Background Tuberculosis (TB) is still a major global health problem with about one-quarter of the global population infected with the causative pathogen, Mycobacterium tuberculosis (Mtb). The role of T-cells in the adaptive immune response against Mtb has been extensively studied with little information on the role of B-cells. B-cells produce antibodies and differentiate into plasma and memory B-cells. Plasmablasts are a subset of plasma cells only present in the peripheral circulation following an ongoing infection or vaccination. Immunoglobulin G’(IgG) especially IgG2 mounts more efficient immune response against bacterial infections, mainly attributed to the high affinity of IgG2 binding to the Fcγ receptor. Therefore, we hypothesised that Mtb-specific IgG +plasmablasts may be a useful biomarker of TB infection status.

Methods Ex-vivo B-cell enzyme-linked immunospot (ELISPOT) was used to identify plasmablasts responses to Mtb-specific antigens ESAT-6/CFP-10 (EC), together with non-specific Mtb purified protein derivative (PPD) and a positive (total IgG) and negative (media only) control from adults with active TB pre- and post-treatment (n=20) or with latent TB infection (LTBI; n=20) in The Gambia.

Results Frequencies of Mtb-specific plasmablasts were significantly higher in active TB cases pre-treatment compared to post-treatment (p<0.0001) and LTBI with no difference seen following PPD stimulation. Interestingly, total IgG +cells were lower in the cases at recruitment but increased following treatment indicating the relative proportion of Mtb-specific responses were also significantly different (p=0.034) prior to therapy.

Conclusion These data show that B-cell responses are differentially modulated during active and latent TB infection, suggesting that plasmablasts may be a useful biomarker for TB infection in TB-endemic settings.