Antibacterial activities of the methanol extracts of ten Cameroonian vegetables against Gram-negative multidrug-resistant bacteria

Background Many edible plants are used in Cameroon since ancient time to control microbial infections. This study was designed at evaluating the antibacterial activities of the methanol extracts of ten Cameroonian vegetables against a panel of twenty nine Gram negative bacteria including multi-drug resistant (MDR) strains. Methods The broth microdilution method was used to determine the Minimal Inhibitory Concentrations (MIC) and the Minimal Bactericidal Concentrations (MBC) of the studied extracts. When chloramphenicol was used as a reference antibiotic, the MICs were also determined in the presence of Phenylalanine-Arginine β-Naphtylamide (PAβN), an efflux pumps inhibitor (EPI). The phytochemical screening of the extracts was performed using standard methods. Results All tested extracts exhibited antibacterial activities, with the MIC values varying from 128 to 1024 mg/L. The studied extracts showed large spectra of action, those from L. sativa, S. edule, C. pepo and S. nigrum being active on all the 29 bacterial strains tested meanwhile those from Amaranthus hybridus, Vernonia hymenolepsis, Lactuca.carpensis and Manihot esculenta were active on 96.55% of the strains used. The plant extracts were assessed for the presence of large classes of secondary metabolites: alkaloids, anthocyanins, anthraquinones, flavonoids, phenols, saponins, steroids, tannins and triterpenes. Each studied plant extract was found to contain compounds belonging to at least two of the above mentioned classes. Conclusion These results confirm the traditional claims and provide promising baseline information for the potential use of the tested vegetables in the fight against bacterial infections involving MDR phenotypes.


Background
Infectious diseases are still a major health concern, accounting for 41% of the global disease burden measured in terms of Disability-Adjusted Life Years (DALYS), close to all noninfectious diseases (43%) and far more than injuries (16%) [1]. One of the main causes of this problem is the widespread emergence of acquired bacterial resistance to antibiotics in such a way that the world is facing today, a serious threat to global public health [2] in the form of not only epidemics, but also pandemics of antibiotic resistance [3]. Several mechanisms have been accounted for, but active efflux plays an important role in this phenomenon [4]. The accumulation of different antibiotic resistance mechanisms within the same strains has led to the appearance of the so called superbugs, or multi-drug resistant bacteria [2]. Due to this problem of resistance to antibiotics, attention is now being shifted towards biologically active components isolated from plant species commonly used as herbal medicine, as they may offer a new source of antibacterial, antifungal and antiviral activities [5]. The potential antimicrobial properties of plants are related to their ability to synthesize several secondary metabolites of relatively complex structures possessing antimicrobial activities [6,7]. Among medicinal plants, vegetables associated to non or less-toxic effects have been shown to possess many medicinal properties [8,9] including antibacterial effects [3]. The present work was therefore designed to investigate the antibacterial effects of ten Cameroonian vegetables namely Amarantus hybridus Linn (Amarantaceae), Vernonia hymenolepis (H.F.) Hook., Lactuca sativa Linn. and Lactuca capensis Thumb. (Asteraceae), Manihot esculenta Crantz (Euphorbiaceae), Phaseolus vulgaris Linn (Fabaceae), Cucurbita pepo Linn and Sechium edule (Jacq) Sw. (Cucurbitaceae), Solanum nigrum Linn. and Capsicum frutescens L. (Solanaceae) against MDR bacteria expressing active efflux pumps

Plant material and extraction
The collected plant materials used in this study were harvested from Dschang, West Region of Cameroon in June 2010 and included the leaves of Amarathus hybridus, Vernonia hymenolepis, Lactuca sativa, Lactuca capensis, Sechium edule, Manihot esculenta, Curcubiata pepo, Solanum nigrum, the cloves of the Green bean (Phaseolus vulgaris), and the fruits of Capsicum frutescens. These plants were identified by Mr Victor Nana of the National Herbarium (Yaoundé-Cameroon) where all the voucher specimens were deposited with the corresponding reference number (Table 1).
Air dried and powdered sample (1 kg) of each plant was extracted with methanol (MeOH) for 48 h at room temperature (25°C), using Whatman Grade No.1 filter paper and concentrated under reduced pressure, then dried to give the crude extracts. All extracts were stored at 4°C until further use.

Bacterial strains and culture media
The studied bacteria included both reference (from the American Type Culture Collection) and clinical strains of Providencia stuartii, Pseudomonas aeruginosa, K. pneumoniae, Escherichia coli, Enterobacter aerogenes and Enterobacter cloacae (See Additional file 1: Table S1 for their features). These clinical strains were obtained from the laboratory "Transporteurs Membranaires, Chimiorésistance et Drug Design, UMR-MD1, IFR 88, UFRs de Médecine et de Pharmacie, Marseille, France". All strains were maintained in Nutrient Broth at 4°C and activated on Mueller Hinton Agar plates 24 h prior to any antimicrobial test. Mueller Hinton Broth (MHB) was used for all antibacterial assays.

Bacterial susceptibility testing
The MICs were determined using the rapid INT colorimetric assay [71,72]. Briefly, test samples were first emulsified in DMSO/MHB (50:50 V/V). The solution obtained was then added to MHB, and serially diluted two fold (in a 96-wells microplate). One hundred microlitres (100 μl) of inoculum (1.5 × 10 6 CFU/ml) prepared in MHB was then added. The plate was covered with a sterile plate sealer, then agitated to mix the contents of the wells using a shaker and incubated at 37°C for 18 h. The final concentration of DMSO was 2.5% and did not affected the microbial growth. Wells containing MHB, 100 μl of inoculum and DMSO at a final concentration of 2.5% served as negative control. The MICs of samples were detected after 18 h incubation at 37°C, following addition of 40 μl of a 0.2 mg/ml INT solution and incubation at 37°C for 30 minutes. Viable bacteria reduce this yellow dye to pink. MIC was defined as the lowest sample concentration that exhibited complete inhibition of microbial growth and then prevented this change [73]. The MBC was determined by adding 50 μL of the suspensions from the wells, which did not show any growth after incubation during MIC assays, to 150 μL of fresh broth. These suspensions were re-incubated at 37°C for 48 hours. The MBC was determined as the lowest concentration of extract which completely inhibited the growth of bacteria [74].
Chloramphenicol, used as reference antibiotic, was tested also in the presence of the PAβN, at 30 mg/L final concentration to confirm the resistance of bacterial strains.

Chemical composition of the vegetable extracts
The results of the qualitative analysis showed that each of the studied plant extract contains at least two classes of secondary metabolites such as alkaloids, anthocyanins, anthraquinones, flavonoids, phenols, saponins, steroids, tannins and triterpenes ( Table 2). Only the extract from A. hybridus contains anthocyanins, while triterpenes were found both in this extract as well as that of C. frutescens. The extract from C. frutescens as well as those from S. edule and M. esculenta contained the highest number of classes of the studied secondary metabolites (five). Alkaloids and phenols were present in all vegetable extracts except that of A. hybridus.

Antibacterial activity of the vegetable extracts
The data summarized in Table 3 show the antibacterial activities of the tested extracts on a panel of twenty-nine Gram-negative bacteria. All extracts were active on at least twelve bacterial strains with MIC ≤ 1024 μg/ml. The extract of C. frutescens showed inhibitory activities against 16 (55.17%) of the 29 tested bacteria whilst that of P. vulgaris inhibited the growth of 12/29 (41.38%) pathogens (narrowest spectrum). None of these two extracts showed any antibacterial activity against Pseudomonas species, but were active against at least one bacterial strain of other studied genus. Extracts from L. sativa, S. edule, C. pepo and S. nigrum displayed the largest spectra of activity, their inhibitory effects being observed on all the 29 Gamnegative bacteria (100% of activity). The extracts from A. hybridus, V. hymenolepis, L. sativa, L. carpensis and M. esculenta also exhibited large spectrum of activity as they were active on 28/29 tested bacteria. The top eight active extracts, with large spectra of activity, showed MIC values generally ranging from 128 to 512 μg/ml. These MIC values were in some of the cases better than those of choramplenicol (Table 3). This was the case with the extract from V. hymenolepis (MIC of 128 μg/ml) against E. aerogenes EA27. The extracts from A. hybridus, S. edule and C. pepo as well as those from L. capensis and M. esculenta were more active than chloramphenicol on at least one of the tested MDR bacteria. The activity of chloramphenicol increased in the presence of PAβN in the majority of the tested bacteria ( Table 3). The best activity was obtained with the extract from A. hybridus with the lowest MIC value of 128 μg/ml observed against 7/29 (25%) tested bacteria. The extracts from P. vulgaris and C. frutescens did not show any MBC value at up to 1024 μg/ml. Concering the eight other vegetable extracts, the MBC results showed values equal to or below 1024 μg/ml in many cases. The extract from C. pepo leaves showed the best MBC spectrum with the values below to 1024 μg/ml recorded on 58,62% (17/29) of the studied microorganisms, followed by those from M. esculenta leaves on 51,72% (15/29), A. hybridus, V. hymenolepis and L. capensis extracts on 44.83% (13/29) and L. sativa on 31.03% (9/29) ( Table 4). Table 4 also shows that M. esculenta exhibited MBC values against all the strains of E. aerogenes and that, in general, the extracts showed values which were not 4-fold greater than the corresponding MICs.

Discussion
In plants, secondary metabolites attract beneficial and repel harmful organisms, serve as phytoprotectants and respond to environmental changes. In animals, such compounds have many beneficial effects including antibacterial and antiviral properties [75,76]. The classes of secondary metabolites detected in the tested vegetables can somehow provide a prelimanry explanation on their activities [77]. In general, the phytochemical contents ( Table 2) were in accordance with the previous reports for some of the vegetables where data were available [11,12,23,38]. It should however be mentioned that the detection of the bioactive phytochemical classes in a plant is not a guarantee for any biological property, as this will depend on the types of compounds, as well as their concentrations and possible interaction with other constituents.
Solanum nigrum has been shown to possess various activities such as antitumorigenic, antioxidant, anti-inflammatory, hepatoprotective diuretic and antipyretic [63]. Though the exact mechanism of action remains to be elucidated in many cases, few are known about its antibacterial properties. In fact, it has been shown that seeds of S. nigrum possess good antimicrobial activity against E. coli on solid medium [63]. We report herein for the first time the antibacterial activity of leaves methanol extract of this plant against a panel of MDR Gram-negative bacterial strains with MIC values varying from 128 to 1024 μg/ml (Table 3). Solanum nigrum possesses various compounds that are responsible for diverse activities. Among these compounds, solanine (found in all parts of the plant [58]),is its major defence product [58].
Many reports have also been published about the biological properties of C. pepo, but these reports are based on the components of the fruits and the seed's oil [54,55,57,78]. To the best of our knowledge, were herein report for the first time its activities against MDR bacteria.
The results of the phytochemical test on P. vulgaris are in accordance with some other reports [48,79]. Phaseolus vulgaris was found to inhibit also the growth of Grampositive bacteria B.subtilis [49]. Amarowicz et al. [80] showed that the acetone extract of P. vulgaris contains tannins with good antimicrobial properties against Listeria monocytogenes. Therefore, the low antibacterial effects of this plant as obtained herein (generally MIC values at 1024 μg/ml) ( Table 3) could be due to the multi-drug resistance ability of the studied bacteria.
The antibacterial effects of the extract from C. frutescens against Staphylococcus aureus as well as K. pneumoniae and P. aeruginosa have been reported [67]. The ethanol extract of this plant was also active against MDR strains of S. aureus [81]. The present study therefore provides additional information on the antibacterial potential of this plant on MDR Gram-negative bacteria with MICs ranging from 256 to 024 μg/ml.
The antibacterial properties of S. edule have already been proved against bacteria of clinical relevance by Ordonez et al. [41] which showed that both fluid extract and tincture of fruits have "very good" antimicrobial activities against MDR staphylococci and enterococci [41]. Herein, the antimicrobial activity of the leaves extract {known to possess high level of secondary metabolites and mostly flavonoids [39]} observed against all the studied bacterial strains (Table 3) is being reported for the first time.

Conclusions
The overall results of the present investigation confirmed the traditional uses of the studied vegetables in the treatment of bacterial infections. This study also provide baseline information for the possible use of the methanol extracts of the tested plant samples in the control of infectious diseases involving Gram-negative MDR bacteria. The arising question is of course which are the active compounds responsible for these effects. Our research group is currently focusing on the characterization of these plants extracts in terms of chemical composition and synergistic effects.