As introduced in previous reports we have developed a murine model which recapitulates the key features of the IRIS disease occuring in ART treated HIV patients with mycobacterial co-infections. In this model originated by Dan Barber in the lab (Barber et al., Blood, 2010) we utilize T cell deficient TCR alpha -/- mice infected with M. avium, an important opportunistic pathogen associated with human IRIS. Following adoptive transfer with CD4+ T cells these animals develop a severe wasting disease with impaired lung function and rapidly succumb. In analyzing this form of immune reconstitution disease (IRD), we found that it requires Ag recognition and IFN-gamma production by the donor CD4 T cells. Last year, we reported that neutralizing IL-6 with monoclonal antibodies against IL-6 or its receptor, CD126, alleviated weight loss and prolonged the survival of mice undergoing IRIS. Using this model we have continued our work on the role of IL-6 in murine IRIS by examining its cellular source in vivo. Adoptive transfer of WT and IL-6 deficient CD4 T cells into M. avium infected TCR alpha KO mice leads to an identical course of wasting and mortality, indicating that IL-6 is not produced by the reconstituting CD4 T cells themselves. To investigate the role of IL-6 production by myeloid cells in the pathogenesis of IRIS, we have developed a novel experimental approach to generate T cell deficient mice selectively lacking IL-6 in the hematopoietic compartment. In this system, athymic nude mice are first lethally irradiated and reconstituted with WT or IL-6 KO bone marrow, and then infected with M. avium. After several months post-infection, we now have T cell deficient mice chronically infected with M. avium and selectively lacking IL-6 from monocytes, macrophages, neutrophils and dendritic cells. Using this approach we have found that the source of pathogenic IL-6 is also not a bone marrow derived cell. This indicates that CD4 T cell driven IRIS involves the activation of non-immune tissues, which are induced to produce proinflammatory cytokines, such as IL-6, that directly contribute to the pathogenesis of IRIS. We are currently working to identify the relevant peripheral tissues that produce IL-6 during T cell driven IRIS. Moreover, we can now use this this nude/chimera approach to investigate the role of additional myeloid inflammatory pathways in IRIS. We have also found that during mouse IRIS the reconstituting CD4 T cells adopt an unusual phenotype characterized by low levels of the Th1 defining transcription factor Tbet and extremely high levels of the transcription factor eomesodermin, which is known as a master regulator of the cytotoxic program in CD8 T cells and NK cells. This ectopic expression of eomesodermin by the vast majority of CD4 T cells may indicate that CD4 T cell adopt a novel differentiation state that has not been observed in any other context. This may be important in understanding the mechanisms of CD4 T cell driven IRIS as well as provide a useful diagnostic marker for IRIS in patients. Indeed, preliminary observations from our collaborative study in India indicates increased expression of eomesodermin in CD4+ T cells from patients developing TB-IRIS. As mentioned in previous reports, our group has also participated in longitudinal clinical studies on IRIS at the NIH Clinical Center with Drs. Irini Sereti and Mario Roeder involving HIV+ patients developing IRIS associated with multiple co-factors in addition to mycobacterial infection. This research showed that in IRIS patients reconstituting CD4+T effector cells display hyperactivation (Antonelli et al. 2010) and that when analyzed at the time of clinical IRIS episode these exuberant T cell responses appear to be driven by specific Ag exposure rather than to a global non-specific alteration in T cell responsiveness (Mahnke et al., 2012). In addition, in collaboration with Dr. Soumya Swaminathan from the National Institute for Research in Tuberculosis (Chennai, India) with support from the NIAID ICER situated in that center, we have attempted to validate these observations in a separate study focusing on the identification of reliable biomarkers for TB-associated IRIS. We found that patients co-infected with HIV and TB who developed IRIS during the follow up presented higher numbers of circulating PD-1+ Tregs and CD4+ T effector memory cells (CD45RO+CD27-) in the blood than those individuals that did not developed IRIS. In addition, at the study baseline (prior to the initiation of ART and anti-TB treatment), we found that in patients that go on to develop IRIS the percentage of CD4+ effector memory T cells expressing PD-1 was positively correlated with increased bacillary burden in sputum. We also found that at the study baseline, increased plasma levels of IL-6, soluble PD-1, soluble Tissue Factor, soluble CD14, soluble CD163, as well as monocyte counts in the blood were all strongly associated with the occurrence of TB-IRIS during the follow up suggesting a major involvement of innate immune factors. M. tuberculosis triggers free radical production in infected macrophages which leads to oxidative burst and lipid peroxidation. Excessive oxidative stress is seen in patients with pulmonary TB and is associated with tissue necrosis and cavitary disease. The host produces a variety of anti-oxidant mediators to circumvent the detrimental effects of oxidative stress. In the lungs, the transcription factor NRF-2 is highly expressed and is thought to be a major regulator of the anti-oxidant responses. One of the most important NRF-2 products is the enzyme heme oxygenase-1 (HO-1) that induces cytoprotection by metabolizing free heme, a highly toxic molecule released as a consequence of tissue damage. To test whether HO-1 expression is associated with human TB, Dr. Bruno Andrade measured the enzyme in plasma samples from a TB cohort study initiated by Dr. Subash Babu (International Center for Excellence in Research, ICER/NIAID, India). He and his collaborators found that concentrations of HO-1 are increased during pulmonary TB and are even higher in those patients with bilateral lung lesions, positive M. tuberculosis in sputum smears or with elevated levels of C-reactive protein, a surrogate of systemic inflammation. Moreover, systemic levels of HO-1 were positively correlated with the bacterial burdens in sputum cultures. Statistical analysis demonstrated that HO-1 levels can be used to discriminate active from latent TB. Importantly, the elevations in plasma HO-1 levels seen in TB patients were completely reversed after successful anti-TB therapy. Interestingly, patients who failed treatment maintained high levels of HO-1, suggesting that the enzyme can be used as surrogate of both active disease and as a biomarker for successful chemotherapy.
Showing the most recent 10 out of 26 publications
