Cardiac dysfunction in Pkd1-deficient mice with phenotype rescue by galectin-3 knockout

Abstract

Alterations in myocardial wall texture stand out among ADPKD cardiovascular manifestations in hypertensive and normotensive patients. To elucidate their pathogenesis, we analyzed the cardiac phenotype in Pkd1(cond/cond)Nestin(cre) (CYG+) cystic mice exposed to increased blood pressure, at 5 to 6 and 20 to 24 weeks of age, and Pkd1(+/-) (HTG+) noncystic mice at 5-6 and 10-13 weeks. Echocardiographic analyses revealed decreased myocardial deformation and systolic function in CYG+ and HTG+ mice, as well as diastolic dysfunction in older CYG+ mice, compared to their Pkd1(cond/cond) and Pkd1(+/+) controls. Hearts from CYG+ and HTG+ mice presented reduced polycystin-1 expression, increased apoptosis, and mild fibrosis. Since galectin-3 has been associated with heart dysfunction, we studied it as a potential modifier of the ADPKD cardiac phenotype. Double-mutant Pkd1(cond/cond):Nestin(cre);Lgals3(-/-) (CYG-) and Pkd1(+/-);Lgals3(-/-) (HTG-) mice displayed improved cardiac deformability and systolic parameters compared to single-mutants, not differing from the controls. CYG- and HTG- showed decreased apoptosis and fibrosis. Analysis of a severe cystic model (Pkd1(V/V); VVG+) showed that Pkd1(V/V);Lgals3(-/-) (VVG-) mice have longer survival, decreased cardiac apoptosis and improved heart function compared to VVG+. CYG- and VVG- animals showed no difference in renal cystic burden compared to CYG+ and VVG+ mice. Thus, myocardial dysfunction occurs in different Pkd1-deficient models and suppression of galectin-3 expression rescues this phenotype.

Keywords: ADPKD; Pkd1-deficiency; apoptosis; cardiac dysfunction; cardiomyopathy; galectin-3.

Figure 1

Strain analyses in Pkd1-deficient mice. Comparative analyses of (A, E, I and M) short-axis circumferential strain; (B, F, J and N) short-axis radial strain; (C, G, K and O) long-axis longitudinal strain; and (D, H, L and P) long-axis radial strain among (A–D) WT, HTG+ and HTG− mice at 5–6 wk of age; (I–L) WT, HTG+ and HTG− animals at 12–13 wk; (E–H) NC, CYG+ and CYG− mice at 5–6 wk; and (M–P) NC, CYG+ and CYG− animals at 22–23 wk. Parametric data were compared using one-way ANOVA, with results presented as mean±SD. A–D and I–L: *p<0.05 vs WT; †p<0.05 vs HTG+; ††p<0.01 vs HTG+; †††p<0.01 vs HTG+. EH and M–P: *p<0.05 vs NC; **p<0.01 vs NC; †p<0.05 vs CYG+; ††p<0.01 vs CYG+; †††p<0.01 vs CYG+.

Figure 2

Strain rate analyses in Pkd1-deficient mice. Comparative analyses of (A, E, I and M) short-axis circumferential strain rate; (B, F, J and N) short-axis radial strain rate; (C, G, K and O) long-axis longitudinal strain rate; and (D, H, L and P) long-axis radial strain rate among (AD) WT, HTG+ and HTG− mice at 5–6 wk of age; (I–L) WT, HTG+ and HTG− animals at 12–13 wk; (E–H) NC, CYG+ and CYG− mice at 5–6 wk; and (M–P) NC, CYG+ and CYG− animals at 22–23 wk. Parametric data were compared using one-way ANOVA, with results presented as mean±SD. A–D and I–L: *p<0.05 vs WT; †p<0.05 vs HTG+; †††p<0.01 vs HTG+. E–H and M–P: *p<0.05 vs NC; **p<0.01 vs NC; †p<0.05 vs CYG+; ††p<0.01 vs CYG+.

Figure 3

Structural and systolic function analyses in Pkd1-deficient mice. Comparative analyses of (A, E, I and M) left ventricular mass normalized to body weight (LVM/BW); (B, F, J and N) left ventricular internal diameter in diastole normalized to body weight (LVIDD/BW); (C, G, K and O) left ventricular ejection fraction (LVEF); and (D, H, L and P) myocardial performance index (MPI) among (A–D) WT, HTG+ and HTG− mice at 5–6 wk of age; (I–L) WT, HTG+ and HTG− animals at 12–13 wk; (E–H) NC, CYG+ and CTG− mice at 5–6 wk; and (M–P) NC, CYG+ and CYG− animals at 22–23 wk. Parametric data were compared using one-way ANOVA, with results presented as mean±SD. A–D and I–L: *p<0.05 vs WT; **p<0.01 vs WT; †p<0.05 vs HTG+; ††p<0.01 vs HTG+; and †††p<0.001 vs HTG+. E–H and M–P: *p<0.05 vs NC; **p<0.01 vs NC; †p<0.05 vs CYG+; ††p<0.01 vs CYG+; and †††p<0.001 vs CYG+.

Figure 4

Analyses of diastolic function in Pkd1-deficient mice. Comparative analyses of (A, E, I and M) E/A ratio; (B, F, J and N) deceleration time (DT); (C, G, K and O) isovolumetric relaxation time (IVRT); and (D, H, L and P) E/e′ ratio among (A–D) WT, HTG+ and HTG− mice at 5–6 wk of age; (I–L) WT, HTG+ and HTG− animals at 12–13 wk; (E–H) NC, CYG+ and CYG− mice at 5–6 wk; and (M–P) NC, CYG+ and CYG− animals at 22–23 wk. Parametric data were compared using one-way ANOVA, with results presented as mean±SD. Non-parametric data were compared by the Kruskal-Wallis test, with results expressed as median (lower to upper quartile). A–D and I–L: **p<0.01 vs WT; and †p<0.05 vs HTG+. E–H and M–P: **p<0.01 vs NC; and ***p<0.001 vs NC.

Figure 5

Polycystin-1, galectin-3 and TGF-β1 expression in heart tissue. Comparative western blot analyses of polycystin-1 expression among (A) WT (n=6), HTG+ (n=6) and HTG− (n=8) mice; and (D) NC (n=7), CYG+ (n=7) and CYG− (n=6) animals. Comparative analyses of galectin-3 expression between (B) WT (n=7) and HTG+ (n=9) mice; and (E) NC (n=5) and CYG+ (n=5) animals. Comparative analyses of TGF-β1 expression among (C) WT (n=7), HTG+ (n=9) and HTG− (n=5) mice; and (F) NC (n=4), CYG+ (n=4) and CYG− (n=4) animals. A, D: 6 wk; B, C: 13 wk; E, F: 23 wk. Immunoblots showing polycystin-1/GAPDH expression ratio in (G) WT, HTG+ and HTG− mice; and (I) NC, CYG+ and CYG− animals. L: lung tissue; K: medullary kidney tissue. Immunoblots showing galectin-3/GAPDH and TGF-β1/GAPDH expression ratios in (H) WT, HTG+ and HTG− mice; and (J) NC, CYG+ and CYG− animals. Parametric data were compared by one-way ANOVA in A and D, with results presented as mean±SD. Non-parametric data were compared by the Mann-Whitney test in B, C, E and F, with results presented as median (lower to upper quartile). A–C: **p<0,01 vs WT; and ***p<0,001 vs WT. D–F: *p<0,05 vs NC; and †p<0,05 vs CYG+.

Figure 6

Apoptosis and fibrosis in heart tissue. Comparative analyses of (A) TUNEL staining among WT (n=7), HTG+ (n=9) and HTG− (n=7) mice; (B) TUNEL staining among NC (n=9), CYG+ (n=7) and CYG− animals (n=8); (C) Sirius red staining among WT (n=8), HTG+ (n=11) and HTG− (n=8) mice; and (D) Sirius red staining among NC (n=9), CYG+ (n=7) and CYG− (n=6) animals. A, C: 13 wk; B, D: 23 wk. (E) Representative images of TUNEL staining in cardiac tissue. Original magnification, x400; inserts x800. Scale bars, 10 μm. (F) Representative images of Sirius red staining in heart tissue. Original magnification, x400; scale bars, 10 μm. Non-parametric data were compared by Kruskal-Wallis test, with results expressed as median (lower to upper quartile). A, C: **p<0.01 vs WT; †††p<0.001 vs HTG+. B, D: **p<0.01 vs NC; ***p<0.001 vs NC; ††p<0.01 vs CYG+; †††p<0.001 vs CYG+.

Figure 7

Renal phenotype in Pkd1-deficient mice. Comparative analyses of SUN among (A) WT, HTG+ and HTG− mice at 10–11 wk of age, and among (B) NC, CYG+ and CYG− animals at 20–21 wk; and comparative analyses of renal cystic index between (C) CYG+ and CYG− mice at 22–23 wk. (D) Representative images of galectin-3 staining in WT, HTG+, HTG−, NC, CYG+ and CYG− kidneys. Original magnification, x400; scale bar, 10 μm. The * symbol indicates a cyst. (E) H&E representative images of kidney cysts in CYG+ and CYG− mice, and absence of cysts in NC mice. Original magnification, x40; scale bar, 50 μm. (F) Representative ultrasound images in longitudinal axis of NC, CYG+ and CYG− kidneys; scale bar, 2.5 mm. Non-parametric data were compared by Kruskal-Wallis test (A, B) and Mann-Whitney test (C), with results expressed as median (lower to upper quartile).

Figure 8

Cell proliferation, apoptosis and fibrosis in renal tissue. Comparative analyses of renal cell Ki-67 staining among (A) WT (n=6), HTG+ (n=10) and HTG− (n=6) organs, and (D) NC (n=7), CYG+ (n=5) and CYG− (n=7) in whole kidney, and in CYG+ and CYG− in cyst epithelia; comparative analyses of renal TUNEL staining among (B) WT (n=5), HTG+ (n=6) and HTG− (n=5) mice in whole kidney, and in CYG+ and CYG− cyst epithelia, and (E) NC (n=7), CYG+ (n=6) and CYG− (n=6) animals; and comparative analyses of area of interstitial fibrosis among (C) WT (n=8), HTG+ (n=11) and HTG− (n=7) kidneys, and (F) NC (n=9), CYG+ (n=7) and CYG− (n=7) organs. (G) Representative images of Ki-67 staining. (H) Representative images of TUNEL staining. (I) Representative images of Sirius red staining. G–H: Original magnification, x400; inserts x800; scale bar, 10 μm. I: Original magnification, x200; scale bar, 10 μm. Non-parametric data were compared by Kruskal-Wallis or Mann-Whitney test, with results expressed as median (lower to upper quartile). A–C: **p<0.01 vs WT; ***p<0.001 vs WT; †p<0.05 vs HTG+. D–F: *p<0.05 vs NC; ***p<0.001 vs NC. †p<0.05 vs CYG+. D, E: ¥¥p<0.01 vs CYG+ in cyst epithelia.

Figure 9

Effects of Lgals3 knockout in survival, cardiac and renal phenotypes in Pkd1^+/+, Pkd1^V/V and Pkd1^V/V;Lgals3^−/− mice. (A) Survival of VVG+ and VVG− mice. The median age of survival was 21 days [20–24] for VVG+ mice (n=22) mice and 30 days [26–33] for VVG− animals (n=22). (B) Left ventricular ejection fraction in WT, VVG+ and VVG− mice at P18. (C) M-mode images used for LVEF calculation in (B). (D) Comparative analyses of renal cystic index between VVG+ and VVG− mice at P18. (E) Representative ultrasound images in transversal axis of VVG+ and VVG− kidneys; scale bar, 2.5 mm. In (A), data were compared with the Log-rank test. ***p<0.001 vs VVG+. In (B), parametric data were compared using one-way ANOVA, with results presented as mean±SD. In (D), non-parametric data were compared using the Kruskal-Wallis test, with results expressed as median (lower to upper quartile). **p<0.01 vs WT; ††p<0.01 vs VVG+.

Figure 10

Cardiac phenotype in Pkd1^+/+, Pkd1^V/V and Pkd1^V/V;Lgals3^−/− mice at P18. (A) TUNEL staining in WT (n=5), VVG+ (n=6) and VVG− (n=7) hearts; (B) area of cardiac interstitial fibrosis in WT (n=5), VVG+ (n=6) and VVG− (n=7) mice. (C) Comparative western blot analysis of galectin-3 expression in heart tissue in WT (n=6) and VVG+ (n=6) animals; (D) comparative analysis of TGF-β1 expression in WT (n=6), VVG+ (n=6) and VVG− (n=5) hearts; (E) comparative analysis of pSmad2/Smad2 ratio in WT (n=6), VVG+ (n=6) and VVG− (n=6) hearts. Representative images in heart tissue of (F) galectin-3 expression; (G) TUNEL staining; and (H) Sirius red staining. In (F–H), original magnification, x400; inserts x800; scale bar, 10 μm. In (H), original magnification, x200; scale bar, 10 μm. Immunoblots showing galectin-3 (I) and TGF-β1 (J) expression relative to GAPDH in WT, VVG+ and VVG− hearts. Parametric data were compared using one-way ANOVA, with results presented as mean±SD. Non-parametric data were compared by Kruskal-Wallis test, with results expressed as median (lower to upper quartile). **p<0.01 vs WT; †p<0.05 vs VVG+.

Figure 11

Renal phenotype in Pkd1^+/+, Pkd1^V/V and Pkd1^V/V;Lgals3^−/− mice at P18. Comparative analyses of (A) SUN among WT (n=5), VVG+ (n=6) and VVG− (n=8) mice; (B) Ki-67 staining in renal tissue among WT (n=5), VVG+ (n=6) and VVG− (n=8) animals; (C) TUNEL staining in renal tissue among WT (n=5), VVG+ (n=5) and VVG− (n=6) mice; and (D) area of interstitial fibrosis in renal tissue among WT (n=5), VVG+ (n=6) and VVG− (n=8) animals. Representative images in kidney tissue of (E) galectin-3 expression; (F) Ki-67 staining; (G) TUNEL staining; and (H) Sirius red staining. In (E–G), original magnification, x400; inserts x800; scale bar, 10 μm. In (H), original magnification, x200; scale bar, 10 μm. Non-parametric data were compared by Kruskal-Wallis test, with results expressed as median (lower to upper quartile). *p<0.05 vs WT; **p<0.01 vs WT; ***p<0.001 vs WT; †p<0.05 vs VVG+. ¥¥p<0.01 vs VVG+ in cyst epithelia.