Diagnosis of the Multiepitope Protein rMELEISH3 for Canine Visceral Leishmaniasis

Abstract

Canine visceral leishmaniasis (CVL) is a major zoonosis that poses a growing challenge to public health services, as successful disease management requires sensitive, specific, and rapid diagnostic methods capable of identifying infected animals even at a subclinical level. The objective of this study was to evaluate the performance of the recombinant chimeric protein rMELEISH3 as an antigen in ELISA assays for the robust diagnosis of CVL. The protein was expressed in a bacterial system, purified by affinity chromatography, and evaluated through a series of serological assays using serum samples from dogs infected with Leishmania infantum. ROC curve analysis revealed a diagnostic sensitivity of 96.4%, a specificity of 100%, and an area under the curve of 0.996, indicating excellent discriminatory power. Furthermore, rMELEISH3 was recognized by antibodies present in the serum of dogs with low parasite loads, reinforcing the diagnostic potential of the assay in asymptomatic cases. It is concluded that the use of the recombinant antigen rMELEISH3 could significantly contribute to the improvement of CVL surveillance and control programs in endemic areas of Brazil and other countries, by offering a safe, reproducible and effective alternative to the methods currently recommended for the serological diagnosis of the disease.

1. Introduction

Canine visceral leishmaniasis (CVL) is an important zoonosis caused by Leishmania infantum [1], an intracellular protozoan that is transmitted mainly by bites of sand flies of the genus Lutzomyia, mainly of the species Lutzomyia longipalpis in Brazil [2,3,4]. Domestic dogs are considered the main reservoir of the parasite in urban and peri-urban areas and, consequently, contribute significantly to maintaining the transmission cycle [5].

In recent decades, CVL has expanded to regions of Brazil that were not previously endemic, posing a growing challenge to public health services and veterinary surveillance agencies. Currently, CVL is endemic in at least 25 Brazilian states, with foci recorded in the North, Northeast, Southeast, and Central-West regions. According to recent data from the Brazilian Ministry of Health, more than 4000 positive cases per year have been recorded in the country, with seroprevalence rates exceeding 20% in some highly endemic municipalities [5,6].

Control of CVL involves integrated actions, including mitigation of vector transmission, canine vaccination, and, most importantly, serological diagnosis with appropriate management of infected dogs. Although the identification of amastigote forms of the parasite in tissue samples by microscopic examination is considered an important parasitological diagnosis of leishmaniasis [7], this method has limited application in large-scale campaigns due to its invasive nature and the need for specialized personnel. The main diagnostic methods recommended by the National Leishmaniasis Surveillance and Control Program [8] for application in public campaigns are the rapid immunochromatographic test based on a dual-pathway platform (DPP^®) and the enzyme-linked immunosorbent assay (ELISA). However, these tests have some limitations, such as low sensitivity in asymptomatic dogs, variations in immune response among animals, and possible cross-reactions with other infectious agents [9,10].

Parasite burden plays a critical role in the progression and clinical presentation of CVL. Dogs with high parasite burdens often develop severe clinical signs and exhibit stronger humoral responses, facilitating detection by serology or PCR. In contrast, a low parasite burden—especially in asymptomatic or early stages—impairs immune activation and results in lower antibody levels, reducing the sensitivity of serological diagnostics such as ELISA. For example, Carvalho et al. [11] found that among asymptomatic dogs with positive PCR results, nearly 18% tested negative on standard ELISA assays. Similarly, Reis et al. [12] showed that dogs with low tissue parasite density exhibit immunological profiles consistent with low antibody titers and altered leukocyte patterns. These findings highlight the diagnostic challenges posed by low-parasitemic infections and justify the development of antigens capable of detecting infections in these cases.

Asymptomatic dogs naturally infected with L. infantum often exhibit a predominance of T-cell-mediated immune responses, with higher levels of IFN-γ and nitric oxide production and lower levels of IL-10 and IL-4. These responses contribute to parasite control while reducing or delaying the production of circulating antibodies, which limits the performance of conventional serological tests in detecting subclinical infections [13,14]. This highlights the need for diagnostic antigens that can increase detection sensitivity, especially in cases of low parasite burden.

In the context of serodiagnostic testing, the use of recombinant antigens is a promising alternative strategy to improve the accuracy of serological assays. Advances in bioinformatics and immunoinformatics have allowed the in silico identification of potentially immunogenic B and T epitopes, thus enabling the development of chimeric proteins that combine different antigenic regions with greater diagnostic potential [15,16]. Such recombinant proteins offer several advantages, including consistent standardization, improved reproducibility, absence of contaminants from other parasite antigens, and the possibility of large-scale production. Furthermore, multi-epitope antigens allow for a broader spectrum of immune response and are especially useful for detecting antibodies, even in dogs with low parasitemia [17].

The aim of this study was to evaluate the diagnostic performance of the chimeric protein rMELEISH3, which combines selected epitopes of immunogenic proteins from L. infantum, by applying an in-house ELISA protocol designed to detect IgG antibodies in serum derived from naturally infected dogs. The rMELEISH3 protein is a novel, independently designed molecule, structurally independent of previously described MELEISH antigens, and was constructed using bioinformatics strategies for epitope prediction and design. This is the first study to investigate its potential for the serological diagnosis of canine visceral leishmaniasis.

2. Materials and Methods

2.1. Construction of rMELEISH3

The recombinant protein rMELEISH3 was previously developed by the Laboratory of Microorganism Biotechnology at the Federal University of São João Del-Rei. The amino acid sequence of rMELEISH3 was submitted to the Protein Homology Recognition Engine V 2.0 (Phyre2) server for intensive modeling, after which the protein structure was visualized using UCSF Chimera software version 1.11.2. The synthetic gene was custom synthesized by Epoch Biosciences using codons for Escherichia coli and cloned as an NdeI/XhoI fragment into pET21a in-frame with a C-terminal histidine tag to allow protein purification by affinity chromatography. The resulting plasmid was used to transform E. coli BL 21(λ) competent cells. DE3), as described below. The DNA and amino acid sequences for the entire synthetic genetic construct are patented (under Brazilian patent no. BR 10 2023 017980 0) and therefore cannot be disclosed at this time.

2.2. Expression and Purification of rMELEISH3

Chemocompetent cells of Escherichia coli BL21(λDE3) strains pLysS and pLysE were prepared by washing with 0.1 M CaCl2 and stored in 15% glycerol at −80 °C. Transformation of cells with the vector was performed by heat shock at 42 °C for 90 s, followed by incubation in Luria—Bertani (LB) broth at 37 °C for 1 h. Treated cells were plated on LB agar medium containing ampicillin (100 µg/mL) and incubated for 16 to 18 h at 37 °C. Transformed colonies were grown to ∼0.6 OD₆₀₀ in LB broth containing ampicillin, and recombinant protein expression was induced with 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) under incubation for 24 h at 17 or 37 °C. Cell suspension samples collected before and after induction were centrifuged, and the pellets analyzed. To evaluate protein expression, cells were lysed by sonication, centrifuged, and the soluble and insoluble fractions were subjected separately to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by Coomassie brilliant blue staining. Recombinant protein expression was confirmed by Western blotting (WB) of the proteins separated from both fractions on nitrocellulose membranes and immunodetection with anti-His antibodies. The His-tagged recombinant tobacco etch virus (TEV) protein, known to contain the histidine tag at its N-terminus, was used as a positive control for the detection of the His epitope.

Recombinant rMELEISH3 was purified by resolubilizing the insoluble protein fraction in denaturing lysis buffer containing 6 M urea and separating the His-tagged protein by affinity chromatography on nickel-nitrilotriacetic acid (Ni-NTA) resin under denaturing conditions. Fractions eluted with 200 mM imidazole were analyzed by SDS-PAGE and dialyzed in 10 mM ammonium carbonate-bicarbonate buffer (pH 8.0) at 4 °C using a 2 kDa pore membrane. The buffer was changed several times until the solution appeared translucent. The concentration of purified rMELEISH3 was determined by the Bradford method with bovine serum albumin (BSA; Sigma-Aldrich, St. Louis, MO, USA) as a standard [18].

2.3. Canine Serum Samples

Serum samples used in the assays were provided by the research group of the Laboratory of Biology of Cellular Interactions of the Federal University of Minas Gerais and were derived from a panel of 47 dogs originating from the municipality of Porteirinha, MG, Brazil, and naturally infected with L. infantum. The animals were selected by convenience sampling, which prioritized dogs that presented clinical signs of CVL identified during home visits. The panel was composed mainly of male dogs (57.4%), with predominantly short hair (93.6%) and presenting peridomiciliary behavior (76.6%). The average weight of the animals in the panel was 12.46 kg (range 1.4–34 kg) and the average age was 6.29 years (range 2–11 years) [19].

Confirmation of L. infantum infection was based on positive diagnoses in DPP and ELISA tests. The parasite load (parasites/μg of tissue DNA) in the bone marrow was determined by quantitative polymerase chain reaction (qPCR) using the kinetoplastid minicircle DNA (kDNA) as a target. The qPCR results allowed the classification of animals into three categories: low load (0.07 to 0.35 parasites/μg of tissue DNA), medium load (2.82 to 14.99 parasites/μg of tissue DNA) and high load (79.6 to 2677.3 parasites/μg of tissue DNA) [19].

2.4. ELISA Protocol

Recognition of the rMELEISH3 antigen by antibodies present in serum samples from dogs infected with L. infantum was evaluated using an in-house ELISA protocol developed at the Laboratory of Biology of Cellular Interactions [20].

Assays were performed using 96-well high-affinity polystyrene plates (Sarstedt, Nümbrecht, Germany) that were sensitized with the rMELEISH3 antigen by overnight incubation at 4 °C, followed by triplicate washing with 1X phosphate-buffered saline (1X PBS) containing 0.05% Tween 20 (PBS-Tween). The wells were subsequently blocked with 200 μL of 1X PBS containing 5% BSA, incubated for 1 h at 37 °C, and washed again. Serum samples for assay were diluted in PBS-Tween (1:80; v/v), distributed into the respective wells, and incubated for 1 h at 37 °C. After further washing, a 100 μL aliquot of horseradish peroxidase-conjugated anti-total IgG secondary antibody (anti-mouse IgG-(H+L)-HRP; Bethyl Laboratories, Montgomery, TX, USA) diluted in PBS-Tween (1:10,000; v/v) was added to each well, followed by incubation for 60 min at 37 °C and further washing. Detection was performed by adding 100 μL of the ready-to-use chromogenic substrate 3,3′,5,5′-tetramethylbenzidine plus hydrogen peroxide (Scienco Biotech, Lages, SC, Brazil) to each well, followed by incubation for 10 min at room temperature. The reaction was stopped by the addition of 2.5 M sulfuric acid and absorbance was measured at 450 nm using a Multiskan™ FC microplate photometer (Thermo Fisher Scientific, Waltham, MA, USA).

2.5. Assessment of the Accuracy of rMELEISH3 Antigen in ELISA

Preliminary ELISA experiments were performed to establish the most effective concentration of rMELEISH3 antigen to differentiate samples derived from L. infantum-infected (n = 16) and uninfected (n = 8) dogs. The assays were performed with rMELEISH3 concentrations of 100, 200, 300, and 400 ng/well, and the accuracies were evaluated by constructing ROC (Receiver Operating Characteristic) curves.

After selecting the best antigen concentration, additional ELISA experiments were performed to evaluate the accuracy of rMELEISH3 antigen as a diagnostic tool for CVL. The assays were performed with 23 L. infantum-positive serum samples from dogs with different parasite loads and comprised nine samples with low load, eight samples with medium load, and six samples with high load. Additionally, five positive and eight negative control samples, as determined by PCR, were used in the experiments [17]. The diagnostic accuracy of rMELEISH3 in detecting antibodies against L. infantum was evaluated using ROC curves.

2.6. Data Analysis

MedCalc^® version 7.3.0.0 (Ostend, West Flanders, Belgium) was used to determine the cutoff point (positivity threshold) and construct the ROC curves. The cutoff point was calculated from the mean absorbance (ABS) at 450 nm of the negative and positive samples, so that the samples were classified as positive (ABS > cutoff point) and negative (ABS < cutoff point). ROC curves were used to evaluate the performance of the serodiagnostic test through sensitivity (ability to correctly detect positive samples) and specificity (ability to correctly identify negative samples), with both parameters ranging from 0 to 100%. Furthermore, the area under the curve (AUC), which ranges from 0 to 1, was used as an overall measure of antigen efficacy. One-way analysis of variance (ANOVA) was used to compare the mean values of the different experimental groups. When ANOVA indicated statistically significant differences (p <0.05), Tukey’s post hoc tests for multiple comparisons were applied to identify which specific groups differed significantly from each other. The results of all analyses were presented in graphs prepared with the aid of GraphPad Prism 9.0 software (La Jolla, CA, USA).

3. Results and Discussion

3.1. Efficiency of E. coli Cell Transformation

Successful bacterial transformation was confirmed by the appearance of colonies on LB agar medium supplemented with ampicillin, a selective marker for recombinant clones carrying the pET21a expression vector. Ampicillin resistance is conferred by the expression of the bla gene in the vector, which encodes β-lactamase, an enzyme that occurs in the periplasm of bacteria and catalyzes the hydrolysis of the antibiotic’s β-lactam ring. Although ampicillin is the most commonly used antibiotic in the selection of recombinant plasmids, kanamycin, chloramphenicol, and tetracycline are also widely used for this purpose [21]. It was observed that transformation of the E. coli strain BL21(λDE 3)pLysE resulted in a higher number of colonies compared to its pLysS counterpart; therefore, all experiments were conducted exclusively with the former strain. E. coli BL21 cells are ideal for use with T7 promoter-based expression vectors because they encode an RNA polymerase required for T7 activation [22]. Due to excellent growth in minimal media, BL21 cells are the preferred choice in biochemical, functional, and structural studies [23].

3.2. Expression of rMELEISH3 in Bacterial Cells

The expression of the rMELEISH3 antigen was induced by the addition of IPTG, a non-metabolizable lactose analogue that promotes the synthesis of T7 RNA polymerase and, consequently, the corresponding gene of interest [24]. Antigen expression was confirmed by SDS-PAGE (Figure 1A) and Western blotting (Figure 1B) using anti-His antibodies against the His tag attached to the C-terminus of the protein. PAGE separates proteins by molecular mass, while the presence of SDS denatures the protein and imparts a negative charge proportional to the mass, thus promoting migration to the positive pole [25].

A discrepancy was observed between the theoretical molecular mass of the rMELEISH3 protein (~32.15 kDa) calculated by the Compute pI/Mw tool (Swiss Institute of Bioinformatics, Basel, Switzerland) and its apparent mass in SDS-PAGE (~55 kDa). This anomalous migration may be related to the presence of disordered or highly negatively charged regions that alter the interaction with SDS and electrophoretic mobility [26]. Other factors may also interfere with protein migration, such as the presence of urea residues from purification under denaturing conditions [26] and the inclusion of His affinity tags that slow down electrophoretic mobility due to changes in the charge/mass ratio or tertiary structure [27,28]. Similar inconsistencies have been reported regarding the expression of recombinant proteins in E. coli, including, for example, the viral antigens of hepatitis B, rubella, and hepatitis C [29].

Comparison between rMELEISH3 expression levels at 17 and 37 °C indicated that protein production is temperature dependent, since higher yields were observed at the lowest temperature tested.

3.3. Characterization of Purified rMELEISH3

The rMELEISH3 protein was readily purified by affinity chromatography on Ni-NTA resin under denaturing conditions using a lysis buffer containing urea and elution with imidazole. The predominance of the band corresponding to rMELEISH3 in fractions E1 to E5 can be clearly observed in Figure 2, indicating that most of the protein was eluted after the washing steps and suggesting that the interaction between the protein’s His-tag and the Ni-NTA matrix was efficient, specific, and stable. The effectiveness of the protocol employed for the purification of rMELEISH3 is demonstrated by the amount of protein recovered (300.72 ng/µL), while the high degree of purity is demonstrated by the absence of additional bands in the eluted fractions.

Although the purified protein shown in Figure 2 migrates slightly below the range indicated in Figure 1A, this variation may result from differences in electrophoretic migration between gels and the presence of bacterial proteins in the expression samples. The identity of the purified protein was confirmed by anti-His Western blotting (Figure 1B). Such migration anomalies by SDS-PAGE have been reported in other studies due to factors such as detergent binding and protein conformation [30,31].

Purification of recombinant proteins expressed in heterologous hosts, such as E. coli, is commonly performed under denaturing conditions to ensure that any inclusion bodies (insoluble aggregates) formed during bacterial expression are solubilized [21,27]. Although denaturation destroys the native three-dimensional structure of the protein, the linear epitopes remain intact and are sufficient for recognition by antibodies in serological tests. This is particularly relevant for diagnostic antigens, since exposure of linear regions of sequence can ensure specific binding with immunoglobulins [32].

3.4. Diagnostic Performance of rMELEISH3 Antigen in Detecting CVL

The evaluation of the antigenic potential of rMELEISH3 in the serodiagnosis of CVL was performed using a proprietary ELISA protocol and serum samples from previously characterized dogs. The results of the preliminary ELISA experiments (Figure 3) show that the rMELEISH3 antigen was efficient in discriminating between serum samples positive and negative for L. infantum at all concentrations (100 to 400 ng per well).

The accuracy of antibody detection against L. infantum was determined from ROC curves generated for each antigen concentration (Figure 4), all showing sensitivities greater than 87.5% and specificities of 100%. However, assays performed with 100 ng of rMELEISH3/well showed slightly superior performance (AUC = 0.969) in terms of sensitivity and specificity compared to the other concentrations, and this antigen concentration was selected for subsequent ELISA experiments.

Multiepitope recombinant proteins are promising alternatives for the serological diagnosis of CVL due to their high sensitivity even in asymptomatic dogs, as previously observed by Faria et al. [33]. The results obtained with rMELEISH3 showed that the performance of this antigen is comparable to that of other recombinant proteins described in the literature for the diagnosis of CVL. For example, the chimeric protein ChimB, composed of B-cell epitopes originating from four L. infantum proteins, achieved 100% sensitivity and specificity in ELISA tests performed with human and canine samples [34]. Similarly, the recombinant L. infantum protein rLc36 used at a concentration of 1.0 μg/mL showed 85% sensitivity and 71% specificity in an ELISA test, and was able to differentiate positive and negative serum samples with 76% accuracy [35]. Furthermore, the rMELEISH protein characterized by Dias et al. [17] achieved 100% sensitivity and specificity, while the LiHyQ protein from L. infantum was reported as a promising diagnostic tool for CVL, even in asymptomatic dogs [36]. These examples highlight the potential of multiepitope antigen constructs in the development of highly sensitive and specific diagnostic tests for CVL.

The results of ELISA tests involving the interaction between the rMELEISH3 antigen (100 ng/well) and serum samples derived from dogs with different parasite loads, together with those from positive and negative control dogs, are shown in Figure 5. ANOVA revealed that there were significant differences between groups [F(4, 31) = 29.96; p < 0.0001], indicating that the absorbance means varied significantly among the five groups tested. Because the F value represents the ratio of the observed between-group variability to the within-group variability, the higher the F value, the stronger the evidence for true differences between groups. Tukey’s post hoc tests confirmed that the differences between the negative and positive control samples were statistically significant (p < 0.05), as well as the differences between positive samples with high, medium and low parasite loads in the bone marrow. These results indicate that the rMELEISH3 antigen not only recognizes anti- L. infantum antibodies but also discriminates between samples with different reactivities.

Antibody detection in dogs with low parasite loads poses a challenge in the diagnosis of CVL using conventional serological tests, such as DPP and ELISA, as sensitivity tends to be reduced at low loads, especially in asymptomatic animals. For example, De Carvalho et al. [11] reported that although ELISA had a sensitivity of 95.2% for symptomatic dogs, the value dropped to 65.6% for asymptomatic dogs. This limitation is associated with low antibody production during the early stages of infection (asymptomatic dogs with low parasite load), when the immune response is predominantly cellular. Furthermore, the heterogeneous distribution of the parasite in tissues may make detection difficult even when molecular techniques are employed [15,36].

The results presented here show that the rMELEISH3 antigen presented high sensitivity, as it was able to detect antibodies in sera from dogs with low parasite loads (0.07 to 0.35 parasites/µg of tissue DNA), demonstrating the effectiveness of the assay even in cases of subclinical infection. This performance is comparable to that of the Leishmania amastigote—specific protein rLiHyp1, which has been shown to induce a robust immune response and has been considered a potential candidate for improving the serodiagnosis of asymptomatic and symptomatic CVL, as well as for the development of a vaccine [15]. The ability of rMELEISH3 to identify infection during the early stages is particularly relevant for the control of CVL, since even asymptomatic dogs are reservoirs of the parasite and perpetuate disease transmission [37].

The ROC curve constructed for the ELISA tests using rMELEISH3 at a concentration of 100 ng per well (Figure 6) shows that the assay presented a sensitivity of 96.4%, a fundamental value that ensures the detection of true positive samples, and a specificity of 100%, ensuring the correct identification of true negative samples. The AUC value of 0.996 indicates a near-perfect diagnostic performance in discriminating between reactive and nonreactive sera.

AUC is widely used to assess the accuracy of diagnostic methods, with values close to 1 indicating the high discriminatory ability of an assay. According to Swets [38], AUC values can be interpreted in terms of discriminatory power as no discrimination (AUC ≤ 0.5), low (0.5 < AUC ≤ 0.7), moderate (0.7 < AUC ≤ 0.9), high (0.9 < AUC < 1), and perfect (AUC = 1.0). While sensitivity represents the ability of an assay to correctly identify infected individuals (positive samples), specificity refers to the ability to prevent detection among uninfected individuals (negative samples) [39].

Based on these findings, it is important to consider the potential clinical implications of rMELEISH3 for the diagnosis of CVL. The use of rMELEISH3 in ELISA-based assays for CVL may represent an important step toward improving diagnostic accuracy, particularly in the detection of asymptomatic infections, which are crucial for controlling disease transmission. However, several challenges remain, including the risk of cross-reactivity with other pathogens, variability in antibody responses among dogs, and the need for multicenter validation studies using field samples. Future work should also focus on comparing rMELEISH3 with currently used recombinant antigens to assess its added value and feasibility for routine veterinary diagnostics.

4. Conclusions

The present study demonstrates that the recombinant rMELEISH3 protein presents high diagnostic performance for the serological detection of CVL by ELISA, as it was efficient in discriminating between dogs infected and uninfected with L. infantum and in detecting antibodies in those with low parasite loads. This characteristic is particularly significant for the identification of asymptomatic carriers who can spread the disease largely undetected and, consequently, undermine transmission control efforts. The test’s near-perfect accuracy was demonstrated by a sensitivity of 96.4%, a specificity of 100%, and an AUC of 0.996. For these reasons, the rMELEISH3 antigen is considered a strong candidate for clinical and epidemiological applications.

5. Patents

A patent application for the recombinant chimeric protein rMELEISH3 was submitted in Brazil under number BR 10 2023 019980-0, filed on 5 September 2023 [40].