Antimicrobial activity of Albizia tulearensis, an endemic Fabaceae from Madagascar

The antimicrobial activity of methanolic extract, butanolic and saponosidic fractions of different organs (leaf, seed, stem bark, root bark) from Albizia tulearensis were tested against 14 pathogenic germs including 5 bacteria Gram-positive (Bacillus cereus, Listeria monocytogenes, Staphylococcus aureus, Streptococcus pneumoniae, Streptococcus pyogenes), 6 Gram-negative (Clostridium perfringens, Enterobacter aerogenes, Escherichia coli, Pseudomonas aeruginosa, Salmonella enterica, Yersinia entericolitica) and 3 yeasts (Candida albicans, Candida tropicalis, Cryptococcus neoformans) using disc diffusion and microdilution methods. At the concentration of 1 mg/disc, the methanolic extracts and fractions of seeds, stem and root bark exhibited selective antimicrobial activity with Inhibition Zone Diameters ranking from 8.67 to 16.5 mm. When using microdilution method, all the methanolic extracts and fractions displayed higher antimicrobial activities while saponosidic fractions were by far the most efficient. Saponosides from leaves (Lsap), seeds (Ssap), stem bark (Bsap) and root bark (Rsap) also exhibited excellent effects with Minimum Inhibitory Concentrations lower than 100 μg/mL against 71.4 %, 21.4 %, 7.14 % and 64.3 % of the germs tested respectively. The most sensitive germs were: S. aureus (4.87 μg/mL) with Rsap and S. pyogenes (9.75 μg/mL) with Lsap in Gram-positive bacteria; E. coli and Y. entericolitica (19.5 μg/mL) with Lsap in Gram-negative bacteria and C. albicans (4.87, 19 and 19.5 μg/mL) with Lsap, Ssap and Rsap respectively in yeasts. The saponosidic fractions had bactericidal and fungicidal effects against the vast majority of the microorganisms tested. The phytochemical screening carried out on plant organs powders revealed the presence of desoxyoses, saponosides, triterpens, unsaturated sterols and phenolic compounds.


Introduction
Antibiotic resistance is currently a major public health problem. Inappropriate and excessive use of synthetic antibiotics have promoted the resistance of pathogenic microorganisms to drugs, making therapeutic treatments less effective [1]. Facing this scourge, many research groups worldwide are screening for plant extracts to detect new active molecules that can eliminate infectious germs without detrimental side effects. Secondary plant metabolites represent a considerable alternative for the exploration of these new healing molecules [2].
In the present study we investigated the antimicrobial activities of the methanolic extracts, butanolic and saponosidic fractions from different organs of Albizia tulearensis, another Malagasy species which has not been studied so far.

Plant and microbial materials
Albizia tulearensis is a tree which is found throughout the southern part of Madagascar and can grow up to 15   The microorganisms used in this study are presented in Table 1. They included 6 Gram-negatives (-) and 5 Grampositives (+) bacteria, as well as 3 pathogenic yeasts.

Preparation of methanolic extracts
The air-dried plant organs were ground into powder which was delipidated with hexane (1/10 w/v) for 24 h under stirring at room temperature. After settling and filtration of the supernatant through Whatmann No. 1 filter paper, the filtrate was set aside while the non-soluble residues were recovered and then treated several times with the same solvent, until the extraction solvent was no longer colored. After drying, they were extracted with methanol under the same conditions. The filtrates were combined and concentrated to dryness by evaporation under reduced pressure to give a residue called methanolic extract.

Preparation of n-butanolic fractions
Each methanolic extract brought into solution in distilled water (150 mL) and an equal volume of n-butanol were vigorously shaken in a separatory funnel. After decanting, the butanolic phase was recovered and the same treatment was repeated twice with the aqueous phase. The butanolic fractions were collected and evaporated to dryness to give a residue called butanolic fraction.

Preparation of saponosidic fractions
Each butanolic fraction was first dissolved in methanol. In an ice bath, a mixture of acetone and diethyl ether (v/v) were gradually added to the resulting solution until the precipitate appeared. After a few minutes of maceration, the resulting suspension was centrifuged (1000 rpm during 10 minutes at +4 °C). The supernatant was treated several times using the same process until there was no more precipitation. All the collected pellets were dissolved in distilled water, and then evaporated to dryness under reduced pressure to give a residue called saponosidic fraction.
The methanolic extracts and the butanolic and saponosidic fractions generated from the different plant organs used in this work are summarized in Table 2.

Phytochemical screening
The reactions of chemical group detection were those developed by Firdouse and Alam [17].

Antimicrobial assays
For all the antimicrobial assays, the methanolic extracts and the butanolic and saponosidic fractions were used as aqueous solutions.
The disc diffusion method described by Pyun and Shin [18], Ngameni et al. [19] and Favel et al. [20] was used to test the antimicrobial activity. Each methanolic extract, and butanolic and saponosidic fraction was prepared by dilution of the stock solutions (100 mg/mL) using distilled water. Dried and sterilized filter paper discs (6 mm diameter from Bio-Rad) were then impregnated with 10 µl/disc of the test substances. The impregnated discs were then carefully deposited on top of the nutrient agar medium, previously seeded uniformly with the pathogenic test microorganisms Standard antibiotic discs (Amoxicillin 25 µg/disc, chloramphenicol 30 µg/disc, penicillin 6 µg/disc, miconazole 50 µg/disc) from Liofilchem (Italy) and blank discs impregnated with distilled water were used as positive and negative controls. The plates were then incubated at 37 °C for bacteria and 30 °C for yeasts to allow growth of the cells. Diffusion of the aqueous test solution generated a gradual change in concentration in the media surrounding discs. A clear and distinct zone of microbial growth inhibition was observed for the test solutions having antibacterial activity. This activity was determined by measuring the Inhibition Zone Diameters (IZD) expressed in millimeters. The results were interpreted using the scale of Ponce et al. [21] and Celikel et al. [22]: meaning that bacteria are were considered as not sensitive for an inhibition zone diameter IZD ≤ 8 mm, sensitive for IZD= 9-14 mm, very sensitive for IZD= 15-19 mm and extremely sensitive for IZD > 20 mm.

MIC, MBC and MFC determination
The Minimum Inhibitory Concentration (MIC), Minimum Bactericidal Concentration (MBC) and Minimum Fungicidal Concentration (MFC) were determined using microdilution method according to Kuete et al. [23]. The initial concentration of each extract was set at 10 mg/mL. This was serially diluted two-fold to obtain concentration ranges from 0.0048 to 10 mg/mL. Each concentration was added in a well (96-well microplate) containing 95 μl of Mueller-Hinton broth (MHB) and 5 μl of microbial inoculum (standardized at 0.5 MacFarland). A positive control containing bacterial culture without extract and a negative control containing only the medium were also analyzed. The microtiter plates were then covered and incubated at 37 °C (bacteria) for 24 h or at 30 °C (yeasts and molds) for 48 h. The MIC of each extract was detected using p-iodonitrotetrazolium chloride colorimetric test which consisted in addition of 40 µl of 0.2 mg/ml mL of p-iodonitrotetrazolium chloride and incubation at appropriate temperature depending on microorganism as discussed above. Viable bacteria generated a change in the medium color from yellow to pink. MIC was defined as the lowest sample concentration that prevented this change and exhibited complete inhibition of bacterial growth.

Statistical analysis
The results were expressed as average values ± standard deviations from three separate determinations. One-way variance analysis (ANOVA) with XLSTAT 2014 software was used for statistical analysis. Statistical estimates were made at the 95 % confidence interval.

Extraction yields and qualitative phytochemical analysis of the fractions
The different extracts from leaf, seed, bark and root of A. tulearensis plant were obtained with yields ranging from 1.7 % for Bbut to 15.6 % for Smet (Table 3). The phytochemical analysis of the different organs powder from A. tulearensis is presented in Table 4.
The results revealed the presence of saponins, polyphenols, desoxyoses, steroids and triterpens and the absence of alkaloids, flavonoids, leucoanthocyanins and anthraquinones.

Antimicrobial activity
The antimicrobial effects of the various extracts normalized at a concentration of 1 mg per disc were investigated against a selection of food pathogenic microorganisms (Table 5).
Except for all the leaf methanolic extract, butanolic and saponosidic fractions that were inactive against all microorganisms tested, the methanolic extracts, butanolic and saponosidic fractions of the three other organs (namely seeds, stem bark and root bark) exhibited a selective antimicrobial activity. The IZD values ranged from 8.67 to 14.5 mm for seed extracts, from 9 to 14 mm for stem bark extracts and from 10 to 16.5 mm for root bark extracts. In most cases, the saponosidic fractions were more efficient than methanolic extracts and butanolic fractions. Bsap impacted the growth of the highest number of microorganisms meaning 11 out of 14 germs with IZD ranking from 9 to 14 mm. Rsap was the most active against all sensitive germs tested with IZD ranking from 12 mm against C. tropicalis to 16.5 mm against E. coli. The yeast C. albicans was sensitive to all the methanolic extracts and fractions of seeds, stem bark and root bark with IZD ranking from 10.3 to 14.5 mm. However, at 1 mg/disc all these methanolic extracts and fractions were less efficient than the reference antibiotics used as positive controls.
In addition, all these methanolic extracts and fractions exhibited broad antimicrobial activity. The germs Y. entericolitica and C. neoformans were resistant to all the plant extracts. The antimicrobial activity was precisely assessed using microdilution method and the results are shown in Tables 6 -8.
In accordance with the results obtained by the disc diffusion method, the saponosidic fractions from the different plant organs had proved to be the most efficient against microbial proliferation.

Discussion
Our results revealed that the A. tulearensis organs contained saponins, polyphenols and triterpens which are known for their antimicrobial properties. Unlike other Malagasy Albizia such as A. polyhylla [15], A. bernieri [13] and A. mahalao) [16], A. tulearensis did not contain alkaloids.
The overall results of the present work provided evidence that the different organs from A. tulearensis possess antibacterial and antifungal properties.
As already noted in the case of extracts from other Albizia species namely A. bernieri [13] and A. mahalao [16], some results from the 2 methods used (Disc diffusion and microdilution methods) were different. For example, Lsap was inactive (IZD < 8 mm) against all the germs tested in solid medium but displayed excellent inhibitory effects (MIC<100 µg/mL) against the majority of strains in liquid medium. We think that this might be to the fact that some bioactive compounds diffused poorly in solid medium whereas in liquid medium as compared to liquid medium where cells are in direct contact with germs. This might also suggest that the molecules involved in the antimicrobial activity were not necessarily the same in the different A. tulearensis parts.
Lmet and Bmet were respectively the highest and the lowest active methanolic extracts against all the germs tested. In comparison with available comparable data on the antimicrobial activities of other Malagasy Albizia we previously studied (Table 9), Lmet (MIC = 156.2 µg/mL) and Smet (MIC = 625 µg/mL) were more efficient than respectively the leaf extract from A. masikororum and A. divaricata. In addition, the saponosidic fractions from seeds (Ssap) were much more effective than those from A. bernieri against Y. entericolitica, B. cereus and C. albicans where the MIC values of Ssap were respectively 78.1, 78.1 and 19.5 µg/mL. Seed extract 2420 [14] Comparison of the methanolic extracts and fractions from A. tulearensis with Albizia species from other countries was not easy because antimicrobial activity was assessed under different conditions namely the plant part, the extraction solvent, the assessment method etc. For illustrative purposes only, we give IZD values of extracts from different Albizia species against the same microorganisms against which A. tulearensis were also tested (Table 10). For instance, A. tulearensis extracts displayed similar activities than A. ferruginea and A. odoratissima against S. aureus but was generally less active against C. albicans. A. zigia and A. lebbeck against P. aeruginosa and E. coli were also more efficient than A. tulearensis axtracts.  Saponosides which were the main secondary metabolites found in all organs extracts and fractions of A. tulearensis are also suspected to be the main responsible for their antimicrobial activity. It is worth noting that saponosides were the main compounds present in numerous Albizia species [32,33] and showed many other pharmacological properties as antioxidant, antidiabetic, anthelmentic, antibacterial, hepato protective, anti-inflammatory and cytotoxic. Our results showed that most of the MBC/MIC ratios observed were lower than 4, highlighting the bactericidal and fungicidal properties of the methanolic extracts, butanolic and saponosidic fractions from A. tulearensis organs. This could probably be attributed to the direct action of saponosides on the cytoplasmic membrane resulting in microorganisms cell lysis and death [16].

Conclusion
This study demonstrated the antimicrobial potential of Albizia tulearensis. This plant might be useful in the development of new drugs against many microorganisms and also help address the problems of access to medicines in developing countries. However, further toxicological and pharmacological studies are still required to confirm these hypotheses. A further extensive chemical and biological study of A. tulearensis secondary metabolites will be necessary to allow determining the structure, the number, the originality, the distribution of the active principles in the different plant parts and their possible role in antimicrobial activity. In addition, these results constitute additional useful informations on the potentials of Malagasy Albizia.