IgG antibodies anti-LcrV of Yersinia pestis: inconsistent responses in confirmed plague patients from Madagascar

Abstract

Objective

Plague (Yersinia pestis infection) is a flea-borne zoonotic disease mainly affecting African countries, with most human cases reported from Madagascar and the Democratic Republic of Congo. Although curable, plague can be fatal without prompt treatment, highlighting the importance of reliable diagnostics. Current tests include bacteriological culture, PCR, and anti-F1 ELISA, however the anti-F1 serology has limitations due to the existence of F1-negative virulent strains. To address this, we developed a serological test detecting IgG antibodies against LcrV or V antigen, the main virulence factor of Y. pestis and further evaluated the developed test on clinical samples.

Results

V antigen was produced from the culture of V-pGEX-6P-2 clones and purified as GST-LcrV, which was functional for ELISA plate coating. The developed anti-V ELISA showed 60% sensitivity and 93.3% specificity when tested on confirmed plague patients’ serum samples from Madagascar. However, evaluation on convalescent sera collected from Day 1 to Month 3 post-infection revealed inconsistent anti-V antibody production. This suggests the anti-V ELISA is best used as a complementary test for plague diagnosis and during outbreak investigations. In addition, this study provides valuable insight into the humoral response diversity following Y. pestis infection, representing crucial information for plague vaccine development.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13104-025-07315-y.

Introduction

Plague is a flea-borne zoonosis caused by the gram-negative bacterium Yersinia pestis, still endemic in some parts of the world but mainly Africa. The World Health Organization has classified plague as a re-emerging disease since the 1990s. Between 2013 and 2018, over 90% of global human plague cases were reported from Madagascar and the Democratic Republic of Congo, accounting for 95% of plague-related deaths worldwide [1]. Although curable by antibiotherapy, plague can be fatal without prompt treatment, emphasizing the need for quick and reliable diagnostics for early detection of the disease [2], especially in Madagascar, where about 40% of the population is exposed to Y. pestis [3].

Several diagnostic tools are available including rapid diagnostic tests [2], bacteriology [4], molecular biology [5], and serology [6]. Since 2021, positive culture (Y. pestis strain isolation) or positive qPCR is confirmatory for plague [7]. However, bacteriology is time-consuming compared to other techniques (requires ≥ 48 h for results) and can be hampered by sample quality or prior antibiotic use, while molecular tools require well-trained staff and expensive equipment often unavailable in resource-limited countries.

An alternative confirmatory test is the enzyme-linked immunosorbent assay (ELISA) showing a four-fold rise in anti-F1 antibody titers in paired sera [7]. It is not sample quality-dependent and easy to implement in low-resource laboratories. Yet, this test has limitations: F1 is not essential for Y. pestis virulence, and rare F1-negative strains from confirmed plague cases have already been reported [8, 9]. Sera from these individuals would not be diagnosed by anti-F1 serology.

To address this, detecting antibodies against other Y. pestis markers is necessary. The low-calcium response V-antigen (LcrV or V antigen) is a promising candidate, and a key protective antigen for bubonic and pneumonic plague pathogenesis [10]. While most anti-V ELISAs have been used for antibody follow-up after experimental vaccination in mice [11, 12], here we developed an anti-V ELISA for IgG detection and evaluated the assay in serum samples from Malagasy confirmed plague patients.

Materials and methods

LcrV production and visualization

Recombinant V antigen was produced from Escherichia coli BL21 clones containing plasmids, which were formed by ligation of the gene encoding LcrV in the pGEX-6-P2 vector [13]. E. coli clones were kindly provided by the Defence Science and Technology Laboratory (DSTL), Porton Down, UK. After induction of gene expression, proteins were released using different methods to disrupt the bacterial cell wall: enzymatic, mechanical, and thermal. For enzymatic disruption, bacterial pellets were resuspended in PBS with 0.05% DNase (Sigma, 11284932001) and lysozyme (Sigma, 9001-63-2). For thermal disruption, bacterial pellets were washed, resuspended in a lysis buffer containing 1.5 M urea (Sigma, U5128), 1% protease inhibitor (Sigma, 539131), 50mM β2-mercaptoethanol (Sigma, M3148), 10% Triton (BioXtra, T9284), and 0.05% DNase, then subjected to freeze-thaw cycles. For mechanical disruption, bacterial pellets were resuspended in the same lysis buffer and homogenized using a dounce homogenizer. Proteins were separated from cellular debris by centrifugation and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Protein solutions were purified on columns specific to the tag (Glutathione-S-transferase or GST) fused with LcrV as previously described [13], allowing the elimination of other proteins and releasing either tagged or untagged LcrV antigen depending on the use of protease lysing the GST-V fusion protein. A LcrV-specific monoclonal antibody, termed MAb 7.3, and reported to protect mice against a fully virulent strain of Y. pestis [14] was used for the detection of LcrV after protein purification (Additional file 1).

Anti-V IgG ELISA

Detection of anti-V IgG was conducted by ELISA as previously described [11, 12] with few modifications. Briefly, microwells were coated overnight at 4 °C with 5 µg/ml of LcrV diluted in PBS. After washing with PBS-0.02% Tween 20, plates were blocked with 1% PBS-skimmed milk for 1 h at 37 °C and washed again. Patients’ sera diluted 1:1 in 1% PBS-skimmed milk were added in duplicate and incubated for 1 h. After washing, an HRP-conjugated secondary antibodies diluted at 1:20000 (Sigma-Aldrich, A8419) was added for 1 h at 37 °C and detected after addition of a chromogenic substrate ABTS (KPL). Results were determined by reading the optical density (OD) at 405 nm using an ELx800 spectrophotometer (Biotek, Winooski, VT) after 20 min of incubation. The positive control was a serum sample from a Malagasy confirmed plague case collected at Month 12 post-infection and positive for anti-F1 IgG. Negative controls were Malagasy human sera samples without plague infection history (n = 3) and negative for anti-F1 IgG. The threshold was predetermined by Receiver Operating Characteristic (ROC) curve analysis using sera from plague-confirmed and non-plague patients, with the best likelihood ratio of 15.38 (Additional file 2). A sample was considered positive when the ratio (net OD/ mean three negative controls + 3SD) ≥ 2.

Sensitivity, specificity and evaluation of the anti-V ELISA

The sensitivity of the anti-V ELISA was assessed by testing 30 sera samples from Malagasy confirmed plague patients (positive on PCR, bacteriology and with anti-F1 seroconversion). These sera were collected between Day 21 and Month 3 after clinical symptoms’ onset.

The specificity was determined with 30 sera from individuals living in a non- endemic plague area of Madagascar (Taolagnaro District). These samples tested negative for anti- F1 IgGs.

The developed anti-V ELISA was evaluated on 63 sera samples collected from 17 plague patients at different time points from the disease onset (Day 1, Day 14, Day 21, Month 3). These patients presented a bubonic form and originated from Ambositra and Manandriana Districts, within the main plague focus in the central highlands of Madagascar. Among the 17 patients, 8 were males and 9 females, with an age range of 4–58 years (median age 12 years). These patients presented a wider range of outcomes across various diagnostic tests (12 positive on PCR, bacteriology and anti-F1 serology; 4 positive on PCR and anti-F1 serology; 1 serum positive on anti-F1 serology only) compared to those tested in the sensitivity test. This sera assessment also allowed the follow-up of anti-V IgG antibodies kinetics.

Statistical analyses

The ROC curve analysis was generated using GraphPad Prism (version 10.3.1).

Specificity and sensitivity of anti-V IgG ELISA were calculated with 95% confidence intervals using R 3.6.2 [15]. The kappa coefficient (κ) was calculated to assess the agreement between the index (anti-V IgG ELISA) and the reference (anti-F1 IgG ELISA) tests and interpreted as follows: poor (< 0), slight (0.01–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), and almost perfect (0.81-1.00) agreement [16].

Logistic regression and Chi-square tests were used to assess anti-V IgG seropositivity with age and sex, respectively. Fisher’s exact test was performed to assess the association between seropositivity and health status. Significance was set at P < 0.05.

Results

LcrV production

Visualization of the protein produced with the three different extraction methods, by SDS-PAGE revealed that thermal method yielded better results than enzymatic method advised for protein release (Fig. 1a). In addition, protein purification using SDS-PAGE and following Ponceau staining showed the presence of GST-LcrV at the expected size of 63 kDa (Fig. 1b).

Fig. 1.

LcrV production using SDS-PAGE and Ponceau staining. a: Visualization of protein extracts from the 3 extraction techniques (thermal, mechanical and enzymatic) with positive bands colored in pink, the thermal method yielded more proteins (thicker band) at 63 kDa (V antigen or LcrV). b: Protein purification also showed the presence of GST-LcrV at the expected size of 63 kDa, M.W.= Molecular Weight

Untagging LcrV using protease digestion was unsuccessful in our hands however, it was recognized by the LcrV-specific monoclonal antibody MAb 7.3. Results were consistent with those obtained with the reference untagged V antigen (DSTL, UK) used at the same concentration (data not shown).

Therefore, a V antigen still fused to GST was used throughout the study.

Characteristics and application of the anti-V IgG ELISA

The sensitivity of anti-V ELISA for plague was 60% (95% CI: 42.3-75.4%). Of the 30 sera collected from individuals living in plague-free area, 28 were both negative on anti-F1 and anti-V IgG ELISA resulting in a specificity of 93.3% (95% CI: 78.7-98.8%) (Table 1). A kappa coefficient of 0.53 (95% CI: 0.3–0.7) was obtained (i.e., moderate agreement).

Table 1.

Sensitivity and specificity results of anti-V ELISA for plague diagnosis. Anti-F1 positivity refers to confirmed plague cases (n = 30) and the negative to non-plague patients (n = 30). The agreement between anti-V and anti-F1 IgG ELISA yielded a κ = 0.53

		Anti-F1 ELISA
		Positive	Negative	Total
Anti-V ELISA	Positive	18	2	20
	Negative	12	28	40
	Total	30	30	60

Among 63 sera samples from 17 plague patients at different follow-up times, 41 (65.0%) were anti-F1 IgG positive whereas only 15 (23.8%) were anti-V IgG positive showing that 10/17 patients (58.8%) did not develop detectable antibodies against LcrV during their infection. We found that neither age (P = 0.6213) nor sex (P = 1) was associated with anti-LcrV IgG positivity.

For anti-V IgG positive sera, kinetics showed that anti-V antibodies increased gradually from Day 1 to Day 14, reached a peak at Day 21 and then decreased progressively but still detectable until Month 3 for 41.2% of them (Fig. 2a). This trend is similar to anti-F1 IgG kinetics observed in these same patients. However, unlike anti-V IgG, 100% of them still have anti-F1 antibodies at Month 3 after symptoms’ onset (Fig. 2b).

Fig. 2.

Evolution of IgG anti-V (a) and anti-F1 (b) days and month after disease onset. Among 17 sera tested, only 47,2% showed seropositivity towards IgG anti-V using ELISA, whereas 100% had IgG anti-F1. A sample was considered as positive if ratio ≥ 2

Discussion

We developed an anti-V IgG ELISA using bacterial clones producing V antigen and further evaluated on suspected plague serum samples from Madagascar, the country most affected by plague. The V antigen used was a tag-fused protein, however previous studies confirmed that its functionality in anti-V ELISA remains intact [17, 18]. The GST tag did not interfere with antibody recognition. The anti-V ELISA had a sensitivity of 60% (95% CI: 42.3-75.4%) and specificity of 93.3% (95% CI: 78.7-98.8%). Assessed on sera from plague confirmed patients, the anti-V ELISA shows a κ of 0.53 (95% CI: 0.3-0.7%). This low sensitivity indicates 40% risk of the anti-V ELISA to misidentify confirmed plague patients while the high specificity supports its reliability in identifying non-plague patients. Although not relevant for plague diagnosis, this anti-V ELISA could still be used for V antigen humoral response assessment.

In this study, the evaluation of 17 bubonic plague patients’ sera samples using the developed anti-V ELISA showed that anti-V IgGs were not detected in more than half of these patients. This observation has already been reported in a previous study using protein microarray technique where 5 out of 7 sera samples from plague patients did not show any detectable anti-V antibodies [19]. This was also the case for human donors immunized with a live plague vaccine. Indeed, only 20.6% o these individuals produced antibodies targeting LcrV [20]. This positivity rate of anti-V antibodies has even been shown to fall to zero in mice vaccinated with Y. pseudotuberculosis containing F1 antigen-based vaccine [21]. Altogether, these results could indicate that in plague immunity, the role of LcrV antigen may be cellular rather than humoral. A T-response-stimulating epitopes within LcrV has already been identified [22]. The location of the V antigen could be one explanation. Unlike F1, LcrV is predominantly intracellular although it is expressed on the tip of the injectisome during infection and therefore may not promote a humoral response because inaccessible to B cell receptors. However, although anti-LcrV antibodies are not always detectable, this antigen is known to be a major asset in protection against plague, as demonstrated in a recent study [23].

While most studies on anti-V immune response have been carried out in mice in the context of plague vaccine development, our study is the first to report the humoral immune response against V antigen within individuals living in a plague-endemic area and where the same biovar of Y. pestis (biovar Orientalis) has circulated since its introduction. A similar study conducted by Bei Li and colleagues [19] suggested that the diversity of humoral responses toward LcrV was associated with patients coming from different regions and infected by different biovars of Y. pestis (biovar Antiqua and Orientalis). Interestingly, individuals with detectable anti-LcrV antibodies were mostly those with altered general condition as per their plague notification form, compared to those in relatively good health at the time of sampling (P = 0.0029). This observation deserves further investigation.

In conclusion, the IgG anti-V ELISA developed in this study can be used as a complimentary test to anti-F1 ELISA and serve as a tool to investigate humoral responses in plague patients. It may also help estimate the prevalence of other pathogenic Yersinioses (Y. enterocolitica, Y. pseudotuberculosis) since LcrV is a shared antigen among these species. Further research is needed to understand the protective role of anti-V and anti-F1 antibodies in plague immunity.

Limitations

The main limitation of our study could be the unavailability of known human positive controls for IgG anti-V at the beginning of the anti-V ELISA development. They were only identified after the process, possibly affecting assay sensitivity. In addition, the ROC curve was established using plague-confirmed samples as positives, not anti-V positive samples, which may have biased sensitivity and specificity estimates.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

DSTL

Defence Science and Technology Laboratory

SDS-PAGE

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

GST

Glutathione-S-transferase

OD

Optical density

IgG

Immunoglobulin G

ELISA

Enzyme-linked immunosorbent assay

ROC

Receiver Operating Characteristic

Author contributions

VA contributed to the conception of the work; AR, VA and MR participated in the design of the work; OHA contributed to the data acquisition and drafted the original manuscript; OHA, AR, MS, MR and VA contributed to the analysis and interpretation of data; EDW, MR and RS contributed to the funding acquisition; EDW and NJW provided the resources; AR, EDW, MS, RS, NJW, MR and VA reviewed the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported in part by the BactiVac Catalyst Project BVNCP5-02, 2021 and the Plague Unit - IPM (PA-14.71).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study has been performed in accordance with the Declaration of Helsinki. The use of human serum samples was approved by the Ethics Committee for Biomedical Research of the Malagasy Ministry of Public Health (N° 086 and 149 MSANP/SG/AMM/CERBM on July 27, 2021, and October 21, 2021, respectively). Informed consent was obtained from each participant or legal guardian.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.