Background
Food-borne illnesses constitute a nagging public health issue that causes considerable impediments to the global health and economy, owing to the globalization and active food trade across countries [
1]. Contamination of foods can lead to food-borne illnesses that can occur at any point of production, processing, distribution, and consumption. This emerging public health problem causes considerable obstruction to socio-economic development as well as contributes significantly to the global burden of disability, morbidity, and mortality. Global estimates correlated foodborne illnesses with nearly 600 million episodes, 4,20,000 mortality, and 33 million disability-adjusted life years [
2].
Enteric bacterial pathogens namely,
Salmonella spp. and
Escherichia coli are important ‘priority’ listed foodborne pathogens [
3]. Non-typhoidal
Salmonella (NTS) serovars and diarrhoeagenic
E. coli (DEC) pathotypes constitute the leading causes of gastrointestinal infections worldwide [
4,
5]. Globally, the NTS serovars are responsible for approximately 153 million cases of gastroenteritis and 7000 deaths annually [
6]. Besides, the DEC pathotypes especially, enteroaggregative
E. coli (EAEC), are long been associated with foodborne outbreaks globally, thereby posing risk to global food safety and public health [
5,
7]. The occurrence of NTS and EAEC has widely been reported from both developed as well as developing countries [
4,
5,
8]. Antimicrobials have recently been employed on a larger scale as prophylactic as well as therapeutic agents to combat infections [
9]. In recent times, the evolution and natural selection of bacteria along with the unprecedented use of antimicrobials have contributed to an alarming increase in antimicrobial resistance (AMR) [
10]. Moreover, it has also been estimated that the mortality rate by way of AMR would increase to the tune of 10 million by 2050, which would further decrease the gross domestic product (GDP) by 3.50 %, resulting in an overall global economic loss of nearly USD 100 trillion [
11]. Of late, multi-drug resistance among the NTS serotypes and EAEC has been reported from various sources [
5,
12]. Hence, the focus has primarily been shifted towards alternative therapeutic strategies to counter the menace of AMR, apart from the routinely employed antibiotics [
13].
Recently, the use of phytochemicals has emerged as one of the promising holistic alternative approaches with minimal side effects [
14]. The bioactive phytochemicals and essential oils were found to have exerted significant antimicrobial activity against
Salmonella spp. and
E. coli [
15]. The screening of molecules with potential bioactivity is quite costly and may consume time. However, computer-aided drug design (CADD) could save time as well as the cost of synthesis of molecules and would ultimately curtail the cost of research [
16]. In silico molecular docking is one such CADD technique that would virtually predict the binding efficacy as well as the structure-based drug design [
16]. Moreover, the molecular docking provides successful insights into the structure-activity relationships, mode of activity, and further analysis from protein-ligand interaction [
17]. Such studies would culminate in the development of novel drug molecules at a faster pace against infectious pathogens. Additionally, the physicochemical properties of the molecule would provide vital information on the initial phase of drug development [
16,
17].
The phytochemical compounds—monoterpenoids (thymol and carvacrol) and phenylpropanoid (cinnamaldehyde) are generally considered safe for human consumption and have been approved by Food and Drug Administration (FDA) for being used as food additives. Moreover, these phytocompounds have been extracted from various indigenous herbs located in different parts of India [
18‐
21]. The objective of the present study was to study in silico absorption, distribution, metabolism, excretion, and toxicity (ADMET) analysis of the identified phytochemicals viz., thymol, carvacrol, and cinnamaldehyde, followed by in silico blind docking approach to check their ligand binding affinities to the outer membrane osmoporin protein (ompC) of
Salmonella Enteritidis and/or
Salmonella Typhimurium, and dispersin (aap) domain of EAEC. Later, the in silico approach was vetted by performing in vitro antimicrobial efficacy of these phytochemicals against the multi-drug resistant (MDR)-strains of NTS and EAEC. Also, in vitro safety and stability aspects of the identified phytochemicals were explored.
Discussion
Considering the pace at which drug resistance is emerging among the pathogens of public health significance, drug discovery, design and their development is the need of the hour [
22]. Researchers across the globe have been searching for the novel alternative therapeutic strategies or even drug re-purposing [
23,
24]. Phytochemicals constitute one such promising alternative [
25]. Phytochemicals have been reported to possess a broad-spectrum of antibacterial activities against various pathogens of public health importance [
26]. However, in vitro screening of phytochemical compounds for their antibacterial efficacy and toxicity studies would consume more time. Hence, in silico computational approaches associated with chemoinformatics, molecular docking, as well as artificial intelligence, have considerably increased during the past decade in the domain of drug design, development, and discovery [
27,
28]. Employing in silico approaches would, therefore, enable virtual screening of molecules which would result in providing better chances of discovering suitable drug candidates in considerably less time and cost. Several structure-based and ligand-based molecular docking approaches are currently available to facilitate high-throughput drug discovery [
16,
27]. In the present study, an in silico molecular docking of three identified phytochemicals (thymol, carvacrol, and cinnamaldehyde) to the ompC and aap protein domains of NTS and EAEC strains, respectively was performed. The in-silico data obtained by docking tools was further validated with in vitro antimicrobial efficacy of these phytochemicals against the MDR-strains of NTS and EAEC.
Obnoxious pharmacokinetic properties and toxicity remain hurdles in the development of drug candidates at clinical trials. Hence, identification of suitable candidates with drug-likeness along with sufficient information regarding absorption, distribution, metabolism, excretion, and toxicity (ADMET) is required during the initial phase of drug discovery [
28]. The identified phytochemical compounds-monoterpenoids (thymol and carvacrol) and phenylpropanoid (cinnamaldehyde) are generally considered safe for human consumption and have been approved by Food and Drug Administration (FDA) for being used as food additives [
18,
19,
21]. However, their ADME profile in connection with in vitro antimicrobial efficacy studies against MDR pathogens has rarely been undertaken. In the present study, the Swiss ADME server was used to analyse various ADME descriptors like, physiochemical properties, pharmacokinetics, solubility, lipophilicity, drug-likeness based on violation of Lipinski’s rule of five and medicinal chemistry [
29]. In silico ADME prediction of the three identified phytochemical compounds established drug-likeness as evidenced by no violation to its Lipinski’s rule of five and the obtained bioavailability score (0.55). Further, the red line of the compound in the bioavailability radar plot must be within the pink area to deem the compound as drug-like. The radar plots of thymol and carvacrol were observed completely within the pink area, while that of cinnamaldehyde was fairly within the pink area justifying its drug-likeness. Besides, the ADME data along with the boiled egg model revealed a better gastrointestinal absorption and permeation of the blood-brain barrier by all the tested phytochemicals [
29].
It is well understood that the P-gp and cytochrome P450 (CYP) help in biotransformation of xenobiotics to protect tissues [
30]. In this study, the phytochemicals exhibited no-P-gp substrate. The P-gp is indicated as the most important member of ATP-binding cassette transporters which is pivotal to evaluate the protective efflux of biological membranes (GI tract or brain) from xenobiotics [
29,
31]. Moreover, all the three phytochemicals tested interacted only with CYP1A2 isoenzyme of CYP family, conferring its effectiveness with minimal toxicity [
31].
Further, to investigate the in-silico toxicity parameters, Protox-II [
32] and StopTox [
33] machine learning apps were used. The predicted LD
50 (mg/kg) for the three phytochemicals tested ranged from 640 to 1850; hence were categorized as toxicity class-4 by Protox-II, indicating that the phytochemicals might be harmful if swallowed (300 < LD
50 ≤ 2000). Further, all the tested phytocompounds were found to be non-cardiotoxic by StopTox software, based on hERG liability prediction, with the confidence levels ranging between 50 and 60 %. However, the obtained toxicity findings need to be correlated with the in vitro safety assays before their assessment in suitable laboratory animal models.
The in silico ADMET analysis encouraged the phytochemicals to be employed further for Autodock-based computational docking studies [
34] to ompC of NTS [
35] and aap of EAEC [
36]. In this study, ligand-based interaction with the protein domains of pathogens was investigated using blind docking employing Autodock software. Osmoporin (ompC), present in the
Salmonella spp., is responsible for its survival and pathogenicity and plays a crucial role in diffusing hydrophilic compounds [
35]. Moreover, dispersin (aap) present in the EAEC strains is highly immunogenic and represents a class of aggregative factors which are responsible for its functional attribute [
36]. The binding free energy values observed for carvacrol, cinnamaldehyde, and thymol were nearly equal for the protein domains of both the pathogens studied. All the studied phytochemical agents interacted firmly with the respective protein motifs and probably have significance as inhibitors of both ompC (NTS) and aap (EAEC). This observed high binding energies obtained during molecular docking would be due to the strong hydrophobic interactions between the phytochemical ligands and the protein motifs-aap and ompC. The in-silico computational data hence obtained needs to be validated with the in vitro antimicrobial efficacy studies.
The antimicrobial efficacy of phytochemicals under study was determined by MIC and MBC values. The MIC values of all the three phytochemicals tested ranged from 0.25 to 0.50 µl/ml for MDR-EAEC strains, 0.06 to 0.50 µl/ml for MDR-NTS strains. However, the MBC values obtained were either equal to or greater than the MIC values. These varied MIC and MBC values of the phytochemicals tested could either be due to strain variation, the difference in the bacterial virulence factors, or structural differences in the bacterial membranes. Moreover, the observed in silico docking results were found to correlate well with the in vitro antimicrobial assays [
31,
37].
Further, the computational toxicity assays need to be correlated with the in vitro safety assay, before being passed on for in vivo clinical trials. Therefore, a haemolytic assay based on sheep and human ‘O’ RBCs was employed to ensure the safety profile of phytochemicals. A concentration-dependent haemolytic assay was observed with all three phytochemicals. At 1X MIC level, minimal hemolysis was observed both in sheep and human RBCs; however, at 2X and 4X MIC levels, the haemolysis observed was minimal to moderate, except in carvacrol. An improved specificity of phytochemicals against the bacterial cells and reduction in the haemolytic activity could further be accomplished by increasing the net charge and/or employing conjugation with nanoparticles [
38]. The nanotechnological interventions are therefore aimed at delivery of drugs (including phytochemicals) with an intention to effect site-directed drug delivery, reduction in the toxicity with non-compromised safety and therapeutic efficacy by minimizing the concentration of the drug with improved bioavailability [
39,
40].
Generally, bioactive phytochemical compounds get degraded in the gastrointestinal tract because of their poor stability [
41]; hence, the in vitro stability assays (pH, and cationic salts) were employed for the identified phytochemical compounds. In this study, thymol and cinnamaldehyde tested retained their antimicrobial activities even after subjecting them to varied stability conditions for MDR-EAEC strains, whereas slight decrease in the antimicrobial efficacy was observed against MDR-NTS strains tested. Moreover, carvacrol exhibited a slight increase in the MIC values for all the MDR strains tested. The stability of phytochemical compounds could be improved by nanobiotechnological interventions using a suitable delivery system [
41] to get it protected from external as well as biological influences.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.