A postgenomic approach to identification of Mycobacterium leprae-specific peptides as T-cell reagents.

To identify Mycobacterium leprae-specific human T-cell epitopes, which could be used to distinguish exposure to M. leprae from exposure to Mycobacterium tuberculosis or to environmental mycobacteria or from immune responses following Mycobacterium bovis BCG vaccination, 15-mer synthetic peptides were synthesized based on data from the M. leprae genome, each peptide containing three or more predicted HLA-DR binding motifs. Eighty-one peptides from 33 genes were tested for their ability to induce T-cell responses, using peripheral blood mononuclear cells (PBMC) from tuberculoid leprosy patients (n = 59) and healthy leprosy contacts (n = 53) from Brazil, Ethiopia, Nepal, and Pakistan and 20 United Kingdom blood bank donors. Gamma interferon (IFN-gamma) secretion proved more sensitive for detection of PBMC responses to peptides than did lymphocyte proliferation. Many of the peptides giving the strongest responses in leprosy donors compared to subjects from the United Kingdom, where leprosy is not endemic, have identical, or almost identical, sequences in M. leprae and M. tuberculosis and would not be suitable as diagnostic tools. Most of the peptides recognized by United Kingdom donors showed promiscuous recognition by subjects expressing differing HLA-DR types. The majority of the novel T-cell epitopes identified came from proteins not previously recognized as immune targets, many of which are cytosolic enzymes. Fifteen of the tested peptides had > or =5 of 15 amino acid mismatches between the equivalent M. leprae and M. tuberculosis sequences; of these, eight gave specificities of > or =90% (percentage of United Kingdom donors who were nonresponders for IFN-gamma secretion), with sensitivities (percentage of responders) ranging from 19 to 47% for tuberculoid leprosy patients and 21 to 64% for healthy leprosy contacts. A pool of such peptides, formulated as a skin test reagent, could be used to monitor exposure to leprosy or as an aid to early diagnosis.

Diagnostic assays for leprosy based on T-cell epitopes.

To date, only a limited number of antigens have been described as specific for Mycobacterium leprae, and in many cases, homologues have subsequently been shown to exist in mycobacteria such as M. avium and M. intracellulare. A Leprosy Synthetic Peptide Skin Test Initiative was established by the Steering Committee on the Immunology of Mycobacteria of the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases, to investigate the potential of synthetic peptides that encode T-cell epitopes as diagnostic tools, which could be used to develop a skin-test reagent specific for leprosy. Such M. leprae-specific peptides should have unique amino acid sequences, or significant sequence-dissimilarity from those in other mycobacteria. Synthetic peptides, 15 amino acids long, were synthesised from 33 genes or open reading frames within the M. leprae genome. Tuberculoid leprosy patients from four leprosy-endemic countries, Brazil, Ethiopia, Nepal and Pakistan, were tested as subjects known to have been infected with M. leprae, and to make good T-cell responses to antigens of M. leprae; UK blood donors were used as non-exposed or non-infected subjects. Peptides inducing potentially specific responses in leprosy patients and not in UK controls, and those inducing cross-reaction responses, present in both leprosy patients and non-exposed, non-infected controls, were identified. A difference from the equivalent M. tuberculosis sequence of five or more amino acid residues did not, by itself, identify peptides that were M. leprae-specific, suggesting that many of these peptides may have homologues in environmental mycobacteria. To date, this approach has identified a number of peptides with greater than 90% specificity and 19-47% sensitivity, which are undergoing further specificity-testing. Such peptides would have great potential as T-cell reagents with which to monitor exposure to M. leprae within communities, formulated either as skin-test reagents, or as antigens for tests in vitro.

Leprosy patients with lepromatous disease recognize cross-reactive T cell epitopes in the Mycobacterium leprae 10-kD antigen.

T cell responses play a critical role in determining protective responses to leprosy. Patients with self-limiting tuberculoid leprosy show high T cell reactivity, while patients with disseminated lepromatous form of the disease show absent to low levels of T cell reactivity. Since the T cell reactivity of lepromatous patients to purified protein derivative (PPD), a highly cross-reactive antigen, is similar to that of tuberculoid patients, we queried if lepromatous patients could recognize cross-reactive epitopes in Mycobacterium leprae antigens as well. T cell responses were analysed to a recombinant antigen 10-kD (a heat shock cognate protein) which is available from both M. tuberculosis (MT) and M. leprae (ML) and displays 90% identity in its amino acid sequence. Lymphoproliferative responses were assessed to ML and MT 10 kD in newly diagnosed leprosy patients (lepromatous, n = 23; tuberculoid, n = 65). Lepromatous patients showed similar, but low, lymphoproliferative responses to ML and MT 10 kD, while tuberculoid patients showed much higher responses to ML 10 kD. This suggests that the tuberculoid patients may be recognizing both species-specific and cross-reactive epitopes in ML 10 kD, while lepromatous patients may be recognizing only cross-reactive epitopes. This was further supported by linear regression analysis. Lepromatous patients showed a high concordance in T cell responses between ML and MT 10 kD (r=0.658; P<0.0006) not observed in tuberculoid patients (r=0.203; P>0.1). Identification of cross-reactive T cell epitopes in M. leprae which could induce protective responses should prove valuable in designing second generation peptide-based vaccines.

Low serum HDL-cholesterol is associated with raised tumor necrosis factor-alpha during ENL reactions.

The concentrations of serum lipids and tumor necrosis factor (TNF) were measured in leprosy patients across the spectrum of the disease and in erythema nodosum leprosum (ENL) patients at the onset of the reaction and after the reaction had clinically subsided. Lepromatous/borderline lepromatous (LL/BL) patients had significantly higher serum triglyceride and lower HDL-cholesterol levels; there was no such change in the tuberculoid/borderline tuberculoid (TT/BT) patients. The household contacts (HC) of the LL/BL patients also had significantly lower serum HDL levels. ENL patients during the acute phase of the reaction had significantly lower total, LDL-, HDL-cholesterol levels compared to the stable LL/BL patients, and these changes were reversible to pre-ENL levels after the reaction had subsided. Serum TNF levels were significantly higher in household contacts and in LL/BL patients but were not statistically different in TT/BT patients. Serum TNF levels were also significantly higher during the acute phase of ENL, and declined after the clinical remission of the reaction to levels comparable with those of LL/BL patients. There was a significant negative correlation between serum TNF and HDL-cholesterol levels during and after ENL reaction. However, there was no such correlation between TNF and total or LDL-cholesterol levels in ENL patients. Our results suggest that the changes in HDL-cholesterol metabolism are a specific part of the host response to lepromatous leprosy and to the ENL reaction and may be mediated by increased TNF production.

Erythema nodosum leprosum is associated with up-regulation of polyclonal IgG1 antibody synthesis.

Erythema nodosum leprosum (ENL) is a serious complication of lepromatous (L) disease in leprosy. We have previously shown that of the four IgG subclasses, IgG1 and IgG3 Mycobacterium leprae-specific antibodies are significantly lower in leprosy patients during ENL reaction compared with untreated L patients. To see if this decrease results from a down-regulation of antibody synthesis during ENL, the frequency of antibody-secreting B cells (ABSC) in the blood compartment was determined by ELISPOT and related to serum immunoglobulin concentrations (microgram/ABSC). Control groups consisted of 16 patients with stable L disease and 32 healthy endemic controls (EC). Paired samples were analysed during acute ENLS (n = 13) and after the reaction had subsided to identify changes associated with ENL. Polyclonal (PC) IgG1 was elevated in L patients compared with EC (325 micrograms versus 180 micrograms). Interestingly, patients during acute ENL showed concentrations higher than L patients (419 micrograms), which decreased after the reaction had subsided (260 micrograms), indicating the transient nature of the antibody response. IgG2 antibodies showed the reverse trend and were lower during ENL and increased after the reaction had subsided. The mean concentrations for PC IgG3 and IgG4 antibodies were similar during ENL and after the reaction had subsided. Thus, decrease in M. leprae-specific IgG1 and IgG3 antibodies is not related to down-regulation of B cell responses. Identification of factors which regulate PC IgG1 antibody synthesis may provide additional insights into determinants of ENL reactions.

Induction of Th1 cytokine responses by mycobacterial antigens in leprosy.

Twelve mycobacterial antigens were compared for induction of gamma interferon (IFN-gamma) secretion by human blood mononuclear cells of patients with leprosy. Fractionated Mycobacterium leprae antigens containing cell wall proteins or cytosolic and membrane proteins induced good IFN-gamma responses in tuberculoid leprosy patients. Lipoarabinomannan from M. tuberculosis Erdman and M. leprae mycolylarabinogalactan peptidoglycan were the poorest IFN-gamma inducers.