Patients with CAD were recruited while hospitalized at the University of New Mexico Hospital for an acute coronary syndrome event (unstable angina, non-ST elevation myocardial infarction, or ST-elevation myocardial infarction) to participate in a health outcomes study designed to assess the benefit of an interdisciplinary cardiovascular risk reduction clinic (CRRC), as compared to usual care. The study was approved by the University of New Mexico Health Sciences Center Human Research Review Committee and all study patients provided informed consent.

As part of the study, blood samples were collected from all patients for assessment of a variety of traditional and non-traditional risk factors (lipids, HbA1c, homocysteine, malondialdehyde [MDA], plasminogen activator inhibitor-1) and a portion of the samples were stored for future use. Patients randomized to the CRRC were evaluated by both a cardiologist and a clinical pharmacist, who devised an appropriate treatment plan based upon identified cardiovascular disease risk factors. Patients randomized to usual care received follow-up at the discretion of their primary care providers and/or specialist providers.

The majority of CAD patients were on standard-of-care pharmacotherapy, including strategies to reduce cholesterol, glucose, blood pressure, platelet aggregation, and tobacco cessation where appropriate. The serum samples used in this analysis were collected at the initial outpatient follow-up study visit to the CRRC.

Healthy control subjects were recruited through the University of New Mexico Clinical and Translational Science Center for routine health screens. Subjects were recruited by newspaper or radio advertisements from the community. Exclusion criteria included - 1) history of diabetes mellitus, atherosclerotic cardiovascular disease, chronic kidney disease, or anorexia nervosa; 2) use of statin class of drugs; 3) current smoking or having quit smoking within previous two months; 4) pregnancy and nursing state; 5) presence of lung diseases other than asthma; 6) stroke in prior 3 months; 7) aortic aneurysm; and 8) failure to expectorate adequate-quality sputum in response to induction. In addition, all tests were delayed in the event of an acute infection or surgery within the prior 4 weeks and respiratory tract infections or asthma exacerbations within the prior 8 weeks to minimize the effect of these factors on measurements. For actively menstruating women, the testing was done within 3–14 days following the cessation of menstrual flow (a period of high estrogen and low progesterone) to standardize the effect of sex hormones.

CAD participants were assessed for body composition in a fasting state using a Quantum II Bioelectrical Impedance Analyzer (RJL Systems, Clinton Township, MI) using the method described by Heyward [19]. Results for bioelectric impedance were analyzed with Comprehensive Body Composition Software (Human Kinetics, 1997, Champaign, Ill) and percent body fat mass was calculated using standard reference equations appropriate for race and gender [19].

The endothelial cell biosensor assay (Figure 1) was conducted as previously described [13,16]. Briefly, hCAECs (Lonza, Allendale, NJ) were seeded in a 24-well plate and grown to confluence in complete media (Lonza). The cells were then serum-starved with Basal Media (Lonza) for 24 hours prior to exposure. Confluent hCAEC were incubated with 10% serum obtained from coronary artery disease (CAD) patients (ie., 50 μl serum in 450 μl media, per well) or 10% serum obtained from otherwise healthy individuals. Each plate of cells was treated with approximately equal numbers of subjects from each group in a randomized, blinded fashion. The samples were incubated for 4 hours at 37°C. RNA was extracted from the cells and cDNA was made using a ThermoCycler (Model # PTC-200; MJ Research). The cDNA was used to determine gene expression via quantitative polymerase chain reaction (LightCycler 480 II, Roche, Indianapolis, IN). Specific targets included interleukin-8 (IL-8), intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1), with TATA-box protein used as a housekeeping gene.Figure 1

Endothelial cell biosensor approach. Primary coronary artery endothelial cells are incubated with 10% serum from individual patients for 4 h, followed by isolation of RNA, and then assayed for transcriptional responses.

Serum samples from healthy and CAD patients were analyzed using an electrochemiluminescence detection system (Meso Scale Discovery, Rockville, MD). The serum samples were analyzed for the presence of CRP, soluble ICAM-1, soluble VCAM-1, serum amyloid A (SAA), TNF-α, IL-6, IL-8, and IL-1β using commercially-available kits.

Gene expression data were log₁₀-transformed for normality. Data were analyzed between groups (CAD versus control) via Satterthwaite’s t-test (SAS v9.4). A multivariate analysis confirmed this difference between CAD and control outcomes after adjusting for age, sex, and BMI. Correlational (Spearman) and multivariate analysis was carried out on the CAD patient outcomes relative to demographic and clinical data (SAS).

A total of 48 patients with CAD and 45 healthy control subjects were included in the study. Subjects were self-selected and inclusion criteria allowed for a wide range of CAD etiology and manifestations. Table 1 summarizes the demographic information. The CAD patients in this unmatched study were expectedly older than control subjects and had a disproportionate prevalence of CV risk factors, including diabetes, body mass index (BMI), hypertension, hypercholesterolemia and current tobacco use. Importantly, a sufficiently overlapping range of demographic factors permitted multivariate assessment to eliminate all potential major effect modifiers, in terms of endothelial cell responses to serum, described below. The CAD cohort was comprised of patients with single and multiple vessel disease, and were recorded on admission as being a mix of non-ST segment elevation myocardial infarction (NSTEMI) and ST segment elevation myocardial infarction (STEMI), as well as unstable angina. Table 2 provides all additional personal health data from the CAD patient population that were available.Table 1

Basic demographics for the healthy and coronary artery disease subjects

hCAECs treated with 10% serum from CAD subjects showed substantial increases in IL-8, ICAM-1, and VCAM-1 expression, compared to hCAECs treated with serum from healthy control subjects (Figure 2). Endothelial mRNA expression of IL-8 and VCAM-1 were approximately 80% higher in hCAECs treated with CAD patient serum compared to controls (p < 0.0001). ICAM-1 expression was more stable at approximately 30% above control (P = 0.0009) and with less overall variability. The distribution patterns of gene expression for each of these markers were notably different and correlations among the marker expression from individuals varied considerably. Cross-correlations ranged from moderate (R = 0.534 for VCAM-1 and ICAM-1) to low (R = 0.201 for ICAM-1 and IL-8), implying that these responses might be influenced by different factors – or cumulative factors - in the serum.Figure 2

Endothelial cell expression of IL-8, ICAM-1, and VCAM-1 mRNA in response to incubation with 10% serum from healthy controls or CAD cohort. Significance was assessed with a Student’s t-test. The differences between groups remained significant after controlling for age, sex, and body mass index, and all potential interactions using a multivariate analysis of variance (general linear model).

As many serum components can be acutely altered as a result of major cardiac events, or may trend with age and/or time, we examined the impact of temporality on serum-treated hCAEC mRNA expression in both the cross-sectional samples, as well as with a sub-cohort of longitudinally followed CAD subjects. All samples were obtained from patients with known CAD, either from having previously suffered an acute myocardial infarction or angina attack. The average time post-event was slightly greater than one year (400 days), but 5 subjects enrolled in the study within 2 months of the precipitating event and 6 subjects were greater than 2 years past their event. The earliest enrolled subject had an event 16 days prior to the initial visit. In general, this cross-sectional data set did not reveal strong trends of temporal resolution of the bioactivity of the serum (Figure 3A-C). For serum-treated hCAEC IL-8 expression, there were no significant trends. hCAEC mRNA for ICAM-1 and VCAM-1 demonstrated modest reductions relative to the time post-event (P = 0.033 and 0.055, R² = 0.10 and 0.08, respectively). As serum samples more proximal to the event than 16 days were not available, it is not known if the serum obtained within hours or days after the event might be more inflammatory. However, from months-to-years after an event, the data suggest considerable stability.Figure 3

Temporal trends on serum inflammatory potential. Cross-sectional samples were obtained from 16d to several years after initial cardiac events (A-C). The time post-event had minimal effect on IL8 mRNA expression, but factors in the serum that drove ICAM-1 and VCAM-1 expression may have been higher acutely following the event. Longitudinal serum samples obtained approximately every 3 months from a limited cohort of CAD subjects were assessed for inflammatory potential independently from healthy subjects (D-F).

Medical records allowed for initial event categorization of CAD subjects. We therefore considered an intra-cohort comparison for subjects whose initial presentation was a myocardial infarction (with or without ST elevation) or unstable angina (Figure 4). A slight trend for increased expression of the gene transcripts was noted for patients admitting with an MI, which was only significant for VCAM-1 (P = 0.0266). No significant differences were observed between subjects admitting with ST-elevation MI or non-ST elevation MI (data not shown). These findings are consistent with the hypothesis that greater inflammatory potential in the serum may lead to more severe vascular pathology, although the results are associative.Figure 4

Serum from subjects with diagnosed myocardial infarction (MI) induced greater VCAM-1 induction compared to serum from subjects with unstable angina (P = 0.027 by one-tailed t-test on natural-log-transformed data). Both IL-8 and ICAM-1 trends were non-significant.

Importantly, the significant differences in endothelial expression of IL-8, ICAM-1, and VCAM-1 in response to patient serum remained significant even after adjusting for the potential major effect modifiers of sex, age, and BMI that were clearly different between cohorts. Direct comparisons across sex, age, and BMI are shown in Figure 5. For each mRNA target, it is clear that CAD subjects were elevated compared to healthy controls, but there was not impact of sex on the endothelial cell responses to serum in either healthy or CAD cohorts (Figure 5A-C). Similarly, age and BMI showed no clear trends across the complete range of either variable (Figure 5D-I).Figure 5

Potential effect modifiers sex, Age, and BMI have no clear influence on serum inflammatory potential. Sex related differences are negligible in determining the differences between healthy and CAD serum-induced endothelial cell responses (A-C). Similarly, Age does not appear to drive any trends for serum-induced IL-8, ICAM-1, or VCAM-1 mRNA expression (D-F). While BMI is limited in the control population, the lack of consistent trends in controls and CAD across various BMI ranges reduces confidence that this factor has a clear role in drive the serum inflammatory potential (G-I).

In exploring potential factors that may contribute to serum-induced endothelial cell responses, we found a number of demographic factors in CAD patients that correlated with IL-8, ICAM-1 and VCAM-1 expression (Table 3). IL-8 expression associated with a number of factors, positively trending with serum levels of VLDL1/2, CRP, and PAI-1, while negatively trending with serum glucose and HDL levels. VCAM-1 expression correlated with both smoking and serum homocysteine levels, while ICAM-1 expression correlated with age, serum homocysteine, and MDA. While serum homocysteine trended negatively with inflammatory potential, it should be noted that out of 51 patients, only 3 values for homocysteine were out of normal range, and only two of these were paired with a measure for VCAM-1 mRNA. Fewer demographic and personal health data were available for our control population, and those few factors analyzed (age, BMI, gender) did not correlate with endothelial cell responses to serum.Table 3

Factors in CAD patients that correlate with hCAECs responses to serum

As IL-8 expression had the most robust correlates, we developed a regression model to better understand the factors that predict this outcome (Table 4). This model confirmed the outcomes in the correlational analysis, providing confidence for the roles of serum VLDL, CRP, and glucose levels in driving the IL-8 related inflammatory potential. However, these variables only explained a small portion of the inflammatory potential, indicating that other unmeasured factors are also mediators.Table 4

Predictive variables for IL8 response using a linear regression model

To assess whether inflammatory cytokines might explain differences in the serum inflammatory potential between healthy and CAD subjects, we measured 8 key inflammatory cytokines in serum from all patients using an electrochemiluminescence technique (Figure 6). No statistically significant differences were noted for IL-1β, IL-6, IL-8, TNFα, C-Reactive Protein, soluble VCAM-1, or Serum Amyloid A, either in terms of mean or distribution, between healthy and CAD cohorts. A modest, but non-significant increase in soluble ICAM-1 was noted however (49.1 ± 8.1 controls versus 84.1 ± 14.4 in CAD patients, P = 0.086).Figure 6

Cytokine and acute phase proteins measured by electrochemiluminescence in serum from healthy controls and CAD cohort. Significance was assessed with a Student’s t-test.