Antimicrobial, anti-inflammatory activity and cytotoxicity of Funtumia africana leaf extracts, fractions and the isolated methyl ursolate

South African Journal of Botany 108 (2017) 126–131

Contents lists available at ScienceDirect

South African Journal of Botany journal homepage: www.elsevier.com/locate/sajb

Antimicrobial, anti-inflammatory activity and cytotoxicity of Funtumia africana leaf extracts, fractions and the isolated methyl ursolate T.E. Ramadwa a, E.E. Elgorashi b, L.J. McGaw a, A.S. Ahmed a, J.N. Eloff a,⁎ a b

Phytomedicine Programme, Department of Paraclinical Sciences, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort 0110, South Africa ARC-Onderstepoort Veterinary Institute, Private Bag X05, Onderstepoort 0110, South Africa

a r t i c l e

i n f o

Article history: Received 1 June 2016 Received in revised form 18 August 2016 Accepted 13 October 2016 Available online xxxx Edited by J Van Staden Keywords: Minimum inhibitory concentration Bioautography Cytotoxicity Traditional knowledge Fungal phytopathogens

a b s t r a c t Funtumia africana is used to treat and manage diverse ailments including fever, inflammation, malaria, cancer and urinary incontinence in South Africa. In this study, the antibacterial, antifungal, anti-inflammatory activities and cytotoxicity of the crude extracts, fractions and an isolated compound were determined. Serial microplate dilution and bioautography methods were used to determine the antimicrobial activities. The bacteria tested were ATCC reference strains of Escherichia coli, Enterococcus faecalis, Pseudomonas aeruginosa and Staphylococcus aureus. The fungal test organisms used were three clinical isolates (Aspergillus fumigatus, Cryptococcus neoformans and Candida albicans), C. albicans ATCC 10231 and three phytopathogenic fungi (Fusarium oxysporum, Penicillium janthinellum and Rhizoctonia solani). The anti-inflammatory activity was determined using cyclooxygenase (COX) enzymes and the MTT assay was used to determine cellular toxicity against Vero and human liver (C3A) cells. The crude extract had MICs as low as 80 μg/ml against both bacteria and fungi. The chloroform fraction had the lowest MIC of 20 μg/ml against P. aeruginosa. The hexane and chloroform fractions had MIC of 40 μg/ml against C. albicans ATCC 10231. The crude extract, hexane and chloroform fractions had moderate activity against both COX-1 and COX-2. The chloroform fraction was more active than the crude extract (59.7%) with an inhibition of 68.2% against COX-1. Bioassay-guided fractionation using column chromatography led to the isolation of methyl ursolate (MU) with an MIC of 62.5 μg/ml against F. oxysporum. It was relatively toxic against Vero cells with an IC50 of 10.4 μg/ml. The antimicrobial and anti-inflammatory activities of the crude extract provide some support for the traditional use of the plant. © 2016 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction As part of a search for plant based antimicrobial agents, a random screening of antimicrobial activities of acetone leaf extracts of more than 500 tree species was undertaken in the Phytomedicine Programme against six bacterial and two fungal pathogens (Pauw and Eloff, 2014). Nine tree species with promising activities were selected from this screening process for further investigation. F. africana had good antimicrobial activity against nosocomial bacterial and fungal pathogens with a minimal inhibitory concentration (MIC) of b100 μg/ml and contained a number of antimicrobial compounds on bioautograms (Ramadwa, 2010). Funtumia africana (Benth.) Stapf belongs to the family Apocynaceae which generally contains alkaloids (Gurib-Fakim, 2006). The genus Funtumia consists of F. africana and F. elastica. The leaves of both species are very similar, glabrous, leathery, elongated, elliptic more or less acuminate, cuneate at the base with short stalks (Keay et al., 1964)

⁎ Corresponding author. E-mail address: [email protected] (J.N. Eloff).

http://dx.doi.org/10.1016/j.sajb.2016.10.019 0254-6299/© 2016 SAAB. Published by Elsevier B.V. All rights reserved.

the flowers and fruits of F. africana are longer than those of F. elastica (Burkill, 1960). Although F. africana is native and widely distributed in the east, central and west Africa, the species is also found in some southern Africa countries like Mozambique (Wagner et al., 1987; Beentje, 1994; Orwa et al., 2009). In West Africa, F. africana is used for the treatment of fever, inflammation, malaria, cancer, amoebic dysentery, urinary incontinence and burns (Adjanohoun and Aké, 1979; Adjanohoun et al., 1986; Wagner et al., 1987; Odugbemi et al., 2007; Ashidi et al., 2010). Many steroidal alkaloids from the genus Funtumia have been reported (Janot et al., 1963; Truong-Ho et al., 1963; Mukam et al., 1973; Zirihi et al., 2005). Previous phytochemical investigation of the stem bark of Funtumia africana led to the isolation of steroidal alkaloids of the conanine group, named 12α-hydroxy norcona-N (18).1.4trienin-3-one, 11α, 12α-dihydroxy norcona-N (18).1.4-trienin-3-one and 11α-hydroxy norcona-N(18).1.4-trienin-3-one (Wagner et al., 1987). The aim of this study was to determine the antibacterial, antifungal, anti-inflammatory activities and the cytotoxicity of the crude extract, fractions and bioactive compounds of F. africana to evaluate its potential use as an antimicrobial agent.

T.E. Ramadwa et al. / South African Journal of Botany 108 (2017) 126–131

2. Materials and methods 2.1. Plant collection and extraction The leaves of F. africana were collected from the Lowveld National Botanical Gardens, Nelspruit, Mpumalanga, in November. The leaves were collected in open weave orange bags, dried at room temperature in the shade and powdered using a Macsalab mill (Model 200 Lab). The powdered materials were then stored in closed honey jars at room temperature in the dark until needed. Voucher specimens (PRU 114782) was prepared and kept at the H.G.W.J. Schweickerdt Herbarium of the University of Pretoria. Acetone was used as extractant because it is the best for investigating antimicrobial activity of plant leaf extracts (Eloff, 1998b). The powdered plant leaf material (400 g) of F. africana were extracted with 4 L of acetone and shaken vigorously for eight hours on a Labotec shaking machine. The supernatant was filtered through Whatman No. 1 filter paper using a Buchner funnel and evaporated under vacuum using a Büchi rotavaporator R-114 (Labotec). The concentrated extract was poured into a pre-weighed beaker. The same procedure was repeated twice on the marc (remainder of plant material). The extract was then left to dry under a stream of cold air. The quantity extracted was 31.34 g. Solvent–solvent fractionation was used to fractionate the acetone extract based on polarity of the compounds (Suffness and Douros, 1979). The acetone extract was reconstituted in 300 ml of chloroform and mixed with an equal volume of distilled water in a separatory funnel to give a chloroform fraction. The water fraction was then mixed with an equal volume of n-butanol to yield the water and butanol fractions. The chloroform fraction was dried in a vacuum rotary evaporator and extracted with an equal volume of hexane and 10% water in methanol mixture, which yielded the hexane fraction. The 10% water:methanol fraction was then further diluted to 30% water in methanol and mixed with chloroform to yield 30% water in methanol fraction and chloroform fraction. A total of five fractions were collected, namely water, butanol, 30% water in methanol, chloroform and hexane. 2.2. Antimicrobial activity 2.2.1. Bacterial and fungal species Four most important nosocomial pathogenic bacterial species were selected for the study, namely the Gramme-positive Staphylococcus aureus (ATCC 29213) and Enterococcus faecalis (ATCC 29212), and the Gramme-negative Pseudomonas aeruginosa (ATCC 27853) and Escherichia coli (ATCC 25922) (Sacho and Schoub, 1993). The selection of the specific bacterial strains was based on the recommendation of the National Committee for Clinical Laboratory Standards (NCCLS, 1990), now known as the Clinical Laboratory Standards Institute (CLSI). All the cultures were maintained on Mueller-Hilton (MH) agar (Fluka, Switzerland) at 4 °C. The cells were inoculated and incubated at 37 °C in MH broth (Fluka, Switzerland) for 12 h prior to determining the activities. The densities of bacterial cultures for use in the screening procedures were as follows; S. aureus (2.6 × 1012 cfu/ml), E. faecalis (1.5 × 1010 cfu/ml), P. aeruginosa, (5.2 × 10 cfu/ml9), E. coli (3.0 × 1011 cfu/ml) (Shai et al., 2008). The fungal pathogens that were used included a Candida albicans reference strain (ATCC 10231) and three clinical isolates from the National Health Laboratory Service (NHLS), Department of Microbiology, Pretoria. Candida albicans, isolated from a Gouldian finch, Cryptococcus neoformans, isolated from a cheetah, and A. fumigatus, isolated from a chicken which suffered from systematic mycosis, were obtained from the Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria. The plant pathogenic fungi, namely Fusarium oxysporum, Penicillium janthinellum and Rhizoctonia solani, which are among the most important fungi of economic significance to plants, were obtained from the Department of Microbiology and

127

Plant Pathology at the University of Pretoria. All the fungal organisms were maintained on Sabouraud (SD) agar (Oxoid Basingstoke, UK) at 4 °C until use. For the antifungal assay, the fungal species were subcultured in SD broth at 35 °C before screening (Mahlo et al., 2010; Suleiman et al., 2010). 2.2.2. Bioautographic antibacterial method Qualitative analyses of the number of antibacterial compounds were determined by the bioautography method (Begue and Kline, 1972). Exactly 100 μg of the crude extract was loaded in a line of 1 cm wide on the thin layer chromatography (TLC) plates and developed in the benzene:ethanol:ammonium hydroxide (90:10:1) (BEA), non-polar/ basic eluent system (Kotze and Eloff, 2002). The TLC plates were dried under a stream of air to evaporate the solvents. Overnight bacterial cultures which were grown in MHin an incubator at 37 °C were centrifuged at 3000 ×g for 10 min. The pellets were resuspended in 10 ml of fresh MH broth. The developed plates were sprayed with the fresh bacterial culture of S. aureus until completely moist using a spraying gun. The moist plates were incubated at 37 °C in a humidified atmosphere for about 18 h. The plates were sprayed with a 2 mg/ml aqueous solution of p-iodonitrotetrazolium violet (INT) (Sigma) and incubated for a further 2–6 h. Bacterial growth led to the emergence of a purple– red colour resulting from the reduction of INT into the corresponding formazan salt. Clear zones indicated the inhibition of the bacteria by the compound present at that Rf value on the chromatogram. For fungi a variation was used by collecting conidia and growing them overnight before spraying on a chromatogram (Masoko and Eloff, 2005). 2.2.3. Minimal inhibitory concentration (MIC) A serial microdilution assay (Eloff, 1998a) with slight modification (Masoko et al., 2005) was used to determine the minimum inhibitory concentration value of the crude extracts, fractions and isolated compound using INT reduction as an indicator. This was determined against all the plant and animal fungi chosen for the study. The samples were tested in triplicate in each assay, and the assays were repeated in their entirety to confirm results. Aliquots of the crude acetone extract and fractions were dissolved in acetone to final concentrations of 10 mg/ml. The isolated compound was dissolved in acetone to a final concentration of 1 mg/ml. Exactly 100 μl of the extracts, fractions and compound were serially diluted with 50% water in 96-well microtitre plates and 100 ml of microbial culture was added to each well. Amphotericin B and gentamicin were used as the positive controls, while serially diluted acetone was used as the negative control. As an indicator of growth, 40 μl of 0.2 mg/ml INT dissolved in hot water was added to the microplate wells (Eloff, 1998a). The covered microplates were incubated and examined after 24 and 48 h at 35 °C at 100% relative humidity after being sealed in a plastic bag to minimize fungal contamination in the laboratory. The MIC was recorded as the lowest concentration of the extract that inhibited antifungal growth. The colourless tetrazolium salt acts as an electron acceptor and is reduced to a red-coloured formazan product by biologically active organisms (Eloff, 1998a). Where fungal growth is inhibited, the solution in the well shows a marked reduction in intensity of colour after incubation with INT or remains clear. 2.3. Anti-inflammatory activity of the crude extract of F. africana and fractions Anti-inflammatory activity was determined using both the COX-1 and COX-2 assays. The COX-1 bioassay was performed as described by White and Glassman (1974) with slight modifications (Jager et al., 1996). All the fractions and the crude extract were reconstituted in ethanol. Cyclooxygenase enzyme (3 units protein prepared from ram seminal vesicles [Sigma Aldrich] was mixed with co-factor solution (0.3 mg/ml adrenalin and 0.3 mg/ml reduced glutathione in 0.1 M Tris

128

T.E. Ramadwa et al. / South African Journal of Botany 108 (2017) 126–131

buffer, pH 8.2) and incubated on ice for 5 min. Sixty microlitres of the enzyme and co-factor solution was added to 20 μl of sample (2.5 μl of ethanolic extract/fraction and 17.5 μl water) and incubated at room temperature for 5 min. Twenty microlitres of [14C] arachidonic acid was added to the assay and incubated in a water bath for 10 min at 37 °C. The reaction was terminated by adding 10 μl of 2 N HCl. After incubation, 4 μl of a 0.2 mg/ml carrier solution of unlabelled prostaglandins was added. The 14C-labelled prostaglandins synthesized during the assay were separated from the unmetabolised [14C] arachidonic acid by column chromatography. Silica gel in eluent 1 (hexane: 1.4-dioxan: acetic acid glacial 350:150:1) was packed in Pasteur pipettes to a height of 3 cm. One ml of eluent 1 was added to each of the assay mixtures and the mixture was applied to separate columns. The arachidonic acid was eluted with 4 ml of eluent 1 and discarded. The prostaglandins were subsequently eluted with 3 ml of eluent 2 (ethylacetate:methanol 85:15) into scintillation vials. After mixing with scintillation liquid, the radioactivity was counted using a scintillation counter (Wallac 1409). The same protocol was followed for COX-2 except that in the preparation of co-factor solution, 0.6 mg adrenaline was used. All the five fractions and crude extract were tested at a concentration of 250 μg/ml with duplicate determinations for each extract per assay. In each test assay, four controls were run. Two were background in which the enzyme was inactivated with hydrochloric acid before the addition of [14C] arachidonic acid, and two were solvent blanks (2.5 μl ethanol and 17.5 μl water). Positive control measurements were carried out by determining the IC50 of indomethacin. The percentage inhibition of the fractions and crude extract was calculated by comparing the amount of radioactivity present in the sample to that in the solvent blank. The percentage inhibition (% I) of prostaglandin synthesis was calculated as follows: % ¼ ½1 ‐ dpm sample ‐ dpm of background divided by

was elucidated using Nuclear Magnetic Resonance (NMR)-vnmrs 600, BIO-Chemtek for both 1H and 13C, using dimethyl sulfoxide (DMSO) as the solvent and Mass Spectrometry. Comparison with published data was used to conclusively interpret the structures. 2.5. Cytotoxicity against Vero and human liver C3A cells The cytotoxicity of the extracts, fractions and isolated compound against Vero monkey kidney cells and human liver (C3A) cells was determined using the MTT (3-[4.5-dimethylthiazol-2-yl]-2.5 diphenyltetrazolium bromide) reduction assay as previously described by Mosmann (1983) with slight modifications by McGaw et al. (2007). Cells were seeded at a density of 1 × 105 cells/ml (100 ml) in 96-well microtitre plates and incubated at 37 °C and 5% CO2 in a humidified environment. After 24 h incubation, test samples (100 μl) at varying final concentrations were added to the wells containing cells. Doxorubicin was used as a positive control. Suitable blanks control with equivalent concentrations of acetone and DMSO were also included and the plates were further incubated for 48 h in a CO2 incubator. Thereafter, the medium in each well were aspirated from the cells, which were then washed with phosphate buffered saline (PBS), and finally fresh medium (200 μl) was added to each well. Then, 30 ml of MTT (5 mg/ml in PBS) was added to each well and the plates were incubated at 37 °C for 4 h. The medium was aspirated from the wells and DMSO was added to solubilize the formed formazan crystals. The lack of purple formazan colour or clear appearance in the wells indicated cytotoxicity of the tested samples on the cells. The absorbance was measured on a BioTek Synergy microplate reader at 570 nm. The percentage of cell growth inhibition was calculated based on a comparison with untreated cells and linear regression was used to calculate the cytotoxicity of the tested samples which was expressed as the concentration leading to 50% deaths (IC50), i.e. inhibition of cell growth by 50%. The selectivity index values were calculated by dividing cytotoxicity IC50 values by the MIC values in the same units (SI = IC50/MIC).

½dpm of blank ‐ dpm of background   100 3. Results and discussion where dpm is the disintegrations · min−1 (Jin et al., 1999). 3.1. Minimal inhibitory concentrations of the crude extracts and fractions 2.4. Isolation of bioactive compounds from chloroform fraction of F. africana crude extract Column chromatography was chosen for separation of compounds using 200 g silica gel column (40 cm × 2.5 cm) as a stationary phase. The chloroform fraction (8.1 g) was dissolved in 50 ml acetone, mixed with 8.1 g silica gel and dried before it was layered on the column bed. Initially, 600 ml of hexane (H) was gradually added into the column to elute non-polar compounds. The polarity was increased by addition of ethyl acetate (EtOAc) at an interval of 5% until 100% EtOAc. One hundred and seven fractions were collected and fractions containing similar constituents were combined (monitored by TLC fingerprinting and bioautography) and labelled fractions 1 to 7. Fractions 6 and 7 from the first column were combined to yield 1.12 g. This was mixed with 10 g of silica gel 60 in hexane. A glass column (40 cm × 2 cm) packed with 100 g silica gel 60 slurry was used. The 1.12 g yield-silica gel mixture was added on top of the column. The column was eluted with 95% H in EtOAc to 40% EtOAc with an increment of 5%. An equal volume (500 ml) was added for each increment. Fractions were collected and developed on TLC using H:EtOAc (7:3) as an eluent, sprayed with vanillin:sulphuric acid in methanol. Fractions containing similar compounds were combined. A glass column (40 cm × 1.5 cm) packed with 70 g silica slurry prepared was used to separate the dried mixture of the resultant yield (320 mg) containing the antimicrobial compound of interest. The column was eluted with 100 ml of 5%, 10%, 15% and 20% EtOAc in hexane until one compound (20 mg) present in a very low concentration was isolated. The structure of the isolated compound

The acetone extract of F. africana had activity against all the tested bacteria, with MIC values as low as 80 μg/ml against both P. aeruginosa and S. aureus, which is considered pharmacologically important for the crude extracts (Table 1). The hexane and chloroform fractions had low MIC values and high total activities compared to the other fractions, as found elsewhere that relatively non-polar compounds play an important role in antimicrobial activity of plant extracts (Kotze and Eloff, 2002). MIC values as low as 20 μg/ml were obtained with the chloroform fraction against P. aeruginosa and the hexane fraction had an MIC value of 80 μg/ml against S. aureus which indicates that non-polar phytochemicals were mainly responsible for the biological activity in the crude extract. The MIC values of the chloroform fraction were higher than those of the crude extract except in the case of S. aureus, where the values were the same. Since there was excellent quantitative recovery of the crude extract after solvent-solvent fractionation, the differences are probably due to synergistic activities of compounds present together in the crude extract but separated from each other in the fractions. The crude extract and the fractions had antifungal activity against all the tested opportunistic yeasts which commonly affect immunocompromised patients, as well as economically important fungal phytopathogens (Table 1). An MIC value as low as 40 μg/ml was obtained for the crude extract, and also with hexane and chloroform fractions, against the standard ATCC strain of C. albicans. Other opportunistic fungal pathogens such as A. fumigatus and C. neoformans also had good activity against the non-polar fractions and crude extract with MIC value of 80 mg/ml. However, all the polar fractions had poor activity

T.E. Ramadwa et al. / South African Journal of Botany 108 (2017) 126–131

129

Table 1 MIC values (mg/ml) of the F. africana crude extract, five fractions and isolated methyl ursolate against four bacterial and seven fungal species. Microorganisms

MIC (μg/ml) of the crude extract, fractions and methyl ursolate Extract

Hexane

CHCl3

WM

Butanol

H2O

MU

Gent

Amp

E.c E.f P.a S.a A.f C.a 1 C.a C.n F.o P.j R.s

240 120 80 80 80 80 80 80 80 320 320

160 160 240 80 80 40 80 160 nt nt nt

160 80 20 80 160 40 80 160 nt nt nt

630 630 320 320 2500 63 2.5 2.5 nt nt nt

630 1250 320 320 1250 630 2500 2500 nt nt nt

2500 1250 2500 1250 1250 630 2500 1250 nt nt nt

N250 N250 N250 N250 nt nt 125 125 62.5 N250 N250

8.0 1.6 0.2 0.3 – – – – – – –

– – – – 0.2 0.16 0.16 0.16 2.5 0.02 0.02

NB: nt; Not tested, CHCl3; Chloroform,; WM; 30% H2O in methanol, MU; Methyl ursolate, E.c; Escherichia coli, E.f; Enterococcus faecalis, P.a; Pseudomonas aeruginosa, S.a; Staphylococcus aureus, F.o; Fusarium oxysporum, P.j; Penicillium janthinellum, R.s; Rhizoctonia solani, A.f; Aspergillus fumigatus, C.a; Candida albicans, Ca1; Candida albicans ATCC 10231, C.n; Cryptococcus neoformans, Gent; Gentamicin, Amp; Amphotericin B.

against all the tested fungi with MIC values as high as 2500 μg/ml, which indicates that mainly non-polar compounds are responsible for the antifungal activity. The results are consistent with other reports on species of the same genus such as F. elastica which has been reported to have antifungal activity (Adekunle and Ikumapayi, 2006). 3.2. Anti-inflammatory activity The percentage inhibition of prostaglandin synthesis by COX-1 and COX-2 of F. africana acetone crude extract and its fractions with indomethacin as the control are shown in Table 2. To classify a plant extract tested at a final concentration of 250 μg/ml as active, the inhibition by aqueous extracts tested must be at least 59% and for ethanolic extracts 70% per solution (Fennell et al., 2004). The crude extract had a moderate activity for both COX-1 (59.7 ± 1.4%) and COX-2 (54.3 ± 0.3%). Of the five F. africana fractions obtained after solvent-solvent fractionation, the hexane and chloroform fractions had moderate activity against COX-1 and COX-2. The other three polar fractions had no inhibitory effect on COX-1 or COX-2. The presence of polyphenols, saponins, certain pigments or fatty acids in crude extracts can lead to false positive results in enzyme assays (O'Neil and Lewis, 1993). The constituents of a plant extract may have an inhibitory effect in the assay by denaturing the enzyme, or by acting on the prosthetic group thus inactivating the enzyme. Tannins and phenols have such effects in in vitro test (Van Wyk et al., 1997). Because only the non-polar fractions were active and polyphenols or saponins would not be expected in the non-polar fractions, it is unlikely that these compounds could have been responsible for the activity. Long chain fatty acids would have been expected in the hexane fraction and may be responsible for the activity in this fraction (O'Neil and Lewis, 1993). Although the crude extract together with hexane and chloroform fractions had moderate activity, neither the crude extract nor the two fractions had selective activity against either COX-1 or COX-2. More emphasis is on finding Non-steriodal anti-inflammatory drugs (NSAIDs) which selectively inhibit COX-2 with little interference of COX-1. Compounds which are selective inhibitors of COX-2 may have anti-inflammatory and non-ulcerogenic activity and are therefore of Table 2 Percentage inhibition of COX-1 and COX-2 prostaglandin synthesis by a crude leaf extract of F. africana and its solvent-solvent fractions (250 μg/ml) from two experiments. Indomethacin (μg/ml) was used as a positive control. Samples

COX-1% inhibition

COX-2% inhibition

Crude extract Hexane fraction Chloroform fraction Indomethacin

59.7 ± 1.4 45.9 ± 3.2 68.2 ± 6.6 3.3 ± 0.008

54.3 ± 0.3 50.4 ± 9.9 59.1 ± 4.5 122 ± 5.7

Values are the mean ± S.E.M percentages of results obtained from two experiments.

considerable interest for therapeutic use (Mantri and Witiak, 1994). However, certain precautions must be taken because COX-2 selective inhibitors can cause thrombosis and hence heart attack (Bombardier et al., 2000; Mukherjee et al., 2001; Mamdani et al., 2004). It is unlikely that the use of F. africana for pain relief can be attributed to COX-1 and COX-2 inhibition. Because pain relief is based on a quick response, one would have expected high activity in determining anti-inflammatory activity. It is however possible that plant extracts without high activity in anti-inflammatory activity in the cyclooxygenase assay might exert their anti-inflammatory activity through other mechanisms. The methanol crude extract and chloroform fraction of the leaf extracts of F. africana have been reported to exhibit larvicidal and antiinflammatory effects in in vivo studies (Adediwura et al., 2015). 3.3. Cytotoxicity of the crude, fractions and isolated methyl ursolate The results of the cytotoxicity effects of the F. africana acetone crude extracts, chloroform fraction and isolated compound methyl ursolate on Vero and C3A cell lines are presented in Table 3. The crude extract had a low toxicity to either the Vero or C3A cells with IC50 values of 139.4 μg/ml and 625.5 μg/ml respectively, considering that extracts with an IC50 value of less than 20 μg/ml threshold are regarded as highly toxic (Zirihi et al., 2005). The fractionation of the crude extract have influenced and potentiated the cytotoxicity of the chloroform fraction with IC50 values of less than 90 μg/ml against the tested cell lines compared to the cytotoxicity of the crude extract. The cytotoxicity of the crude extracts of F. africana may explain the use of the plant species to treat different ailments. 3.4. Structure elucidation and biological activity of the isolated compound One antimicrobial compound was isolated from the chloroform fraction of the crude extract of F. africana using a bioassay guided fractionation. The structure of the compound was established using 1 H, 13C NMR and mass spectrometry data as methyl ursolate (Fig. 1), isolated previously from almond hulls. The structure was confirmed by comparison of the nuclear magbetic resonance and mass

Table 3 Cytotoxicity (IC50 in μg/ml) of the F. africana leaf extracts, chloroform fraction and isolated compound methyl ursolate on Vero monkey kidney and human liver (C3A) cells. Samples

Crude extract Chloroform fraction Methyl ursolate Doxorubicin (μM)

IC50 (μg/ml) Vero cells

Human liver C3A cells

139.4 ± 0.00035 85.5 ± 0.00087 10.4 ± 0.25 4.8 ± 0.32

629.5 ± 0.18 80.9 ± 0.0021 38.0 ± 1.2 0.12 ± 0.08

130

T.E. Ramadwa et al. / South African Journal of Botany 108 (2017) 126–131

30 29 20

19 12 25

11

10 3

HO 23

4

5

9

18 17

26

1 2

13 14

8

16

21 22

COOMe

28

and anti-inflammatory effects in animal models (Fabri et al., 2014; Ma et al., 2005; Pádua et al., 2014). Methyl ursolate had some degree of toxicity with an IC50 value of 10.4 μg/ml against Vero cells (Table 3). According to established cytotoxicity criteria, an IC50 of less than 4 μg/ml is generally considered toxic for compounds (Boik, 2001; Brahemi et al., 2010; Kuete and Efferth, 2015). Methyl ursolate has also been reported to be toxic to tumoral cell lineages and resulted in no inhibition of nitric oxide production of macrophages stimulated with lipopolysaccharide (Ma et al., 2005; Fabri et al., 2014).

15 7

27

6

24

Fig. 1. The structure of methyl ursolate isolated from the leaves of F. africana.

spectroscopy spectra with those in literature (Seo et al., 1975; Takeoka et al., 2000). The MIC values of the isolated methyl ursolate tested against four bacteria are shown in Table 1. Plant extracts with antimicrobial activities of less than 100 μg/ml are accepted as potentially having clinical relevance (Eloff, 2004; Gibbons, 2004)). Furthermore, compounds with MIC values of 10 μg/ml or less are noteworthy (Rios and Recio, 2005). Ojinnake and Kenne (1985) also reported the isolation of methyl ursolate from the stem of Myrianthus arboreus. The isolated methyl ursolate from the leaves of F. africana had a low activity with MIC values of N250 μg/ml against all the four tested bacteria. It is important to keep in mind the complexity of plant extracts and that a single compound may not be responsible for the activity but rather a combination of compounds (either major or minor) interacting in an additive or synergistic manner (Van Vuuren, 2007). As a result, the possibility that the higher antibacterial activity of the crude extract compared to the isolated methyl ursolate may be due to synergistic interactions of different compounds in the crude extract should not be ruled out. It was interesting that the isolated methyl ursolate was still detected on TLC bioautograms when tested against E. coli, E. faecalis, P. aeruginosa and S. aureus at 25 μg and 12.5 μg with clear zones indicating bacterial inhibition by the compound This also demonstrated that methyl ursolate was not an artefact of the isolation process (Results not shown). Three alkaloids have been isolated from the stem bark of the same plant species (Wagner et al., 1987). Methyl ursolate differs from ursolic acid only due to the presence of a methyl group at C-28, instead of the carboxyl group. Ursolic acid had poor antibacterial activity against E. coli ATCC 25922 (N 256 mg/L), P. aeruginosa ATCC 27853 (256 mg/L), and E. faecalis ATCC (4 mg/L), S. aureus ATCC 29213 (8 mg/L) (Fontanay et al., 2008). The esterification of 28-COOH is essential for activity of triterpenoids (Ma et al., 2005). MIC values of the isolated methyl ursolate from F. africana against opportunistic yeasts and phytopathogenic fungi are presented in Table 1. The methyl ursolate was generally more active against the tested fungi with MIC values ranging from 63 μg/ml to N250 μg/ml. The isolated compound had a higher activity against F. oxysporum with an MIC value of 63 μg/ml. The MIC value for C. albicans of 250 μg/ml is consistent with the results of Haraguchi et al. (1999) for the same fungal pathogen with MIC value of N200 μg/ml. The crude extract had good antifungal activity against C. neoformans compared to the pure isolated compound, which might be due to synergistic effects of the active and non-active compounds in the crude extract. Funtumia elastica which belongs to the same genus was reported to possess antifungal activity (Adekunle and Ikumapayi, 2006). Methyl ursolate has also been reported to have in vitro biological activities against tumour cell lines

4. Conclusion The F. africana crude extract and non-polar fractions had varying antibacterial and antifungal activities. The low toxicity of the crude extracts to Vero cells may support the use of the plant in traditional medicine. Especially against P. aeruginosa the chloroform fraction had excellent activity with an MIC of 20 μg/ml leading to a selectivity index of N4. The isolated methyl ursolate was relatively toxic and inactive against pathogenic fungi and bacteria with noteworthy activity against only F. oxysporum. Yet the MIC of the chloroform fraction was more than a hundred times lower than the MIC of methyl ursolate, which provides strong evidence for synergistic activities. Compounds that are inactive when separated from others in bioautography may have an effect on the absorption and/or metabolism of the active compound in the microorganism. In this study bioautography was used as a measure of antibacterial activity in the bioassay guided fractionation. If the MIC values of fractions were determined the large loss of activity would have been noticed. Although there are some difficulties to determine the MIC values of many fractions, this appears to be a reasonable approach to isolate antimicrobial compounds. It is unlikely that the use of F. africana for pain relief can be ascribed to COX-1 or COX-2 inhibition because most traditional healers usually only have water available as an extractant and only the non-polar fractions of the crude extract had some activity against these enzymes. It is however possible that plant extracts without high activity in the cyclooxygenase assay might still exert their anti-inflammatory activity through other mechanisms (McGaw et al., 1997). To the best of our knowledge, this is the first report of the isolation of methyl ursolate from F. africana. Acknowledgements The National Research Foundation (IPPR 95991) and the University of Pretoria provided funding for the research. Dr. Lita Pauw helped and the curator Willem Froneman gave us permission to collect plant material from the Lowveld National Botanical Gardens, Nelspruit. Dr. Nivan Moodley from the Council for Scientific and Industrial Research also helped to resolve the structure of the compound. References Adediwura, F.A., Tochukwu, D., Oluwanisola, O., 2015. Larvicidal and anti-inflammatory activities of Funtumia africana (Benth) Stapf leaf and stem. International Journal of Phytomedicine 7, 55–61. Adekunle, A.A., Ikumapayi, A.M., 2006. Antifungal property and phytochemical screening of the crude extracts of Funtumia elastic and Mallotus oppositifolius. West Indian Medical Journal 55, 219. Adjanohoun, E.J., Aké, A.L., 1979. Contribution au recensement des plantes médicinales de Côte d'Ivoire. Centre National de Floristique, Abidjan, p. 142. Adjanohoun, E.J., Ahyi, M.R.A., Aké Assi, L., Akpagana, K., Chibon, P., El Hadj Watara, A., Eymé, J., Garba, M., Gassita, J.N., Gbéassor, M., Goudoté, E., Guinko, S., Hodouto, K.K., Houngnon, P., Kéita, A., Kéoula, Y., Kluga, O.W.P., Lo, I., Siamevi, K.M., Taffame, K., 1986. Contribution aux Études Ethnobotaniques et Floristiques du Togo. ACCT, Paris, pp. 122–135. Ashidi, J.S., Houghton, P.J., Hylands, P.J., Efferth, T., 2010. Ethnobotanical survey and cytotoxicity testing of plants of South-western Nigeria used to treat cancer, with isolation of cytotoxic constituents from Cajanus cajan millsp. leaves. Journal of Ethnopharmacology 128, 501–512.

T.E. Ramadwa et al. / South African Journal of Botany 108 (2017) 126–131 Beentje, H.J., 1994. Kenya Trees, Shrubs and Lianas. National Museums of Kenya. Begue, W.J., Kline, R.M., 1972. The use of tetrazolium salts in bioautographic procedures. Journal of Chromatography 64, 182–184. Boik, J., 2001. Natural Compounds in Cancer Therapy. Oregon Medical Press, Princeton, MN, USA. Bombardier, S., Cattani, P., Ciabattoni, G., Di Munno, O., Pasero, G., Patrono, C., 2000. The synovial prostaglandin system in chronic inflammatory arthritis: differential effects of steroidal and nonsteroidal anti-inflammatory drugs. Brazilian Journal of Pharmacology 75, 893–901. Brahemi, G., Kona, F.R., Fiasella, A., Buac, D., Soukupova, J., Brancale, A., Burger, A.M., Westwell, A.D., 2010. Exploring the structural requirements for inhibition of the ubiquitin E3 ligase breast cancer associated protein 2 (BCA2) as a treatment for breast cancer. Journal of Medicinal Chemistry 53, 2757–2765. Burkill, H., 1960. The Useful Plants in Nigeria. Nigerian College of Arts, Science and Technology, Lagos, p. 90. Eloff, J.N., 1998a. A sensitive and quick method to determine the minimum inhibitory concentration of plant extracts for bacteria. Planta Medica 60, 1–8. Eloff, J.N., 1998b. Which extractant should be used for the screening and isolation of antimicrobial components from plants? Journal of Ethnopharmacology 60, 1–8. Eloff, J.N., 2004. Quantifying the bioactivity of plant extracts during screening and bioassay-guided fractionation. Phytomedicine 11, 370–371. Fabri, R.L., Garcia, R.A., Florêncio, J.R., Carvalho, L.O., Pinto, N.C.C., Coimbra, E.S., SouzaFagundes, S., Ribeiro, A., Scio, E., 2014. Pentacyclic triterpenoids from Mitracarpus frigidus (Willd. Ex Roem. & Schult.) K. Shum: in vitro cytotoxic and leishmanicidal and in vivo anti-inflammatory and antioxidative activities. Medicinal Chemistry Research 12, 5294–5304. Fennell, C.W., Lindsey, K.L., MacGaw, L.J., Sparg, S.G., Stanfford, G.I., Elgorashi, E.E., Grace, O.M., Van Staden, J., 2004. Assessing African medicinal plants efficacy and safety: pharmacological screening and toxicology. Journal of Ethnopharmacology 94, 205–217. Fontanay, S., Grare, M., Mayer, C.F., Duval, R.E., 2008. Ursolic, oleanolic and betulinic acids: antibacterial spetra and selectivity indexes. Journal of Ethnopharmacology 120, 272–276. Gibbons, S., 2004. Anti-Staphylococcal plant natural products. Natural Product Reports 21, 263–277. Gurib-Fakim, A., 2006. Medicinal plants: traditions of yesterday and drugs of tomorrow. Molecular Aspects of Medicine 27, 1–93. Haraguchi, H., Kataoka, S., Okamoto, S., Hanafi, M., Shibata, K., 1999. Antimicrobial activity of triterpenoids from Ilex integra and the mechanisms of antifungal action. Phytotherapy Research 13, 151–156. Jager, A.K., Hutchings, A., Van Staden, J., 1996. Screening of Zulu medicinal plants for prostanglandin- inhibitors. Journal of Ethnopharmacology 52, 95–100. Janot, M.M., Truong-Ho, M., Khuong-Huu, Q., Goutarel, R., 1963. Bulletin de la Société Chimique de France 3, 1977. Jin, J., Opoku, A.R., Geheed-Keller, M., Hutchings, A.D., Terblanche, S.E., Jager, A.K., Van Staden, J., 1999. Preliminary screening of some traditional Zulu medicinal plants for anti-inflammatory and antimicrobial activities. Journal of Ethnopharmacology 68, 267–274. Keay, R.W.J., Onochie, C.F.A., Stanfield, D.F., 1964. Nigerian Trees. Nigerian National press Ltd, p. 495. Kotze, M., Eloff, J.N., 2002. Extraction of antibacterial compounds from Combretum microphyllum (Combretaceae). South African Journal of Botany 68, 62–67. Kuete, V., Efferth, T., 2015. African flora has the potential to fight multidrug resistance of cancer. BioMed Research International 914813, 2015. Ma, C.M., Cai, S.Q., Cui, J.R., Tu, P.F., Hattori, M., Daneshtalab, M., 2005. The cytotoxic activity of ursolic acid derivatives. European Journal of Medicinal Chemistry 40, 582–589. Mahlo, S.M., McGaw, L.J., Eloff, J.N., 2010. Antifungal activity of leaf extracts from South African trees against plant pathogens. Crop Protection 29, 1529–1533. Mamdani, M., Juurlink, D.N., Lee, D.S., Rochon, P.A., Kopp, A., Naglie, G., 2004. Cyclooxygenase-2 inhibitors versus non-selective non-steroidal anti-inflammatory drugs and congestive heart failure outcomes in elderly patients: a population-based cohort study. Lancet 363, 1751–1756. Mantri, P., Witiak, D.T., 1994. Inhibitors of cyclooxygenase and 5-lipoxygenase. Current Medicinal Chemistry 1, 123–129. Masoko, P., Eloff, J.N., 2005. The diversity of antifungal compounds of six South African Terminalia species (Combretaceae) determined by bioautography. African Journal of Biotechnology 4, 1425–1431. Masoko, P., Picard, J., Eloff, J.N., 2005. Antifungal activities of six South African Terminalia species (Combretaceae). Journal of Ethnopharmacology 99, 301–308. McGaw, L.J., Jäger, A.K., Van Staden, J., 1997. Prostaglandin synthesis inhibitory activity in Zulu, Xhosa and Sotho medicinal plants. Phytotherapy Research 11, 113–117.

131

McGaw, L.J., Steenkamp, V., Eloff, J.N., 2007. Evaluation of Athrixia bush tea for cytotoxicity, antioxidant activity, caffeine content and presence of pyrrolizidine alkaloids. Journal of Ethnopharmacology 110, 16–22. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival — application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55–63. Mukam, L., Charles, G., Hentchoya, J., Njimi, T., Ourisson, G., 1973. Le cyclofuntumienol, methyl-4α sterol en C 31 de Funtumia elastica. Tetrahedron Letters 2779. Mukherjee, D., Nissen, S.E., Topol, E.J., 2001. Risk of cardiovascular events associated with selective COX-2 inhibitors. Journal of the American Medical Association 286, 954–959. NCCLS, 1990. The National Comitte for Clinical Laboratory Standards. Performance Standards for Antimicrobial Disk Susceptibility Teste-Fourth Edition; Approved Standard. NCCLS Documant M2-A4. Villanova, Pa. USA. Odugbemi, T.O., Akinsulire, O.R., Aibinu, I.E., Fabeku, P.O., 2007. Medicinal plants useful for malaria therapy in Okeigbo, Ondo State, Southwest Nigeria. African Journal of Traditional, Complementary, and Alternative Medicines 4, 191–198. Ojinnake, C.M., Kenne, L., 1985. Studies of Nigerian medicinal plants: components of the stem of Myrianthus arboreus. Journal of Natural Products 481, 1002–1003. O'Neil, M.J., Lewis, J.A., 1993. The renaissance of plant research in the pharmaceutical industry. In: Kinghorn, A.D., Balandrin, M.F. (Eds.), Human Medical Agents from PlantsACS Symposium Series 534. American Chemical Society, Washington. Orwa, C., Mutua, A., Kindt, R., Jamnadass, R., Simons, A., 2009. Agroforestree Database:a Tree Reference and Selection Guide. Pádua, T.A., de Abreu, B.S.S.C., Costa, T.E.M.M., Nakamura, M.J., M.J., Valente, L.M.M., Henriques, M.G., Siani, A.C., Rosas, E.C., 2014. Anti-inflammatory effects of methyl ursolate obtained from a chemically derived crude extract of apple peels: potential use in rheumatoid arthritis. Archives of Pharmacal Research 37, 1487–1495. Pauw, E., Eloff, J.N., 2014. Which tree orders in southern Africa have the highest antimicrobial activity and selectivity against bacterial and fungal pathogens of animals? BMC Complementary and Alternative Medicine 14, 317. Ramadwa, T.E., 2010. Isolation and Characterization of Antimicrobial Compounds from Funtumia africana (Apocynaceae) Leaf Extracts MSc dissertation University of Pretoria, South Africa. Rios, J.L., Recio, M.C., 2005. Medicinal plants and antimicrobial activity. Journal of Ethnopharmacology 100, 80–84. Sacho, H., Schoub, D.B., 1993. Current Perspectives on Nosocomial Infections, [Sponsored by Glaxo Wellcome]. Natal Witness Printing and Publishing, Pietermaritzburg, p. 100. Seo, S., Tomita, Y., Tori, K., 1975. Biosynthesis of ursine-type triterpenes from sodium (1.2-13C) acetate in tissue cultures of Isodon japonicas Hara and re-assignments of 13C NMR signals in urs-12-ene. Journal of the Chemical Society, Chemical Communications 23, 954–955. Shai, L.J., McGaw, L.J., Masoko, P., Eloff, J.N., 2008. Antifungal and antibacterial activity of seven traditionally used South African plant species active against Candida albicans. South African Journal of Botany 74, 677–684. Suffness, M., Douros, J., 1979. Drugs of plant origin. Methods in Cancer Research 26, 73–126. Suleiman, M.M., McGaw, L.J., Naidoo, V., Eloff, J.N., 2010. Evaluation of several tree species for activity against the animal fungal pathogen Aspergillus fumigatus. South African Journal of Botany 76, 64–71. Takeoka, G., Dao, L., Teranishi, R., Wong, R., Flessa, S., Harden, L., Edwards, R., 2000. Identification of three triterpenoids in almond hulls. Journal of Agricultural and Food Chemistry 48, 3437–3439. Truong-Ho, M., Khuong-Huu, Q., Goutarel, R., 1963. Goutarel: Alcaloïdes stéroïdiques XV. Les irehdiamines A et B, alcaloïdes du Funtumia elastica (Preuss) Stapf. Bulletin de La Societe Chimique de France 3 p. 594. Van Vuuren, S.F., 2007. The Antimicrobial Activity and Essential Oil Composition of Medicinal Aromatic Plants Used in African Traditional Healing Ph.D. Thesis University of the Witwatersrand, South Africa. Van Wyk, B.-E., Van Oudtshoorn, B., Gericke, N., 1997. Medicinal Plants of South Africa. Briza Publications, Pretoria. Wagner, H., Seegert, K., Sonnenbichler, H., Ilyas, M., 1987. Steriods alkaloids of Funtumia africana. Planta Medica 53, 444–449. White, H.L., Glassman, A.T., 1974. A simple radiochemical assay for prostaglandins synthetase. Prostaglandins 7, 123–129. Zirihi, G.N., Grellier, P., Gue'de´-Guina, F., Bodo, B., Mambu, L., 2005. Isolation, characterization and antiplasmodial activity of steroidal alkaloids from Funtumia elastica (Preuss) Stapf. Bioorganic and Medicinal Chemistry Letters 15, 2637–2640.