Prediction of Th1 and Cytotoxic T Lymphocyte Epitopes of Mycobacterium tuberculosis and Evaluation of Their Potential in the Diagnosis of Tuberculosis in a Mouse Model and in Humans.

Abstract

ABSTRACTLatent tuberculosis infection (LTBI) is the primary source of tuberculosis (TB) but there is no suitable detection method to distinguish LTBI from active tuberculosis (ATB). In this study, five antigens of Mycobacterium tuberculosis belonging to LTBI and regions of difference (RDs) were selected to predict Th1 and cytotoxic T lymphocyte (CTL) epitopes. The immunodominant Th1 and CTL peptides were identified in mouse models, and their performance in distinguishing LTBI from ATB was determined in mice and humans. Ten Th1 and ten CTL immunodominant peptides were predicted and synthesized in vitro. The enzyme-linked immunosorbent spot assay results showed that the combination of five Th1 peptides (area under the curve [AUC] = 1, P < 0.0001; sensitivity = 100% and specificity = 93.33%), the combination of seven CTL peptides (AUC = 1, P < 0.0001; 100 and 95.24%), and the combination of four peptide pools (AUC = 1, P < 0.0001; sensitivity = 100% and specificity = 91.67%) could significantly discriminate mice with LTBI from mice with ATB or uninfected controls (UCs). The combined peptides or peptide pools induced significantly different cytokine levels between the three groups, improving their ability to differentiate ATB from LTBI. Furthermore, it was found that pool 2 could distinguish patients with ATB from UCs (AUC = 0.6728, P = 0.0041; sensitivity = 72.58% and specificity = 59.46%). The combination of Th1 and CTL immunodominant peptides derived from LTBI-RD antigens might be a promising strategy for diagnosing ATB and LTBI in mice and patients with ATB and uninfected controls.IMPORTANCE Latent tuberculosis infection (LTBI) is a challenging problem in preventing, diagnosing, and treating tuberculosis (TB). The innate and adaptive immune responses are essential for eliminating or killing the mycobacteria. Antigen-presenting cells (APCs) present and display mycobacterium peptides on their surfaces, and recognition between T cells and APCs is based on some essential peptides rather than the full-length protein. Therefore, the selection of candidate antigens and the prediction and screening of potential immunodominant peptides have become a key to designing a new generation of TB diagnostic biomarkers. This study is the first to report that the combination of Th1 and CTL immunodominant peptides derived from LTBI-RD antigens can distinguish LTBI from active TB (ATB) in animals and ATB patients from uninfected individuals. These findings provide a novel insight for discovering potential biomarkers for the differential diagnosis of ATB and LTBI in the future.