Follicular helper T (Tfh) cells are a distinct subset of CD4+ helper T (Th) cells that regulate the development of antigen-specific B cell immunity. Upon exposure to a foreign antigen, Tfh cells help B cells generate antibody-producing plasma cells and long-lived memory B cells. While the specific triggers required for Th1, Th2 or Th17 polarization are well defined, those responsible for Tfh differentiation are incompletely understood. In order to identify signals that could be used to promote Tfh responses, we have utilized an experimental model in which C57BL/6 or gene KO mice are immunized with either peptide or protein antigen emulsified in the oil and surfactant-only containing adjuvant IFA (Incomplete Freunds Adjuvant). Using this immunization protocol in wild-type mice we find that at the peak of the response in the draining lymph nodes, half of the antigen specific CD4+ T cells have differentiated into Tfh based on their expression of the chemokine receptor CXCR5 and the lineage defining transcription factor Bcl6 while little or no increase in Th1 or Th17 cells is observed. Similar results were observed with other oil and surfactant-only adjuvants. Thus, use of these relatively inert non-microbial adjuvants specifically promotes Tfh differentiation. Interestingly, both the overall number and the frequency of antigen specific Tfh were substantially decreased in IFA immunized MyD88-/- mice, but not in animals lacking TRIF, TLR4 or IL-1R signaling. A similar defect in Tfh polarization was observed in the absence of type I IFN signaling. Together, our results show that Tfh responses can be promoted by adjuvants that lack exogenous PAMPs through a pathway involving type I IFN and, paradoxically, the microbial sensing related adaptor molecule MyD88. Studies during the past decade have revealed a prominent influence of the commensal microbiota in determining immune responsiveness and resistance to infectious and inflammatory as well as malignant disease. In the case of M. tuberculosis (Mtb) infection such an influence could impact not only susceptibility to the pathogen but also re-activation of latent infection and the expression of clinical disease. Moreover, the standard treatment for tuberculosis involves the oral administration of antibiotics which themselves might be expected to alter the normal distribution of gut microflora. For these reasons we have launched a new research project examining in both experimental animal models and eventually in humans (1) the influence of the gut and pulmonary microbiota on Mtb infection and vice-versa (2) the effect of Mtb antibiotic treatment on the host microbiota and the possible effects of the resulting dysbiosis on immunological responsiveness to infection as well as the long range pharmacokinetics of the TB drugs themselves. In our work during the report period, we began by asking whether disruption of the gut microflora with orally administered broad spectrum antibiotics would influence aerosol administered Mtb infection. Interestingly, we found that when given orally these antibiotics which can reduce viability of Mtb in vitro failed to reduce pulmonary Mtb infection loads and if anything caused a small increase in bacillary burden. Perhaps more interesting, we found that antibiotic treatment of the gut flora caused a significant delay in the clearance of Mtb mediated by a standard (HRZ) regimen of TB drugs. In parallel studies we performed 16s rRNA sequencing of gut bacteria in fecal pellets from MTb infected mice following standard HRZ chemotherapy. We found that the TB drugs induced major changes in the distribution of gut bacterial species as early as 2 wks post-treatment. These changes included the expansion of species barely detectable before chemotherapy. Together these findings have provided us with platforms for studying both Mtb microbiota interactions as well as TB drug induced dysbioses of the gut flora.
Showing the most recent 10 out of 30 publications
