Towards serological diagnosis of tuberculosis infection using a combination of in silico and empirical approaches

Objective:

Serological tests have been used successfully for the diagnosis of many infectious diseases, however, serological diagnosis of tuberculosis (TB) infection, whether in cattle or humans, is considered to be inconsistent and imprecise. Numerous serological based assays have been developed using a range of pure antigens, purified protein derivative, culture filtrates and sonicated antigens, and although potential candidate antigens have been identified, no single antigen was found to achieve sufficient sensitivity. Combining multiple antigens was found to increase both sensitivity and specificity. Advances in peptide research has led to the development of serological assays that use peptide mimotopes to diagnose TB infection. The use of peptide mimics rather than the whole immunogen is more cost effective and standardised, and has been found to improve the low specificity of ELISA generally observed in tuberculosis diagnosis. The aim of this research is to use in silico and empirical approaches to map key epitopes of Mycobacterium tuberculosis complex proteins and cell surface structures and to determine if identified mimitopes could be used for serological diagnosis of TB infection.


Key results/ findings:
Gamma irradiated whole cells of M. tuberculosis H37Rv and M. bovis AF2122/97, ethanol-extracted surface antigens of these cells, and identified secreted antigens were used as targets for binder production. Selected binders from each target were tested for sensitivity and specificity by ELISA and then reversed biopanned. The peptides identified by reverse biopanning were characterised by phage ELISA and computational biology techniques. Potential candidate mimitopes were synthesised and tested for activity by ELISA. A selection of peptide mimitopes of mycobacterial proteins were then used to evaluate cattle serum samples previously characterised by a range of traditional TB detection assays. The results indicated that peptide mimics could be used to identify infected cattle but that further work is required to refine the assay.

Significance/Impact of study:
Sensitive and specific serological diagnosis of TB infection would provide a stepping stone to controlling tuberculosis disease not only in cattle but in humans. The nature of the cell surface of pathogenic mycobacteria combined with their inherent pathogenicity has resulted in gaps in knowledge in relation to the complex immuno-biology of TB infection. Precise and consistent serological diagnosis relies on identification of key antigens covering all stages of infection. In this study a novel method to identify important antigens of pathogenic mycobacteria combined with in silico and empirical approaches have identified a promising route for investigating serological diagnosis of TB infection.