Research Journal of Medicinal Plants

Year: 2012 | Volume: 6 | Issue: 5 | Page No.: 370-382
DOI: 10.17311/rjmp.2012.370.382

Antimicrobial Activity and Safety of two Medicinal Plants Traditionally used in Bomet District of Kenya

R.K. Korir, C. Mutai, C. Kiiyukia and C. Bii

Abstract: Indigenous rural communities in the tropics manage microbial diseases using herbal concoctions. The efficacy and safety of most of the herbal preparations have not been proven scientifically. Extracts from two selected plant species were evaluated for in vitro antimicrobial activities and toxicity. Extraction was done using hexane, dichloromethane, methanol and water. Hexane extracts of Senna didymobotrya was active against Microsporum gypseum with 16.0 mm inhibition zone diameters and no activity against other test isolates. Dichloromethane extracts had good activities against Trichophyton mentagrophyte and Microsporum gypseum with 18.0 and 15.0 mm inhibition zones, respectively. Minimum inhibition concentration of Cyathula polycephala extracts against bacterial isolates ranged from 25.0 to 3.124 mg mL^-1. Methanol extracts of C. polycephala had a CC₅₀ of 100.0%. The water extracts recorded CC₅₀ of 23.75 and 31.56% at 1000 and 100 mg mL^-1 concentration, respectively. The methanol extracts recorded a CC₅₀ of 99.34 and 100.0% at 1000 and 100 mg mL^-1 concentration, respectively. The DCM extract of S. didymobotrya was very toxic at 5000 mg kg^-1 and killed 80% of mice. In both the cell and acute toxicity, it was observed that extracts at high concentration and at a high dose tend to be toxic. Most extracts were active against a wide range of bacteria than fungi.

Keywords: toxicity, antimicrobials, in vivo, in vitro, Ethnobotanical and inhibition zones

INTRODUCTION

People all over the world have used plants as medicines from time immemorial. It is estimated by WHO that 80% of the population, majority of this in developing countries, still rely on plant-based medicine for primary health care (Evans, 1997). Herbal drugs are prepared from various parts of the plant such as leaves, stem, roots, seeds, tubers or exudates (Mukherjee, 2002). There are more than 1,340 plants known to be potential sources of antimicrobial compounds but few have been systematically studied scientifically (Wilkins and Board, 1989). Some plant like Heteromorpha trifoliata (Wendl.) Eckl. and Zeyh. is used traditionally to treat stomach problems. It has been shown to have strong activity against several species of bacteria including Escherichia coli, Klebsiella pneumoniae, Proteus vulgaris, Pseudomonas aeruginosa, Salmonella gallinarum and Staphylococcus albus (Desta, 1993; Baker et al., 1995). Although, plants have been extensively used, specific evaluation on toxicity has not been done and could lead to serious complications (Rahman et al., 1996).

Indiscriminate exposure to antibiotics in patients hastens bacteria mutations that bring about drug resistance. This situation produces bacteria with greater ability to survive even the strongest antibiotics (WHO, 2002). More than 70% of the bacteria causing these infections are resistant to at least one of the drugs commonly used to treat them. Persons infected with drug resistant strains are more likely to have longer hospital stay and require treatment with second or third line drugs that may be effective but more toxic and expensive (Brun-Buisson et al., 1987). Drug resistant bacteria have mainly been of nosocomial origin (Brun-Buisson et al., 1987).

The antimicrobial resistance problem has been worsened by HIV pandemic due to emergence of opportunistic infection thereby leading to indiscriminate use of antibiotics (WHO, 2004). The antimicrobial resistance to the available drugs is driving up healthcare costs, increasing the severity of disease, morbidity and mortality. The situation calls for urgent research on any potential source of new effective antimicrobials. Plants products are some of such sources which have not been exhaustively utilized. Many plant preparations are used by the Kipsigis community of Kenya without any available scientific study on efficacy or safety. In this study, selected medicinal plants used by the Kipsigis ethnic community of Kenya, were subjected to a systematic screening for antimicrobial activity, in vitro and in vivo toxicity.

Senna didymobotrya (Fresen) Irwin and Barneby (Caesalpinaceae) is used traditionally by the Kipsigis community for the control of Malaria as well as diarrhea. It is used to treat skin conditions in humans and widely used to treat livestock infections in Kenya as documented by Njoroge and Bussmann (2007). Antimicrobial properties of this plant from Bomet district have not been documented.

Cyathula polycephala auct. belongs to the family Amaranthaceae. It is a genus of medicinal and ornamental plants and the species are distributed in Africa, Asia, Oceania and the Americas. The leaves are burned into ashes and are used to treat diabetes among the Mbeere and Ameru communities in Kenya (Kareru, 2007). Brine shrimp toxicity as well as antiplasmodial properties of methanol extracts of Cyathula polycephala has been documented by Wanyoike et al. (2004). The leaves are burned and the ashes are applied to skin infection while the roots are boiled and used to treat pneumonia by the Kipsigis community. The antimicrobial properties of this plant have not been documented.

MATERIALS AND METHODS

Collection of the plants from the field: The Senna didymobotrya and Cyathula polycephala plants were collected from Bomet District. Selection of the plants was based on available ethnobotanical information from traditional health practitioners consulted during the pilot study as well as available literature. The plants are used to treat skin conditions and stomach problem. The herbalist advised on the samples to be collected and the plant part used for different kinds of ailments. The plants parts commonly used and collected were from the stem barks. The study plant materials were photographed and collected for authentification at the East African Herbarium (Table 1). The samples for extractions were collected in paper bags and delivered to the Medicinal Chemistry Laboratory of the Kenya Medical Research Institute (KEMRI), Centre for Traditional Medicine and Drug Research for processing.

Extraction of plant materials: Plants materials were dried at room temperature, grounded into a fine powder and stored in a cool and dry place. Total extraction of each plant material was prepared by mixing with solvent in the ratio of 1:10 (plant material/solvents).

This procedure was done successively using hexane, dichloromethane and methanol where, the plant materials were soaked in 1000 mL conical flasks for 12 h. The extracts were filtered using Whatman No. 1 filter paper and solvents removed through evaporation under reduced pressure at 45°C. The extracts were kept in stoppered sample vials at 4°C until they were used (Stein et al., 2005; Chhabra et al., 1984). Total water extract of each plant material was done by soaking a weighed amount of the dry powder (27-50 g) in distilled water and shaken for two hours with an electric shaker at 65°C. The suspension was filtered and the filtrate was kept in a deep freezer before it was evaporated to dryness by freeze drying. The lyophilized dry powder was collected in stoppered sample vials, weighed and kept in a desiccator, to avoid absorption of water, until they were used. Table 1 and 2 presented the plants and reagents collected for study.

Test strains: Test strains were chosen based on the ethnobotanical exploitation of the plants. The following are the standard organisms and clinical isolates used in this study.

Bacterial isolates:

Preparation of test organisms: Stocked bacterial strains were sub-cultured on Muller Hinton agar No. CM0337. (Oxoid Ltd, Basingstoke, Hampshire, England). Incubation was done at 37°C for 12-18 h to obtain freshly growing strains. Yeast and molds were subcultured onto Sabouraud Dextrose Agar No. CM 004 (Oxoid Ltd, Basingstoke, Hampshire, England). Each media was prepared according to the manufacturers instructions. Yeasts were incubated for 24 h while molds were incubated for 72 h at 30°C to obtain freshly growing culture (Rajakaruna et al., 2002).

Preparation of McFarland standard: Exactly 0.5 McFarland equivalent turbidity standards was prepared by adding 0.6 mL of 1% barium chloride solution (BaCl₂.2H₂O) to 99.4 mL of 1% sulphuric acid solution (H₂SO₄) and mixed thoroughly. A small volume of the turbid solution was transferred to cap tube of the same type that was used to prepare the test and control innocula. It was then stored in the dark at room temperature (25°C). Exactly 0.5 McFarland gives an equivalent approximate density of bacteria 1x10^-6 Colony Forming Units (CFU) mL^-1 (Stein et al., 2005).

Antimicrobial screening test: From an overnight growth of the test organism, 4-6 colonies were emulsified and the suspension was adjusted to match the 0.5 McFarland's standard. Respective plates were inoculated using a sterile cotton wool swab. Antimicrobial susceptibility test was done using disk diffusion methods. Briefly, 100 mg of each extract was dissolved in 1 mL of the appropriate solvents and 10 μL of the mixture was impregnated onto 6 mm sterile filter paper disk and air dried. The disk was placed aseptically onto the inoculated plates. The bacterial and yeast cultures were incubated at 37°C for 24 and 48 h, respectively. Dermatophytes (molds) cultures were incubated at 25°C for 72 h after which the inhibition zone diameter was measured in millimeters and recorded against the corresponding concentrations as described by Elgayyar et al. (2001). Positive controls were set against standard antibiotics gentamicine 30 μg mL^-1 and antifungal drugs fluconazole 30 μg mL^-1 while negative controls were set using disk impregnated with extraction solvents.

Determination of minimum inhibitory concentration: Broth micro dilution method was used to determine minimum inhibitory concentration for the active crude extracts against the test microorganisms. The procedures were done as recommended by the National Committee for Clinical Laboratory Standards now Clinical Laboratory Standard Institute (CLSI) (Ferraro, 2003). The tests were performed in 96 well-micro-titer plates. Plants extracts dissolved in respective solvents were transferred into micro-titer plates to make serial dilutions ranging from 10¹, 10², 10³……..10¹⁰. The final volume in each well was 100 μL. The wells were inoculated with 5 μL of microbial suspension. The yeast and bacteria were incubated at 37°C for 24 h while molds were incubated at 25°C for 3-7 days in ambient air. The MIC was recorded as the lowest extract concentration demonstrating no visible growth as compared to the control broth turbidity (Michael et al., 2003). Wells that were not inoculated were set to act as control. All the experiments were done in triplicates and average results were recorded.

Cell toxicity: The cytotoxic concentration causing 50% cell lysis and death (CC₅₀) was determined for the extracts by a method described by Kurokawa et al. (2001). The extracts of the active plants were tested for in vitro cytotoxicity, using Human Embryonic Lung Fibroblast (HELF) Vero-199 cells using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Mosmann, 1983). The HELF cells were cultured and maintained in Minimum Essential Medium (MEM) supplemented with 10% Featal Bovin Serum (FBS). The cells were cultured at 37°C in 5% CO₂, harvested by trypsinization, pooled in a 50 mL vial. Approximately 100 μL cell suspension (1x10⁵ cells mL^-1) added to each well in a 96-well micro-titer plate. Each sample was replicated 3 times and the cells incubated at 37° in 5% CO₂ for 24 h for attachment. 150 μL of the highest concentration of each of the test samples (a serial dilution, prepared in MEM) added into the same row and a serial dilution done. The experimental plates were incubated further at 37°C for 48 h. The cells in media without drugs were used as controls. After 48 h of incubation, MTT (10 μL) was added into each well. The cells were incubated for 4 h or until a purple precipitate was clearly visible under a microscope. The medium together with MTT was aspirated off from the wells and dimethlysulfoxide (DMSO) (100 μL) was added and the plates shaken for 5 min. The absorbance for each well was measured at 562 nm in a micro-titre plate reader (Mosmann, 1983) and percentage Cell Viability (CV) calculated via an excel program with the formula:

Determination of acute toxicity: Fifteen Swiss albino mice used in the study were sourced from KEMRI animal house facility. Healthy mice (weight 20-22 g) were divided into two groups (control and treatment group) each cage with five mice. The mice were allowed to have access to water and food, except for a short fasting period of 12 h before oral administration of single dose of the test sample. The active extracts suspension was administered orally at logarithmic dose ranging between 1000-5000 mg kg^-1 b.wt. In this study, two dose levels of 1000.0 and 5000.0 mg kg^-1 body weight were used. The general behavior of mice was observed continuously for 1 h after the treatment and then intermittently for 4 h and thereafter over a period of 24 h (Twaij et al., 1983). The mice were observed further for up to 14 days following treatment for any sign of restlessness and the latency of death. The LD₅₀ value was determined according to a method described by Thompson (1985). During experiment all the dead mice were disposed according to KEMRI biosafety guidelines. After the experiment, all the mice were sacrificed using chloroform and the carcasses incinerated.

Phytochemical evaluation: Phytochemical screening was done on the active extracts to identify the active chemical principles. Based on the initial bioassays the plant extracts exhibiting biological activity were screened for groups of chemical constituents. Thin Layers Chromatography (TLC) plates were developed with Chloroforms: Methanol (98:2) with five drops of glacial acetic acid before spraying with thin layer chromatography visualization reagents giving specific reactions (Harborne, 1998).

Statistical data analysis: The results were subjected to statistical analysis for qualification of variability. Statistical Packages for Social Scientist (SPSS) Version 12.0 was utilized which enabled the Analysis of Variance by one way (ANOVA) to establish the significance variability between and within groups (Plants, Solvents and organisms). Bioactivity was used as an independent variable to establish significance p value of 0.05 at 95% confidence interval. The cytotoxic concentration causing 50% cell lysis and death (CC₅₀) was determined for the extracts. Three experiments were performed for each of the in vivo study. In vivo toxicity was checked for significant differences in lethal dose resulting to fifty percent deaths (LD₅₀) values between the tests and the controls groups.

RESULTS

Yields for sequential extractions: The percentage yields of Senna didymobotrya Hexane bark extract was 0.166% while Cyathula polycephala was 0.987%. Water extracts produced 2.836 and 7.023% yields for Senna didymobotrya and Cyathula polycephala, respectively. The results are as shown in Table 3.

Bioactivities of the plants extracts: The methanol and dichloromethane extracts of Senna didymobotrya had the highest activity compared to water and hexane. Hexane extracts had no activity against S. aureus, MRS. aureus, Escherichia coli and Klebssiela pneumoniae and the water extract had activity of 10.0 mm against Pseudomonas aeruginosa and 11.0 mm against K. Pneumoniae, respectively.

Methanol extract recorded relatively a moderate to low minimum inhibition concentration. The plant extracts had different MIC range for different solvents against different fungal isolates. The results are as shown in Table 4 and 5, respectively.

The methanol extract of Cyathula polycephala was active against methicillin resistant Staphylococcus aureus, S. aureus and P. aeruginosa with inhibition zone diameters of 16.0, 14.0 and 10.0 mm, respectively. The MIC for the plant in different solvents against different bacterial isolates ranged from 100.0 to 3.12 mg mL^-1. Methanol extract of C. polycephala recorded the lowest MIC of 3.12 mg mL^-1 against MRSA and 6.25 mg mL^-1 against S. aureus. Water extract showed less activity against Microsporum gypseum with 7.0 mm zone of inhibition. The plants extracts had good activities against dermatophytes while yeasts isolates recorded poor activity judged by the size of zones of inhibition. The results are as shown in Table 4 and 5, respectively.

Phytochemical screening: The plants were screened using thin layer chromatography visualization reagents. The plants extracts were found to show a positive test for the presence of phenols. Terpenoids and anthraquinones were present on Senna didymobotrya dichloromethane extract and absent in the other plants extracts screened. The results are as shown in Table 6.

Cytotoxicity: Cyathula polycephala methanol extracts was very safe with a CC₅₀ of 100% while water extracts were toxic with CC₅₀ of 23.75 and 31.56%. The two extracts C. polycephala water and S. didymobotrya DCM were very toxic as compared to the positive control Chloroquine with CC₅₀ of 25.28 and 51.94% at concentration 1000 and 100 mg mL^-1, respectively. The results are as shown in Table 7.

Acute toxicity: Senna didymobotrya DCM extract killed 80% of mice at a dose of 5000 mg kg^-1 and 40% at a dose of 1000 mg kg^-1 therefore, it LD₅₀ is between 5000 and 1000 mg kg^-1. This suggests that the Cyathula polycephala extract could be safe as an antimicrobial agent at both doses although this was done on an animal model but not human beings. Other indicators of toxicity were restlessness, general body weakness, loss of appetite and fever. The results are as shown in Table 8.

DISCUSSION

Yields for sequential extractions: Water extraction produced the highest yields of as compared to all the solvents used. Methanol extracts produced higher yields compared to the other organic solvents used. In general, the percentage yields for water extracts were more than the percentage yields for organic solvents. This is due to the fact that water is very polar than organic solvents hence it is able to extract more compounds from a plant material. Kigondu (2007) also found that water extracts produced the highest yields as compared to organic solvents.

Bioactivities of the plants extracts: The results obtained in this study indicate a considerable difference in antimicrobial activity of different extracts from the two plant species. Pseudomonas aeruginosa was the most susceptible bacterium of all the bacterial test strains. Similarly, T. mentagrophyte was found to be the most sensitive fungus of all the tested fungal strains. From all the fungal strains included in the test, C. albicans, C. parapsilosis, C. krusei, C. neoformans were found to be virtually resistant to all plant extracts. Cryptococcus neoformans was found to be the least susceptible fungus probably because of the presence of a mucopolysacharrid capsule. The polysaccharide capsuler material in some of the pathogenic microorganism is responsible for virulence and antimicrobial resistance (Hooper, 2001). Some crude extracts were more active on fungal strains as compared to bacteria while a wide range of the extracts were moderately active against a wide range of bacteria. The least activity or no activity was evidence in several extracts with inhibition zone diameters being below 10 and 6 mm indicating no activity. In terms of inhibition zone diameters most of the extracts displays antibacterial activity amounting to 10.0 mm this can be considered moderately active in consideration that the highest is at 16.0 and the lowest is 7.0 mm. These findings are in line with the investigation of Bii et al. (2010) which indicates that the lowest activity was at 7.0 mm and the highest was at 18 mm in diameters. The solvents used for extraction were used as negative control while fluconazoles and gentamycin were used as positive controls. The standards drugs (gentamycin and fluconazole) inhibition zone diameters was between 9 and 21 mm which compared well with that of the plant extracts. When the antimicrobial activities of these plants extracts were compared to that of the positive controls, only S. didymobotrya dichloromethane extract at a concentration of 100 mg mL^-1 was found to have almost comparable activity to the standard gentamycin against P. aeruginosa. Similarly, S. didymobotrya and C. polycephala at 100 mg mL^-1 showed comparable activity against the fungi T. menthagrophytes and Microsporum gypseum to that of fluconazole.

In general, among the tested microbial strains, bacteria were found to be more sensitive to many of the test agents than fungi. The antibacterial activity was more pronounced on the gram-negative bacteria (K. pneumonia) than the gram-positive bacteria (S. aureus and MRSA). The reasons for variant activities between gram negative and gram positive bacteria could be due to their morphological difference. Gram negative bacteria have an outer phospholipidic membrane carrying the structural lipopolysaccharide components. This makes the cell wall more permeable to lipophilic solutes while porins constitute a selective barrier to the hydrophilic. Gram positive bacterial should be more susceptible since they have only an outer peptidoclycans layer which is not an effective permeability barrier as reported by Nostro et al. (2000). But in this study, we found contradicting results. Some extracts are active on both gram positive and gram negative like Senna didymobotrya DCM which was active against MRSA and P. aeruginosa, hence this can be considered as having broad spectrum of activity and present a potential agent for control of drug resistant strains. It is also good to note that, Senna didmobotrya is widely used among the Kipsigis for a wide range of conditions including; malaria, diarrhoea and skin infections hence, this practice is supported by the demonstrated broad spectrum of activity of its different extracts. Water extracts of Senna didmobotrya had good activity and support the traditional methods of water extraction of most of the medicinal plants.

Extracts from organic solvents had good activities as compared to water extracts, interestingly most medicinal plants are used as water extracts despite the low activities observed in this study (Gachathi, 1989; Kokwaro, 1976). This result agrees with the findings of Kigondu et al. (2009) which indicated that methanol extracts were more active as compared to water extracts. The activity could be explained by the ability of Methanol to extract both polar and non polar compounds hence more active compounds in the extracts.

The preliminary screening assays for antimicrobial activity can largely be considered as qualitative assays and are used for identifying the presence or absence of bioactive constituents in the extracts. However, these methods of assay offer little information on these compounds. The MIC is a quantitative assay and provides more information on the potency of the compounds present in the extracts. Thus, the MIC values of crude extracts of the two medicinal plants were determined so as to demonstrate the potency of the extracts against the selected strains of bacteria and fungi. The MIC of the crude extracts were has high as 100 mg mL^-1 as compared to the standard drugs which is 0.5 mg mL^-1 for both yeast and bacterial and 1.0 mg mL^-1 for dermatophytes at 95% confidence interval (p = 0.05 level of significance). Although this was significantly lower than that of fluconazole (p<0.01) the extracts are promising since they are crude extract compared to pure compound of gentamycin and fluconazole. This is a clear indication that the active ingredient is present in low quantities which necessitate the use of large amounts of crude extracts to gain the desired therapeutic effects. The differences in bioactivities could be attributed to the facts that plants differ phytochemically and the extraction procedures could also affect or alter their composition. Absences of bioactivity does not warrant disapproval of ethnobotanical utilization of the plants, simply because it may suggest that the extracts are acting in an indirect way where active ingredients exist as a precursor requiring activation in vivo. Another reason is that since the tradition herbal practitioners are more often than not using plants in combination to attain therapeutic effects, the extracts could be acting synergistically. Another reason is that, the herbalist is using the extracts in large quantities hence having therapeutic activities since the concentration of the active ingredients increases with the increasing quantity of the crude extracts. From this study, it is clear that activities vary greatly from plant to plant, solvent used for extraction and also on different. Among the Kipsigis community the extracts of Senna didymobotrya is widely used for skin conditions which may be caused by dermatophytes and bacteria. This study therefore provides scientific evidence of its efficacy against skin disease coursing microorganisms.

Thin layer chromatography: Phytochemical screening is important in determining the type of chemical constituents found in a given herbal drug or extract. The plants were screened using thin layer chromatography visualization reagents. According to the results of the phytochemical screening study, phenols were present in the two extracts. The S. didymobotrya methanol extract showed positive test for the presence of alkaloids and flavonoids. Terpenoids and Anthraquinones were present on S. didymobotrya dichloromethane extract and absent in the other plants extracts screened. The presences of flavonoids, phenolics or terpenoids in plants have been reported to be responsible for antimicrobial activity in plants (Nostro et al., 2000). Flavonoids and flavonoid-derived plant natural products have long been known to function as antimicrobial defense compounds (Kazmi et al., 1994). The different in vitro studies have also shown that they are effective antimicrobial substances against a wide spectrum of microorganisms (Schewe and Sies, 2003). For example, the quercetin and naringenin (Meng et al., 2000) and the mixture of catechin compounds (found in oolong and green teas) (Rauha et al., 2000) have been shown to exert antimicrobial activity against a wide range of microorganisms. Recent findings have also showed that ronane diterpenes, hugorosenone [3 β-hydroxyrosa-1, 15-dien-2-one and 18-hydroxy hugorosenone among others] from H. casteneifolia have both larvicidal and antifungal activity (Baraza et al., 2008). The antifungal and antibacterial activity of the plants studied was attributed to the presences of these secondary metabolites. In plants, their role is to protect plants against microorganisms and insects, (Cowen, 2008).

In vitro toxicity (IC₅₀) of the two medicinal plant extracts: Toxicity studies are very important in screening medicinal plants for safety. The extracts that had MIC of 50.0 mg mL^-1 and below were considered potent and their cell toxicity were determined. The CC₅₀ for the plants extracts was tested on 100 and 100 mg mL^-1 concentrations to determine their safety. Cell toxicity of the extracts were categorized into; Toxic CC₅₀ below 50% and safe CC₅₀ above 50% (Zirihi et al., 2005). Except C. polycephala, water extracts of the test plants generally had lower cytotoxicities than both methanol and DCM. The results are in agreement with other work done by Muthaura et al. (2007) on antimalaria properties of Boscia angustifolia. Cepleanu et al. (1994) also reported that the stem bark extracts from this plant had neither cytotoxicity nor brine shrimp lethality. These results seem to validate the traditional practice where herbs are extracted in boiled water (Gessler et al., 1995). Most of the extracts tested had low toxicity (CC₅₀ above 50%) against HELF cells, suggesting that they may be safe as antimicrobials. This result agrees with other studies done by Kigondu et al. (2009) on cell toxicity of medicinal plants.

Acute toxicity of Senna didymobotrya and Cyathula polycephala extracts: Acute toxicity was done on Senna didymobotrya DCM extracts and Cyathula polycephala water extracts since they were found to be toxic to the cells. The extracts were considered safe when the death of mice was less than 50% which was observed at a lower dose of less than 5000 mg kg^-1 (Zirihi et al., 2005). It was found that C. polycephala extract is safe as an antimicrobial agent at both doses. These results are in agreement with the findings of Muthaura et al. (2007) who found that, the water extract was less toxic than the DCM extract. Senna didymobotrya is widely used among the Kipsigis community in treating and managing skin and diarrhoea infections. However, the method of preparation such as burning into ashes and then mixing with margarine for skin conditions and mixing with hot water and milk for stomach problems may be a way of reducing toxicity otherwise the population may be continuously exposed to cytotoxic plant.

CONCLUSION

Senna Didymobotrya was active against both gram positive and gram negative bacterial as well as fungal isolates; hence it has a broad spectrum of activity. Some extracts in this study has high activity against fungal strains as compared to bacteria but generally the bioactivities of all the extracts tested were generally good on bacterial isolates as compared to fungi. In both the cell and acute toxicity, it was observed that extracts at high concentration and at a high dose tend to be toxic as compared to lower concentration and lower doses. In particular, the Senna didymobotrya DCM extracts was toxic with LD₅₀ between 1000 and 5000 mg kg^-1, therefore, for safety measures, herbs should be taken at lower concentration and lower doses. Senna didymobotrya is generally applied topically and mixed with milk or margarine for oral use; this is basically to reduce toxicity.

RECOMMENDATIONS

It is worth recommending that since some plants are potent both antibacterial and antifungal, further work should be done especially analysis and isolation of the compounds present as well as determination of their bioactivity. Bioassay of combinations of plant extracts that exhibited moderate and low activity should be carried out to establish any synergism between them. Since, this study was based on the plants part used by herbalist, it is worth recommending that other parts be studied. The plants that were found to be toxic to the mice should not be used as antimicrobial agents unless the herbal practitioners have other means of reducing their toxicity, alternatively they can be used at very low concentration and doses. Since some plants have good activities they should be formulated into different consumable forms as well as being incorporated into food as supplements. The plants should be conserved both in situ and ex-situ.

ACKNOWLEDGMENT

The authors acknowledge the research grants from International Foundation of Science which enabled this work to be done. The Director Kenya Medical Research Institute is also acknowledged.

REFERENCES