Tamarindus indica Extract as a Promising Antimicrobial and Antivirulence Therapy

The worldwide crises from multi-drug-resistant (MDR) bacterial infections are pushing us to search for new alternative therapies. The renewed interest in medicinal plants has gained the attention of our research group. Tamarindus indica L. (T. indica) is one of the traditional medicines used for a wide range of diseases. Therefore, we evaluated the antimicrobial activities of ethanolic extract of T. indica. The inhibitions zones, minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC), and fractional inhibitor concentration indices (FICI) against Gram+ve and −ve pathogens were detected. The bioactive compounds from T. indica extract were identified by mass spectroscopy, thin-layer chromatography, and bio-autographic assay. We performed scanning electron microscopy (SEM) and molecular docking studies to confirm possible mechanisms of actions and antivirulence activities, respectively. We found more promising antimicrobial activities against MDR pathogens with MIC and MBC values for Staphylococcus aureus (S. aureus) and Pseudomonas aeruginosa (P. aeruginosa), i.e., (0.78, 3.12 mg/mL) and (1.56, 3.12 mg/mL), respectively. The antimicrobial activities of this extract were attributed to its capability to impair cell membrane permeability, inducing bacterial cell lysis, which was confirmed by the morphological changes observed under SEM. The synergistic interactions between this extract and commonly used antibiotics were confirmed (FICI values < 0.5). The bioactive compounds of this extract were bis (2-ethylhexyl)phthalate, phenol, 2,4-bis(1,1-dimethylethyl), 1,2-benzenedicarboxylic acid, and bis(8-methylnonyl) ester. Additionally, this extract showed antivirulence activities, especially against the S. aureus protease and P. aeruginosa elastase. In conclusion, we hope that pharmaceutical companies can utilize our findings to produce a new formulation of T. indica ethanolic extract with other antibiotics.


The Possible Antimicrobial Mechanisms of Actions
The leakage of K + ions in response to the T. indica ethanolic extract at MIC concentration was measured by an atomic absorption spectrophotometer. T. indica ethanolic extract caused a rapid increase in ion leakage in both S. aureus (Gram+ve) ( Figure 1A) and P. aeruginosa (Gram−ve) ( Figure 1B) during the first 60 min after exposure. This was still increasing after 60 min, but at a low rate. The nucleotide leakage in both S. aureus ( Figure 1C) and P. aeruginosa ( Figure 1D) gradually increased upon treatment with the MIC concentration. Thus, Gram+ve and Gram−ve bacteria possess a strong sensitivity to T. indica extract, which causes great cell membrane damage.
Moreover, SEM was performed to observe the morphological effects of T. indica on the MDR P. aeruginosa and S. aureus isolates. In contrast to untreated P. aeruginosa cells, which displayed typical bacilliform with uniformity in size and distribution (Figure 2A), cells treated with the extract had irregular, withered, coarse surfaces; lysis of cell membranes; and leakage of cellular contents, forming aggregations and adhesions ( Figure 2B). Similar alterations were observed in cells of S. aureus treated with the same extract ( Figure 2C,D).

Phytochemical Analysis of the Bioactive Molecules
Phytochemical analysis revealed that T. indica ethanolic extract includes phenols, flavonoids, alkaloids, quinones, tannins, saponins, and terpenoids. Paper TLC was used to purify active antibacterial compounds, which was followed by a bio-autography assay. The dried fraction of T. indica ethanolic extract with an Rf value of 0.4, among all other ethanolic extract fractions, demonstrated antibacterial activity against P. aeruginosa and S. aureus. This fraction was subjected to UV, IR, and MS scanning. From the UV profile, two major peaks were evidenced at λ = 264.5 and 214.5 nm ( Figure 3A). In addition, the IR spectrum ( Figure 3B  Moreover, SEM was performed to observe the morphological effects of T. indica on the MDR P. aeruginosa and S. aureus isolates. In contrast to untreated P. aeruginosa cells, which displayed typical bacilliform with uniformity in size and distribution (Figure 2A

Phytochemical Analysis of the Bioactive Molecules
Phytochemical analysis revealed that T. indica ethanolic extract includes phenols, flavonoids, alkaloids, quinones, tannins, saponins, and terpenoids. Paper TLC was used to purify active antibacterial compounds, which was followed by a bio-autography assay. The dried fraction of T. indica ethanolic extract with an Rf value of 0.4, among all other

GC-MC Analysis
According to mass spectroscopy analysis, the chemical composition of purified ethanolic extract from T. indica contained the following molecules: (i) bis (2-ethylhexyl) phthalate (C24H38O4) with an MW of 390, 277 (56.5%), as illustrated in Figure

GC-MC Analysis
According to mass spectroscopy analysis, the chemical composition of purified ethanolic extract from T. indica contained the following molecules:

Molecular Docking Studies
From all the virulence proteins expressed by S. aureus and P. aeruginosa which were tested by the molecular docking studies, only the S. aureus protease and P. aeruginosa elastase could bind with bioactive compounds of T. indica. Regarding S. aureus, it was discovered that the investigated compounds (Bis (2-ethylhexyl) phthalate (1), phenol, 2,4-bis (1,1-dimethylethyl) (2), and 1,2-benzenedicarboxylic acid, bis(8-methylnonyl) ester (3) could interact with the S. aureus protease through pi-H bond with PRO153, at binding scores of −6.52, −4.46, and −7.36 Kcal/mol and with RMSD values of 1.74, 1.52, and 1.58 Å, respectively. It is worth noting that the benzene ring was responsible for pi-H bond formation with PRO153 for the three investigated compounds, as shown in Table 1 and Figure 5. However, regarding P. aeruginosa, it was shown that compound 1 could interact with P. aeruginosa elastase at a binding score of −7.47 Kcal/mol, and the value of RMSD was 2.38 Å. Obviously, the carboxylate group at the phthalate moiety in compound 1 had the ability to form H-bond with TYR155 at 2.85 Å. The terminal side chain of compound 1 could form two H-pi bonds with HIS140, at distances of 3.70 and 4.40 Å. In addition, compound 2 exhibited a binding score of −4.74 Kcal/mol, with an RMSD value of 0.98 Å. Notably, the tert-butyl moiety in compound 2 was able to form an H-pi bond with HIS140 at a distance of 4.25 Å. Furthermore, compound 3 displayed a binding score equal to −7.45 Kcal/mol and an RMSD value equal to 1.90 Å. Notably, the phenyl ring of compound 3 was able to form a pi-H bond with ASN112 at a distance of 4.79 Å. The terminal side chain of compound 3 was able to form an H-pi bond with HIS140 at a distance of 4.43 Å. Compounds 1, 2, and 3 were able to form metal bonds with GLU141 at a distance of 1.72 Å, as illustrated in Table 1 and Figure 6.

Molecular Docking Studies
From all the virulence proteins expressed by S. aureus and P. aeruginosa which were tested by the molecular docking studies, only the S. aureus protease and P. aeruginosa elastase could bind with bioactive compounds of T. indica. Regarding S. aureus, it was discovered that the investigated compounds (Bis (2-ethylhexyl) phthalate (1), phenol, 2,4-bis (1,1-dimethylethyl) (2), and 1,2-benzenedicarboxylic acid, bis(8-methylnonyl) ester (3) could interact with the S. aureus protease through pi-H bond with PRO153, at binding scores of −6.52, −4.46, and −7.36 Kcal/mol and with RMSD values of 1.74, 1.52, and 1.58 Å, respectively. It is worth noting that the benzene ring was responsible for pi-H bond formation with PRO153 for the three investigated compounds, as shown in Table 1 and Figure 5. However, regarding P. aeruginosa, it was shown that compound 1 could interact with P. aeruginosa elastase at a binding score of −7.47 Kcal/mol, and the value of RMSD was 2.38 Å. Obviously, the carboxylate group at the phthalate moiety in compound 1 had the ability to form H-bond with TYR155 at 2.85 Å. The terminal side chain of compound 1 could form two H-pi bonds with HIS140, at distances of 3.70 and 4.40 Å. In addition, compound 2 exhibited a binding score of −4.74 Kcal/mol, with an RMSD value of 0.98 Å. Notably, the tert-butyl moiety in compound 2 was able to form an H-pi bond with HIS140 at a distance of 4.25 Å. Furthermore, compound 3 displayed a binding score equal to −7.45 Kcal/mol and an RMSD value equal to 1.90 Å. Notably, the phenyl ring of compound 3 was able to form a pi-H bond with ASN112 at a distance of 4.79 Å. The terminal side chain of compound 3 was able to form an H-pi bond with HIS140 at a distance of 4.43 Å. Compounds 1, 2, and 3 were able to form metal bonds with GLU141 at a distance of 1.72 Å, as illustrated in Table 1 and Figure 6.

Discussion
The wide spread of MDR bacterial and fungal pathogens has created several health problems [15,16], especially throughout those countries that did not follow up on the infection control guidance. The discovery and development of antimicrobial therapies are normally conducted with very slow steps, which is not in line with the rate of evolution of antimicrobial resistance mechanisms to commonly used antibiotics. In the same context, the therapeutic switching of already-used medicine [17] and the renewed interest in medicinal plants [18] may compensate for the wide gap in solutions for this issue. Therefore, the use of complementary and alternative medicines, especially natural compounds and essential oils, with certain precautions, are the perfect choice to prevent the compounding of this crisis. The use of medicinal plants must occur under full medical supervision, without any self-medication, to avoid drug interactions, in addition to other adverse effects [19].
T. indica extracts from various plant parts have been used for several therapeutic purposes [20]. In this study, T. indica ethanolic extract was selected, and its antimicrobial and antivirulence activities were evaluated against Gram+ve and Gram−ve resistant pathogens. Generally, a broad spectrum of antibacterial activity with low MIC and MBC values was observed for the tested ethanolic extract compared to ordinary aqueous extract [20]. In this study, the promising use of ethanolic extract from T. indica as an alternative and complementary therapy for resistant pathogens was confirmed by the large zones of inhibitions. Additionally, the MIC and MBC values were detected for S. aureus (0.78, 3.12 mg/mL) and P. aeruginosa (1.56, 3.12 mg/mL), respectively. Parallel to our finding, several authors reported the antimicrobial activities of ethanolic extract of T. indica through MIC and MBC values, thus confirming our hypothesis [20][21][22]. Therefore, the success of our postulates regarding the antimicrobial activities of T. indica makes reconsidering the use of other medicinal plants an urgent necessity.
In fact, the in vitro antimicrobial potential of these natural compounds did not reflect the overall bacterial response in vivo, since it was tested in broth rather than in a physiological human body, in addition to the bioavailability problems [23]. Therefore, we cannot suggest the use of T. indica extract as the sole drug for treating resistant pathogens. In the same context, resistance to commonly used antibiotics such as imipenem, amikacin, and ofloxacin were previously reported. For that, we suggest the use of a combination of any of these antibiotics and T. indica extract. Synergistic interactions between these combinations were detected (FICI > 0.5) against both Gram+ve and Gram−ve bacteria. Confirming our finding, the co-admixing of antibiotics with natural compounds and/or essential oils had huge success in treating MDR bacterial and fungal infections, in contrast to the use of each one alone [24][25][26]. Furthermore, the use of medicinal plants can reduce the duration of use, dose, and toxicity hazards associated with antibiotics, and decrease the possibility of the emergence of new resistant strains [27].
It has also been reported that a huge number of bioactive compounds were found in various parts of medicinal plants. These bioactive compounds were diverse in their chemical structure, and their concentration was not the same in each part of the medicinal plants. Therefore, it is essential to determine the exact bioactive compounds of medicinal plant extract. In this report, GC-mass and other spectrophotometer analyses of the ethanolic extract of T. indica revealed several phytochemicals, including phenolic content, flavonoids, alkaloids, quinones, tannins, saponins, and terpenoids. This finding was in agreement with other T. indica-related phytochemical studies [28]. The antimicrobial activity of this extract may be attributable to its phenolic compounds [28]. Chemical analysis of the extract revealed three major compounds: (i) Bis . The DEHP, a major bioactive compound in this extract, showed a broad spectrum of antibacterial activity against both G+ve and G−ve bacteria compared to other secondary metabolites [29]. The amount of these phenolic compounds, which are present in almost every part of this medicinal plant, varied according to the extraction method, geographical location, and climatic conditions [30].
The novelty of this study is the determination of antimicrobial mechanisms by various methods, in addition to the assessment of the antivirulence activity of T. indica extract by molecular docking. In this study, the microbial cytoplasmic membrane is the main target site of the bioactive compounds of this extract. Similar studies documented the mechanism of action via inhibition of protein and DNA synthesis, increasing cell membrane and wall permeability as well as lysing the cells [31,32]. The results obtained in this study revealed that T. indica ethanolic extract caused a rapid increase in ion leakage, especially of K + ions, and nucleotides in both S. aureus (Gram+ve) and P. aeruginosa (Gram−ve); this was confirmed by an atomic absorption spectrophotometer. Additionally, the treated isolates showed irregular, withered, and coarse surfaces; lysis of the cell membrane; and leakage of cellular contents, forming aggregations and adhesions under SEM. Parallel to our findings, it was confirmed that the phenolic compounds acted on the bacterial cytoplasmic membrane as the essential intercellular materials of the treated pathogens, such as nucleic acids and other ions, released into the extracellular solution by cellular leakage [33], and similar observations have been documented by other studies [31]. Furthermore, the antivirulence activities of the T. indica ethanolic extract were assessed by a molecular docking study. All of the bioactive compounds showed good binding capacities with the S. aureus protease and P. aeruginosa elastase. The measurement of binding scores, RMSD values, and amino acid interactions of the investigated compounds of the tested extract with the S. aureus proteases and P. aeruginosa elastase confirmed these antivirulence activities. The DEHP affected the intercellular communication in the bacteria and resulted in a significant reduction in biofilm, extracellular polysaccharide, prodigiosin, lipase, haemolysin, and protease, thus increasing the susceptibility of bacteria to conventional antibiotics when administered synergistically [29]. In addition, 1,2-benzenedicarboxylic acid, bis (8-methylnonyl) ester is one of putative compounds found in many plants, and is known to have antimicrobial activity [34].

Microorganisms, Plant Materials, and Extraction
The MDR bacterial isolates which were used in this study, such as S. aureus ATCC25923 and Pseudomonas aeuginosa ATCC 27853 were kindly provided from the microbiological units of Zagazig University Hospitals. Furthermore, these isolates were confirmed by molecular detection of specific 16S RNA genes using the previously described primers. Additionally, the MDR patterns for these isolates were confirmed by the Kirby Bauer Disc Diffusion Method according to CLSI, 2020 [35].
Tamarind (Tamarindus indica L.) is a medicinal plant used commonly in Egyptian folk medicine. Fresh fruit of T. indica, which was planted in Southern Egypt (Aswan city), was purchased from a local supplier in Zagazig City, Egypt. This fresh fruit was used to prepare the antimicrobial ethanolic extract used in the investigation as follows. A 50-g sample of dry fruit powder was added to 500 mL of methanol 80% and continuously shaken for 48 h at room temperature. The ethanolic solution was then centrifuged at 5000 rpm for 10 min and filtered through 1 layer of Whatman No. 1 filter paper. The supernatant was evaporated using a rotary vacuum evaporator under 34-36 kPa pressure at 45 • C. The pellet was dissolved in distilled water containing 2% dimethylsulfoxide to form stock solutions with 25 mg/mL concentration [36].

Agar Diffusion Assay by Filter Paper Disc Method
The antibacterial activity of ethanolic Tamarindus indica extract was evaluated in triplicate by the disc diffusion method [37]. Pure bacterial isolates were sub-cultured in Muller-Hinton agar medium at 37 • C for 4 h. The density of the bacterial suspension was adjusted to 10 6 CFU/mL, equivalent to standard barium sulfate (0.5 McFarland). Then, 3 layers of sterile filter paper discs (Whatman No. 3, 6 mm diameter) were saturated with the fruit extract, left to dry for 1 h, and then placed on the surface of the agar plate and incubated for 24 h at 37 • C. Antibacterial activity was evaluated by measuring the entire diameter of the inhibition zone in mm.

Estimation of Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC)
MIC is the lowest concentration that inhibits visible bacterial growth in liquid media, whereas MBC is the lowest concentration at which no growth occurs in solid media. The MICs and MBCs of the ethanolic extract from T. indica were determined in triplicate by the broth microdilution method [38]. To obtain the appropriate suspensions needed for each experiment, the stock solution of tamarind extract (25 mg/mL) was diluted in nutrient broth to obtain twofold serial dilutions ranging from 0.195 to 12.5 mg/mL. The bacterial broth suspensions were prepared at 10 8 CFU/mL from overnight cultures. Each experiment used positive and negative controls, the first consisting of tubes containing a bacterial suspension and nutrient broth, and the second of tubes containing the extract from T. indica and nutrient broth. All tubes were incubated for 24 h at 37 • C and examined for turbidity (λ = 600 nm) to detect the MIC value. Regarding the determination of MBC, 100 µL of MIC concentration and the 2 highest concentrations were introduced onto the nutrient agar plate and incubated at 37 • C for 24 h to determine their MBC values.

Evaluation of Co-Admixture of the Ethanolic Extract from T. indica with the Commonly Used Antibiotics by Checkerboard Method
The degree of interaction between the tested extract and other antibiotics (imipenem, amikacin, and ofloxacin) was assessed by determination of the fractional inhibitory concentration (FICI) values in triplicate according to [39,40]. The MIC values of the tested extract alone and in the presence of other antibiotics were detected. Additionally, the MIC values of the antibiotics alone and in the presence of tested extract were measured. The FICI values were detected according to the following equation: FICI = (MIC of tested extract in combination/MIC of tested extract alone) + (MIC of antibiotic in combination/ MIC of antibiotic alone). The synergistic and antagonistic interactions were obtained for FICI ≤ 0.5 and FICI ≥ 4, respectively. On the other hand, additive and indifference effects were expected when 0.5 < FICI ≤ 1 and 1 < FICI < 4, respectively.

Assessment the Possible Antibacterial Mechanisms of T. indica Ethanolic Extract
We assessed the cell membrane integrity by measuring both the K + level and nucleotide leakage. The K + leakage was determined following the method reported previously [41], with minor modifications. Briefly, bacteria were allowed to grow overnight in nutrient broth in a shaking incubator at 37 • C. Then, normal saline was used to wash the cells three times, and then cells were resuspended in 1 mmol/L glycylglycine (Sigma, USA) buffer solution with a pH value of 6.8 [42]. Bacteria were treated with the studied extract at the detected MIC, and incubated in a shaking incubator at 37 • C. After that, we took the bacterial cell suspensions after 0, 10, 20, 40, 60, 80, 100, and 120 min and filtered them through a membrane (0.22 µm pore-size membrane, Sartorius, Gottingen, Germany) to remove any bacteria. We determined the K + concentration in the supernatant by applying an atomic absorption spectrophotometer (900T, Perkin-Elmer Ltd., Beaconsfield, UK) at λ = 766.5 nm. In reference to previously established standard K+ solutions, the absorbance was converted to K + concentration (ppm). The experiments were conducted in triplicate, and the obtained data value averages are reported herein.
The bacterial nucleotide leakage was measured upon treatment with the studied extract [43]. After incubation of the bacterial suspensions with MIC concentrations of the extracts at 37 • C and 150 rpm, samples were taken after 1, 2, 4, 6, and 8 h and filtered through a 0.22 µm pore-size membrane for the removal of the bacterial cells. The absorbance of the filtrate was detected utilizing a UV-spectrophotometer at λ = 260 nm. The nucleotide leakage was confirmed to be a valid indicator of cytoplasmic membrane damage.
Scanning electron microscopy (SEM) was also employed to observe morphological changes caused in S. aureus and P. aeruginosa by the ethanolic extract of T. indica. Cultures of tested microorganisms were treated with the detected MIC, then incubated for 6 h at 37 • C. After incubation, bacterial cells were pelleted by low-speed centrifugation (4000 rpm for 15 min); washed with sterile, distilled water; and fixed with 3% glutaraldehyde in 0.1 M phosphate buffer for 4 h at 4 • C. Then, cells were exposed to a secondary fixation with 2% aqueous solution of osmium tetroxide for 60 min at room temperature, and were then serially dehydrated with 75, 95, and 100% ethanol. The last drying step was performed over anhydrous CuSO 4 for 15 min. After finishing the drying step, we mounted the cells on stubs of 12.5 mm diameter, attached them with sticky tabs, and then coated them in an Edwards S150B sputter coater with 25 nm thickness. Non-treated cells were used as negative controls. Small cell samples of the treated bacteria and the relative controls were examined with SEM (JEOL, Japan) at an accelerating voltage of 20 kv [44].

TLC and Bio-Autographic Assays
Thin layer chromatography (TLC) was carried out to identify the bioactive fractions of T. indica extract. First, its powder was dissolved in ethyl acetate and spotted by capillary tubes on TLC paper (20 × 20 cm) using running solvents chloroform/methanol (6:4, v/v). The detected fractions were then dissolved in methanol and dried. The retention factor (Rf) values of each fraction were calculated, and the antimicrobial activity of the dried fractions was re-tested against the selected pathogenic bacteria using bio-autographic assay. TLC-dried fractions were placed on the surface of a Mueller-Hinton agar plate seeded with each microbe and incubated at 37 • C for 24 h. After incubation, the clear zone that appeared on the media was taken as proof of the antibacterial efficacy of the tested extract [45,46].

Phytochemicals Analysis
The active ingredients of the T. indica ethanolic extract were analyzed for the presence of different phytochemicals according to standard procedures [47]. The structure of the purified active components of T. indica was analyzed using data from a wide range of spectroscopic techniques, such as ultraviolet (UV), infrared (IR), and mass spectroscopy (MS), at the Regional Centre for Mycology and Biotechnology, AL-Azhar University, Cairo (Egypt).

GC-MS Analysis
Identification of the bioactive substances from T. indica was conducted at the National Research Center, Cairo, Egypt. A GC/MS-QP -1000 -Mass spectrophotometer (SHIMADU, Kyoto, Japan) instrument was used for analyses. For interpretation of the mass spectroscopy (GC-MS), we used the database of the Chemical Abstracts Service (CAS). The spectrum of unknown components was compared with the spectrum of known molecules stored in the CAS and Wiley 6 N libraries [48,49]. We recorded the retention time, molecular weight (M.Wt), molecular formula, and composition percentage in the sample material, following a previously published method [50].

Molecular Docking (In Silico) Studies
A molecular docking study, which afforded us further insights into the inhibitory potential of the investigated compounds, was used in this study. The nuclei of the detected bioactive compounds of T. indica were evaluated against all virulence proteins expressed by the tested pathogens. The potential of the investigated compounds for the virulence proteins S. aureus and P. aeruginosa was pursued via molecular docking using an MOE 2019 suite [51].

Preparation of the Investigated Compounds
By using PerkinElmer ChemOffice Suite 2017, the bioactive compounds of the tested extract were chemically drawn to make them ready for the molecular docking program [52,53]. We uploaded the investigated compounds to one database and saved them as an MDB extension file.

Preparation of the Proteases of S. aureus and P. aeruginosa
All virulence proteins of the X-ray structure of S. aureus and P. aeruginosa were detected from an online protein data bank website, and downloaded with PDB entries 4INK [54] and 1EZM [55]. Accordingly, the sequence of the target protein chain was identified and protonated; then, the broken bonds were connected and fixed. Before beginning the docking process, the virulence proteins of the tested pathogens were energetically minimized [52,53].

Conclusions
This study revealed that T. indica ethanolic extract had a variety of in vitro antibacterial activities against MDR Gram+ve and Gram−ve isolates. It also had synergistic effects with conventional antibiotics (imipenem, amikacin, and ofloxacin) and reduced their MICs. The mechanism of activity showed that the extract was able to influence the cellular membrane permeability, as evidenced by potassium and nucleic acid leakage, resulting in cell lysis and death. Additionally, the bioactive compounds showed a good binding capacity with the S. aureus protease and P. aeruginosa elastase confirming the antivirulence activities of this extract. Therefore, we introduced a new combination of T. indica ethanolic extract with other antibiotics to fight MDR pathogens and to avoid treatment failure. Further studies are still needed to unequivocally determine the activity of these molecules and to evaluate their antimicrobial effects individually.  Institutional Review Board Statement: Our study did not require ethical approval, as all isolates were kindly provided from the microbiological laboratories of Zagazig University Hospital.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data generated or analyzed during this study are included in the published article, and there are no supplementary information files.

Conflicts of Interest:
The authors declare that they have no known competing financial interests nor personal relationships that could have appeared to influence the work reported in this paper.