In chronic kidney disease altered cardiac metabolism precedes cardiac hypertrophy

Abstract

Conduit arterial disease in chronic kidney disease (CKD) is an important cause of cardiac complications. Cardiac function in CKD has not been studied in the absence of arterial disease. In an Alport syndrome model bred not to have conduit arterial disease, mice at 225 days of life (dol) had CKD equivalent to humans with CKD stage 4-5. Parathyroid hormone (PTH) and FGF23 levels were one log order elevated, circulating sclerostin was elevated, and renal activin A was strongly induced. Aortic Ca levels were not increased, and vascular smooth muscle cell (VSMC) transdifferentiation was absent. The CKD mice were not hypertensive, and cardiac hypertrophy was absent. Freshly excised cardiac tissue respirometry (Oroboros) showed that ADP-stimulated O₂ flux was diminished from 52 to 22 pmol/mg (P = 0.022). RNA-Seq of cardiac tissue from CKD mice revealed significantly decreased levels of cardiac mitochondrial oxidative phosphorylation genes. To examine the effect of activin A signaling, some Alport mice were treated with a monoclonal Ab to activin A or an isotype-matched IgG beginning at 75 days of life until euthanasia. Treatment with the activin A antibody (Ab) did not affect cardiac oxidative phosphorylation. However, the activin A antibody was active in the skeleton, disrupting the effect of CKD to stimulate osteoclast number, eroded surfaces, and the stimulation of osteoclast-driven remodeling. The data reported here show that cardiac mitochondrial respiration is impaired in CKD in the absence of conduit arterial disease. This is the first report of the direct effect of CKD on cardiac respiration.NEW & NOTEWORTHY Heart disease is an important morbidity of chronic kidney disease (CKD). Hypertension, vascular stiffness, and vascular calcification all contribute to cardiac pathophysiology. However, cardiac function in CKD devoid of vascular disease has not been studied. Here, in an animal model of human CKD without conduit arterial disease, we analyze cardiac respiration and discover that CKD directly impairs cardiac mitochondrial function by decreasing oxidative phosphorylation. Protection of cardiac oxidative phosphorylation may be a therapeutic target in CKD.

Keywords: CKD-mineral bone disorder; cardiac mitochondrial function; cardiorenal syndromes; chronic kidney disease; vascular calcification.

Graphical abstract

Figure 1.

Serum and plasma chemistry in Alport mice. Experimental groups include 1) WT, normal littermates of Alport mice; 2) CKD + IgG, Alport mice treated with an isotype-matched IgG; and 3) CKD + Ab, Alport mice treated with a mAb to activin A beginning at 75 days of life (dol) and proceeding through 225 dol at 25 mg/kg subcutaneous delivery twice weekly. A: blood urea nitrogen (BUN) is significantly elevated in Alport mice at 225 dol compared with WT littermates (n = 22–27). B: serum phosphorus (Pi) doesn’t show significant change (n = 18–20). Serum Ca (n = 18–20) (C) and log-transformed plasma PTH (D) are both significantly elevated in CKD (n = 16–19). E: the log-transformed plasma FGF23 levels are increased in CKD. Treatment with antibody (CKD + Ab) did not reduce significance compared with WT (n = 9–11). F: real-time RT-PCR of membrane klotho in renal homogenates (n = 7–9). Significance was determined using ANOVA with Tukey’s post hoc test (B–D) or Kruskal–Wallis H test with Bonferroni correction for multiple pairwise comparisons (A, E, and F). Includes data limited by animals with associated BUN < 35 for WT and >60 for CKD (B–F). Data are represented as means ± SD. CKD, chronic kidney disease; WT, wild type.

Figure 2.

Alport CKD. A: diseased kidneys show increased fibrosis by Picrosirius red histological staining. Fibrosis in medulla is more significant than outer cortex. Bar is 100 µm. B: quantification of relative fibrosis determined by color-based quantification of staining (n = 9–11). C: Pearson correlation between relative levels of fibrosis and the kidney function biomarker, BUN (n = 22). D: kidney Inhba mRNA expression (n = 7–9). Kidney smad3 (E), smad2 (F), and activin receptor IIA (acvr2a) (G) mRNA expression in CKD (n = 7–9). H: plasma activin A not significantly changed between groups (n = 16–19). All three groups contain mice with markedly increased activin A values with this assay. Observation of increase of these points in CKD as compared with WT not statistically supported in this dataset. Significance was determined using ANOVA with Tukey’s post hoc test (E–G) or Kruskal–Wallis H test with Bonferroni correction for multiple pairwise comparisons (B, D, and H). Includes data limited by animals with associated BUN <35 for WT and >60 for CKD (D–H). Data are represented as means ± SD. BUN, blood urea nitrogen; CKD, chronic kidney disease; smad3, suppressor of mothers against decapentaplegic homolog 3; WT, wild type.

Figure 3.

Vascular calcification and vascular stiffness. A: aortic calcium was not increased in CKD in these animals (n = 3–6). B and C: Tagln and alpha smooth muscle actin (Acta2) are expressed in vascular smooth muscle cells and remain unchanged with disease or treatment (n = 6–8). D: Runx2, a marker of osteoblastic transition, was not increased in CKD (n = 6–8). E: very low level Sost mRNA expression in the aorta was unaffected by Alport CKD (n = 6–8). Significance was determined using ANOVA with Tukey’s post hoc test (B) or Kruskal–Wallis H test with Bonferroni correction for multiple pairwise comparisons (A, C–E). Includes data limited by animals with associated BUN <35 for WT and >60 for CKD (A–E). F: Alport CKD mice were hypotensive compared with WT littermates. n = 7 for WT and 11 for CKD, significance was determined using Student’s t test. G: aortic compliance was unchanged in CKD. n = 3 WT, n = 4 CKD + IgG. Significant differences were examined using two-way ANOVA with Dunnett’s multiple comparison test and variables of pressure and genotype. Data are represented as means ± SD. BUN, blood urea nitrogen; CKD, chronic kidney disease; WT, wild type.

Figure 4.

Cardiac mass, oxidative phosphorylation, activin signaling, and mitochondrial electron microscopy. A: heart weight was not elevated in Alport CKD 225-day-old (do) mice (CKD + IgG) or Alport CKD mice treated with activin A antibody (CKD + Ab) compared with WT littermates (n = 19 or 20). B: left ventricular mass as determined by echocardiography was not increased with disease (n = 6–9). Significance was examined using ANOVA with Tukey’s post hoc test (A and B). Includes data limited by animals with associated BUN <35 for WT and >60 for CKD (A and B). C: addition of palmitoyl carnitine and malate to permeabilized cardiac fibers to measure LEAK respiration (n = 8 or 9). D: sequential addition of ADP to cardiac muscle fibers to measure fatty acid oxidative phosphorylation (OXPHOS) capacity (n = 8 or 9). Cardiac fibers from 225-day-old Alport mice (CKD + IgG) had significantly reduced rates of fatty acid oxidative phosphorylation compared with cardiac muscle fibers from WT mice. Significance was determined using Kruskal-Wallis H test with Bonferroni correction for multiple pairwise comparisons (C and D). Inclusion criteria for the respirometry cohort were WT mice with BUN (mg/dL) <35 and CKD + IgG and CKD + antibody (CKD + Ab) treatment mice with BUN >60. Data were limited to cardiac fibers with cytochrome c-stimulated oxygen flux increase <35%. E: immunoblot of OXPHOS complexes: (CI) subunit NDUF8B, CII, (CIII) core protein 2, (CIV) subunit I, and (CV) alpha subunit. There is no significant change in protein quantities normalized to GAPDH. F: immunoblots showing no significant change in total SMAD2 or phosphorylated SMAD2 normalized to GAPDH. The order of samples used for each lane is repeated in all blots shown. G: electron microscopy of cardiac myocyte mitochondria. Compared with same-age WT mice, 225-do Alport mice (CKD + Ig) and activin A antibody treatment group (CKD + Ab) had more areas of increased interlamellar mitochondrial rounding and spacing (white arrows). All images were in ×5,000 magnification, scale bar is 2 μm. Data are represented as means ± SD. BUN, blood urea nitrogen; WT, wild type.

Figure 5.

RNA sequencing of Alport CKD hearts. Differential gene expression analysis of 225-day-old Alport (CKD + IgG) (n = 6) compared with WT mice (n = 8) with global perturbations in signaling and metabolism pathways in the KEGG pathway (A) database and Hallmark gene set from the Mouse Molecular Signatures Database (MSigDB) collection (B). Significantly upregulated (red) or downregulated pathways (blue) are listed with false discovery rate (FDR) <0.05 and P value <0.01. C: immunoblots showing no significant change in cardiac PGC1-α protein levels (normalized to GAPDH) between WT, Alport (CKD + IgG), and Alport-treated (CKD + Ab) mice. The biological samples used for each lane are the same as all blots in Fig. 4. CKD, chronic kidney disease; KEGG, Kyoto Encyclopedia of Genes and Genomes; PGC1-α, peroxisome proliferator-activated receptor-gamma coactivator-1 alpha; PPAR, peroxisome proliferator-activated receptor; WT, wild type.

Figure 6.

Skeletal histomorphometry of Alport mice. A: bone volume was not significantly changed in CKD. Osteoclast number (N.Oc/B Pm × 100) (B) and eroded surfaces (ES/BS) (C) were increased with disease (CKD + IgG), and activin A antibody (CKD + Ab) eliminated the significant change. D: osteoclast surface (Oc.S/BS) was not significantly changed. E: mineralizing surfaces (MS/BS) were increased in Alport IgG (CKD + IgG) mice. F: bone formation rates were increased in Alport IgG mice and decreased toward WT levels in Alport Ab treated mice. n = 12 WT, 7 CKD + IgG, and 8 CKD + Ab. Significance was determined using ANOVA with Tukey’s post hoc test (A–D and F) or Kruskal–Wallis H test with Bonferroni correction for multiple pairwise comparisons (E). Data are represented as means ± SD. CKD, chronic kidney disease; WT, wild type.

Figure 7.

Plasma sclerostin. CKD increases sclerostin levels in plasma (n = 16–18), significance was determined using Kruskal–Wallis H test with Bonferroni correction for multiple pairwise comparisons. Includes data limited by animals with associated BUN <35 for WT and >60 for CKD. Data are represented as means ± SD. BUN, blood urea nitrogen; CKD, chronic kidney disease; WT, wild type.