Endothelial Pannexin 1 Regulates Cardiac Response to Myocardial Infarction

Cardiovascular disease is a leading cause of morbidity and mortality in the world. A recent cohort study identified a significant reduction in cardiovascular disease events, especially myocardial infarction (MI) and stroke, in patients taking probenecid compared to allopurinol¹. Probenecid is a potent pannexin 1 (Panx1) channel blocker. In studies from our group, we have identified endothelial cell (EC) Panx1 as an important regulator of ischemic outcome in multiple organs, specifically playing a role in regulating leukocyte infiltration². Here we examine the functional effects of MI with probenecid pharmacological intervention or genetic deletion of EC Panx1.

Male C57Bl6J mice, 12–20 weeks old, underwent coronary ischemia/reperfusion (IR) surgeries (blinded study; approved by University of Virginia and Tufts Medical Center Animal Care and Use Committees)³. Anesthetized (isoflurane) mice were intubated and the left anterior descending artery was ligated 1 mm distal to the left atrial appendage, confirmed by myocardium blanching, visual tachycardia, and ST elevation. After 60 min, the ligation was removed and saline or 1.1 mg/kg probenecid (Millipore Sigma, St. Louis, MO), randomly assigned, was intraperitoneally injected. Hearts were collected 14-days post-IR and fixed (neutral buffered formalin). Probenecid-treated mice had a significantly higher ejection fraction than saline treated mice 14-days post-IR (Figure 1A). Infarct sizes, quantified using polar plots from stained sections apex to base, and vascular densities, measured using isolectin (data not shown (DNS)) were not significantly different (Figure 1A).

Figure 1: Pharmacological inhibition or genetic deletion of EC Panx1 improves cardiac function post-MI.

Ejection fraction (A), measured via echocardiogram, in saline (n=9; baseline vs 1-day: p=0.000002; vs 14-day p=0.0000005) or probenecid (n=14; baseline vs 1-day: p=0.00005; vs 14-day p=0.0004) (Saline vs probenecid 14-day p=0.0002) treated mice. (A) Infarct size (Mason’s Trichrome representative images; saline: n=5, probenecid: n=14) 14-day post-IR hearts. (B) Ejection fraction and infarct size in Cre^- (n=9; baseline vs 1-day: p=0.000003; vs 14-day p=0.00003) and Cre⁺ mice (n=8; baseline vs 1-day: p=0.0003; vs 14-day p=0.0002) (Cre^- vs Cre⁺ 14-day p=0.009). (C) PV loop data from Cre^- (n=3) and Cre⁺ (n=5) mice (p=0.03). (D) Cytokine release (change from control siRNA) from human aortic EC treated with control or Panx1 siRNA, confirmed with qPCR (p=0.007), post-hypoxia/reoxygenation. (E) Total Ly6C, Ly6C^HI+ (p=0.04) and Ly6C low (Ly6C^LOW+) expressing monocytes/macrophages infiltration in Cre^- and Cre⁺ (n=5) mice. (F) Distribution of CCR2⁺ cells (magenta) (% of CCR2⁺ area within given region; Cre^- vs Cre⁺: p=0.03; Cre^- (n=3): infarct vs border: p=0.02; Cre⁺ (n=4): non-infarct vs infarct: p=0.003, vs border: p=0.03). Infarct area, reduced autofluorescence area (teal), and infarct border area (~0.5mm around the infarct) are delineated by the dashed line. Two-way ANOVA with Sidak’s multiple comparisons test (Prism 9) (A, B): vs. baseline: ***p<0.001 and ****p<0.0001; Saline/Cre^- vs Probenecid/Cre⁺: ### p<0.001 and ## p<0.01; (F) *p<0.05 and **p<0.01. Student’s t-test (A, B, C, D, E): *p<0.05. All data passed Shapiro-Wilk normality tests (p>0.05), performed within each group. Scale bar: 1 mm.

To examine if EC Panx1 deletion recapitulated the functional effects seen with probenecid treatment, we utilized 12–20 week inducible EC Panx1 deleted mice (Cdh5-Cre^ERT2+/Panx1^fl/fl; Cre⁺) and control mice (Cdh5-Cre^-/Panx1^fl/fl; Cre^-) injected with tamoxifen². Ejection fractions were significantly increased in Cre⁺ mice at 14-days post-IR (Figure 1B). Infarct sizes and vascular densities (DNS) were not significantly different (Figure 1B). Pressure-volume (PV) loops at 14 days post-IR showed decreased average rate of pressure decline suggesting improved diastolic function in Cre⁺ mice. dPdt max, contractility index, and ESPVR were not significantly different, although all trend towards increased contractility (Figure 1C). Thus, acute pharmacological blockade of Panx1 or genetic deletion of EC Panx1 was functionally beneficial in mouse models of MI.

Because Panx1 is a key regulator of inflammatory cell infiltration², we examined cytokine release from ECs following in vitro 3 h hypoxia/24 h reoxygenation (R&D Systems, Minneapolis, MN). RANTES, TNFα, and GM-CSF, which promote a pro-inflammatory status, were decreased in Panx1 knockdown cells; and M-CSF, an anti-inflammatory macrophage cytokine, was modestly increased (Figure 1D). Perfused and digested hearts, using the Langendorff preparation, 2-days post-IR were examined for infiltration of neutrophils (CD45-FITC: 1:100 and Ly6G-APC-Cy7: 1:200) and macrophages (CD45-FITC and Ly6C-PE: 1:200) using flow cytometry^{2, 3}. Neutrophils were not different (DNS). In Cre⁺ mice we found a significant reduction in the number of Ly6C high (Ly6C^HI+) expressing macrophages and no difference in total Ly6C⁺ macrophages (Figure 1E). The Ly6C^HI+ macrophages, identified by CCR2⁺ (1:100; Abcam, Cambridge UK), distribution differed between the Cre^- and Cre⁺ mice at 2-day post-IR, with significantly reduced accumulation in the non-infarct region in Cre⁺ mice (Figure 1F).

Ly6C^HI+ macrophages are pro-inflammatory and predominately found during the initial stages of injury post-MI⁴. Our data suggest an early reduction of Ly6C^HI+ macrophages. Whether this is due to a greater transition of Ly6C^HI+ to Ly6C^LOW+ or a change in the infiltration of these monocytes is unclear. This difference in Ly6C^HI+ expressing monocytes/macrophages number and localization within the non-infarct region could impact recovery in the heart in numerous ways, including improved tissue remodeling or facilitate cardiac electrical conduction, and remains an active area of study^{4, 5}. Macrophages are regulated by purinergic signaling providing a potential point of crosstalk for Panx1-mediated ATP release and early tissue inflammatory response.

In summary, we found that pharmacological inhibition or genetic deletion of EC Panx1 significantly improved cardiac function following MI possibly through an early shift to an anti-inflammatory status. These data provide strong evidence for the benefit of post-reperfusion inhibition of Panx1 using FDA-approved pharmacological drugs and suggest that Panx1 may contribute to the reduced mortality and incidence of cardiovascular disease in patients taking probenecid¹.

Data Availability.

The data that support the findings of this study are available from the corresponding author upon reasonable request.