Potential T cell epitopes of Mycobacterium tuberculosis that can instigate molecular mimicry against host: implications in autoimmune pathogenesis

Abstract

Background: Molecular mimicry between microbial antigens and host-proteins is one of the etiological enigmas for the occurrence of autoimmune diseases. T cells that recognize cross-reactive epitopes may trigger autoimmune reactions. Intriguingly, autoimmune diseases have been reported to be prevalent in tuberculosis endemic populations. Further, association of Mycobacterium tuberculosis (M. tuberculosis) has been implicated in different autoimmune diseases, including rheumatoid arthritis and multiple sclerosis. Although, in silico analyses have identified a number of M. tuberculosis specific vaccine candidates, the analysis on prospective cross-reactive epitopes, that may elicit autoimmune response, has not been yet attempted. Here, we have employed bioinformatics tools to determine T cell epitopes of homologous antigenic regions between M. tuberculosis and human proteomes.

Results: Employing bioinformatics tools, we have identified potentially cross-reactive T cell epitopes restricted to predominant class I and II alleles of human leukocyte antigens (HLA). These are similar to peptides of mycobacterial proteins and considerable numbers of them are promiscuous. Some of the identified antigens corroborated with established autoimmune diseases linked with mycobacterial infection.

Conclusions: The present study reveals many target proteins and their putative T cell epitopes that might have significant application in understanding the molecular basis of possible T cell autoimmune reactions during M. tuberculosis infections.

Figure 1

Identification of HLA class I and II restricted T cell epitopes from host proteins that shared similarity with M. tuberculosis antigens. Protein sequences of M. tuberculosis were subjected to BLAST search with the human proteome for identifying similarity regions. The peptides were analyzed for HLA binding for predominantly occurring HLA class I and II alleles using NetMHC server as described in methods. Peptides binding to HLA class I and class II molecules were considered to be CD8 T cell and CD4 T cell epitopes respectively. The peptides were classified based on predicted IC₅₀ value as strong binders (IC₅₀ ≤ 500), weak binders (500 ≤ IC₅₀ ≤ 5000) and non-binders (IC₅₀ ≥ 5000). The total number of binders (strong and weak) for each allele is represented as bar diagrams for (A) HLA class II and (B) HLA class I molecules.

Figure 2

CD4 T cell epitopes of host proteins share resemblance with different classes of M. tuberculosis antigens. The absolute number of host-peptides restricted to different HLA class II alleles that showed similarity with the different classes of mycobacterial proteins is represented as pie charts for (A) antigenic; (B) structural; (C) metabolic proteins of M. tuberculosis.

Figure 3

The relative fractions of autoreactive CD4 T cell epitopes exhibiting similarity with different categories of antigens of M. tuberculosis. Bar diagram indicates the relative percentages of HLA class II binders from host-antigens sharing sequence corresponding to different categories of antigenic proteins of M. tuberculosis.

Figure 4

A considerable fraction of peptides from autoantigens demonstrate strong HLA class II binding. Bar diagram depicts the percentage of strong binders (IC₅₀ ≤ 500) among the total binders from host peptides that shared similarity to M. tuberculosis antigens.

Figure 5

CD8 T cell epitopes of host antigens display similarity with different classes of M. tuberculosis antigens. The absolute number of host-peptides restricted to different HLA class I alleles that showed similarity with the different classes of M. tuberculosis proteins is represented as pie charts for (A) antigenic; (B) structural; (C) metabolic proteins.

Figure 6

The relative fractions of autoreactive CD8 T cell epitopes exhibiting cross-reactivity to different categories of antigens of M. tuberculosis. Bar diagram indicates the relative percentages of HLA class I binders of host-antigens sharing sequence similarity with the different categories of antigenic proteins of M. tuberculosis.

Figure 7

The peptides identified from host proteins reveal strong HLA class I binding. Bar diagram depicts the percentage of strong binders (IC₅₀ ≤ 500) among the total identified binders from host-peptides that shared cross-reactivity with M. tuberculosis antigens.