Non-Classical Monocytes and Monocyte Chemoattractant Protein-1 (MCP-1) Correlate with Coronary Artery Calcium Progression in Chronically HIV-1 Infected Adults on Stable Antiretroviral Therapy

Background Persistent inflammation and immune activation has been hypothesized to contribute to increased prevalence of subclinical atherosclerosis and cardiovascular disease (CVD) risk in patients with chronic HIV infection. In this study, we examined the correlation of peripheral monocyte subsets and soluble biomarkers of inflammation to coronary artery calcium (CAC) progression, as measured by cardiac computed tomography scan. Methods We conducted a longitudinal analysis utilizing baseline data of 78 participants with HIV infection on stable antiretroviral therapy (ART) in the Hawaii Aging with HIV-Cardiovascular study who had available baseline monocyte subset analysis as well as CAC measurement at baseline and at 2-year follow up. Monocyte phenotypes were assessed from cryopreserved blood by flow cytometry and plasma was assayed for soluble biomarkers using antibody-coated beads in a high sensitivity Milliplex Luminex platform. Change in CAC over 2 years was analyzed as the primary outcome variable. Results Of all monocyte subsets and biomarkers tested, higher non-classical monocyte percentage (ρ = 0.259, p = 0.022), interleukin (IL)-6 (ρ = 0.311, p = 0.012), and monocyte chemoattractant protein (MCP)-1 (ρ = 0.524, p = <0.001) were significantly correlated to higher 2-year CAC progression in unadjusted Spearman’s correlation. Non-classical monocyte percentage (ρ = 0.247, p = 0.039), and MCP-1 (ρ = 0.487, p = <0.001), remained significantly correlated to 2-year CAC progression, while IL-6 was not (ρ = 0.209, p = 0.120) after adjustment for age, hypertension, diabetes mellitus, total/HDL cholesterol ratio, smoking history, and BMI. Conclusion The percentage of non-classical monocytes and plasma MCP-1 levels were independently associated with CAC progression and may be related to the progression of atherosclerosis and increased CVD risk associated with chronic HIV infection on stable ART.


Introduction
Patients with human immunodeficiency virus (HIV) infection, even those with well-suppressed HIV infection on antiretroviral therapy (ART), are at increased risk of cardiovascular disease (CVD) events [1,2]. Paralleling clinical observation, imaging studies have demonstrated increased prevalence of subclinical atherosclerosis among HIV-infected patients [3,4]. Inflammation has been increasingly recognized as a key pathologic process in the development and progression of atherosclerosis [5,6]. As antiretroviral-treated HIV infection remains associated with persistent immune activation and inflammation, these processes are hypothesized to promote atherosclerosis and contribute to increased atherosclerotic cardiovascular disease (ASCVD) risk in HIV-infected patients on ART [7,8]. However, the precise immunologic mechanisms that promote atherosclerosis in these patients remains uncertain.
In viremic HIV-infected patients, the expansion of both intermediate and non-classical monocytes has been reported [13]. However, only the expansion of non-classical monocyte persisted during 1 year of treatment with ART [13]. A few studies have evaluated the association between monocyte subsets and atherosclerosis in HIV-infected patients. Among these studies, intermediate monocytes [14] and CD16+ monocytes expressing CX3CR1 [15] have been associated with subclinical atherosclerosis. In addition, our group has observed that a fourth monocyte subset, termed the "transitional" monocytes, characterized by low levels of CD14 and negative CD16 expression (CD14 dim CD16 -) was associated with carotid artery intima-media thickness (CIMT) [16,17].
Soluble biomarkers are related to several integral processes of atherosclerosis, including endothelial activation, immune cells recruitment, as well as production of other cytokines and acute phase proteins [5,6]. In patients with HIV infection, CRP and IL-6 has been independently associated with CVD events in some [18,19] but not all studies [20]. Similarly, independent association between monocyte chemoattractant protein (MCP)-1 and subclinical atherosclerosis in HIV-infected patients has been reported inconsistently [21][22][23][24][25]. Thus, the relationship between these biomarkers, traditional CVD risk factors, and atherosclerosis remains uncertain in HIV-infected patients.
In this study, we evaluated the association of monocyte subsets and plasma soluble biomarkers with the progression of subclinical atherosclerosis as measured by coronary artery calcium (CAC) in participants with HIV infection on stable ART. CAC has been demonstrated to correlate with coronary atherosclerotic plaque burden [26] and its progression has been associated with coronary heart disease (CHD) events [27,28].

Study Participants
This is a longitudinal analysis utilizing the baseline data and 2-year follow up data of participants from the Hawaii Aging with HIV-Cardiovascular (HAHC-CVD) study. In this analysis, we included only participants who had available baseline monocyte subsets analysis and coronary artery calcium (CAC) measurement at baseline and at 2-year follow up. HAHC-CVD is a 5-year longitudinal cohort study of the role of oxidative stress and inflammation in HIV cardiovascular risk. The details of the cohort study design and enrollment have been published previously [29]. Briefly, the study enrolled adults, age 40 years old with documented HIV infection who were on stable ART for 3 months, from 2009 to 2012. The study was approved by the Committee on Human Subjects at our institution and was performed in accordance with the Declaration of Helsinki, all International Conference on Harmonization Good Clinical Practice guidelines, and applicable local regulatory requirements and laws. Written informed consents were obtained from all participants.

Clinical assessment
General medical history with special emphasis on CVD and HIV infection was obtained. Clinical parameters including blood pressure (BP), height, weight, body mass index (BMI), and waist to hip ratio were measured. Smoking was defined as a dichotomous variable of ever smoked or never smoked. Blood tests, including CD4+ T-cell count, HIV RNA, fasting (nothing by mouth for 12 hours) total, high-density lipoprotein (HDL), directly measured low-density lipoprotein (LDL) cholesterol, triglycerides, and glucose were performed.
In this study, hypertension was defined as systolic BP 140 mmHg, or diastolic BP 90 mmHg on entry visit, self-reported history of hypertension, or use of anti-hypertensive medications. Diabetes mellitus was defined by a fasting blood sugar (FBS) 126 mg/dL, 2-hour oral glucose tolerance test (OGTT) 200 mg/dL, or self-reported history of diabetes mellitus. Tenyear coronary heart disease (CHD) risk was calculated by Framingham risk score (FRS) based on a model comprised of age, gender, total cholesterol, HDL cholesterol, systolic blood pressure, treatment of hypertension and any cigarette smoking in the past month using the National Cholesterol Education Program website (http://hp2010.nhlbihin.net/atpiii/calculator. asp) [30]. Participants with diabetes (as a CVD equivalent) or clinical CVD (history of myocardial infarction, angina, coronary disease-related cardiac surgery, or ischemic stroke) were automatically classified as having 10-year CHD risk by FRS of 20%. Clinical CVD events were adjudicated by 2 physician-researchers (C.M.S and D.C.C.). Undetectable plasma HIV RNA was defined as HIV RNA 50 copies/mL or less.

Coronary Artery Calcium (CAC) measurements
Computed tomography (CT) examinations for CAC were performed locally in Honolulu, Hawaii at Accuimaging, a free standing imaging center, using a dual source CT (DSCT) scanner (Siemens 64-slice Somatom) following previously published methods [31]. Technical quality assurance and centralized imaging analyses were provided by CT reading center at the Los Angeles Biomedical Research Institute (M. Budoff). A radiologist or cardiologist, who were blinded to clinical data, quantified CAC using an interactive scoring system to calculate Agatston score, with any Agatston score >0 defining the presence of CAC [32]. All participants were scanned twice at entry and after 2 years, with total Agatston score used for all analyses.
The total monocyte count was calculated using white blood cell count (WBC) and percent monocyte values on the CBC conducted as part of entry evaluations on the same blood specimen as that utilized for flow cytometry in line with our previous reports [17].

Statistical analyses
Demographic, cardiovascular, and HIV-related characteristics as well as monocyte subsets, soluble biomarkers, and CAC measurements were described using the median, first quartile (Q1), and third quartile (Q3) for continuous variables and frequency and percent for categorical variables.
Given that 2-year change in CAC is not normally distributed and a substantial percentage of patients had zero 2-year change in CAC, Spearman's correlation was conducted to evaluate baseline characteristics, including demographic, cardiovascular, and HIV-related characteristics, as well as monocytes subsets and soluble biomarkers for their correlation to 2-year change in CAC.
To further explore the monocyte subsets and soluble biomarkers that were found to be significantly correlated to 2-year change in CAC, we conducted partial Spearman's correlation with sequential incremental adjustment for 1) Age, 2) Hypertension, 3) Diabetes mellitus, 4) Total/HDL cholesterol ratio, 5) Smoking history, and 6) Body mass index (BMI). Adjusted Spearman's correlations were conducted using SAS (SAS Institute Inc., Version 9.4, Cary, NC). All other statistical analyses were conducted using SPSS (IBM, Version 21, Armonk, NY). A two-sided probability of p-value less than 0.05 was considered statistically significant.

Participant characteristics
Of all participants enrolled into the Hawaii Aging with HIV-Cardiovascular cohort, a total of 78 had available baseline (entry) monocyte subsets analysis as well as coronary artery calcium (CAC) measurement at baseline (entry) and at 2-year follow up and were included in this analysis. Soluble biomarkers at baseline were available in 64 of 78 participants included in this analysis. The demographic, cardiovascular risks, and HIV-related characteristics of the participants are presented in Table 1. Participants were predominantly white (59%) and male (88.5%), with median age of 50.5 years. Almost 8% of patients had a history of clinical CVD events. The median 10-year CHD risk estimated by FRS was 5%. Most participants (84.6%) had undetectable HIV RNA and the median CD4+ T-cell count of all participants was 507.5 cells/mm 3 . Nucleoside Reverse Transcriptase Inhibitors (NRTI), Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTI), Protease Inhibitors (PI), and Integrase Inhibitor (II) were used at entry by 98.7%, 47.4%, 50%, and 11.5% of participants respectively. The baseline monocyte subset percentage and soluble biomarkers are reported as median (Q1, Q3) in Table 1.
The correlations of non-classical monocyte percentage, IL-6, and MCP-1 to 2-year CAC Agatston score progression were further evaluated in adjusted Spearman's correlation. After adjusting for age, hypertension, diabetes mellitus, total/HDL cholesterol ratio, smoking history, and BMI, non-classical monocyte percentage (ρ = 0.247, p = 0.039), and MCP-1 (ρ = 0.487, p = <0.001), remained significantly correlated to 2-year CAC Agatston score progression, while IL-6 was not (ρ = 0.209, p = 0.120) ( Table 3). The correlations of non-classical monocyte percentage and MCP-1 to 2-year CAC Agatston score progression were also significant in a model adjusting for 10-year CHD risk estimated by FRS and BMI instead of each individual traditional CVD risk factors (data not shown).

Discussion
Persistent inflammation and immune activation have been hypothesized to promote atherosclerosis in patients with HIV infection [7,8]. In this cohort of HIV-infected participants on stable ART, we found that higher baseline percentage of non-classical monocytes and level of MCP-1, a chemokine that regulates monocytes, were associated with higher 2-year CAC   progression independent of traditional CVD risk factors. These associations suggest a potential relation between monocyte-related immune activation and increased CVD risk associated with HIV infection in these patients. At the site of atherosclerosis, monocytes adhere to the endothelial surface and migrate into the subendothelial space in response to chemokines, such as MCP-1 [8,9,33]. In the subendothelial space, they scavenge oxidized lipid particles, transform into foam cells, and secrete inflammatory cytokines and proteases, such as matrix metalloprotease (MMP), contributing to the progression of atherosclerosis [8,9]. Non-classical monocytes express high level of adhesion molecule expression and exhibit a characteristic "patrolling movement" on the endothelial surface [10,34]. They have been proposed to continually survey the endothelium for signs of damage and inflammation [10,34] and have been demonstrated to extravasate rapidly in response to inflammation in a murine model [35]. It is easy to speculate that these characteristics of non-classical monocytes may facilitate their recruitment into atherosclerotic lesions. In the general population, the expansion of non-classical monocytes has been reported in patients with stable coronary artery disease [36] and non-classical monocyte frequency has been associated with CIMT in renal transplant recipients [37].
The expansion of the non-classical monocyte subset has been reported in HIV-infected patients, even after receiving treatment with ART [13]. Given the aforementioned characteristics of non-classical monocytes and their association with CAC progression observed in our study, it is possible that non-classical monocytes may be actively involved in the pathogenesis of atherosclerosis in HIV-infected patients. However, it remains inconclusive if they contribute to progression of atherosclerosis and increased CVD risk in these patients as there is also some evidence to suggest that they may functionally be atheroprotective, representing a mechanistic defense against increased ASCVD risk [34].
We did not find any associations between other monocyte subsets and CAC progression in this study. Intermediate monocyte subset have been reported to be independently associated with CAC progression in the SUN (Study to Understand the Natural History of HIV/AIDS in the Era of Effective Therapy) study cohort of HIV-infected patients [14]. We have also previously reported an independent association between transitional monocyte subset and CIMT in our entry specimens [16]. The differences observed may be attributable to the limited sample size of our study, difference between subclinical atherosclerosis assessment by CAC and CIMT, and the difference in the characteristics of participants in the two studies.
MCP-1 regulates migration of monocytes via its effect on chemokine receptor CCR2 [33]. It can be produced by several cell types including endothelial cells and monocytes both constitutionally and in response to inflammatory cytokines [33]. Evidence from murine models supports the role of MCP-1 in attracting monocytes to atherosclerotic lesions [38,39]. The association between MCP-1 and CAC progression observed in this study is consistent with a previous report by our group demonstrating a cross-sectional association between MCP-1 and CAC in our entry specimens [23]. Together, the results of both studies suggest that excess CVD risk in HIV-infected patients on ART may be related to increased monocyte recruitment into atherosclerotic plaques. However, the role of MCP-1 in the recruitment of human nonclassical monocytes is unclear at present. In a murine study, the lack of CCR2 receptor did not significantly affect the distribution of Ly6Cmonocytes (equivalent to human non-classical monocytes) [34].
This study is limited by the small sample size, and the inability of the study to demonstrate cause and effect due to the observational nature of the study. We also did not assess the potential impact on ASCVD risk from other infections such as cytomegalovirus (CMV) co-infection, as well as from translocation of gut microbial products. Host genetics were also not examined in this study. The duration and prior history of specific antiretroviral medications were not extensively analyzed due to the limited number of subjects on each medication. Although PIs as a class of antiretrovirals were associated with increased CAC, no specific PI was significantly correlated with CAC. This was also noted for other antiretroviral classes and specific NRTIs, NNRTIs, and IIs. The lack of a HIV negative comparison group also limited the interpretation of the soluble biomarkers. The ranges to the biomarkers were broad. We compared our values to a set of eighty-six HIV negative controls evaluated in different study populations evaluated by our group and observed that HIV-infected subjects have higher measures of CRP, SAA, and SAP. Median measures were 6.0% higher in SAP, 57.0% higher in CRP, and 69.3% higher in SAA compared to the HIV negative group (data unpublished) [40]. This comparison is consistent with a presentation of increased inflammatory tendencies attributable to HIV infection despite being on stable ART. Despite these limitations, the strength of this study is the careful clinical characterization and CAC measurement performed in the cohort in association with detailed monocyte subsets and biomarker assays. The identification of the non-classical monocyte subset and MCP-1 as being related to CAC progression provides potential new therapeutic targets for intervention in chronic HIV infection.