Subtractive genome mining in Xanthomonas citri pv. citri strain 306 for identifying novel drug target proteins coupled with in-depth protein-protein interaction and coevolution analysis - A leap towards prospective drug design

doi:10.21203/rs.3.rs-3933880/v2

Citrus canker poses a serious threat to a highly significant citrus fruit crop, this disease caused by one of the most destructive bacterial plant pathogens Xanthomonas citri pv. citri (Xcc). Bacterial plant diseases significantly reduce crop yields worldwide, making it more difficult to supply the growing food demand. The high levels of antibiotic resistance in Xcc strains are diminishing the efficacy of current control measures, necessitating the exploration of novel therapeutic targets to address the escalating antimicrobial resistance trend. Genome subtraction approach along with protein-protein network and coevolution analysis were used to identify potential drug targets in Xcc stain 306. In this study we used genome mining. 750 essential proteins were identified that are nonhomologous to citrus plant. Subsequent analyses included metabolic pathway assessment, subcellular localization prediction, and druggability analysis. Protein network analysis, coevolution analysis, protein active site identification was also performed. In conclusion, this study identified eight potential drug targets, highlighting RpoN, FleQ, and SecB as unprecedented targets for Xcc. These findings may contribute to the development of novel antimicrobial agents in the future that can efficiently control citrus canker disease.

Citrus canker

Xanthomonas citri pv. citri

Subtractive genomics

Coevolution

Druggability

Drug target proteins

Citrus is one of the world’s most important fruit crops and is cultivated throughout the world having substantial value in both fresh and processed fruit industries (Martins et al. 2020). Instead of having good flavour, Citrus fruits are rich resources of secondary metabolites like flavonoids, including hesperetin, naringenin, isosakuranetin, and eriodictyol, having potential anti-inflammatory and antibacterial and anticancer properties (Khan et al. 2014; Rathod et al. 2023). The global population anticipated to reach 9.6 billion by 2050 and consequently needs more food, requiring a potential 70% increase in the global food supply to meet up the demand (Searchinger et al. 2019). Therefore, it necessitates sustainable food sources and an around 80–110% more crop production to feed the entire population (Albahri et al. 2023). The main constraint of food production and its actual utilisation rely on many factors among which crop loss is of great concern. Different plant diseases alone attribute to approximately 10% of global food production loss and therefore, warrants efficient disease management system to avoid the wastage (Košćak et al. 2023). Citrus fruits, cultivated and produced roughly around 140 countries, play a very important role to the global economy (Syed-Ab-Rahman et al. 2022). Citrus cultivation in India spans 1055 thousand hectares of land, yielding approximately 12746 thousand tonnes of production (Kumar et al. 2020). Citrus canker has emerged as a serious threat to Indian citrus cultivation causing substantial production losses and deterioration of quality (Singh et al. 2019). The growth and yield of citrus cultivation are affected by many plant pathogens and Citrus Bacterial Canker (CBC) is the most dreaded and contagious disease of this fruit crops, which has a significant negative economic impact on the agricultural industry and impedes trade as well as cultivation. It affects many related rutaceous plants and all commercial citrus types (Naqvi et al. 2022). As of now there are mainly three distinct types of CBC have been reported such as type A, B, C and each of the type of diseases caused by different pathovar and variants of the strains of Xanthomonas. Xanthomonas citri subsp. citri (Xcc), formerly known as X. axonopodis pv. citri, is the causative agent of CBC type A and it the most devastating and widespread form of the disease (Yasuhara-Bell et al. 2023). Xanthomonas fuscans subsp. aurantifolii strain B and Xanthomonas fuscans subsp. aurantifolii strain C are the causative agent of CBC type B and C respectively (Ali et al. 2023). Xanthomonas citri pv. citri is a rod-shaped, slender, gram-negative bacterium with a single polar flagellum, member of the family Xanthomonadaceae, one of the most important groups of bacterial phytopathogens. It usually ranges in size from 1.5–2.0 × 0.5–0.75 μm (Ference et al. 2018). Xanthomonas is a large genus, which belongs to the class of Gammaproteobacteria, containing at least 27 official species, many of which also possess several pathovars. The host range of this genus is broad and capable of infecting around 400 plant (124 monocot species and 268 dicot species) hosts including citrus fruits, rice, tomato, cabbage, beans, banana, pepper and provide excellent case studies for understanding plant-microbe interactions (Ference et al. 2018). Biofilm formation during disease development by Xanthomonas citri pv. citri appears to be very frequent (Conforte et al. 2019) and it facilitates in the infection process of citrus canker (An et al. 2020). Furthermore, the development of biofilm formation is regulated by the two-component regulatory system, which is essential for managing their pathogenicity (Yan and Wang 2011). Copper-based bactericides (i.e., copper hydroxide, copper oxychloride, and copper oxide) have been used extensively for treating Citrus canker for quite a long time (Ference et al. 2018). Applying streptocycline with copper oxychloride (0.1%) ideally every 7 to 15 days has been found to be particularly effective against citrus canker disease (Talibi et al. 2014). As a result, various copper-resistant (Cur) strains from different bacterial species that damage plants have evolved. Application of neem powder solution with 100 ppm of streptomycin and 0.3% of copper oxychloride to trimmed infected twigs have been a strategic disease management in Citrus (Luiz et al. 2021). Recently, the emergence of both copper resistant (Behlau et al. 2011; Izadiyan and Taghavi 2023) and streptomycin resistant mutants (Sundin and Wang 2018) rendered the conventional management ineffective to control the diseases. Moreover, European authorities have banned streptomycin due to the potential of antibiotic resistance developing in non-target bacteria, in addition to the emergence of strains resistant to the antibiotic as a result of frequent spraying (Rabbee et al. 2019). The rising incidence of antimicrobial resistance to the conventional antibiotics used to treat Citrus canker therefore offering a potential threat to Citrus production. Since the incidence of citrus canker is expanding together with its high cost of management necessitates to find a more efficient alternative way to control the diseases (Naqvi et al. 2022). In this study, a subtractive proteomic approach on the whole proteome of Xanthomonas citri pv. citri strain 306 has been applied to identify essential and non-homologous proteins with unique metabolic pathways in the pathogen that could serve as novel drug targets offering a strong foundation for the bacterial disease management in the future. A number of databases, computational software tools, and bioinformatics approaches have been employed to leverage omics data such as proteomics, metabolomics, and genomics to determine the key proteins essential for the pathogen's survival. Rather than focusing on just one drugs target protein, the study incorporated all of the top-ranked target proteins identified by computational analysis. which is a very unique case of protein mining study in Xanthomonas citri pv. citri, for therapeutic purpose.

Drug targets were identified in Xanthomonas citri pv. citri strain 306 utilizing a comprehensive computational approach. The process of identification has been described in detail.

Retrieval of the bacterial proteome

The whole proteome of Xanthomonas citri pv. citri strain 306 was retrieved from the UniProt (Universal Protein Resource) database (The UniProt Consortium 2021) in FASTA format. Uniprot is world’s most prominent, complete and freely available resource of protein sequences with functional annotation.

Removal of paralogous proteins

Paralogous proteins were eliminated by using CD-HIT tool. Sequence identity cut-off was maintained at 0.8 (80% identity) as sequence having identity > 80% may exhibit similar/related structure and functions (Huang et al. 2010).

Selection of non-homologous proteins to host

To identify nonhomologous sequences of all the selected proteins were screened by using BLASTp (Basic Local Alignment for Protein) against the NCBI (National Centre for Biotechnology Institute) database with a threshold of an e-value of <0.005 (Altschul et al. 1997).

Screening of essential proteins

The non-homologous proteins comprising essential genes responsible for the biological activities of the bacteria considered to be potential druggable targets. With an e-value cutoff of 0.00001, hypothetical proteins were searched using BLAST against the Database of Essential Genes (DEG) (Luo et al. 2014).

Analysis of metabolic pathway

The Kyoto Encyclopaedia of Gene and Genome (KEGG) is a pathway database that enables the identification of the pathogen's non-homologous proteins' metabolic pathways (Kanehisa et al. 2002). To find the pathways that are unique to the pathogen, the identified metabolic pathways of the host and pathogen were manually compared using this database. Additionally, the proteins were divided into groups based on their roles sharing common pathways found in both pathogens and hosts and unique pathways specific to pathogens. Specific proteins found exclusively in the pathogen's distinctive metabolic pathways were meticulously selected for additional analysis, while those associated with shared pathways or solely involved in host pathways were discarded.

Table 1 List of potential drug targets and the pathways involved to these target proteins

Sl. No.	UniProt Accession no.	Name	Pathways involved	Function
1.	Q8PQV0	secB (Pre-protein translocase subunit)	Protein export (xac02060), Bacterial secretion system(xac02070), Quorum sensing (02024)	Post-translational translocation of proteins.
2.	Q8PL39	fleQ (Transcriptional regulator)	Two component system (02020), Flagellar assembly (xac02040)	Involved in flagellar biosynthesis, EPS production, biofilm formation.
3.	Q8PL37	rpoN (RNA polymerase sigma-54 factor)	Flagellar assembly (xac02040), Two component system (02020)	Transcription regulation of many proteins in pathogenic bacteria.
4.	Q8PEX8	gspE (General secretion pathway protein)	Bacterial secretion system (03070)	Translocation of unfolded precursor proteins across the cytoplasmic membrane.
5.	Q8PGH2	glmU (UDP-N-acetylglucosamine pyrophosphorylase)	O-Antigen nucleotide sugar biosynthesis (xac00541), Biosynthesis of nucleotide sugars (xac01250), Metabolic pathways (xac01100), Amino sugar and nucleotide sugar metabolism (xac00520)	Catalyzes the conversion of uridine monophosphate from uridine-triphosphate to acetylglucosamine-1-phosphate, playing a crucial role in the biosynthesis of lipopolysaccharide and peptidoglycan.
6.	Q8PL99	cheA (histidine kinase)	Bacterial chemotaxis (xac02030), Two-component system (02020), Biosynthesis of secondary metabolites (xac01110), Metabolic pathways (xac01100)	Plays a crucial role in bacterial chemotaxis.
7.	Q8PGN4	rmlD (dTDP-4-dehydrorhamnose reductase)	Streptomycin biosynthesis(xac00521), O-Antigen nucleotide sugar biosynthesis (xac00541), Biosynthesis of nucleotide sugars (xac01250), Polyketide sugar unit biosynthesis(xac00523), Metabolic pathways (01100), Biosynthesis of secondary metabolites (01110)	Plays a vital role in the biosynthesis of dTDP-L-rhamnose and involves in the biosynthesis of various cell surface polysaccharides and glycolipids.
8.	Q8PME1	Shk (Sensor histidine kinase)	Bacterial chemotaxis(xac02030), Two component system (xac02020)	Involved in primary signal transduction pathway, two component system.

Prediction of subcellular localization

To predict subcellular localization of the shortlisted proteins, PSORTb v3.0.2 server (Yu et al. 2010) was used and the result was also compared using another server called CELLO v.2.5 (http://cello.life.nctu.edu.tw/) (Yu et al. 2006). Cytoplasmic, inner membrane, periplasmic, outer membrane, and extracellular are the five main localizations of the proteins of gram-negative bacteria. Subcellular location is crucial for a particular protein to understand its function and physicochemical characteristics. Proteins found in the cytoplasm and at the membrane channel were chosen for this study because they are more likely to be effective and serve as good druggable targets (Barh et al. 2011).

Druggability analysis

The selected proteins were further analysed with the database of DrugBank 5.1.0 (Wishart et al. 2018) using default parameters. 14,666 drug entries are included in the most recent release of DrugBank Online (version 5.1.9, released 2022-01-03), including 2,722 approved small molecule drugs, 1,533 approved biologics (proteins, peptides, vaccines, and allergenics), 132 nutraceuticals, and more than 6,692 experimental (discovery-phase) drugs.

Protein interaction network analysis

In this study, a network-oriented approach was utilized for the identification of drug targets, aiming to reconstruct the regulatory, endogenous metabolic, and signalling networks that interface with these potential drug targets. By using protein network database STRING, previously selected non-homologous proteins were checked for any potential metabolic connections between them so that we were able to find the suitable therapeutic targets (Szklarczyk et al. 2015).

Co-evolution analysis

In the exploration of co-evolution among nonhomologous proteins, the web-based CoeViz tool (Baker and Porollo 2016) was utilized. The server employs three metrics-Mutual Information, Chi-Square Statistic, and Pearson Correlation as well as one conservation metric, joint Shannon entropy, to calculate pairwise coevolution scores. Subsequently, CastP 3.0 (Tian et al. 2018) server was employed to identify the active sites within proteins or the amino acids located in various protein pockets. Additionally, the centrality betweenness plot of specific target proteins including GlmU, CheA, RmlD, GspE, FleQ, RpoN, Shk, SecB were generated using ProsNex server (Aydınkal et al. 2019).

The aim of this study was to identify potential druggable target of the bacterial plant pathogen Xanthomonas citri pv. citri strain 306. Total proteome containing 4372 proteins were retrieved from UniProt database. The proteins were further analysed using subtractive genomic approaches with other comprehensive computational analysis. The summary of workflow and analysis is illustrated in Fig. 3.

Non homology analysis

Non-homology analysis was the initial step in the subtractive genomics approach. Homologous proteins that are present in the plant host may interact with chemicals and convey undesirable toxic effects. To avoid this problem non homologous protein were selected for further analysis. At first, paralogous proteins were removed by CD-HIT tool and non-paralogous 4321 proteins were selected at 80% threshold. By subjecting the protein sequences to BLASTp against Rutaceae proteome with an e-value of 10^-3, non-homologous proteins were chosen as potential therapeutic targets.

Essential genes of the organism

The BLASTp tool of DEG database was used to find out the essential genes of the organism at default parameter settings. It is also crosschecked using NetGenes (Senthamizhan et al. 2021) and ePath (Kong et al. 2019), databases and total 750 essential genes were chosen for further analysis.

Metabolic pathway analysis

The analysis involved all 750 essential proteins, utilizing the KEGG pathway database. Out of these, 96 proteins were identified to be associated with distinctive pathways (Table S1) specific to plant pathogenic bacteria, and notably, these unique proteins were absent in the plant host.

Subcellular localization

The 96 non homologous unique proteins those were selected screened for subcellular localization prediction using pSORTb server. Out of these, 30 proteins were found to be located in cytoplasm of the pathogenic organism.

Druggability analysis

Thirty non-homologous unique proteins were assessed for their druggability through similarity searches against targets present in the Drug Bank database. Following a comprehensive examination of these parameters, a total of eight proteins were finally identified as potential drug targets, and these selected proteins were advanced for additional analysis, as outlined Table 1.

Table 2 Presence of potential drug targets in the drug bank database

Potential Drug Targets	Presence in Drug Bank	Drug IDs	Drug Group
Bifunctional protein (glmU)	Yes	DB03397	Experimental
General secretory pathway protein (gspE)	Yes	DB04395	Experimental
RNA polymerase sigma-54 factor (rpoN)	No	*	*
Transcriptional regulator (fleQ)	No	*	*
Histidine kinase (cheA)	Yes	DB04414, DB04494	Experimental
dTDP-4-dehydrorhamnose reductase (rmlD)	Yes	DB03723	Experimental
Sensor histidine kinase (shk)	Yes	DB03758	Experimental
Export protein (secB)	No	*	*

The asterisk (*) denotes the absence of these drug targets in the DrugBank database.

Protein network analysis

A protein interaction network analysis was carried out for the eight selected target proteins using the STRING database to determine the functional relationships of target proteins with other important proteins. Proteins having a higher confidence score (c score) were considered to be highly interacting metabolic proteins. The strength of the interactions between proteins increases with a greater correlation value. A confidence score of 1 indicates the highest level of confidence, with increased reliability corresponding to higher c scores. The c scores are on a scale from 0 to 1. In this study, twenty-six highly interacting enzymes or proteins were found to have a confidence score greater than 0.8. These 26 enzymes, detailed in Table S2, play crucial roles and are linked to selected eight target proteins through diverse pathways (Table 3). Potential drug target proteins including SecB, FleQ, and RpoN exhibit extensive interactions with various metabolically significant proteins in the pathogenic organism Xcc (Fig. 1)

Table 3 List of 26 proteins that exhibit high interaction with eight selected drug target proteins

Sl.No.	Non-homologous proteins	Functional partners	Name of the proteins	C score
1	GlmU	mrsA	Phosphoglucosamine mutase	0.989
		murA	UDP-N-acetylglucosamine 1-carboxyvinyltransferase	0.975
		bplD	UDP-N-acetylglucosamine 2-epimerase	0.937
		glmS	Glucosamine fructose-6-phosphate aminotransferase	0.931
		lpxA	UDP-N-acetylglucosamine acyltransferase	0.917
2	GspE	pilC	Fimbrial assembly protein	0.952
		xpsF	General secretion pathway protein F	0.876
		xcsF	Type II secretion system protein F.	0.872
		pilQ	Fimbrial assembly protein.	0.832
		xpsE	Type II secretion system protein	0.826
		xcsE	Type II secretion system protein	0.824
3	RpoN	fleQ	Sigma-54 dependent transcriptional regulator,	0.978
		pilR	Two-component system regulatory protein.	0.951
		ntrC	Two-component system, nitrogen-regulated protein	0.857
		styS	Histidine kinase-response regulator hybrid protein.	0.823
		flbD	Transcription regulator.	0.812
		prpR	Propionate catabolism regulatory protein	0.805
		pilS	Two component system	0.804
4	FleQ	rpoN	RNA polymerase sigma-54 factor; initiation factor.	0.978
		filC	Flagellar protein.	0.871
		vioA	Nucleotide sugar transaminase	0.812
		fliA	RNA polymerase sigma factor	0.803
5	CheA	cheW	Chemotaxis protein.	0.999
		cheZ	Chemotaxis related protein;	0.999
		cheV	Chemotaxis protein.	0.999
		CheB1	Two component system, chemotaxis family.	0.996
		cheY	Two component system, chemotaxis family.	0.996
6	RmlD	rmlC	DTDP-4-dehydrorhamnose 3,5-epimerase	0.999
		rmlB	DTDP- glucose 4,6 dehydratases.	0.999
		rmlA	Glucose 1 phosphate thymidylyltransferase.	0.998
7	Shk (XAC1488)	XAC2054	Two-component system sensor protein.	0.952
		vieA	Response regulator.	0.865
		XAC2492	Two-component system sensor protein.	0.860
		XAC1992	C-di-GMP phosphodiesterase A	0.845
8	SecB	secA	Preprotein translocase subunit secA	0.999
		secG	Preprotein translocase subunit secG	0.981
		secE	Preprotein translocase subunit secE	0.980
		ftsY	Membrane translocase protein	0.979
		secY	Preprotein translocase subunit secY	0.973
		gpdA	Glycerol 3 phosphate dehydrogenase	0.938
		ffh	Signal recognition particle subunit srp54	0.930
		tig	Involved in protein export.	0.832

Coevolution Analysis

Coevolution analysis was carried out by using a web-based tool, CoeViz that provides a versatile analysis and visualization of pairwise coevolution of amino acid residues. Using this server, heatmaps were generated for all the target proteins (Fig. S1 and Fig. S2). In the heatmap, individual squares depict the coevolution score between two amino acid sites, often arranged as a grid of squares or rectangles. The colour of each square corresponds to the strength of the coevolutionary relationship, typically employing shades of blue in colour schemes. Darker hues signify stronger coevolution, while lighter shades indicate weaker coevolution, suggesting potential independence in both structure and function for those positions. Shannon's Entropy calculation gauges how quickly evolution occurs but doesn't unveil details about the pairwise interactions among residues over time. To maintain the protein's overall structure, mutations in non-conserved locations might prompt compensatory substitutions elsewhere. These compensatory mutations establish interdependencies between positions. The degree of positional interdependency, indicating a stronger co-evolutionary signal in the sequence space, correlates with larger Mutual Information (MI) values, as highlighted by the findings of Sanyal et al. (2022). The region of the MI heat map of FleQ shown in Fig. 2 that is heavily dense (positions 200-320, denoted by the deep coloured region) represents the sub-region with the highest positional inter-dependent residues. Intrigued by the MI heat map we then intended to correlate with the active site amino acids residue for a comprehensive understanding of evolutionarily relationship. We went on to predict the active sites of the proteins using CastP 3.0 server (Table S2). Interestingly, it has been observed that the active site amino acid positions 220, 221, 222, 223, 227, 228, 230, 237, 306, 307, 310, 311, 312, 313, 314, 315, 319 of protein FleQ as predicted by CastP aligns to the MI heat map sub region (Figure 3a). Furthermore, an understanding of amino acid sequence conservation in the context of protein-protein interaction networks or structural characteristics can be gained from centrality betweenness plots created from a protein structure. By linking together several parts of a network, betweenness centrality in network analysis measures the significance of a node or an amino acid residue in the protein which may unveil its intricacy in the functionality of a protein. A protein is thought to interact with other molecules, such as ligands, DNA, or RNA, at conserved interaction sites represented by amino acids with high betweenness centrality. These positions are most likely both functionally significant and conserved throughout evolution. Betweenness centrality may also accentuate residues lying at the interfaces or boundaries of protein domains or structural motifs. High betweenness centrality residues may occasionally be implicated in allosteric control or conformational changes. Seamless transition of the protein to different functional states, thought to be attributed by the conserved residues which also contribute to the integrity of the structural features. Therefore, High betweenness centrality (Figure 4) amino acids are considered as functional hotspots. Highly conserved sequences in these areas are likely significant to the overall function of the protein. In this study, the amino acids of protein FleQ, located at positions 224, 225, 229, 232-239, 252, 253, 262, 272, 275, 276, 277, 278, 285, 286, 299, 300, 306, 308, 309, 313, 315, 319 exhibit higher centrality betweenness value (more than 0.5) and it was observed that there is a coherence between amino acids of the coevolution heatmap generated by CoeViz and centrality betweenness plot using ProSNex server (Aydınkal et al. 2019) and notably these amino acids were also present in the active sites of the proteins (Table S3) or pocket number generated by CastP web server (Tian et al. 2018) further confirming their role in metabolic functioning. Similar results were observed for the remaining seven target proteins, with their respective amino acids demonstrating elevated centrality betweenness values. The centrality betweenness plots for the remaining proteins is available in the supplementary files provided (Fig. S3 and Fig. S4).

The identification of potential drug targets by subtractive genome analysis has been a recent trend and has been widely used to identify putative drug targets in variety of plant and human pathogens. Many target proteins have been identified in Ralstonia solanacearum, Bacillus anthracis, Mycobacterium tuberculosis, Moraxella catarrhalis and to Helicobacter pylori through subtractive genome analysis (Keshri et al. 2014; Rahman et al. 2014; Ibrahim et al. 2020; Mandal et al. 2021; Kaur et al. 2021b; Ashraf et al. 2022). The target proteins are crucial and unique for the pathogen's survival inside the host plant and share no homology with the host plant protein. This will reduce the negative effects of a drug on the host (Solanki et al. 2019). Up till now most of the approaches pertaining to novel antimicrobial targets of plant pathogenic bacteria mainly focused on peptidoglycan and lipopolysaccharide biosynthesis, two component system, secretion system, bacterial motility (Keshri et al. 2014; Mandal et al. 2021). This limited therapeutic strategy targeting to specific pathways of pathogenic bacteria prompted antibiotic resistance in the bacterial pathogens (Stephenson and Hoch 2002; Sundin and Wang 2018). Therefore, considering the limitation in current therapeutics, our study employed a wider approach by including all nonhomologous proteins and tried to tap into their therapeutic potentiality as prospective drug targets. Total 30 nonhomologous unique proteins were selected in this present study and after druggability analysis finally eight proteins are selected for protein network analysis and coevolution study. In the current study, a total of eight proteins were ultimately selected from a pool of 30 nonhomologous unique proteins and being subjected to protein network analysis and coevolution study after assessing druggability analysis. Keshri et al. 2014 and Mandal and Sinha (2021) also used similar approach for the identification of potential antimicrobial targets in Xanthomonas oryzae pv. oryzae PXO99A and Ralstonia solanacearum GMI 1000 strain respectively. Nevertheless, they did not carry out coevolution study. Proteins found in the cytoplasm are more likely to play specific roles in essential cellular processes and tend to be highly conserved, making them attractive drug targets. However, membrane bound, extracellular proteins are more variable among strains and are suitable for vaccine targets (Butt et al. 2012; Sudha et al. 2019). In this present study, the location of the eight potential drug target proteins were found to be cytoplasmic and this result also compared with CELLO v.2.5 server (Table 4). We subsequently observed that the two proteins viz., SecB export protein and Sensor histidine kinase had shown discrepancy in their cellular localizations. According to pSORTb both the proteins predicted to be cytoplasmic proteins, while cello identified them as periplasmic proteins.

Table 4 Cytoplasmic location of eight target protein using pSORTb and CELLO

Gene names	Name of the proteins	Cellular Location
Gene names	Name of the proteins	pSORTb	CELLO
glmU	Bifunctional protein	Cytoplasmic	Cytoplasmic
gspE	General secretory pathway related protein	Cytoplasmic	Cytoplasmic
rpoN	RNA polymerase sigma-54 factor	Cytoplasmic	Cytoplasmic
fleQ	Transcriptional regulator	Cytoplasmic	Cytoplasmic
cheA	Histidine kinase	Cytoplasmic	Cytoplasmic
rmlD	dTDP-4-dehydrorhamnose reductase	Cytoplasmic	Cytoplasmic
shk	Sensor histidine kinase	Cytoplasmic	Cytoplasmic and Inner membrane
secB	Export protein	Cytoplasmic	Periplasmic

Out of these, 30 non homologous unique proteins were further examined by comparing them to targets listed in the Drug Bank database via a similarity search. Following a manual search in the Drug Bank it was revealed that target molecules against the GlmU, GspE, CheA, RmlD, and Shk proteins (Table 2) are available and these drug molecules are all in the experimental stage of the drug development process. However, there are no homologs for the other three proteins (SecB, FleQ and RpoN) in the Drug Bank, therefore these three proteins were designated as novel drug targets. SecB engages nascent proteins in all cellular compartments, facilitating membrane retargeting with SRP-mediated ribosomal docking. SecB remains involved throughout translation and membrane integration. Initially, TF or trigger factor engages nascent chains in the cytoplasm, periplasm, and outer membrane, after which SecB takes over, promoting either folding or translocation (Eismann et al. 2022). FleQ plays a critical role in regulating the expression of genes involved in motility, attachment, and biofilm formation in Xanthomonas citri pv. citri (An et al. 2020). A key regulator in many pathogenic bacteria, sigma-54 has been connected to various cellular functions, including motility, pathogenicity, and the production of biofilms, quorum sensing and other virulence factors as well as crucial activities like nitrogen assimilation (Francke et al. 2011). In a previous deletion mutation study by a group of researchers proved that RpoN protein is required for full virulence in citrus plants (Gicharu et al. 2016). SecB, FleQ and RpoN have been identified as potential drug targets in phytopathogenic bacteria for the first time in this study. To garner more insights about the proteins which are highly interacting with the selected drug target proteins, we identified 26 important proteins with a C-score exceeding 0.8 by using STRING database. The functions of these proteins are crucial for the organism's survival. Therefore, blocking these targets could impede the functions of the associated proteins. The understanding of protein interactions was carried out through a coevolution study conducted using the CoeViz server. Proteins with high C-scores and identified coevolving residues are prospective targets for additional experimental validation (Baker and Porollo 2016).

Proteins participating in unique pathways may exhibit homology with human and host citrus plant (Khan et al. 2022). Previously, it has been observed that these proteins are non-homologous to citrus host plant proteins. Comparing protein sequences to human protein databases became an essential next step in the development of new drug against the plant pathogenic bacterium. By employing BLASTP tool in NCBI, we identified all the eight final drug target proteins have no similarities with human proteome. These findings may illuminate that subsequent to the development of specific antimicrobial compounds, they might exhibit a lack of binding affinity with human proteins this holds significant implications for drug discovery. Adopting the similar vein, Kaur et al. 2021 had identified the potential drug target proteins in highly virulent uropathogenic Escherichia coli strain CFT073 by protein-protein interaction network analysis. These methods, along with the development of computational biology and a wide range of bioinformatics applications, have completely transformed the drug design field by cutting off the time and cost associated with the in vivo and wet-lab testing that is essential for effective drug development (Lin et al. 2020). Therefore, the selection of eight drug target proteins in this study offers streamline options for subsequent experimental validation.

The current investigation employs subtractive genome analysis to identify potential drug targets in phytopathogenic bacteria, adopting a comprehensive approach that considers all nonhomologous proteins. We selected 30 unique proteins, subjected them to druggability analysis, and ultimately identified eight potential drug targets (GlmU, CheA, RmlD, GspE, FleQ, RpoN, Shk, SecB). Notably, three proteins- SecB, FleQ, and RpoN emerged as novel drug targets with no homologs in the Drug Bank. The study emphasizes the importance of addressing limitations in current therapeutic strategies, which often focus on specific pathways and contribute to antibiotic resistance. By broadening the scope to include all nonhomologous proteins, we aim to tap into their therapeutic potentiality and mitigate the negative effects on host organisms. The cytoplasmic localization of selected targets and their unique characteristics, as validated by previous studies, reinforce the novelty of our findings. The potential drug targets have also been analysed through coevolution studies, presenting promising avenues for further experimental validation. Our study revealed potential therapeutic targets that are essential for treating citrus canker. This methodology can be used to other multi drug resistant plant diseases to identify potential targets for treatment.

Author Contributions

Tumpa Mahato, Jayanta Mandal, Sangram Sinha designed the study. Tumpa Mahato performed most of the analysis. Jayanta Mandal assisted in coevolution analysis. Eilita Chatterjee and Satyabrata Bhattacharya helped to prepare the tables. Tumpa Mahato and Sangram Sinha wrote and edited the manuscript. All authors contributed to the interpretation of results and critical revision.

Declarations and statements

Acknowledgements The authors would like to express their sincere gratitude to the University of Burdwan and Vivekananda Mahavidyalaya, Haripal, West Bengal, India for offering essential support and encouragements.

Data availability Available on request.

Funding No funding was received for conducting this study from any funding agencies.

Competing interests The authors affirm that they have no conflicts of interest that could potentially influence the content of this article.

Albahri G, Alyamani AA, Badran A, Hijazi A, Nasser M, Maresca M, Baydoun E (2023) Enhancing essential grains yield for sustainable food security and bio-safe agriculture through latest innovative approaches. Agronomy 13:1709. https://doi.org/10.3390/agronomy13071709

Ali S, Hameed A, Muhae-Ud-Din G, Ikhlaq M, Ashfaq M, Atiq M, Ali F, Zia ZU, Naqvi SA, Wang Y (2023) Citrus Canker: A Persistent Threat to the Worldwide Citrus Industry—An Analysis. Agronomy 13:1112. https://doi.org/10.3390/agronomy13041112

Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389-3402. https://doi.org/10.1093/nar/25.17.3389

An SQ, Potnis N, Dow M, Vorhölter FJ, He YQ, Becker A, Teper D, Li Y, Wang N, Bleris L, Tang JL (2020) Mechanistic insights into host adaptation, virulence and epidemiology of the phytopathogen Xanthomonas. FEMS Microbiol Rev 44:1-32. https://doi.org/10.1093/femsre/fuz024

Ashraf B, Atiq N, Khan K, Wadood A, Uddin R (2022) Subtractive genomics profiling for potential drug targets identification against Moraxella catarrhalis. PloS One 17:e0273252. https://doi.org/10.1371/journal.pone.0273252

Aydınkal RM, Serçinoğlu O, Ozbek P (2019) ProSNEx: a web-based application for exploration and analysis of protein structures using network formalism. Nucleic Acids Res 47:W471-W476. https://doi.org/10.1093/nar/gkz390

Baker FN, Porollo A (2016) CoeViz: a web-based tool for coevolution analysis of protein residues. BMC Bioinform 17:1-7. https://doi.org/10.1186/s12859-016-0975-z

Barh D, Tiwari S, Jain N, Ali A, Santos AR, Misra AN, Azevedo V, Kumar A (2011) In silico subtractive genomics for target identification in human bacterial pathogens. Drug Dev Res 72:162-177. https://doi.org/10.1002/ddr.20413

Behlau F, Canteros BI, Minsavage GV, Jones JB, Graham JH (2011) Molecular characterization of copper resistance genes from Xanthomonas citri subsp. citri and Xanthomonas alfalfae subsp. citrumelonis. Appl. Environ. Microbiol 77:4089-4096. https://doi.org/10.1128%2FAEM.03043-10

Butt AM, Tahir S, Nasrullah I, Idrees M, Lu J, Tong Y (2012) Mycoplasma genitalium: a comparative genomics study of metabolic pathways for the identification of drug and vaccine targets. Infect Genet Evol 12:53-62. https://doi.org/10.1016/j.meegid.2011.10.017

Conforte VP, Malamud F, Yaryura PM, Toum Terrones L, Torres PS, De Pino V, Chazarreta CN, Gudesblat GE, Castagnaro AP, R. Marano M, Vojnov AA (2019) The histone‐like protein HupB influences biofilm formation and virulence in Xanthomonas citri ssp. citri through the regulation of flagellar biosynthesis. Mol Plant Pathol 20:589-598. https://doi.org/10.1111/mpp.12777

Eismann L, Fijalkowski I, Galmozzi CV, Koubek J, Tippmann F, Van Damme P, Kramer G (2022) Selective ribosome profiling reveals a role for SecB in the co-translational inner membrane protein biogenesis. Cell Rep 41:111776. https://doi.org/10.1016/j.celrep.2022.111776

Ference CM, Gochez AM, Behlau F, Wang N, Graham JH, Jones JB (2018) Recent advances in the understanding of Xanthomonas citri ssp. citri pathogenesis and citrus canker disease management. Mol Plant Pathol 19:1302. https://doi.org/10.1111%2Fmpp.12638

Francke C, Groot Kormelink T, Hagemeijer Y, Overmars L, Sluijter V, Moezelaar R, Siezen RJ (2011) Comparative analyses imply that the enigmatic Sigma factor 54 is a central controller of the bacterial exterior. BMC Genom 12:1-21. https://doi.org/10.1186/1471-2164-12-385

Gicharu GK, SUN DL, Xun HU, FAN XJ, Tao ZH, WU CW, ZOU HS (2016) The sigma 54 genes rpoN1 and rpoN2 of Xanthomonas citri subsp. citri play different roles in virulence, nutrient utilization and cell motility. J Integr Agric 15:2032-2039. https://doi.org/10.1016/S2095-3119(15)61317-X

Huang Y, Niu B, Gao Y, Fu L, Li W (2010) CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinform 26:680-682. https://doi.org/10.1093/bioinformatics/btq003

Ibrahim KA, Helmy OM, Kashef MT, Elkhamissy TR, Ramadan MA (2020) Identification of potential drug targets in helicobacter pylori using in silico subtractive proteomics approaches and their possible inhibition through drug repurposing. Pathogens 9:747. http://dx.doi.org/10.3390/pathogens9090747

Izadiyan M, Taghavi SM (2023) Diversity of copper resistant Xanthomonas citri subsp. citri strains, the causal agent of Asiatic citrus canker, in Iran. Eur J Plant Pathol 1-14. https://doi.org/10.1007/s10658-023-02786-w

Kanehisa M, Goto S, Kawashima S, Nakaya A (2002) The KEGG databases at GenomeNet. Nucleic Acids Res 30:42-46. https://doi.org/10.1093/nar/30.1.42

Kaur H, Kalia M, Singh V, Modgil V, Mohan B, Taneja N (2021a) In silico identification and characterization of promising drug targets in highly virulent uropathogenic Escherichia coli strain CFT073 by protein-protein interaction network analysis. Inform Med Unlocked 25:100704. https://doi.org/10.1016/j.imu.2021.100704

Kaur H, Kalia M, Taneja N (2021) Identification of novel non-homologous drug targets against Acinetobacter baumannii using subtractive genomics and comparative metabolic pathway analysis. Microb Pathog 152:104608. https://doi.org/10.1016/j.micpath.2020.104608

Keshri V, Singh DP, Prabha R, Rai A, Sharma AK (2014) Genome subtraction for the identification of potential antimicrobial targets in Xanthomonas oryzae pv. oryzae PXO99A pathogenic to rice. 3 Biotech 4:91-95. https://doi.org/10.1007/s13205-013-0131-7

Khan K, Jalal K, Khan A, Al-Harrasi A, Uddin R (2022) Comparative Metabolic Pathways Analysis and Subtractive Genomics Profiling to Prioritize Potential Drug Targets Against Streptococcus pneumoniae. Front Microbiol 12:796363. https://doi.org/10.3389/fmicb.2021.796363

Khan MK, Dangles O (2014) A comprehensive review on flavanones, the major citrus polyphenols. J Food Compos Anal 33:85-104. https://doi.org/10.1016/j.jfca.2013.11.004

Kong X, Zhu B, Stone VN, Ge X, El-Rami FE, Donghai H, Xu P (2019) ePath: an online database towards comprehensive essential gene annotation for prokaryotes. Sci Rep 9:12949. https://doi.org/10.1038/s41598-019-49098-w

Košćak L, Lamovšek J, Đermić E, Prgomet I, Godena S (2023) Microbial and plant-based compounds as alternatives for the control of phytopathogenic bacteria. Horticulturae 9:1124. https://doi.org/10.3390/horticulturae9101124

Kumar D, Jarial K, Jarial RS, Banyal SK, Jandaik S (2020) Prevalence of citrus canker caused by Xanthomonas axonopodis pv. citri in subtropical zone of himachal Pradesh. Int j bio-resour stress manag 11:89-94. https://doi.org/10.23910/IJBSM/2020.11.1.2061

Lin X, Li X, Lin X (2020) A review on applications of computational methods in drug screening and design. Molecules 25:1375. https://doi.org/10.3390/molecules25061375

Luiz MH, Takahashi LT, Bassanezi RC (2021) Optimal control in citrus diseases. Comput. Appl. Math 40:191. https://doi.org/10.1007/s40314-021-01581-9

Luo H, Lin Y, Gao F, Zhang CT, Zhang R (2014) DEG 10, an update of the database of essential genes that includes both protein-coding genes and noncoding genomic elements. Nucleic Acids Res 42:D574-D580. https://doi.org/10.1093/nar/gkt1131

Mandal J, Sinha S (2021) In Silico Identification of Protein in Ralstonia solanacearum, A Bacterial Wilt Pathogen for Drug Target by Subtractive Genomic Analysis. Biosc Biotech Res Comm 14:291-297. http://dx.doi.org/10.21786/bbrc/14.1/41

Martins PM, de Oliveira Andrade M, Benedetti CE, de Souza AA (2020) Xanthomonas citri subsp. citri: host interaction and control strategies. Trop Plant Pathol 45:213-236. https://doi.org/10.1007/s40858-020-00376-3

Naqvi SA, Wang J, Malik MT, Umar UU, Hasnain A, Sohail MA, Shakeel MT, Nauman M, Hassan MZ, Fatima M, Datta R (2022) Citrus canker—distribution, taxonomy, epidemiology, disease cycle, pathogen biology, detection, and management: A critical review and future research agenda. Agronomy 12:1075. https://doi.org/10.3390/agronomy12051075

Rabbee MF, Ali MS, Baek KH (2019) Endophyte Bacillus velezensis isolated from Citrus spp. Controls streptomycin-resistant Xanthomonas citri subsp. citri that causes citrus bacterial canker. Agronomy 9:470. http://dx.doi.org/10.3390/agronomy9080470

Rahman MA, Noore MS, Hasan MA, Ullah MR, Rahman MH, Hossain MA, Ali Y, Islam MS (2014) Identification of potential drug targets by subtractive genome analysis of Bacillus anthracis A0248: An in silico approach. Comput Biol Chem 52:66-72. http://dx.doi.org/10.1016/j.compbiolchem.2014.09.005

Rathod NB, Elabed N, Punia S, Ozogul F, Kim SK, Rocha JM (2023) Recent developments in polyphenol applications on human health: a review with current knowledge. Plants 12:1217. https://doi.org/10.3390/plants12061217

UniProt Consortium (2019) UniProt: a worldwide hub of protein knowledge. Nucleic Acids Res 47:D506-D515. http://dx.doi.org/10.1093/nar/gky1049

Sanyal D, Banerjee S, Bej A, Chowdhury VR, Uversky VN, Chowdhury S, Chattopadhyay K (2022) An integrated understanding of the evolutionary and structural features of the SARS-CoV-2 spike receptor binding domain (RBD) Int J Biol Macromol 217:492-505. https://doi.org/10.1016/j.ijbiomac.2022.07.022

Searchinger T, Waite R, Hanson C, Ranganathan J, Dumas P, Matthews E, Klirs C (2019) Creating a sustainable food future: A menu of solutions to feed nearly 10 billion people by 2050. Final report In WRI.

Senthamizhan V, Ravindran B, Raman K (2021) NetGenes: A database of essential genes predicted using features from interaction networks. Front genet 12:722198. https://doi.org/10.3389/fgene.2021.722198

Singh R, Ramniwas, Kumar M (2019) Development of citrus canker in citrus in relation to weather parameters. J Agrometeorol 21:134-139.

Solanki V, Tiwari M, Tiwari V (2019) Prioritization of potential vaccine targets using comparative proteomics and designing of the chimeric multi-epitope vaccine against Pseudomonas aeruginosa. Sci Rep 9:5240. https://doi.org/10.1038/s41598-019-41496-4

Stephenson K, Hoch JA (2002) Two-component and phosphorelay signal-transduction systems as therapeutic targets. Curr Opin Pharmacol 2:507-512. https://doi.org/10.1016/S1471-4892(02)00194-7

Sudha R, Katiyar A, Katiyar P, Singh H, Prasad P (2019) Identification of potential drug targets and vaccine candidates in Clostridium botulinum using subtractive genomics approach. Bioinform 15:18. https://doi.org/10.6026%2F97320630015018

Sundin GW, Wang N (2018) Antibiotic resistance in plant-pathogenic bacteria. Annu Rev Phytopathol 56:161-180. https://doi.org/10.1146/annurev-phyto-080417-045946

Syed-Ab-Rahman SF, Hesamian MH, Prasad M (2022) Citrus disease detection and classification using end-to-end anchor-based deep learning model. Appl Intell 52:927-938. https://doi.org/10.1007/s10489-021-02452-w

Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, Simonovic M, Roth A, Santos A, Tsafou KP, Kuhn M (2015) STRING v10: protein–protein interaction networks, integrated over the tree of life. Nucleic Acids Res 43:D447-D452. https://doi.org/10.1093/nar/gku1003

Talibi I, Boubaker H, Boudyach E, Ait Ben Aoumar A (2014) Alternative methods for the control of postharvest citrus diseases. J Appl Microbiol 117:1-17. https://doi.org/10.1111/jam.12495

The UniProt Consortium (2021) UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res 49: D480–D489. https://doi.org/10.1093/nar/gkaa1100

Tian W, Chen C, Lei X, Zhao J, Liang J (2018) CASTp 3.0: computed atlas of surface topography of proteins. Nucleic Acids Res 46:W363-W367. https://doi.org/10.1093/nar/gky473

Wishart DS, Feunang YD, Guo AC, Lo EJ, Marcu A, Grant JR, Sajed T, Johnson D, Li C, Sayeeda Z, Assempour N (2018) DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res 46:D1074-D1082. https://doi.org/10.1093/nar/gkx1037

Yan Q, Wang N (2011) The ColR/ColS two-component system plays multiple roles in the pathogenicity of the citrus canker pathogen Xanthomonas citri subsp. citri. J Bacteriol 193:1590-1599. https://doi.org/10.1128%2FJB.01415-10

Yasuhara-Bell J, Santillana G, Robène I, Pruvost O, Nakhla M, Mavrodieva V (2023) Genome-informed multiplex conventional PCR for identification and differentiation of Xanthomonas citri pv. citri subpathotypes, the causal agents of Asiatic citrus canker. PhytoFront 3:235-245. https://doi.org/10.1094/PHYTOFR-04-22-0044-FI

Yu CS, Chen YC, Lu CH, Hwang JK (2006) Prediction of protein subcellular localization. Proteins: Struct Funct Genet 64:643-651. https://doi.org/10.1002/prot.21018

Yu NY, Wagner JR, Laird MR, Melli G, Rey S, Lo R, Dao P, Sahinalp SC, Ester M, Foster LJ, Brinkman FS (2010) PSORTb 3.0: improved protein subcellular localization prediction with refined localization subcategories and predictive capabilities for all prokaryotes. Bioinform 26:1608-1615. https://doi.org/10.1093/bioinformatics/btq249

The authors declare no competing interests.

Supplementarytables.pdf
Supplementaryfigures.pdf
GraphicalAbstract.png
Graphical Abstract Overview of the workflow showing different server and tools to find out the potential drug targets

Subtractive genome mining in Xanthomonas citri pv. citri strain 306 for identifying novel drug target proteins coupled with in-depth protein-protein interaction and coevolution analysis - A leap towards prospective drug design

Status:

Version 2

Abstract

Figures

Introduction

Methodology

Result

Discussion

Conclusion

Declarations and statements

References

Additional Declarations

Supplementary Files

Status:

Version 2