Cytotoxic and antimicrobial activity of selected Cameroonian edible plants

Abstract

Background

In Cameroon, the use of edible plants is an integral part of dietarybehavior. However, evidence of the antimicrobial as well as the cytotoxiceffects of many of them has not been investigated. In the present study,aqueous and methanol extracts from barks, seeds, leaves and roots of threeCameroonian edible plants namely Garcina lucida, Fagara heitzii andHymenocardia lyrata were evaluated for their cytotoxic andantimicrobial activities.

Methods

Antibacterial and antifungal activities were assessed by the brothmicro-dilution method meanwhile the cytotoxicity was performed usingsulphorhodamine B assay (SRB) against the human leukemia THP-1, the alveolarepithelial A549, prostate cancer PC-3, breast adenocarcinoma MCF-7 andcervical cancer HeLa cell lines.

Results

The minimum inhibitory concentration (MIC) values of the seven testedextracts ranged from 62.5 μg/ml to 1000 μg/ml. Themethanol (MeOH) extract from the roots of H. lyrata showed thehighest antibacterial activity against Gram-positive bacteria S.aureus and S. epidermitis. The best antifungal activitywas obtained with the MeOH extract from the leaves of G. lucidaagainst C. tropicalis (MIC value of 62.5 μg/ml). Thein vitro antiproliferative activity revealed that, extract fromthe bark of F. heitzii and extract from H. lyrata rootshad significant cytotoxic activity on THP-1 (IC₅₀8.4 μg/ml) and PC-3 (IC₅₀ 9.5 μg/ml)respectively.

Conclusion

Our findings suggest that Cameroonian spices herein studied could bepotentially useful for the development of therapeutic agents againstbacterial infections as well as for prostate and leukemia cancer.

Background

Plants have served humans well as valuable components of seasonings as well asmedicines, and have played a significant role in maintaining human health andimproving the quality of human life for thousands of years. There is no doubt thatincreasing our intake of spices is one of the most effective, convenient andeconomical ways in which we can fortify ourselves against infectious diseases andrelated cancers [1]. In the area of cancer prevention, plants consumption such as spices andtheir constituents as potential chemopreventive agents remains an extensive researchtopic. Numerous studies have been published in regards to the relation betweenplants consumption, cancer prevention, antimicrobial effects, and overall protectionof human health [2]. In Cameroon, medicinal plants consumption is an integral part of dietarybehavior, but relatively little is known about their antimicrobial potential andanti-cancer effects. The three selected edible plants studied herein, namelyGarcinia lucida Vesque, Fagara heitzii (Guill and Perr) Engl.and Hymenocardia lyrata Tul are recognized for their medicinal virtues andhave been reported to possess various biological activities. The barks, seeds andleaves of G. lucida are used to treat gastric infections, gynecologicalinfections, diarrheas and as anti-poison [3, 4]. It has been reported to have antileishmanial and anti-trypanosomalproperties [5, 6]. Seeds and barks of Fagara heitzii (Guill and Perr) Engl havebeen used to treat abdominal pains, asthma, appendicitis and toothache [7]. H. lyrata is well known in Cameroon, where roots and barks areused as antiparasitic, against gastric ulcer and in high blood pressure regulation [8]. There are few numbers of reports about the potential of Cameroonianedible plants in terms of preventing and treating microbial diseases and cancer [9, 10]. The objective of this study was to evaluate aqueous and methanolextracts from barks, seeds, leaves and roots of three Cameroonian plants edibleplants for their anticancer activity on human cancer cell lines of various tissuesincluding as cervix, leukemia, prostate, lung and breast; as well as theirantimicrobial activities against bacterial and fungal pathogens.

Methods

Plant material and extraction

F. heitzii ( Rutaceae), G. lucida (Guttiferae) and H.lyrata (Euphorbiaceae)were harvested in the East region, Ebolowa andBatchingou in Cameroon. The plants were identified at the Cameroon NationalHerbarium where voucher specimens were deposited under the reference number1441/HNC, 53354/HNC and 32301/HNC respectively for F. heitzi, G. lucida andH. lyrata.

Each collected sample (leaves and barks for G. lucida, roots and barksof H. lyrata, and fruits and roots for F. heitzii) was driedat room temperature (28 ± 3°C), pulverized and powdered. Each powder(50 g) was macerated in 500 ml of methanol or water for 72 h atroom temperature. After 72 h, the mixture was filtered using a paper filterWhatman No. 1. Each filtrate was then concentrated under vacuum (Rotaryevaporator, Heidolph WB 200) to obtain the crude extract. Each crude extractobtained was then weighed and stored at 4°C.

Cytotoxic activity assay

Cell lines and treatment

The effect of the extracts and compounds on cell growth was determined in apanel of human tumor cells including lung A549 adenocarcinoma, breastcarcinoma MCF-7, prostate carcinoma PC-3, cervical carcinoma HeLa and acutemonocytic leukemia cell line THP-1, obtained from National Cancer Institute,USA. THP-1, A-549 and PC-3 were maintained in RPMI medium while MCF-7 andHeLa were cultured in MEM medium. All media used were supplemented with 10%fetal bovine serum (FBS), 100 IU/ml penicillin. The cell lines weremaintained under standard cell culture conditions at 37°C and 5%CO₂ in a humidified environment.

Cytotoxic activity by SRB assay

In vitro cytotoxicity against above mentioned five human cancercell lines was determined using sulphorhodamine B assay (SRB) as describedpreviously [11]. Briefly, cells were harvested in log phase using trypsin (0.05%trypsin, 0.02% EDTA, in PBS). The cell suspensions were diluted withappropriate growth medium to obtain the cell densities depending on the cellline: (10⁴ cells/well for HeLa, 10⁴ cells/well forA549, 10⁴ cells/well for THP-1, 1, 5×10⁴cells/well for MCF-7 and 10⁴ cells/well for PC-3). An aliquot of100 μl of each suspension were seeded in 96 wells cell cultureplates. The cells were incubated at 37°C in an atmosphere of 5%CO₂ and 95% relative humidity in a CO₂ incubator.After 24 h incubation, test materials (100 μl/well) atvarying concentrations (1, 10, 30 and 50 or 100 μg/ml) were addedto the wells containing cells. Paclitaxel 0.1, 1 and 10 μM wasused as positive control. Suitable controls with equivalent concentration ofDMSO were also included. The plates were further incubated for 48 h ina CO₂ incubator after addition of test material. After incubationcells were fixed by gently layering trichloroacetic acid(50 μl/well, 50% w/v) on top of the medium in all the wells andincubated at 4°C for 1 h. The plates were washed five times withdistilled water and air-dried. Cell growth was measured by staining withsulforhodamine B dye (0.4% w/v in 1% acetic acid,100 μl/well). The unbound dye was washed 3–5 times with 1%acetic acid and plates were air dried. The adsorbed dye was dissolved inTris-Buffer (100 μl/well, 0.01 M, pH 10.4) and plateswere gently shaken for 10 min on a mechanical shaker. The opticaldensity (OD) was recorded using a 96 well plate reader. Growth inhibitionwas calculated by subtracting mean OD values of respective blank from themean OD value of experimental set. Percentage growth in presence of testmaterial was calculated considering the growth in absence of any testmaterial as 100% and in turn percentage growth inhibition in presence oftest material was calculated. The viability and growth in the presence oftest material is calculated by following formula.

\begin{array}{l}\mathit{Growth}\mathit{inhibition}\left(\%\right)=\left[O{D}_{\left(\mathit{test}\mathit{sample}\right)}-O{D}_{\left(\mathit{blank}\right)}/\right.\\ \left.O{D}_{\left(\mathit{control}\right)}-O{D}_{\left(\mathit{blank}\right)}\right]\times 100\end{array}

IC₅₀ value is the concentration of sample required to inhibit 50%of the cell proliferation and was calculated by plotting the percentagesurvival versus the concentrations, using Microsoft Excel. For all samples,each compound concentration was tested in triplicates in a singleexperiment.

Antimicrobial assays

Microbial growth conditions

A total of ten microbial strains were tested obtained from the American TypeCulture Collection for their susceptibility to extracts and compounds. Thesestrains comprised of three yeasts: Candida albicans (ATCC 90028),Candida krusei (ATCC 6258), and Candida tropicalis(ATCC 750); one filamentous fungi: Aspergillus fumigatus (MTCC1811); three Gram-negative bacteria: Pseudomonas aeruginosa ATCC27853, Escherichia coli ATCC25292, vancomycin-resistantEnterococcus faecalis (VRE) and three Gram-positive bacteria:Staphylococcus aureus ATCC 29213, methicillin-resistantStaphylococcus aureus (MRSA, ATCC 33591) and Staphylococcusepidermidis (ATCC 12228). They were maintained on agar slant at4°C and sub-cultured on a fresh appropriate agar plates 24 h priorto any antimicrobial test. The Mueller Hinton Agar (MHA) and Sabourauddextrose Agar (SDA) were used for the activation of bacteria and fungirespectively. The Mueller Hinton Broth (MHB) and RPMI 1640 were used for theMIC determinations.

Inoculum preparation

Suspensions of bacteria and yeasts were prepared in sterile normal saline(0.85%) from 24 h grown on SDA or MHA at 37°C. The turbidity ofthe microbial suspension was adjusted with a densitometer to a McFarlandstandard of 0.5 for bacteria and 0.9 for yeast, which are equivalent to1–5 × 10⁸ CFU/ml and 1–5 ×10⁷ CFU/ml respectively.

Inoculum suspensions of Aspergillus species were prepared fromfresh, mature (3 to 5 days old) cultures grown on Sabouraud agar orpotato dextrose agar slants. The colonies were covered with approximately5 ml of distilled containing 5% Tween 20. Then, the suspensions weremade by gently probing the colony with the tip of a Pasteur pipette andtransferred to a sterile tube; the resulting suspensions were homogenizedfor 15 s with a vortex mixer at 2000 rpm. The suspension wasfiltered and collected in a sterile tube. The inoculum size was adjusted to1-5 × 10⁶ spores/ml by microscopic enumeration with acell-counting hematocytometer. Adjusted suspensions were checked by plating0.01 ml of a 1:100 dilution onto PDA plates to determine the viablenumber of CFU/ml. The plates were incubated at 37°C and observed dailyfor the presence of fungal colonies. The colonies were counted as soon aspossible after the observation of visible growth.

MIC determination

The MIC was performed by broth microdilution method, with Mueller HintonBroth (MHB) for bacteria and RPMI 1640 medium (containing L-glutamine,without sodium bicarbonate and buffered to pH 7.0 with 0.165 Mmorpholine propanesulfonic acid) for fungi. Stock solutions of extracts wereprepared in 100% dimethylsulfoxide (DMSO; Sigma) and twofold serialdilutions were prepared in media in amounts of 100 μl per well in96-well. The above-mentioned microbial suspensions were further diluted to1:100 in media, and a 100 μl volume of this diluted inoculum wasadded to each well of the plate, resulting in a final inoculum of1.5×10⁶ cfu/ml for bacteria,1.5×10⁴ cfu/ml for A. fumigatus and1.5×10⁵ cfu/ml for yeasts. The final concentrationof samples ranged from 7.8-1000 μg/ml. The medium without theagents was used as a growth control and the blank control used containedonly the medium. Ciprofloxacin and Amphotericin B served as the standarddrug controls. The microtiter plates were incubated at 37°C for24 h, 48 h and 72 h respectively for bacteria, yeasts andAspergillus species. The plates were read visually, and the MICwas defined as the lowest concentration of the antifungal agents thatprevented visible growth with respect to the growth control.

Statistical analysis

The one-way ANOVA at 95% confidence level was used for statistical analysis.

Results and discussion

In the present study, we have evaluated the antimicrobial and antiproliferativeactivity of MeOH and aqueous extracts from three edible plants used in Cameroon. Ina preliminary screening, the antiproliferative activity of extracts was assessed ona panel of five human cancer cell lines at a single concentration of100 μg/ml. The five human cell lines used were representative of tumorsfrom a five types of human tissue including blood, lung, breast, prostate and cervixtissues. Result of the growth inhibition effects are shown in Figure 1. A perusal of this figure revealed that all the extractstested induced more than 50% cell death on at least one of the five cell lines. Themost potent extract identified, the methanol extract from Fagara heitziibarks (FHB) were found to induce over 60% cell death in 4 out of 5 cell lines. Withthe respective highest growth inhibition percentage of 95.55%, 69.30% and 83.11%,the acute monocytic leukemia cell line THP-1, breast carcinoma MCF-7 and cervicalcarcinoma HeLa) were more sensitive to the extracts than the two others tumor celllines tested (lung A549 adenocarcinoma, prostate carcinoma PC-3). This observationindicated that the extracts selectively inhibited the growth of different tumorcells. To some extent, these results were similar to those of previous studies thatextracts from some spices selectively inhibit the growth of human cancer cell [10, 12, 13]. In contrast, all the extracts had low inhibitory effects on the growthof A549 suggesting the resistance of this cell line towards the tested samples.

Figure 1

Cytotoxic activity of Extracts and paclitaxel determined by the percentageof growth inhibition. Extracts were tested at 100 μg/mL.Data with different alphabetic letters are significantly different (P <0.05). FHF: Methanol extract from Fagara heitzii fruits; FHB:Methanol extract from Fagara heitzii barks; GLB: Methanol extractfrom Garcina lucida barks; GLL: Methanol extract from Garcinalucida leaves; HLR: Methanol extract from Hymenocardialyrata roots; HLBb: Methanol extract from Hymenocardialyrata barks.

All the extracts were initially screened at one concentration (100 μg/ml);then dose-dependant inhibition was further performed to determine IC₅₀(Table 1). Results indicated that almost all plantextracts exerted cytotoxic activity against at least one of the five cell lines used(IC₅₀ varying from 8.4 μg/ml to 99 μg/ml).Significant antiproliferative properties were observed with FHB and the methanolextract from H. lyrata roots (HLR) with IC₅₀ of8.4 μg/ml and 9.5 μg/ml against THP-1 and PC-3 respectively.Interesting cytotoxic activity was also observed against MCF-7 cell line, for themethanol extract from F. heitzii fruits and barks (FHF and FHBrespectively) and the methanol extract from H. lyrata roots (HLR) withIC₅₀ of 26 μg/ml, 42 μg/ml and 32 μg/mlrespectively, while extract from Garcinia lucida barks and leaves (GLB andGLL respectively) demonstrated weak antiproliferative ability with IC₅₀of 73 μg/ml and 82 μg/ml respectively. According to the criteriaof the American National Cancer Institute the anticancer activity of a crude extractpromising for further purification based in the IC₅₀ values is lower than30 μg/mL [14]. Considering this criteria, IC₅₀ values of FHF(IC₅₀ of 26 μg/ml), FHB (IC₅₀. of8.4 μg/ml) and HLR (IC₅₀. of 9.5 μg/ml) respectivelyon MCF-7, THP-1 and PC-3 are well within the limit. Therefore, FHF, FHB and HLRcould be considered as promising sources for new natural products with cytotoxicproperties. Extracts from F. lepreuiri, Fagara macrophylla andGarcinia lucida (fruits) have earlier been reported to containpotentially antiproliferative activity [10]. These results are consistent with previous work that has been carriedout on dietary plants confirming their ability to prevent cancer [2].

Table 1 IC ₅₀ values of the extracts from threeCameroonian edible plants and paclitaxel against human cancer celllines

Results of the antimicrobial assay are depicted in Table 2. All the extracts tested in this study exhibited antimicrobial activitiesagainst bacteria and/or fungi. The minimum inhibitory concentration (MIC) valuesranged from 62.5 μg/ml to 1000 μg/ml. HLR had the highestantibacterial and antifungal activity against Gram-positive bacteria S.aureus and S. epidermitis. According to Kuete [15], the antimicrobial activity of extracts can be classified as follows:significant if MIC values are below 100 μg/ml, moderate when100<MIC<625 μg/ml and weak if MIC>625 μg/ml. Therefore,the overall antibacterial activity exhibited in this study varied from weak tosignificant. Among the six bacterial strains tested, the two Gram-positive bacteria(S. aureus and S. epidermitis) were the most sensitive to theextracts, while the two Gram-negative bacteria strains (E. coli and P.aeruginosa) were the most resistant. These results were consistent with thecommon observation that Gram-positive bacteria are generally more sensitive to thespice and herb extracts than Gram-negative bacteria. The resistance of Gram-negativebacteria towards antibacterial substances is due to their outer membrane thatcontributes to the intrinsic resistance by acting as an efficient permeabilitybarrier [16]. The antibacterial activity of other edible plants was previouslyreported by many authors [17, 18]. It is noteworthy that HLR showed moderate growth inhibiting activityagainst MRSA and VRE which are two bacterial strains expressingMDR phenotypes. Probably the activity exerted by these extracts is due to thepresence of natural bioactive compounds with either a new mode of action or whichare able to escape or to inhibit the mechanism of resistance of these MDR bacteria.Previous study reported the ability of some plant extracts to inhibit efflux pumpsresistance mechanism in bacteria [9, 19]. The tested extracts differed greatly in their activity against fungi andthe best inhibition was observed with GLL with MIC of 62.5 μg/ml againstC. tropicalis. In a similar study, Hamza et al. [20] reported that extracts having MIC of 0.5 mg/ml or less as beingstrong inhibitors of fungal growth. Their report was based on classification of MICearlier proposed by Aligiannis et al. [21] who classified plant extracts having MIC of 0.5 mg/ml as stronginhibitors of fungal growth; MIC between 0.6 and 1.5 mg/ml as moderatedinhibitors and extracts having MIC above 1.6 mg/ml considered as weakinhibitors.Taking in consideration this classification, our extracts shown strongand moderate antifungal activity.

Table 2 Minimal Inhibitory Concentrations (MIC) results for antimicrobialactivity of extracts from some Cameroonian dietary spices(μg/ml)

Conclusions

It comes out from this study that the methanol extract of H. lyrata rootssignificanly prevent the proliferation of THP-1 and PC-3 cancer cell lines while themethanol extract of F. heitzii barks inhibit the growth of Gram-positivebacteria. These extracts could be potentially useful for the development oftherapeutic agents against bacteria infections as well as for prostate and leukemiacancer. In addition these results bring supporting data that consumption of plantscould reduce our susceptibility to some cancers. These extracts are good candidatesfor further activity-guided fractionation in the search for new active therapeuticcompounds.