Systemic Inflammation in Metabolic Syndrome: Increased Platelet and Leukocyte Activation, and Key Role of CX3CL1/CX3CR1 and CCL2/CCR2 Axes in Arterial Platelet-Proinflammatory Monocyte Adhesion

Background: Metabolic syndrome is associated with low-grade systemic inflammation, which is a key driver of premature atherosclerosis. We characterized immune cell behavior in metabolic syndrome, its consequences, and the potential involvement of the CX3CL1/CX3CR1 and CCL2/CCR2 chemokine axes. Methods: Whole blood from 18 patients with metabolic syndrome and 21 age-matched controls was analyzed by flow cytometry to determine the leukocyte immunophenotypes, activation, platelet-leukocyte aggregates, and CX3CR1 expression. ELISA determined the plasma marker levels. Platelet-leukocyte aggregates adhesion to tumor necrosis factor-α (TNFα)-stimulated arterial endothelium and the role of CX3CL1/CX3CR1 and CCL2/CCR2 axes was investigated with the parallel-plate flow chamber. Results: When compared with the controls, the metabolic syndrome patients presented greater percentages of eosinophils, CD3+ T lymphocytes, Mon2/Mon3 monocytes, platelet-eosinophil and -lymphocyte aggregates, activated platelets, neutrophils, eosinophils, monocytes, and CD8+ T cells, but lower percentages of Mon1 monocytes. Patients had increased circulating interleukin-8 (IL-8) and TNFα levels and decreased IL-4. CX3CR1 up-regulation in platelet-Mon1 monocyte aggregates in metabolic syndrome patients led to increased CX3CR1/CCR2-dependent platelet-Mon1 monocyte adhesion to dysfunctional arterial endothelium. Conclusion: We provide evidence of generalized immune activation in metabolic syndrome. Additionally, CX3CL1/CX3CR1 or CCL2/CCR2 axes are potential candidates for therapeutic intervention in cardiovascular disorders in metabolic syndrome patients, as their blockade impairs the augmented arterial platelet-Mon1 monocyte aggregate adhesiveness, which is a key event in atherogenesis.


Cell Culture
Human umbilical arterial endothelial cells (HUAEC) were isolated by collagenase treatment [1] and maintained in human endothelial cell specific medium (EBM-2, Lonza, Barcelona, Spain), supplemented with endothelial growth media (EGM-2, Lonza) and containing 10% fetal bovine serum (FBS; Biowest, Nuaillé, France). Cells were grown to confluence up to passage 1 to preserve endothelial features. Prior to every experiment, cells were incubated for 24 h in medium containing 2% FBS.

Flow Cytometry
Blood samples for flow cytometry studies were collected in BD Vacutainer® blood collection tubes containing 3.2% sodium citrate, or in BD Vacutainer® PST™ II tubes with lithium/heparin (17 IU/mL) as anticoagulant agents (both from BD Biosciences, San Jose, CA). All samples were run in a FACSVerse™ flow cytometer (BD Biosciences) and all flow cytometry data were analyzed with FlowJo® v10.0.7 software (FlowJo LLC, Ashland, OR).
To determine the grade of leukocyte activation, the expression of CD11b was analyzed on circulating neutrophils, eosinophils and monocyte subsets, or CD69 expression on CD3 + T cells and lymphocyte subsets. Heparinized whole blood samples were incubated in the dark for 30 min with saturated amounts of a phycoerythrin (PE)-conjugated mAb against human integrin CD11b (clone CBRM1/5, IgG1; Biolegend, San Diego, CA) or a PE-conjugated mAb against human CD69 (clone FN50, IgG1; Immunostep). Fractalkine/CX3CL1 receptor (CX3CR1) expression was determined using a PE-conjugated rat mAb against human CX3CR1 (clone 2A9-1, IgG2B; Biolegend). In some experiments, heparinized blood samples were incubated with ethylenediaminetetraacetic acid (EDTA, 10 mM, for 15 min at 37ºC) to promote platelet dissociation as described [2]. This disaggregation was measured using the marker CD41 in circulating leukocyte subsets. To do this, heparinized whole blood or EDTA samples were incubated in the dark for 30 min with saturated amounts of a PE/Cy™7-conjugated mouse mAb against human CD41 (clone HIP8, IgG1; Biolegend) or a CF-Blue™-conjugated mAb against human CD41 (clone HIP8, IgG1; Immunostep).
Red blood cells were lysed using a commercial lysis buffer (BD FACS™ lysing solution 10× concentrate; BD Biosciences).
Results were expressed as pg or ng/mL of mediator in plasma.

Leukocyte-Endothelial Cell interactions Under Flow Conditions
Before starting each assay, whole blood was In all experiments, leukocyte interactions were determined after 7 min at 0.5 dyn/cm 2 .
Cells interacting on the surface of the endothelium were visualized and recorded (×20 objective, ×10 eyepiece) using a phase-contrast microscopy (Axio Observer A1 Carl Zeiss microscope; Carl Zeiss, Thornwood, NY). For each determination, at least 5 fields were recorded for 10 s and then averaged. Finally, recorded images were saved on a computer for further analysis.

Immunofluorescence studies
To visualize adherent CX3CR1 and CCR2 expressing-platelet-leukocyte complexes, platelet-monocyte aggregates, leukocytes or monocytes with endothelial cells we performed an immunofluorescence analysis. Confluent endothelial cells were grown on glass coverslips and stimulated with 20 ng/mL TNFα, for 24 h. Heparinized blood from patients with metabolic syndrome and age-matched controls was incubated without or with EDTA. After the flow chamber assay, cells were fixed with 4% paraformaldehyde and blocked in PBS containing 1% BSA (Sigma-Aldrich, Madrid, Spain). Subsequently, cells were incubated at room temperature for 2 h with an Alexa Fluor 594-conjugated antibody against human CD45, to detect leukocytes     ). In heparinized blood, neutrophil-platelet-aggregates were selected as a CD16 + CD41 + population and eosinophil-platelet aggregates as a CD16 -CD41 + population. Figure S3. Gating strategy for human monocyte detection in whole blood by flow cytometry. Monocytes were selected by CD14 labelling and morphology (medium SSC-A). For the detection of monocyte subpopulations, CD16 and CCR2 markers were used. Monocytes-platelets complexes were selected as CD14 + CD41 + populations in heparinized whole blood, and platelet-free monocytes were gated as CD14 + CD41 − from blood incubated with EDTA. Figure S4. Gating strategy for human T lymphocyte detection in whole blood by flow cytometry. T lymphocytes were selected as a CD3 + population and with a low SSC-A. T helper (Th) lymphocytes were selected as the CD3 + CD4 + population. In heparinized blood, Th lymphocyte-platelet complexes were selected as the CD3 + CD4 + CD41 + population, whereas platelet-free Th lymphocytes were gated as CD3 + CD4 + CD41 − from blood incubated with EDTA. Cytotoxic lymphocytes were selected as CD3 + CD8 + . In heparinized blood, cytotoxic lymphocyte-platelet complexes were selected as the CD3 + CD8 + CD41 + population, whereas platelet free cytotoxic lymphocytes were selected as CD3 + CD8 + CD41 -from blood incubated with EDTA. Figure S5. Flow cytometry detection and morphologic gating of human monocytes for the immunophenotype study of the different monocyte subsets. Classical monocytes (Mon1) were selected as CD14 + CD16 -population from whole blood. In order to study the monocyte immunophenotypic changes, eight gates were designed (S1-8) from CD14 + CD16 -to CD14 low CD16 + . Figure S6. Comparison of CD14, CD16, CCR2 and CX3CR1 expression in the different monocyte subsets of MS patients and healthy volunteers. Immunophenotypic changes of the different monocyte subsets, relative to CD14 (A), CD16 (B), CCR2 (C) and CX3CR1 (D) expression in metabolic syndrome patients and agematched controls. **p <0.01 relative to values in the control group.