Bioactive constituents with antibacterial, resistance modulation, anti-biofilm formation and efflux pump inhibition properties from Aidia genipiflora stem bark

Antimicrobial resistance is a global health challenge. The involvement of bacterial biofilms and efflux pumps in the development of multidrug resistance (MDR) is well established. Medicinal plants have been proposed as alternatives for combating MDR focusing on their bioactive constituents with resistance modulatory activities. This study was aimed at investigating the stem bark of Aidia genipiflora for bioactive constituents with anti-biofilm, efflux pump inhibition and resistance modulatory activities. The crude methanol extract was purified by column chromatography and isolated compounds characterized by mass and nuclear magnetic resonance spectrometry. Antibacterial activity was determined by the High-throughput spot culture growth inhibition and the broth micro-dilution assay. The ethidium bromide accumulation assay was used to determine efflux pump inhibition property. Biofilm inhibition was determined in a microplate crystal violet retention assay. Purification of the ethyl acetate fraction led to the isolation of oleanonic acid (1), 4-hydroxy cinnamic acid docosyl ester (2), β-stigmasterol/β-sitosterol (mixture 3a/b) and D-mannitol (4). The minimum inhibitory concentrations (MICs) ranged from 250 to > 500 μg/mL for extracts and fractions and from 15 to 250 μg/mL for compounds. In the presence of sub-inhibitory concentrations of the compounds, the MIC of amoxicillin against E. coli (20 μg/mL) and P. aeruginosa (320 μg/mL) was reduced by 32 and 10 folds respectively. The whole extract demonstrated anti-biofilm formation and efflux pump inhibition in E. coli, S. aureus and P. aeruginosa. The sterol mixture (3a/b) at concentration of 100 μg/mL caused the highest inhibition (73%) of biofilm formation in S. aureus. Oleanonic acid (1) demonstrated remarkable efflux pump inhibition at MIC of 7.8 μg/mL in E. coli better than the standard drugs verapamil and chlorpromazine. This study confirms the prospects of A. genipiflora as a source of new antibacterial agents and adjuvants that could interact with some resistance mechanisms in bacteria to enhance the activity of hitherto ineffective antibiotics. “A small portion of the study has been presented in a conference in the form of poster”.


Introduction
The increasing frequency of antimicrobial resistance (AMR) has resulted in lower rates of antibiotic efficacy and hence therapeutic failure [1]. Today common infections that were hitherto easy to cure have become increasingly difficult and sometimes impossible to treat resulting in an increased morbidity and mortality from infectious diseases [2]. Coupled with this problem is the depleting pipeline of new antibiotics which has necessitated the search for new potent antimicrobial agents [3].
Among the various mechanisms of antimicrobial resistance, bacteria biofilms and efflux pumps have been shown as main contributors to AMR [4]. The over expression of efflux pumps enable bacteria to extrude broad spectrum of antibiotics to their exterior rendering the drugs ineffective [5]. Biofilm formation is a vital microbial survival strategy through which they exhibit higher resistance than planktonic forms. An enlarged gene pool, more efficient quorum sensing systems, passive resistance and metabolic cooperation are ways by which biofilm forming bacteria protect themselves from antimicrobial agents and host immune responses [6]. Research has shown that efflux pumps are highly active in bacterial biofilms making these two attractive targets for the pharmacological development of new antibacterial agents against resistant pathogens [4].
Medicinal plants are constantly faced with attack from pathogenic microorganisms in the environment, prompting them to produce a wide range of metabolites to protect themselves [7]. Studies have shown that these plant metabolites may be of clinical significance in combatting resistant bacteria due to their highly diversified chemical structures and mechanisms of action [8,9]. The objective of this study was to investigate the crude extract, fractions and some constituents of the stem bark of Aidia genipiflora for antimicrobial, anti-biofilm formation, efflux pump inhibition and resistance modulation activities as part of a continuing effort to identify bioactive constituents of tropical medicinal plants [10,11].
Aidia genipiflora (DC.) Dandy (Rubiaceae) also referred to as Randia genipiflora is distributed in the wild forests of African countries including Ghana, Sierra Leone, Liberia, Nigeria, Ivory Coast, Guinea-Bissau, Sudan and Cameroon. Its leaves and stem bark are used for treating gout, oedema and boils in traditional medicine [12]. A search in literature however revealed little or no scientific investigations on it biological activity and phytochemistry, thus prompting this research.

Plant material collection
The stem bark of A. genipiflora was collected from Kwahu Asakraka, a town in the Eastern Region of Ghana (06°36.704′N/ 000°42.659′W) in November, 2018. The plant material was authenticated by a botanist, Dr. George Henry Sam of the Department of Herbal Medicine, Faculty of Pharmacy and Pharmaceutical Sciences, Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. A voucher specimen (KNUST/HM/2017/SB016) was kept at the herbarium of the Faculty.

Preliminary phytochemical screening
The presence of major secondary metabolites in the powdered stem bark of A. genipiflora was determined by simple qualitative phytochemical screening methods [13].

Preparation of extracts and fractions
The stem bark was washed under running water, chopped into pieces, air dried for a week and pulverized into coarse powder. Three kilograms of the powdered stem bark was extracted by Soxhlet extraction with methanol/chloroform (4:1) for 6 h. The extract obtained was concentrated on a rotary evaporator under reduced pressure and temperature to obtain a brown solid extract subsequently referred to as AG or 'the whole extract' in this report. A yield of 5.1% w / w (153.2 g) was obtained. The whole extract (100 g) was adsorbed onto silica gel 60, packed onto a glass column and eluted successively with petroleum ether (200 mL× 3), ethyl acetate (200 mL× 3) and methanol (200 mL× 3) to afford pet ether (AGPE, 4.3 g), EtOAc (AGEt, 38.8 g) and MeOH (AGM, 53.2 g) fractions. The extract and fractions were kept in a dessicator until required for use.

Isolation and characterization of bioactive constituents
The ethyl acetate fraction (AGEt, 28.3 g) was subjected to purification by column chromatography (CC). CC was performed using silica gel 60 (70-230 mesh). Elution was done using mixtures of pet-ether, EtOAc and MeOH by gradient elution. Thin layer chromatography (TLC) was done using pre-coated silica gel 60 TLC plates (GF254 0.25 mm, Alpha laboratories, UK). Three pure compounds and one phytosterol mixture were isolated ( Fig. 1). Characterization of the compounds was achieved by comparing their 1 H, 13 C NMR and mass spectral data with published data. Details of the isolation procedure are presented in supplementary material (See Additional File 1). A standardized bacteria culture was prepared from overnight cultures (prepared by inoculating sterile nutrient broth with test organism and incubating at 37°C for 24 h). The organisms were then collected and standardized by serial dilution in sterile normal saline to achieve an initial cell count of approximately 1 × 10 5 CFU/mL.

Evaluation of antibacterial activity of crude extracts and fractions
The antibacterial activity of the whole extract and major fractions was tested by the high-throughput spot culture growth inhibition assay (HT-SPOTi) as previously described [14]. Amoxicillin was included as the positive control and 2% DMSO, negative control. The experiment was performed in triplicate.

Determination of the antibacterial activity and antibiotic modulation effect of compounds
The MIC of isolated compounds against two WHO high priority (S. aureus and E. faecalis) and WHO critical priority organisms (P. aeruginosa and E. coli) was   determined by the broth dilution method as previously described by Cos, 2006 [15]. Amoxicillin was included as positive control and 2% DMSO as negative control.
For the modulation test, the MIC of amoxicillin was determined in the absence or presence of the isolated compounds against P. aeruginosa and E. coli using the broth dilution assay as previously described [16]. The modulation factor (MF) was used to express the modulation effect of the compound on the MIC of amoxicillin and was calculated as the ratio of the MIC of the antibiotic alone and the MIC of antibiotic in the presence of compounds.

Biofilm inhibition assay
The effect of the crude extract and isolated compounds on biofilm formation by S. aureus, E. coli and P. aeruginosa was investigated using the microplate crystal violet stain retention method as previously described [17]. As controls, wells containing only medium without test samples were included. The experiment was performed in triplicate.

Ethidium bromide (EtBr) accumulation assay (efflux pump inhibition assay)
The crude extract and isolated compounds were investigated for their ability to enhance the accumulation of EtBr (a substrate for efflux pumps) into bacterial cells according to a previously described method [16]. The known efflux pump inhibitors (EPI) verapamil and chlorpromazine were included as comparative probes and DMSO as the negative control. The experiment was performed in triplicate.

Data management and analysis
All experimental results were analyzed using Graph Pad Prism (Version 5 for windows, San Diego, USA).

Preliminary phytochemical investigation of the stem bark of A. genipiflora
The result of preliminary phytochemical screening performed on the dried powdered stem bark of A. genipiflora revealed the presence of some classes of plant secondary metabolites as presented on Table 1.

Isolated compounds from the stem bark of A. genipiflora
Chromatographic fractionation and purification of the bioactive EtOAc fraction resulted in the isolation of five known compounds identified based on their 1 H and 13 C NMR data. All spectral and physicochemical data obtained for the compounds matched those reported in literature for oleanonic acid (1) [18], 4-hydroxy cinnamic acid docosyl ester (2) [19], β-stigmasterol and βsitosterol (3a/3b; 2:3) [20] and D-mannitol (4) [21] (Fig.  1). These compounds are known to occur in several  Antibacterial activity of A. genipiflora stem bark extract and major fractions The whole extract, MeOH, EtOAc and pet-ether fractions exhibited varying degrees of growth inhibition towards Gam-positive and Gram-negative bacteria in the HTSPOTi assay. For susceptible bacteria, the MIC ranged between 250 and 500 μg/mL and varied with respect to the different bacteria and solvent extract ( Table 2). P. mirabilis was the most susceptible organism being inhibited at 250 μg/mL by most extracts. Amoxicillin had variable inhibitory activities depending on the organisms tested with the highest inhibitory effect (MIC < 10 μg/mL) against the Gram positive bacteria i.e. S. aureus, S. pyogens and E. faecalis. P. aeruginosa was the least susceptible to amoxicillin.
Antimicrobial activity and resistance modulation effect of isolated compounds The antimicrobial activity of the isolated compounds was investigated against S. aureus, P. aeruginosa, E. coli and E. faecalis (WHO high priority and critical priority organisms, WHO, 2019) in the broth dilution assay. The minimum inhibitory concentrations are presented on Table 3. All compounds demonstrated antibacterial activity at MIC range between 15 and 250 μg/mL ( The isolated compounds at sub-inhibitory concentrations (1/4 MIC) were investigated for antibacterial resistance modulatory effect in amoxicillin against E. coli and P. aeruginosa. From the results, all compounds when coadministered with amoxicillin notably potentiated its antibacterial activity against P. aeruginosa and E. coli ( Table 4). The MIC of amoxicillin in the presence of the isolated compounds was reduced from 320 to 31.25 μg/ mL i.e. modulation factor (MF) of 10 for P. aeruginosa. In E. coli, the MIC of amoxicillin was reduced from 20 to 0.625 μg/mL i.e. MF = 32.

Biofilm formation inhibitory effect of A. genipiflora stem bark extract and isolated compounds
The whole extract of A. genipiflora stem bark (AG) was investigated for biofilm formation inhibitory effect against S. aureus, E. coli and P. aeruginosa in the crystal violet retention assay. The extract exhibited remarkable concentration-dependent inhibition of biofilm formation between 15.6-500 μg/mL. The highest anti-biofilm formation effect was expressed against E. coli followed by S. aureus then P. aeruginosa (Fig. 2). The percentage biofilm inhibition against E. coli at its sub-inhibitory concentration (15-125 μg/mL) ranged between 60 and 79%. For S. aureus and P. aeruginosa, the percentage biofilm inhibition ranged between 60 and 76% and 3-57% respectively at sub-inhibitory concentration of 15-500 μg/ mL.
The effect of compounds 1, 2 and 3a/b (3.1-100 μg/ mL) on biofilm formation in S. aureus, E. coli and P. aeruginosa is presented on Fig. 3. The highest inhibition of biofilm formation was given by compound 3a/b against S. aureus (73% inhibition) in a concentration-dependent manner. The percentage biofilm inhibition for the other compounds against the test organisms ranged between 17 and 66% and was not concentration dependent. P. aeruginosa was the least susceptible to all test compounds.

Effect of A. genipiflora stem bark extract and isolated compounds on ethidium bromide (EtBr) accumulation in bacterial cells
The potential of the crude extract (AG) and isolated compounds (1, 2, 3) to act as efflux pump inhibitors was investigated by the EtBr accumulation assay. EtBr, a substrate for several multidrug resistant efflux pumps, emits a strong fluorescent signal when bound to DNA intracellularly and only has a weak signal when present extracellularly. Thus, the activity of putative EPIs can be measured fluorometrically due to the retention of fluorescence overtime if efflux is reduced [22]. Figure 4 shows the EtBr accumulation behaviour in S. aureus, E. coli and P. aeruginosa in the presence of the extract (AG) compared to the action of two standard EPIs, verapamil (VP) and chlorpromazine (CP) measured over 60 mins. From the results it can be inferred that in E. coli, the crude extract (AG) exhibited remarkable efflux pump inhibition resulting in much higher EtBr fluorescence than both standard EPIs (Fig. 4b). However, in S. aureus and P. aeruginosa, the standard EPIs proved much effective in causing higher accumulation of EtBr in the bacterial cells than the test extract (Fig. 4a and c respectively). Figure 5 shows the EtBr accumulation behaviour in E. coli in the presence of isolated compounds. Compound 1 at ½ MIC showed a superior EtBr accumulation than both VP and CP. At ¼ MIC however, its effect was higher than CP but lower than VP (Fig. 5a). For compound 2, EtBr accumulation was higher at ¼ MIC than ½ MIC. VP was superior to compound 2 at all concentrations tested. At ¼ MIC, compound 2 showed a higher effect than CP but lower than VP. At ½ MIC however its effect was similar to that of CP (Fig. 5b). Compound 3 at ½ MIC showed little or no effect on EtBr accumulation activity as its effect was similar to the negative control. At ¼ MIC however, it showed a better potential to accumulate EtBr than CP but lower than VP (Fig. 5c).
The EtBr accumulation enhancement behaviour of compounds 1, 2, and 3 in P. aeruginosa is demonstrated in Fig. 6. Compound 1 at ½ MIC enhanced accumulation of EtBr slightly higher than CP but lower than VP. At ¼ MIC, its effect was lower than both VP and CP (Fig. 6a). Compounds 2 and 3 at the test concentrations showed no obvious increase on the accumulation of EtBr in P. aeruginosa over 60 min (Fig. 6b and c).

Discussion
This study investigated the antimicrobial activity of A. genipiflora stem bark extract (MeOH/CHCl 3 ), three solvent fractions (i.e. pet-ether, EtOAc and MeOH fractions) and some isolated constituents. The effect of the whole extract and isolated compounds on bacteria biofilm formation and efflux-pump inhibition as well as resistance modulation potential of isolated compounds on the antibacterial activity of amoxicillin was also evaluated.
Preliminary phytochemical studies revealed the presence of classes of secondary metabolites ( Table 1) that exhibit a wide range of biological activities including antimicrobial, anti-inflammatory and antioxidant effects [23] supporting the use of the plant in traditional medicine. Other Aidia species were reported to contain coumarins, phenolic acids (catechol) and fatty acid esters [24]. From antibacterial screening of the whole extract and fractions, the MIC ranged from 250 to > 500 μg/mL. In general, the extract and fractions had quite high MICs for most of the organisms tested (MIC ≥500). Nevertheless, the antibacterial activity observed for AG, AGPE and AGEt towards P. mirabilis (MIC = 250 μg/ mL), AG towards E. coli (MIC = 250 μg/mL), AGET towards K. pneumonia (MIC = 250 μg/mL) and AGM towards S. aureus (MIC = 250 μg/mL) are notable ( Table  2). The antimicrobial activity of plant extracts can be classified as significant (MIC < 100 μg/mL), moderate (100 ≤ MIC ≤625 μg/mL) or weak (MIC > 625 μg/mL) [25]. By this criterion, the whole extract and fractions can be said to have demonstrated moderate antibacterial activity. In the genus Aidia, antibacterial activity against B. subtilis, E. coli, P. aeruginosa and S. aureus was reported for the MeOH and aqueous extracts of the leaf of Aidia boorneensis [24].
Detailed phytochemical investigation led to the isolation of oleanonic acid (1) [18], 4-hydroxy cinnamic acid docosyl ester (2) [19], β-stigmasterol and β-sitosterol (3a/3b; 2:1) [20] and D-mannitol (4) [21] (Fig. 1). These compounds are known to occur in several plant species but are being reported for the first time from A. genipiflora. All compounds demonstrated antibacterial activity at MIC range between 15 and 250 μg/mL ( Table 3). The MICs of isolated compounds being lower than the parent fraction implies that fractionation led to more active samples. For pure compounds, antimicrobial activity may be classified as significant (MIC < 10 μg/mL), moderate (10 ≤ MIC ≤100 μg/mL) or weak (MIC > 100 μg/ mL) [25,26]. Though none of the compounds had MIC < 10 μg/mL, the antibacterial activity of compound 1 against all organisms (MIC 15.6 μg/mL) and compounds 2 and 3a/b against S. aureus and E. faecalis (MIC = 31.25 μg/mL) are remarkable. In previous studies, oleanonic acid (1) demonstrated significant antibacterial activity against pathogenic Gram positive and Gram negative bacteria at MIC range between 10 and 50 μg/ mL [27] and against methicillin resistant S. aureus (MRSA) at 10 μg/mL [28]. The anti-tubercular activity of oleanonic acid (1) against Mycobacterium tuberculosis was previously reported [18]. β-stigmasterol and βsitosterol (3a/b) have demonstrated broad spectrum antibacterial activity against several pathogenic bacteria in previous studies [29][30][31][32]. This is the first report on the antibacterial activity of 4-hydroxy cinnamic acid docosyl ester (2) though some cinnamic acid derivatives have been shown to possess very potent antibacterial activity [33]. The presence of these compounds in A. genipiflora may thus contribute to its antibacterial activity. In modulatory tests, the MIC of amoxicillin in the presence of the isolated compounds was reduced by 10 folds for P. aeruginosa and 32 folds in E. coli. This suggests that the isolated compounds have the ability to potentiate the antibacterial effect of amoxicillin against these Gram negative bacteria. In previous studies, oleanolic acid, the parent derivative of compound 1 was found to synergistically potentiate the antibacterial activity of ampicillin and oxacillin towards S. aureus, P. aeruginosa, L. monocytogenes, S. epidermidis [34]. A combination of stigmasterol (3a) and ampicillin resulted in a significant increase in the effect of ampicillin against clinical isolates of S. aureus, E. coli, P. aeruginosa and S. pyogens [35]. The synergistic action of plant-derived compounds and antibiotics has been proposed to be by the interaction of these agents with bacterial resistance mechanisms [36] such as destruction of the bacterial membrane structure, increasing the influx of antibiotics into the bacteria cell [37] inhibition of bacterial efflux pumps, inhibition of quorum sensing and some gene expression modulations [22,38].
The whole extract and isolated compounds further showed remarkable inhibition of biofilm formation in S. aureus, E. coli and P. aeruginosa. Several plant extracts and plant derived compounds have been shown to prevent the adhesion or implantation of planktonic forms of bacterial cells on abiotic surfaces thus inhibiting the formation of biofilms [39][40][41]. Though this is the first report on the anti-biofilm formation potential of A. genipiflora and its constituents, previous studies have reported the biofilm inhibitory activity of similar bioactive constituents. Cinnamic acid derivatives of natural origin exhibited significant quorum sensing and biofilm inhibition in Chromobacerim violaceum increasing its susceptibility to tobramycin [42]. Pentacyclic triterpenes and some sterols including βsitosterol (3b) also demonstrated remarkable anti-biofilm activities [43,44]. Extract and plant derived compounds may inhibit biofilm formation by damaging microbial membrane structures, inhibiting peptidoglycan synthesis [45], inhibition of nucleic acid synthesis [46], quorum sensing inhibition or disruption [47] and anti-cell-adhesion properties [48].
The whole extract demonstrated potential to inhibit efflux pumps in E. coli and to a lesser extent in S. aureus and P. aeruginosa. For pure compounds as well, efflux ump inhibition was higher in E. coli than P. aeruginosa. It was also interesting to note that lower concentrations (1/4 MIC) of compounds 2 and 3 exhibited far greater EPI activity in E. coli than more concentrated samples (1/2 MIC) (Figs. 5 and 6). This observation is consistent with a previous reports on the EPI activity of some Berberis spp [49] which was attributed to the possibility of putative EPIs binding at only high affinity sites of the efflux pump causing greater inhibition at only low concentrations whereas higher concentrations would result in low affinity binding sites being occupied.
The antibacterial activity demonstrated by A. genipiflora and its constituents in this study has revealed the prospects of this plant as a potential source of new antibacterial agents and adjuvants that could possibly interact with some resistance mechanisms in bacteria to enhance the activity of antibiotics that have been rendered inactive by resistant organisms.

Conclusion
This study has provided the first evidence of the antimicrobial, resistance modulation, anti-biofilm formation and efflux pump inhibitory potentials of A. genipiflora stem bark and its constituents. This gives scientific credence to the traditional uses of the stem bark of A. genipiflora for managing infections. Although oleanonic acid, β-stigmasterol, β-sitosterol, 4-hydroxy cinnamic acid docosyl ester and D-mannitol are known plant bioactive compounds, this is the first report of their isolation from A. genipiflora as well as the genus, Aidia.
Additional file 1. Methods of isolation of compounds and spectroscopic data of isolated compounds from Aidia genipiflora stem bark.