Ablation of both Cx40 and Panx1 results in similar cardiovascular phenotypes exhibited in Cx40 knockout mice

Abstract

Connexins (Cxs) and pannexins (Panxs) are highly regulated large-pore channel-forming proteins that participate in cellular communication via small molecular exchange with the extracellular microenvironment, or in the case of connexins, directly between cells. Given the putative functional overlap between single membrane-spanning connexin hemichannels and Panx channels, and cardiovascular system prevalence, we generated the first Cx40^-/-Panx1^-/- mouse with the anticipation that this genetic modification would lead to a severe cardiovascular phenotype. Mice null for both Cx40 and Panx1 produced litter sizes and adult growth progression similar to wild-type (WT), Cx40^-/- and Panx1^-/- mice. Akin to Cx40^-/- mice, Cx40^-/-Panx1^-/- mice exhibited cardiac hypertrophy and elevated systolic, diastolic, and mean arterial blood pressure compared with WT and Panx1^-/- mice; however assessment of left ventricular ejection fraction and fractional shortening revealed no evidence of cardiac dysfunction between groups. Furthermore, Cx40^-/-, Panx1^-/-, and Cx40^-/-Panx1^-/- mice demonstrated impaired endothelial-mediated vasodilation of aortic segments to increasing concentrations of methacholine (MCh) compared with WT, highlighting roles for both Cx40 and Panx1 in vascular endothelial cell (EC) function. Surprisingly, elevated kidney renin mRNA expression, plasma renin activity, and extraglomerular renin-producing cell populations found in Cx40^-/- mice was further exaggerated in double knockout mice. Thus, while gestation and gross development were conserved in Cx40^-/-Panx1^-/- mice, they exhibit cardiac hypertrophy, hypertension, and impaired endothelial-mediated vasodilation that phenocopies Cx40^-/- mice. Nevertheless, the augmented renin homeostasis observed in the double knockout mice suggests that both Cx40 and Panx1 may play an integrative role.

Keywords: Pannexin 1; connexin; gap junctions.

Figure 1. Characterization of Cx40/Panx1 knockout mice

(A) qPCR revealed that Panx1 was expressed in the heart atria and ventricles, and the brain of WT mice, but absent from Panx1 null mice. (B) PCR genotyping confirmed that both Panx1 and Cx40 were ablated in double knockout mice as revealed in lanes 2–5 representing different mice. Cx43 was present in all mice and the insertion of the Cx40 neocassette (Cx40-NC) reaffirmed the ablation of Cx40. (C) WT and knockout mouse mean litter size at birth (number of litters from 10–17 dams, WT, n=30; Cx40^−/−, n=29; Panx1^−/−, n=50; Cx40^−/−Panx1^−/−, n=15; Dams). (D) Representative photograph revealing that mice of all four genotypes are of similar size. (E) Assessment of mouse weight over 1 year revealed that all WT and knockout mice have similar weights and weight gain (n=3–6 litters, 10 mice/time point).

Figure 2. Cx40 localization and expression levels are unchanged in Panx1-ablated mice

(A) Fluorescent micrographs reveal similar localization of Cx40 (green) in WT and Panx1^−/− mouse atria (arrows = Cx40 gap junctions). The absence of Cx40 gap junction plaques in Cx40^−/− and Cx40^−/− Panx1^−/− confirms the ablation of Cx40. (B) Western blot and (C) quantitation reveal similar levels of Cx40 in WT and Panx1^−/− mice, and its ablation in Cx40^−/− and Cx40^−/−Panx1^−/− atria (n=3). Immunoblotting for GAPDH was used as a loading control. Nuclei in (A) were stained with Hoechst dye. Scale bar = 20 μm.

Figure 3. Increased heart mass found in Cx40^−/− and Cx40^−/−Panx1^−/− mice

(A) Photographic examples of representative hearts from 3-month-old mice across all genotypes highlighting the enlarged hearts in Cx40^−/− and Cx40^−/−Panx1^−/− mice. When heart weights were examined, it was found that Cx40^−/− and Cx40^−/−Panx1^−/− mice had increased heart weight relative to body weight at the ages of (B) 3 weeks (n=4) and (C) 3 months (n=6) as compared with WT and Panx1^−/− mice. (D) Mean kidney weights were compared in 3-month-old mice and found to be similar (n=6). ***, P<0.001; ****, P<0.0001; ns, not significantly different.

Figure 4. Cx40^−/− and Cx40^−/−Panx1^−/− mice have hypertrophic cardiomyocytes

Wheat germ agglutinin staining (red) of the cardiomyocyte cell surface in left (A) atrial and (B) ventricular cross-sections taken from 3-month-old WT and knockout mice. Quantitation of the average cardiomyocyte cell area revealed greater atrial (n=10–13; (C)) and ventricular (n=5; (D)) cardiomyocyte size in Cx40^−/− and Cx40^−/−Panx1^−/− mouse hearts. *, P<0.05; **, P<0.01, ***, P<0.001. Scale bar = 20 µm.

Figure 5. Masson’s Trichrome staining and assessment of extracellular matrix proteins in the hearts of knockout mice

Light micrographs of Masson’s Trichrome staining in the left (A) atria and (B) ventricle of 3-month-old mice revealed minimal fibrosis in all knockout mice compared with WT. Western blotting revealed that the levels of collagen I (col I) and fibronectin (Fib) were similar between mouse genotypes in (C) atria and (D) ventricle tissue protein lysates (n=3). (E) Phalloidin (red) was used to visualize F-actin in the ventricle of WT and knockout mice. Immunoblotting for GAPDH was used as a loading control. Trichrome staining images; scale bar = 40 μm; fluorescent phalloidin images; scale bar = 20 μm.

Figure 6. Junctional proteins are unaltered in the hearts of knockout mice

In 3-month-old WT, Cx40^−/−, Panx1^−/−, and Cx40^−/−Panx1^−/− mice Cx43 (green) and N-cadherin (N-Cad, orange) localization was assessed by immunofluorescence in the left (A) atria and (B) ventricles. Western blots of Cx43 and N-cadherin in atria (C) and ventricle (D) lysates revealed no difference in the abundance of these junctional proteins between mouse genotypes (n=3). Nuclei (A,B) were stained with Hoechst dye. Arrows and arrowheads in (A,B) denote Cx43 gap junction plaques and N-cadherin at the ICDs. Dashed lines in the Western blots shown in (D) reflect using the mirror image of original Western blot to maintain consistent loading order of samples. Scale bar = 20 µm.

Figure 7. Cx40^−/− and Cx40^−/−Panx1^−/− mice are hypertensive

Millar pressure catheter recordings of arterial blood pressure revealed that Cx40^−/− and Cx40^−/−Panx1^−/− mice have significantly higher (A) systolic, (B) diastolic, and (C) mean arterial blood pressure than WT and Panx1^−/− mice (n=4–6). (D) Mean heart rate is similar amongst all four genotypes (n=4–6). **, P<0.01; ***, P<0.001, ****, P<0.0001.

Figure 8. Left ventricular contractility is conserved in Cx40^−/− and Cx40^−/−Panx1^−/− mice, despite cardiac hypertrophy and hypertension

Representative echocardiographic M-Mode images (A). Left ventricular anterior and posterior wall dimensions during systole and diastole (B–E; *, P<0.05, **, P<0.01). Left ventricular diameter during systole and diastole were similar across mouse genotypes as measured using M-Mode echocardiography, thus calculated left ventricular ejection fraction (F) and fractional shortening (G) were comparable amongst WT, Cx40^−/−, Panx1^−/−, and Cx40^−/−Panx1^−/− mice (n=6–8).

Figure 9. Endothelium-dependent vasodilatory responses are blunted in aortic segments of knockout mice, where responses to vasoconstriction are unaltered

Dose–response curves to (A) PE, (B) KCl, (C) MCh, and (D) SNP were performed to assess thoracic aorta contractility and dilation, amongst 2–4-month-old WT, Cx40^−/−, Panx1^−/−, and Cx40^−/−Panx1^−/− mice. MCh-evoked vasodilatory responses of aortic segments from knockout mice were compromised compared with WT mice at concentrations 1 µM and greater. * represents difference from WT, *P<0.05; **P<0.01; ****P<0.0001. Error bars = SEM.

Figure 10. Ectopic renin-producing cell localization, and elevated kidney renin mRNA expression and plasma renin activity in Cx40^−/− and Cx40^−/−Panx1^−/− mice

(A) Kidney renin mRNA levels were elevated in Cx40^-/- and Cx40^−/−Panx1^−/− mice compared with WT and Panx1^−/− mice (n=7–8). (B) Plasma renin activity was elevated in Cx40^−/− mice compared with WT, where plasma renin activity levels of Cx40^−/−Panx1^−/− mice surpassed that of Cx40^−/− mice (n=6–7). (C) Populations of renin-producing cells (arrows) were observed in the periglomerular space in Cx40^−/− mice and to a greater extent in Cx40^−/−Panx1^−/− mice, compared with normal juxlaglomerular positioning at the afferent arteriole proximal to the glomerulus (asterisks) in WT and Panx1^−/− mice. *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. Scale bar = 20 μm.