Small Molecular Antimicrobial Ligands of YspD are Potential Therapeutic Agents Against Yersinia enterocolitica Infection

Abstract

YspD is a hydrophilic translocator forming the platform for assemblage of functional translocon. Exposure to the extra-cellular milieu makes YspD a potential therapeutic target. DoGSiteScorer predicted best druggable pocket (P0) within YspD, encompassing predominantly the C-terminal helical bundles and the long helices-9 & 5. COACH metaserver also identified ligand binding residues within the aforementioned druggable pocket mapping to helix-9. Amino acids of helix-9 are involved in oligomerization of YspD. Interaction of helix-9 and parts of C-terminal of YspD with hydrophobic translocator protein (YspB), is essential for translocation of bacterial effectors to initiate an infection. Helices-9 & 5 form an intramolecular coiled-coil structure, required for protein–protein interaction. Targeting intramolecular coiled-coil and parts of C-terminal would be important for functional inactivation of YspD. Solvent exposed surface in YspD, particularly in P0, enhances its accessibility to ligands. Nine small molecular inhibitors of TIIISS were identified and retrieved from ZINC15 database (drug-library) as putative drug candidates. Molecular docking of potential ligands with P0 was done using SwissDock server and Achilles Blind Docking server. Considering the “Significance” threshold of binding score and region of interaction, Salicylidene Acyl Hydrazide derivatives (INP0400) and Phenoxyacetamide derivative (MBX1641) were found to bind effectively with YspD. These potential ligands interact with functional domains of YspD including parts of C-terminal and the intramolecular coiled-coil, which may affect the oligomerization of YspD and disrupt the interaction of YspD with YspB, inhibiting formation of functional translocon. The identified small molecular antimicrobial ligands of YspD could be tested in vivo to attenuate Y. enterocolitica infection by deregulation of Ysa-Ysp TIIISS.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40011-022-01443-2.

Introduction

Different species of Gram-negative bacterium Yersinia show pathogenicity to humans. Yersinia pestis infection leads to pneumonic and bubonic plague. Yersinia pseudotuberculosis causes enteric disease, whereas Yersinia enterocolitica is specifically responsible for diseases like gastroenteritis and mesenteric adenitis. Opportunistic pathogen Y. enterocolitica enters the body through food and water and causes Yersiniosis in immune-compromised individuals. Although Yersiniosis is a self-limiting condition, the mortality rate is high in infants and in children below 5 years of age. Cosmopolitan distribution, resistance to antibiotics like penicillin, ampicillin, cephalotin and tendency to cause nosocomial infections make Y. enterocolitica a dreaded pathogen. Its infection is treated with antibiotics like chloramphenicol, ceftriaxone, sulfamethoxazole and fluoroquinolones [1–4]. Once Y. enterocolitica migrates to the gastrointestinal tract, Yersinia secretion apparatus-Yersinia secretion protein (Ysa-Ysp) Type III Secretion System (TIIISS) gets activated and the gastrointestinal phase of the infection commences. Then, the bacterium invades the M-cells overlying the Peyer’s patches and migrates to visceral organs like spleen or lymph node, to initiate the systemic phase of the infection. In this stage, Y. enterocolitica combats host immune responses using Yop (Yersinia outer protein) effectors of Yersinia secretion component-Yersinia outer protein (Ysc-Yop) TIIISS. A 199 Kb Pathogenicity Island in the chromosome encodes Ysa-Ysp TIIISS, which is activated under low temperature, high salts (LTHS) condition. Ysc-Yop TIIISS is encoded by a 70 Kb pYV plasmid [5–8]. Bent et al. clearly established that Ysa-Ysp TIIISS is activated under physiological condition within the host and this TIIISS is absolutely essential for the virulence of Y. enterocolitica [9].

TIIISS is a secretion apparatus formed by assemblage of different types of protein to transport bacterial toxin from bacterial cell to the host cell cytoplasm. Effector proteins, Structural proteins, Translocator proteins, Chaperones and Regulator proteins are the components of TIIISS. After the activation of TIIISS, a needle complex is formed with a translocation apparatus (translocon) at its tip acting like a nano-syringe. The basal structure encompasses the inner and the outer membranes and the periplasmic space. YscF is the needle protomer in Y. enterocolitica, which oligomerizes to form a narrow substructure with outer diameter 6–10 nm, inner conduit of 2–3 nm and length 60–120 nm [10–12]. At the needle tip, hydrophilic translocator proteins like YopD, LcrV (Yersinia), PcrV (Pseudomonas) form a platform on which major hydrophobic translocators like YopB, YspB, PopB and minor hydrophobic translocators like YopD, YspC and PopD assemble to form the functional translocon, interacting with the host cell membrane, thereby transporting the effector toxins to the host cell cytoplasm. Hydrophilic translocators are capable of creating pores in the host cell membrane resulting in the death of the host cells. The strategic location of translocator proteins in the extracellular milieu and their functional significance makes them potential therapeutic targets. Deregulation of translocator assemblage attenuates Y. enterocolitica infection [11–14].

In the downstream region of Ysa-Ysp TIIISS, syc-ysp operon is present which codes for proteins like SycB, YspB, YspC, YspD, YspA and AcpY. YspB and YspC are sequestered in the bacterial cytoplasm by Class II chaperone SycB. YspD is the annotated hydrophilic translocator which forms the functional scaffold for assemblage of YspB and YspC. Localization studies by Foultier et al. and fluorescence studies of Young et al. depicted the secretory nature of YspD [7, 15]. Translocator and chaperone proteins YspB, YspC, YspD and SycB of syc-ysp operon of Ysa-Ysp TIIISS are homologous to IpaB, IpaC, IpaD and IpgC of Mxi-Spa TIIISS of Shigella flexneri and SipB, SipC, SipD and SicA of SPI-1 TIIISS of Salmonella enterica, respectively. In Psc (Pseudomonas), Ysc (Yersinia) and Asc (Aeromonas) TIIISS, regulator proteins like PcrG, LcrG sequester hydrophilic translocators like PcrV, LcrV, respectively, in the secretion competent state. However, hydrophilic translocator YspD lack cytoplasmic regulators and like the translocators of Mxi-Spa and SPI-1 TIIISS possesses an N-terminal domain with chaperoning activity. Long intramolecular coiled-coil domain in YspD plays an important role in protein–protein interaction. The structural aspects of dimerization of YspD and its interaction with YspB have been elucidated in our previous study [6, 7, 15–17]. In this study, the most druggable sites in YspD protein is predicted and screened along with some small molecular antimicrobial ligands of YspD, which could be putative drug candidates against Y. enterocolitica infection.

Material and Methods

Prediction of Druggable Sites in YspD

In our previous study, a reliable homology model of  (29 N-terminal amino acid deleted YspD) YspD, was generated by I-TASSER with high C-Score (− 0.49) and TM-score (0.65 ± 0.13) [17]. DoGSiteScorer is a web-based automated pocket detection and analysis tool [18]. DoGSiteScorer was used to predict the druggable pockets in the YspD model based on the Drug score, Simple score, Volume and Surface Area.

Prediction of Ligand Binding Sites in YspD

YspD model was loaded into COACH server, which predicts the ligand binding sites within a protein using a metaserver approach and also proposes the probable ligands interacting with protein of interest. COACH server predicts complementary ligand binding sites by TM-SITE and S-SITE, two comparative methods, which are further combined with prediction from methods like FINDSITE, COFACTOR and ConCavity to predict the final sets of ligand binding sites [19].

Calculation of Solvent Accessible Surface Area

Solvent Accessible Surface Area (SASA) of YspD and Pocket-0 of YspD was calculated using web-based server GETAREA. PDB files of YspD and Pocket-0 of YspD were loaded as inputs. The radius of the water probe was selected to be 1.4 Å. Total Area/Energy and Area/Energy per Residue were calculated. In this way, the solvent accessibility of specific amino acids and total number of surface and buried atoms were estimated [20].

Selection of Potential Ligands of YspD

From the existing literature, after elimination of redundancies, nine ligands were selected which were potential TIIISS inhibitors. Five ligands were selected from the review article of Gu et al. and four ligands were selected from the study of Dey et al. [21, 22]. All these ligands were retrieved from ZINC15 database (Ligand library) for molecular docking studies [23]. Online molecular converter powered by ChemAxon JChem was used to convert ligand file format from .sdf to .mol2, which is a pre-requisite for molecular docking. Three therapeutic agents were selected as control molecules, which does not specifically inhibit TIIISS. These are Glycolic acid, Resorcinol and Acetohydroxamic acid. Glycolic acid and Resorcinol are actively used in dermatological treatments and Acetohydroxamic acid is used as adjunct in chronic urea splitting urinary tract infection. These control molecules were also subjected to molecular docking with YspD.

Molecular Docking Analysis of YspD with Antimicrobial Ligands

Molecular docking analysis was performed using SwissDock and Achilles Blind Docking server. The SwissDock web server is based on the docking software EADock DSS developed by Swiss Institute of Bioinformatics (SIB) [24]. The homology model of YspD was loaded as the receptor (pdb file) and docked with different ligands and control molecules (.mol2 files). Docking was done without allowing flexibility for the side chains within atoms of ligand in its reference binding mode. From the output files, the receptor ligand interaction was analysed on the basis of $\mathrm{\Delta }G$ value. Subsequently, the interaction between the antimicrobials and YspD was also analyzed from structural perspective. Achilles Blind Docking server is based on Autodock Vina of version Vina Vision developed at Universidad Católica San Antonio de Murcia [25]. The homology model of YspD was loaded in pdb format and ligands and control molecules were loaded in .mol2 format. The server automatically converted the receptor and ligands files into pdbqt format and different docking simulations were performed on each alpha carbon of the protein. New binding modes were detected on the basis of best binding affinities. The ligands and receptor interacting residues were also predicted by the server.

Representation of Models

UCSF-Chimera was used to mark the Solvent Accessible Surface Area (SASA), the location of Pocket-0 (P0) in YspD and representation and analysis of the molecular docking results. PyMOL was used to represent the ligand binding residues within YspD.

Results and Discussion

Identification of Potential Druggable Pockets within YspD

DoGSiteScorer predicted nine potential druggable pockets (P0, P1, P2, P3, P4, P5, P6, P7 and P8) in YspD (Fig. 1). Web-based server DoGSiteScorer uses a Grid-based method, based on Gaussian filter to detect potential drug binding pockets on the basis of three dimensional structure of the protein [18]. The nine potential drug binding pockets detected within YspD are further divided into sub-pockets. Based on the Drug Score, Simple Score, Volume and Surface Area determined by DoGSiteScorer, Pocket-0 (P0) is the potentially best druggable site within YspD (Table 1) (Fig. 2). COACH server from Zhang Lab uses a metaserver approach for the prediction of ligand binding sites within proteins. TM-SITE and S-SITE methods of COACH server identified ligand binding templates from the BioLiP protein function database. Combining these results with other methods like COFACTOR, FINDSITE and ConCavity, final ligand binding sites are predicted within YspD [19]. Comparing all these results, based on the C-Score and Cluster size, amino acids 80, 84, 295, 298, 299, 302 are consensus residues present in the ligand binding sites (Fig. 3). This corroborates the finding of DoGSiteScorer as amino acid residues 84, 295, 298, 299 and 302 also maps to the pocket P0 (Figs. 2, 3). Amino acids 295, 298, 299 and 302 belong to helix-9 of YspD and residues 298 and 302 play an important role in formation of YspD-dimer [17]. P0 predominantly encompasses C-terminal helical bundles and long helix-9 and helix-5 of YspD. Helix-9 is essential for the multimerization of YspD. It forms an intermolecular coiled-coil structure with helix-5 of YspD, which is significant for protein–protein interaction. This intramolecular coiled-coil region is the functionally most significant and evolutionarily most conserved region within YspD. The consensus ligand binding residues obtained from COACH metaserver predominantly lies in the intramolecular coiled-coil of YspD. The helix-9 and parts of C-terminal domain of YspD interact with YspB. This interaction is essential for formation of functional translocon, the sub-structure needed for transport of bacterial toxin to the host cell. Therefore, targeting Pocket-0 may disrupt the formation of functional translocon and attenuates Ysa-Ysp TIIISS, thereby inhibiting Y. enterocolitica infection [13, 17, 26].

Fig. 1.

Nine Druggable pockets within YspD identified by DoGSiteScorer server. Adjoining colour code represents the location of these pockets. The cartoon representation of YspD model is shown in deep blue

Table 1.

Druggability of different pockets within YspD predicted by DoGSiteScorer based on the drug score, simple score, volume and surface area

	Volume Å3	Surface Å2	Drug score	Simple score
P0	1045.62	1308.3	0.81	0.64
P1	884.64	1285.34	0.79	0.59
P2	457.63	848.65	0.72	0.29
P3	392.99	477.06	0.8	0.19
P4	291.2	451.0	0.61	0.17
P5	180.1	236.34	0.32	0.03
P6	167.17	338.69	0.37	0.04
P7	159.96	477.68	0.51	0.0
P8	130.7	366.13	0.26	0.0

Fig. 2.

Location of best druggable pocket—Pocket-0 (P0) within YspD. The cartoon representation of YspD model is shown in deep blue and Pocket-0 is shown in dark grey colour

Fig. 3.

The Ligand binding residues within YspD predicted by COACH metaserver. The cartoon representation of YspD model is shown in deep blue. The Ligand binding residues are marked in red and shown by Sticks

Sufficient Solvent Accessible Surface Within YspD Facilitates Binding of Ligands

YspD is a hydrophilic translocator protein and a potential drug target. Its location in the extracellular milieu enhances its accessibility to drugs, as the potential drug candidates need not cross any membrane barrier to target YspD. The polar nature of drug molecules often affects their permeability across lipid bilayer membranes. Web-based server GETAREA predicted that there is sufficient solvent exposed surface in YspD in general and in Pocket-0 in particular, which enhances its accessibility to drugs [20]. The red region marks solvent exposed surfaces of YspD and P0 of YspD (Fig. 4). The number of surface atoms in YspD is 1488 and the number of buried atoms is 1025. The ratio of surface atoms and buried atoms significantly increases in P0 of YspD, which contains 234 surface atoms and 12 buried atoms. Therefore, Pocket-0 is sufficiently solvent exposed as corroborated by Fig. 4, which further establishes its potentiality as a drug binding site. The solvent accessibility of individual amino acid residues was also predicted by GETAREA server [20].

Fig. 4.

YspD and Pocket-0 within YspD have sufficient Solvent Accessible Surface Area. A The surface representation of YspD in different orientations. B The surface representation of Pocket-0 within YspD in different orientations. Red region marks the Solvent Accessible Surface Area (SASA) and Blue region marks the Solvent Excluded Surface Area (SESA)

Identification of Small Molecular Antimicrobial Ligands of YspD

Use of antibiotics is a time-tested and efficient method to combat bacterial infections. Y. enterocolitica infection is also treated with antibiotics like chloramphenicol, ceftriaxone, sulfamethoxazole and fluoroquinolones [3]. However, this therapy has side effects and facilitates the evolution of antibiotic resistant strain of the bacteria. Y. enterocolitica uses arsenal of proteins of Ysa-Ysp TIIISS to initiate an infection. Therefore, targeting specific proteins of TIIISS (especially the proteins exposed to the extracellular milieu) will be a preferred choice to curb Y. enterocolitica infection as it will not confer any resistance towards the putative drug candidates. Amongst the large number of potential drugs available in different databases, we have narrowed down our choice to nine potential drug candidates. Four ligands (2-Methylquinoline-4-Amine, 4-Morpholin-4-ylaniline, 4-(2-Pyrrolidin-1ylethyl) Aniline and 1H-Indole-5-ol) were selected from a study by Dey et al. [22]. Methylquinoline, Pyrrolidin-Aniline, Hydroxyindole and Morpholinoaniline scaffolds successfully binds to IpaD, a hydrophilic translocator protein and a close homolog of YspD. Five ligands (Phenoxyacetamide derivative (MBX1641), Salicylidene Acyl Hydrazide derivative (INP0400), 2-Imino-5-Arylidenethiazolidinone derivative (TTS29), Salicylidene Acyl Hydrazide derivative (INP0010) and Benzoic acid) were identified from a review published by Gu et al. [21]. Salicylidene Acyl Hydrazide derivatives target Yop effectors like YopE in Yersinia. Phenoxyacetamide derivative (MBX1641) inhibits TIIISS effector protein ExoS in P. aeruginosa. 2-Imino-5-Arylidenethiazolidinone derivative (TTS29) disrupts the needle assembly of Yersinia spp., P. aeruginosa and Francisella novicida. Plant phenolic compounds like benzoic acid inhibit the TIIISS of Erwinia amylovora [21]. Analysing the broad classes of molecular inhibitors of TIIISS and their efficacy, these nine TIIISS inhibitors were selected as potential ligands of YspD. These ligands were retrieved from ZINC15 database (Drug library) for molecular docking analysis [23]. Three therapeutic small molecules were selected as control, as they do not specifically inhibit TIIISS. Glycolic acid and Resorcinol are used in dermatological treatments and Acetohydroxamic acid is used in treatment of chronic urinary tract infections. These control molecules were also retrieved from ZINC15 database for molecular docking with YspD. The structures of the aforementioned small molecular antimicrobial inhibitors of TIIISS are provided in Fig. 5.

Fig. 5.

Structure of therapeutic molecules (non-specific towards TIIISS) and small molecular inhibitors of Type III Secretion System used in this study. These inhibitors are identified as potential ligands of YspD. A Structure of three therapeutic (control) molecules. B Structure of nine small molecular inhibitors of TIIISS identified as putative drug candidates against YspD

Molecular Docking Predicted Four Antimicrobial Ligands Binding Effectively with YspD

In vitro and in vivo testing of drugs is a costly affair. Therefore, it is supplemented with in silico screening to narrow down the choices for putative drug candidates. SwissDock and Achilles Blind Docking server were used for molecular docking analysis of YspD with the nine potential ligands, which are inhibitors of TIIISS and three therapeutic (control) molecules (Figs. 6, 7 and Supplementary Fig. 1). SwissDock uses EADock DSS and the results are categorised based on $\mathrm{\Delta }G$ value [24]. Achilles Blind Docking server is based on Autodock Vina, which detects new binding modes on the basis of best binding affinities [25, 26]. To screen specific high affinity ligands from non-specifically binding molecules, we have assumed − 7.5 kcal/mol $\mathrm{as}\mathrm{\Delta }G$ value ‘significance’ thresholds for SwissDock and − 8.0 kcal/mol as Autodock Vina binding energy ‘significance’ threshold. This is based on previous references of binding affinities shown by docking servers towards specific high affinity ligands and binding affinities of TIIISS inhibitors towards homologous hydrophilic translocator [27–29]. From the docking analysis, four potential ligands of YspD were identified based on the aforementioned criteria. Salicylidene Acyl Hydrazide derivative (INP0400) exhibited best $\mathrm{\Delta }G$ value (− 8.24 kcal/mol) and highest negative binding energy (− 10.7 kcal/mol), followed by Phenoxyacetamide derivative (MBX1641)—$\mathrm{\Delta }G$ value (− 7.57 kcal/mol) and binding energy (− 9.10 kcal/mol), Salicylidene Acyl Hydrazide derivative (INP0010)—$\mathrm{\Delta }G$ value (− 8.01 kcal/mol) and binding energy (− 8.80 kcal/mol), 2-Imino-5-Arylidenethiazolidinone derivative (TTS29)—$\mathrm{\Delta }G$ value (− 7.84 kcal/mol) and binding energy (− 8.40 kcal/mol) (Tables 2 and 3). Based on $\mathrm{\Delta }G$ value and binding energy, the three control molecules showed much less binding affinity towards YspD compared to the identified TIIISS inhibitors. Hydrophilic translocators like LcrV and PcrV (distant homologs of YspD) have been successfully targeted with small molecular antimicrobials for suppression of Yersinia and Pseudomonas infection, respectively [26].

Fig. 6.

Molecular Docking analysis of therapeutic (control) molecules and potential ligands with YspD using SwissDock server. The cartoon representation of YspD model is shown in deep blue and Pocket-0 is shown in dark grey colour. The control molecules and the ligands are marked in red colour—A Glycolic acid, B Resorcinol, C Acetohydroxamic acid, D Phenoxyacetamide derivative (MBX1641), E Salicylidene Acyl Hydrazide derivative (INP0400), F 2-Imino-5-Arylidenethiazolidinone derivative (TTS29), G Salicylidene Acyl Hydrazide derivative (INP0010), H Benzoic acid, I 4-(2-Pyrrolidin-1ylethyl) Aniline, J 4-Morpholin-4-ylaniline and K 2-Methylquinoline-4-Amine L 1H-Indole-5-ol

Fig. 7.

Molecular docking analysis of YspD with screened putative drug candidates using Achilles Blind Docking server. The residues of YspD interacting with the putative drug candidates were also shown. The cartoon representation of YspD is shown in magenta and the interacting residues of YspD are marked in blue. The complex of YspD and A Salicylidene Acyl Hydrazide derivative (INP0400), B Phenoxyacetamide derivative (MBX1641), C Salicylidene Acyl Hydrazide derivative (INP0010), D 2-Imino-5-Arylidenethiazolidinone derivative (TTS29), and E Specific interactions between YspD and Salicylidene Acyl Hydrazide derivative (INP0400)

Table 2.

∆G value for binding of control molecules and potential ligands with YspD, estimated by molecular docking analysis using SwissDock server

	Therapeutic (control) molecules and TIIISS-inhibitors (scaffold/ligand)	∆G value for binding (kcal/mol)
1	Glycolic acid (Control 1)	− 4.70
2	Acetohydroxamic acid (Control 2)	− 5.14
3	Resorcinol (Control 3)	− 5.72
4	Salicylidene Acyl Hydrazide derivative (INP0400)	− 8.24
5	Salicylidene Acyl Hydrazide derivative (INP0010)	− 8.01
6	2-Imino-5-Arylidenethiazolidinone derivative (TTS29)	− 7.84
7	Phenoxyacetamide derivative (MBX1641)	− 7.57
8	Benzoic acid	− 7.25
9	4-(2-Pyrrolidin-1ylethyl) Aniline	− 6.87
10	4-Morpholin-4-ylaniline	− 6.32
11	2-Methylquinoline-4-Amine	− 6.18
12	1H-Indole-5-ol	− 6.00

Table 3.

Binding energy, and amino acids of YspD interacting with control molecules and potential ligands, estimated by molecular docking analysis using Achilles Blind Docking server

	Therapeutic (control) molecules and TIIISS-inhibitors (scaffold/ ligand)	Binding energy (kcal/mol)	Amino acid residues of YspD interacting with the ligand
1	Glycolic acid (Control 1)	− 3.70	Gly 124, Pro 175, Gln 177, Asp 178
2	Acetohydroxamic acid (Control 2)	− 4.00	Ala 120, Gly 124, Phe 174, Gln 177, Asp 178
3	Resorcinol (Control 3)	− 5.30	Ala 120, Lys 123, Phe 174, Gln 177, Asp 178
4	Salicylidene Acyl Hydrazide derivative (INP0400)	− 10.7	Ile 29, Asn 33, Ser 46, Phe 50, Gln 115, Ala 119, Ala 120, Lys 123, Phe 174, Trp 189, Asn 293, Phe 289
5	Phenoxyacetamide derivative (MBX1641)	− 9.10	Ile 29, Asn 33, Leu 36, Phe 42, Glu 43, Ser 46, Phe 50, Lys 116, Ala 119, Ala 120, Phe 289
6	Salicylidene Acyl Hydrazide derivative (INP0010)	− 8.80	Leu 36, Glu 43, Ser 46, Ala 119, Lys 123, Phe 174, Pro 175, Gln 177, Trp 189
7	2-Imino-5-Arylidenethiazolidinone derivative (TTS29)	− 8.40	Lys 123, Phe 174, Pro 175, Gln 177, Lys 183, Trp 189, Pro 190
8	Benzoic acid	− 6.30	Leu 136, Phe 158, Met 162, Ile 240, Ile 243, Ile 271, Ile 278
9	4-(2-Pyrrolidin-1ylethyl) Aniline	− 7.60	Lys 116, Ala 119, Lys 123, Phe 174, Gln 177, Asp 178, Trp 189
10	4-Morpholin-4-ylaniline	− 6.80	Ala 120, Lys 123, Phe 174, Gln 177
11	2-Methylquinoline-4-Amine	− 7.30	Ala 120, Lys 123, Gly 124, Trp 189, Phe 174, Pro 175, Gln 177, Asp 178
12	1H-Indole-5-ol	− 6.60	Leu 136, Phe 158, Ile 240, Ile 243, Ile 271, Lys 275, Ile 278

Salicylidene Acyl Hydrazide Derivative (INP0400) and Phenoxyacetamide Derivative (MBX1641) Bind to Functional Domains of YspD

A potential ligand can be considered as a putative drug candidate when it binds to the functional domains of the target protein. Achilles Blind Docking server clearly marks the amino acid residues of YspD involved in interaction with the potential ligands [25]. It is observed that the control molecules only interact with a localized region within YspD. However, the interaction of YspD with other TIIISS inhibitors involves a much wider region of YspD, predominantly mapping to the most druggable Pocket-0 [Supplementary Fig. 1 and Table 3]. P0 predominantly comprises of helix-5, helix-9 and C-terminal loop. Helix-5 extends from Ile 99 to His 141, whereas helix-9 consists of amino acid residues Ala 260 to Met 314. The most functionally significant domain of YspD is helix-9, which is responsible for multimerization of YspD, its interaction with hydrophobic translocator YspB and forms an intramolecular coiled-coil with helix-5. Parts of C-terminal domain of YspD are also responsible for interaction with YspB [17]. Among the functional regions of YspD, Salicylidene Acyl Hydrazide derivative (INP0010) and 2-Imino-5-Arylidenethiazolidinone derivative (TTS29) bind to helix-5 and parts of C-terminal domain only. However, Salicylidene Acyl Hydrazide derivative (INP0400) and Phenoxyacetamide derivative (MBX1641) bind to almost the entire Pocket-0 encompassing N-terminal and C-terminal domains, and structurally and functionally conserved amino acid residues of helix-9 and helix-5 (like Gln 115, Lys 116, Ala 119, Ala 120, Lys 123, Phe 289, Asn 293). Phe 289 of helix-9, Ala 119, Ala 120 & Lys 123 of helix-5 and Trp 189 of loop of C-terminal domain are the amino acid residues interacting with most of the ligands of YspD [Fig. 7A–D and Table 3]. INP0400, MBX1641, INP0010 and TTS29 predominantly show hydrophobic interaction and hydrogen bonding with YspD. In case of INP0010, cation-pi interaction and halogen bonds are also observed [25]. Achilles Blind Docking server predicted that Salicylidene Acyl Hydrazide derivative (INP0400) participate in hydrophobic interaction with Trp 189, which is a part of the loop of C-terminal domain. Ala 119, Ala 120, Lys 123, Gln 115 of helix-5 and Phe 289, Asn 293, Gln 297 of helix-9 constitute the patch of hydrophobic interactions with Salicylidene Acyl Hydrazide derivative (INP0400). Lys 116 of helix-5 forms hydrogen bond with the ligand molecule. Lysine is a basic amino acid with a positively charged side chain in physiological pH 7.4 and its polar nature exposes the residue on the surface of the protein. This facilitates the access of the ligand. The polar uncharged residues like Glutamine and Asparagine will also showcase similar pattern for their interactions. Non-polar residues like Alanine and Phenylalanine are more likely to reside in the helical core of YspD and favours hydrophobic interactions (Fig. 7A–E). Most of the residues present in helix-5 and helix-9 are exposed on the surface of YspD which makes them more available for ligand binding. The multimerization of YspD takes place in the extracellular milieu making it more accessible for the drugs. The targeting of helix-9, intramolecular coiled-coil and parts of C-terminal bundle of YspD with putative drug candidates might inhibit the multimerization potential of YspD and disrupt its interaction with hydrophobic translocator YspB, thereby inhibiting the formation of functional translocon [13, 17, 26]. Salicylidene Acyl Hydrazide derivative (INP0400) and Phenoxyacetamide derivative (MBX1641) not only have high binding affinities for YspD, but also bind to the functionally important domains of YspD and might lead to the functional inactivation of YspD. Therefore, these two putative drug candidates could be selected for in vivo trials. It is established that Salicylidene Acyl Hydrazide derivative (INP0400) has been successful in blocking effector secretion in Shigella flexneri, Salmonella enterica and Chlamydia tracomatis and Phenoxyacetamide derivative (MBX1641) inhibits effector function in Pseudomonas aeruginosa [21, 30]. Further, hydrophilic translocators like LcrV and PcrV are established therapeutic targets for successful attenuation of Y. enterocolitica and Pseudomonas aeruginosa infections, respectively [26]. Hydrophilic translocators like LcrV, PcrV, IpaD and YspD are potentially better drug targets than toxic effector proteins, as they are exposed to the extracellular milieu for considerable period of time during activation of TIIISS. The sets of effector proteins of TIIISS show functional overlap, however, a single hydrophilic translocator must perform its unique function for activation of TIIISS.

Conclusion

The essential role of YspD in translocation of Yersinia toxins and its strategic location in the extracellular milieu makes YspD a significant drug target. Development of novel antimicrobials against bacterial virulence factors is an alternative to antibiotic therapy (which show significant collateral damages). Salicylidene Acyl Hydrazide derivative (INP0400) and Phenoxyacetamide derivative (MBX1641), both TIIISS inhibitors, interact with functionally important regions of YspD with high affinity. Therefore, Salicylidene Acyl Hydrazide derivative (INP0400) and Phenoxyacetamide derivative (MBX1641) is worth for a trial as a putative drug candidate against Y. enterocolitica infection. The objective of this research is to screen some TIIISS inhibitors which could bind with YspD and potentially inhibit Y. enterocolitica infection. The COVID-19 pandemic has depicted the significance of small molecular antimicrobials against viral and bacterial infection. However, to design a clinical drug, many more in vitro and in vivo experiments have to be performed.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

Department of Biotechnology, Government of West Bengal, India, provided the funding for this research [Grant No. 60-(sanc)/BT(Estt)/1P-4/2013].

Availability of data and materials

Not applicable for this study.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethics approval

Not applicable.