Immunopathogenic Mechanisms of Human Immunodeficiency Virus (HIV) Disease

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications & trials

Abstract

HIV infection leads to CD4 lymphopenia and immunosuppression, which can be successfully improved with antiretroviral therapy (ART) in the majority of people. Approximately 20 percent of HIV patients starting ART, especially with lower CD4 counts, may develop an aberrant immune response known as immune reconstitution syndrome (IRIS). IRIS encompasses the paradoxical worsening of the manifestations of underlying opportunistic diseases (paradoxical IRIS) or the abrupt presentation of previously occult opportunistic disease (unmasking IRIS) in patients who recently started ART. Symptoms of IRIS can range from uncomplicated localized reactions to severe and systemic manifestations and, more rarely, autoimmune phenomena. Clinical studies have identified two main risk factors for IRIS: severe CD4 lymphopenia and the presence of opportunistic infections even if clinically silent (M. tuberculosis, M. avium complex, C. neoformans or other fungi, or viral pathogens) prior to ART initiation. The pathogenesis of IRIS in HIV infection remains unclear and there is no animal model that adequately mimics the clinical observations. In order to study the clinical predictors, biomarkers and pathogenesis of IRIS we conducted a large prospective international multi-site study of 506 ART-naive HIV+ patients with low CD4 (<100 cells/L) who were followed after initiation of ART in US, Kenya and Thailand. Participants with lower hemoglobin at baseline were at higher IRIS risk amd IRIS was independently associated with increased risk of death after adjustment for known risk factors (HR, 3.2; P = .031). Lower levels of hemoglobin was found predictive of IRIS, and high C-reactive protein (CRP) >106 g/mL with low BMI was predictive of death. Adding other inflammatory mediators such as cytokines led to an inflammatory score that was highly predictive of mycobacterial (but not viral) IRIS. With respect to pathogenesis, we previously reported that T cells from IRIS patients bore a highly activated phenotype (high PD-1 expression) showing evidence of profound antigenic stimulation with a slower recovery of naive CD4 T cells following IRIS events. In continuation of that work, we showed that the activated CD4 T cells predominantly recognize the underlying opportunistic pathogen and mount polyfunctional cytokine responses while the T cell responses to other pathogens, including HIV itself, remain intact. We later showed that monocytes play a prominent role in TB-IRIS and may serve as important predictors as well as potential therapeutic targets. Finally, comparing HIV persons and HIV-seronegative with the same pathogen (MAC) we found that the CD4 responses against MAC during IRIS are far greater than those of HIV-seronegative patients with MAC with unique features including expression of high levels of EOMES, low Tbet and evidence of cytotoxic potential. In a currently ongoing study, we also studied patients with fluorodeoxyglucose (FDG)-PET imaging and found that higher glucose uptake (total glycolytic activity and total glycolytic volume) and higher standardized max uptake pre-ART were associated with IRIS incidence after therapy initiation. PET measurements correlated with plasma inflammatory cytokines and higher Glut-1 expression (main glucose transporter) was observed on CD4 T cells and monocytes of patients who developed IRIS. These observations highlight a seminal role of metabolism in general, and glycolysis more specifically, in IRIS events which may help identify new therapeutic targets. Finally our most recent work clearly delineated the role of inflammasome activation in monocytes in people with TB-IRIS and the important role of complement activation in monocytes with evidence of inflammasome activation. More recently we showed the spectrum of lack of immune response with overwhelming infection to intense IRIS in a case series of M. genavense in HIV patients. Importantly, we also described that similar IRIS phenomena can occur in people without HIV infection, specifically some marrow recipients with primary immunodeficiencies who had disseminated mycobacterial infections and recovered their pathogen-specific immune responses. With respect to therapeutics, in a small case series we showed that TNF blockade may be a useful means to treat refractory to corticosteroids IRIS and in a phase I pilot study of an oral phosphodiesterase 4 inhibitor (CC-11050) in people with HIV who had suppressed plasma viremia we found the drug was well tolerated and also lowered slightly plasma IL-8 levels. Despite the significant improvement of morbidity and mortality in HIV infection in the ART era, mortality in HIV+ patients is still in excess of what is expected based on age and strongly relates to the degree of immunodeficiency as measured by CD4 T cell counts and to the degree of residual inflammation measured by IL-6, CRP, sCD14 and D-dimer. These markers appear to be linked more to activation of the innate system. We had previously evaluated these critical biomarkers at a cohort of HIV infected persons diagnosed and treated at the earliest possible stages of HIV infection and found that despite the early treatment, inflammatory markers did not completely normalize after two years of ART although D-dimer levels did normalize. In another study we found that tissue factor expression on monocytes, a protein that triggers activation of the extrinsic pathway of the coagulation cascade, is heightened in HIV despite ART and is linked to both coagulation and increased inflammatory cytokines. The pathogenic role of tissue factor was further confirmed in an SIV model. Treatment with Ixolaris (an inhibitor of tissue factor) blocked activation of coagulation in vitro with human and non-human primate cells. Newer strategies for decreasing residual inflammation in chronically treated patients are being pursued including commonly used medications such as statins and aspirin or losartan. These observations are being followed by further evaluation of the coagulopathy in HIV and potential interventions to improve it. More recently, we also evaluated the role of gut microbiome in a HIV+/HIV- carefully matched cohort and found that dysbiosis in HIV is independent of sexual practices and directly associates with prevalence of noncommunicable diseases. In our Idiopathic CD4 lymphopenia (ICL) work, we developed a humanized mouse model and we were able to show that there is heterogeneity in ICL patients with approximately half being able to reconstitute the empty mouse host just as healthy controls. The ones who could not, showed evidence of either depressed proliferative capacity of PBMC or increase death. In contrast hematopoietic cells were able to develop normally into T cells. This mouse model provides us with a tool to classify and further study ICL patients evaluating potential defects of T cell development or peripheral expansion and survival. Following up on the observation of impaired T cell survival, we have found that all ICL patients have evidence of autoantibodies compared to healthy controls. Some of these antibodies appear able to coat human lymphocytes and at times can cause cytotoxicity either complement or antibody-mediated. These observations suggest a potentially important role of autoantibodies in ICL pathogenesis and open the path for new therapeutic strategies. Finally we also identified important genetic defects manifesting with CD4 lymphopenia either due to CD4 gene mutation with absence of CD4 translation and also a novel gc mutation with reversion in T-cells manifesting with HPV diseases that regressed after hematopoietic transplant.