HIV and cardiovascular disease: the role of inflammation

Abstract

Purpose of Review:

HIV and antiretroviral therapy (ART) use are linked to an increased incidence of atherosclerotic cardiovascular disease (ASCVD). Immune activation persists in ART-treated people with HIV (PWH), and markers of inflammation (i.e. IL-6, C-reactive protein) predict mortality in this population. This review discusses underlying mechanisms that likely contribute to inflammation and the development of ASCVD in PWH.

Recent Findings:

Persistent inflammation contributes to accelerated ASCVD in HIV and several new insights into the underlying immunologic mechanisms of chronic inflammation in PWH have been made (e.g. clonal hematopoiesis, trained immunity, lipidomics). We will also highlight potential pro-inflammatory mechanisms that may differ in vulnerable populations including women, minorities and children.

Summary:

Mechanistic studies into the drivers of chronic inflammation in PWH are ongoing and may aid in tailoring effective therapeutic strategies that can reduce ASCVD risk in this population. Focus should also include factors that lead to persistent disparities in HIV care and co-morbidities, including sex as a biological factor and social determinants of health. It remains unclear if ASCVD progression in HIV is driven by unique mediators (HIV itself, ART, immunodeficiency), or if it is an accelerated version of disease progression seen in the general population.

Introduction

Human immunodeficiency virus (HIV) infection is associated with increased atherosclerotic cardiovascular disease (ASCVD) risk, potentially due to chronic immune activation and inflammation(1, 2). Plasma markers of inflammation, including interleukin (IL)-6, high-sensitivity C-reactive protein (hs-CRP), D-dimer, tumor necrosis factor receptor (TNFR) 1 and 2, and soluble CD14 (sCD14) are independent predictors of mortality, including CVD-related deaths, in people with HIV (PWH)(3). Recent findings from the mechanistic substudy of the Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE), a large ongoing primary prevention trial in PWH, demonstrate that several biomarkers of innate immune activation and inflammation were significantly higher in participants with, compared to levels in participants without, coronary artery plaques as measured by Computed Tomography Angiography (CTA)(4). These findings highlight that even in virally suppressed PWH with low CVD risk scores, there is a high plaque burden that is associated with increased immune activation and arterial inflammation. Multiple mechanisms likely contribute to persistent immune activation in antiretroviral therapy (ART) treated PWH, including: HIV persistence, co-pathogen infection, microbial translocation, and pro-inflammatory lipids(3). Many of these drivers activate innate immune pathways and the production of cytokines (TNF, IL-6, IL-1β) that contribute to vascular inflammation and progression of atherosclerotic cardiovascular disease (ASCVD). In this article, we will focus on potential mediators of inflammation and the underlying biological mechanisms that may contribute to ASCVD progression in PWH.

Mechanisms contributing to inflammation in PWH

Clonal Hematopoiesis of indeterminate potential (CHIP) and CVD

Migration of immune cells into the blood vessel wall and the pro-inflammatory mediators these cells produce contribute to ASCVD progression(5, 6). Inflammation may modify hematopoietic stem and progenitor cells (HSPCs) in the bone marrow, resulting in the production of immune cells with altered functional and proliferative capacity; these cells may contribute to inflammation and ASCVD(7, 8). Recent studies have explored relationships among clonal hematopoiesis of indeterminate potential (CHIP), inflammation, and CVD risk in PWH. The prevalence of somatic mutations associated with CHIP are enriched among PWH and are related to ART duration, levels of IL-6 and hs-CRP in the Age-Related Clonal hematopoiesis in an HIV Evaluation cohort (ARCHIVE)(9). The increased prevalence of CHIP mutations in PWH has been confirmed in other studies; however, associations between CHIP and inflammation are inconsistent. While levels of inflammation and monocyte activation in PWH (e.g. IL-6, CRP, sCD14, sCD163) were not related to CHIP in a recent publication by van der Heijden et al., levels of von Willebrand factor and D-dimer were increased in CHIP carriers among PWH(10). The HIV reservoir, low nadir CD4 counts and an increased CD4/CD8 T cell ratio were also related to CHIP prevalence in this study(10). In the Swiss HIV Cohort study and the Atherosclerosis Risk in Communities study (ARIC), HIV was associated with a two-fold increase in CHIP prevalence, and there was a trend towards an association between CHIP and coronary artery disease in PWH(11). While age is a known contributor to CHIP mutations in the general population, multiple factors, including immunodeficiency, inflammation, coagulation, and ART, may contribute to CHIP prevalence in PWH, but further exploration into the underlying mechanisms related to CHIP, and whether systemic and vascular inflammation are cause or consequence of CHIP in PWH is needed.

Microbial Translocation

The impaired intestinal mucosal barrier and the resultant translocation of microbial products into the circulation likely contribute to HIV-associated immune activation(12–14). Recognition of microbial products by innate immune receptors results in cellular activation and production of inflammatory cytokines. Markers of microbial translocation and gut barrier dysfunction have been associated with biomarkers of inflammation and mortality in PWH(12, 15–17). Previous attempts to reduce microbial translocation, however, have yielded mixed results: sevelamer, a phosphate-lowering drug that decreases circulating LPS, did not reduce microbial translocation or inflammation in PWH, however, it did decrease levels of the procoagulant molecule tissue factor (18). Additionally, prebiotics, symbiotics and probiotics, have been studied in HIV and the administration appears to be safe and able to modify microbiota composition, but with limited to no effect on systemic inflammatory markers, pathology or gut permeability (19–26). None of these studies measured changes in CVD specific markers. Recent findings from a proof of concept trial found that teduglutide, a glucagon-like peptide-2 analog, decreased arterial inflammation after 6 months of treatment in PWH. The reduction in arterial inflammation was associated with a reduction in activated monocytes and CD8 T cells(27). These findings suggest that improving intestinal integrity may reduce inflammation and CVD risk in PWH.

The Lipidome

Alterations in lipid profiles may also contribute to chronic inflammation and exacerbate ASCVD in PWH. Technological advances have enabled investigators to examine the concentration and fatty acid composition of multiple lipid classes and individual lipid species (“the lipidome”). Lipid profiles have been associated with CVD risk and atherosclerosis in people with(28, 29) and without(30) HIV. Treatment with ART alters lipid levels, and various classes of ART drugs (e.g. protease inhibitors versus integrase inhibitors) may have differential effects on the lipidome(31). Several lipid classes and species that are associated with CVD in people without HIV(30) are increased in PWH and are related to levels of inflammatory biomarkers (e.g. sCD14, IL-6, D-dimer) that are predictive of CVD risk in this population(32, 33).

Lipids may activate inflammatory cascades and drive CVD progression(34, 35). Plasma concentrations of saturated fatty acids (SaFA) have been associated with ASCVD(36, 37), including non-calcified coronary plaque burden(38), in HIV- populations and these lipids likely contribute to inflammation. In vitro studies demonstrate that SaFAs can activate the inflammasome within myeloid cells(35), and induce NFκB signaling in macrophages(39). Exposure of monocytes from HIV- individuals to the SAFAs stearic (18:0) or palmitic acid (16:0), molecules that are increased in PWH, induced activation marker expression on monocyte subsets and the production of inflammatory cytokines (IL-6 and TNFα) in PBMCs(40). Monocytes from PWH also have altered lipid handling capabilities(41, 42), priming them for foam cell formation and contributing to atherosclerotic plaque progression.

While several lipids may contribute to inflammation and atherosclerosis, ceramides (CER) may be of particular interest in PWH. Ceramides can activate the inflammasome and increase IL-1β production and levels of CERs can be increased by inflammation (43). Elevated levels of CER species (CER16:0, CER22:0, CER24:0, and CER 24:1) are associated with ART use, carotid atherosclerosis, and levels of sCD14 in PWH(28). Increased CER levels were also directly related to pro-coagulant, pro-atherosclerotic gene profiles in monocyte derived macrophages (MDMs) in PWH(42). Recently, in a substudy of SATURN HIV(44–46), we reported that treatment with rosuvastatin decreased levels of CERs associated with atherosclerosis in PWH and improved pro-inflammatory gene expression profiles in MDMs(47). Whether or not statin induced improvements in lipid and inflammatory profiles are related to a reduction in CVD events in PWH is being explored in the REPRIEVE trial (48).

Trained Immunity

Acute exposure of immune cells to microbial products and pro-inflammatory lipids contributes to low grade inflammation, persistent exposure to these molecules exacerbates the inflammatory response. Trained Immunity describes the mechanisms whereby cells of the innate immune system (i.e. monocytes, macrophages) respond more robustly to re-exposure of pathogen associated molecular patterns (PAMPs) or other “danger signals” (i.e. lipopolysaccharide, flagellin, oxidized LDL)(49–53). Innate immune cells undergo transcriptional, epigenetic, and metabolic changes in response to Toll-like Receptor (TLR) ligation, resulting in innate immune memory and enhanced functional capabilities on repeated exposure(50, 53). Trained immunity of monocytes and macrophages has been proposed as a mediator of vascular inflammation and atherosclerosis in the ageing general population(51). Several studies have demonstrated increased responsiveness of immune cells from PWH to TLR ligands(54, 55). Recent work shows that monocytes, but not Tcells, from PWH have increased cytokine production in response to exposure to a number of TLR ligands, including bacterial and viral products. This enhanced monocyte response was related to priming of the IL-1β pathway and increased levels hsCRP, sCD14, and β-glucan, a known inducer of trained immunity(56). Chronic exposure of immune cells to TLR ligands, including molecules derived from HIV-1 itself, proinflammatory lipids, or bacterial products from the damaged GI-tract, may all exacerbate innate immune responses in PWH. How the myriad drivers of trained immunity contribute to inflammatory profiles and CVD risk in PWH requires further investigation.

ART and Mitochondrial Function

Treatment with ART has improved the lifespans of PWH, but ART may also decrease mitochondrial function and contribute directly to inflammation and immune cell activation. Mitochondrial dysfunction increases production of reactive oxygen species (ROS) and activation of redox sensitive transcription factors (including NFκB)(57–59). Products derived from damaged mitochondria can activate the inflammasome and increase IL-1β production(60, 61). Singh et al reported that combination ART can induce a senescent phenotype (e.g., telomere shortening, increased ROS production and inflammatory cascade activation, and decreased efferocytosis) in monocytes and macrophages(62). Recent work from Korencak et al. demonstrated that exposure of peripheral blood mononuclear cells (PBMCs) to dolutegravir (DLG) or elvitegravir (EVG) resulted in decreased proliferation and polyfunctionality of CD4+ T cells and promoted a TNF-α focused stress response in these cells(63). We have also shown that exposure to tenofovir disoproxil fumerate (TDF) or emtricitibine (FTC) resulted in decreased mitochondrial function in PBMCs and increased ROS production from T cells and monocyte subsets and induced changes in the functional (i.e. lipid uptake, efferocytosis) and transcriptomic profiles of monocyte derived macrophages (MDMs)(64). Exposure to TDF or FTC resulted in profound changes in MDM gene expression, including changes in genes associated with immune signaling (i.e. STAT2, MyD88, NLRP3) and genes related to histones and chromatin/nucleosome structure. Exposure of MDMs to ART regimens containing TDF, FTC, and efavirenz (EFV) or abacavir (ABC), lamivudine (3TC) and DTG resulted in decreased proliferation of these cells and increased ROS and inflammatory cytokine production(65). ART induced mitochondrial dysfunction may contribute to persistent inflammation in PWH and as newer ART regimens are developed, an understanding of how these drugs influence immune cell function is needed.

Social Determinants of health and vulnerable populations

Sex as a biological factor

Racial and ethnic minority, sexual and gender minority, and low-income people continue to experience persistent disparities in HIV incidence, prevalence, access to care, and other HIV-related health outcomes(66). Specifically, large epidemiologic studies have highlighted increased CVD risk in women living with HIV (WLHIV) compared to men living with HIV (MLHIV)(67–69). Several studies have identified that WLHIV have higher systemic immune activation and decreased gut integrity markers compared to MLHIV(70–74). Similar results were found in a cohort of WLHIV in rural Uganda(75). Recent data presented from REPRIEVE(76), suggest that although WLHIV had lower prevalence of coronary artery plaques, and lower prevalence of plaques of non-calcified portions and/or vulnerable plaque features, they had higher levels of systemic inflammatory markers, including D-Dimer, and a higher percentage of inflammatory and patrolling monocytes. In addition, immune-plaque relationships differed by sex for D-dimer. Sex-related differences in immune activation in HIV may be influenced by hormonal effects on immune cell function, specifically related to sex-specific expression of TLRs(77, 78), sex-specificity of the microbiome(79) and behavioral factors which may not become relevant until puberty since these sex differences have not been found in pre-pubertal children(80). Similarly, the transgender population is disproportionately affected by HIV and is at higher risk of CVD due to HIV-related risks and potentially, due to hormone therapy(81–83). The role of inflammation in the transgender population with HIV and how it may differ from the cisgender population has not been well studied. Social threats, minority stress and HIV may all be drivers of inflammation in this population(84).

Socioeconomic factors

Socio-economic inequality, structural racism and vulnerability may also influence chronic inflammation and downstream co-morbidities. In the Women’s Interagency HIV Study (WIHS), food insecurity was associated with inflammatory markers IL-6 and sTNFR1 in a cross-sectional analysis of WLHIV in the United States, even after adjusting for viral suppression and nutritional factors[26]. Similarly, we found that lack of electricity and clean water were associated with IL-6 and D-dimer in children with HIV, but not in those without HIV in Uganda even after adjusting for demographic and HIV-related factors(85). Careful consideration of biological and social factors are especially important when enrolling comparison cohorts of people with and without HIV in order to appreciate potential differences in inflammatory profiles and CVD risk in these populations.

Youth

ART scale-up has dramatically reduced rates of pediatric HIV mortality and morbidity. Recent findings suggest that children, despite contemporary ART, continue to be susceptible to subclinical cardiovascular and metabolic complications at an early age and that mechanisms may differ from adults. In a study conducted at our US site (Cleveland)(86), McComsey et al showed higher intima-media thickness (IMT) in children with HIV, with a subsequent follow up showing that increased hsCRP was independently associated with thicker IMT(87). On the other hand, other reports from children in Spain and Uganda have not found a relation between IMT and inflammation or immune activation(88, 89). These findings suggest that although youths with HIV also have persistent immune dysfunction (90–92), the factors driving cardiometabolic complications in this population may differ from adults.

The microbiome:

Early childhood is a transitional period that is key for acquisition of human microbiome. HIV/ART exposure in utero, early HIV acquisition and ART initiation likely all play a role in alteration in the gut microbiome that has been described in children with perinatally acquired HIV (PHIV) (93–95). Similarly, to adults with HIV, a decrease in alpha diversity (a measure of bacterial richness and evenness within a community), and a shift from Bacteroides to Prevotella predominance are associated with markers of immune and endothelial cell activation markers (94, 95).

The lipidome:

We have also highlighted difference in the lipidome in this population which may be distinct from adults living with HIV. Biomarkers associated with CVD risk including hsCRP, sCD163, and CD8+ T cell activation were directly correlated with lipid species in PHIV(89). When comparing serum fatty acids concentrations in PHIV to those in HIV+ adults on ART(40), we find that levels of SaFA are nearly two fold higher in PHIV, suggesting that this population likely has a unique lipidomic signature.

Mitochondrial function:

Mitochondrial function and substrate utilization are perturbed in young children with PHIV(96, 97). Despite early ART treatment and viral suppression, cross-sectional analyses suggest that young PHIV children (mean age= 6 years) have mitochondrial impairment as measured by complex IV, citrate synthase and mitochondrial DNA content(97). It is unclear, however, whether decades of ART will allow for recovery or further declines in mitochondrial function.

Conclusion

Cardiovascular disease is the leading cause of death in the United States; and there is ~2-fold greater risk for CVD events for PWH(1, 2). The mechanisms underlying increased CVD prevalence in PWH are not fully understood, but likely involve persistent systemic and vascular inflammation. Even when ART is initiated during acute infection, some, but not all, inflammatory biomarkers remain elevated in PWH compared to levels in people without HIV(98, 99). It will be interesting to assess whether early ART initiation improves CVD risk in PWH in the future. Multiple potential mediators of inflammation have been identified in PWH and recent studies have sought to link patterns of biomarkers to specific drivers of immune activation and to clinical outcomes in this population (100–102). Potentially, certain “drivers of inflammation” or the patterns of inflammatory mediators they elicit will be more strongly associated with specific comorbidities (e.g. CVD, cancer, neurological impairments). Understanding potential mediators of inflammation in PWH and the biological processes they regulate, may inform future clinical interventions. Additionally, understanding the causal effect of HIV on co-morbidities in the context of age, sex, race and social determinants of health deserves special consideration. These factors may not only predict HIV infection, but also CVD risk, and they remain poorly measured and investigated in co-morbidity research. There is a need for research focused on identification of mechanistic pathways contributing to excess risk of CVD among HIV-infected women, sexual minorities, and the transgender population. In addition, considering the consequences of lifelong HIV, ART, and inflammation on children living with perinatally acquired HIV may provide insights into immune health and function in a vulnerable population with implications for treatment of comorbid conditions, even though they lack other traditional risk factors such as hypertension, smoking and substance use that contribute to CVD in adults with HIV. The complex mechanisms that underpin how chronic immune dysregulation and inflammation may contribute to ASCVD risk in PWH remain an important area of clinical and scientific investigation.

• Cardiovascular disease is increased in people with HIV (PWH) and is likely driven by chronic inflammation.

• Trained immunity may exacerbate inflammation in PWH and microbial products and pro-inflammatory lipids likely play a role in initiating pathways related to trained immunity.

• Early treatment with antiretroviral therapy (ART) likely improves inflammatory profiles in PWH, but ART may also decrease mitochondrial function and alter immune cell function.

• Cardiovascular disease is especially important in the aging population of PWH, but consideration of the cardiometabilic complications of lifelong exposure to HIV and ART in perinatally infected children is also warranted.