IgM and IgG Epitope Mapping of the Porin Outer Membrane Protein-2a from Brucela abortus: Potential Biomarkers for Detecting Exposure to Brucellosis

1. Introduction

Brucellosis is one of the most common and endemic zoonotic diseases affecting both animals and humans in many countries; however, this disease remains neglected in human medicine, and nationwide control activities and surveillance programs have never been adequately structured [1,2,3]. Among twelve known Brucella spp., only B. melitensis, B. abortus, B. canis, and B. suis (except biovar 2) are associated with human infection and are facultative intracellular pathogens [4,5].

Brucellosis is an occupational disease affecting mostly people in contact with animals or animal products, particularly in agricultural countries where most of the population is involved in livestock farming and land cultivation.

The clinical symptoms of human brucellosis are nonspecific and can be mistaken for those of other diseases that cause fever [6,7]. Infection in humans is mainly associated with undulant fever, severe back pain, headache, joint pain, fatigue, myalgia, weight loss, loss of appetite, depression, and night sweats. Arthritis, orchitis, endocarditis, hepatomegaly, and splenomegaly have been seen in patients who suffered from brucellosis [8,9,10].

The disease is transmitted from unhealthy animals to humans via infected fluids and discharges such as blood, vaginal secretions, placental fluids, and aborted fetuses [11]. Bacterial transmission from person to person is rare, but sexual contact and breastfeeding have been proven to transmit Brucella [12]. Common routes of infection include direct inoculation through cuts, inoculation through the conjunctival sac of the eyes, abrasions of the skin, and inhaling infectious aerosols [13,14]. Occupational exposure to animals and direct contact with sick animals, as well as consuming contaminated milk and milk products, are the most common risk factors for human brucellosis [15,16]. The prevalence of brucellosis in livestock, particularly in low- and middle-income countries, has significant socioeconomic consequences [17,18] and poses substantial public health challenges [19,20]. In some parts of the world, the incidence of brucellosis is increasing daily due to a lack of awareness [21].

Early diagnosis of infected animals and humans is a crucial step in reducing the incidence of brucellosis. Thus, laboratory assays such as culturing, serology, and molecular analysis can frequently be used to diagnose brucellosis [22,23]. Although serological markers and PCR-based techniques are extremely sensitive, and extensive experience has been gained with these techniques in the laboratory diagnosis of brucellosis [24], a culture is still considered to be the "gold standard" due to the importance of this aspect of public health and clinical care.

In endemic regions, however, serological tests remain the primary method of diagnosis due to their low cost, user-friendliness, and strong ability to provide a negative prediction, so they are commonly used. A nucleic acid amplification assay, which is highly sensitive, specific, and safe, enables rapid disease diagnosis [25,26,27]. Patients who have reportedly fully healed may continue to have positive molecular test results for an extended period [28]. Therefore, cultures and serological methods will continue to be the primary tools for diagnosing and monitoring human brucellosis if no commercial tests or studies demonstrate adequate interlaboratory reproducibility.

Over the past few years, several studies have been conducted to develop Brucella vaccines [29,30,31,32,33,34], but the problem of controlling brucellosis in both humans and animals remains challenging [28]. Therefore, rapid and specific diagnosis is still necessary for both the treatment and prevention of infectious diseases, especially in humans, since all animal species that test positive for the disease must be slaughtered [35,36].

Classical serological tests, such as the tube agglutination test [37,38], rose Bengal plate test [39,40], and complement fixation test [41], as well as commercial enzyme-linked immunosorbent assay kits [42], are based on the detection of antibodies to the cell wall polysaccharide antigens of Brucella spp smooth strains. As a result, they do not exclude cross-reactions with related bacteria [43,44,45] and fail to differentiate between infected and vaccinated animals. Over the past decades, many attempts have been made to identify immunoreactive and pathogen-specific protein antigens. To date, several studies have investigated Brucella spp full recombinant proteins and chimeric cell wall proteins, as the best antigens for diagnosing brucellosis in humans [46,47,48,49,50] and animals [51,52,53,54,55,56]. However, the available results on the specificity and sensitivity of serological tests based on cell wall proteins are ambiguous and sometimes contradictory [28,57].

Therefore, this study aimed to identify the linear B-epitopes of the B. abortus Omp-2a protein, specifically those recognized by IgM and IgG, and to develop multiepitope polypeptides to explore their utility in the development of a rapid IgM and/or IgG diagnostic assay. The Omp-2a belongs to the alphaproteobacterial porin family, a multi-pass β-barrel outer membrane protein that induces a large passive diffusion pore at the center of each barrel, allowing small, hydrophilic molecular materials to cross the outer membrane [58,59].

2. Materials and Methods

2.1. Human Sera

In this study, blood samples from 40 patients diagnosed with brucellosis, confirmed by a commercial ELISA, were analyzed. Sera were obtained between 2012 and 2022 by the Central Public Health Laboratory of Rondônia (LACEN-Ro) and the Infectiology Service at the Oswaldo Cruz Policlinic in Rondônia, Brazil.

The other twenty-one sera samples used were from healthy individuals (21) obtained from the blood bank donor HEMORIO-RJ. For this study, patients' identities were kept anonymous.

2.2. Spot Synthesis

Two libraries of 163 peptides, each covering the entire sequence of Omp-2a (UniProt Q44620) from B. abortus (biovar 1, strain 9-941), were synthesized. These peptides were 15 residues in length, with a 10-residue overlap, and were obtained using a synthesizer (Auto-Spot ASP222, Intavis Bioanalytical Instruments AG, Köln, Germany) and the F-moc (9-fluorenylmethoxy carbonyl) strategy, as previously described [60].

As positive controls, the following peptides were used: GYPKDGNAFNNLDRI (Clostridium tetani; G20), KEVPALTAVETGATN (Poliovirus; G21), YPYDVPDYAGYP YDV (Hemagglutinin, Influenza; G22), GDFIDYEELREQLGG (Influenza A virus H3N2; G23), and YPGEFADYEELREQL (Influenza A virus Jakarta H1N1; G24).

2.3. Screening and Measurement of Spot Signal Intensities

Cellulose membranes (amino-PEG500-UC540; Intavis Bioanalytical Instruments, Köln, Germany) were equilibrated with TBS (50 mM Tris-buffered saline, pH 7.0) and blocked overnight at 4°C with TBS containing 3% casein and 0.1% Tween 20 (TBS-CT). After extensive washing with TBS-T (Tris-buffered saline, 0.1% Tween 20, pH 7.0), the membranes containing the peptide libraries were incubated for 2 hours with a sera pool from twenty different patients (1:250) in TBS-CT and then washed again with TBS-T.

Next, the membranes were incubated with alkaline phosphatase (AP)-labeled goat anti-human IgM (huIgM, 1:5000 in TBS-T; KPL, Gaithersburg, MD, USA; Lot #070466) or goat anti-human IgG (huIgG, 1:5000 in TBS-CT; Thermo, Lot #JUA1121836) for 1 hour, washed with TBS-T, and given a final wash in CBS (50 mM citrate-buffered saline, pH 7.0). Chemiluminescent CDP-Star® Substrate (0.25 mM) with Nitro-Block-II™ Enhancer (Applied Biosystems, Waltham, MA, USA) was then added to complete the reaction.

As described previously [61], chemiluminescent signals were detected using an Odyssey FC (LI-COR Bioscience, Lincoln, NE, USA). Briefly, a digital image file was generated at a resolution of 5 MP, and the signal intensities were quantified using the TotalLab TL100 software (v. 2009, Nonlinear Dynamics, Newcastle, Tyne, UK). The signal intensity (SI) used as a background was set by negative controls spotted on each membrane. Finally, a comparative analysis of the reactivity index of the spots, normalized on a dimensional hierarchical level, was conducted using the approach previously described [62].

2.4. Preparation of the Chimeric Peptides

Eight multi-antigen peptides (MAP4) epitope [PP225, PP226, PP227, PP228, PP229, PP231, PP232, PP233 (15 residues)] were synthesized using the F-moc protocol and the tetrameric TentaGel-S-NH2 resin (Intavis Bioanalytical Instruments AG, Köln, Germany) (Table S1) [63]. The synthesis of the two chimeric peptides, PP230-IgG (60 residues) and PP234 (50 residues), was conducted using XXX resin. Constructs were synthesized on a MultiPep-1 automated peptide synthesizer (CEM Corp, Charlotte, NC, USA), employing tetrafunctional Fmoc-amino acids with TFA-labile side chain protection where necessary. Residues of the monovalent (tail') segment, starting with bis-Fmoc Lys, were coupled using single-step protocols. After sequence assembly, F-moc groups were removed, and the peptide-resin was cleaved and deprotected with TFA/H₂O/EDT/TIS (94:2.5:2.5:1.0 v/v) for 90 min. The peptides were precipitated with cold diethyl ether, centrifuged three times for 10 minutes at 4°C, then dissolved in 10% aqueous Acetic Acid, dried, and stored as a lyophilized powder. MAP4 was dissolved in water, centrifuged at 10,000 g for 60 min at 15°C, and the supernatant filtered with a Centricon 10 filter. The single peptides were utilized without prior purification, and their identity was verified by MS (MALDI-TOF or electrospray).

2.5. Enzyme-Linked Immunosorbent Assay (ELISA)

The in-house ELISA was performed as previously described [64]. Briefly, 96-well plates (Immulon 2HB; Thermo Fisher, Waltham, MA, USA), were incubated (12 h at 4 °C) with 50 µl synthetic peptide solution (100 µg/mL in Na2CO3-NaHCO3 buffer, 0.1 M, pH 9.6). After three washes with PBS pH 7.2, the free sites on the plate were blocked for 90 min with 200 μL of a 2% skim milk solution in PBS containing 0.1% Tween 20, pH 7.2. After three further washes in PBS, 50 µl of patient serum (1:50) was added to the wells and the plate incubated for 90 min at 37 °C. After further washes with PBS, 50 µl of goat anti-human IgG (H+L) conjugated to alkaline phosphatase (AP) (1: 5000) (Thermo Fisher, Waltham, MA, USA) was added to the wells and the plate incubated for 90 min at 37 °C. After a series of three washes in PBS pH 7.2, the pNPP substrate was added, and the plates were incubated for 15 min. Next, 50 µl of 3N NaOH were added, and the reading was done after 30-120 min at 405 nm on a FlexStation3 (Molecular Devices, Sunnyvale, Ca, USA).

For comparison, a commercial ELISA assay (classic kit, Serion, Curitiba, Brazil) was used. Briefly, precoated micro test plates were incubated with patients' sera, and after removal of unbound material, anti-human Ig-AP conjugate was allowed to react with the immune complex. After washing away excess conjugation, pNPP was added, and specific antibody binding was measured photometrically.

2.6. Bioinformatics and In-Silico Analysis Model

The complete sequence of Omp-2a (Q44620) of B. abortus was retrieved from the National Center for Biotechnology Information, USA (http://www.ncbi. nlm.nih.gov). Epitope locations within the three-dimensional molecular structure of Omp-2ae were identified by generating in silico protein models using the I-TASSER server (http://zhanglab.ccmb, accessed April 10, 2024). Models were selected based on their C-score and TM-score, and the 3D models were validated using the AlphaFold v3 database [65].

Multiple sequence alignments were performed using the programs ClustalW (http:// www.ebi.ac.uk/clustalw) and BioEdit (http://www.mbio.ncsu.edu/BioEdit/bioedit.html).

2.7. Statistical Analysis

ELISA tests were statistically analyzed using Med Calc software version 20.218 [66]. A statistical difference was considered significant if the p-value was ≤ 0.05. Initially, the outcomes for each peptide were reported as a reactivity index (RI), determined by the optical density (OD) ratio of a particular sample to the cut-off OD values for each test. All RI values were classified as positive (>1.00) or negative (< 0.005); those with a value of 0.005 were statistically significant.

2.8. Ethical Standards

This work was approved by the FIOCRUZ Ethics in Research Committee (CAAE: 23784114.8.0000.5248, 15 May 2014) and Federal Fluminense University (23/04/21 UFF)

3. Results

3.1. Epitope Mapping of the Omp-2a

IgM and IgG linear B epitopes of the full Omp-2a (321 aa) protein recognized by the patient’s sera were identified using a library of 163 synthetic peptides covalently linked to a cellulose membrane (Figure 1A and Figure 2A) and developed as described before. Panel B shows the chemiluminescence intensity measurement using the patient’s serum, and Panel C shows the hierarchical distribution of the peptides according to their reaction intensity. For IgM, the most reactive peptide was C1, followed by C2, D5, and others. For IgG, the most reactive peptides were C1, C2, D7, and B18.

The intensities were normalized using the reactivity of the positive control as 100%. The ratio of the synthesized peptides and their respective positions in the membranes is shown in Table S2.

The overlapping sequences of the positive peptides defined 10 epitopes for Omp-2a, comprising 3 IgM and 7 IgG epitopes recognized by the patient’s sera pool (Table 1).

3.2. Shared Epitopes and Selection of Specific Brucella sp Epitopes

Three common structural overlapping epitopes recognized by both IgM and IgG were identified. These include the sequence NNSRHDGQYG, shared by epitopes Omp-2a/1M and Omp-2a/4G; NGFSAVIALE, found in Omp-2a/2M and Omp-2a/5G; and KFGGE, present in Omp-2a/3M and Omp-2a/10G (Figure 3). The remaining four IgG epitopes (6G, 7G, 8G, and 9G) were IgG isotype-specific and did not overlap with IgM epitopes.

Regarding specificity based on sequence database analysis, only the 3M and 10G epitope was exclusively found in the Brucella genus, indicating genus specificity, ie. are present in Omp-2a of the structure of Brucella spp. but absent in other bacteria such as E. coli, Shigella boydii, and Yersinia spp.. The other eight epitopes exhibited sequence homology with proteins from various non-pathogenic genera, including Rhodotorula, Bartonella, Ochrobactrum, and Falsochrobactrum (Table 1). However, none of these genera are known to include human pathogens.

Figure 4 presents the alignment results of the B. bovis Omp-2a sequence with autotransporter proteins from various genera, demonstrating a general identity of 70% or more. In the figure, the colored rectangles demarcate only the unique linear B-epitopes (either IgM or IgG) that have been identified. No correlation was identified with Yersinia sp proteins.

3.3. Structural Studies

The secondary structure elements of Omp-2a (Q44620) were identified using the TMpred tool, as described in the previous section. The protein has 11 alpha-helical transmembrane (TM) and two conserved domains (signal peptide and autotransporter) and tandem repeats, indicating that Omp-2a is an authentic OmpA (Figure 5) but shares no sequence homology with any other porin and a higher number of negatively changed residues in the exposed loops than Omp-2b [45].

The Omp-2a 3D model was constructed by homology modeling using as a template the available data toxin (ABC protein complex) Yersinia entomophaga, which presented the highest "score". The models generated by the I-Tasser server presented reliability parameters with C-score values of 0.48 and RMSD = 8.6 (± 4.5 Ǻ). The 3D model, illustrating the spatial distribution of epitopes, is shown in Figure 5. The amino acids that define the 10 (IgM and IgG) epitopes were located and marked on the model of secondary structure. From this analysis, we can conclude that all epitopes are facing the external surface of the protein and are thus accessible to the immune system. All are "linear" structures with "coils" characteristics of linear B epitopes.

The protein presents 16-beta-barrel sheets, with large surface-exposed loops, that form a TM pore at the center of each barrel. The pore is partially occluded by a peptide loop that folds into the pore lumen. The larger pore formed by Omp-2a may be advantageous for intracellular growth when the bacterium competes with the host cell for nutrients whose concentration is particularly low within the phagosome.

3.4. Enzyme Immunoassay with Human Serum

From the sera of forty brucellosis suspect patients, nineteen were identified as positive with a higher optical density (OD) using a commercial kit and were included in this study. The results are depicted in Figure 6A. Among the nineteen analysed sera, only 1-3 did not exceed the cutoff values for all eight peptides analysed (PP225-PP229 and PP231-PP233), whereas sera from twenty-one apparently healthy individuals consistently fell 3-5 units above the cutoff point (Figure 6B) .

Cutoffs for each peptide (PP225-PP227 and PP231-233) were established using statistical analysis with ROC curve (Figure 7), as detailed in Table S3.

The ROC curves illustrate the diagnostic test’s discriminatory ability across different cutoff values, highlighting the trade-offs between optical sensitivity and specificity. The points closest to the top left corner of the ROC curve indicate the most effective diagnostic thresholds,

Sensitivity, defined as the proportion of true positive results (TP) relative to all affected patients (TP + false negative [FN], ranged from 0.88 to 0.95, indicating a strong statistical correlation between sera of patients with brucellosis and healthy individual sera results.

Specificity, representing the proportion of true negatives (TN) among controls (TN + false positive results [FPR]), ranged from 90% to 95%. The control group analyzed in the ROC curve analysis consisted of healthy individuals, demonstrating that the in-house ELISA test peptide exhibits 90-95% specificity for the diagnosis of Brucellosis in humans, a similar value obtained by the commercial kit.

ROC curves provide quantitative assessments of diagnostic test accuracy, with the area under the curve indicating overall efficiency. Peptide PP225-PP228 and P230 demonstrated higher immunogenicity (90% specificity and 94% sensitivity) (Table S2).

4. Discussion

Brucellosis diagnosis requires further refinement in terms of sensitivity and specificity, particularly during the acute phase [67]. Research indicates that diagnosing brucellosis is challenging due to difficulties with serological examinations. Previous studies have reported variable accuracy in ELISA-based detection of Brucella-specific IgM and IgG antibodies, with some suggesting a combined specificity of up to 100% [68,69]. However, the persistence of antibodies in cases without active infection complicates clinical interpretation.

Therefore, identifying epitope-specific immune responses may enhance diagnostic accuracy by targeting epitopes associated with active and recent infection. This study focused on the Omp-2a protein for epitope mapping due to its antigenic properties, membrane localization [70], and critical role in immune responses. Therefore, it was used in association with other proteins in the process of vaccination and diagnosis as a recombinant protein [71].

Our results identified ten immunoreactive epitopes on the B. abortus Omp-2a protein that reacted with anti-IgM and anti-IgG antibodies (Table 1). As no cross-reactivity was found with other human pathogens, the finding supports the molecular identification of Omp-2a from B. abortus as a suitable target for diagnostic use. Numerous studies have highlighted the physicochemical properties of B cell epitopes, such as the surface accessibility of membrane-bound or free antibodies [72,73]. This characteristic enables antibodies to effectively bind to and neutralize their biological targets [74]. Additionally, understanding the dynamics of humoral immune responses during Brucella infection is crucial for enhancing serodiagnostic tools and developing effective vaccine candidates. In human brucellosis, IgM antibodies typically appear within the first week of infection and peak around two months later, whereas IgG antibodies emerge after the second week and reach their maximal levels between six- and eight-weeks post-infection. Although IgG responses align more closely with disease progression, the detection of IgM in the absence of IgG, or vice versa, can lead to diagnostic ambiguity [75]. This highlights the need for refined diagnostic markers that can distinguish between the early and late phases of infection with high specificity and sensitivity.

This pattern reflects both the immunological stage of infection and the antigenic architecture of the Omp-2a protein, a 16-layer β-barrel secondary structure with large surface-exposed loops, which can naturally make accessible unique linear B cell epitopes. [76].

The relative scarcity of IgM epitopes identified in this work may be attributed to isotype switching during infection, as most patients included in our study were in a later phase of disease or under treatment. The process of isotype switching is a hallmark of adaptive immunity, involving recombination of constant regions in immunoglobulin genes to generate antibody classes with different effector functions and half-lives. This transition, while essential for developing high-affinity IgG antibodies, can obscure the presence of short-lived IgM responses, which are typically useful markers of recent or acute infection.

Interestingly, three of the IgM epitopes (Omp-2a/1M, 2M, and 3M) shared sequences with corresponding IgG epitopes (Omp-2a/4G, 5G, and 10G), indicating a sustained immune response against these antigenic determinants across different stages of infection. In contrast, four IgG epitopes (Omp-2a/6G to 9G) were uniquely recognized, possibly reflecting epitopes that induce IgG responses with no preceding or concurrent IgM reactivity, suggesting either rapid class switching or late-stage immune memory development. These findings support the notion that certain epitopes may be more relevant for the early diagnosis of brucellosis, while others could serve as markers of chronic or convalescent phases.

The validation of these immunodominant epitopes/peptides via ELISA confirmed their reactivity with sera from Brucella-infected patients. To assess their diagnostic potential and specificity, four epitopes with no homology to proteins from unrelated microorganisms were further evaluated. Notably, peptides PP225–PP229, mapped to the loops, were recognized by both IgM and IgG, indicating their capacity to elicit long-lasting immune responses. However, cross-reactivity observed within the epitopes 2M/5G and 7G and proteins from Bartonella tamiae, a Gram-negative bacterium isolated from the blood of Tailândia patients, suggests that these epitopes must be avoided when preparing a specific and universal diagnostic test for Brucellosis.

On the other hand, Omp-2a and Omp-2b are homologous antigens [77], but do not share the same linear epitopes.

In summary, this study advances our understanding of human humoral responses to Brucella Omp-2a by identifying epitope-specific IgM and IgG reactivity patterns [78]. The identification of three epitopes (1M/4G, 2M/5G, and 3M/10G) recognized by both IgM and IgG isotypes and five recognized only by IgG (exhibiting high specificity highlights their potential utility in the development of improved serological IgM/IgG or only IgG assays for brucellosis. These findings underscore the importance of epitope-level analysis in distinguishing between infection stages and improving diagnostic precision, thereby providing valuable insights into both clinical management and future vaccine design.

5. Conclusions

This study demonstrates the potential of synthetic peptides as effective substitutes for native proteins in immunodiagnostic assays for Brucellosis. Using peptide arrays, we established a high-throughput and cost-efficient platform for the identification of disease-specific linear B-cell epitopes. A total of three IgM and seven IgG epitopes were identified from the Omp-2a protein. Subsequently, eight peptides were selected for further validation through ELISA-based assays. These peptides demonstrated strong diagnostic performance, with ROC curve analyses confirming high sensitivity, specificity, and overall accuracy in distinguishing individuals infected with Brucellosis from healthy controls. Statistical analyses further reinforced the significant immunoreactivity of these peptides. Together, these findings highlight the diagnostic value of peptide-based assays and support their application in the development of reliable, specific, and scalable serological tests for Brucellosis.