Genome Investigation of a Cariogenic Pathogen with Implications in Cardiovascular Diseases

Abstract

The proportion of people suffering from cardiovascular diseases has risen by 34% in the last 15 years in India. Cardiomyopathy is among the many forms of CVD s present. Infection of heart muscles is the suspected etiological agent for the same. Oral pathogens gaining entry into the bloodstream are responsible for such infections. Streptococcus mutans is an oral pathogen with implications in cardiovascular diseases. Previous studies have shown certain strains of S. mutans are found predominantly within atherosclerotic plaques and extirpated valves. To decipher the genetic differences responsible for endothelial cell invasion, we have sequenced the genome of Streptococcus mutans B14. Pan-genome analysis, search for adhesion proteins through a special algorithm, and protein–protein interactions search through HPIDB have been done. Pan-genome analysis of 187 whole genomes, assemblies revealed 6965 genes in total and 918 genes forming the core gene cluster. Adhesion to the endothelial cell is a critical virulence factor distinguishing virulent and non-virulent strains. Overall, 4% of the total proteins in S. mutans B14 were categorized as adhesion proteins. Protein–protein interaction between putative adhesion proteins and Human extracellular matrix components was predicted, revealing novel interactions. A conserved gene catalyzing the synthesis of branched-chain amino acids in S. mutans B14 shows possible interaction with isoforms of cathepsin protein of the ECM. This genome sequence analysis indicates towards other proteins in the S. mutans genome, which might have a specific role to play in host cell interaction.

Electronic supplementary material

The online version of this article (10.1007/s12088-019-00823-z) contains supplementary material, which is available to authorized users.

Introduction

India is witnessing a dramatic increase in the incidences of cardiovascular diseases (CVD). The global scenario follows suit as CVDs are a major death toll-takers, according to the WHO in 2017 [1]. Most of the CVD related morbidities are related to strokes and heart attacks. Obstruction in blood flow to the heart is considered the leading etiological agent for cardiac dysfunction. Though the possible causes for restricted blood flow to the heart, mainly stem from fat depositions, there are other rising causes too for the same. Transient bacteremia is one such cause. Transient bacteremia following cleaning procedures and treatments are speculated etiological factors for atherosclerosis, infective endocarditis (IE), and vegetation on implantation devices [2]. Oral bacteria gain temporary entry into blood vessels, where they may adhere to previously present depositions or start a new one. Such events lead to immunological reactions and subsequent rupturing of plaques. The relation between oral hygiene and cardiac problems has started gaining attention only from the very recent past. Dental caries is a looming threat to a majority of the world population with far-reaching systemic consequences. Streptococcus species play a significant role in both oral health and other CVD conditions like IE [3]. S. mutans is a Gram-positive cariogenic organism, and its virulence properties include glucan production, acid tolerance, and acidogenicity [4]. S. mutans is capable of withstanding sudden drastic changes in its ambient environment, especially pH. S. mutans produces glucans from sucrose with the help of glucosyltransferases viz., GtfB, GtfC, and GtfD, which are primary components of biofilm [5]. Serologically S. mutans is classified into four serotypes based on the rhamnose polysaccharide composition of the cell envelope. Serotype c is mostly found in the oral cavity, followed by the other serotypes e, f, and k in decreasing proportions. It is also well known that the latter three serotypes are the ones predominantly found from patient samples of diseased heart valves [6]. With the increase in cardiovascular diseases and their plausible link with oral bacteria, there have been umpteen numbers of instances of isolation of bacterial DNA from extirpated valve samples from cardiovascular patients [7]. The correlation between these two factors stems from the fact that oral bacteria can gain transient entry into the bloodstream during routine dental procedures. This transient, short-lived event can also lead to the accumulation of bacteria on platelet aggregated depositions on the endothelial lining [8]. This kind of interaction with the endothelial lining of blood vessels has a long-term effect on the architecture of the blood vessel. Depositing bacteria not only have a slow, gradual impact on the lumen size, but they also raise immunological reactions against the host. These reactions from the immune system further exacerbate the condition and lead to the conversion of macrophages into foam cells [9].

Non c serotypes usually are found in more abundance in extirpated heart valve and from atheromatous plaques. The aim of the analysis of the draft genome of S. mutans B14 is to find genetic components unique to itself or in totality to the invasive serotypes (e, f, and k), since these are the ones that seem to have implications in cardiovascular pathologies. S. mutans B14 has been investigated for its invasive nature and cariogenic properties for over three decades now [10]. S. mutans B14 was first isolated from the dental plaque of a patient by A.S. Bleiweiss and has been maintained at the Dept of Oral Biology, Microbiology and Cell Science, University of Florida, USA.

Materials and Methods

Genome Sequencing and Assembly

The serotype e S. mutans B14 was kindly shared by Prof. Jacqueline Abranches, University of Florida and was routinely cultured in Todd Hewitt broth containing 0.2% yeast extract at 37 °C supplemented with 5% CO₂, under static conditions. Extraction of genomic DNA was accomplished by using the DNeasy blood and tissue kit (Qiagen, Hilden, Germany). Whole genome sequencing was performed using the Ion Torrent Personal Genome Machine (Life Technologies, Carlsbad, CA). Genomic DNA library preparation was done using the Ion Torrent library preparation kit. The de novo assembly was performed using Mimicking Intelligent Read Assembly (MIRA) version 3.9.18 [11].

Gene/protein Function Prediction and Annotation

S. mutans B14 genome was analyzed using the genome annotation tool WebMGA [12]. Annotation of the derived genomic sequence was performed using the Rapid Annotation using Subsystem Technology (RAST) program, and the annotated proteins were categorized based on function using the Cluster of Orthologous proteins (COG) database [13]. Delineation of the various proteins into functional metabolic pathways was done using the KEGG database [14].

Comparative Genomics

A total of 187 S. mutans genome sequences (whole genome and assemblies) covering the four serotypes were available through NCBI at the time of analysis. We performed comparative genomic analysis for the 187 genomes using the ROARY pan-genome software [15]. The output from the pangenome analysis software revealed information regarding the evolution of the genome of S. mutans. A phylogenetic tree was constructed based on the similarity of orthologous groups of genes and visualized using the Molecular Evolutionary Genetics Analysis (MEGA) version X [16].

Genome-Wide Screening for Adhesion Proteins

Analysis of the annotated gene set of S. mutans B14 was carried out to find possible candidate proteins, which might have a role in adhesion to host proteins. The SPAAN algorithm has been reported previously for testing and validating proteins having roles in adhesion [17]. The algorithm calculates the probability of a protein being an adhesion protein, using five attributes, namely, charge composition, dipeptide frequency, multiple frequency, and hydrophobic composition. The calculated probability P_ad value of > 0.7 is considered as a putative adhesion protein.

Protein–Protein Interactions of Predicted Adhesion Proteins

Adhesion proteins function in a myriad number of ways to gain access to the host system. Understanding protein function and their physiological role require information regarding its interaction with other proteins as well. Based on both experimental data like co-expression, immunoprecipitation, Two-hybrid systems and through prediction based approaches, the STRING database brings to the forefront all possible connections between proteins. Adhesion proteins fulfilling the required P_ad criterion were tested using the STRING database for possible protein–protein interactions [18].

Protein–Protein Interactions of S. mutans B14 Proteins with Human Extracellular Matrix Components

To gain insights into the proteins involved in communications with the host, we performed a search for protein–protein interactions of pathogen proteins with the extracellular matrix components of the human proteome. We used the HPIDB database for checking the presence of homologous interactions between the ECM components and adhesion proteins. The results were analyzed and visualized using the CYTOSCAPE tool, and subcellular localization of the proteins was predicted using pSORT-B [19] and TargetP [20].

Results

Genome Sequencing, Assembly, and Annotation

De novo sequencing assembly of the S. mutans B14 genome was done through Ion Torrent Personal genome machine. The genome sequence has been submitted to NCBI and is available through the accession number GCA_002179995. Sequenced S. mutans B14 chromosome consists of 1985,972 bp with a GC content of 36.8%. Supplementary Table 1 enlists the essential features of the genome. The distribution of genes under the different functional categories is depicted in Fig. 1.

Fig. 1.

Functional categorization of the genes present in Streptococcus mutans B14. Numbers corresponding to each slice denote the number of genes present in the particular COG under consideration

Carbohydrate Assimilation, Defense Mechanism, and Signal Transduction

Functional annotation of the coding sequences showed that B14 devotes a considerable portion of its genetic machinery to carbohydrate transport and metabolism. Our analysis was directed towards finding novel and previously unknown genetic predispositions which could attribute for the enhanced virulence of S. mutans B14. We investigated the genome sequence in the directions of most represented gene families, orthologous genes responsible for trans membrane transport, defense mechanism related genes, and genes functioning for membrane maintenance and recycling. The cluster of orthologous genes (COG) analysis reveals that the phosphotransferase system (PTS) components are the third highly represented genes in number. S. mutans adapts itself to the oral environment through many routes. One of the ways it can survive is through the production of lantibiotics against other microbes. The genome sequence of B14 reveals that the ABC type multidrug transport system (COG1131) is the most represented (in terms of number), which covers 43 such transporters and permeases. Two-component signal (TCS) transducing systems and quorum sensing are well-established paradigms in S. mutans. The genetic annotation for signal transducing components identifies the sensor histidine kinase VraS, TCS regulator VraR, and transporter associated with VraSR in the draft genome sequence of S. mutans B14.

Pan-Genome Analysis

Complete pan-genome analysis comprising all the available genome sequences (187) at the time of this analysis was carried out for S. mutans. We used six complete genome sequences and 181 draft or contig assemblies. Supplementary Figure 1 depicts the pangenome matrix generated from the alignment of all the genome sequences based on similarity in the gene content (Roary phylogenetic analysis output). The phylogenetic tree reconstructed through the MEGA, for better visualization of the grouped taxa is depicted in Supplementary Figure 2. The output from the comparative genomics tool Roary showed the presence of 24 unique genes in S. mutans B14. Some of the genes with probable involvement in virulence and pathogenicity include CAV53_00795, CAV53_05385, both annotated as IS982 family transposases, CAV53_09040 designated to be an ABC transporter substrate-binding protein, CAV53_02265, CAV53_02225 both encoding for bacteriocins and CAV53_08565 a collagen-binding protein. The progressive analysis of the complete genetic components of all the genomes compared the occurrence of new genes with the number of genomes. The graphical representation revealed that there, is no noteworthy increase in new genes with an increase in the number of genomes analyzed. However, the number of unique genes follows a steady ascending trend, as seen through the positive slope of the curve. The number of unique genes steadily increases with an increase in the number of assemblies compared. The gene pool consisted of the total gene content of 6965 genes, which accounted for 918 core genes shared by all the 187 drafts/complete assemblies, and 395 soft core genes present in more than 95% of the genome assemblies analyzed. It is noteworthy that the majority of the genes are categorized as cloud genes (present in less than 15% of the total genomes).

Prediction of Adhesion Proteins by SPAAN Algorithm

S. mutans B14 has been tested for its presence of the cnm gene previously [21], we wanted to check for the presence of other extra cellular matrix (ECM) binding proteins. Sequences of the putative ECM binding genes were retrieved from NCBI, and their presence was tested in our draft genome sequence through stand alone BLAST tool [22]. The presence of the well-characterized four adhesion proteins, viz Cnm, Cbm, WapA, and SpaP was verified. All the four genes are represented in the genome, albeit to different degrees of homology (Supplementary Figure 3). In addition to the known LPXTG motif proteins, we further screened for the presence of other putative adhesion proteins using the SPAAN software. Based on the recommended threshold values, we shortlisted proteins with a P_ad-value ≥ 0.7 to be putative adhesion proteins and a comprehensive list of the proteins is given in Supplementary Table 2. The complete list of the proteins with a P_ad value of ≥ 0.7, which can potentially function as adhesion proteins is provided in Supplementary Table 3. Though the set criterion of a P_ad value ≥ 0.7, for a protein, is essential to be considered as an adhesion protein, there is the probability for proteins with a lower value as well, of functioning as adhesion proteins. A total of 74 proteins have been predicted with putative adhesion function based on the P_ad value criterion. The predicted putative adhesion proteins include proteins with distinct physiological functions other than adhesion. These include glucosyltransferases, transporters, cell wall protein precursors, and even enzymes. Of these, 23 proteins have unassigned functions. Penicillin-binding proteins and ABC transporters are represented in the list. Two collagen-binding proteins (OVF1580.1 and OVF1581.1) have been predicted. These 2 proteins are homologous to the cnm gene and cnaB gene of S. mutans LJ23, respectively. Wall-associated proteins, surface-expressed glucosyltransferases, and endoglucanases are also among the predicted adhesion proteins.

Homology Based Protein–Protein Interactions of the Predicted Adhesion Proteins

We predicted the protein–protein interactions amongst the adhesion proteins, to have a deeper understanding of the regulations underlying them using the STRING software. The interaction network of the predicted adhesion proteins (Supplementary Figure 4) reveals an important feature. Node 1 (highlighted in the figure) consists of 9 protein partners. The distinguishing feature about this node is that most of the members of this group are localized to the cytoplasmic membrane, cell wall, or extracellular (localization verified with pSORT-B, data not shown). SpaP and WapA are cell wall-associated proteins characterized previously for their role in biofilm formation [23]. These are among the 6 LPXTG anchor proteins encoded by S. mutans. The WapA protein is designated collagen-binding protein (OVF01581.1) with cell wall localization. Since S. mutans is capable of invading endothelial cells in vitro, we also sought to study the protein–protein interactions of the pathogen with the human proteome. We restricted the human proteome dataset by selectively limiting our study to only the matrisome and its associated factors.

Possible Interactions of Human Extracellular Matrix Components with S. mutans B14 Proteins

We performed a protein–protein interaction study for ECM components and its associated factors, with the protein sequences of S. mutans B14. We tested for the presence of all possible interactions between the core matrisome proteins in humans [24] (proteoglycans, collagens, and ECM glycoproteins) and matrisome associated proteins (ECM affiliated proteins, ECM regulators and secreted factors) with the protein sequences of S. mutans B14. Interaction between the human and pathogen proteins was tested using the HPIDB [25]. The results were visualized through Cytoscape [26]. We analyzed all possible interactions of the B14 proteins with the core-components of the human ECM. The data is shown in Supplementary Figure 5. The analysis with the core matrisome protein components, viz., collagens, and other structural proteins did not reveal any significant interactions between the pathogen and host protein sequences. We wanted to analyze further and test if there was any possible interaction between the proteins and ECM associated factors. The ECM associated factors, regulators, and secreted factors also have a role to play in mediating pathogen attachment. Based on previous studies, we sought to test if the proteins of S. mutans B14 have any possible interactions with ECM associated factors. In our analysis through HPIDB (Supplementary Figure 6), we found that approximately 1% of the total proteins of S. mutans B14 have some form of interaction with 3% of the overall ECM associated factors. This further implies that a single protein of S. mutans B14 is perhaps engaging 2 or > 2 ECM associated factors simultaneously. Since the interaction between host and pathogen occurs in the extracellular milieu before physical attachment with the host membrane, we further screened the results based on the localization of the proteins. Tools pSORT-B and TargetP were used to predict the subcellular localization of the interacting pathogen and eukaryotic host proteins, respectively. S. mutans B14 proteins acetolactate synthase AlsS (OVF00222.1) and ATP synthase subunit β (OVF01497.1), were two proteins having a localization at the cytoplasmic membrane. Being surface-exposed, we wanted to find out if the interacting partners of these two proteins were also localized external to the cell. We found that the two proteins interacted with isoforms of cathepsin and plasminogen respectively, which have secretory nature (proteins possess signal peptide sequence, verified through TargetP). Our analysis of possible interactions between the core matrisome components (collagen, proteoglycans, and glycoproteins) did not yield any strong interactions with pathogen proteins. However, the matrisome associated factors displayed significant interactions with pathogen proteins. Thus, the pathogen proteins may use affiliated factors and regulatory factors of ECM, instead of direct physical interaction with structural components of the matrisome. A comprehensive summary of the interacting pathogen and ECM associated proteins, with their subcellular localization is given in Table 1. A detailed list of all the interactions between pathogen proteins and ECM associated factors is provided in Supplementary Table 4. The main focus of our study was to find genes and proteins with binding specificity for human extracellular matrix components. We have predicted a list of proteins through various analyses, which might have the capacity to bind ECM proteins, and thus influence host–pathogen interactions. The list of these proteins is given in Table 2.

Table 1.

Table summarizing interaction between matrisome associated factors and S. mutans B14 protein sequences

Pathogen ID	Gene/protein name	Localization	Localization	Gene/protein name	Host ID
OVF00222.1	Acetolactate synthase AlsS	Cytoplasmic membrane	Secretory	Cathepsin L1 isoform 1 preproprotein	NP_001903.1
			Secretory	Cathepsin L2 preproprotein	NP_001188504.1
			Secretory	Cathepsin L2 preproprotein	NP_001324.2
			Secretory	Cathepsin L1 isoform 1 preproprotein	NP_001244900.1
			Secretory	Cathepsin L1 isoform 1 preproprotein	NP_666023.1
			Secretory	Cathepsin L1 isoform 1 preproprotein	NP_001244901.1
OVF01497.1	ATP synthase subunit beta	Cytoplasmic membrane	Secretory	Pigment epithelium-derived factor precursor	NP_002606.3
			Secretory	Plasminogen isoform 2 precursor	NP_001161810.1
			Secretory	Plasminogen isoform 1 precursor	NP_000292.1
OVF01499.1	ATP synthase subunit alpha	Cytoplasmic	Secretory	Pigment epithelium-derived factor precursor	NP_002606.3
			Secretory	Plasminogen isoform 2 precursor	NP_001161810.1
			Secretory	Plasminogen isoform 1 precursor	NP_000292.1

Localization of the proteins is highlighted, to signify the location of the probable interaction

Table 2.

Compilation of proteins predicted as adhesion factors or associated to extra cellular matrix components

Pathogen ID	Gene/protein name	Localization	Earlier known function according to studies	Newly assigned function
OVE99678.1	Glycoside hydrolase 68 family protein—fructosyltransferase	Extracellular	Incorporation of fructose in extra cellular polymers	Adhesion (SPAAN)
OVE99865.1	Glucosyltransferase gtfB	Extracellular	Secreted enzyme acting on sucrose
OVE99688.1	Dextransae dexA	Cell wall	Enzyme acting on dextran, and controlling polymer structure
OVE99977.1	Glucan binding protein gbpC	Cell wall	Glucan binding protein
OVF00523.1	Biofilm regulatory protein brpA	Cell membrane	Involved in biofilm formation
‘OVE99594.1’	Glucan binding protein gbpB	Extracellular	Glucan binding protein
OVF00848.1	Glucan binding protein gbpD		Glucan binding protein
OVF01581.1’	Wall associated protein wapA	Cell wall	Involved in biofilm formation
OVF00139.1	Cell surface antigen spaP		Involved in biofilm formation
OVF00022.1’	Wall associated protein wapE	Cell wall	Cell wall biogenesis
OVF00171.1	Foldase prsA	Cell membrane	Involved in cell wall associated phenotypes and protein export
OVF00017.1	ABC transporter metal binding lipoprotein sloC	Cell membrane	Codes for lipoprotein receptor antigen(LraI)
OVF00222.1	Acetolactate synthase alsS	Cell membrane	Part of operon governing acetoin metabolism	ECM interaction (HPIDB)
OVF01497.1	ATP synthase subunit beta	Cell membrane	ATP metabolism
OVF01499.1	ATP synthase subunit alpha	Cytoplasmic	ATP metabolism

The proteins have been previously annotated to have other functions. The results from the SPAAN algorithm and HPIDB homologous interaction searches indicate new function

Discussion

Considering the immense implications of oral pathogens in CVD s, we have analyzed the genome sequence of S. mutans B14. S. mutans has many pathogenic traits, and our annotation of the functional genes showed that carbohydrate assimilation genes, lantibiotic production genes, and signal transduction systems are amongst the highly represented COGs. The phosphotransferase system is responsible for the transport of carbohydrates across the membrane for energy production. In S. mutans B14 the IIC, IIB, and IIA (OVF01207, OVE99686) components of the PTS are found in higher numbers. The first two are cytoplasmic components while the last forms a membrane channel for sugar transport. Arabinose efflux permease (OVE99908) and fructose 2,6 bisphosphatase (OVF00031) are the highest found genes with respect to numbers as revealed by the COG analysis. S. mutans is capacitated to defend itself from the antibiotics produced by other microbes as well, through the presence of lantibiotic exporters (OVF01235). The gene cluster responsible for the production of colicin V bacteriocin is present in the annotated genome sequence of S. mutans B14, and this gene is also present in other complete genome sequenced such as the S. mutans strain LAR01 [27] (accession ID of colicin homolog CO204_01600). One of the most well-characterized TCS is the LiaSR. Similarly, the VraSR regulon is well studied in Staphylococcus aureus, with special emphasis on the MRSA strains. The VraSR operon is responsible for the resistance towards antibiotics and mutations in the constituent genes viz., VraS and VraR decrease resistance towards the cell wall manipulating enzymes.

Microbes inhabiting the oral niche find novel routes of entry into the bloodstream, during routine procedures of cleaning. These transient events lead to bacteremia and formation of vegetation on already present lipid depositions, in certain cases. The molecular biology behind the interaction of oral microbes, with cells of the host, is multifaceted. Among the many mechanisms employed by microbes for adhesion and invasion to host tissue, the pili system (E. coli, Salmonella enterica, Pseudomonas aeruginosa), adhesion proteins (Streptococcus pyogenes) and invasive strategies (Listeria monocytogenes, Shigella sp.) have already been investigated in quiet detail [28]. The molecular mechanisms employed by S. mutans in host cell interaction are not fully understood. It was only recently discovered that the presence of adhesion proteins such as cnm is mandatory for the invasive capacity of S. mutans [21]. Adhesion proteins have a varied nature, and it is seen that there is considerable ambiguity in the nature of proteins serving the function of adhesion [29]. Wall associated proteins, envelope linked enzymes, and membrane-localized proteins can serve the function of attaching macromolecules of the host, such as the basement membrane in blood vessels. During our search for candidate genes with potential for adhesion to host tissue, we observed that genes, as predicted through the SPAAN algorithm, were also responsible for other functions such as biofilm formation. The protein–protein interactions amongst these adhesion genes revealed that surface localized genes were perhaps functioning in concert to bring about adhesion to host tissue (Supplementary Figure 4). BrpA has been investigated in much detail in the past, for its role in cell well envelope stress response, and biofilm formation. The role of BrpA in virulence and pathogenicity was also tested likewise, and its abolition drastically reduced the efficacy of the organism in its virulent capacity against an animal model too [30]. The other co-occurrence genes viz., ftf, gtfB, dexA, gbpC, gbpD, and gbpB of the node have been studied earlier as well. Fructosyltransferase (ftf) and Glucosyltransferase(gtfB) are secreted enzymes that act on fructose and glucose, respectively for incorporation into polymers constituting the biofilm. The cariogenic potential of the glucan binding proteins and their extracellular nature have been well documented [31]. The close cooperation amongst these proteins for cariogenic traits is not surprising. Among the adhesion proteins predicted by the SPAAN algorithm, the newly found adhesion candidates included collagen-binding protein, partial (OVF01580.1). This partial collagen-binding protein has a homolog in S. mutans LJ23 (serotype k) and is annotated to be a disrupted form of the cnm gene. Another collagen-binding protein (OVF01581.1), was also predicted as an adhesion protein, which is a homolog of the Cna protein B-type domain, found in S.mutans LJ23. It is worthwhile noting that the new adhesion proteins of S. mutans B14, have homologs in another serotype k strain. Thus, these strains may share some important virulence factors. Surface exposed proteins are considered the ideal choice as vaccine candidates [32]. The list of new adhesion proteins has surface-exposed proteins such as cell wall protein (OVF00022.1) and wall-associated protein (OVE99552.1). OVF00022.1 is an annotated cell wall protein found in S.mutans LAR01 (serotype e) and other strains of S. mutans. Apart from the proteins discussed above, there were certain other novel candidates as well. As listed in the Supplementary Table 3, bacteriocins were also predicted as possible adhesion candidates. It is well known that bacteriocins are peptides produced by a certain class of microbes, for out competing other bacteria, present in their niche. S. mutans is known to produce avid amounts of mutacins (bacteriocins) for its survival. Recent studies in other microbes such as Burkholderia, reveals adhesion proteins as promising vaccine candidates, against recurrent infection [33]. The study revolves around finding bacterial adhesion proteins, linking to host tissue and which could further be used for developing a vaccine, against the complex group of bacteria, causing cystic fibrosis-associated infections. Since the study identified novel adhesion functions for previously characterized antimicrobial proteins, it is not surprising to note, the predicted adhesion proteins in our study viz, OVF01247.1 bacteriocin, OVF01244.1 bacteriocin, OVF01678.1 bacteriocin-type signal sequence, OVF01249.1 bacteriocin, can also function in adhesion to host tissue as well. These candidate genes, have better clinical applications, as they can confer immunity against other harmful bacteria as well. If any such bacteriocin (mutacin) is administered as a vaccine, it will also help in curtailing infection load of S. mutans itself, against which the vaccine is designed. Among the other adhesion proteins were OVF01413.1 peptidase S8 and OVF01683.1 peptidase S16. There have been previous reports of proteolytic proteins displaying non-proteolytic functions [34]. Such proteins display and use their protein degrading functions for adhesion to host tissue and in dissemination throughout host [35].

Extracellular matrix of the host is a megastructure, and pathogens interact with its components for their dissemination through the host. Binding of Leptospira to ECM components collagen and laminin and the effect of this event in establishing host–pathogen interaction has been reviewed previously [36]. In a similar rationale, we have checked for all possible interactions between the ECM components of the human host and the proteome of S. mutans B14. Analysis to test the presence of such pathogen proteins in B14 genome, displayed some proteins having surface localization, and these proteins interact with affiliated factors of the ECM (Supplementary Figure 6). Cathepsin is a known lysosomal cysteine protease with functions of protein catabolism in the host. Its main substrates include collagen and elastin. Also, there are reports of cathepsin being involved in increasing pathogen viability (Mycobacterium) during infection of human monocytes in vitro [37]. From these evidences, it might be possible that the pathogen protein (Acetolactate synthase, OVF00222) is involved in engaging the host enzyme isoforms (CathepsinNP_666023.1,NP_001903.1,NP_001324.2,NP_001244900.1,NP_001244901.1,NP_001188504.1) and thus preventing the catabolism of Extracellular components like collagen and elastin. These ECM components are crucial for adhesion to host tissue. Staphylococcus aureus manganese transport protein (MntC) has been shown to bind both ECM core components and plasminogen. Recruitment of plasminogen and subsequent conversion to plasmin at the pathogen surface is usually accompanied by degradation of ECM proteins to facilitate pathogen movement and dissemination [38]. The possible interaction between plasminogen and pathogen protein (ATP synthase) as predicted through HPIDB, could perhaps also lead to pathogen dissemination, in a similar manner, which needs to be validated through experiments. Our study to find novel adhesion factors, and other proteins with the capability to interact with ECM components, has comprehensively given us few targets. These highly probable candidate genes/protein products have been summarized in Table 2.

Conclusion

Our analysis of possible interactions between the core matrisome components (collagen, proteoglycans, and glycoproteins) did not yield any strong interactions with pathogen proteins. However, the matrisome associated factors displayed significant interactions with pathogen proteins. We have predicted several proteins with the capacity to bind ECM proteins, and thus influence host–pathogen interactions.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author Contributions

PG and JR designed the study. SS, USV, and RK performed genome sequencing. SS and JS analyzed the genome sequence.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they do not have any conflict of interests in the publication of this manuscript.

Human and animal rights

This study does not contain the involvement of human subjects or animals, and no experiments have been performed by any of the authors on either human subjects or animals.