Genome-Wide Investigation Reveals Potential Therapeutic Targets in Shigella spp.

Shigella stands as a major contributor to bacterial dysentery worldwide scale, particularly in developing countries with inadequate sanitation and hygiene. The emergence of multidrug-resistant strains exacerbates the challenge of treating Shigella infections, particularly in regions where access to healthcare and alternative antibiotics is limited. Therefore, investigations on how bacteria evade antibiotics and eventually develop resistance could open new avenues for research to develop novel therapeutics. The aim of this study was to analyze whole genome sequence (WGS) of human pathogenic Shigella spp. to elucidate the antibiotic resistance genes (ARGs) and their mechanism of resistance, gene-drug interactions, protein-protein interactions, and functional pathways to screen potential therapeutic candidate(s). We comprehensively analyzed 45 WGS of Shigella, including S. flexneri (n = 17), S. dysenteriae (n = 14), S. boydii (n = 11), and S. sonnei (n = 13), through different bioinformatics tools. Evolutionary phylogenetic analysis showed three distinct clades among the circulating strains of Shigella worldwide, with less genomic diversity. In this study, 2,146 ARGs were predicted in 45 genomes (average 47.69 ARGs/genome), of which only 91 ARGs were found to be shared across the genomes. Majority of these ARGs conferred their resistance through antibiotic efflux pump (51.0%) followed by antibiotic target alteration (23%) and antibiotic target replacement (18%). We identified 13 hub proteins, of which four proteins (e.g., tolC, acrR, mdtA, and gyrA) were detected as potential hub proteins to be associated with antibiotic efflux pump and target alteration mechanisms. These hub proteins were significantly (p < 0.05) enriched in biological process, molecular function, and cellular components. Therefore, the finding of this study suggests that human pathogenic Shigella strains harbored a wide range of ARGs that confer resistance through antibiotic efflux pumps and antibiotic target modification mechanisms, which must be taken into account to devise and formulate treatment strategy against this pathogen. Moreover, the identified hub proteins could be exploited to design and develop novel therapeutics against MDR pathogens like Shigella.


Introduction
Shigella spp.are ubiquitous, nonmotile, non-spore-forming, rod-shaped Gram-negative bacteria that thrive in a wide range of conditions [1,2].Shigella spp.are the causative organism of shigellosis, which is an acute gastroenteritis infection [3,4].Shigella infections continue to be unacceptably high creating a global human health problem [4,5].According to the World Health Organization (WHO), each year shigellosis causes ~700 000 deaths worldwide [6].The Shigella spp.consist of four species such as S. dysenteriae (serogroup A), S. flexneri (serogroup B), S. boydii (serogroup C), and S. sonnei (serogroup D) and >50 serotypes [7,8].A shift in the epidemiology of Shigella serogroups has been observed in recent years.S. flexneri is mostly prevalent in developing countries, whereas S. sonnei is mostly prevalent in developed countries [9].S. boydii is most commonly isolated in Bangladesh, India, and other Southeast Asian countries, though S. dysenteriae (especially serotype 1) plays only a minor role in the endemicity of shigellosis in recent years [7,8].Despite that, humans are the sole reservoir for Shigella spp., but sometimes, natural infections can also occur in captive nonhuman primates (such as macaques) [7].Shigellosis affects individuals of all ages worldwide; however, children younger than five years bear the greatest disease burden [10].Shigella spp.are transmitted predominantly through the fecal-oral route [10].Moreover, Shigella spp.have been involved in several foodborne and waterborne outbreaks [4,11].Because of its low infectious dose (10-100 bacteria are sufficient to produce disease), Shigella has been a significant global health threat for many years, especially in Third World countries [4,10].About 5% to 10% of the diarrheal and dysentery cases are caused by different strains of Shigella worldwide [1,12,13], with the highest incidence among children ages 1 to 4 years [14,15].Like other diseasecausing bacteria, Shigella spp.are the most common and first-line contagious pathogens in both affluent and developing countries in recent years [2,11].This infectious pathogen thrives in crowded environments, such as prisons, slums, congested colonies, and daycare centers [10,16], and has become one of the biggest threats to public health worldwide [2,11].
Over the course of the outbreak, diverse antibiotic resistance patterns emerged in various Shigella strains, complicating treatment, including recent outbreaks of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains [7,17].The escalating prevalence of drug-resistant Shigella underscores an increasingly urgent public health challenge, necessitating a unified and coordinated response.Several recent reports have affirmed that pathogenic bacteria, such as Shigella spp., Klebsiella spp., and E. coli, stand out as the most crucial pathogens contributing to fatalities linked to AMR [10,18,19].Contiguously, antimicrobial resistance (AMR) has become a significant global concern in recent years, which is a consequent threat to public health worldwide [13,20,21].Over the past few decades, the uncontrolled use of antibiotics has resulted in the emergence of MDR in bacteria [13].MDR in Shigella spp.has recently been considered a significant concern in food safety and public health.The mainstay of shigellosis treatment involves antibiotic therapy, aiming to decrease the probability of complications and death while expediting clinical recovery [22].However, it is crucial to identify the suitable antibiotics for shigellosis treatment by thoroughly comprehending the evolving resistance patterns.This entails examining the phenotypic and genotypic profiles of AMR or antimicrobial resistance genes (ARGs) in Shigella spp.Notably, Shigella serotypes are found to carry conjugative plasmids and several ARGs such as mdtG, mdtE, ampH, beta-lactamase, emrR, emrK, acrB, tet(D), and tet(A) that encode resistance to several antibiotics routinely used by the healthcare professionals [23,24].Moreover, Shigella spp.may exhibit resistance to various antibiotics through multiple mechanisms, including efflux of antibiotics, inactivation of antibiotics, diminished permeability to antibiotics, and alteration of antibiotic targets [24,25].The presence of such antibiotic resistance genes in Shigella strains is a significant public health concern, as it limits the effectiveness of commonly used antibiotics.This can complicate the treatment of Shigella infections and potentially lead to more severe or prolonged illness.Thus, it is essential for healthcare professionals and researchers to monitor the prevalence of antibiotic resistance in bacterial pathogens like Shigella and develop strategies to mitigate the spread of resistant strains.This includes prudent use of antibiotics, improved sanitation and hygiene practices, and the development of new antimicrobial agents.A series of studies through system biology approaches have been conducted to identify the ARGs by annotating bacterial whole genome sequences (WGS) and further constructed gene-drug interaction networks, proteinprotein interaction (PPI) networks, and pathway enrichment analysis plots.For instance, E. coli O157:H7 [26], Proteus mirabilis [27], K. oxytoca [19], and Pseudomonas aeruginosa PA01 [28] genomes were explored through system biology and in silico approaches for MDR genes.However, there is no comprehensive study on gene-drug interactions, PPI, and pathway enrichment analysis among the globally circulating Shigella serotypes.Therefore, this study investigated 45 WGS of four Shigella spp. to detect ARGs and their possible mechanisms, gene-drug interactions, PPIs, and associated functional pathways.We built a phylogenetic tree to better comprehend the evolutionary divergence of the Shigella strains.We anticipate that the findings of this study will be useful in future clinical and pharmacological studies to design and formulate novel antimicrobials against Shigella like MDR and XDR pathogens.S1).The WGS data were obtained from the NBCI database as of June 30, 2023.We selected the genomes focusing on four Shigella spp. reported to cause shigellosis, a significant cause of diarrheal disease worldwide.The selection criteria are also based on their epidemiology, prevalence, antimicrobial resistance, clinical relevance, and impact on global health.

BioMed Research International
Moreover, all of the genomes were sequenced from the human stool samples of the hospitalized patients from seven countries (e.g., USA, China, Tanzania, Australia, South Korea, Sweden, and Germany) of the world (Table S1).We extracted only those WGS with a read coverage depth of ≥40x.We used only assembled whole genome sequences excluding draft genomes, scaffolds, and contigs (Table S1). ) tools were used to construct the phylogenetic tree applying ML method with 1000 bootstraps.The phylogenetic tree was subsequently imported to interactive tree of life (iTOL) v. 3.5.4(http://itol.embl.de/,accessed on July 10, 2023) for better visualization, and bootstrap values were reported for each branch [35].Each node in the phylogenetic tree represents an isolate, while the edge represents the Hamming distance between two isolates.
2.4.Functional Domains, Protein Featuring Pathway, and Protein-Protein Interaction Analysis.GO (Gene Ontology) database [36] and KEGG (the Kyoto Encyclopedia of Genes and Genomes) [37] databases were used to predict the functions of the detected ARGs.We utilized Enrichr [38] with Fisher's exact test to conduct the functional enrichment analysis of the study genomes.We retrieved GO keywords and KEGG pathway's data from the STRING v11.[41].To eliminate inconsistent PPIs from the dataset, we set a cut-off criterion to a high confident interaction score of ≥0.7.Thereafter, we incorporated the results from the STRING database into Cytoscape v3.9.1 (https://cytoscape.org/,accessed on July 13, 2023) to create a visual representation of the target network for molecular interactions [42] and to envisage the PPIs within the statistically significant ARGs.A p value < 0.01 was considered statistically significant.
2.5.Cluster Analysis and Selection of Hub Proteins.We utilized the molecular complex detection (MCODE) plugin from Cytoscape to identify the interconnected regions or clusters from the PPI network [43].The cluster finding parameters were adopted, such as a degree cut-off of 2, a node score cut-off of 0.2, a kappa score (K-core) of 2, and a maximum depth of 100 in the MCODE, which limits the cluster size for coexpressing networks and verifies the efficacy of interactive collaborators in the context of ARG expression [43].We further utilized the Cytoscape plug-in CytoHubba to find highly interconnected protein nodes and investigate the network topology.Our analysis included five approaches using the CytoHubba plugin, including three locally ranked methodologies, degree, MNC (maximum neighbourhood component), and MCC (maximum clique centrality), as well as two globally ranked methodologies, closeness and betweenness [44].Subsequently, collected ARGs from the CytoHubba were submitted to jvenn (interactive Venn diagram analyzer) for further analysis.The ARGs that were intersected among the six approaches of CytoHubba were considered as significant hub proteins [44].

Network Construction for Antimicrobial Resistance
Genes, Antibiotic Classes, and Resistance Mechanisms.To briefly visualize the ARG resistance mechanism, ARG antibiotic class, and antibiotic class resistance mechanism interaction networks, we constructed a sunburst plot and bar diagrams using Python.We used high-level programming languages to create extraordinary Python libraries, including Pandas, NumPy, SciPy, Matplotlib, and Plotly, for data analysis and visualization [45].

Statistical
Analyses.The majority of statistics for this project are descriptive in nature.For comparison of the distribution of ARGs, resistance mechanisms, gene-drug interactions, and PPI traits across the genomes of different Shigella spp., Student's t-tests were used with statistical significance observed at p < 0 05.

Genome Sequences, Phylogenetic Relationship, and
Resistance Repertoire of Shigella spp.We found 25,002 complete genomes of Shigella spp. in the NCBI GenBank database up to June 30, 2023.Through comprehensive filtering 3 BioMed Research International in terms of genome coverage, we selected only 45 complete genomes with a genome coverage of ≥40x, and size of the selected genomes ranged from 4.5 to 5.2 mega base pairs (Table S1).The selected genomes belonged to four species of Shigella, of which 37.78% (17/45), 31.11%(14/45), 24.44% (11/45), and 6.67% (3/45) genomes were from S. flexneri, S. dysenteriae, S. boydii, and S. sonnei, respectively.To determine the evolutionary relationship across the genomes of S. flexneri, S. dysenteriae, S. boydii, and S. sonnei, we established a maximum-likelihood phylogenetic tree (Figure 1).The phylogenetic tree showed three main clades among 45 genomes, with each supported by 100% bootstrapping (Figure 1).In this phylogenetic tree, clade 1 consisted of only three strains (two S. flexneri and one S. boydii) of Shigella, whereas clade 2 and clade 3 possessed 21 strains of Shigella (in each).We also observed that every ramification of nodes describes each strain of Shigella (Figure 1), implying no significant heterogeneity among the circulating strains of four Shigella spp.Furthermore, the genetic homogeneity of these clusters was evidenced by the high degree of overall nucleotide sequence similarity in the genomes.Through a comprehensive annotation of these genomes, we detected 2,146 ARGs including 835 ARGs in S. flexneri strains, 620 ARGs in S. dysenteriae strains, 519 ARGs in S. boydii strains, and 172 ARGs in S. sonnei strains.Out of 2,146 detected ARGs, 91 genes were found to be shared among the genomes of four Shigella spp.Therefore, these 91 shared ARGs were implemented    BioMed Research International to explore this current study further.Among the selected genomes, ARGs counts ranged from 38 to 61 with an average count of 47.69 ARGs per genome (Figure S1).In this study, we identified a higher proportion of ARGs coding for fluoroquinolones, penems, cephalosporins, tetracyclines, aminoglycosides, and macrolide resistance in related Shigella spp.(Figure S2).By calculating the number of ARGs against particular antibiotics, we found that 30 genes showed resistance against fluoroquinolone antibiotics followed by 27 genes against cephalosporins and penems (Figure S2).Besides, 21, 16, 15, 13, and 11 ARGs showed resistance to tetracyclines, aminoglycosides, phenolics, and rifamycin antibiotics, respectively (Figure S2).

Protein-Protein Interaction and Clustering Show Potential
Hub Proteins in Shigella Genomes.To predict typical PPI network connectivity, protein interactions and hub proteins were explored through the STRING database and visualized using Cytoscape (Figure 2).In a PPI network, a node typically represents a protein name, identifier, and other attributes depending on the context of the network analysis.On the other hand, edge represents the biochemical or physical interaction between the two proteins and also the strength of interaction.The PPI network possessed 38 nodes and 90 edges for S. boydii (Figure 2 1).The score in MCODE represents the "cliquishness" or the tightness of connections within a cluster.Higher scores generally indicate denser and more significant clusters, meaning that the nodes within those clusters are more tightly interconnected compared to nodes outside the cluster.Within these clusters, cluster 1 comprised 8 nodes and 94 edges (Figure 3(a) and Table 1) and cluster 2 comprised 7 nodes and 39 edges (Figure 3(b) and Table 1), whereas cluster 3 consisted of 5 nodes and 16 edges (Figure 3(c) and Table 1).
In this study, we detected hub proteins by employing five different methodologies such as betweenness (Figure 4 4(e)).A total of 13 proteins such cpxA, soxR, emrB, gyrA, marR, marA, acrR, acrE, tolC, mdtA, mdtC, acrB, and acrA were identified as hub proteins (ten proteins in each method) (Figure 5(a) and Table 2).Among these proteins, only four hub proteins (tolC, acrR, mdtA, and gyrA) were common in the five methods (Figure 5(b)).Importantly, marR and acrE proteins were detected in four methods except for betweenness (Figure 4(b)), while emrB was found in four methods except for the MCC algorithm (Figure 4(d)).The topological properties of unique hub proteins are shown in Table 2.Moreover, the identified hub proteins were also detected in cluster 1, cluster 2, and cluster 3 (as mentioned in Figure 3), indicating that they were the most crucial hub proteins.Our PPI networking and clustering analysis revealed that these hub proteins might have the potential to function in antibiotic-resistant mechanisms and/or pathways.

Gene Ontology, KEGG Pathway, and Protein Domain Analyses Reveal Significant Pathway and Domain Features.
To analyze the functions of the detected ARGs and further understand their resistance mechanism regulation, we performed GO assignment and enrichment and KEGG functional analyses.The detected ARGs were assigned to and enriched in one or more of three categories: biological process (BP), molecular function (MF), and cellular component (CC) with 3 to 45 gene counts (Figure 6(a)).The ARGs were significantly enriched (p < 0 05 and strength 0.11 to 1.54) in 24 GO terms.Of these enriched GO terms, 17 were related to BP whereas five to CC and two to MF, respectively (Figure 6(a)).The most significant and enriched GO terms of the BP were cellular process, response to stimulus, regulation of biological process, regulation of cellular process, and response to chemicals and antibiotics, and those of MF were transmembrane transporter activity and xenobiotic transmembrane transporter activity (Figure 6(a)).Besides, the top four significantly enriched GO terms of the CC were cellular anatomical entities, membrane, cell periphery, and cellular component : plasma membrane, suggesting that AMR genes are regulated by multiple cellular processes (Figure 6(a)).The two-component system, cationic antimicrobial peptide (CAMP) resistance, and beta-lactam resistance were the most enriched KEGG pathways related to the citrate cycle (TCA cycle), the breakdown of RNA, the metabolism of carbon, and the production of secondary metabolites (Figure 6(a)).These results and observations suggested that Shigella spp.were involved in regulating various metabolic pathways to develop antimicrobial resistance.
In order to understand the importance of domain associations in Shigella spp., high-frequently co-occurrent domains and their interactions were systematically identified, which were then interpreted for their contributions to bacterialspecific antimicrobial resistance (AMR) features.We calculated and filtered out the distinct pathways to generate a bubble plot to present the significant correlation.In this study, gene count ranged from 2 to 18, strength values varied from 0.   S3).Further investigation into the shared ARGs (n = 91) revealed that these ARGs conferred resistance against 24 distinct antibiotics belonging to 16 drug classes.We found a higher number of shared ARGs conferring resistance through efflux pump mechanism (n = 46) followed by antibiotic inactivation (n = 21), antibiotic target alteration (n = 16), antibiotic target replacement (n = 5), reduced permeability to antibiotics (n = 2), and antibiotic 7 BioMed Research International target protection (n = 1) mechanisms (Figure S4).In this study, we figured out four hub genes, e.g., tolC, acrR, mdtA, and gyrA, from 91 shared ARGs, and of these hub genes, tolC and mdtA operate antibiotic efflux mechanisms.In contrast, acrR and gyrA genes wield antibiotic target alteration mechanisms to inhibit antibiotic sensitivity (Figure S4).

Discussion
Shigella is the leading global etiological agent of shigellosis, a potential public health catastrophe globally.The prevalence of MDR and/or XDR Shigella spp. is increasing and becoming globally dominant in shigellosis [2,10,46].Moreover, Shigella spp.are on the priority list of the World Health Organization regarding antimicrobial resistance [47].Therefore, shigellosis is becoming critical day by day and necessitate new interventions for prevention, treatment, and control.The current global landscape of shigellosis is domi-nated by S. flexneri and S. sonnei, with S. sonnei and S. boydii clearly in the ascendency in many low-and middle-income countries [2,10,46].The advancement in WGS and the application of online bioinformatics tools for real-time detection of AMR determinants, ARGs, and mechanism of resistance are essential to identify effective control and prevention strategies to combat the increasing threat of AMR especially in bacterial diseases caused by MDR or XDR lineages [47,48].However, until now, no studies have explored ARGs and their mechanism of resistance, gene-drug interactions, PPI, and functional pathways in Shigella spp.By utilizing 45 WGS of four species of Shigella spp.(e.g., S. flexneri, S. dysenteriae, S. boydii, and S. sonnei), we identified four hub proteins (e.g., tolC, acrR, mdtA, and gyrA) that could be exploited to design and develop novel therapeutic candidate(s) against MDR or XDR lineages of Shigella spp.
The phylogenetic analysis with 45 genomes showed three distinct clades among the circulating strains of Shigella worldwide.However, no significant heterogeneity was observed 8 BioMed Research International among the circulating strains of Shigella.The genetic homogeneity of these clusters was evidenced by the high degree of overall nucleotide sequence similarity in the study genomes.
Our results are consistent with a previous study based on inferred bacterial coconserved networks based on phylogenetic profiles [49].By conducting an extensive search and utilizing bioinformatics analysis, we identified 2,146 ARGs across 45 genomes, averaging 47.69 ARGs per genome.Among these ARGs, only 91 ARGs were found to be shared by the four species of Shigella.We found that higher proportion of ARGs was related to fluoroquinolones, penems, cephalosporins, tetracyclines, aminoglycosides, and macrolide resistance.Antimicrobial resistance to Shigella spp. is a growing international concern, specifically with the international dominance of the MDR or XDR lineages [2,48].AMR determination by WGS approaches can complement traditional laboratory-based surveillance and provide direct insights into their evolution and transmission from one strain to another [50,51].The high throughput WGS data can help reveal the ARGs and their possible mechanism for drugs not being tested routinely or where the mechanisms of antimicrobial resistance are not yet identified [52,53].
We also sought to predict PPI network connectivity, ARG interactions, and hub genes which revealed significant clustering among the strains of four Shigella spp., with highly connected networks and node interactions.The PPI networks seem to be the most well-connected one with around 30 nodes, at least 60 connections per species, and relatively high clustering coefficient.Understanding the PPI networks among Shigella genomes and the strength of Four hub proteins (e.g., tolC, acrR, mdtA, and gyrA) were found to be shared in the five methods.9 BioMed Research International association between two or more proteins makes it easier to select drug targets.The high clustering coefficient indicates highly connected networks, while node degree denotes the number of interactions the network proteins have on average [54].The merged network, subjected to cluster analysis, showed that the proteins fall into three clusters with significant variations.Therefore, we suggest that the PPI networks that show a higher number of interactions could be more critical in analyzing the molecular-level interactions of ARGs as also reported earlier [28].The PPI network analysis performed using these ARGs showed that there were several unique genes conferring AMR in different strains of Shigella through different resistance mechanisms.We also detected four hub genes such as tolC, acrR, mdtA, and gyrA showing significant interactions.These genes play numerous essential roles in the metabolic, cellular, and biological processes that occur in microbes [55].Previously, some researchers described that tolC, acrR, mdtA, and gyrA genes in Shigella strains plied different resistant mechanisms to survive [2,4,7].The overexpression of efflux pump-related genes (e.g., acrR, TolC, and mdtA) may cause an overall decreased accumulation of antibiotics inside the bacterial cells, further resulting in decreased susceptibility and development of MDR or XDR phenomena [2,56].Moreover, mdtA is a novel immunogenic Shigella protein responsible for the efflux pump-mediated AMR in Shigella [22,57].In addition, acrR and gyrA gene expression implied an antibiotic target alteration and reduced antibiotic sensitivity [58].The mutations in bacterial targets gyrA and gyrB encoding for DNA gyrase and topoisomerase IV cause a change either in the target structure or its binding strength, resulting in less susceptibility, and increased minimum inhibitory concentrations [4].We found that there was a strong link between these hub genes and different types of drug resistance mechanisms and drug classes.These mechanisms can be carried out by changing the drug, changing the antimicrobial targets, limiting access to the target, or a combination of these things [4,13].Moreover, predicted ARGs can easily spread through their host bacteria to different host inhabitants of other ecosystems [53,59].
One of the hallmark findings of this study is the prediction of both ARG repertoire and correlated biological functions (e.g., biological processes (BP), cellular components (CC), and molecular functions (MF)) through which different strains of Shigella develop AMR.Gene Ontology enrichment analysis of the ARGs in Shigella genomes showed that majority of the shared ARGs represented biological functions involved mainly in BP, MF, and CC.The cellular metabolic process of bacteria determines their resistance to antibiotics; hence, the metabolic condition of bacteria could be modified to boost therapeutic efficacy [60].Disruptions to the bacterial metabolic balance also significantly affect treatment plans [60,61].We found that two-component system, primary metabolic process, and nitrogenous compound metabolism were the most enriched pathways in different strains of Shigella.Our results therefore suggest that Shigella spp.were involved in regulating various metabolic pathways, which might play a significant role in the development of resistance to multiple antibiotics.For instance, the twocomponent regulatory system plays a substantial role in the pathogenicity, virulence, biofilm formation, and drug resistance in bacterial pathogens, including Shigella [62].
Another important finding of the current study is that we found significant correlation across the detected ARGs, drug class, and resistance mechanisms.We identified that Shigella spp.developed AMR through efflux pump mechanism, antibiotic inactivation, antibiotic target alteration, antibiotic target replacement, reduced permeability to antibiotics, and antibiotic target protection mechanisms.Various antibiotic resistance mechanisms in bacterial pathogens include poor drug penetration into the cell, efflux of antibiotics by efflux pumps, target modification by mutation, and hydrolysis of antibiotics [13,63].It has been reported that efflux pumps regulated by two-component systems in several bacterial pathogens provide multidrug resistance, which may limit the treatment options against bacterial infections like shigellosis.Furthermore, antibiotic accumulation is known to be one prominent feature of bacterium tolerance.Our findings showed that efflux pump, antibiotic inactivation,   11 BioMed Research International antibiotic target alteration, replacement, protection, and reduced permeability to antibiotics were the major activities playing significant roles in AMR development in Shigella spp.Our findings also suggest that specific pumps are involved in tolerance maintenance, which is a new concept in antibiotic tolerance studies.Therefore, a crucial aspect is the thorough characterization of sources, reservoirs, mechanisms, and networks involved in the potential transmission of AMR and ARGs among humans, animals, and the environment [64].This comprehensive understanding is pivotal for devising interventions that are both efficient and effective in addressing pathogens resistant to antimicrobials [13].

Conclusions
Antimicrobial resistance in human pathogenic bacteria such as Shigella is a complex and multifaceted challenge.We unveiled noteworthy genomic characteristics in MDR Shigella spp., including S. flexneri, S. dysenteriae, S. boydii, and S. sonnei.These findings have implications for devising novel therapeutic approaches aimed at preventing, treating, and controlling shigellosis.Evolutionary phylogenetic analysis revealed three distinct clades in 45 strains of Shigella, with less genomic diversity.We found a significant correlation among ARGs, their resistance mechanisms, and drug classes in the genomes of four Shigella species.Several GO keywords and KEGG pathways correlated with MDR were identified during the functional enrichment analysis of the ARGs.In Shigella spp., ARGs and their functional interactions most commonly expressed included antibiotic resistance through the inactivation of antibiotics, antibiotic efflux pump, target alteration, reduced permeability to antibiotics, and target replacement of antibiotics.Numerous antibiotics such as cephalosporins, penems, fluoroquinolone antibiotic, tetracyclines, aminoglycoside antibiotics, monobactam, carbapenems, macrolides, and sulfonamides are found to be ineffective to Shigella spp.Besides, PPI clustering revealed that some of the ARG sets are closely connected to develop resistance in Shigella spp.We identified four unique hub genes (e.g., tolC, acrR, mdtA, and gyrA) among the ARG repertoire of Shigella, and these hub genes could be used as potential therapeutic candidates and aid in developing new drugs.Therefore, Shigella spp.harbored a higher number of ARGs encoding for different resistance mechanisms, which must be considered for further research.Our findings will thereby furnish researchers with a firm foundation to develop hypothesis to predict clinically significant determinants of antibiotic resistance.This will, in turn, support investigations into innovative therapeutic strategies for effectively managing shigellosis outbreaks on a global scale.

Figure 1 :
Figure 1: The evolutionary phylogenetic relationships among different strains of Shigella.Genomes of 45 Shigella strains (S. flexneri, n = 17;S.dysenteriae, n = 14; S. boydii, n = 11; and S. sonnei, n = 13) were used to construct the phylogenetic tree applying maximum-likelihood method with 1000 bootstraps.The tree topology partitions the isolates into three distinct clades (clade 1-clade 3).Each node in the phylogenetic tree represents an isolate while the edge represents the hamming distance between two isolates.

Figure 2 :
Figure 2: Protein-protein interaction (PPI) network connectivity and gene interaction network of the most significant shared ARGs with their interacting partners.The network contains (a) 38 nodes with 90 edges for S. boydii, (b) 29 nodes with 57 edges for S. dysenteriae, (c) 35 nodes with 108 edges for S. flexneri, and (d) 33 nodes with 67 edges for S. sonnei.Edges represent the protein-protein associations.Node size indicates node degree value, and edge colour and shape represent Shigella spp.

Figure 3 :
Figure 3: Clustering of protein-protein interaction (PPI) networks.The PPI networks were merged into three clusters: (a) the 1 st cluster comprised 8 nodes with 94 edges, (b) the 2 nd cluster comprised 7 nodes with 39 edges, and (c) the 3 rd cluster consisted of 5 nodes with 16 edges.These clusters had the highest scores among the clusters.

Figure 5 :
Figure 5: Visualization of the hub genes using five different calculation methods.(a) Bar plot to identify significant top 13 hub genes using five intersecting algorithms (betweenness, closeness, degree, MCC, and MNC) for differentially expressed ARGs in Shigella genomes.(b) Four hub proteins (e.g., tolC, acrR, mdtA, and gyrA) were found to be shared in the five methods.
RND efflux pump, membrane fusion protein, barrel-sandwich domain Multidrug efflux transporter AcrB TolC docking domain, DN/DC subdomains HTH domain AraC-type, conserved site DNA binding HTH domain, AraC-type Acriflavin resistance protein Major facilitator superfamily domain Major facilitator superfamily HlyD membrane-fusion protein of T1SS Helix-turn-helix domain Major facilitator superfamily MMPL family AcrB/AcrD/AcrF family Bacterial regulatory helix-turn-helix proteins, AraC family Barrel-sandwich domain of CusB or HlyD membrane-

Table 1 :
Protein-protein clustering information of Shigella genomes.

Table 2 :
Topological information of hub proteins as determined by five distinct algorithms.