The Fractalkine Receptor CX3CR1 Links Lymphocyte Kinetics in CMV-Seropositive Patients and Acute Myocardial Infarction With Adverse Left Ventricular Remodeling

Abstract

Aims: Latent cytomegalovirus (CMV) infection is associated with adverse cardiovascular outcomes. Virus-specific CX₃CR1⁺ effector memory T-cells may be instrumental in this process due to their pro-inflammatory properties. We investigated the role of CX₃CR1 (fractalkine receptor) in CMV-related lymphocyte kinetics and cardiac remodeling in patients with ST-elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (pPCI).

Methods and results: We retrospectively analysed lymphocyte count, troponin, and survival in 4874 STEMI/pPCI patients, evaluated lymphocyte kinetics during reperfusion in a prospective cohort, and obtained sequential cardiac MRI (cMRI) to assess remodeling. Pre-reperfusion lymphopenia independently predicted mortality at 7.5 years. Prior to reperfusion, CCR7⁺ T-lymphocytes appeared to be depleted. After reperfusion, T-lymphocytes expressing CX₃CR1 were depleted predominantly in CMV-seropositive patients. During ischaemia/reperfusion, a drop in CX₃CR1⁺ T-lymphocytes was significantly linked with microvascular obstruction in CMV+ patients, suggesting increased fractalkine-receptor interaction. At 12 weeks, CMV+ patients displayed adverse LV remodeling.

Conclusion: We show that lymphopenia occurs before and after reperfusion in STEMI by different mechanisms and predicts long-term outcome. In CMV+ patients, increased fractalkine induction and sequestration of CX₃CR1⁺ T-cells may contribute to adverse remodeling, suggesting a pro-inflammatory pathomechanism which presents a novel therapeutic target.

Keywords: CX3CR1; T-lymphocytes; acute myocardial infarction; cardiac MRI; cytomegalovirus; remodeling.

Figure 1

Overview of patients with acute MI used in this study. Three of the four prospective cohorts were recruited between 2008 and 2015, so these patients are also included in the retrospective analysis. The TACTIC trial recruited patients after 2015, so these are not included in the retrospective cohort. cMRIs were unanalysable either because of technical problems with the scanner or poor image quality (e.g. due to atrial fibrillation). Acute cMRIs from Boag et al. and CAPRI were analysed together. We also describe data from 10 healthy control patients, and 21 explanted human hearts, not shown in this diagram.

Figure 2

CMV serostatus is associated with adverse remodeling 12 weeks after reperfusion. (A) In the acute phase (1-8 days post-reperfusion, n=101), CMV serostatus has no effect on infarct size, LV ejection fraction, end-diastolic volume or end-systolic volume. At 12-week follow-up (n=48), CMV seropositive patients displayed significantly more deterioration in end-diastolic volume (+10.7mL vs -6.1mL, p=0.02). P values determined using the unpaired t-test. (B) Relationship between amount of MVO (zero, low or high) and change in T cell subsets between 15-30 minutes post-reperfusion, separately for CMV seropositive and seronegative patients. Box plots display median (central line), 25^th and 75^th centiles (limits of box), and range (error bars). Statistics refer to differences between MVO groups as indicated (Kruskal-Wallis test with Dunn’s multiple comparisons test). Total n=47; CMV positive n=25 [8 zero MVO, 6 low, 11 high], CMV negative n=22 [9 zero MVO, 7 low, 6 high]). * p<0.05, ***p<0.001; ns, not significant.

Figure 3

Lymphocyte dynamics during ischaemia/reperfusion depending on CMV serostatus. (A) Absolute count of lymphocyte subsets by time point. 0 minutes represents pre-reperfusion, other time points are post-reperfusion. CD4⁺, CD8⁺ and CD16⁺56⁺ cells dropped between 0 and 90 minutes post-reperfusion, and CD4⁺ and CD8⁺ cells rebounded at 24 hours. CMV seropositive patients had more effector memory cells and showed sharper drops by 90 minutes. Dynamics were significantly affected by CMV serostatus in CD8⁺ (p=0.05) CD4⁺ T_EM (p=0.002), CD4⁺ T_EMRA (p=0.002) and CD8⁺ T_EMRA (p=0.003). n=52. Horizontal lines display statistical significance across time points for CMV seropositive (red) and seronegative (blue) patients. *=p<0.05; ** = p<0.01; *** = p<0.001; *** = p<0.0001. p values determined using 2-way ANOVA. (B) Mean CX₃CR1 expression on each T-cell subset at 24 hours post-reperfusion.

Figure 4

Lymphopenia prior to reperfusion predicts mortality in STEMI and is associated with admission troponin. (A) Kaplan-Meier survival curves of 4874 consecutive STEMI/pPCI patients discharged alive following pPCI (mean follow-up time of 7.5 years), divided into four quartiles of pre-reperfusion lymphocyte count, with the lowest quartile having worse survival. Survival of each quartile was compared with unadjusted Kaplan-Meier analysis. (B) The same 4874 patients, with the mean admission high-sensitivity cardiac troponin T value (ng/ml) for each lymphocyte quartile, showing that patients with the lowest lymphocytes had the highest troponin values (statistical comparison by unpaired t-tests). (C) Correlation between pre-reperfusion lymphocyte count and left-ventricular ejection fraction between 2-7 days post-reperfusion. Lower lymphocyte count was predictive of worse ejection fraction only in CMV seronegative patients.

Figure 5

Infiltration of CD3 T-lymphocytes in myocardium. (A) Abundance of T-lymphocytes in myocardial tissue from human explanted hearts. n=21 (non-failing hearts: 8 seropositive, 7 seronegative; failing hearts: 6 seronegative). Mann Whitney non-parametric test. (B, C) Representative immunohistochemistry to demonstrate CD3⁺ T-lymphocyte infiltration (red arrows) in myocardium from non-failing hearts.