Rational combination of peptides derived from different Mycobacterium leprae proteins improves sensitivity for immunodiagnosis of M. leprae infection

Abstract

The stable incidence of new leprosy cases suggests that transmission of infection is continuing despite the worldwide implementation of multidrug therapy programs. Highly specific tools are required to accurately diagnose asymptomatic and early stage Mycobacterium leprae infections which are the likely sources of transmission and cannot be identified by using the detection of antibodies against phenolic glycolipid I. One of the hurdles hampering T-cell-based diagnostic tests is that M. leprae antigens cross-react at the T-cell level with antigens present in other mycobacteria, like M. tuberculosis or M. bovis bacillus Calmette-Guerin (BCG). Using comparative genomics, we previously identified five candidate proteins highly restricted to M. leprae which showed promising features with respect to application in leprosy diagnostics. However, despite the lack of overall sequence homology, the use of recombinant proteins includes the risk of detecting T-cell responses that are cross-reactive with other antigens. To improve the diagnostic potential of these M. leprae sequences, we used 50 synthetic peptides spanning the sequences of all five proteins for the induction of T-cell responses (gamma interferon) in leprosy patients, healthy household contacts (HHC) of leprosy patients, and healthy controls in Brazil, as well as in tuberculosis patients, BCG vaccinees, and healthy subjects from an area of nonendemicity. Using the combined T-cell responses toward four of these peptides, all paucibacillary patients and 13 out of 14 HHC were detected without compromising specificity. The peptides contain HLA binding motifs for various HLA class I and II alleles, thereby meeting an important requirement for the applicability of diagnostic tools in genetically diverse populations. Thus, this study provides the first evidence for the possibility of immunodiagnostics for leprosy based on mixtures of peptides recognized in the context of different HLA alleles.

FIG. 1.

IFN-γ production of PBMC derived from Brazilian leprosy patients (PB, n = 6; Rx, n = 4; MB, n = 4) induced by five recombinant M. leprae unique candidate proteins or 20-mer peptides overlapping the entire sequence of each corresponding protein. Horizontal bars indicate median responses. Dotted line shows 100-pg/ml cutoff value defining a positive response. PHA, phytohemagglutinin.

FIG. 2.

IFN-γ production of PBMC derived from Brazilian HHC of MB leprosy patients (n = 14) induced by five recombinant M. leprae unique candidate proteins or 20-mer peptides overlapping the entire sequence of each corresponding protein. Horizontal bars indicate median responses. Dotted line shows 100-pg/ml cutoff value defining a positive response. PHA, phytohemagglutinin.

FIG. 3.

IFN-γ production of PBMC derived from Dutch TB patients (n = 10) induced by five recombinant M. leprae unique candidate proteins or 20-mer peptides overlapping the entire sequence of each corresponding protein. Horizontal bars indicate median responses. Dotted line shows 100-pg/ml cutoff value defining a positive response. PHA, phytohemagglutinin.

FIG. 4.

IFN-γ production of PBMC derived from Dutch BCG vaccinees (n = 11) induced by five recombinant M. leprae unique candidate proteins or 20-mer peptides overlapping the entire sequence of each corresponding protein. Dotted line shows 100-pg/ml cutoff value defining a positive response. Horizontal bars indicate median responses. PHA, phytohemagglutinin.

FIG. 5.

IFN-γ production against four selected peptides derived from ML2283, ML1989, ML1990, or ML2567 in PBMC derived from PB leprosy patients (n = 6); MB leprosy patients (n = 4); patients with leprosy Rx (n = 4); HHC (n = 14); healthy controls from the area of endemicity without (EC Mlep−) (n = 5) or with (EC Mlep+) (n = 7) in vitro T-cell responses to M. leprae extracts; Dutch healthy, PPD-negative control individuals (NEC) (n = 10); Dutch TB patients (n = 10); and Dutch BCG-vaccinated individuals (n = 11). For each test group, the number of IFN-γ responders versus the total number of individuals in the group is indicated below the x axis. Horizontal bars indicate median responses. Dotted line shows 100-pg/ml cutoff value defining a positive response.

FIG. 6.

IFN-γ production patterns of four selected M. leprae peptides from ML2283, ML1989, ML1990, or ML2567 induced in PBMC derived from all test groups. For each test group, the number of IFN-γ responders versus the total number of individuals in the group is indicated below the x axis. Horizontal bars indicate median responses. Dotted line shows 100-pg/ml cutoff value defining a positive response. P values for comparison between PB/TB, HHC/TB, and EC Mlep⁺/TB groups are shown. The MB group did not differ significantly from the control groups. Patient group abbreviations are defined in the text or in the Fig. 5 legend.