This study was conducted exclusively using publicly available genomic and proteomic datasets retrieved from the National Center for Biotechnology Information (NCBI) database and other open-access bioinformatics resources. No live animals, human participants, or biological samples were involved at any stage of the study. Therefore, ethical approval from an institutional animal care and use committee or human ethics committee was not required.

The study was conducted using publicly available genomic and proteomic data retrieved from the NCBI database. All analyses were performed in silico between April 2024 and February 2026.

The study design and analytical workflow are presented in Figure 1. The selection of Brucella species in this study was guided by biological, epidemiological, and phenotypic diversity considerations rather than random inclusion. The five selected species were chosen based on their established relevance to animal health and, in several cases, human infections, as well as their representation across multiple host species, including cattle, sheep, goats, pigs, and dogs.

Furthermore, both smooth and rough phenotypes were intentionally included to reflect structural variability in LPS composition and the associated immunological differences. This strategy ensured a broader representation of biological and epidemiological diversity across major Brucella lineages. The inclusion of both phenotypes also enhanced diversity in LPS structures, which are known to influence immune recognition, virulence, and serological behavior.

In this study, the objective was to explore conserved molecules with potential relevance as vaccine or therapeutic targets across B. melitensis bv. 1 str. 16M, B. suis 1330, B. canis ATCC 23365, B. ovis ATCC 25840, and B. abortus 2308.

The selection of nine virulence-associated genes (virB3, virB7, virB5, wbkC, wbkB, wbkA, fabZ, gmd, and lpxC) was based on their functional roles in key pathogenic mechanisms. In addition, the availability of sequence data for all selected species in open-access databases was considered as an important methodological criterion. These proteins were selected due to their established involvement in virulence, LPS biosynthesis, and host-pathogen interactions.

The genes (virB3, virB7, virB5, wbkC, wbkB, wbkA, fabZ, gmd, and lpxC) and their corresponding amino acid sequences from B. melitensis bv. 1 str. 16M, B. suis 1330, B. canis ATCC 23365, B. ovis ATCC 25840, and B. abortus 2308 were retrieved from the NCBI database (https://www.ncbi.nlm.nih.gov/gene/) (Accessed April 17, 2024).

The physicochemical properties of the nine selected virulence proteins were analyzed using the ProtParam tool available on the ExPASy server (http://web.expasy.org/protparam/) (Accessed April 17, 2024). The parameters evaluated included the number of amino acids, molecular weight, theoretical isoelectric point, instability index, and the grand average of hydropathicity (GRAVY).

The predicted subcellular localization of the virulence proteins was determined using the DeepLoc-1.0 online server (https://services.healthtech.dtu.dk/services/DeepLoc-1.0/) (Accessed April 18, 2024). Protein sequences were submitted in FASTA format for analysis.

Information regarding conserved domains of the virulence proteins was obtained from the Conserved Domains Database (CDD) at NCBI (https://www.ncbi.nlm.nih.gov/guide/domains-structures/) (Accessed June 29, 2024). Protein sequences in FASTA format were submitted and analyzed at an E-value threshold of 0.01. The output displayed the best-scoring domain model along with its associated domain superfamily for each query region.

Conserved motif identification was performed using the Multiple Em for Motif Elicitation (MEME) server (Version 5.5.9) (https://meme-suite.org/meme/tools/meme) (Accessed Feb 19, 2026) for VirB3, VirB7, VirB5, WbkC, WbkB, WbkA, FabZ, gmd, and lpxC proteins.

Multiple sequence alignments were conducted using the amino acid sequences of the selected virulence proteins in Geneious Prime version 2024.0 (Biomatters; https://www.geneious.com). These analyses were used to evaluate the degree of similarity and divergence among virulence proteins.

Phylogenetic relationships were inferred using the Maximum Likelihood method with the Jones–Taylor–Thornton (JTT) matrix-based model [22]. The dataset consisted of 24 amino acid sequences, and evolutionary analyses were performed using Molecular Evolutionary Genetics Analysis (MEGA11).

Protein–protein interaction networks were constructed using the STRING database (https://string-db.org/cgi/input?sessionId=bgzpUl8u2HmH) (Accessed May 03, 2024). The proteins VirB3, WbkB, WbkC, FabZ, gmd, and lpxC from B. abortus 2308 were selected to identify their interaction partners and functional connectivity.

The bioinformatics tools and software resources used in this study, along with their respective versions and access details, are summarized in Table 1.

Gene and protein sequence information for B. melitensis bv. 1 str. 16M, B. suis 1330, B. canis ATCC 23365, B. ovis ATCC 25840, and B. abortus 2308 were obtained from the NCBI Gene database. The selected virulence-associated genes included virB3, virB7, virB5, wbkC, wbkB, wbkA, fabZ, gmd, and lpxC, together with the amino acid sequences of the proteins they encode. For each gene and its corresponding protein, key annotation details such as Gene ID, gene symbol, functional description, gene type, protein accession number (NCBI RefSeq), chromosomal location, and genomic reference sequence were compiled and organized. These features are summarized in Table 2 and served as the reference dataset for the subsequent motif, localization, phylogenetic, and protein–protein interaction analyses.

The physicochemical properties of nine virulence proteins were predicted and analyzed via using PortParam to obtain their number of amino acids, molecular weight, theoretical isoelectric point, instability coefficient, and the grand average of hydropathicity (GRAVY) (Table 3).

Using the DeepLocPro 1.0 server, prokaryotic proteins can be assigned to six localities: cytoplasm, cytoplasmic membrane, periplasm, outer membrane, cell wall and surface, and extracellular space. The potential subcellular location predictions of virulence proteins were obtained for lpxC, gmd, FabZ, WbkA, and WbkC as cytoplasmic, WbkB, and VirB3 as cytoplasmic membrane, VirB7 as outer membrane, and VirB5 as extracellular Table 4.

Information on conserved domains of virulence proteins was obtained from the CDD at NCBI and is presented in Table 5. The online MEME server was used to identify conserved motifs for the VirB3, VirB7, VirB5, WbkC, WbkB, WbkA, FabZ, gmd, and lpxC proteins and showed Table 6 and Figure 2.

When the resulting VirB3, VirB7, VirB5, WbkC, WbkB, WbkA, FabZ, gmd, and lpxC proteins multiple sequences alignment was visualized in Geneious Prime version 2024.0 (Figure 3) conserved and distinct amino acids were identified in the B. melitensis bv. 1 str. 16M, B. suis 1330, B. canis ATCC 23365, B. ovis ATCC 25840 and B. abortus 2308 virulence proteins sequences. The evolutionary history was inferred by using the Maximum Likelihood method and Jones–Taylor–Thornton (JTT) matrix-based model [22]. The tree with the highest log likelihood (-6454.69) is shown. The percentage of trees in which the associated taxa clustered together is shown above the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with the superior log likelihood value. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. This analysis involved 24 amino acid sequences. There were 424 positions in the final dataset. Evolutionary analyses were conducted in MEGA11 [23] (Figure 4).

The interactive protein study unfolds the numerous interacting partners of VirB3, WbkB, WbkC, FabZ, gmd, and lpxC (Figure 5) and their interactive protein partners (Table 7). The VirB2-4-5-6-8-9-10-11 had very high interactions with reference protein VirB3 (Figure 5a). The reference VirB3 protein was also highly associated with the VirB1 and VirB7. The BAB1_0999, BAB1_0058, BAB1_0542, RfbD, gmd, BAB1_0544, BAB1_0561, and BAB2_0856 had high interactions with reference protein WbkB (BAB1_0541) (Figure 5b). BAB1_0999 is a conserved hypothetical protein that shows glycosyltransferase activity and functions in the LPS biosynthetic process. The activities of BAB1_0058, BAB1_0542, RfbD, gmd, BAB1_0544, BAB1_0561, and BAB2_0856 proteins are summarized in Table 7. In addition, BAB1_0802 ve BAB1_0540 had medium interactions with reference protein WbkB [24, 25, 26]. The BAB2_0695 had very high interactions with reference protein WbkC (BAB1_0540) (Figure 5c). BAB2_0695 has NAD-dependent epimerase/dehydratase activity [24]. NusB, BAB1_0544, FolD, Sun, BAB2_0845, BAB1_0188, BAB1_0542, RfbD, whose activities are summarized in Table 7, had high interactions with reference protein WbkC. The lpxA, lpxD, AcpP, AccA, FabD, AccD, lpxC, FabB, FabH, and BAB1_0486 had very high interactions with reference protein FabZ (Figure 5d). The BAB1_0802 had a very high interaction with the reference protein gmd (Figure 5e). Also, BAB1_0544, BAB1_0542, BAB2_0856, BAB1_0561, RfbD, BAB1_0541, Apt, RplD, and BAB1_0534 had high interactions with the reference protein gmd. The lpxA, lpxD, FabZ, lpxB, lpxK, and KdsA had very high interactions with the reference protein lpxC (Figure 5f). Also, the KdtA, KdsB, FtsZ, and SecA had high interactions with the reference protein lpxC.

Brucella species are facultative intracellular pathogens known to cause serious zoonotic diseases around the world [27]. The virulence of these pathogens depends on their ability to evade or modulate the host immune response, as well as their capacity to survive and replicate within a vacuole derived from the endoplasmic reticulum [28].

Proteins with an instability index above 40 indicate that the protein is unstable and has a short half-life [29]. The GRAVY value is commonly used to evaluate the hydrophilicity and hydrophobicity of proteins. Moreover, the GRAVY value ranges between −2 and 2, where a negative or positive value indicates that the protein possesses hydrophilic or hydrophobic properties, respectively [30]. VirB3, VirB7, and FabZ proteins are predicted to be hydrophobic, whereas VirB5, WbkC, WbkB, WbkA, gmd, and lpxC proteins are predicted to be hydrophilic.

Considering the properties of the conserved domains of VirB3, VirB7, VirB5, WbkC, WbkA, FabZ, gmd, and lpxC proteins, respectively, the VirB3 protein contains the VirB3 superfamily covering amino acids 18-113, the VirB7 protein contains the putative periplasmic lipoprotein domain between amino acids 1-43, and the VirB5 protein contains the VirB5_like domain covering amino acids 1-217 [31]. There are main genes in the wbk locus, and one of the proteins encoded by these genes, the WbkC protein, contains the formyltransferase catalytic core domain encompassing amino acids 9-182 [17]. When the conserved domain for the WbkA protein was examined, it was observed that it belongs to glycosyltransferase family 4 (GT4) between amino acids 6-362 [32]. The gmd protein contains the GDP-D-mannose dehydratase domain between amino acids 4-343 [33]. The FabZ protein contains the 3-hydroxyacyl-ACP dehydratase FabZ domain between amino acids 6-150 [24]. The lpxC protein contains the UDP-3-O-acyl-N-acetylglucosamine deacetylase domain between amino acids 4-279.

Protein sequence motifs are patterns of residues that exhibit distinct structural and functional characteristics. They have been identified using multiple sequence and structural alignment methods as conserved signatures typical of protein families [34]. An important aspect of biological sequence characterization is the identification of motifs and domains [35]. Three conserved motifs were obtained:”YNSIIEIDPRFIREAEVVLPV” (Motif 1, red), “YSNLPHNYRDLYEAVMSGGIL” (Motif 2, blue), and “MKKIILSFAFALTVTSTAHAQLPVTDAGS” (Motif 3, green). B. melitensis bv. 1 str. 16M, B. suis 1330, B. canis ATCC 23365, B. ovis ATCC 25840, and B. abortus 2308 had all three motifs in VirB5 and WbkB virulence proteins. Motif 1 (red) and Motif 2 (blue) were detected in WbkA, FabZ, gmd, and lpxC virulence proteins. No motif was detected in VirB3 and VirB7 virulence proteins. All virulence proteins except VirB3, VirB7, and WbkC have the “YNSIIEIDPRFIREAEVVLPV” motif represented in Motif 1 (red). The conserved motif analysis and the phylogenetic tree are consistent with each other.

Understanding biological processes requires a well-organized understanding of the physical interactions between proteins [36]. Proteins are essential molecules in living organisms, performing a variety of critical functions within cells, such as signaling, regulating metabolism, and maintaining cell structure [37]. Proteins interact with one another, regulating each other’s functions and binding together to fulfill their roles in biological processes [36]. Investigating protein–protein interactions not only enhances our understanding of vital cellular functions but also plays a significant role in the development of disease treatments and the design of drugs [37].

Virulence factors are essential for pathogen infection, making the interaction of virulence factors particularly important. LPS is an essential virulence factor of Brucella [5]. The virulence factors lpxA, lpxC, KdsB, and KdsA are involved in different stages of lipid A biosynthesis, an essential component of the LPS structure [38].

Although lpxC has been extensively validated as a drug target in other Gram-negative pathogens such as E. coli, Pseudomonas aeruginosa, and Acinetobacter baumannii [39, 40], its role as a therapeutic target in Brucella has not been widely addressed. Protein-protein interaction analyses have revealed that lpxC interacts significantly with a variety of proteins. Inhibition of lpxC may influence multiple interconnected components of the LPS biosynthetic pathway rather than definitively destabilizing the entire machinery. To the best of our knowledge, lpxC has not been widely emphasized as a rational drug target in Brucella, and our in silico network, motif, and localization data provide a complementary perspective for anti-Brucella drug development.

The interactive network analysis in this study highlights that inhibiting one of the proteins such as VirB3, VirB5, VirB7, WbkC, WbkA, FabZ, gmd, or lpxC may affect several functionally associated proteins and biological processes rather than guaranteeing a direct cascade failure. This synergistic vulnerability suggests that these proteins are not only potential drug targets but may also represent promising vaccine candidates when surface-exposed or secreted (e.g., VirB5, VirB7). Moreover, the conserved motifs identified, particularly those shared across Brucella species, may provide a basis for designing multi-epitope subunit vaccines aimed at inducing broad protective immunity [19, 41–44].

Brucella T4SS proteins play critical roles in the intracellular life cycle of the bacterium and may serve as targets for the immune system. In the literature, studies on VirB proteins have mostly focused on single antigens. For example, Tan et al. [45] reported that recombinant VirB5 protein could serve as a strong antigen for the serological diagnosis of bovine brucellosis, demonstrating high specificity (97.8%). Similarly, Pollak et al. [46] showed that immunization with VirB7 in combination with VirB9 in mice and dogs elicited a Th1-type immune response and reduced splenic bacterial load by approximately 1 log.

VirB3 is located in the cytoplasmic membrane, which limits its immune accessibility. The importance of VirB3 does not stem from sequence conservation or surface exposure, but rather from its central functional role within the T4SS and its strong connections with multiple VirB components in the protein–protein interaction network. This network centrality supports prioritizing VirB3 as a functional drug target rather than as a primary vaccine candidate.

By comparison, although VirB5 and VirB7 are not absolute central hubs in the interaction network, they are functionally associated with the T4SS and show significant interactions with VirB3, indicating biological relevance within the system. When their immune accessibility advantage is considered, these two proteins emerge as complementary and potentially synergistic candidates for subunit vaccine design.

Our findings indicate that the combined evaluation of VirB5 and VirB7 may provide a rational basis for a potential broad-spectrum subunit vaccine perspective. VirB5 was shown to harbor conserved motifs across all Brucella species in this study and to be extracellularly localized, while VirB7 was predicted to be an outer membrane-associated lipoprotein with strong interactions within the T4SS network. These features suggest that both proteins could represent promising core immunogens for a multi-epitope vaccine strategy rather than isolated antigen selection.

Nevertheless, considering the motif distribution observed in our study, the likelihood of stronger cross-protective potential appears particularly high for five Brucella spp., where VirB5 motifs were consistently conserved. These conclusions are consistent with previous reports highlighting conserved outer membrane and T4SS proteins as cross-protective antigens and align with recent reviews emphasizing multi-antigen subunit or motif-based vaccines as promising next-generation strategies [19, 20, 41–44, 46, 47].

To the best of our knowledge, the VirB5-VirB7 combination has received limited attention in previous Brucella vaccine studies, and our findings support its consideration as a potential vaccine target. It should be noted that the distinct contribution of the present study does not arise from the individual bioinformatic tools employed, which are widely used in molecular analyses, but from their integrative interpretation across multiple complementary analytical dimensions. This combined evaluation enabled a comparative prioritization of conserved virulence factors rather than a purely descriptive characterization, thereby providing a more rational and evidence-guided framework for translational vaccine and drug target exploration.

Since the Brucella species included in this study are known to infect different hosts, the conserved proteins identified here may have relevance across multiple species rather than being limited to a single host. From a One Health perspective, recognizing shared virulence mechanisms between animal and human infections can contribute to more integrated prevention and control strategies. Although these findings are based on computational analyses, the cross-species conservation observed suggests a possible value for broader zoonotic surveillance, vaccine research, and antimicrobial target exploration in both veterinary and human health contexts.