Myocardial dysfunction occurs prior to changes in ventricular geometry in mice with chronic kidney disease (CKD)

Abstract

Uremic cardiomyopathy is responsible for high morbidity and mortality rates among patients with chronic kidney disease (CKD), but the underlying mechanisms contributing to this complex phenotype are incompletely understood. Myocardial deformation analyses (ventricular strain) of patients with mild CKD have recently been reported to predict adverse clinical outcome. We aimed to determine if early myocardial dysfunction in a mouse model of CKD could be detected using ventricular strain analyses. CKD was induced in 5-week-old male 129X1/SvJ mice through partial nephrectomy (5/6Nx) with age-matched mice undergoing bilateral sham surgeries serving as controls. Serial transthoracic echocardiography was performed over 16 weeks following induction of CKD. Invasive hemodynamic measurements were performed at 8 weeks. Gene expression and histology was performed on hearts at 8 and 16 weeks. CKD mice developed decreased longitudinal strain (-25 ± 4.2% vs. -29 ± 2.3%; P = 0.01) and diastolic dysfunction (E/A ratio 1.2 ± 0.15 vs. 1.9 ± 0.18; P < 0.001) compared to controls as early as 2 weeks following 5/6Nx. In contrast, ventricular hypertrophy was not apparent until 4 weeks. Hearts from CKD mice developed progressive fibrosis at 8 and 16 weeks with gene signatures suggestive of evolving heart failure with elevated expression of natriuretic peptides. Uremic cardiomyopathy in this model is characterized by early myocardial dysfunction which preceded observable changes in ventricular geometry. The model ultimately resulted in myocardial fibrosis and increased expression of natriuretic peptides suggestive of progressive heart failure.

Keywords: Animal model, experimental; blood pressure; cardiomyopathy; chronic kidney failure; echocardiography; fibrosis, endomyocardial; mice; uremia; ventricular dysfunction.

Figure 1

Partial nephrectomy (5/6Nx) results in chronic uremia and hypertension. Experimental design is presented in (A) where S₁ represents surgery 1, S₂ surgery 2, “e” echocardiography, “b” blood pressure measurement, “p” plasma studies. Trends in kidney function (plasma urea [B] and cystatin c [C]), systolic blood pressure (D), and body weight (E) in Sham versus chronic kidney disease (CKD) mice. *P < 0.05, **P < 0.01 between Sham and CKD groups at the same time point.

Figure 2

Temporal course of left ventricle structural and functional changes in chronic kidney disease (CKD) mice as measured by longitudinal echocardiography. Functional changes include global longitudinal strain (A) and mitral flow E/A ratio (B). Measures of structural remodeling include relative wall thickness (C) and left ventricular (LV) mass [median with 95% CI] (D) over time. Mean and standard deviations presented unless otherwise indicated; *indicates P < 0.05, **P < 0.01, ***indicates P < 0.001 between Sham and CKD groups at the same time point.

Figure 3

Chronic kidney disease (CKD) mice develop diastolic dysfunction with preserved ejection fraction. Invasive hemodynamic measures show preserved systolic function with no change in ejection fraction (A) or the rate of pressure increase during systole, dP/dt_Max (B), but signs of impaired relaxation as evidenced by prolonged relaxation time, Tau (C) and decreased rate of pressure decrease during diastole, dP/dt_Min (D). Hearts from CKD mice have increased expression of natriuretic peptides type A (Nppa, E) and type B (Nppb, F) compared to sham‐operated mice. *Indicates P < 0.05, **P < 0.01 between Sham and CKD groups at the same time point.

Figure 4

Chronic kidney disease (CKD) mice develop cardiac fibrosis. Representative photomicrographs of Picosirius red‐stained hearts from Sham at 8 weeks (A), CKD at 8 weeks (B), Sham at 16 weeks (C), and CKD at 16 weeks (D). Upper panel represents low‐power view (2× magnification, scale bar represents 2 mm) and bottom panel represents 10× magnification view of each picture above (scale bar represents 200 μm). Fibrosis quantification is presented in (E). Relative expression of fibrosis‐related transcripts (F), type 1 collagen (Col1a1), and connective tissue growth factor (Ctgf). *P < 0.05, **P < 0.01 between Sham and CKD groups at the same time‐point.

Figure 5

Calcium handling in isolated myocytes. Cardiomyocytes isolated from chronic kidney disease (CKD) mice at 8 weeks have shorter resting sarcomere length (A upper) but no appreciable accumulation of diastolic calcium (A lower). Measurements are from individual cells isolated from 3 to 4 mice per group are presented and are reflective of three separate experiments with comparative results. Hearts were collected at 8 and 16 weeks following surgery (n = 7–10 per group per time‐point) and mRNA expression quantified using qRT‐PCR for genes encoding proteins involved in sarcolemmal calcium handling (B). Atp2a2 (sarcoplasmic reticulum Ca²⁺ ATPase, SERCA2a), Slc8a1 (sodium/calcium exchanger, NCX), Pln (phospholamban), and Ryr2 (ryanodine receptor 2). **P < 0.01 between Sham and CKD.