). Jan ua ry Fe bru ary M arc h Ap ril M ay Ju ne Ju ly Au gu st Se pte mb er Oc tob er No ve mb er De ce mb er 0 10 20 30 40 50 60 70 80 90 100 ?-Pinene Myrcene ?-Eudesmol ?-Eudesmol ?-Gurjunene Spathulenol ?-3-Carene 1,8-Cineole p-Cymene o-Cymene cis-?-Ocimene ?-Phellandrene ?-Cadinene Limonene OthersR el at iv e % o il c om po si ti on Figure 12.3 Variation of major components of Salvia africana-lutea essential oil over a one year period. 180 The major components in S. lanceolata showing considerable changes include ?- caryophyllene (trace-12.90%), ?-humulene (0.1-10.0%), spathulenol (0.7-10.6), ?- caryophyllene oxide (1.4-21.2%) and ledol (0.9-11.70) (Figure 12.4). The components of S. lanceolata oil reported here represent close to 50% of the total oil in April and only 15% in December (Figure 12.4). The identified compounds of S. africana-lutea over a one year period are displayed in Appendix F (Table F-3). Jan ua ry M ar ch Ap ril M ay J u ne Jul y Au gu st Se pt em be r Oc tob er No ve mb er De ce mb er 0 10 20 30 40 50 60 70 80 90 100 ?-Caryophyllene oxide ?-Humulene ?-Caryophyllene Ledol ?-Humulene Bicyclogermacrene Myrcene Spathulenol ?-Eudesmol ?-Gurjunene Geranyl acetate Globulol Others R el at iv e % o il c om po si ti on Figure 12.4 Variation of major components of Salvia lanceolata essential oil over a one year period. 12.4.2 TLC finger print of the solvent extracts over seasons Like the essential oil composition, the TLC chromatograms of the solvent extracts showed a similar range of compounds for each extract prepared from plant material collected in different seasons of the year (Figure 12.5). The sample collected in the spring, however showed less compounds. Some compounds which cannot be seen in the spring sample include A, B and C (S. africana-caerulea), compound D (S. africana-lutea) and compound E and F (S. lanceolata). 181 S. africana-caerulea S. africana-lutea S. lanceolata Figure 12.5 TLC chromatograms of the solvent extracts of three Salvia species collected at various seasons of the year: compounds A, B, C, D, E and F are those absent in the samples collected during spring. 12.4.3 Seasonal variation in biological activities 12.4.3.1 The antibacterial activity The MIC values of each plant across seasons are given in Figure 12.6. The best activity against Gram-negative bacteria was obtained from the spring sample (November), with the exception of S. africana-caerulea against K. pneumoniae. With the Gram-positive bacteria, the best activity was exhibited against B. cereus from the sample harvested in spring and S. aureus from the collection done in autumn, respectively for S. africana-lutea. The solvent extract of the winter collection of S. lanceolata was more active (S. aureus) and the same activity was obtained against B. cereus in the collection done in summer and winter. S. africana-caerulea exhibited its best activity against B. cereus in winter and S. aureus in summer (Figure 12.6). Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn Winter Spring 182 Summer Autumn Winter Spring 0.0 2.5 5.0 7.5 10.0 E. coli K. pneumoniae B. cereus S. aureus S. africana-caerulea M IC v al ue s (m g/ m l) Summer Autumn Winter Spring 0.0 2.5 5.0 7.5 10.0 S. africana-lutea M IC v al ue s (m g/ m l) Summer Autumn Winter Spring 0 . 0 2 . 5 5 . 0 7 . 5 1 0 . 0 1 2 . 5 1 5 . 0 1 7 . 5 2 0 . 0 S. lanceolata M IC v al ue s (m g/ m l) Figure 12.6 Antibacterial activity of the solvent extracts of three Salvia species obtained across four seasons of the year (n = 3). 12.4.3.2 The antimalarial activity of the solvent extracts The variations in antimalarial activity of Salvia africana-caerulea, S. africana-lutea and S. lanceolata as influenced by the seasons are displayed in Figure 12.7. An ANOVA showed significant variation between the antimalarial activities of samples collected at different seasons (P < 0.05). The multiple comparison tests (Tukey test) showed that the activity of S. africana-caerulea from the sample collected in winter was the most favourable (IC50 183 value: 11. 61 ? 4.53 ?g/ml), whilst the autumn sample exhibited the lowest activity (IC 50 value: 26.41 ? 3.17 ?g/ml) (Figure 12.7). The activity of S. africana-lutea also varied over seasons with the spring sample giving the most favourable activity (IC50 value: 15.86 ? 5.04 ?g/ml) and winter the lowest activity (IC50 value: 32.86 ? 2.01 ?g/m l). In contrast to S. africana-caerulea, which exhibited its best activity with sample collected in winter, the lowest activity of S. lanceolata was obtained with plant harvested in winter (IC50 value: 42.60 ? 6.34 ?g/ml), while no variation was obser ved between the antimalarial activity of S. lanceolata in spring, summer and autumn (P > 0.05) (Figure 12.7). Summer Autumn Winter Spring 0 10 20 30 40 50 S. africana-caeruleaS. africana-lutea S. lanceolata A nt im al ar ia l a ct iv it y (I C 5 0 va lu es in ? g /m l) Figure 12.7 Antimalarial activity of three Salvia species obtained at different seasons of the year (mean ? s.d., n = 3). 12.4.3.3 The anti-oxidant activity of the solvent extracts The anti-oxidant activity of the three species over four seasons is presented in Figure 12.8. An ANOVA showed significant variation in the anti-oxidant activity of each species across seasons (P < 0.05). The multiple comparison tests (Tukey test) showed that the activity of S. africana-caerulea obtained in autumn was statistically higher compared to the activity obtained in other seasons (P < 0.05). S. africana-lutea exhibited its best activity with the extract harvested in winter, which was statistically higher than that of other seasons (P < 0.05). In contrast, the best activity of S. lanceolata was obtained in summer (P < 0.05) with the spring extract yielding the lowest activity (Figure 12.8). 184 Summer Autumn Winter Spring 0 20 40 60 80 S. africana-caerulea S. africana-lutea S. lanceolata A nt i- ox id an t ac ti it y (I C 5 0 va lu es in ? g /m l) Figure 12.8 Anti-oxidant activity of three Salvia species obtained at different seasons of the year (mean ? s.d., n = 3). 12.4.3.4 Seasonal variation in the toxicity profile of the solvent extracts and their essential oils The toxicity of the essential oils and solvent extracts was evaluated against human kidney epithelial cells. The analysis of variance showed variation in the toxicity over the four seasons, both for the solvent extracts and essential oils (P < 0.05) (Figure 12.9 and 12.10). The solvent extracts exhibited the lowest toxicity for all three species with sample harvested in winter (higher IC50 values) (Figure 12.9), while the highest toxicity profile (lower IC50 values) was obtained with spring sample for S. africana-caerulea and summer samples for S. africana-lutea and S. lanceolata. Summer Autumn Winter Spring 0 20 40 60 80 S. africana-caerulea S. africana-lutea S. lanceolata T ox ic it y pr of il e of t he so lv en t ex tr ac ts (I C 5 0 va lu es in ? g /m l) Figure 12.9 Toxicity profile of the solvent extracts of three South African Salvia species across seasons (mean ? s.d., n = 3). 185 Summer Autumn Winter Spring 0 2 4 6 8 S. africana-caerulea S. africana-lutea S. lanceolata T ox ic it y pr of il e of t he es se nt ia l o il s (I C 5 0 va lu es in ? g /m l) Figure 12.10 Toxicity profile of the essential oils of three South African Salvia species across seasons (mean ? s.d., n = 3). The oils samples of the three species collected during autumn were significantly more toxic compared to other seasons. The lowest toxicity of the essential oil was obtained with samples collected during spring for S. africana-caerulea and S. africana-lutea, and sample harvested in summer for S. lanceolata (Figure 12.10). 12.5 Discussion The essential oil yields varied greatly for the same species collected at different periods of the year (Figure 12.1). However, a continuous increase in the yield was noted for all species during summer (December to February, with exception of S. africana-caerulea). Pitarevic et al. (1984) recorded variation in essential oil yield of S. officinalis collected in Yugoslavia in various seasons, with July (flowering period) giving the highest yield. The essential oils yield varied from 0.07% (July) to 0.19% (August) for S. africana-caerulea. In S. africana-lutea, the highest yield (0.18%) was obtained in September while July gave the lowest yield (0.09%). Finally, in S. lanceolata, the lowest yield was obtained in June and July (0.03%) with September giving the highest yield (0.11%). The higher essential oil yield obtained in early spring, may be explained by the fact the plants remain in flower from August to January, a period of full vegetation. During this period, plants might produce substantial amounts of essential oils in order to attract more pollinators (Pal?-pa?l et al., 2001). The yield of S. africana-caerulea and S. africana-lutea decreased from May to July which corresponds to dry season and the period of latent vegetation. 186 Quantitative seasonal variation in chemical composition was found in the essential oil of each species and those showing remarkable fluctuations include ?-pinene, viridiflorol, limonene, terpinen-4-ol and trans-calamanene in S. africana-caerulea. Two compounds namely, ?-caryophyllene oxide and spathulenol were detected in the November sample of S. africana-caerulea; however were absent during the rest of the year (Figure 12.2). ?- Pinene, myrcene, o-cymene, spathulenol and ?-eudesmol showed fluctuations in S. africana-lutea between seasons (Figure 12.3), while ?-caryophyllene, ?-humulene, spathulenol, ?-caryophyllene oxide and ledol showed major variations in S. lanceolata (Figure 12.4). Although no pattern in the fluctuation of the components of an individual oil component was observed, the most conspicuous variation was recorded for ?- caryophyllene and ?-caryophyllene oxide in S. lanceolata. The variation of these two compounds was inversely related: a rise in the ?-caryophyllene oxide content was accompanied by a decrease in the content of ?-caryophyllene (Figure 12.11). The content of ?-caryophyllene oxide started to increase from January and reached its maximum in April when the content of ?-caryophyllene was at its lowest level (Figure 12.11). 0 5 10 15 20 25 No ve mb er De ce mb er Jan ua ry M ar ch Ap ril M ay Jun e July Au gu st Se pt em be r Oc tob er % R el at iv e am ou nt ?-Caryophyllene ?-Caryophylle oxide Figure 12.11 Variation in ?-caryophyllene and ?-caryophyllene oxide contents in S. lanceolata over seasons. Researchers have reported seasonal variation in essential oil yields and composition for many species. Pitarevic et al. (1984) also noted variation in essential oil composition of S. officinalis collected at various seasons of the year. The components of the oil with significant variation included 1,8-cineole (5.8-13.2%), ?-thujene (10.9-23.1%), ?-thujene 187 (25.0-40.1%) and camphor (7.1-16.4%). The qualitative variation in essential oil composition of S. libanotica was also demonstrated (Farhat et al., 2001). The major components of the oil including 1,8-cineole (47.7-57.4%) and camphor (7.7-12.3%) fluctuated from one season to another. The seasonal variation in the composition of the essential oil of Crithmum maritimum L. (Umbelliferae) also showed a great degree of variation in the essential oil composition of the main components, namely sabinene (7- 42%) and ?-terpinene (26-55%) (Barroso et al., 1992), but the essential oil yield of this species did not show remarkable change. This study has demonstrated that the essential oil yield and change in the amount of each component varied from one season to another. It is known that many factors affect the essential oil yield and composition such as season, temperature and reproductive stage (Putievsky et al., 1986). The biological activities of the same plant collected at different periods of the year showed significant variation, which could be attributed to changes in the amounts of the active compounds present in each plant in different months. Changes in chemical composition could be explained in terms of a thermoregulatory action of the hydrophobic compounds which could protect the plant from desiccation. Furthermore, polar solids (triterpenes, flavonoids, diterpene acids) present in the plants might act as a physical barrier to prevent water permeation and dehydration (Harbone et al., 1975). Many studies have reported differences in the biological activity of the same plant collected in different seasons. McGaw et al. (2002) investigated the variation in antibacterial activity of Schotia species and noted slight fluctuations. They also noted small variations in the chromatograms obtained from one season to another. Mar?chal et al. (2004) also demonstrated seasonal variation in antibacterial activity of solvent extracts of Bifurcaria bifurcata (Cystoseiraceae) with the highest level of activity obtained from spring to summer. In the current study, major fluctuations in the antibacterial activity of the three Salvia species investigated were observed. Furthermore, the variation obtained in the antibacterial activity depended on both the plant species and the test organism involved. Variation in the anti-oxidant activity and toxicity profile of the three species across seasons was exhibited. Farhat et al. (2001) investigated the essential oil of S. libanotica across four seasons of the year and found that the winter sample was the most toxic, while the spring sample was the least toxic. The toxicity of the solvent extracts of all three species was low during winter (Figure 12.9), while the three essential oils were 188 significantly (4 to 142 times) more toxic during autumn (Figure 12.10). An attempt to find a relationship between the amount of major components of each essential oil and the toxicity profile was not successful. Thus it can be assumed that the concentration of the essential oil components is not directly involved in the toxicity of the essential oils. To account for this, two possibilities can be proposed: either the toxicity profile was due to other minor components or there was a synergistic effect between the major components of the essential oils and other molecules. A negative correlation was observed between the toxicity and the major components [Pearson?s correlation coefficient r = -0.98, P < 0.05, (S. africana-lutea); r = - 0.97, P < 0.05, ( S. lanceolata)] denoting that an increase in the amounts of the major components of the essential oil are accompanied by an increase in the toxicity. However, S. africana-caerulea showed no significant correlation (r = 0.45, P > 0.05) between the toxicity of the oil and it s major components. This suggests that both major and minor components of the essential oil may interact to contribute to the toxicity of the oil. This investigation has demonstrated the importance of seasonal changes on biological activity and essential oil yield. 12.6 Conclusions ? Quatitative variation in the essential oil composition and yields exists over seasons and this also depends on the species studied. ? Variation in the chemical profile of the solvent extracts was observed on the TLC plates as for the various seasons. ? Seasonal/climatic changes play an important role in the biological activities and the toxicity of the plants. ? The best activity of the solvent extract of S. africana-caerulea was obtained in summer (Gram-positive), winter (antimalarial activity) and autumn (anti-oxidant activity). ? The best activity of S. africana-lutea was obtained in spring (antibacterial activity), spring (antimalarial activity) and winter (anti-oxidant activity). ? S. lanceolata exhibited the best activity in winter (against Gram-positive bacteria), spring (against Gram-negative bacteria), winter (antimalarial activity) and summer (anti-oxidant activity). 189 ? The solvent extracts of all three species exhibited the lowest toxicity in winter, while the essential oils of the three species were less toxic in spring. The toxicity of the essential oils is not always related to presence of the major components, but the influence of minor components to act synergistically with other components needs to be evaluated to understand the overall toxicity of the essential oils. ? It is difficult to determine which season will give the best activity because this varies with the species involved as well as the biological activity studied. However, it is important to use more plants in order to have a good pattern. 190 Chapter 13: The Influence of Essential Oils on the Biological Activities of Solvent Extracts Abstract The solvent extracts of aromatic plants contain both the volatile (essential oils) and non- volatile compounds. The biological activities of the solvent extracts of three Salvia species (Salvia africana-caerulea, S. africana-lutea and S. lanceolata) were evaluated in the presence and absence of essential oils. The biological activities assessed included the antibacterial, anti-oxidant, antimalarial and anticancer activity. In addition, the toxicity profile was also investigated. The solvent extract of S. africana-caerulea without essential oil exhibited the best activity against Gram-positive bacteria (MIC value: 0.1 mg/ml), while the solvent extract containing essential oil of S. africana-lutea was the most active against Gram-negative bacteria (MIC value: 3 mg/ml). No significant difference was obtained in the antimalarial activity for the two extracts of S. africana-caerulea and S. lanceolata, while the activity of the solvent extract without essential oil was significantly higher than that of the solvent extract containing essential oil in S. africana-lutea (P < 0.05). The toxicity profile of all three species was significantly higher (P < 0.05) with the solvent extracts containing essential oils. 191 13.1 Introduction Aromatic plants have traditionally been used in folk medicine as well as to extend the shelf life of foods. In the previous Chapters, it was observed that the essential oils and the solvent extracts of indigenous Salvia species exhibited a broad range of biological activities including anti-oxidant (Chapter 6), anti-inflammatory (Chapter 7), antimalarial (Chapter 8), antibacterial (Chapter 9) and anticancer (Chapter 10) activity. The essential oils and solvent extracts also displayed various levels of toxicity against normal cells with the toxicity of the essential oil being greater than that of the solvent extracts (Chapter 11). The solvent extracts contain both the essential oils (volatiles) and non-volatile compounds (e.g. phenolics). In order to investigate whether the removal of essential oils could reduce the toxicity of the solvent extract without affecting the biological properties, the investigation of the solvent extracts in the presence and absence of the essential oils was conducted. The objective of this study was to evaluate the antibacterial, anti-oxidant, anticancer and antimalarial activities of two solvent extracts (with and without essential oils) of three Salvia species and to compare the toxicity of these solvent extracts. 13.2 Materials and methods 13.2.1 Plant material Three randomly selected species namely Salvia africana-caerulea, S. africana-lutea and S. lanceolata, all collected from the South Western Cape (Table 2.1) were investigated. 13.2.2 Isolation of the essential oils and preparation of the solvent extracts The essential oils were isolated by hydrodistillation as previously described in Section 2.3.1. In order to investigate the effect of the essential oils on selected biological activities, the residue of the plant material in the Clevenger apparatus (with water) was air-dried at room temperature and then extracted as described in Section 2.3.2. To provide a better understanding of the terminology used in this Chapter, the plant collected from the field and air-dried is referred to ?as the solvent extracts containing essential oils or solvent extracts with essential oils?. The plant material remaining in the Clevenger apparatus, 192 which was then air-dried and also extracted with methanol:chloroform, are referred to as ?the solvent extracts without th e essential oils or solvent extracts in the absence of the essential oils?. The same plant material was used and three-quarters of fresh plant material of each species was hydrodistilled and the remaining one-quarter air-dried at room temperature to constitute the extracts without essential oils and the extracts containing essential oils, respectively. 13.2.3 Comparison of the TLC plates of the solvent extract with and without essential oils In order to verify whether the high temperature used to isolate the essential oils would affect the plant extract, a TLC plate representing the solvent extract with essential oil and solvent extract without essential oil were prepared. Hence, 5 ?l (2 mg/ml) of each methanol:chloroform extract from S. africana-caerulea, S. africana-lutea and S. lanceolata were separated by TLC (Alugram? Sil G/UV254) and developed using ethyl acetate:toluene:acetic acid (10:5:0.01). The separated spots were examined under UV light (254 and 366 nm) and also visualised by spraying the plates with 0.5% anisaldehyde sulfuric acid reagent in glacial acetic acid, methanol, concentrated sulfuric acid (10:85:5) and heated at 105 ?C for five minutes. 13.2.4 Evaluation of the biological activities The biological activities evaluated included the antibacterial, anticancer, antimalarial and anti-oxidant activities, as well as the toxicity profile being determined for the extracts. 13.2.4.1 The antibacterial activity The MIC values of the solvent extracts with and without the essential oils were determined against Gram-positive B. cereus and S. aureus and Gram-negative E. coli and K. pneumoniae test organisms using the micro-dilution method (Section 9.2.4.1) and compared to activity of the essential oils alone (Table 9.1). 193 13.2.4.2 The anticancer activity The anticancer activity of the solvent extracts in the presence and absence of the essential oils was determined using the SRB method (Section 10.2.3) against the HT-29 and MCF-7 cells. 13.2.4.3 The antimalarial activity The antimalarial activity of the solvent extracts with and without essential oils was assessed using the [3H]-hypoxanthine radiometric method (Section 8.2.5) and compared to activity of the essential oils alone (Table 8.1). 13.2.4.4 The anti-oxidant activity The effect of the essential oils on the anti-oxidant activity of the two solvent extracts was determined using the DPPH method as previously described (Section 6.2.2.2) and compared to activity of the essential oils alone (Table 6.1). 13.2.4.5 The toxicity profile The toxicity profile of the two solvent extracts was determined using the MTT proliferation colourimetric method on human kidney epithelial cells (Section 11.2.4.3) and compared to activity of the essential oils alone (Table 11.1). 13.3 Data analysis The IC50 values are given as a mean ? s.d. of th ree replicate experiments. The difference between the activity of the solvent extracts with and without essential oils was determined using the paired sample t-tests. All statistical analyses were performed using Statistica? version 5.0 software and P < 0.05 was considered significant. 13.4. Results 13.4.1 Comparison of the TLC plates of the solvent extracts with and without the essential oils There was not obvious variation in the chromatographic profiles in the solvent extracts of the three species studied in the presence or absence of the essential oils (Figure 13.1). 194 Similar compounds were noted and it appeared that the high temperature did not alter the composition of the phenolic compounds based on TLC plate results. SacW SacWT SalW SalWT SlaW SlaWT Figure 13.1 TLC chromatogram of the solvent extracts of three Salvia species obtained in the presence and absence of the essential oils. SacW: solvent extract of S. africana- caerulea with essential oils; SacWT: solvent extract of S. africana-caerulea without essential oils, SalW: solvent extract of S. africana-lutea with essential oils; SalWT: solvent extract of S. africana-lutea without essential oils; SlaW: solvent extract of S. lanceolata with essential oils; SlaWT: solvent extract of S. lanceolata without essential oils. 13.4.2 The antibacterial activity A comparison between the activity of the solvent extracts containing essential oils and the solvent extracts without essential oils is displayed in Figure 13.2. The best activity against Gram-positive bacteria for all three species was obtained with the solvent extracts without essential oils. In contrast, the best activity against Gram-negative bacteria was obtained with the solvent extracts containing essential oils with the exception of S. africana- caerulea against K. pneumoniae (Figure 13.2). 195 E. co li K. pn eu mo nia e B. ce reu s S. au reu s 0 8 16 24 32 Extracts containing essential oils Extracts in the absence of essential oils Essential oils S. africana-caerulea > 32 M IC v al ue s (m g/ m l) E. co li K. pn eu mo nia e B. ce reu s S. au reu s 0 8 16 24 32 > 32 M IC v al ue s (m g/ m l) S. africana-lutea E. co li K. pn eu mo nia e B. ce reu s S. au reu s 0 8 16 24 32 S. lanceolata > 32 M IC v al ue s (m g/ m l) Figure 13.2 Antibacterial activity of the solvent extracts with and without essential oils of three Salvia species (n = 3). 13.4.3 The anticancer activity The results of the anticancer activity of the two solvent extracts (with and without essential oils) against two cancer cell lines are displayed in Figure 13.3. A paired sample t-test showed no difference between the solvent extracts containing essential oil and the solvent extract without essential oil against the HT-29 cell line (P > 0.05). In S. africana-lutea, the solvent extract containing essential oil was significantly less active than the solvent extract 196 without essential oil against the MCF-7. In contrast, S. africana-caerulea exhibited the most favourable activity with the solvent extract containing essential oil, while in S. lanceolata there was no difference between the two solvent extracts against the MCF-7 cell line. S. afr ica na -ca eru lea S. afr ica na -lu tea S. lan ce ola ta 0 1 0 2 0 3 0 4 0 HT-29 Solvent extracts containingessential oils Solvent extracts without essential oils A nt ic an ce r ac ti vi ty (I C 5 0 va lu es in ? g /m l) S. afr ica na -ca eru lea S. afr ica na -lu tea S. lan ce ola ta 0 1 0 2 0 3 0 4 0 5 0 6 0 MCF-7 A nt ic an ce r ac ti vi ty (I C 5 0 va lu es in ? g /m l) Figure 13.3 Anticancer activity of the solvent extracts of three Salvia species evaluated in the presence and in the absence of the essential oils (mean ? s.d., n = 3). 197 13.4.4 The antimalarial activity The antimalarial activity of the two solvent extracts against the chloroquine-resistant P. falciparum strain is depicted in Figure 13.4. A paired sample t-test showed no variation in the antimalarial activity of S. africana-caerulea and S. lanceolata using the extracts with and without the essential oils (P > 0.05). The solvent extract of S. africana-lutea without essential oils exhibited higher activity (P < 0.05) compared to the solvent extract with essential oils (Figure 13.4). In all cases, the antimalarial activity of the solvent extracts (with and without essential oils) was significantly lower than the activity of the essential oils (Figure 13.4). S. afr ica na -ca eru lea S. afr ica na -lu tea S. lan ce ola ta 0 10 20 30 40 Solvent extract containing essential oils Solvent extracts without essential oils Essential oils A nt im al ar ia l a ct iv it y (I C 5 0 va lu es in ? g /m l) Figure 13.4 Variation in the antimalarial activity of the solvent extracts with and without their respective essential oils of three Salvia species (mean ? s.d., n = 3). 13.4.5 The anti-oxidant activity The anti-oxidant activity of the two solvent extracts is depicted in Figure 13.5. A paired sample t-test showed that the activity of the solvent extracts containing essential oils was significantly lower (P < 0.05) compared to the activity of the solvent extracts without essential oils for all three species. This suggests that the essential oils contributed little to the anti-oxidant activity of the solvent extracts. 198 S. afr ica na -ca eu lea S. afr ica na -lu tea S. lan ce ola ta 0 20 40 60 80 100 120 Solvent extract containing essential oils Solvent extract in the absence of essential oils Essential oils > 100 > 100 > 100 A nt i- ox id an t ac ti vi ty IC 5 0 va lu es ( ? g /m l) Figure 13.5 Anti-oxidant activity of the solvent extracts with and without their respective essential oils of three Salvia species (mean ? s.d., n = 3). 13.4.6 The toxicity profile The outcome of the toxicity profile of the two solvent extracts is displayed in Figure 13.6. A paired sample t-test indicated that the toxicity of the solvent extracts was significantly lower in the absence of the essential oils (P < 0.05) implying that the essential oils have contributed significantly to the toxicity of the solvent extracts of these three species. The toxicity of the essential oil was always higher than that of the two solvent extracts (with and without essential oils). In order to determine whether there is a direct correlation between the toxicity of the oil and solvent extract of the same plant, a regression analysis was performed. A poor relationship (r2 = 0.25) between the activit y of the essential oil and the solvent extract of the same plant was observed. This suggests that the high toxicity of an essential oil does not imply that its solvent extract is also highly toxic. 199 S. afr ica na -ca eru lea S. afr ica na -lu tea S. lan ce ola ta 0 15 30 45 60 75 Solvent extracts containing essential oils Solvent extract without essential oils Essential oils T ox ic it y pr of il e (I C 5 0 va lu es i n ?g /m l) Figure 13.6 Toxicity profile of the solvent extracts of three Salvia species with and without essential oils (mean ? s.d., n = 3). 13.5 Discussion There is a variation in the in vitro biological activities of the solvent extracts when tested with and without the essential oils. The essential oils seem to have beneficial or adverse effects depending on the biological activities implicated. In this study, it was found that the essential oils do not have a negative influence on the activity of the solvent extracts against Gram-positive bacteria, while they increase the activity of the solvent extracts against Gram-negative bacteria. The essential oils also slightly increase the activity of the solvent extracts against cancer cells. The toxicity of the solvent extracts was significantly decreased when the essential oils were removed implying that the essential oils contribute to the toxicity of the solvent extracts. A poor relationship obtained between the toxicity of the essential oils and the solvent extracts may be due to the fact that essential oils were present in relatively small amounts in the solvent extracts (Figure 4.2). Although the mechanism of action of the essential oil is not clearly understood, the hydrophobic compounds present in essential oils are proposed to disrupt cell membranes resulting in increased permeability to ions which will cause cell death (Sikkema et al., 1995; Cowan, 1999). 200 Although the activity of the essential oils was generally greater than that of the solvent extracts, this study has demonstrated that the removal of the essential oils did not significantly reduce the antimalarial activity of the solvent extract (for S. africana-caerulea and S. africana-lutea) as one may have assumed. Solvent extracts contain both volatile and non-volatile fractions. In Chapter 6 it was shown that the essential oils possessed poor anti-oxidant activity (IC50 values > 100 ?g/ml), thus it coul d be expected that the removal of essential oils would not affect the anti-oxidant activity of the solvent extracts. This assumption was correct as the solvent extracts still exhibited favourable anti-oxidant activity once the essential oil had been removed (Figure 13.5). It is known that there is a direct relationship between the anti-oxidant activity and the total phenolic content (Chapter 6; Cuvelier et al., 1994, Cuvelier et al., 1996; Hohmann et al., 1999). The dilution of total phenolic content after the removal of the essential oils affects the anti-oxidant capacity, as well as other biological activities, of the solvent extracts. 13.6 Conclusions ? The solvent extracts have varying degrees of activities, and depend on the presence or absence of essential oils. ? The solvent extracts without the essential oils exhibited the best activity against Gram-positive bacteria, while those containing essential oils displayed the most favourable activity against Gram-negative bacteria. The hydrophobic essential oil compounds may therefore be important to disrupt the cell membrane and increasing the activity of the solvent extracts. ? The presence of the essential oils in the solvent extracts may have antagonistic, synergistic or additive interactions depending on the biological activity investigated. ? The presence of essential oils in the solvent extracts reduces the concentrations of phenolic compounds which are the main contributors of the anti-oxidant activity of the solvent extracts. ? S. africana-caerulea and S. lanceolata showed no significant variation in antimalarial activity between the two extracts. However, the solvent extract of S. africana-lutea containing essential oil was lower than that of the solvent extract without essential oil. 201 ? The solvent extract containing essential oils was more toxic to ?normal cells? compared to the solvent extracts without essential oils. ? The essential oils contribute to the toxicity of the solvent extracts. 202 Chapter 14: General Conclusions and Recommendations 14.1 General conclusions The aims of the study were to evaluate the biological activities of the solvent extracts and essential oils of indigenous Salvia species, to isolate and identify compounds which may be responsible for the biological activities and to determine any possible relationship between the species investigated using the essential oil composition and the leaf trichomes type data. In addition, the effect of seasons on the chemical composition of essential oil was studied as well as the influence of the essential oils on the biological activities of the solvent extracts. The aims of the study were achieved and the major findings are listed below. 14.1.1 Leaf trichome types and essential oil composition ? Both glandular (peltate and capitate) and non-glandular trichomes were observed in all the investigated species. ? The peltate trichome consists of eight to sixteen cells in the head arranged in one or two concentric circles. ? Four types of capitate trichomes were recorded based on the number of cells in the head and the length of the stalk cell and this helped to divide the investigated species into four groups. ? Three species namely S. aurita, S. chamelaeagnea and S. repens have multiple cells at the base of the non-glandular trichomes, while others only have one cell at the base. ? The essential oil yield of the eleven species studied was low (0.004-0.5%) and varied between species with S. muirii and S. radula having the highest and the lowest yield, respectively. ? Quantitative and qualitative variations in essential oil were recorded for the essential oils investigated. ? Ninety-three components composed mainly of monoterpene and sesquiterpene hydrocarbons, oxygen-containing monoterpenes and oxygen-containing sesquiterpenes were recorded. 203 ? Major components identified include ?-pinene, 1,8-cineole, linalool, limonene, myrcene, ?-caryophyllene, spathulenol, ?-caryophyllene oxide, viridiflorol, ?-3- carene and ?-bisabolol. ? S. runcinata was found to particularly rich in ?-bisabolol (65% of the total oil) and may therefore be considered as a natural source of ?-bisabolol. ? Cluster analysis data revealed that the essential oils of S. africana-caerulea and S. lanceolata were highly correlated due to their high amount of spathulenol and ?- caryophyllene oxide, which supports the morphological resemblance of the two species. The correlation between the essential oil of S. dolomitica and the rest of the species was very low. 14.1.2 Identification of phenolic compounds in the solvent extracts ? Indigenous Salvia species contain various phenolic compounds such as rosmarinic acid, carnosol, ursolic acid, oleanolic acid, salvigenin, 7-O-methylepirosmanol, betulafolientriol oxide, caffeic acid and carnosic acid. ? Betulafolientriol oxide isolated from S. radula was detected in all the 17 species investigated. Various populations need to be investigated in order to determine whether this compound could be used as taxonomic marker for indigenous Salvia species. ? Rosmarinic acid was detected in all the indigenous species, except S. verbenaca. ? S. muirii, S. runcinata and S. albicaulis have high levels of rosmarinic acid (> 35%), while the levels of carnosol and/or carnosic acid are high in S. aurita, S. chamelaeagnea, S. namaensis and S. stenophylla (> 20%). 14.1.3 Biological activities and isolated compounds from two Salvia species ? The solvent extracts of indigenous Salvia species are good anti-oxidant but exhibited poor anti-inflammatory activity. ? In contrast, the essential oils possessed favourable anti-inflammatory activity, but poor anti-oxidant activity. ? Qualitative analysis by TLC indicated that the same compounds possessed the ability to scavenge the ABTS?+ and DPPH? radicals. 204 ? The anti-oxidant capacity of the S. schlechteri extract was higher than that of the reference compound, vitamin C which indicates that this plant may be used as Natural anti-oxidant. ? The solvent extracts and essential oils inhibited the in vitro growth of P. falciparum and in most of cases the activity of the essential oils was greater than that of its counterpart solvent extract. ? Betulfolientriol oxide and salvigenin isolated from S. radula exhibited comparable or lower activity than the crude extract and chloroquine and quinine. This suggests that the activity of the crude extract is determined by the combination of various compounds. ? The solvent extracts and/or essential oils displayed varing degrees of activity against bacterial, human cancer cells and human kidney epithelial kidney cells. ? The solvent extracts of S. chamelaeagnea, S. disermas, S. dolomitica and S. muirii exhibited good activities in various assays, while their toxicity to human epithelial cells were low in comparison to other solvent extracts. ? S. africana-caerulea was the only species displaying anticancer activity on all three human cancer cell lines with an IC50 less than 30 ?g/ml. ? The antibacterial and anticancer activities of the essential oils and/or solvent extracts against the test patogens provide a scientific rationale of the use of the taxa in traditional herbal remedies for the treatment of respiratory ailments, dermal infections, tuberculosis and some cardio-vascular diseases. ? The in vitro combination of S. chamelaeagnea and L. leonurus demonstrated a synergistic interactions against B. cereus and S. aureus, providing support for the co-administration use of these two species. ? Carnosol, ursolic acid/oleanolic acid, 7-O-methylepirosmanol were isolated from S. chamelaeagnea and tested against bacterial pathogens. Their activities were in most cases higher than that of the crude extract, but less active than ciprofloxacin. ? No direct correlation could be observed between the chemical composition of the essential oils and the solvent extracts and the biological activities. Hence, the activity of the essential oil or solvent extract is most likely the result of the combination between the major and minor compounds. 205 14.1.4 Seasonal variation in essential oil composition and biological activities and the influence of the essential oil in the biological activities of the solvent extracts ? Seasonal variation is an important factor that has an influence on essential oil yield. In addition, the chemical profile of the solvent extracts and essential oils were found to vary between seasons. ? It is difficult to determine a specific time of the year when the solvent extract and essential oil provide optimal biological activity as the activity varies with the species involved, as well as the biological activity. However, it was noted that, collection and preparationsof the solvent extracts in winter and essential oils in spring were generally less toxic than plants collected during other seasons. ? The presence of essential oils in the solvent extract had a variable effect on the the activity of the solvent extract and was dependent on the biological activity investigated. ? The solvent extracts containing essential oils were more toxic to human kidney epithelial cells than the solvent extracts devoid of essential oils. ? The results of the in vitro pharmacological activities justify why Salvia species in general are considered as a ?panacea? in traditional medicine. 14.2 Recommendations ? The collection of the same species from different localities is important in order to determine chemotypic variation for each species. ? The leaf trichome types, biological activity and phytochemistry of indigenous Salvia species not included in this study should be completed. ? Investigate the mechanisms by which the solvent extracts and essential oils exert their anti-inflammatory and anti-oxidant activities. ? Isolate the active compounds from the most promising plant which displayed anticancer, antimycobacterial and anti-oxidant activity. ? Further research should consider testing various fractions of the solvent extracts for different biological activities and disregard the fraction with highly toxic effects. ? Perform in vivo tests with those plants that exhibited specific activity in vitro. 206 References Adzet, T., Canigueral., S., Iglesias, J. 1988. A chromatographic survey of polyphenols from Salvia species. Biochemical Systematics and Ecology 16, 29?32. Ahmed, A.A., Mohamed, A.E-H., Karchesy, J., Asakawa, Y. 2006. Salvidorol, a nor- abietane diterpene with a rare carbon skeleton and two abietane diterpene derivatives from Salvia dorrii. Phytochemistry 67, 424?428. Ahmed, M., Ting, I.P., Scora, R.W. 1994. Leaf oil composition of Salvia hispanica L. from three geographical areas. Journal of Essential Oil Research 6, 223?228. Ai, C.-B., Li, L.-N. 1988. Stereostructure of salvianolic acid b and isolation of salvianolic acid c from Salvia milthiorrhiza. Journal of Natural Products 51, 145?149. Ai, C.-B., Li, L.-N. 1992. Salvianolic acids D and E: two new dipsides from Salvia milthiorrhiza. Planta Medica 58, 197?199. Alitonou, G.A., Avlessi, F., Sohounhloue, D.K., Agnaniet, H., Bessiere, J.-M., Menut, C. 2006. Investigations on the essential oil of Cymbopogon giganteus from Benin for its potential use as an anti-inflammatory agent. International Journal of Aromatherapy 16, 37?41. Amabeoku, G.J., Eagles, P., Scott, G., Mayeng, I., Springfield, E. 2001. Analgesic and antipyretic effects of Dodonaea angustifolia and Salvia africana-lutea. Journal of Ethnopharmacology 75, 117?124. Ara?jo, C., Sousa, M.J., Ferreir a, M.F., Le?o, C. 2003. Activ ity of essential oils from Mediterranean Lamiaceae species against food spoilage yeasts. Journal of Food Protection 66, 625?632. Areias, F.M., Valent?o, P., Andrade, P.B., Ferreres, F., Seabra, R.M. 2000. Flavonoids and phenolic acids of sage: influence of some agricultural factors. Journal of Agricultural and Food Chemistry 48, 6081?6084. Areias, F.M., Valent?o, P., Andrade, P.B., Ferreres, F., Seabra, R.M. 2001. Phenolic fingerprint of peppermint leaves. Food Chemistry 73, 307 ?311. Arnol, W.N. 1988. Vincent Van Gogh and thujone connection. Journal of the American Medical Association 260, 3042?3044. Arnold, T.H., De Wet, B.C. 1993. Plants of Southern Africa: Names and Distribution. Memoirs of the Botanical Survey of South Africa. 62, National Botanical Institute, Pretoria. 207 Aruoma, O.I. 1998. Free radicals, oxidative stress and anti-oxidants in human health and disease. Journal of the American Oil Chemists? Society 75, 199?212. Ascens?o, L., Marques, N., Pais, M.S. 1995. Glandular trichomes on vegetative and reproductive organs of Leonotis leonurus (Lamiaceae). Annals of Botany 75, 615?626. Ascens?o, L., Pais, M.S. 1987. Glandular trichomes of Artemisia campestris (ssp. maritima): ontogeny and histochemistry of the secretory product. Botanical Gazette 148, 221?227. Atindehou, K.K., Schmid, C., Brun, R., Kon?, M.W., Traore, D. 2004. Antitrypanosomal and antiplasmodial activity of medicinal plants from C?te d?Ivoire. Journal of Ethnopharmacology 90, 221?227. Azcan, N., Ertan, A., Demirci, B., Baser, K.H.C. 2004. Fatty acid composition of seed oils of Twelve Salvia species growing in Turkey. Chemistry of Natural Compounds 40, 218?221. Baratta, M.T., Dorman, H.J.D., Deans, S.G., Figueiredo, A.C., Barroso, J., Ruberto., G. 1998. Antimicrobial and anti-oxidant properties of some commercial essential oils. Flavour and Fragrance Journal 13, 235?244. Baricevic, D., Sosa, S., Della, L.R., Tubaro, A., Simonovska, B., Krasna, A., Zupancic, A. 2001. Topical anti-inflammatory activity of Salvia officinalis L. leaves: the relevance of ursolic acid. Journal of Ethnopharmacology 75, 125?132. Barnes, S. 1995. Effect of genistein on in vitro and in vivo models of cancer. Journal of Nutrition 125, 777?783. Barroso, J.G., Pedro, L.G., Figueiredo, A.C., Pais, M.S.S. 1992. Seasonal variation in the composition of the essential oil of Crithmum maritimum L. Flavour and Fragrance Journal 7, 147?150. Baylac, S., Racine, P. 2003. Inhibition of 5-lipoxygenase by essential oils and other natural fragrant extracts. The International Journal of Aromatherapy 13, 138?142. Bellomaria, B., Arnold, N., Valentini, G., Arnold, H.J. 1992. Contribution to the study of the essential oil from three species of Salvia growing wild in the eastern Mediterranean region. Journal of Essential Oil Research 4, 607?614. Benoit-Vical, F., Imbert, C., Bonfils, J.P., Sauvaire, Y. 2003. Antiplasmodial and antifungal activities of iridal, a plant triterpepenoid. Phytochemistry 62, 747?751. 208 Benrezzouk, R., Terencio, M.C., Ferrandiz, M.L., Hernandez-Perez, M., Rabanal, R., Alcaraz, M.J. 2001. Inhibition of 5-lipoxygenase activity by the natural anti- inflammatory compound aethiopinone. Inflammation Research 50, 96?101. Berenbaum, M.C. 1978. A method for testing for synergy with any number of agents. The Journal of Infectious Diseases 137, 122?130. Bilia, A.R., Lazari, D., Messori, L., Taglioti, V., Temperini, C., Vincieri, F.F. 2002. Simple and rapid physical-chemical methods to examine action of antimalarial drugs with hemin. Its application to Artemisia annua constituents. Life Sciences 70, 769?778. Bisio, A., Corallo, A., Gastaldo, P., Romussi, G., Cairallo, G., Fontana, N., De Tommasi, N., Profumo, P. 1999. Glandular hairs and secreted material in Salvia blepharophylla Brandegee ex Epling grown in Italy. Annals of Botany 83, 441?452. Blumenthal, M., Goldberg, A., Brinckmann, J. 2000. Herbal Medicine: Expanded Commission E Monographs. Integrative Medicine Communications, USA. Bodeker, G. 2000. Searching for antimalarials in plants. Journal of Alternative and Complementary Medicine 6, 127?129. Boik, J. 2001. Natural Compounds in Cancer Therapy. 1st edition. Oregon Medical Press, LLC, USA. Bourett, T.M., Howard, R.J., O?Keefe, D.P., Hallahan, D.L. 1994. Gland development on leaf surface of Nepeta racemosa. International Journal of Plant Science 155, 623? 632. Boyom, F.F., Ngouana, V., Zollo, A.P.H., Menut, C., Bessiere, J.M., Gut, J., Rosenthal, P. 2003. Composition and anti-plasmodial activities of essential oils from some Cameroonian medicinal plants. Phytochemistry 64, 1269?1275. Brand-Williams, W., Cuvelier, M.E., Berset, C. 1995. Use of a free radical method to evaluate anti-oxidant activity. Food Science and Technology 28, 25?30. Brunke, E.J., Hammerschmidt, F.J. 1985. Constituents of the essential oil of Salvia stenophylla-first identification of the (+)-?-bisabolol in Nature. In: Svendsen, A.B., Scheffer, J.J.C. (eds), Essential Oils and Aromatic Plants. Martinus Nijhoff/Dr. W. Junk Publishers, Dordrecht, Netherlands, pp 37?43. Buchbauer, G. 2000. The detailed analysis of essential oils leads to the understanding of their properties. Perfumer and Flavorist 25, 64?67. 209 Burkhard, P.R., Burkharddt, K., Haenggeli, C.A., Landis, T. 1999. Plant-induced seizures: reappearance of an old problem. Journal of Neurology 246, 667?670. Butler, M.S. 2004. The role of natural products chemistry in drug discovery. Journal of Natural Products 67, 2141?2153. Cabo, Y., Crespo, M.E., Jimenez, J., Navarro, C., Perani, E., Zarzuelo, A. 1987. Sur l?huile ?ssentielle de Salvia lavandulifolia Vahl subsp. oxydon. Plantes M?dicinales et Phytoth?rapie 21, 116?121. Campbell, W.E., Gammon, D.W., Smith, P., Abrahams, M., Purves, T.D. 1997. Composition and antimalarial activity in vitro of the essential oil of Tetradenia riparia. Planta Medica 63, 270?272. Canfield, C.J., Pudney, M., Gutteridge, W.E. 1995. Interactions of atovaquone with other antimalarial drugs against Plasmodium falciparum in vitro. Experimental Parasitology 80, 367?381. Cantrell, C.L., Franzblau, S.G., Fischer, N. 2001. Antimycobacterial plant terpenoids. Planta Medica 67, 2685?694. Carta, C., Moretti, M.D.L., Peana, A.T. 1996. Activity of the oil of Salvia officinalis L. against Botrytis cinerea. Journal of Essential Oil Research 8, 399?404. Chalchat, J.C., Garry, R.P., Menut, C., Mamaty, G., Malhuret, R., Chopineau, J. 1997. Correlation between chemical composition and antimicrobial activity VI activity of some African essential oils. Journal of Essential Oil Research 9, 67 ?75. Chandler, S.F., Dodds, J.H. 1983. The effects of phosphate, nitrogen and sucrose on the production of phenolics and salasidine in callus cultures of Solanum lacinitum. Plant Cell Report 2, 105. Changwe, K. 2002. Taxonomy Studies of Thorncroftia (Lamiaceae) in Africa. MSc dissertation, Faculty of Science, University of the Witwatersrand, South Africa. Chen, X., Ding, J., Ye Y.M., Zhang, J.S. 20 02. Bioactive abietane and seco-abietane diterpernoids from Salvia prionitis. Journal of Natural Products 65, 1016?1020. Cimanga, K., Kambu, K., Tona, L., Apers, S., De Brunne, T., Harmans, N., Tott?, J., Pieters, L., Vlietinck, A.J. 2002. Correlation between chemical composition and antibacterial activity of essential oils of some aromatic medicinal plants growing in the Democratic Republic of Congo. Journal of Ethnopharmacology 79, 213?220. Clark, A.M. 1996. Natural products as a resource for new drugs. Pharmaceutical Research 13, 1133?1141. 210 Clarkson, C., Maharaj, V.J., Crouch, N.R., Grace, O.M., Lillay, P., Matsabisa, M., Bhagwandin, N., Smith, P.J., Folb, P.I. 2004. In vitro antiplasmodial activity of medicinal plants native to or naturalized in South Africa. Journal of Ethnopharmacology 92, 177?191. Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Collins, M.A., Charles, H.P. 1987. Antimicrobial activity of carnosol and ursolic acid. Two anti-oxidant constituents of Rosmarinus officinalis. Food Microbiology 4, 311?315. Cook, N.C., Samman, S. 1996. Flavonoids, chemistry, metabolism, cardioprotective effect and dietary sources. Nutritional Biochemistry 7, 66?76. Corbett, T.H., Valeriote, F.A., Polin, L., Panchapor, C., Pugh, S., White, K., Lowichik, N., Knight, J., Bissery, M.C., Wozniak, A., LoRusso, P., Biernat, L., Polin, D., Knight, L., Biggar, S., Looney, D., Demchik, L., Jones, J., Jones, L., Blair, S., Palmer, K., Essenmacher, S., Lisow, L., Mattes, K.C., Cavanaugh, P.F., Rake, J.B., Baker, L. 1990. In: Valeriote, F.A., Corbett, T.H., Baker, L.H. (eds), Cytotoxic Anticancer Drugs: Models and Concepts for Drug Discovery and Development. Kluwer Academic Publishers, Boston, pp 35?87. Cordell, G.A., Beecher, C.W., Pezzuto, J.M. 1991. Can ethnopharmacology contribute to the development of new anticancer drugs? Journal of Ethnopharmacology 32, 117?133. Couladis, M., Tzakou, O., Verykokidou, E., Harvala, C. 2003. Screening of some Greek aromatic plants for anti-oxidant activity. Phytotherapy Research 17, 194?195. Cowan, M.M., 1999. Plant products as antimicrobial agents. Clinical Microbiology Review 12, 564?582. Cragg, G.M, Newman, D.J. 2005. Plants as a source of anticancer agents. Journal of Ethnopharmacolgy 100, 72 ?79. Cragg, G.M., Newmann, D.J., Snader, K.M. 1997. Natural products in drug discovery and development. Journal of Natural Products 60, 52?60. Cutler, D.F. 1978. Applied Plant Anatomy. Longman, New York. Cutler, D.F., Brandham, P.E. 1977. Epidermal evidence of the genetic control of leaf surface characters in hybrid Aloineae (Liliaceae). Kew Bulletin 32, 23?32. Cuvelier, M.E., Berset, C., Richard, H. 1994. Anti-oxidant constituents in sage (Salvia officinalis). Journal of Agricultural and Food Chemistry 42, 665?669. 211 Cuvelier, M.E., Richard, H., Berset, C. 1996. Anti-oxidant activity and phenolic composition of pilot-plant and commercial extracts of sage and rosemary. Journal of the American Oil Chemists? Society 73, 645?652. Davis, N.M., Good, R.L., Roupe, K.A., Y? ?ez, J.A. 2004. Cyclo-oxygenase-3: axiom, dogma, anomaly, enigma or splice error? Not as easy as 1, 2, 3. Journal of Pharmacy and Pharmaceutical Sciences 7, 217?226. Davis, P.H., Heywood, V.H. 1973. Principles of Angiosperm Taxonomy. R. E. Krieger, New York. De Feo, V., Bruno, M., Tahiri, B., Napolitano, F., Senatore, F. 2003. Chemical composition and antibacterial activity of essential oils from Thymus spinulosus Ten. (Lamiaceae). Journal of Agricultural and Food Chemistry 51, 3849?3853. De La Puerta, R., Saenz, M.T., Garcia, M.D. 1993. Cytostatic activity against HEp-2 cells and antibacterial activity of essential oils from Helichrysum picardii. Phytotherapy Research 7, 378?380. Deighton, N., Brennan, R., Finn, C., Davies, H.V. 2000. Anti-oxidant properties of domesticated and wild Rubus species. Journal of the Science of Food and Agriculture 80, 1307?1313. Delaquis, P.H., Stanich, K., Girard, B., Mazza, G. 2002. Antimicrobial activity of individual and mixed fractions of dill, cilantro, coriander and eucalyptus essential oils. International Journal of Food Microbiology 74, 101?109. Desjardins, R.E., Canfield, C.J., Haynes, J.D., Chulay, J.D. 1979. Quantitative assessment of antimalarial activity in vitro by semi-automated micro-dilution technique. Antimicrobial Agents and Chemotherapy 16, 710?718. Dilaka, F., Meyer, J.J.M. 1996. Antimicrobial activity of Helichysum pedunculatum used in circumcision rites. Journal of Ethnopharmacology 53, 51?54. Dilcher, D.L. 1974. Approaches to the identification of angiosperm leaf remains. Botanical Review 40, 1?157. Dimayuga, R. E., Garcia, S. K., Nielsen, P. H., Christophersen, C. 1991. Traditional medicine of Baja California Sur (Mexico) III. Carnosol: a diterpene antibiotic from Lepechinia hastate. Journal of Ethnopharmacology 31, 43?48. Dorman, H.J.D., Surai, P., Deans, S.G. 2000. In vitro anti-oxidant activity of a number of plant essential oils and phytoconstituents. Journal of Essential Oil Research 12, 241?248. 212 Duan, S., Weng, X.C., Dong, X.W., Liu, Y.P ., Li, H.P., Jin, J.R. 1998. Anti-oxidant properties of butylated hydroxytoluene refluxed in ferric chloroide solution. Food Chemistry 61, 101?105. Eloff, J.N. 1998. A sensitive and quick microplate method to determine the minimal inhibitory concentration of plant extracts for bacteria. Planta Medica 64, 711?713. Eloff, J.N. 1999. It is possible to use the herbarium specimens to screen for antibacterial activity components in some plants. Journal of Ethnopharmacology 67, 355?360. Exarchou, V., Nanadis, N., Tsimidou, M., Gerothanassis, I.P. Troganis, A., Boskou, D. 2002. Anti-oxidant activities and phenolic composition of extracts from Greek oregano, Greek sage and summer savory. Journal of Agricultural and Food Chemistry 50, 5294?5299. Fabry, W., Okemo, P.O, Ansorg, R. 1998. Antibacterial activity of East African medicinal plants. Journal of Ethnopharmacology 60, 79?84. Fahn, A. 2001. Structure and function of secretory cells. In: Hallahan, D.L., Gray, J.C. (eds), Advances in Botanical Research: Plant Trichomes. Academic Press, London, pp. 37?66. Falciani, L., Bini Maleci, L., Mariotti Lippi, M. 1995. Morphology and distribution of trichomes in Italian species of the Stachys germania group (Labiatae): a taxonomic evaluation. Botanical Journal of the Linnean Society 119, 245?254. Farhat, G.N., Affara, N.I., Gali-Muhtasib, H. 2001. Seasonal changes in the composition of the essential oil extract of East Mediterranean sage (Salvia libanotica) and its toxicity in mice. Toxicon 39, 1601?1605. Farnworth, N.R., Bingel, A.S. 1977. Problems and propects of discovering new drugs from high plants by pharmacological screening. In: Wangner, H., Wolff, P. (eds), New Natural Products and Plant Drugs With Pharmacological, Biological or Therapeutical Activity. Springer-Verlag, Berlin, pp. 1-22. Fatope, M.O., Al-Kindi, S.M.Z., Abdulnour, A.O. 2000. Research trend: natural products as pest, microbial disease as tumour control agents. Science and Technology, Special Review 55?71. Feder, N., O?Brien, T.P. 1968. Plant microt echnique: some principles and new methods. American Journal of Botany 55, 123?142. Foley, M., Tilley, L. 1997. Quinoline antimalarials: mechanisms of action and resistance. International Journal of Parasitology 27, 231?240. 213 Foray, L., Bertrand, C., Pinguet, F., Soulier, M., Astre, C., Marion, C., Pelissier, Y., Bessi?re, J. M. 1999. In vitro cytotoxic activity of three essential oils from Salvia species. Journal of Essential Oil Research 11, 522?526. Frankel, E.N., Waterhouse, A.L., Teissedre, P.L. 1995. Principal phenolic phytochemicals in selected California wines and their anti-oxidant activity inhibiting oxidation of human low-density lipoprotein. Journal of Agricultural and Food Chemistry 43, 890?894. Gali-Muhtasib, H., Hilan, C., Khater, C. 2000. Traditional uses of Salvia libanotica (East Mediterranean sage) and the effects of its essential oils. Journal of Ethnopharmacology 71, 513?520. Gerhardt, U., Schroeter, A. 1983. Rosmarinic acid ? a naturally occurring anti-oxidant in spices. Fleischwirtschaft 63, 1628?1630. Gessler, M.C., Nkunya, M.H.H., Mwasumbi, L.B., Heinrick, M., Tanner, M. 1994. Screening Tanzanian medicinal plants for antimalarial activity. Acta Tropica 56, 65?77. Gono-Bwalya, A. 2003. The Chemotaxonomy and Biological Activity of Salvia stenophylla (Lamiaceae) and Related Taxa. MSc dissertation, Faculty of Health Sciences, University of the Witwatersrand, South Africa. Grange, J.M., Davey, R.W. 1990. Detection of anti-tuberculosis activity in plant extracts. Journal of Applied Bacteriology 68, 587?591. Hallahan, D.L. 2000. Monoterpenoid biosynthesis in glandular trichomes of Labiatae plants. In: Hallahan, D.L., Gray, J.C. (eds), Advances in Botanical Research: Plant Trichomes, Academic Press, London, pp.77?120. Hanlidou, E., Kokkinis, S., Bosabalidis, A.M., Bessiere, J.M. 1991. Glandular trichomes and essential oil constituents of Calamintha menthifolia (Lamiaceae). Plant Systematics and Evolution 117, 17?26. Harbone, J.B. 1993. Introduction to Ecological Biochemistry. 4th edition, Academic Press, London. Harbone, J.B., Mabry, T.J., Mabry, H. 1975. The Flavonoids. Chapman and Hall, London. Harraz, F.M., Ulubelen, A., ?ks?z, S., Tan, N. 1995. Betulafolientriol oxide triterpenes from Cleome amblocarpa. Phytochemistry 39, 175?178. Hartwell, J.L. 1982. Plant Used Against Cancer. Quarterman, Lawrence, MA. 214 Hayes, A.J., Markovic, B. 2002. Toxicity of Australian essential oil Backhousia citriodora (Lemon myrtle), Part 1. Antimicrobial activity and in vitro cytotoxicity. Food and Chemical Toxicology 40, 1409?1416 Hedge, I.C. 1974. A Revision of Salvia in Africa including Madagascar and the Canary Islands. Royal Botanic Garden, Edinburgh. Heifets, L.B., Iseman, M.D., Cook, J.L., Lindholm-Levy, P.J., Drupa, I. 1985. Determination of in vitro susceptibility of M. tuberculosis to cephalosporins by radiometric and conventional methods. Antimicrobial Agents and Chemotherapy 27, 11?15. Heinonen, M., Lehtonen, P.J., Hopla, A. 1998. Anti-oxidant activity of berry and fruit wines and liquor. Journal of Agricultural and Food Chemistry 43, 890?894. Hernandez-Perez, M., Rabanal, R.M, De la Torre M.C., Rodriguez B. 1995. Analgesic, anti-inflammatory, antipyretic and haematological effects of aethiopinone, an O- naphthoquinone ditepernoid from Salvia aethiopis roots and two hemisynthetic derivatives. Planta Medica 61, 505?509. Hinou, J.B., Harvala, C.E., Hinou, E.B. 1989. Antimicrobial activity of 32 common constituents of essential oils. Pharmazie 44, 302?330. Hohmann, J., Zupk?, I., R?dei, D., Cs?nyi, M, Falkay, G., M?the, I., Janics?k, G. 1999. Protective effects of the aerial parts of Salvia officinalis, Melissa officinalis and Lavandula angustifolia and their constituents against enzyme-dependent and enzyme-independent lipid peroxidation. Planta Medica 65, 576?578. Houghton, P.J. 1999. New approaches to the search for novel coumpounds. Acta Horticulture 501, 49?56. Houghton, P.J. 2001. Old yet new-pharmaceuticals from plants. Journal of Chemical Education 78, 175?184. Houghton, P.J. and Raman, A. 1998. Laboratory Handbook for the Fractionation of Natural Extracts. Chapman and Hall, London. Hristea, E., Hillebrand, M., Caproiu, T.M., Caldararu, H., Constantinescu, T., Balaban, A.T. 2002. Scavenging the hydroxyl radical by 2,2-diphenyl-1-picrylhydrazyl. Arkivoc ii 123?132. IDRC (International Development Research Centre, Canada), 2006. La production d??ssences v?g?tales au Togo: Une autr e alternative de cr?ation d?emplois. www.idrc.ca/en/ev-72993-201-1-DO_TOPIC.html (25th June 2006). 215 Ito, N., Fukushima, S., Tsuda, H. 1985. Carcinogenicity and modification of the carcinogenic response by BHA and BHT and other anti-oxidants. CRC Critical Review in Toxicology 15, 109?150. Iwu, M.M. 1994. African medicinal plants in the search for new drugs based on ethnobotanical leads. In: Chadwick, D.J., Marsh, J. (eds), Ethnobotany and the Search for New Drugs. John Wiley and Sons. Chichester, Ciba Foundation Symposium, pp. 116?129. J?ger, A.K., van Staden, J. 2000. Salvia in Southern Africa. In: Kintzios, S.E. (ed), Sage: The Genus Salvia. Harwood Academic Publishers, Netherlands, pp. 47?53. Janics?k, G., M?the, I. 1997. Parallel determin ation of rosmarinic acid and caffeic acids by TLC-densitometry. Chromatogrophia 46, 322?324. Janics?k, G., M?the, I., Mikl?ssy-V?ri, M., Bl unden, G. 1999. Comparative studies of the rosmarinic and caffeic acid contents of Lamiaceae species. Biochemical Systematics and Ecology 27, 733?738. Jequier, C., Nicollier, G., Tabacchi, R., Garnero, J. 1980. Constituents of the essential oils of Salvia stenophylla. Phytochemistry 19, 461?462. Johansen, D.A. 1940. Plant Microtechnique. McGraw Hill, New York. Joubert, P.H. 1990. Poisoning admission of Black South Africans. Journal of Toxicology- Clinical Toxicology 28, 85?94. Juergens, U.R., St?ber, M., Schmilling, L ., Kleuver, T., Vetter, H. 1998. Anti- inflammatory effects of eucalyptol (1,8-cineole) in bronchial asthma: inhibition of arachidonic acid metabolism in human blood monocytes ex vivo. European Journal of Medical Research 3, 407?412. K?hk?nen, M.P., Hopia, A.I., Vuorela, H.J. , Rauha, J.-P., Pihlaja, K., Kujala, T.S., Heinonen, M. 1999. Anti-oxidant activity of plant extracts containing phenolic compounds. Journal of Agricultural and Food Chemistry 47, 3954?3962. Kalemba, D., Kunicka, A. 2003. Antibacterial and antifungal properties of essential oils. Current Medicinal Chemistry 10, 813?829. Karou, D.D., Dicko, M.H., Sanon, S. 2003. Antimalarial activity of Sida acuta Burm. f. (Malvaceae) and Pterocarpus erinaceus Poir. (Fabaceae). Journal of Ethnopharmacology 89, 291?294. Karousou, R., Bosabalidis, A.M., Okinis, S. 1992. Sideritis syriaca ssp. syriaca: glandular trichome structure and development in relation to systematics. Nordic Journal of Botany 12, 31?37. 216 Kaya, A., Demirci, B., Ba?er, K.H.C. 2003. Glandular trichomes and essential oils of Salvia glutinosa L. South African Journal of Botany 69, 422?427. Kehrer, J.P. 1993. Free radicals as mediators of tissue injury and disease. Critical Reviews in Toxicology 23, 21?48. Kim, D.O., Lee, K.W., Le, H.J., Lee, C.Y. 2002. Vitamin C equivalent anti-oxidant capacity (VCEAC) of phenolic phytochemicals. Journal of Agricultural Food Chemistry 50, 3713?3717. Kobayashi, T., Nakata, T., Kuzumaki, T. 2002. Effect of flavonoids on cell cycle progression in prostate cancer cells. Cancer Research 176, 17?23. Koenen, E. V. 2001. Medicinal Poisonous and Edible Plants in Namibia. Klaus Hess Publishers/Verlag, Berlin. Kraft, C., Jennet-Siems, K., K?ehler, I., Tofem- Reblin, B., Siems, K., Bienzle, U., Eich, E. 2002. Antiplasmodial activity of sesquilignans and sesquineolignans from Bonamia spectabilis. Phytochemistry 60, 167?173. Lahlou, M. 2004. Methods to study the phytochemistry and bioactivity of essential oils. Phytotherapy Research 18, 435?448. Lall, N., Meyer, J.J.M. 1999. 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. Lawrence, B.M. 2000. Essential oils: from agriculture to chemistry. International Journal of Aromatherapy 10, 82?98. Lee, D., Lee, S., Noh, J., Hong, S. 1999. Antibacterial activities of crytotanshinone and dihydrotanshinone I from a medicinal herb, Salvia miltiorrhyza Bunge. Bioscience Biotechnology and Biochemistry 63, 2236?2239. Lee, K.W., Kim, Y.J., Lee, H.J., Lee, C.Y. 2003. Cocoa has more phenolic phytochemicals and a higher anti-oxidant capacity than teas and red wine. Journal of Agricultural and Food Chemistry 51, 7292?7295. Li, J., Guo, W.-J., Yang, Q.-Y. 2002. Effects of ursolic acid and oleanolic acid on human colon carcinoma cell line HCT-15. World Journal of Gastroenterology 8, 493?495. Lin, C-N., Chung, M-I., Gan, K-H., Chiang, J-R. 1987. Xanthones from Formosan gentia- naceous plants. Phytochemistry 26, 2381?2384. 217 Lis-Balchin, M., Deans, S.G., Eaglesham, E. 1998. Relationship between bioactivity and chemical composition of commercial essential oils. Flavour and Fragrance Journal 13, 98?104. Liu, G.T., Zhang, T.M., Wang, B.E., Wang, Y.W. 1992. Protective action of seven natural phenolic compounds against peroxidative damage to biomembranes. Biochemical Pharmacology 43, 490?495. Liu, J. 1995. Pharmacology of oleanolic acid and ursolic acid. Journal of Ethnopharmacology 49, 57?68. Liu, J. 2005. Oleanolic acid and ursolic acid: research perspectives. Journal of Ethnopharmacology 100, 92?105. Liu, J., Shen, H.M., Ong, C.N. 2000. Salvia miltiorrhiza inhibits cell growth and induces apoptosis in human hepatoma (HepG-2) cells. Cancer Letter 153, 85?493. Liu, K.S.C., Yang, S.L., Roberts, M.F., Elford, B.C., Phillipson, J.D. 1989. The contribution of flavonoids to antimalarial activity of Artemisia annua. Planta Medica 55, 654?655. Lopes, N.P., Kato, M.J., Andrade, E.H.A., Maia, J.G.S., Yoshida, M., Planchart, A.R., Katzin, A.M. 1999. Antimalarial use of volatile oil from leaves of Virola surinamensis (Rol.) Warb. By Wai?pi Amazon Indians. Journal of Ethnopharmacology 67, 313?319. Lourens, A.C.U., Reddy, D., Ba?er, K.H.C., Viljoen, A.M., van Vuuren, S.F. 2004. In vitro biological activity and essential oil composition of four indigenous South African Helichrysum species. Journal of Ethnopharmacology 95, 253?258. Lu, Y., Foo, L.Y. 1999. Rosmarinic acid derivatives from Salvia officinalis. Phytochemistry 51, 263?267. Lu, Y., Foo, L.Y. 2001. Ani-oxidant activities of polyphenols from sage (Salvia officinalis). Food Chemistry 75, 197?202. Lu, Y., Foo, L.Y. 2002. Polyphenolics of Salvia: a review. Phytochemistry 59, 117?140. Maffei, M., Chialva, F., Sacco, T. 1989. Glandular trichomes and essential oils of developing peppermint leaves. New Phytologist 111, 707?716. Mahato, S.B., Kundu, A.P. 1994. 13C NMR spectra of pentacyclic triterpenes ? a compilation and some salient features. Phytochemistry 37, 1517?1575. Mambro, V.M., Azzolini, A.E.C.S., Valim, Y.M.L., Fonseca, M.J.V. 2003. Comparison of anti-oxidant activities of tocopherols alone and in pharmaceutical formulations. International Journal of Pharmaceutics 262, 93?99 218 Mann, C.M., Markham, J.L. 1998. A new method for determining the minimum inhibitory concentration of essential oils. Journal of Applied Microbiology 84, 538?544. Mar?chal, J-P., Culioli, G., Hellio, C., Thomas-Guyon, H., Callow, M.E., Clare, A.S., Ortalo-Magn?. A. 2004. Seasonal variation in antifouling activity of crude extracts of the brown alga Bifurcaria bifurcata (Cystoseiraceae) against cyprids of Balanus amphitrite and the marine bacteria Cobetia marina and Pseudoalteromonas haloplanktis. Journal of Experimental Marine Biology and Ecology 313, 47?62. Markham, K.R. 1982. Techniques of Flavonoids Identification. Academic Press, London. Mascarenhas, M. 1994. Structure-activity characterization, a quick method to screen mushrooms for the presence of antitumour glucans. Mushrooms Research 3, 791? 795. Mativandlela, S.P.N., Lall, N., Meyer, J.J.M. 2006. Antibacterial, antifungal and antitubercular activity of (the roots of) Pelargonium reniforme (CURT) and Pelargonium sidoides (DC) (Geraniaceae) roots extracts. South African Journal of Botany 72, 232?237. Matu, E.N., van Staden, J. 2003. Antibacterial and anti-inflammatory activities of some plants used for medicinal purposes in Kenya. Journal of Ethnopharmacology 87, 35?41. McGaw, L.J., J?ger, A.K., van Staden, J. 2002. Variation in antibacterial activity of Schotia species. South African Journal of Botany 68, 41?46. Metcalfe, C.R., Chalk, L. 1950. Anatomy of the Dicotyledons, Leaves, Stem and Wood in Relation to Taxonomy with Notes on Economic Uses. Clarendon Press, Oxford. Middlebrook, G., Reggiards, Z., Tigertt, W.D. 1977. Automated radiometric detection of growth of Mycobacterium tuberculosis in selective media. American Review of Respiratory Disease 115, 1066?1069. Milhau G., Valentin A., Benoit F., Mallie M., Bastide J.M., Pelissier Y., Bessiere J.M., 1997. In vitro antimalarial activity of eight essential oils. Journal of Essential Oil Research 9, 329?333. Miski, M., Ulubelen, A., Johansson, C., Mabry, T.J. 1983. Antibacterial activity studies of flavonoids from Salvia paleastina. Journal of Natural Products 46, 874?875. Monks, A., Scudiero, D., Skehan, P., Shoemaker, R., Paull, K.D., Vistica, D., Hose, C., Langley, J., Cronise, P., Vaigro-Wolff, A., Gary-Goodrich, M., Campbell, H., Mayo, J., Boyd, M.R. 1991. Feasibility of a high-flux anticancer drug screen using 219 a diverse panel of cultured human tumour cell lines. Journal of the National Cancer Institute 183, 757?766. Moretti, M.D.L., Peana, A.T., Satta, M., 1997. A study on anti-inflammatory and peripheral analgesic action of Salvia sclarea oil and its main components. Journal of Essential Oil Research 9, 199?204. Mosmann, T. 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55?63. Mueller, S. 1966. The taxonomic significance of cuticular patterns within the genus Vaccinium (Ericaceae). American Journal of Botany 53, 633?640. Mulholland, D., Drewes, S. 2004. Global phytochemistry: indigenous medicinal chemistry on track in southern Africa. Phytochemistry 65, 769?782. National Center of Infectious Disease 2006. www.cdc.gov/malaria/biology/life cycle.htm (10 th August 2006). Nikaido, H., Vaara, M. 1985. Molecular basis of bacterial outer membrane permeability. Microbiological Reviews 49, 1?32. Njenga, E.W., van Vuuren, S.F., Viljoen, A.M. 2005. Antimicrobial activity of Eriocephalus L. species. South African Journal of Botany 71, 81?87. Nyiligira, E. 2004. Phytochemical and Pharmacological Investigation of South African Vitex Species. MSc dissertation, Faculty of Health Sciences, University of the Witwatersrand, South Africa. Okoli, A.S., Iroegbu, C.U. 2004. Evaluation of extracts of Anthocleista djalonensis, Nauclea latifolia and Uvaria afzalii against bacterial isolates from cases of non- gonococcal urethritis. Journal of Ethnopharmacology 92, 135?144. Oliveira, F.Q., Andrade-Neto, V., Kretti, A.U., Brand?o, M.G.L. 2004. New evidences of antimalarial activity of Bidens pilosa roots extract correlated with polyacetylenes and flavonoids. Journal of Ethnopharmacology 93, 39?42. Olliaro, P.L., Nair, N.K., Sathasivam, K., Mansor, S.M., Navaratman, V. 2001. Pharmacokinetics of artesunate after single oral administration to rats. BMC Pharmacology 1, 1?4. Pal?-Pa?l, J., P?rez-Alonso, M.J., Velasco- Negueruela, A., Pal?-Pa?l, R., Sanz, J., Conejero, F. 2001. Seasonal variation in chemical constituents of Santolina rosmarinifolia L. ssp. rosmarinifolia. Biochemical Systematics and Ecology 69, 663?672. 220 Paton, A. 1991. Salvia lanceolata, Labiateae: In the Kew Magazine incorporating Curtis?s Botanical Magazine. The Royal Botanic Gardens, Kew. Pauli, A. 2001. Antimicrobial properties of essential oil constituents. The International Journal of Aromatherapy 11, 126?133. Pavela, R. 2004. Insecticidal activity of certain medicinal plants. Fitotherapia 75, 745? 749. Peana, A.T., D?Aquila, P. S., Panin, F., Serra , G., Pippia1, P., Moretti, M. D. L. 2002. Anti-inflammatory activity of linalool and linalyl acetate constituents of essential oils. Phytomedicine 9, 721?726. Pedro, L.G., Santos, P.A.G., Da Silva, J.A., Figueiredo, A.C., Barroso, J.G., Deans, S.G., Looman, A., Scheffer, J.C. 2001. Essential oils from Azorean Laurus azorica. Phytochemistry 57, 245?250. Perry, N.S., Bollen, C., Perry, E.K., Ballard, C. 2003. Salvia for dementia therapy: review of pharmacological activity and pilot tolerability clinical trial. Pharmacology, Biochemistry and Behavior 75, 651?659. Pestana, M.H., Gageiro, J.N. 2000. An?lise de Dados para Ci?ncias Sociais. A complementaridade do SPSS. Edi??es S?labo, Lisboa. Petersen, M., Simmonds, M.S.J. 2003. Molecule of interest: Rosmarinic acid. Phytochemistry 62, 121?125. Pitarevic, I., Kuftinec, J., Blazevic, N., Kustrak, D. 1984. Seasonal variation of essential oil yield and composition of Damatian sage, Salvia officinalis. Journal of Natural Products 47, 409?412. Pitarokili, D., Tzakou, O., Loukis, A., Harvala, C. 2003. Volatile metabolites from Salvia fruticosa as antifungal agents in soilborne pathogens. Journal of Agricultural and Food Chemistry 51, 3294?3301. Popoca, J., Aguilar, A., Alonso, D., Villarreal, M.L. 1998. Cytotoxic activity of selected plants used as antitumorals in Mexican traditional medicine. Journal of Ethnopharmacology 59, 173?177. Price, R., Luxemburger, C., van Vught, M., Nosten, F. 1998. Artesunate and mefloquine in the treatment of uncomplicated multidrug-resistant hyperparasitaemic falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 92, 207?211. 221 Prozesky, E.A., Meyer, J.J.M., Louw, A.I. 2001. In vitro antiplasmodial activity and cytotoxicity of ethnobotanically selected South African plants. Journal of Ethnopharmacology 76, 239?245. Putievsky, E., Ravid, U., Dudai, N. 1986. The influence of season and harvest frequency on the essential oil and the herbal yield from a pure clone of sage (Salvia officinalis) grown under cultivated conditions. Journal of Natural Products 49, 326?329. Qian, T-X., Li, L-N. 1992. Isolavianolic acid, a depside possessing a dibenzooxepin skeleton. Phytochemistry 31, 1068?1072. Raffauf, R.F. 1996. Plant Alkaloids: a Guide to their Discovery and Distribution. The Haworth Press, New York. Rasoanaivo, P., Ratsimamanga-Urverg, S. 1993. Biological evaluation of plants with reference to the Malagasy flora. Napreca, Madagascar. Reddy, M.K., Reddy, V.B.M., Gunasekar, D., Murthy, M.M., Caux, C., Bodo, B. 2003. A flavone and unusual 23-carbon terpenoids from Andrographis paniculata. Phytochemistry 62, 1271?1275. Roberts, G.D., Goodman, N., Heifets, L., Larsh, H.W., Lindner, T.H., McClatchy, J.K., McGinnis, M.R., Siddiqi, S.H., Wright, P. 1983. Evaluation of the BACTEC radiometric method for recovery of mycobacteria and drug susceptibility testing of Mycobacterium tuberculosis from acid-fast smear-positive specimens. Journal of Clinical Microbiology 18, 689?696. Rohlf, J.F. 1992. NTSYS-pc, Numerical Taxonomy and Multivariate Analysis System, Applied Biostatistics Inc., New York. Rouf, A.S.S., Ozaki, Y., Rashid, M.A., Rui, J. 2001. Betulafolientriol oxide derivatives from the dried fruits of Forsythia suspense. Phytochemistry 56, 815?818. Ruberto, G. Baratta, M.T. 2000. Anti-oxidant activity of selected essential oil components in two lipid models systems. Food Chemistry 69, 167?174. Ryu, S.Y., Lee, C.O., Choi, S.U. 1997. In vitro cytotoxicity of tanshinones from Salvia miltiorrhiza. Planta Medica 63, 339?342. Safayhi, H., Rall, B., Sailer, E.-R., Ammon, H. P. T. 1997. Inhibition by boswellic acids of human leukocyte elastase. Pharmacology and Experimental Therapeutics 282, 460?463. Safayhi, H., Sailer, E.R., Ammon, H.P.T. 1995. Mechanism of 5-lipoxygenase inhibition by acetyl-11-keto-?-boswellic acid. Molecular Pharmacology 47, 1212?1216. 222 ?ahin, F., G?ll?ce, M., Daferera, D., S?kmen , A., S?kmen, M., Polissiou, M., Agar, G., ?zer, H. 2006. Biological activities of th e essential oils and methanol extract of Origanum vulgare ssp. vulgare in the Eastern Anatolia region of Turkey. Food Control 15, 549?557. Sailer, E.-R., Subramanian, L. R., Rall, B., Hoernlein, R. F., Ammon, H. P. T., Safayhi, H. 1996. Acetyl-11-keto-?-boswellic acid (AKBA): Structure requirements for binding and 5-lipoxygenase inhibitory activity. British Journal of Pharmacology 117, 615?618 Sala, A., Recio, M.C., Schinella, G.R., M?nez, S., Giner, R.M., Cerd?-Nicol?s, M., R?os, J.-L. 2003. Assessment of the anti-inflammatory activity and free radical scavenger activity of tiliroside. European Journal of Pharmacology 461, 53?61. Salmon, S.E. 1984. Human tumor colony assay and chemosensitivity testing. Cancer Treatment Reports 68, 117?125. Sanchez-Moreno, C., Larrauri, J.A., Saura-Calixto, F. 1998. A procedure to measure the antiradical efficiency of polyphenols. Journal of the Science of Food and Agriculture 76, 270?276. Sanon, S., Ollivier, E., Azas, N., Mahiou, V., Gasquet, M., Ouattara, C.T., Nebie, I., Traore, A.S., Esposito, F., Balansard, G., Timon-David, P., Fumoux, F. 2003. Ethnobotanical survey and in vitro antiplasmodial activity of plants used in traditional medicine in Burkina Faso. Journal of Ethnopharmacology 86, 143?147. Sattar, A., Bankova, V., Kujumgiev, A., Galabov, A., Ignatova, A., Todorova, C., Popov, S. 1995. Chemical composition and biological activity of leaf exudates from some Lamiaceae plants. Pharmazie 50, 62-65. Savelev, S., Okello, E., Perry, N.S.L., Wilkins, R.M., Perry, E.K. 2003. Synergistic and antagonistic interactions of anticholinesterase terpenoids in Salvia lavandulaefolia essential oils. Pharmacology, Biochemistry and Behavior 75, 661?668. Scarparo, C., Coli, P., Rigon, A., Ruggiero, G., Ricordi, P., Piersimoni, C. 2002. Evaluation of the BACTEC MGIT 960 in comparison with BACTEC 460 TB for detection and recovery of mycobacteria from clinical specimens. Diagnostic Microbiology and Infectious Disease 44, 157?161. Schwikkard, S., van Heerden, F.R. 2002. Antimalarial activity of plant metabolites. Natural Products Report 19, 1?19. Seaman, T. 2005. The Antimicrobial and antimycobacterial Activity of Plants Used for the Treatment of respiratory Ailments in Southern Africa and the Isolation of 223 Anacardic Acid from Ozoroa paniculosa. MSc dissertation, Faculty of Helath Sciences, University of the Witwatersrand, South Africa. Serrato-Valenti, G., Bisio, A., Cornara, L., Ciarallo, G. 1997. Structural and histochemical investigation of the glandular trichomes of Salvia aurea L. leaves and chemical analysis of the essential oil. Annals of Botany 79, 329?336. Shoemaker, M., Hamilton, B., Dairkee, S.H., Cohen, I., Campbell, M.J. 2005. In vitro anticancer activity of twelve Chinese medicinal herbs. Phytotherapy Research 19, 649?651. Siddiqi, S.H., Libonati, J.P., Middlebrook, G. 1981. Evaluation of a rapid radiometric method for drug susceptibility testing of Mycobacterium tuberculosis. Journal of Clinical Microbiology 13, 908?912. Sikkema, J., De Bont, J.A.M., Poolman, B. 1995. Mechanisms of membrane toxicity of hydrocarbons. Microbiological Reviews 259, 2201?2222. Sindiga, I. 1994. Indigenous (medical) knowledge of the Maasai. Indigenous Knowledge and Development Monitor 2, 16?18. Slinkard, K., Singleton, V.L., 1977. Total phenol analyses. Automation and comparison with manual methods. American Journal of Enology and Virticulture 26, 49?55. Spring, O. 2000. Chemotaxonomy based on metabolites from glandular trichomes. Advances in Botanical Research 31, 153?174. Stace, C.A. 1965. Cuticular studies as an aid to plant taxonomy. British Museums Natural History (Botany) 4, 1?78. Steenkamp, V. 2000. Analytical and Clinical Studies of Toxicity Caused by Traditional South African Medicines. PhD thesis, Faculty of Health Sciences, University of the Witwatersrand, South Africa. Steenkamp, V., Gouws, M.C. 2006. Cytotoxicity of six South African medicinal plant extracts used in the treatment of cancer. South African Journal of Botany 72, 630?633. Suffness, M., Pezzuto, J.M. 1990. Assays related to cancer drug discovery. In: Hostettmann, K. (ed), Methods in Plant Biochemistry: Assays for Bioactivity. Academic Press, London, pp. 71?133. Svoboda, K.P., Svoboda, T.G. 2000. Secretory Structures of Aromatic and Medicinal Plants: A Review and Atlas of Micrographs. Microscopix Publications, U.K. Tanaka, T., Morimoto, S., Nonaka, G., Nishioka, I., Yokozawa, T., Chung H.Y., Oura, H. 1989. Magnesium and ammonium-potassium lithospermates B, the active 224 principles having a uremia-preventive effect from Salvia miltiorrhiza. Chemical and Pharmaceutical Bulletin 37, 340?344. Tepe, B., Daferera, D., Sokmen, A., Sokmen, M., Polissiou, M. 2005. Antimicrobial and anti-oxidant activities of the essential oil and various extracts of Salvia tomentosa Miller (Lamiaceae). Food Chemistry 90, 333?340. Tepe, B., Donmez, E., Unlu, M., Candan, F., Daferera, D., Vardar-Unlu, G., Polissiou, M., Sokmen, A. 2004. Antimicrobial and anti-oxidative activities of the essential oils and methanol extracts of Salvia cryptantha (Montbret et Aucher ex Benth.) and Salvia multicaulis (Vahl). Food Chemistry 84, 519?525. Tepe, B., Sokmen, M., Akpulat, H.A, Sokmen, A. 2006. Screening of the anti-oxidant potentials of six Salvia species from Turkey. Food Chemistry 95, 200?204. Theobald, W.L., Krahulik, J.L., Rollins, R.C. 1979. Trichome description and classification. In: Metcalfe, C.R., Chalk, L. (eds), Anatomy of Dicotyledons. Clarendon Press, London, pp 5?53. Thompson, K.F., Wilkinson, S.M. 2000. Allergic contact dermatitis to plant extracts in patients with cosmetic dermatitis. British Journal of Dermatology 142, 84?88. Thorsen, M.A., Hildebrandt, K.S. 2003. Quantitative determination of phenolic diterpenes in rosmary extracts, aspects of accurate quantification. Journal of Chromatography A 995, 119?125. Tom?s-Barber?n, F., Iniesta-Sanmart?n, E ., Tom?s-Lorente, F., Rumbero, A. 1990. Antimicrobial compounds from three Spanish Helichrysum species. Phytochemistry 29, 1093?1095. Topcu, G., Tan, N., Kokdil., G., Ulubelen, A. 1997. Terpenoid from Salvia glutinosa. Phytochemistry 45, 1293?1294. Tosun, F., Aky?z Kizilay, C., ?ener, B., Vural, M., Palittapongarnpim, P., 2004. Antimycobacterial screening of some Turkish plants. Journal of Ethnopharmacology 95, 273?275. Trager, W., Jensen, J.B. 1976. Human malaria parasites in continuous culture. Science 190, 792?794. Tyler, V.E., Brady, L.R., Robbers, J.E. 1988. Pharmacognosy. 9th edition. Lea and Febiger, Philadelphia, USA. Tzakou, O., Pitarokili, D., Chinou, I.B., Harvala, C. 2001. Composition and antimicrobial activity of the essential oil of Salvia ringens. Planta Medica 67, 81?83. 225 Ulubelen, A., Evren, N., Tuzlaci, E., Johansson, C. 1988. Diterpenoids from the roots of Salvia hypargeia. Journal of Natural Products 51, 1178?1183. Ulubelen, A., Oksuz, Kolak U., Johansson, C., Celik, C., Voelter, W. 2000. Antibacterial diterpenes from roots of Salvia viridis. Planta Medica 66, 458?462. Ulubelen, A., Oksuz, S., Topcu, G., Goren, A.C. and Voelter, W. 2001. Antibacterial diterpenes from the roots of Salvia blepharochlaena. Journal of Natural Products 64, 549?551. Ulubelen, A., Topcu, G., Chai, H-B., Pezzuto, J.M. 1999. Cytotoxic activity of diterpenoids isolated from Salvia hypargeia. Pharmaceutical Biology 37, 148?151. Ulubelen, A., Tuzlaci, E. 1990. Flavonoids and triterpenoids from Salvia euphratica and S. longipedicellata. Fitoterapia 61, 185. Urones, J.G., Marcos, I.S., Diez, D., Cubilla R, L. 1998. Tricyclic diterpenes from Hyptys dilate. Phytochemistry 48, 1035?1038. Valenti, A., Pellissier, Y., Benoit, F., Marion, C., Kone, D., Mallie, M., Bastide, J.M., Bessiere, J.M. 1995. Composition and antimalarial activity in vitro of volatile components of Lippia multiflora. Phytochemistry 40, 1439?1442. Valeriote, F., Corbett, T., Baker, L. 1990. Drug discovery: In Valeriote, F., Corbett, T.H., Baker, L.H. (eds), Cytotoxic Anticancer Drugs: Models and Concepts for Drug Discovery and Development. Kluwer Academic Publishers, Boston, pp. 1?9. Valeriote, F., Grieshaber, C.K., Media, J., Pietraszkewlcz, H., Hoffmann, J., Pan, M., McLaughlin, S. 2002. Discovery and development of anticancer agents from plants. Journal of Experimental Therapeutics and Oncology 2, 228?236. van Vuuren, S.F., Viljoen, A.M., van Zyl, R.L., van Heerden, F.R., Ba?er, K.H.C. 2006. The antimicrobial, antimalarial and toxicity profiles of helihumulone, leaf essential oil and extracts of Helichrysum cymosum (L.) D. Don subsp. cymosum. South African Journal of Botany 72, 287?290. van Wyk, B.E., van Oudtshoorn, B., Gericke, N. 2000. Medicinal Plants of South Africa. 2nd edition. Briza Publications, Pretoria. van Zyl, R.L., Seatlholo, S.T., van Vuuren, S.F., Viljoen, A.M. 2006. The biological activities of 20 nature identical essential oil constituents. Journal of Essential Oil Research 18, 129?133. van Zyl, R.L., Viljoen, A.M. 2002. In vitro activity of Aloe extracts against Plasmodium falciparium. South African Journal of Botany 68, 106?110. 226 van Zyl, R.L., Viljoen, A.M. 2003. Antimalarial activity and essential oil composition of South African medicinal aromatic plants. South African Journal of Botany 69, 265. Velioglu, V.S., Mazza, G., Gao., L, Oomah, B.D. 1998. Anti-oxidative activity and total phenolics in selected fruits, vegetables and grain products. Journal of Agricultural and Food Chemistry 46, 4113?4117. Viljoen, A., van Vuuren, S., Ernst, E., Klepser, M., Demirci, B., Ba?er, H., van Wyk, B.E. 2003. Osmitopsis asteriscoides (Asteraceae): the antimicrobial activity and essential oil composition of a Cape-Dutch remedy. Journal of Ethnopharmacology 88, 137?143. Villa?o, D., Fern?ndez-Pach?n, M.S., Troncoso, A.M., Garc?a-Parrilla, M.C. 2004. The anti-oxidant activity of wines determined by the ABTS?+ method: influence of sample dilutions and time. Talanta 64, 501?509. Voigt, W. 2005. Sulforhodamine B assay and chemosensitivity. Methods in Molecular Medicine 110, 39?48. Voirin, B. 1983. UV spectral differentiation of 5-hydroxy- and 5-hydroxy-3- methoxyflavones with mono-(4?), di-(3?,4?) or tri-(3?,4?,5?)-substituted B. rings. Phytochemistry 22, 2107?2145. Wagner, G.J. 1991. Secreting glandular trichomes: more than just hairs. Plant Physiology 96, 675?679. Wagner, H. Bladt, S. 1996. Plant Drug Analysis. A Thin Layer Chromatography Atlas. 2nd edition, Spring-Verlag, New York. Wagstaff, A. 2002. ?Poverty and health sector inequalities? 80, Bulletin of the World Health Organization. Wall, M.E., Wani, M.C. 1996. Camptothecin and taxol: from discovery to clinic. Journal of Ethnopharmacology 51, 239?254. Wanasundara, U.N., Shahidi, F. 1998. Anti-oxidant and pro-oxidant activity of green tea extract in marine oils. Food Chemistry 63, 335?342. Wanyoike, G.N., Chhabra, S.C., Lang?at-Thoruw a, C.C., Omar, S.A. 2004. Brine shrimp toxicity and antiplasmodial activity of five Kenyan medicinal plants. Journal of Ethnopharmacology 90, 129?133. Watt, J.M., Breyer-Brandwijk, M.G. 1962. The Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E and S. Livingstone, Edinburgh. Weng, X.C., Wang, W. 2000. Anti-oxidant activity of compounds isolated from Salvia plebeia. Food Chemistry 71, 489?493. 227 Wenkert, E., Fuchs, A., McChesney, J.D. 1964. Chemical artifacts from the family Labiatae. Journal of Organic Chemistry 30, 2931?2934. Werker, E, Ravid, U., Putievsky, E. 1985. Structure of glandular hairs and identification of the main components of their secreted materials in some species of the Labiatae. Israel Journal of Botany 34, 31?45. Werker, E. 2000. Trichome diversity and development. In: Hallahan, D.L., Gray, J.C. (eds), Advances in Botanical Research: Plant Trichomes. Academic Press, London, pp. 1?35. White, A.I., Jenkins, G.L. 1942. Salvia carnosa (Dougl.): a phytochemical study. Journal of the American Pharmaceutical Association, Scientific Edition 31, 33?37. WHO (World Health Organization) 2005a. World Malaria Report 2005. www.who/htm/mal (13th February 2006). WHO (World Health Organization) 2005b. TB emergency declaration issued by WHO regional office for Africa. www.who.int/tb/features_archives (20th January 2005). WHO (World Heath Organization) 1998. Roll Back Malaria. www.rbm.who.int/ (15th January 2006). WHO (World Heath Organization) 2006a: Communicable diseases: tuberculosis. WHO, regional office for South East Asia. WHO (World Heath Organization) 2006b. Cancer. www.who.int/cancer/ (29th January 2006). Williamson, E.M. 2001. Synergy and other interactions in phytomedicines. Phytomedicine 8, 401?409. Wink, M. 2003. Evolution of secondary metabolites from an ecological and molecular phylogenetic perspective. Phytochemistry 64, 3?19. Wollenweber, E., D?rr, M., Rustaiyan, A., Roitman, J.N., Graven, E.H. 1992. Exudate flavonoids of some Salvia and a Trichostema species. Zeitschrift fur Naturforschung 47c, 782?784. Woolf, A. 1999. Essential oil poisoning. Clinical Toxicology 37, 721?727. Wu, F.Y.H., Liao, W.C., Chang, H.M. 1993. Comparison of antitumour activity of vitamin K1, K2, and K3 on human tumour cells by two (MTT and SRB) cell viability assays. Life Sciences 52, 1797?1804. Wu, J.W., Lee, M-H., Ho, C-T., Chang, S. 1982. Elucidation of the chemical structures of natural anti-oxidants isolated from rosemary. Journal of the American Oil Chemists' Society 59, 339?344. 228 Yamazaki, K., Yamamoto, T., Kawai, Y., Inoue, N. 2004. Enhancement of antilisterial activity of essential oil constituents by nisin and diglycerol fatty acid ester. Food Micobiology 21, 283?289. Ye, F., Xui, L., Yi, J., Zhang, W., Zha ng, D.Y. 2002. Anticancer activity of Scutellaria baicalensis and its potential mechanism. Journal of Alternative and Complementary Medicine 8, 567?572. Yen, G.C., Chang, Y.C., Su, S.W. 2003. Anti-oxidant activity and active compounds of rice Koji fermented with Aspegillus candidus. Food Chemistry 83, 49?54. Yuan, J.-P., Chen, C.P., Chen, F. 1998. Simultaneous determination of rosmarinic acid, lithospermic acid b and related phenolics in Salvia milthiorrhiza by HPLC. Journal of Agricultural and Food Chemistry 46, 2651?2654. Zainol, M.K., Abd-Hamid, A., Yusof, S., Muse, R. 2003. Anti-oxidative activity and total phenolic compounds of leaf, root and petiole of four accessions of Centella asiatica (L.) Urban. Food Chemistry 81, 575?581. Zakarya, D., Fkih-Tetouani, S., Hajji, F. 1993. Chemical composition, antimicrobial activity relationship of eucalyptus essential oils. Plantes M?dicinales et Phytoth?rapie 26, 331?339. 229 Appendix A: Abstracts of papers published/submitted from this thesis 230 231 232 233 234 235 236 Appendix B: Abstracts of conferences/presentations Appendix B1: (South African Association of Botanists, Durban 2004) Biological activities and essential oil composition of Salvia repens , S. runcinata and S. stenophylla G.P.P. KAMATOU1, A.B. GONO-BWALYA1, A.M. VILJOEN1, A. LOURENS, Y. FRUM, R.L. VAN ZYL1, S.F. VAN VUUREN1, K.H.C. BA?ER2 1Department of Pharmacy and Pharmacology, University of the Witwatersrand, 7 York Road, Parktown 2193, Johannesburg, South Africa 2Department of Pharmacognosy, Faculty of Pharmacy, Anadolu University, 26470 Eski?ehir, Turkey Salvia species have been used medicinally throughout the word. There are about 900 Salvia species of which 26 are found in South Africa. The methanol extracts and essential oils from S. stenophylla, S. repens and S. runcinata were investigated for anti-oxidant (?,?- diphenyl-?-picrylhydrazyl), anti-inflammatory (5-lipoxygenase), antimalarial (tritiated hypoxanthine incorporation), antimicrobial (disc diffusion and microplate methods) and toxicity properties (MTT assay). The results indicated that some extracts and/or essential oils were active with the anti-oxidant IC50 values ranging from 8.98 to 15.07 ppm (methanol extracts) and anti-inflammatory activity from 22.84 to 30.24 ppm (essential oils). Essential oils and extracts of plants possessed antimalarial properties but are potentially toxic. S. stenophylla exhibited the highest activity with the MIC values ranging from 0.10 mg/ml (S. aureus) to 3.13 mg/ml (S. epidermidis). Essential oils have shown little antimicrobial activity against all tested organisms. Analytical techniques (GC-MS, HPLC) were employed to complete chromatographic profiles for the essential oils and methanol extracts, respectively. HPLC results indicate the presence of rosmarinic acid and carnosic acid derivatives. Using bioassay guided fractionation the antimicrobial activity is ascribed to carnosic acid. The results of the study gave the evidence of the traditional use of the members of the S. stenophylla species complex to treat different ailments in some regions of South Africa. 237 Appendix B2 (Indigenous Plant Use Forum, Grahamstown 2005) Biological activities of solvent extracts and essential oils of four related Salvia species G.P.P. KAMATOU1, A.M. VILJOEN1, R.L. VAN ZYL1, S.F. VAN VUUREN1, D.H. DAVIDS1, F.R. VAN HEERDEN2, T. SEAMAN3 1Department of Pharmacy and Pharmacology, University of the Witwatersrand, 7 York Road, Parktown 2193, Johannesburg, South Africa 2School of Chemistry, University of KwaZulu-Natal, Private Bag X01, Scottsville, 3209, Pietermaritzburg, South Africa 3Division of Pharmacology, Faculty of Health Sciences, University of Cape Town, observatory, 7925, cape Town, South Africa Salvia species (sage) are widely used in traditional medicine throughout the world to treat various ailments. In South Africa, they have been used to treat microbial infections, malaria, sore throat and cancer. The solvent extracts and essential oils of four related species; S. africana-lutea, S. africana-caerulea, S. lanceolata and S. chamelaeagnea were investigated for antibacterial, antimalarial, anti-inflammatory, anticancer and anti-oxidant activities. The essential oils were obtained by hydrodistillation of the fresh aerial plant parts. The solvent extract (methanol:chloroform) displayed antibacterial activity against Gram-positive and Gram-negative bacteria with MIC values ranging from 0.03 to 4.67 mg/ml. The extracts also exhibited good anti-oxidant (12.75 ? 1.04 < IC 50 < 68.09 ? 3.69 ?g/ml.), antimalarial (15.86 ? 5.04 < IC 50 < 28.68 ? 2.80 ?g/ml) and anticancer activities (8.72 ? 1.52 < IC 50 < 54.40 ? 4.42 ?g/ml) but poor anti-inflammatory activity (IC50 > 100 ?g/ml). The essential oils exhibited better antimalarial activity (4.76 ? 0.48 < IC 50 < 8.63 ? 0.95 ?g/ml) and anti-inflammatory activity (43.38 ? 0.48 < IC 50 < 77.2 ? 6.27 ?g/ml), antibacterial and poor anti-oxidant activity. The in vitro assessment of the toxicity against human kidney epithelium cells indicated that both essential oils and solvent extracts displayed some degree of toxicity with essential oils being more toxic than the solvent extracts. However, the toxicity of the solvent extract was reduced when tested after extraction of the essential oil. A study of S. chamelaeagnea and Leonotis leonurus in combination showed that when the two plants are used in combination (as reported in the 238 ethnobotanical literature) they have synergistic effects against Gram-positive bacteria and additive or antagonistic effects against Gram-negative bacteria. The bio-autographic- guided fractionation method was employed to isolate the active compound from S. chamelaeagnea against Staphylococcus aureus; and was identified by NMR to be carnosol. 239 Appendix B3 (University of Joh annesburg Symposium, 2005) Trichome morphology and the role of leaf essential oils in the pharmacological activity recorded for indigenous Salvia species G.P.P. KAMATOU1, A.M VILJOEN2, P.M. TILNEY3, S.F. VAN VUUREN1, R.L. VAN ZYL1 1Department of Pharmacy and Pharmacology, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown 2193 2School of Pharmacy, Tshwane University of Technology, Private Bag X680, Pretoria, 0001 3Department of Botany, University of Johannesburg, PO Box 524, Auckland Park, 2006 South Africa is home to 26 Salvia species of which many are used in traditional medicine to treat various ailments. We investigated the leaf trichome morphology of four related species; S africana-caerulea, S. africana-lutea, S. chamelaeagnea and S. lanceolata. The SEM and the LM show both glandular and non-glandular trichomes. Glandular trichomes are single- or multi-celled with short to medium stalk (1-3 cells) and present in all four species. The non-glandular trichomes are unbranched with at least two stalk cells. The leaf trichome morphology cannot be used to set up boundaries among the four species studied. The present work also investigated the in vitro biological activities of extracts in the presence and absence of essential oil to establish the role of the essential oil fraction in the pharmacological activity. The MIC values for the antibacterial activity were best after removal of essential oils against Gram-positive bacteria while the opposite results were obtained for the Gram-negative bacteria. Superior anti-oxidant activity (lower IC50 values) and lower toxicity (high IC50 values) were obtained after the essential oils were removed from the plants. The total phenolic content was also high after removal of the essential oil fraction. The results suggest that essential oils may play an important role (antagonistic or synergistic) in the biological activity of medicinal aromatic plants. This work forms part of a greater project to summarise the chemistry, trichome morphology and biological activity of indigenous aromatic plants in monograph format. 240 Appendix B4 (South African Association of Botanists, Port Elizabeth, 2006) Biological activities, essential oils composition and leaf trichome morphology of four selected Salvia species indigenous to southern Africa G.P.P. KAMATOU1, A.M. VILJOEN2, P.M. TILNEY3, R.L. VAN ZYL1, D.H. DAVIDS1, T. SEAMAN4, A.C. FIGUEIREDO5 1 Department of Pharmacy and Pharmacology, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown 2193 2 School of Pharmacy, Tshwane University of Technology, Private Bag X680, Pretoria, 0001 3 Department of Botany, University of Johannesburg, PO Box 524, Auckland Park, 2006 4Division of Pharmacology, Faculty of Health Sciences, University of Cape Town, Observatory, 7925, Cape Town, South Africa 5Universidade de Lisboa, Faculdade de Ci?ncias de Lisboa, Departamento de Biologia Vegetal, Centro de Biotechnologia Vegetal, C2, Campo Grande, 1749-016 Lisbon, Portugal Salvia species are widely used in traditional medicine to treat various ailments. Twenty- three species are indigenous to southern Africa and have been poorly investigated. The biological activities of the essential oils and the solvent extracts of four related species namely S. africaca-caerulea, S. africana-lutea, S. chamelaeagnea and S. lanceolata were studied. The essential oils were extracted by hydrodistillation and analyzed by GC and GC/MS. The essential oils displayed good antimalarial (IC50 < 8. 6 ?g/ml) and promising anti-inflammatory properties (IC50 < 77 ?g/ml) in comparison to respective controls. The solvent extracts also exhibited potent antimalarial (IC50 < 26 ?g/ml), anticancer (IC50 < 50 ?g/ml), antituberculosis (MIC ? 0.5 mg/ml) activities but poor anti-inflammatory properties (IC50 > 100 ?g/ml). The toxicity of the esse ntial oils was higher compared to that of solvent extracts (P < 0.05) with the IC50 values ranging from 1.87 to 7.24 ?g/ml and from 14.38 to 25.01 ?g/ml respectively. The analysis of the leaf trichome morphology with SEM and LM revealed the presence of the non-glandular and glandular trichomes in all four species. The non-glandular trichome, unbranched, composed of 2 to 3 stalk cells 241 was common to all the species investigated. The glandular trichomes, were single or multicelled in the head, 2-4 short/medium stalk cells were also common to the four species. The composition of the essential oils analysis is also reported and the taxonomical relationship using the essential oils data discussed. 242 Appendix B5 (4th International Conference on Pharmaceutical and Pharmacological Sciences, Gauteng, 2006) The Antimalarial Activity and Cytotoxic Effects of Solvent Extracts of South African Salvia Species and Isolated Compounds from S. radula G.P.P. KAMATOU1, R.L. VAN ZYL1, A.M. VILJOEN2, H. DAVIDS1, F.R. VAN HEERDEN3 1Department of Pharmacy and Pharmacology, University of the Witwatersrand, 7 York Road, Parktown 2193, Johannesburg, South Africa. 2School of Pharmacy, Tshwane University of Technology, Private Bag X680, Pretoria, 0001 3School of Chemistry, University of KwaZulu-Natal, Private Bag X01, Scottsville, 3209, Pietermaritzburg, South Africa Purpose: About 80% of the population in developing countries, including South Africa rely on traditional medicine for their primary care needs. The objectives of this paper are to investigate the in vitro antimalarial and cytotoxic effects of Salvia species which are currently used in traditional medicine in South Africa and to isolate the compounds responsible for the antimalarial activity from the most active species. Method: The hypoxanthine radioisotope and the sulforhodamine B methods used to investigate the antimalarial and anticancer activity respectively. The cytotoxicity of the extracts (methanol:chloroform) were tested on a human kidney epithelium cell line using the 3-(4,5-dimethylthiazol-2-yl)-2,5 dimethyl tetrazolium bromide (MTT) method and on three human cancer cell lines (SF-268, MCF-7 and HT-29). The active compounds were isolated using a combination of TLC and column chromatography and their characterization was done using NMR and mass spectrometry. Results: Solvent extracts displayed antimalarial activity with the IC50 values ranging from 3.91 ? 0.52 to 26.01 ? 2.95 ?g/ml. Among the plants screened, S. radula displayed the most favorable activity, while S. lanceolata was the least active. Since the most active 243 extract was that obtained from S. radula, this species was selected to isolate the compound(s) which may account for the observed biological activity. Two compounds were isolated, identified and characterized as; 5-hydroxy-6,7-dimethoxy-2-(4- methoxyphenyl)-4H-1-benzopyran-4-one and betulafolientriol oxide and tested for antimalarial activity. The plant extracts also exhibited the ability to inhibit cell proliferation against the human kidney epithelial cells (12.12 ? 2.02 < IC 50 < 53.34 ? 3.90 ?g/ml) and three human cancer cell lines. The concentration required to inhibit 50% of cell growth (IC50 values) ranged between 9.69 ? 0.92 and 43.65 ? 8.38 and between 8.72 ? 1.52 and 54.40 ? 4.20 ?g/ml against the MCF-7 and SF-268 cell lines, respectively. Against the HT-29 cell line, the IC50 values ranged from 17.05 ? 3.50 to 57.00 ? 11.67 ?g/ml; with S. lanceolata being the most active plant. S. africana-caerulea was the most active against SF-268 and S. radula showed the best activity against the MCF-7 cell line. Results obtained in this in vitro study may support the traditional use of indigenous Salvia species as coveted ingredient in traditional headlining. 244 Appendix C: Monographs of Salvia species investigated A monograph of all the investigated species are presented in alphabetical order and includes: (i) a brief description, (ii) the geographical distribution, (iii) the traditional uses, (iv) the main components of the essential oil, (v) the phenolic compounds identified in each solvent extract, (vi) the biological activities of the essential oils and solvent extracts, (vii) the leaf trichome types, and (viii) the HPLC profile of the solvent extract (only peak with percentage area greater than 3% is recorded). 245 C1. Salvia africana-caerulea L. Synonyms: S. africana L., S. lanuginose Burm.f., S. integerrima Mill., S. barbata Lm. Common names: ?Bloublomsalie?, wild sage. Botanical description: Shrub up to 2 m tall, often branching at the base with several erect, usually sparingly branched stems. Stems are gland-dotted and occasionally with glandular hairs. Leaves are petiolated, blade simple, obovate-ellicptic to broadly obovate. The calyx is somewhat funnel-shapped, glandular-villous and the corolla is light blue to bluish purple or pinkish, the lower lip usually with a paler blue margin and white to yellowish in the centre (Codd, 1985). Distribution: This plant is distributed from the Vanrhynsdorp district to Cape Town and eastwards to Montagu and Caledon with an old record from Peddi district in the Eastern Cape Province, common in coastal fynbos and rocky slopes (Codd, 1985). Petiolate and obovate leaves and flowers Distribution of S. africana-caerulea. of S. africana-caerulea. Traditional uses: The infusion or decoction is used as a remedy for coughs, colds, chest problems, urine troubles, kidney infections, stomach ailments (Watt and Breyer-Brandwijk 1962; Hutchings et al.,1996; van Wyk et al. (1997). Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. 246 Scanning electron microscopy and light microscopy micrographs of S. africana-caerulea. Essential oil composition: The essential oil yield of this plant collected in the South Western Cape was 0.12% (w/w, wet biomass) and the major components of this oil include spathulenol (29.1%), ?-caryophyllene oxide (14.6%) and ledol (6.5%). Spathulenol ?-Caryophyllene oxide Ledol HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, 7-O- methylepirosmanol, salvigenin, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were detected. The HPLC chromatogram and the tentative of identification are displayed in below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. africana-caerulea. AU 0.00 2.00 4.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 H H OH R O H HO H 247 A summary of the HPLC-UV data of the methanol:chloroform extract of S. africana- caerulea. Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS?+ DPPH? MTT FCM SE 8.72 23.36 27.10 > 100.0 22.68 39.72 33.40 14.38 EO nd nd nd 73.47 4.76 > 100.0 > 100.0 1.87 115.06 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are displayed below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 4.67 6.00 3.00 0.50 EO 16.00 8.00 0.75 4.00 nd nd: not determined References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Hutchings, A., Scott, A.H., Lewis, G., Cunningham, A. 1996. Zulu Medicinal Plants ? an Inventory. University of Natal Press. Pietermaritzburg. van Wyk, B.E., van Oudtshoorn, B., Gericke, N. 1997. Medicinal Plants of South Africa. Briza, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 15.55 218, 324 Caffeic acid 5.29 2 23.27 329 Rosmarinic acid 22.90 3 37.08 204, 283 11.55 248 C2. Salvia africana-lutea L. Synonyms: S. aurea L., S. colorata L., S. eckloniana Benth. Common names: ?Bruinsalie?, ?beach Salvia?, ?Dune Salvia?, ?golden Salvia? Botanical description: Branched shrub up to 2 m tall; stems are densely leafy. Leaves are petiolate; blade simple, grey-tomentosed, gland dotted, apex rounded to obtuse. The calyx is purplish and membranous with short spreading glandular and non-glandular hairs dotted with orange-red gland-dots. The corolla is golden brown, reddish brown khaki or sometimes purplish. The flowering begins in early spring and the bright yellow flowers soon fade to rusty-orange and then reddish brown (Codd, 1985). Distribution: The distribution of S. africana-lutea extends from the coast of Namaqualand to the Cape Peninsula and eastwards from there to Port Alfred. In its natural state, it grows not far from the sea and is often a common constituent of the vegetation on coastal sand dunes (Codd, 1985). Golden brown flowers of S. africana-lutea. Distribution of S. africana-lutea. Traditional uses: An infusion of the plant is used for coughs, colds and female ailments (Watt and Breyer-Brandwijk, 1962). Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. 249 Scanning electron microscopy and light microscopy micrographs of S. africana-lutea. Essential oil composition: The essential oil yield of this plant collected in the South Western Cape was 0.15% (w/w, wet biomass) and the major components in the oil include myrcene (11.5%), p-cymene (7.6%) and ?-pinene (6.0%). Myrcene p-Cymene ?-Pinene HPLC analysis of the solvent extract: Rosmarinic acid, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. 250 HPLC chromatogram of the solvent extract of S. africana-lutea. A summary of the HPLC-UV data of the methanol:chloroform extract of S. africana-lutea. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 10.81 281, 374 3.47 2 16.03 202, 281, 391 3.12 3 19.79 273 3.51 4 21.30 255, 283, 345 8.58 5 23.21 311, 326 Rosmarinic acid 23.82 6 24.28 286, 310 3.10 7 30.31 202, 285, 328 9.20 8 35.68 205, 282 7.46 9 36.53 284, 391 3.2 10 37.05 208, 283 Carnosic acid 14.22 11 41.17 245, 277, 374 4.49 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS?+ DPPH ? MTT FCM SE > 100.0 43.65 20.00 > 100.0 15.86 30.35 47.58 25.01 EO nd nd nd 77.32 5.45 > 100.0 > 100.0 7.24 67.83 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 1.00 2.00 3.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 5 6 7 8 9 10 11 251 The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are displayed in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 3.00 3.00 0.75 0.75 0.50 EO > 32.00 8.00 2.33 8.00 nd nd: not determined References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. 252 C3. Salvia albicaulis Benth. Synonym: S. dregeana Benth. Botanical description: Shrub or woody herb up to 0.3-0.6 m tall and branched from the base. Stems erect, densely and shortly tomentose with occasional multicelular hairs. Leaves are petiolated, blade simple and coriaceous. The inflorescence is a panicle. Calyx is densely villous with long white hairs mainly along the nerves and the corolla is purplish (Codd, 1985). Flowering time is November to May. Distribution: This species is distributed from Clanwilliam southwards to Ceres and Wellington districts, common in fynbos on rocky slopes (Codd, 1985). Distribution of S. albicaulis Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. albicaulis. 253 Essential oil composition: The essential oil yield of this plant collected at the Kirstenbosch Botanical Garden was 0.06% (w/w, wet biomass) and the major components include viridiflorol (24.5%), 1,8-cineole (9.4%), limonene (9.4%), ledol (6.6%) and ?- caryophyllene oxide (5.6%). Viridiflorol 1,8-Cineole Limonene Ledol ?-Caryophyllene oxide HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, salvigenin, carnosol, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table which follows. HPLC chromatogram of the solvent extract of S. albicaulis. A summary of the HPLC-UV data of the methanol:chloroform extract of S. albicaulis. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 10.67 282, 327, 357 9.54 2 22.48 228, 284, 327 6.36 3 23.24 330 Rosmarinic acid 35.23 4 24.25 230, 256, 286 14.58 5 41.08 373 10.23 AU 0.00 0.20 0.40 0.60 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 3 4 2 5 1 O R O H HO H H OH H 254 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table, which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE 27.50 35.25 25.33 > 100.0 15.83 24.22 19.85 37.29 EO nd nd nd 39.22 6.41 > 100.0 > 100.0 2.75 100.27 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method, nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are depicted in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 4.67 1.00 1.00 0.50 EO > 32.00 8.00 2.33 8.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 255 C4. Salvia aurita L. f. Synonyms: S. sylvicola Burch., S. lasiostachys Benth., S. peglerae Skan Botanical description: Perennial herb with few to several stems from a subwoody base, stems ascending or straggling, pilose, usually with longish multicelular hairs, sometimes dense, or short and dense and occasionally with capitate glandular hairs. Leaves are petiolated, blade simple to lyrate or runcinate and variable in shape. Apex obtuse to rounded, base truncate to auriculate. Calyx pilose, tubular campanulate, corolla pale blue, lilac or white pinkish (Codd, 1985). Distribution: This plant is distributed from the southern and south Eastern and KwaZulu- Natal to the Southpansberg in Gauteng, on grassy slopes, stream banks and wooded places (Codd, 1985). Pilose stem and petiolate leaves of S. aurita. Distribution of S. aurita. Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. aurita. HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are given in the table. 256 HPLC chromatogram of the solvent extract of S. aurita. A summary of the HPLC-UV data of the methanol:chloroform extract of S. aurita. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 10.67 282, 341, 355 5.63 2 23.29 330 Rosmarinic acid 24.44 3 32.44 235, 285, 333 3.06 4 34.90 206, 284, 319 Carnosol 27.61 5 37.05 285 Carnosic acid 6.37 6 37.74 205, 283 20.74 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in table, which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE 44.87 17.28 24.58 > 100.0 8.92 22.94 16.59 28.31 119.09 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.50 1.00 1.50 2.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 6 5 257 The MIC values (mg/ml) of the solvent extract against various bacteria are given in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 4.00 0.20 0.05 0.50 Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 258 C5. Salvia chamelaeagnea Berg. Synonym: S. paniculata L. Common name: ?Afrikaansesalie? Botanical description: Salvia chamelaeagnea forms a dense shrub, growing almost 2 meters high. The stems are square-shaped, a typical feature of the Lamiaceae. The leaves are dotted with glands which emit a very strong scent. The large calyx is usually an attractive reddish-purple colour. The corolla usually called the flower varies in colour from blue, mauve, pink to pure white. Flowers with a dark blue top lip and white lower lip are the most common (Codd, 1985). Flowering time is from October to May. Distribution: This plant is distributed from Clanwilliam to Cape Town and eastwards to Ladismith and Riversdale districts, but also found in fynbos along watercourses, in sandy soil among rocks and along roadsides (Codd, 1985). Petiolate and elliptical leaves of Large panicle inflorescence Distribution of S. S. chamelaeagnea. of S. chamelaeagnea. chamelaeagnea. Traditional uses: Like many of the other sages, Salvia chamelaeagnea is used to treat a wide variety of ailments such as coughs, symptoms associated with cancer, colds, diarrhoea, colic, heartburn and flatulence. A tea, to be sipped slowly a few times a day, is usually made from the fresh or dried leaves and flowers (Watt and Breyer-Brandwijk, 1962). Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. 259 Scanning electron microscopy and light microscopy micrographs of S. chamelaeagnea. Essential oil composition: The essential oil yield of the plant collected at the Kirstenbosch Botanical Garden was 0.39% (w/w, wet biomass) and the major components for the oil include 1,8-cineole (40.5%) followed by ?-pinene (10.4%) and limonene (9.7%). 1,8-Cineole ?-Pinene Limonene HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, oleanolic acid, 7-O- methylepirosmanol, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table which follows. HPLC chromatogram of the solvent extract of S. chamelaeagnea. O AU 0.00 0.50 1.00 1.50 2.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 260 A summary of the HPLC-UV data of the methanol:chloroform extract of S. chamelaeagnea. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 10.63 282, 339, 354 4.71 2 23.23 330 Rosmarinic acid 26.36 3 34.81 206, 284, 324 Carnosol 22.48 4 36.95 203, 285 Carnosic acid 27.25 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS?+ DPPH ? MTT FCM SE 34.98 18.12 29.53 > 100.0 8.71 14.61 12.75 24.76 EO nd nd nd 48.58 8.63 > 100.0 > 100.0 6.00 211.78 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against various bacteria are depicted in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 1.00 3.00 0.03 0.06 0.50 EO > 32.00 12.00 2.00 > 32.00 nd nd: not determined References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. 261 C6. Salvia disermas L. Synonyms: S. rugosa Ait., S. fleckii Gurke, S. fleckii G?rke Common name: Grootblousalie. Botanical description: Herbaceous perennial shrub, up to 1.2 m tall with one or more stems from a woody rootstock. Leaves are often crowded and larger near the base of the plant and petiolated. The inflorescence is composed of 15 or more verticils. The calyx is glandular-hispid to villous, while the corolla is whitish, pale blue or mauve (Codd, 1985). The flowering time is from August to May. Distribution: This plant is distributed in the south-west of Gauteng and Northern Cape through the western Free State and Namaqualand (Codd, 1985) Glandular stems and inflorescence of S. disermas. Distribution of S. disermas. Traditional uses: Treatment of heart problems, high blood pressure, rheumatism and sores; used as a tea (Watt and Breyer-Brandwijk, 1962). Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. 262 Scanning electron microscopy and light microscopy micrographs of S. disermas. HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, salvigenin, ursolic acid and betulafolientriol oxide were identified. Various unidentified flavonoids were detected in the solvent extract. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table which follows. HPLC chromatogram of the solvent extract of S. disermas. A summary of the HPLC-UV data of the methanol:chloroform extract of S. disermas. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.24 330 Rosmarinic acid 3.62 2 32.40 284, 332 3.35 3 34.58 218, 277, 330 Salvigenin 23.76 4 40.40 271 19.85 AU 0.00 0.10 0.20 0.30 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 263 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in the table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS?+ DPPH ? MTT FCM SE 59.12 38.56 26.87 > 100.0 24.17 33.05 55.07 53.34 69.01 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) against various bacteria are given in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 2.00 3.00 0.75 0.50 References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. 264 C7. Salvia dolomitica Codd Common name: ?Kruipsalie? Botanical description: It is a shrub that may reach 2 m in height and branched from the base. Leaves are petiolated, blade simple, elliptic to obovate densely greyish and gland- dotted. The inflorescence is compact. The calyx is broadly campanulate, often purple- tinged, glandular hirsute. The corolla is light pink or lilac with cream or yellow markings on the lower lip (Codd, 1985). Distribution: The plant is restricted to Gauteng, between 1000 and 1500 m altitude (Codd, 1985). Leaves and flowers of S. dolomitica. Distribution of S. dolomitica. Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Essential oil composition: The essential oil yield obtained from the plant collected at the Kirstenbosch Botanical Garden was 0.34% (w/w, wet biomass) and the major components for the oil include geraniol (19.6%), linalyl acetate (19.6%), linalool (16.6%) and ?- terpineol (6.2%). Scanning electron microscopy and light microscopy micrographs of S. dolomitica. 265 Geraniol Linalyl acetate Linalool ?-Terpineol HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table which follows. HPLC chromatogram of the solvent extract of S. dolomitica. A summary of the HPLC-UV data of the methanol:chloroform extract of S. dolomitica. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.23 231 Rosmarinic acid 14.05 2 30.62 252, 268, 341 4.48 3 33.36 252, 269, 344 3.44 4 34.53 237, 276, 329 9.16 5 35.23 242, 326, 373 4.36 6 43.27 249, 273, 355 5.08 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in table below. OH O A c O H O H AU 0.00 0.05 0.10 0.15 0.20 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 5 6 266 AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 37.05 > 100.0 > 100.0 7.62 49.94 > 100.0 40.26 EO nd nd nd 64.58 4.81 > 100.0 > 100.0 7.74 53.04 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are given in table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 3.33 2.00 0.75 0.03 0.10 EO > 32.00 8.00 9.33 12.00 nd nd: not determined References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. 267 C8. Salvia garipensis E. Mey. ex Benth. Synonyms: S. steingroeveri Briq., S. dinteri Briq. Botanical description: Shrub mostly branched up to 1.2 m tall. Stems are glandular- pubescents with longish non-glandular hairs. Leaves are green, glandular-hispid and glandular-dotted. The leaves vary considerably in size, degree of crenation. The calyx is glandular-pubescent and the corolla is white or pale blue to mauve (Codd, 1985). Distribution: This species is distributed from the southern half of Namibia to the adjoining Northern Cape Province (Codd, 1985). Flowers of S. garipensis. Distribution of S. garipensis. Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. garipensis. HPLC analysis of the solvent extract: Rosmarinic acid, salvigenin and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while 268 the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. garipensis. A summary of the HPLC-UV data of the methanol:chloroform extract of S. garipensis. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.24 331 Rosmarinic acid 29.22 2 23.63 252, 347 Flavone 4.76 3 34.57 277, 332 Salvigenin 7.38 4 37.02 273, 400 14.09 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in the table which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 39.44 > 100.0 > 100.0 13.95 41.66 74.15 42.44 45.56 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.10 0.20 0.30 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 269 The MIC values (mg/ml) of the solvent extract (SE) against bacteria are given in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 1.88 2.81 0.75 4.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 270 C9. Salvia lanceolata Lam. Synonyms: S. nivea Thunb., S. hastifolia Benth., S. diversifolia Benth. Botanical description: Branched shrub up to 2 m tall, stems glabrescent often reddish brown. Leaves petiolate, blade simple, linear elliptic to ovate-oblong or oblanceolate with a short dense greywish tomentum or with a sparse indumentum of short broad hairs mainly on the veins and leaf margin, gland-doted, apex acute. Inflorescence usually branched. Calyx fairly densely glandular-hispid. Corolla dull rose to brownish crimson or grey-blue (Codd, 1985). Flowering time September to November. Distribution: This plant is distributed from Namaqualand to the Cape Peninsula and eastwards to Montagu. Found mainly in coastal sandveld or arid fynbos, on sandy soil and rocky hillsides at altitudes of about 300 m (Codd, 1985). Linear-elliptic leaves, inflorescence and distribution map of S. lanceolata Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. lanceolata. 271 Essential oil composition: The essential oil yield for this plant collected in the South Western Cape was 0.10% (w/w, wet biomass) and the major components include spathulenol (18.3%), ?-caryophyllene oxide (14.3%), ?-caryophyllene (5.7%) and ledol (5.2%). Spathulenol ?-Caryophyllene-oxide ?-Caryophyllene Ledol HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, carnosol, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. lanceolata. A summary of the HPLC-UV data of the methanol:chloroform extract of S. lanceolata. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 11.66 279, 329, 355 20.58 2 12.22 229, 278, 329 3.35 3 15.44 242, 298, 324 Caffeic acid 3.29 4 20.20 226, 282, 376 3.70 5 22.49 233, 284, 329 3.77 6 23.26 235, 329 Rosmarinic acid 10.67 7 30.66 244, 269, 340 7.57 H H O H R O H H H HO H AU 0.00 0.05 0.10 0.15 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 7 1 2 3 4 5 6 9 10 8 272 Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 8 32.37 239, 284, 332 6.34 9 41.09 273, 374 Ursolic acid 14.28 10 43.26 248, 373, 374 3.83 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 26.15 17.05 > 100.0 26.01 25.63 68.09 26.71 EO nd nd nd 43.38 7.83 > 100.0 >100.0 3.66 54.22 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay ,while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are displayed below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 2.00 3.00 2.00 0.50 EO 32.00 16.00 1.00 8.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 273 C10. Salvia muirii L. Bol. Synonym: S. muirii var. gradiflora L. Bol. Common names: ?Wild sage?, ?wildesalie?. Botanical description: Salvia muirii is a small, twiggy, evergreen shrub that grows only about 30 cm high. The leaves are leathery, oval in shape and light green to almost grey in colour. When crushed, they are slightly sticky and release a light scent that is reminiscent of Vicks. The calyx is densely and shortly antrose pubescent, dotted with orange-red glands (Codd, 1985). Distribution: This species is found in the Langberg in the Riversdale district to Great Brak River in the Mossel Bay district (Codd, 1985). Oval leaves and flowers of S. muirii. Distribution of S. muirii Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. muirii. Essential oil composition: The essential oil yield of this plant collected at the Kirstenbosch Botanical Garden was 0.50% (w/w, wet biomass) and the major components include 1,8-cineole (23.2%), ?-pinene (22.3%) followed by limonene (11.6%) and camphene (7.6%). 274 1,8-Cineole ?-Pinene Limonene Camphene HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, salvigenin, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table which follows. HPLC chromatogram of the solvent extract of S. muirii. A summary of the HPLC-UV data of the methanol:chloroform extract of S. muirii. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.23 330 Rosmarinic acid 52.82 2 34.82 232, 284, 327 Carnosol 4.83 3 36.98 285, 327, 371 Carnosic acid 11.80 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are displayed in the table below. O AU 0.00 0.50 1.00 1.50 2.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 275 AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 39.07 55.63 > 100.0 11.87 11.88 11.06 37.00 EO nd nd nd > 100.0 5.93 > 100.0 > 100.0 3.09 186.15 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are given below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 2.00 3.00 0.25 0.36 0.50 EO > 32.00 6.00 9.33 > 32.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 276 C11. Salvia namaensis Schinz Synonym: S. burchellii N.E.Br. Botanical description: Shrub or bushy shrub up to 1.2 m tall, often herbaceous above and woody below. Leaves are shortly petiolate, blade irregular. The inflorescence is simple, of up to 14 verticils. The calyx and glandular-hispidulous and the corolla is white, mauve or blue (Codd, 1985). Distribution: This species is found in the in the Northern Cape and in the Free Sate on rocky slopes (Codd, 1985). Leaves and flowers of S. namaensis. Distribution of S. namaensis. HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, 7-O- methylepirosmanol, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. namaensis. AU 0.00 1.00 2.00 3.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 4 5 3 277 A summary of the HPLC/UV data of the methanol:chloroform extract of S. namaensis. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.21 330 Rosmarinic acid 15.60 2 31.46 206, 289 9.99 3 34.29 288, 326 4.66 4 34.84 206, 284, 332 Carnosol 30.97 5 36.98 206, 285 Carnosic acid 20.25 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in table which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 36.36 24.39 > 100.0 25.38 16.57 10.63 21.91 190.5 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 278 C12. Salvia radula Benth. Botanical description: Perennial herb with one or more erect stems from a woody rootstock and can reach up to 1 m in height or more. Leaves are often crowded and larger near the base of the plant. The calyx is gland-dotted and the corolla is white or pale mauve to blue. This plant is closely related to S. disermas but can usually be distinguished by its denser lanate indumentum on lower stems and the undersides of the leaves (Codd, 1985). Distribution: This species is found in Gauteng and North-West Province aided by road building and overgrazing of veld (Codd, 1985). Glandular stem and flowers of S. radula. Distribution of S. radula Essential oil composition: The essential oil yield of this plant collected in North-West was 0.004% (w/w, wet biomass) and the major components include ?-caryophyllene oxide (22.6%), humulene epoxide (12.6%) followed by isocaryophyllene (8.6%) and spathulenol (7.7%). ?-Caryophyllene oxide Humulene epoxide Isocaryophyllene Spathulenol HPLC analysis of the solvent extract: Rosmarinic acid, salvigenin and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. R O H H OH H H O 279 HPLC chromatogram of the solvent extract of S. radula. A summary of the HPLC-UV data of the methanol:chloroform extract of S. radula is displayed below. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.26 238, 329 Rosmarinic acid 3.36 2 32.40 269, 338 Flavone 5.65 3 34.56 235, 277, 329 Salvigenin 8.52 4 36.36 272, 338 4.84 5 40.38 271, 373, 392 20.67 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE 27.55 9.69 32.10 78.78 13.50 69.26 > 100.0 20.12 EO nd nd nd > 100.0 3.91 > 100.0 > 100.0 22.90 55.73 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.10 0.20 0.30 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 2 1 3 4 5 280 The MIC values (mg/ml) of the solvent extract (SE) against bacteria are depicted in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE > 16.00 4.00 1.00 0.06 0.10 Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 281 C13. Salvia repens Burch. ex Benth. Synonyms: S. rudis Benth., S. incisa Benth., S. natalensis Benth., S. cooperi Skan, S. subssillis Benth. Common name: ?Kruipsalie?. Botanical description: An easy and fast-growing herb, Salvia repens has rhizomes, which spread to form small clumps of soft; hairy stems about 400 mm tall. Simple or branched, the stems are square (typical of Salvia) and the leaves are formed opposite each other along the stem. The small, two-lipped flowers vary in colour from pure white to pale blue, mauve and purple (Codd, 1985). The main flowering season is during the summer from about October to March. Distribution: Salvia repens is widespread in eastern part of South Africa. It is most common in the grasslands of the highveld and in open woodland (Codd, 1985). Large leaves and flowers of S. repens. Distribution of S. repens. Traditional uses: The leaves have been added to the bath for treating sores, and a decoction of the root has been used for both humans and cattle for treating stomachache and diarrhoea. Smoke from burning the plant can be used as a fumigant (Watt and Breyer- Brandwijk, 1962). Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. 282 Scanning electron microscopy and light microscopy micrographs of S. repens. Essential oil composition: The essential oil yield of this plant collected from Lady Grey was 0.13% (w/w, wet biomass) and the oil has a predominance of 1,8-cineole (12.8%), p- cymene (9.5%) followed by isocaryphyllene (8.6%) and terpinen-4-ol (7.1 %) 1,8-Cineole p-Cymene Isocaryphyllene Terpinen-4-ol HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. repens. H H HO AU 0.00 0.20 0.40 0.60 0.80 1.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 2 1 4 5 3 283 A summary of the HPLC-UV data of the methanol:chloroform extract of S. repens. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.24 329 Rosmarinic acid 23.66 2 26.50 328 13.29 3 32.40 240, 285, 331 Flavone 3.10 4 34.86 206, 284, 321 Carnosol 13.40 5 37.02 285, 372 Carnosic acid 13.01 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in table, which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 23.36 43.62 > 100.0 8.25 18.24 15.47 23.24 EO nd nd nd 28.04 1.65 > 100.0 > 100.0 6.66 178.17 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) against bacteria are displayed in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 2.00 3.00 0.03 0.25 0.50 References Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. Watt, J.M., Breyer-Brandwijk, B.N. 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd edition. E. and S. Livingstone, Edinburg, UK. 284 C14. Salvia runcinata L.f. Synonyms: S. scabra Benth., S. monticola Benth. Botanical description: Perennial erect herb up to 0.7 m tall with one to several stems from a taproot. Leaves, shortly petiolate or the upper ones sessile, blade runcinate-pinnatipartite to lyrate. The inflorescence has several verticils. The calyx is hispid-scabrid, gland-dotted and the corolla is white or pale blue to mauve or purplish (Codd, 1985). Distribution: Distributed to the Free State and Eastern Cape as far south as Bredasdorp district, rare in the Eastern Cape, KwaZulu-Natal and Lesotho, but also found in Gauteng and North-West Province (Codd, 1985). Runcinate-pinnatipartite leaves and pale-blue Distribution of S. runcinata. flowers of S. runcinata. Essential oil composition: The essential oil yield of this plant collected at the Klerkskraal Dam was 0.20% (w/w, wet biomass) and this oil was largely dominated by ?-bisabolol (65.5%) and ?-caryophyllene (10.5%). HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, salvigenin, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is H H O H ?-Bisabolol ?-Caryophyllene 285 displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. runcinata. A summary of the HPLC-UV data of the methanol:chloroform extract of S. runcinata. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 23.29 329 Rosmarinic acid 38.40 2 35.25 237, 288, 373 3.55 3 37.10 236, 283, 328 Carnosic acid 3.36 4 37.83 244, 282 14.34 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 43.42 55.37 > 100.0 16.61 19.42 19.34 22.00 EO nd nd nd 22.81 1.20 > 100.0 > 100.0 1.79 149.34 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.50 1.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 286 The MIC values (mg/ml) of the solvent extract (SE) and essential oil (EO) against bacteria are depicted in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 2.00 4.00 0.03 0.25 0.50 EO > 32.00 > 32.00 2.00 > 32.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 287 C15. Salvia schlechteri Briq. Synonym: S. monticola Benth. Botanical description: Herbaceous or perennial herb up to 0.3 m tall usually branched from a woody base. Leaves are sessile or subsessile, pinnatisect or pinnatifid and relatively large. The inflorescence is composed of several to many verticils. Calyx is gland-dotted and the corolla, wide-throated is pale blue and white (Codd, 1985). The flowering time starts in January and ends in March. Distribution: This species is generally found in disturbed area and well drained soil, and narrow endemic to Eastern Cape (Codd, 1985). Pinnatifid or pinnatisect leaves and flowers Distribution of S. schlechteri. of S. schlechteri. Leaf trichome types: Both non-glandular and glandular trichomes are present on the leaf. The glandular trichomes are either peltate or capitate. Scanning electron microscopy and light microscopy micrographs of S. schlechteri. HPLC analysis of the solvent extract: Caffeic acid, rosmarinic acid, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. 288 HPLC chromatogram of the solvent extract of S. schlechteri. A summary of the HPLC-UV data of the methanol:chloroform extract of S. schlechteri. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 20.75 269, 341 Flavone 3.75 2 23.24 330 Rosmarinic acid 26.58 3 26.19 329 31.09 4 34.87 206, 284, 321 Carnosol 10.88 5 37.02 206, 285 Carnosic acid 10.38 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in the table which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE 54.40 18.37 57.00 > 100.0 17.51 17.51 1.61 28.16 209.18 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.50 1.00 1.50 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 4 5 289 The MIC values (mg/ml) of the solvent extract (SE) against bacteria are depicted in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 3.00 0.75 0.25 0.50 Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 290 C16. Salvia stenophylla Burch. ex Benth. Synonyms: S. xerobia Briq., S. chlorophylla Briq, S. pallida Dinter Botanical description: It is a perennial erect herb that can reach 0.6 m tall and usually much branched from a woody taproot. The leaves are shortly petiolated, linear oblong to oblong-lanceolate. The calyx is hispidulous, gland dotted and the corolla is pale blue or mauve (Codd, 1985). Distribution: The plant is common in Gauteng, Free State and rarely found in KwaZulu- Natal (Codd, 1985). Pinnatifid or pinnatisect leaves of S. stenophylla. Distribution of S. stenophylla. Essential oil composition: The essential oil yield of this plant collected at the East of Clarens was 0.25% (w/w, wet biomass) and the major components of this oil were ?- bisabolol (26.1%) and ?-3-carene (22.6%). ?-Bisabolol ?-3-Carene HPLC analysis of the solvent extract: Rosmarinic acid, carnosol, carnosic acid, ursolic acid and betulafolientriol oxide were identified. The HPLC chromatogram is displayed in the figure below, while the retention time, the absorption maxima and the percentage area of the major peaks are depicted in the table below. O H 291 HPLC chromatogram of the solvent extract of S. stenophylla. A summary of the HPLC-UV data of the methanol:chloroform extract of S. stenophylla. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 10.65 282, 316, 344 5.31 2 23.22 330 Rosmarinic acid 22.17 3 24.23 287, 254, 310 4.10 4 34.86 206, 284, 323 Carnosol 23.60 5 37.00 203, 285 Carnosic acid 25.17 6 37.77 282 6.33 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the essential oil (EO), as well as the total phenolic content are given in the table below. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE 43.86 23.74 17.41 > 100.0 6.50 20.84 14.92 12.12 EO nd nd nd 49.17 4.13 > 100.0 > 100.0 1.98 161.28 - AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method; nd: not determined. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. AU 0.00 0.50 1.00 1.50 2.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 4 5 3 6 292 The MIC values (mg/ml) of the solvent extract and the essential oil against bacteria are given in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 4.00 2.00 0.03 0.06 0.50 EO > 32.00 8.00 1.50 > 32.00 nd nd: not determined Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 293 C17. S. verbenaca L., Synonyms: Horminum verbenaca (L.) Mill., S. cladestina L., S. controversa Ten., S. cleistogama De Bary and Paul Botanical description: It is a perennial short-lived plant with stems arising from woody taproot. The leaves form a dense basal rosette, shortly petiolate to subsessile. The inflorescence is usually branched with many verticils. The calyx is pilose and gland- dotted. The corolla is light blue to purple. Its nearest affinity is with S. disermas and S. radula. Distribution: This plant is distributed in the central and Western Cape Province and it also occurs in the Free State (Codd, 1985). Rosette leaves, inflorescence and glandular stem of S. Distribution of S. verbenaca. verbenaca. HPLC analysis of the solvent extract: Salvigenin, ursolic acid and betulafolientriol oxide were identified. S. verbenaca was the only species devoid of rosmarinic acid. The HPLC chromatogram is displayed in the figure below, while the retention time, the UV spectra and the percentage area of the major peaks are depicted in the table below. HPLC chromatogram of the solvent extract of S. verbenaca. AU 0.00 0.50 1.00 Minutes 0.00 10.00 20.00 30.00 40.00 50.00 60.0 0 1 2 3 294 A summary of the HPLC-UV data of the methanol:chloroform extract of S. verbenaca. Peak number Retention time (min) ?max (nm) Tentative identification Area (%) 1 32.41 220, 285, 332 7.46 2 36.77 272, 333 8.17 3 49.82 272, 327 3.09 Biological activities and total phenolic content: The IC50 values (in ?g/ml) of the in vitro biological activities of the solvent extract (SE) and the total phenolic content are given in the table which follows. AC AI AM AO TO TP SF-268 MCF-7 HT-29 5-LOX 3[H]-hyp ABTS? + DPPH ? MTT FCM SE > 100.0 31.50 50.04 > 100.0 23.97 37.47 45.83 20.85 73.29 AC: anticancer activity; AI: anti-inflammatory activity; AM: antimalarial activity; AO: anti-oxidant activity; TO: toxicity profile: TP: total phenolic content; 5-LOX: 5-lipoxygenase assay; [ 3H]-hyp; hypoxanthine isotope method; MTT: 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay; FCM: Folin- Ciocalteu method. NB: the anticancer activity was determined using the SRB assay, while the anti-oxidant activity was investigated using the ABTS and the DPPH methods. The MIC values (mg/ml) of the solvent extract (SE) against bacteria are displayed in the table below. Pathogens E. coli K. pneumoniae B. cereus S. aureus M. tuberculosis SE 8.00 2.00 2.00 3.00 0.10 Reference Codd, L.E.W. 1985. Lamiaceae: Flora of Southern Africa. 28, Botanical Research Institute, Pretoria. 295 Appendix D: 1H and 13C spectra of compounds isolated from Salvia radula Figure D1 1H spectrum of betulafolientriol oxide. Figure D2 13C spectrum of betulafolientriol oxide. 296 Figure D3 1H spectrum of salvigenin. Figure D4 13C spectrum of salvigenin. 297 Appendix E: 1H and 13C spectra of compounds isolated from Salvia chamelaeagnea Figure E1 1H spectrum of carnosol. Figure E2 13C spectrum of carnosol. 298 Figure E3 1H spectrum of 7-O-methylepirosmanol. v Figure E4 13C spectrum of 7-O-methylepirosmanol. 299 Figure E5 1H spectrum of ursolic/oleanolic acid. Figure E6 13C spectrum of ursolic/oleanolic acid. 300 Appendix F: Seasonal variation in essential oil composition of Salvia africana-caerulea , S. africana-lutea and S. lanceolata Table F-1 Percentage of the components identified in the essential oil of Salvia africana-caerulea over a one year period. Components RI Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec ?-Thujene 924 tr 0.1 1.6 0.1 0.1 tr 1.2 0.3 0.1 0.4 0.1 0.1 Benzaldehyde 927 tr tr tr tr tr tr tr tr tr tr tr tr ?-Pinene 930 1.7 4.2 tr 3.4 4.0 2.6 tr 3.9 3.7 8.2 4.1 4.7 Camphene 938 0.7 0.1 0.1 0.1 0.1 tr 0.1 0.1 0.1 0.1 0.1 0.1 Sabinene 958 0.2 0.2 tr 0.3 0.4 0.2 tr 0.2 0.3 0.1 tr 0.4 1-Octen-3-ol 961 0.2 0.2 tr 0.3 0.4 0.2 0.2 0.2 tr 0.1 tr 0.4 ?-Pinene 963 tr 0.3 tr tr tr tr 0.2 0.3 1.0 2.9 1.3 tr Myrcene 975 0.6 0.8 0.6 0.7 0.6 0.6 tr 0.8 0.6 10.7 0.2 0.8 ?-Phellandrene 995 0.6 0.5 0.4 0.5 0.5 0.3 0.3 0.3 0.2 0.8 - 0.6 p-Cymene 1003 tr 0.1 0.1 0.1 tr 0.2 0.1 0.6 0.2 0.2 1.5 0.2 ??Cymene 0.2 0.4 0.7 0.4 tr 0.2 0.1 0.3 0.2 0.2 - 0.6 1,8-Cineole 1005 tr tr tr tr tr tr tr tr tr tr 0.5 tr ?-Phellandrene 1005 0.4 0.8 tr 0.4 1.0 0.4 tr 0.1 1.2 7.8 - 0.8 Limonene 1009 24.9 31.4 23.4 27.3 28.5 21.4 10.2 29.2 31.5 5.8 1.5 32.6 cis-?-Ocimene 1017 0.1 0.1 0.1 0.2 tr 0.1 tr 0.3 0.2 3.1 tr 0.2 trans-?-Ocimene 1027 0.3 tr tr 0.1 tr tr tr 0.1 0.2 1.0 tr tr ?-Terpinene 1035 tr 0.1 tr 0.1 0.1 0.1 tr 0.1 0.1 0.2 - 0.1 trans-Sabinene hydrate 1037 tr 0.1 tr 0.1 0.1 0.1 tr 0.1 0.2 0.2 tr 0.1 cis-Sabinene hydrate 1066 tr tr tr tr tr tr tr tr tr tr 0.1 tr Linalol 1074 1.5 0.4 1.6 0.6 0.4 0.3 0.4 0.4 0.4 0.3 0.3 0.4 trans-p-2-Menthen-1-ol 1074 tr tr tr tr tr tr tr tr tr tr 0.6 tr Camphor 1095 0.4 0.2 0.1 0.3 0.2 0.3 0.5 tr tr 0.3 - 0.2 ?-Terpineol 1134 - - - - - - - - - - tr - 301 Components RI Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Terpinen-4-ol 1148 8.3 1.4 0.3 0.4 3.2 2.2 7.3 1.7 1.1 1.1 0.5 1.1 ?-Terpineol 1159 1.2 0.4 1.5 1.1 0.4 0.2 1.4 0.2 0.1 0.5 0.7 0.4 trans-Carveol 1189 tr tr tr tr tr tr tr tr tr tr tr tr Carvone 1206 tr tr tr tr tr tr tr tr tr tr tr tr cis-Jasmone 1372 tr tr tr tr tr tr tr tr tr tr tr tr ??Gurjunene 1400 tr tr tr tr tr tr tr tr tr tr tr tr ?-Caryophyllene 1414 0.7 0.4 0.5 0.4 0.5 0.2 0.7 0.3 0.4 0.3 1.9 0.4 ?-Gurjunene* 1426 6.1 1.4 1.3 2.3 3.8 3.9 6.7 2.0 1.3 4.3 0.7 1.0 Aromadendrene 1428 tr 0.5 0.7 tr tr tr 2.2 tr tr tr 0.9 0.6 Eudesmadiene* 1435 tr tr tr tr tr tr tr tr tr tr 0.8 tr ?-Humulene 1447 tr 0.3 0.6 1.4 tr 0.9 tr 1.3 0.5 1.8 1.2 0.4 allo-Aromadendrene 1456 1.5 0.9 0.7 0.1 2.5 0.7 1.3 0.2 1.4 0.5 tr 0.9 Viridiflorene 1487 0.1 1.1 2.2 1.3 1.0 1.8 2.3 1.0 1.5 1.1 0.2 1.2 ??Muurolene 1494 1.0 0.4 1.4 0.6 2.2 2.8 1.1 0.4 0.5 1.1 - 0.4 ??Cadinene 1500 0.6 1.0 0.9 1.2 0.9 0.9 tr 2.4 1.3 0.3 1.5 2.0 trans -Calamenene 1505 0.2 0.8 2.8 3.4 4.4 8.9 tr 0.7 0.3 3.4 0.3 1.0 trans-Nerolidol 1549 1.2 1.9 2.2 2.3 2.5 2.4 2.0 2.4 2.3 0.7 1.4 1.9 ?-Caryophyllene alcohol 1550 1.8 2.3 2.9 3.0 2.6 3.1 1.6 4.0 3.6 1.4 1.4 2.2 Spathulenol 1551 - - - - - - - - - - 29.1 - ?-Caryophyllene oxide 1561 - - - - - - - - - - 14.6 - Globulol 1566 - - - - - - - - - - tr - Viridiflorol 1569 11.1 21.0 24.4 22.8 18.9 20.1 14.2 22.8 21.2 1.6 6.5 20.6 Ledol 1580 6.4 2.6 3.7 3.7 3.7 5.6 7.5 3.4 3.1 1.9 tr 2.2 epi-Cubenol 1600 1.3 0.8 1.3 1.3 1.4 2.0 1.1 1.2 1.2 0.8 tr 0.8 T- Cadinol 1616 0.7 3.4 2.5 3.8 3.4 3.4 2.5 3.9 3.5 1.3 3.0 3.3 ?-Eudesmol 1620 0.4 0.8 3.2 1.0 0.8 0.7 0.7 0.5 0.6 3.8 0.7 0.8 ??Eudesmol 1634 1.4 1.4 3.0 2.3 2.2 2.3 2.7 1.5 1.4 7.7 2.0 1.5 % Identification 75.8 82.8 84.8 87.4 90.8 89.1 68.6 87.2 85.4 76.0 78.0 85.0 RI: Relative retention indices calculated against n-alkanes. - : not present, tr: trace (< 0.05 %); *: determined by mass spectra only. 302 Table F-2 Percentage of the components identified in the essential oil of Salvia africana-lutea over a one year period. Components RI Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec ?-Thujene 924 0.5 0.1 0.4 0.4 0.3 0.3 0.5 0.4 0.2 0.2 0.6 0.1 ?-Pinene 930 11.0 0.5 7.1 8.6 11.6 5.0 11.9 9.8 4.6 3.2 6 1.1 Camphene 938 tr 0.1 tr tr 0.1 tr 0.1 0.1 0.1 0.1 0 0.1 Sabinene 958 tr 0.3 1.4 1.9 3.9 2.7 0.9 0.1 0.1 0.7 0.9 0.1 1-Octen-3-ol 961 tr 0.3 1.4 1.9 3.8 2.7 0.9 0.1 0.1 0.7 0.9 0.1 ?-Pinene 963 3.4 0.2 tr tr tr 1.8 2.9 3.8 1.9 tr 0.2 0.9 Myrcene 975 10.7 3.7 5.0 2.6 4.3 5.9 5.3 8.4 2.7 5.4 11.5 2.2 ?-Phellandrene 995 0.6 0.2 0.8 0.4 0.9 1.1 1.0 0.8 0.8 0.7 - 0.6 ??3?Carene 1000 1.9 3.9 1.1 7.8 1.5 8.3 8.0 3.5 4.4 0.1 - 6.8 ??Cymene 1000 10.3 0.4 8.1 3.0 7.1 0.6 3.4 0.7 0.5 0.3 - 1.1 p-Cymene 1003 2.6 0.2 1.3 0.7 0.4 0.3 0.4 0.2 0.2 0.1 7.6 0.3 1,8-Cineole 1005 5.3 0.5 3.2 2.6 2.4 4.0 4.9 3.5 2.4 2.5 1.9 3.3 ?-Phellandrene 1005 5.3 0.5 3.2 2.6 2.4 4.0 4.9 3.5 2.4 2.5 - 3.3 Limonene 1009 tr 3.4 tr 2.8 tr 6.0 3.7 5.4 3.7 3.3 1.6 5.3 cis-?-Ocimene 1017 1.4 3.0 1.3 tr 0.5 0.5 0.5 1.7 0.2 2.1 5.4 0.2 trans-?-Ocimene 1027 0.5 1.1 0.6 tr 0.1 0.2 0.1 0.6 0.2 0.7 1.8 0.1 ?-Terpinene 1035 tr 0.4 0.1 tr 0.1 0.2 0.2 0.1 0.2 0.1 - 0.1 trans-Sabinene hydrate 1037 tr 0.4 0.1 tr 0.1 0.2 0.2 0.1 0.2 0.1 0 4.7 Linalol 1074 0.6 0.2 0.3 0.2 tr tr tr 0.4 0.7 0.4 0.8 0.8 Camphor 1095 0.2 0.4 0.5 0.4 tr tr tr 0.3 0.2 0.4 0.2 0.3 Terpinen-4-ol 1148 1.3 0.1 0.5 0.7 tr tr tr 0.2 0.2 tr 0.6 1.6 ?-Terpineol 1159 tr 0.3 0.1 tr tr tr tr tr 0.2 tr 0.8 0.3 1-Decanol 1259 0.8 tr 0.4 0.4 0.4 0.6 0.6 0.8 0.5 0.5 - 1.4 Bornyl acetate 1265 0.5 tr tr 0.6 tr tr tr tr 0.1 tr - 0.2 ??Gurjunene 1400 0.5 0.6 4.7 5.0 4.6 4.3 2.5 0.7 0.8 0.4 1.1 0.9 ?-Caryophyllene 1414 2.7 2.0 2.4 1.5 1.1 0.8 0.9 2.1 4.3 2.6 3.4 3.4 ?-Gurjunene* 1426 1.0 3.3 tr tr 0.2 tr 0.2 2.8 1.5 1.0 0.3 1.2 Aromadendrene 1428 1.0 1.1 1.2 0.7 1.4 0.6 0.8 1.4 1.4 0.9 2.7 0.6 ?-Humulene 1447 1.1 1.1 2.4 2.4 1.3 1.8 1.8 1.1 1.9 1.5 1.9 1.7 trans-?-Farnesene 1455 tr 0.4 tr tr 0.7 tr 0.4 0.9 0.2 0.4 tr 0.1 303 Components RI Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec allo-Aromadendrene 1456 tr 0.4 tr tr 0.7 tr 0.4 0.9 0.2 0.4 1.3 0.1 Germacrene-D 1474 0.5 0.9 1.4 0.5 0.4 0.3 0.7 0.5 0.6 1.0 tr 0.8 Bicyclogermacrene 1487 2.2 2.3 1.0 0.8 0.4 0.5 0.2 0.1 0.2 0.5 - 0.4 Viridiflorene 1487 - tr tr tr tr - - - - 1.3 - ?-Bisabolene 1495 0.5 1.8 0.9 0.5 0.5 0.4 0.5 0.5 0.8 3.4 1.3 0.8 ??Cadinene 1505 0.4 2.3 0.6 0.5 0.5 0.4 0.4 0.5 0.6 0.4 4.7 0.5 trans-Calamenene 1505 tr tr tr tr tr tr tr tr tr tr 0.9 tr Elemol 1530 1.6 0.9 1.2 1.3 0.6 0.7 0.8 0.7 1.4 1.8 3.5 2.2 trans-Nerolidol 1549 0.2 0.7 0.7 0.3 0.5 0.2 0.5 0.7 0.6 0.6 2.0 0.3 ?-Caryophyllene alcohol 1550 0.2 0.7 0.4 0.3 0.5 0.2 0.5 0.7 0.6 0.6 2.0 0.3 Spathulenol 1551 3.3 2.8 6.3 5.6 10.9 2.3 4.3 6.8 6.7 3.8 1.9 3.1 ?-Caryophyllene oxide 1561 2.3 2.7 4.1 2.8 2.4 1.9 1.9 2.8 3.5 3.9 0.8 2.6 Humulene epoxide* 1579 0.5 2.6 0.8 0.9 0.4 0.7 0.7 tr 1.3 3.3 0.7 epi-Cubenol 1600 - - - - - - - - - - 1.8 - T- Cadinol 1616 1.3 2.2 1.7 2.7 2.3 3.2 1.8 1.5 2.3 2.2 1.9 1.9 ?-Eudesmol 0.9 4.0 2.4 1.5 2.6 2.9 2.8 2.4 2.2 3.3 - 3.6 ?-Eudesmol 1620 3.2 8.9 2.1 4.6 4.7 1.4 5.3 1.8 9.9 2.1 1.7 12.7 ??Eudesmol 1634 5.0 10.6 8.1 9.5 10.4 10.9 9.3 8.5 12.6 9.1 tr 12.9 % Identification 85.3 72.5 79.3 79.0 87.1 77.9 87.1 81.9 79.1 65.3 78.6 85.8 RI: Relative retention indices calculated against n-alkanes. - : not present, tr: trace (< 0.05 %); *: determined by mass spectra only. 304 Table F-3 Percentage of the components identified in the essential oil of Salvia lanceolata over a one year period. Components RI Jan Mar Apr May Jun July Aug Sep Oct Nov Dec Tricyclene 921 ?-Thujene 924 0.4 0.5 0.5 0.4 0.1 0.1 0.2 0.4 0.4 - 0.2 Benzaldehyde 927 - - - - - - - - - 0.5 - ?-Pinene 930 3.8 tr tr 0.3 1.2 0.8 1.4 0.5 2.7 0.3 Camphene 938 0.1 tr tr tr tr 1.1 tr tr 0.1 2.6 0.1 Sabinene 958 1.9 0.3 0.3 0.2 0.3 0.7 0.8 0.1 1.9 tr 0.1 1-Octen-3-ol 961 1.9 0.3 0.3 0.2 0.3 0.7 0.8 0.1 1.9 0.1 ?-Pinene 963 1.8 tr tr 0.6 0.8 0.8 2.2 0.7 1.3 1.4 0.1 Myrcene 975 3.9 0.7 0.1 1.6 1.0 3.0 1.5 2.0 6.1 - 0.1 ?-Phellandrene 995 0.1 tr 0.1 tr 0.2 0.3 0.2 tr 0.1 - 0.1 ??3?Carene 1000 0.1 tr tr 0.1 0.2 0.1 0.5 0.1 0.1 - 0.4 ??Cymene 0.9 tr tr tr 0.5 0.8 0.3 tr 1.0 - tr p-Cymene 1003 1.1 1.0 0.4 0.9 0.5 0.2 0.7 1.4 0.5 - 0.2 1,8-Cineole 1005 0.7 1.0 0.5 1.3 1.2 1.0 0.5 1.0 1.0 1.1 0.6 ?-Phellandrene 1005 0.7 1.0 0.5 0.7 1.2 1.0 1.4 1.0 1.0 1.8 0.6 Limonene 1009 3.6 tr tr tr tr 1.6 2.1 tr 1.9 tr cis-?-Ocimene 1017 0.8 0.1 tr 1.1 1.5 1.9 3.1 1.6 2.5 1.8 tr trans-?-Ocimene 1027 0.4 tr tr 0.7 1.0 1.3 1.6 1.5 1.9 tr tr ?-Terpinene 1035 tr tr tr tr 0.6 tr 0.1 0.1 0.1 tr tr trans-Sabinene hydrate 1037 tr tr tr tr 0.4 tr 0.2 0.1 0.1 - tr Linalol 1074 0.6 tr 0.3 0.8 0.7 0.3 0.2 tr 1.1 tr 0.2 Camphor 1095 0.4 tr 0.1 0.4 0.7 0.3 0.2 0.3 tr - 0.4 Terpinen-4-ol 1148 0.2 1.0 0.7 0.5 0.6 0.6 0.8 0.7 0.8 - 0.4 ?-Terpineol 1159 tr 0.2 0.5 0.1 0.4 0.5 0.9 0.1 0.2 1.0 0.3 Bornyl acetate 1265 tr tr 1.1 tr tr 0.5 tr tr tr - tr ?-Bourbonene 1373 0.3 1.5 1.1 0.8 2.1 1.2 1.2 1.0 1.3 - 0.8 ??Elemene 1388 0.7 2.0 1.3 2.2 1.8 1.0 1.1 2.6 0.9 - 0.6 ??Gurjunene 1400 0.3 0.4 0.9 0.2 0.3 tr tr 0.2 0.3 - 0.9 ?-Caryophyllene 1414 12.9 13.1 3.1 19.2 18.6 17.8 17.0 18.4 12.7 - 0.7 ?-Gurjunene* 1426 1.6 1.8 1.1 1.5 1.2 2.3 1.4 1.5 0.7 5.7 1.8 Aromadendrene 1428 - 2.9 0.5 3.2 0.2 1.0 0.4 0.1 0.6 ?-Elemene* 1430 0.5 2.0 0.9 0.6 2.0 1.0 1.3 1.2 3.4 - 0.6 305 Components RI Jan Mar Apr May Jun July Aug Sep Oct Nov Dec Geranyl acetone 1434 0.4 0.8 1.9 0.6 2.0 tr 0.9 5.2 5.2 - - ?-trans-Bergamotene 1434 - - - - - - - - - 0.8 - ?-Humulene 1447 5.9 8.4 2.7 1tr 9.7 8.5 8.6 8.5 5.5 tr 1.6 trans-?-Farnesene 1455 - - - - - - - - - 4.7 - allo-Aromadendrene 1456 0.9 0.5 0.5 0.8 0.9 1.0 1.1 0.9 0.6 - 0.6 Germacrene-D 1474 2.2 1.4 0.3 2.7 2.2 1.9 4.3 1.2 3.0 - 1.2 ar-Curcumene 1475 - - - - - - - - - 2.4 - Bicyclogermacrene 1487 4.8 1.3 1.0 3.4 6.0 5.4 7.9 2.5 4.7 - 1.8 ?-Bisabolene 1495 0.9 1.3 0.3 0.9 0.8 1.0 0.9 0.9 0.8 0.2 0.6 ??Cadinene 1500 1.1 0.7 0.6 1.0 0.8 1.0 1.1 0.6 0.8 - 0.3 trans-Nerolidol 1549 0.7 0.9 0.8 0.8 0.7 3.1 0.9 4.6 0.8 - 0.8 Spathulenol 1551 6.5 4.3 10.6 3.1 2.6 2.8 4.0 1.9 2.4 2.2 0.7 ?-Caryophyllene oxide 1561 3.4 7.1 21.2 3.3 2.2 2.0 1.8 2.4 1.4 18.3 10.3 Globulol 1566 - - - - - - - - - 14.3 - Viridiflorol 1569 - - - - - - - - - 0.6 - Ledol 1580 - 4.1 2.9 11.7 4.5 3.6 3.7 4.1 4.3 5.9 - Humulene epoxide* 1579 2.8 6.7 3.6 4.6 4.5 4.4 4.0 4.3 6.5 5.2 10.8 epi-Cubenol 1600 0.8 1.4 0.4 0.9 0.7 0.8 tr 1.0 tr 0.8 T- Cadinol 1616 0.7 0.8 1.3 0.6 0.7 0.9 1.6 0.2 tr 0.6 1.4 ?-Cadinol 1618 - - - - - - - - - 1.9 - ?-Eudesmol 1620 0.9 1.3 tr 1.1 1.0 0.8 3.4 0.2 3.9 tr 1.7 Cubenol 1624 - - - - - - - - - 1.3 - ??Eudesmol 1634 1.2 3.8 1.3 3.3 3.4 2.7 tr 3.3 tr - 1.0 ??Bisabolol 1656 - - - - - - - - - 0.4 - % Identification 77.0 73.4 72.5 79.2 81.4 80.9 86.3 79.6 87.9 69.8 44.7 RI: Relative retention indices calculated against n-alkanes. - : not present, tr: trace (< 0.05 %); *: determined by mass spectra only. NB: Sample from February collection was not done due to the lack of plant material.