Innate Determinants of Microbial Immunity

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

We have previously demonstrated (Namasivayam et al, 2017, 2019) that anti-tuberculosis therapy (ATT) causes a distinct alteration in the composition of the intestinal microbiota that is long-lasting in both mice and humans, and which could potentially impact host health. In this regard, it has been noted that individuals who have received ATT display an increased risk for re-infection and a previous clinical study using PBMCs reported that certain Mtb specific T cell epitopes were recognized poorly by individuals treated for TB in comparison to an untreated, latently infected group. Interestingly, these same epitopes had higher homology to peptides identified in the microbiome (Scriba et al. Am J Respir Crit Care Med, 2017). Based on this evidence, it was hypothesized that depletion of microbiota taxa with peptides that cross-react with Mtb following ATT might affect the ability to maintain long-term resistance to TB. To test this hypothesis, we generated a peptide pool restricted to mouse MHC-IAb in collaboration with Cecilia Lindestam Arlehamn at the La Jolla Institute of Immunology. The pool consisted of 288 peptides (MTB288) representing 79 Mtb antigens. Of these, 68/288 peptides were found to have high sequence homology to microbiota epitopes (MTB68). We found that the MTB288 peptide pool produced a potent IFN gamma+TNFalpha+ response following ex-vivo stimulation of Mtb-infected murine lung cells. However, mice that had completed ATT did not display a diminished response to MTB68 pool in comparison to ESAT-6, an immunodominant Mtb specific peptide widely used to read out Mtb-specific T cell responses. This data from the mouse model argues against the hypothesis that the increased susceptibility of previously antibiotic treated patients to Mtb infection is due to a diminished response to epitopes in the pathogen that cross-react with those in the microbiome. Interestingly, the MTB288 as well as MTB68 pools elicited stronger T cell responses than that observed with ESAT-6 following Mtb-infection alone. Moreover, several peptides in the MTB68 pool independently elicited a T cell response that, while lower than that seen with ESAT-6, was comparable to that observed following stimulation with other widely used Mtb-specific peptides such as Ag85B and EsxG. Thus, we have identified novel murine Mtb epitopes with high sequence homology to microbiome peptides and in recent work have successfully developed and tested tetramers for a number of these peptides. These resources can now be utilized to further elucidate the influence of microbiota peptides in the T cell response to Mtb infection, such as in the investigation of the presence and role of primed tolerogenic T cells that rapidly mount a response following Mtb infection. In a second project investigating the consequences of ATT on the microbiome performed in collaboration with Veronique Dartois (Center for Discovery and Innovation, Hackensack Meridian Health), we asked if TB drug absorbance is altered in mice with ATT induced dysbiosis. In the experimental model employed, C57BL/6 mice were pre-treated with HRZ (Isoniazid/Rifampin/Pyrazinamide- a drug combination used clinically for ATT) or a cocktail of broad-spectrum antibiotics, vancomycin, ampicillin, metronidazole and neomycin (VANM). VANM treatment has previously been shown to deplete the gut microbiota and has been used as a surrogate germ-free mouse model and this group was included as a control to test whether the microbiota as a whole plays any role in the absorbance of oral TB antibiotics. Following a pre-treatment period of 4 weeks, bio-availability of the rifampicin, moxifloxacin, isoniazid and pyrazinamide was assessed in mouse plasma for a period of 12 hours. We observed a significant decrease in the plasma concentration of rifampicin and moxifloxacin in the VANM-treated mice, but no significant difference in the levels of these drugs following administration to HRZ-treated animals when each condition was compared with control mice not previously treated with antibiotics. Interestingly no alterations were observed in the plasma concentrations of isoniazid and pyrazinamide when tested in the same protocol. Parallel experiments in germ-free animals confirmed the observations made in the VANM-treated mouse model supporting the role of the microbiota in the observed effects. Together, these data indicated that while HRZ-induced dysbiosis by itself does not affect the bioavailability of TB antibiotics, the microbiota is important for the absorbance of some but not all of these drugs. This observation may be clinically relevant for patients on broad-spectrum antibiotics who acquire TB and are treated with ATT. In a related COVID-19 project, we asked whether the microbiome influences immune responses and host resistance to SARS-COV-2 in our mouse infection models. Following viral challenge, mice treated with broad spectrum antibiotics to disrupt the microbiota showed a significant increase in viral loads. This increase was further pronounced in older antibiotic treated animals in comparison to young and age-matched controls. Further, older mice receiving antibiotic treatment displayed reduced survival following viral challenge. Immunologically antibiotic-treatment animals, young and old, displayed diminished inflammatory responses in their pulmonary non-haemopoietic, myeloid and lymphoid compartments and especially in their viral specific CD8+ T responses. The latter observation is of particular interest as it suggests that the microbiome may influence the induction of adaptive immunity to secondary infection and possibly to COVID-19 vaccination. Together, these data support an important role for the microbiome in mounting an optimal response to SARS-CoV-2 infection.