Altered atherosclerotic-related gene expression signature in circulating mononuclear leukocytes from hypercholesterolemic patients with low HDL cholesterol levels

Abstract

HDL exerts atheroprotective effects through modulation of theimmune system physiology [1,2]. An abnormal function of immunecells including mononuclear leukocytes has been associated with thepathogenesis of cardiovascular disease (CVD) and their complications[3,4]. Nevertheless, the impact of blood HDL-C levels on leukocytefunctionality in populations at high cardiovascular risk has not beenpreviouslyexplored.Anadditionalapproachseemstobenecessary.Thepurpose of the current investigation was to compare Peripheral BloodMononuclearCells(PBMCs)atherosclerotic-relatedgeneexpressioninhypercholesterolemic patients with optimal and low HDL-C levels.Atotalof75subjectsover30 yearsoldwererecruitedatthetimeofattending the outpatient clinic for vascular risk assessment at theUniversity Clinic of Navarra, Spain (Table 1S, Supplementary content).Attendingtocurrentcardiovascularguidelines,hypercholesterolemiawasdefined as: total cholesterol ≥ 200 mg/dL and/or LDL-C ≥ 130 mg/dL.LowHDL-Clevelswerede finedasHDL-C ≤ 40 mg/dLformalesandHDL-C≤ 50 mg/dLforfemales.Exclusioncriteriawerethepresenceofseverelyimpaired renal function, chronic in flammatory conditions, and adminis-tration of anti-inflammatory, antithrombotic or hormonal therapy in theprevious 2 weeks. Patients with signifi cant acute infection, according toclinical criteria applied by the attending physician, were also excluded.The institutional ethics committee approved this study, and writteninformed consent was obtained from all patients.PBMCs were isolated from the blood using the Ficoll separationmethod. Total RNA was extracted using the RNeasy mini kit (Qiagen,Hilden,Germany)asspecifiedbythemanufacturer'sinstructions.Geneexpression analyses were performed at mRNA level by TaqMan Low-Density Array(TLDA). Pre-designed TaqManprobeand primer sets fortargetgeneswerechosenfromanon-linecatalog(AppliedBiosystems,Foster City, CA, USA). Genes were chosen based on literature reviewand previous data from our group (Table 2S, Supplementary content).Multiplelogisticregressionanalyseswereperformedtoexploretheassociations between HDL-C levels and PBMC gene expression inhypercholesterolemic patients. Previous investigations have demon-stratedthatTG-andcholesterol-richlipoproteinsotherthanHDLhavearegulatoryeffectonimmunecellgeneexpression[5].Todeterminetheinfluenceof theselipidparametersonthe associationsbetweenHDL-Clevels and PBMC gene expression, we made an analysis in duplicateadjusting by non-HDL-C or TG levels. The logistic regression modelsshowed an inverse and close association between HDL-C levels andPBMC CD36, CD40 and IL4 gene expression in models adjusted byage,sexandnon-HDL-ClevelsandPBMCCD36andCD40geneexpressioninmodels adjusted by age, sex and TG levels (Table 3S, Supplementarycontent).Sincethemodeladjustedbyageandsexshowedasignificantassociation between IL4 expression and low HDL-C level in hyperch-olesterolemia, these data suggest that this relation depends, at least inpart,onTGlevels.PreviousstudiessuggestedalinkbetweenstatinsandHDL-C levels [6]. We thus included this covariable in the logisticregression models. For all non-HDL-C and TG models, the associationsbetween gene expression and HDL-C levels remained unchangedincluding statin treatment (Table 3S, Supplementary content).Importantly, many investigations have reported an altered geneexpression of circulating leukocytes in relationwith conditions linkedto CVD [7,8]. Therefore, we explored whether the association betweenlow HDL-C levels and PBMC gene expression in hypercholesterolemicpatients was influenced by subclinical femoral atherosclerosis,hypertension or type 2 diabetes. Our results show that the relationbetween HDL-C and CD36/CD40 mRNA levels remains statisticallysignificant even after adjusting by age, sex, non-HDL-C or TG levels,statin intake and prevalence of subclinical femoral atherosclerosis,hypertension or type 2 diabetes (Tables 1 and 2). In contrast, theassociation with IL4 expression, observed in models adjusted by non-HDL-C levels and hypertension or type 2 diabetes, was lost in a modeladjustedbysubclinical femoralatherosclerosis(Tables1 and2).Theseresultssupportpreviousunpublisheddatafromourgroupshowinganinverse association between PBMC IL4 mRNA levels and the presence