Antituberculous, antimicrobial, cytotoxicity and phytochemical activity study of Piliostigma thonningii extract fractions

Department of Medicine, Kisii University, PO Box, 408-40200, Kisii-Kenya. Fritz Lipmann Institute (FLI) – Leibniz Institute for Age Research, D-07745, Jena-Germany. Department of Biochemistry & Molecular Biology, Egerton University, P.O Box. 536-20155, Egerton, Kenya. Department of Chemistry, Egerton University, P.O Box. 536-20155, Egerton, Kenya. Centre of Microbiology Research, KEMRI, PO Box 19464-00202, Nairobi, Kenya. Friendrich Loeffler Institute (FLI), Institute for Bacterial Infections and Zoonosis, D-07743, Jena-Germany. Department of Physical Sciences, Chuka University, PO Box 109-60400, Chuka, Kenya.


INTRODUCTION
There is an increased and concerted efforts to the adoption and integration of traditional medicine and medical practices referred to as complementary or/and alternative medicine in both developing and developed countries in their health system (World Health Organization [WHO], 2005). This is after studies by WHO have demonstrated that 80% of the world's population depends on medicinal plants for their primary health care (Mothana et al., 2008;Gupta et al., 2010).
Piliostigma thonningii (Schum.) Milne-Redh locally known as mukuura in Mbeere, Embu County in Kenya has been used and reported for different and varied medicinal purposes. For example, it has been reported that in many African countries, various parts of the plant (root, bark, seed, and fruit) are used to treat wounds, ulcers, gastric and heart pain, gingivitis, and as an antipyretic (Silva et al., 1997;Maroyi, 2011Maroyi, , 2013. In addition, a cough remedy is prepared from the root bark (Akinpelu and Obuotor, 2000). Its roots mixed with roots of Elephantorrhiza goetzei are used as bilharzia (schistosomiasis) medicine (Maroyi, 2011). Certain compounds (alkaloids, flavonoids, saponins and tannins) isolated from some of its parts have been reported to bear anti-inflammatory, analgesic and antibacterial activities (Paiva et al., 2010;Akinpelu and Obuotor, 2000).
However, reported scientific work is rather incomprehensive, in relating the phytochemicals present in the antibacterial activities and cytotoxicity of the same extracts from any part of P. thonningii. For example, the reported studies deal distinctly in quantification of active substances (Babajide et al., 2010;Deshi et al., 2014) or antibacterial and anti-inflammatory activity (Etsuyankpa et al., 2013;Aderogba et al., 2006;Ibewuike et al., 1997) or effects of ethanolic extracts on kidney function (Dasofunjo et al., 2012) and lipid profile (Ighodaro and Omole, 2012) in Wistar rats. Other studies independently show phytochemical evaluation, antibacterial activity (Tor-Anyiin and Anyam, 2013) and anti-malaria activity (Kwaji et al., 2010).
No literature was found on antitubercular activity of the plants extracts and therefore this study was designed to investigate anti-TB activity of the root extract. In addition, this study sought to investigate its antibacterial and antifungal activity, determine the phytochemicals present and relate these findings with the observed activities and cytotoxicity to Vero cells.

Plant
The plant used in this study was identified through ethnobotanical approach. The information of its use and preparation in Mbeere community, Kenya was gleaned from local herbalist and confirmed from documentation by Riley and Brokensha (1988) in The Mbeere in Kenya (ii), Botanical identity and use.. This plant is not an endangered species and it was collected in open community field and therefore no prior permission was required. The location for collection was around 0°46'27.0"S 37°40'54.9"E; -0.774156, 37.681908 of GPS co-ordinates. The identity was also confirmed by a Botanist at Egerton University where voucher specimen number NSN4 was deposited.

Plant extract preparation
Root samples were chopped into small pieces of 2 to 3 cm and airdried in dark at room temperature (23±2°C) to constant weight. Using a mechanical grinder, the dried root specimens were ground to powder. The powder (50 g) was cold extracted in water with intermittent shaking to mimic the traditional local method of extraction and later lyophilized to obtain a dry powder. Methanol extraction was done by macerating 50 g in 200 ml of methanol for 48 h and then filtered using a filter paper (Whatmann 1) and the residue obtained was further re-extracted using similar amount of methanol. The two volumes of filtrate were pooled together and thereafter concentrated in vacuo using a rotary evaporator. Afterwards, the product was allowed to air dry and the yields recorded.
Fractionation of powdered root part of P. thonningii was done using different solvents of increasing polarity. The root powder (50 g) was macerated in 200 ml of Petro ether with intermittent shaking for 48 h (two days) after which they were filtered using Whatman No. 1 filter paper. The residue was further re-extracted using the same fresh solvent for 24 h and thereafter the filtrates pooled together. The resulting residue was air dried and further extracted with dichloromethane followed by ethyl acetate and lastly methanol using the same procedure carried out for petrol ether. Using a rotary evaporator, the solvent was removed from each filtrate under conditions of reduced temperature and pressure. The resulting dry extract was weighed and stored in air tight sample bottles at -20°C until the next use (Njeru et al., 2015).

Disc diffusion test
The antibacterial activity was assayed by disc diffusion method according to Clinical and Laboratory Standards Institute (CLSI, 2007) and Mbaveng et al. (2008) with slight modifications. Fresh inoculum was prepared by suspending activated colonies in physiological saline water (0.85% NaCl). Using 0.5 McFarland turbidity standard, the bacteria and fungi suspensions were adjusted to 1.5 × 10 6 CFU/ml after which they were inoculated aseptically by swapping the surfaces of the Muller-Hinton agar (MHA plates and Sabouraud dextrose agar (SDA) plates. Whatmann filter paper (No. 1) discs of 6 mm diameter were made by punching the paper, and the blank discs sterilized in the hot air oven at 160°C for 1 h. They were then saturated with 10 μl of various stock extract solution. The methanolic and water crude extracts stock solution was at 1.0 g/ml. For fractions, petro ether, dichloromethane, and methanol fractions stock solutions were made at 500 µg/ml, while ethyl acetate at 250 µg /ml. This afforded disc extract concentration of 1.0 × 10 4 µg/disc for water and methanol crude extracts, 5 µg/disc for petro ether, dichloromethane, and methanol fractions and 2.5 µg/disc for ethyl acetate. Three standard drugs were used as positive controls: Oxacillin 10 µg/disc (Oxoid Ltd, Tokyo-Japan) and Gentamycin 10 µg/disc (Oxoid Ltd, Tokyo-Japan) for Gram positive and Gram negative bacteria, respectively. Nystatin 100 µg/disc (Oxoid Ltd, Tokyo-Japan) was used as the standard drug for all fungi while discs loaded with 10 µl of DMSO was used as negative controls. Using a sterile forceps, the impregnated dry discs were carefully placed on the agar plates at equidistance points. A positive control as well as a negative control was incorporated in each plate and the plates incubated at 4°C for 2 h so as to allow the extract to diffuse into the media after which they were incubated at 37°C for 18 h. Antimicrobial activity was determined by measuring the size of the inhibition zone to the nearest mm and the results recorded. Extracts fractions that gave an inhibition zone of more than 10 mm were considered to be active (CLSI, 2007) and therefore their Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) were determined (Mothana et al., 2008).

Determination of MIC and MBC
The MIC and MBC of the plant P. thonningii extracts was determined for all the organisms in triplicates using broth microdilution assay. The petro ether, dichloromethane, and methanol fractions stock solutions were made at 500 µg/ml while ethyl acetate at 250 µg /ml with dimethyl sulfoxide (DMSO). To 100 µl of nutrient broth in a sterile 96 well plate, 50 µl of varying plant concentration (petro ether, dichloromethane, and methanol fractions at 500 to 3.91 µg/ml while ethyl acetate at 250 to 1.95 µg/m) was added followed by 50 µl of test organisms previously diluted to the equivalent of 0.5 McFarland standard. Addition of the test organisms was done in all the wells except for wells of column 11 which contained neat DMSO and broth, this served as control to check for purity. The adequacy of the media to support the growth of the test organism was evaluated by putting the broth and the test organism in wells of column 12. The plates were then covered with a sterile "cling-on" sealer and incubated for 24 h at 37°C. Bacterial growth was evaluated by addition of 40 µl of 0.2 mg/ml piodonitroterazolium chloride (INT, Sigma) to each well and incubated for 30 min. Growth of bacteria was detected by formation of a pink-red coloration while inhibition of growth was signaled by persistence of a clear coloration. The lowest concentration that exhibited color change was considered as the MIC. MBC was determined by streaking a loopful of broth from wells that exhibited no color change onto sterile nutrient agar and Sabouraud dextrose agar for bacteria and fungi, respectively and thereafter incubated at 37°C for 24 h. The lowest concentration that exhibited no growth was considered as the MBC (Lai et al., 2010).

Antitubercular activity
Prior to its use, the M. tuberculosis was revived on Lowenstein Jensen (LJ) slants for 14 days at 37°C following standard procedures (Gupta et al., 2010;Mariita et al., 2010). The efficacy of the plant extracts against M. tuberculosis was carried out using the BACTEC MGIT 960 system (BD, New York-U.S.A). This is a fully automated, high volume, non-radiometric instrument that offers continuous monitoring of culture growth. The dry crude extract (water and methanolic) was first dissolved in DMSO to a final concentration of 1 g/ml for preliminary screening. Growth supplement (0.8 ml) containing a mixture of oleic acid, bovine albumen, dextrose and catalase (OADC) was added to five 7 ml BBL ™ MGIT ™ tube labeled growth control (GC), streptomycin (STR), isonaizid (INH), rifampicin (RIF), and ethambutol (EMB) to provide essential substrates for rapid growth of Mycobacteria. 100 μl of BBL ™ MGIT ™ SIRE (streptomycin, isonaizid, rifampicin, ethambutol) prepared aseptically according to the manufacturers' instruction was added to corresponding labeled BBL ™ MGIT ™ tube followed by addition of 0.5 ml of 1% Mycobacterium suspension. Mycobacterium suspension was prepared by pipetting 0.1 ml Middlebrook 7H9 Broth containing Mycobacterium adjusted to 0.5 McFarland standard into 10 ml sterile saline aseptically. The BACTEC MGIT ™ 960 system (BD, New York-U.S.A) was then loaded following the manufacturer's instructions and incubated at 37°C. Streptomycin at 1.0 µg/ml, isoniazid at 0.1 µg/ml, rifampicin at 1.0 µg/ml and ethambutol at 5.0 µg/ml served as the positive control whereas DMSO was used as a negative control. The procedure was repeated using plant water and methanolic crude extracts at 1.0 g/ml in place of SIRE. The process was also repeated with petro ether, dichloromethane, ethyl acetate and methanol solvent fractions. The fractions were tested at concentrations ranging from 50 to 6.25 µg/ml (petro ether, dichloromethane and methanol) or 25 to 3.125 µg/ml (ethyl acetate) to determine the MIC.

Cytotoxicity screening
MTT assay was used to determine the toxicity of the extracts obtained from the plant. This is a colorimetric assay hinged on the ability of mitochondrial enzyme (succinate dehydrogenase) to reduce yellow water soluble MTT to an insoluble colored substance, (formazan) that is spectrophotometrically measurable (Njeru et al., 2015). The level of formazan is directly proportional to the measure of cell viability because only metabolically active cells can reduce MTT. The test cell line used was Vero cells from African green Monkey Kidney cells (Cercopithecus aethiops epithelial cell line; ATCC CCL-81). The test cells were grown in growth media comprising of 100 ml DMEM, 10 ml fetal bovine serum (FBS), 1 ml penstrep, 1 ml amphotericin B, 1 ml L-glutamine. The test cells were incubated at 37°C in 5% CO 2 until they attained confluency after which they were passaged by adding 2 ml of 0.25% trypsin and further incubated at room temperature until the cell were detached. Growth media (6 ml) was introduced to the test cells to inactivate trypsin off action, while the cell crumps formed were broken gently by sucking and releasing the cell suspension using a pipette. 2 ml of the harvested cells were then transferred into a 50 ml vial and topped up to 50 ml mark using growth media. A cell suspension of 100 μl (1 × 10 5 cell/ml) was seeded into two rows of wells in a 96-well micro-titer plate for one sample. The test cells were then incubated in 100 μl growth media at 37°C and 5% CO 2 for 48 h to form a confluent monolayer. The growth medium was then aspirated off and replaced with 100 μl of maintenance medium comprising of 100 ml DMEM, 2 ml fetal bovine serum (FBS), 1 ml penstrep, 1 ml amphotericin B, 1 ml L-glutamate and 0.1 ml gentamycin. Afterwards, cells were exposed to decreasing concentrations of respective plant extracts (from 250 to 0.24 μg/ml for petro ether and dichloromethane fractions and 500 to 0.49 μg/ml for ethyl acetate and methanolic fraction) and incubated at 37°C for 48 h. This was followed by a further incubation period of 4 h in 10 µl of 5 mg/ml MTT solution after aspirating off the plant extracts. This was followed by addition of 100 µl acidified isopropanol (0.04 N HCl in isopropanol). The well plate was gently shaken for 5 min to dissolve the formazan and then optical density measured using ELISA Scanning Multiwell Spectrophotometer (Multiskan Ex labssystems) at 562 nm using 690 nm as reference wave length. The last column containing medium without plant extracts was included to act as negative control. Cell viability (%) was calculated at each concentration as follows using the formula of Ngeny et al. (2013).

Phytochemical tests
Phytochemical tests were done to determine the class of compounds present in the active fractions that could be responsible for activity and/or cytotoxicity. They were identified by characteristic colour changes based on standard procedures according to Houghton and Raman (1998), Edeoga et al. (2005), Ngoci et al. (2011) and Somboro et al. (2011). The results were reported as either (+) for presence, and/or (-) for absence.

Alkaloids
Six to eight drops of Dragendorf reagent was mixed with 2 ml of the extract. Formation of brownish-red precipitate indicated the presence of alkaloids. The Dragendorf reagent was prepared by mixing two reagents: reagent 1 and reagent 2 in equal parts. Reagent 1 was made by dissolving 8.5 g of Bismuth subnitrate in a solution of 10 ml acetic acid and 40 ml of distilled water while as Reagent 2 was prepared by dissolving 8 g of potassium iodide in 20 ml of water (Harborne, 1973;Somboro et al., 2011).

Phenols
Phenols were detected using ferric ferichloride which was prepared by dissolving 0.1 g of ferric ferichloride in 10 ml of water. Equal volumes (2 ml) of both ferric ferichoride and the plant extract were mixed. Formation of a violet-blue color or greenish color was evidence that phenols is present (Harborne, 1973;Somboro et al., 2011).

Terpenoids
One gram of Vanillin was mixed with 100 ml of concentrated sulphuric acid after which 2 ml of the resultant solution was mixed with 2 ml of the plant extract. Formation of a blue-green ring or pinkpurple coloration signified the presences of terpenoids (Harborne, 1973;Somboro et al., 2011). Anthraquinones 0.5 ml of the plant extract was mixed with 0. 5 ml of 10% methanolic potassium hydroxide. Red coloration indicated the presences of anthroquinones. 10% methanolic potassium hydroxide was prepared by dissolving 0.5 g of potassium hydroxide pellets in 50 ml of methanol (Harborne, 1973;Somboro et al., 2011).

Flavonoids
Five milliliters of dilute aqueous ammonia solution was added to a portion of the aqueous filtrate of the plant extract, followed by concentrated sulphuric acid. A positive test result was confirmed by the formation of a yellow coloration that disappeared instantly (Edeoga et al., 2005;Ngoci et al., 2011).

Statistical analysis
Ms Excel 2010 data sheets and Graphpad Prism version 6 were used to analyze the data. The data on cytotoxicity was expressed as a percentage of the untreated controls. CC 50 values, which is the concentration that kills 50% of the Vero cells, was determined by regression analysis. A particular fraction's extract was considered cytotoxic if it had CC 50 of less than 90 μg/ml (Irungu et al., 2007). Furthermore, unpaired student's t-test was used to test for statistical significance in the differences between the treatments and the control in this study. A p value of less than 0.05 was considered to indicate statistical significance. Values were expressed as mean ± standard error of mean (S.E.M).

Antibacterial, antifungal and antitubercular activity of water and methanolic crude extract
This study first sought to identify whether the plant bark extract had any general antibacterial activity by mimicking the indigenous extraction methods. Traditionally, the plant bark extract was prepared in water or in combination with alcoholic beverages. It was found out that both water and methanolic crude extract had general antimicrobial activity. The diameter of inhibition zones was in all cases higher than 10 mm (Table 1) and therefore this formed the basis for further fractionation.
The test was however performed at higher concentration and we needed to bring this down as well as determine MIC and MBC. Unexpectedly, both crude extract had no antituberculous activity, contrary to long held believe by members of Mbeere community that this plant can be used to treat tuberculosis. Both extract gave 400 GU which is the same as the one given by the negative control (Table 2). Therefore, the crude extract had a remarkable general antimicrobial activity but no antitubercular activity.

Antibacterial and antifungal activity of fractions
Further analysis on fractions demonstrated very weak activity on petro ether, dichloromethane and ethyl acetate fractions with all giving diameter of zones of inhibition of   WT: Water crude extract at 1 g/ml; MOH: methanol crude extract at 1 g/ml; SIRE: positive control of streptomycin at 1.0 µg/ml, isonaizid at 0.5 µg/ml, rifampicin at 1.0 µg/ml and ethambutol at 5.0 µg/ml; GC: growth control; NC: negative control of media treated with DMSO; R: resistant; S: sensitive.  Indicates there is no significant difference between the test fraction and standard control (P>0.05).

Diameter of zone of inhibition in mm
less than 10 mm. Therefore, these 3 fractions were not considered for MIC and MBC determination. Interestingly, methanolic extract had broad spectrum activity inhibiting the growth of Gram positive bacteria (the highest was 20.3 mm against S. aureus), Gram negative bacteria (the highest was 14.0 mm against E. coli) and fungi (the highest was 13 mm against C. albicans). Activity of methanolic fraction against E. coli (14.0) was remarkable since there was no significant statistical difference in activity with that of positive control (p>0.05) ( Table 3). The MIC of methanolic fraction against 4 bacteria was less than 200 µg/ml. S. aureus and S. sonnei had MIC of 31.25 µg/ml; Methicillin resistant S. aureus had MIC of 62.5 µg/ml and E. coli had MIC of 125 µg/ml. However, all four had MBC >500 µg/ml and hence bacteriostatic in action (Table 4).

Antituberculous, cytotoxicity activity and phytochemical profile of the fractions
When screening with crude water and methanolic extract revealed no antituberculous activity, we sought to investigate whether fractionation could improve on activity in some of the fractions. This is based on the knowledge that solvents of different polarities will retain certain compounds of equivalent polarity. It was also hypothesized that different bioactive compound combinations could bear antagonistic effects on each other once lumped together in crude extract, but their separation could render certain fractions active against M. tuberculosis (MTB). Interestingly, methanolic extract had MIC of 12.5 µg/ml and hence significant activity while petro ether fraction had MIC of 50 µg/ml, dichloromethane and ethyl acetate had MIC of >50 and 25 µg/ml (Table 5). All the fractions had CC 50 > 90 µg/ml  and hence were all within the acceptable toxicity limit ( Table 6). The varied activity in the extract could be attributed to various phytochemicals that tested positive. These include alkaloids, flavonoids, phenols, terpenoids and anthraquinones (Table 7).

DISCUSSION
Both crude extract and methanolic fraction extract had broad spectrum activity. They inhibited both Gram positive and Gram negative bacteria and fungi giving remarkable zones of inhibition. The activity against Gram positive was the highest. For example, methanolic fraction gave inhibition zone diameter of 20.3 mm and MIC of 31.25 µg/ml. This could be attributed to the cytoplasmic membrane of Gram-positive species which are simple lipid bilayer and hence not a barrier for most amphipathic compounds and can be readily traversed by antibacterials. In contrast, Gram-negative bacteria have evolved a sophisticated permeability barrier with an additional outer membrane comprising a highly hydrophilic lipopolysaccharide layer and this restricts penetration of hydrophobic and amphipathic compounds, which encompasses many drug compounds (Kim and Frederick, 2006). Difference in activity was also evidence among tested strains in both crude and fraction extracts. This could be due to genetic differences between different strains and this provides proof for the necessity of antibiogram prior to prescription as a precautionary measure in mitigating drug resistance development (Yimta et al., 2014). Even though crude extract had no antituberculous activity, fractionation gave us two fraction (petro ether and methanolic fraction) which remarkably inhibited growth of MTB with MIC of 50 and 12.5 µg/ml, respectively. Moreover, the general antibacterial and antifungal activity was higher with methanolic fraction with four bacteria having MIC of less than 200 µg/ml; the set threshold for herbal extract activity (Kuete, 2010). This may suggest that fractionation enhanced antitubercular activity, implying either that there is higher concentration of active principle better at certain level of polarity or possibility of antagonism of various active compounds when lumped together as in crude extracts thus diminishing the activity. Equally important to note is that both crude and fraction extracts had CC 50 >90 µg/ml and therefore were within the acceptable toxicity limit. The bioactivity in P.thonningii bark sample can to some extent be attributed to the tested phytochemicals, especially terpenoids which previous studies have associated with antibacterial, antifungal and antitubercular activities. The mechanism of action of terpenoids is not entirely understood but is speculated to involve membrane disruption by the lipophilic compounds (Kuete, 2010).

Conclusion
A major output of the current study is the identification of the methanolic fractions which yielded the best antituberculous activity (MIC of 12.5 µg/ml) as well as the highest antibacterial activity (with zones of inhibition of 20.3 mm and MIC of 31.25 µg/ml (S. aureus), 18.3 mm and MIC of 62.5 µg/ml (MRSA), 14 mm and MIC of 125 µg/ml (E. coli) and 13.3 mm and MIC of 31.25 (S. sonnei) which was within the acceptable toxicity limit (CC 50 >500 µg/ml). For us the fraction provides the best candidates for further investigation due to its selective activity. Of particular relevance is its high activity against MRSA, S. aureus, E. coli, S. sonnei, C. albicans and MTB which are currently posing great public health challenge due to drug resistance development and as major sources of community and hospital based infections. To the best of our knowledge, this is the first report exploring the antituberculous activity of P. thonningii and thence a major output in search of new safe drug leads to mitigate the global tuberculosis threat.