Development of a recombinant protein based ELISA for the diagnosis of canine leptospirosis

Background

Leptospirosis is an emergent zoonotic disease caused by pathogenic Leptospira spp., and humans are regarded as an accidental or dead-end host [1]. The global incidence of severe human leptospirosis was estimated to be approximately 500.000 cases worldwide, although this is likely an underestimate due to the lack of a reliable diagnostic test. Mortality ranges from 10 to > 50% for Weil's disease or severe pulmonary hemorrhage syndrome (SPHS), respectively [2]. The microscopic agglutination test (MAT) is the standard test for this disease, however, it requires paired serum samples, is laborious, difficult to analyze and subjective [3]. Apart from the danger connected with rodents, which are the main vectors of leptospires, occurrence of the disease in dogs can generate a higher risk of infection for humans. Thus, a novel sensitive and specific serological test providing a rapid and secure diagnosis is urgently required for the laboratorial diagnosis of leptospirosis. In this study, 15 recombinant proteins from L. interrogans were expressed, purified and evaluated in an indirect ELISA using canine sera that were characterized by the MAT and a whole-cell Leptospira ELISA.

Methods

Fifteen recombinant polypeptides based on LigB, LigA LipL32, LemA, OmpL37, FlaA1 and FlaB1 were cloned and expressed in Escherichia coli, purified by Ni²⁺affinity chromatography, quantified and stored at -20 ºC until use. The recombinant proteins were assessed in a checkerboard ELISA to evaluate their capability discriminate between infected and non-infected canine sera, and to determine the optimal combinations of antigen, canine sera and anti-dog IgG and IgM HRP conjugate antibodies. After the initial screening, two recombinant proteins were selected and tested against a serum panel, 10 positive and 10 negative serum samples.

Results and conclusions

Following the preliminary screening using pooled sera, rLigBrep and rLipL32 were evaluated using a panel of individual canine serum samples. The optimal concentrations of antigen for the coating step was 100 ng/well. The optimal sera dilution was 1:200 for rLigBrep, 1:100 for rLipL32, 1:2500 for the IgG and 1:5000 for the IgM conjugates. Based on the negative sera, the average OD for rLigBrep was 0.24 ± 0.11 for the negative pool and the cut-off was calculated as OD = 0.47. 100% of the negative sera were negative and 70% of the characterized positive were identified as positive under these conditions. The average OD based on rLipL32 was 0.2 ± 0.08 for the negative pool, equivalent to a cut-off of OD = 0.36. 100% of negative sera were negative and 50% positive sera were identified as positive. The anti-dog-IgG conjugate exhibited the best specificity and sensitivity. The recombinant protein rLigBrep demonstrated potential as a diagnostic tool for canine leptospirosis, 70.0% sensitivity and 100% specificity. The native protein, LigB, is highly conserved in pathogenic Leptospira spp. and therefore represents an ideal candidate for further studies.