Soluble Klotho Protects against Uremic Cardiomyopathy Independently of Fibroblast Growth Factor 23 and Phosphate

Abstract

Cardiac hypertrophy occurs in up to 95% of patients with CKD and increases their risk for cardiovascular death. In the kidney, full-length membranous Klotho forms the coreceptor for fibroblast growth factor 23 (FGF23) to regulate phosphate metabolism. The prevailing view is that the decreased level of Klotho in CKD causes cardiomyopathy through increases in serum FGF23 and/or phosphate levels. However, we reported recently that soluble Klotho protects against cardiac hypertrophy by inhibiting abnormal calcium signaling in the heart. Here, we tested whether this protective effect requires changes in FGF23 and/or phosphate levels. Heterozygous Klotho-deficient CKD mice exhibited aggravated cardiac hypertrophy compared with wild-type CKD mice. Cardiac magnetic resonance imaging studies revealed that Klotho-deficient CKD hearts had worse functional impairment than wild-type CKD hearts. Normalization of serum phosphate and FGF23 levels by dietary phosphate restriction did not abrogate the aggravated cardiac hypertrophy observed in Klotho-deficient CKD mice. Circulating levels of the cleaved soluble ectodomain of Klotho were lower in wild-type CKD mice than in control mice and even lower in Klotho-deficient CKD mice. Intravenous delivery of a transgene encoding soluble Klotho ameliorated cardiac hypertrophy in Klotho-deficient CKD mice. These results suggest that the decreased level of circulating soluble Klotho in CKD is an important cause of uremic cardiomyopathy independent of FGF23 and phosphate, opening new avenues for treatment of this disease.

Keywords: heart failure; renal failure; signaling.

Figure 1.

Comparable classic risk factors between WT and heterozygous Klotho-deficient CKD mice. Similar effects of 5/6 nephrectomy on (A) systolic BP (millimeters Hg), (B) creatinine clearance (CrCl; microliters per minute), (C) serum Pi (milligrams per deciliter), (D) and hematocrit (percent) of WT mice and mice Het for hypomorphic klotho allele. Mice were studied 4 weeks after 5/6 nephrectomy (CKD) or sham surgery. All values are means±SEMs. *P<0.05 versus sham.

Figure 2.

Serum-soluble Klotho levels are decreased in CKD mice. (A) Serum samples (100 μl each) from WT and Het klotho-hypomorphic (kl/+) mice 4 weeks after CKD or sham surgery were immunoprecipitated by anti-Klotho antibodies and analyzed for the abundance of Klotho by Western blotting. For comparison, serum samples (100 μl each) from mice homozygous for the klotho-hypomorphic allele (Homo; kl/kl) prespiked with 0, 0.2, 0.5, or 1 ng purified recombinant Klotho protein (rKl; thus giving concentrations of Klotho of 0, 15, 37, and 74 pM) were similarly immunoprecipitated and analyzed by Western blotting. (B) Means±SEMs (n=5–7 per group) of soluble Klotho levels determined as described above. Concentrations of soluble Klotho in unknown serum samples were determined by comparing densities of Klotho bands with standards curves obtained using known serum concentrations of Klotho between 0 and 74 pM as shown in A. Note that serum concentrations of soluble Klotho in CKD mice fall progressively postsurgery (data not shown). Results shown here are at 4 weeks postsurgery. MW, molecular mass. *P<0.05 versus sham. ^#P<0.05 between indicated groups.

Figure 3.

Het klotho-deficient CKD mice have aggravated cardiac hypertrophy compared with WT mice. (A) Heart weight-to-body weight (HW/BW; milligrams per gram) ratio and (B) brain natriuretic peptide (BNP) expression of WT and Het-klotho mice 4 weeks after CKD or sham surgery. All CKD mice included in the study were in general good health condition with a stable body weight during the 4 weeks postsurgery. Body weight of WT and Het-klotho CKD mice were not significantly different but slightly less than the respective sham controls (32.3±1.5, 31±2, 29±1.5, and 28.4±1.2 g for WT sham, Het sham, WT CKD, and Het CKD, respectively; mean±SEM, n=6–7 each). The abundance of mRNA for BNP was measured by quantitative RT-PCR and is shown as fold relative to WT sham mice (given the value of one). All values are means±SEMs (n=6–7). *P<0.05 versus sham. ^#P<0.05 between indicated groups.

Figure 4.

Klotho-deficient CKD mice have aggravated cardiac fibrosis compared with WT mice. (A) Representative trichrome staining of heart sections from WT and Het-klotho mice 4 weeks after CKD (shown in duplicate) or sham surgery (shown one time). Blue staining reflects collagen fibers. (B) Mean±SEM (n=6 per group) of the area of fibrosis relative to the total area of heart sections. *P<0.05 versus sham. ^#P<0.05 between indicated groups.

Figure 5.

Het klotho-hypomorphic mice have aggravated cardiac dysfunction. (A) Representative images shown in long-axis and cross-section views during end diastolic and end systolic phases. Left ventricle (LV) and right ventricle (RV) are indicated. (B) LVEDV (calculated by three-dimensional reconstruction of serial parallel cross-sectional images) is increased in Klotho-deficient CKD hearts. (C) Ejection fraction is decreased in Klotho-deficient CKD hearts. (D) Heart mass–to-LVEDV ratio is increased in WT CKD hearts. All values are means±SEMs (n=5 each). Heart mass was 140±5, 176±6, 127±5, and 202±7 mg for WT-sham, WT CKD, Het-sham, and Het-CKD, respectively (P<0.01 CKD versus sham; P<0.05 WT CKD versus Het-CKD; not significant WT-sham versus Het-sham).*P<0.05 versus sham.

Figure 6.

Klotho deficiency aggravates cardiac hypertrophy in CKD independently of effects on BP. (A) Systolic BP (SBP) of WT mice 4 weeks after CKD or sham surgery treated with or without antihypertensive drugs. Drugs were given in drinking water 5 days postsurgery. The treatment normalizes elevated BP in CKD mice to the level of sham mice. *P<0.05 versus sham; ^#P<0.05 between indicated groups. (B) Heart weight-to-body weight (HW/BW) ratio of the mice described in A. The antihypertensive treatment attenuates but does not completely abrogate cardiac hypertrophy in CKD mice. All values are means±SEMs (n=6–7 each). *P<0.05 versus sham; ^#P<0.05 between indicated groups. (C) SBP of WT and Het-klotho mice 4 weeks after CKD surgery treated with or without antihypertensive drugs. Treatment with antihypertensive drugs normalizes BP in both WT and Het-klotho CKD mice to the level of sham mice. All values are means±SEMs (n=7–8 each). *P<0.05 versus no drug treatment. (D) HW/BW ratio of mice described in C. Note that, although antihypertensive treatment significantly decreases HW/BW ratios in both WT and Het-Klotho CKD mice, the ratio in Het-Klotho mice remains elevated compared with WT mice. All values are means±SEMs (n=7–8 each). *P<0.05 Het versus WT; ^#P<0.05 treatment versus no drug treatment.

Figure 7.

Normalization of serum phosphate and FGF23 levels by dietary phosphate restriction does not prevent Klotho deficiency–induced aggravated cardiac hypertrophy in CKD mice. WT and Het-klotho mice after CKD or sham surgery were fed a low-Pi (0.2% inorganic phosphate) or a normal Pi (nl Pi; 0.35% inorganic phosphate) diet starting on day 5 postsurgery. (A) Serum phosphorus level, (B) serum FGF23 level (picograms per milliliter), and (C) heart weight-to-body weight (HW/BW) ratio were analyzed at 4 weeks after surgery. Note that, although feeding a low-Pi diet normalizes serum phosphate and FGF23 levels in both WT and Het-klotho CKD mice to the same levels of sham mice fed an nl Pi diet, it does not prevent aggravated CKD-induced cardiac hypertrophy in Het-klotho versus WT mice. All values are means±SEMs (n=7–8 each). In panel A, *P<0.05 CKD versus sham; ^#P<0.05 CKD on low-Pi diet versus CKD on nl Pi diet; not significant (ns) for WT versus Het in low-Pi diet as indicated. In panel B, *P<0.05 WT or Het on low-Pi diet versus WT or Het on nl Pi diet. In panel C, *P<0.05 Het versus WT.

Figure 8.

Transgenic expression of soluble Klotho attenuates cardiomyopathy in Klotho-deficient CKD mice. (A, upper panel) Experimental protocol for transgenic expression of Klotho by tail-vein injection of soluble Klotho–expressing (Kl-Tg) plasmid. Het klotho-deficient mice received CKD or sham surgery. The pilot study showed that tail-vein injection of a transgene encoding soluble Klotho in mice produced circulating Klotho detectable from 24 hours to about 2 weeks after injection and peaked at about days 2–4 after injection (data now shown). Thus, cytomegalo virus promoter (pCMV)–driven Kl-Tg plasmid or empty vector (V) was injected through the tail vein in mice at days 5, 15, and 25 after surgery. Mice were euthanized and studied at day 35 postsurgery. (A, lower panel) A representative image of immunoblot analysis of serum samples. Serum samples from Het-klotho CKD (C); mice at day 28 after (3 days after last injection of vector [V] or Klotho-transgene [Kl]) were analyzed for Klotho levels by immunoprecipitation and Western blotting as described in Figure 2. For comparison, the following serum samples were also included in the analysis: samples from homozygous klotho (Homo) mice (no surgery) that received either no injection or a single injection of Kl-Tg or vector for 3 days and samples from WT and Het klotho mice that received sham surgery but no transgene injection. Note that serum Klotho levels in Het-klotho CKD mice are barely detectable (Figure 2A), but levels are increased by injection of Klotho-expressing transgene. Inj, injection. (B) Heart weight-to-body weight (HW/BW) ratios of Het klotho-hypomorphic CKD mice injected with Klotho-expressing plasmid (Kl-Tg) or vector (Vec) compared with that of sham mice without injection. Note that transgenic expression of Klotho partially rescues cardiac hypertrophy in CKD mice. (C) Hypertrophic marker brain natriuretic peptide (BNP) mRNA levels in the same groups of mice in B were measured by quantitative RT-PCR and plotted as fold relative to that in sham mice (given value of one). (D) Representative trichrome-stained heart sections from mice as described in B and quantification of the area of fibrosis relative to the total area. (E) Creatinine clearance (CrCl) and (F) BUN (milligrams per deciliter) of mice as described in B. All values are means±SEMs (n=7–8 each). Gene deliv, gene delivery. *P<0.05 versus sham. ^#P<0.05 between indicated groups.

Figure 9.

Transgenic expression of Klotho does not alter serum FGF23 and phosphate levels in CKD mice. (A) Serum FGF23 and (B) phosphate levels of WT and Het-klotho mice receiving tail-vein injection of Klotho-expressing transgene (Kl-Tg) or empty vector (Vec) or no injection after 4 weeks of CKD or sham surgery. Protocol for injection of Klotho-expressing transgene is as described in Figure 8A. Note that FGF23 and phosphate levels are both elevated in CKD mice, but there is no difference between Het-Klotho CKD mice injected with Kl-Tg and Vec. All values are means±SEMs (n=7–8 each). Gene deliv, gene delivery. *P<0.05 versus sham.

Figure 10.

CKD induces upregulation of cardiac TRPC6, which is aggravated by Klotho deficiency. (A) TRPC6 mRNA levels (measured by RT-PCR) in the hearts of WT and Het-klotho mice who had sham or CKD surgery. All values are means±SEMs (n=7–8 each). (B) TRPC6-mediated currents were recorded from cardiac ventricular myocytes freshly isolated from mice as indicated. Currents were recorded in ruptured whole-cell mode. Voltage protocol consists of holding at −40 mV and repetitive descending ramp pulses from +120 to −120 mV. This protocol diminishes the activation of Na_v. Bath solution contains inhibitors for L-type Ca_v channel (Nifedipine; 1 μM) and Na⁺-Ca²⁺ exchanger (NiCl₂; 3 mM). Currents were recorded before and after the addition of endothelin-1 (ET1; 20 nM), and ET1-activated currents are shown. Upper panel shows representative current-voltage relationships of ET1-activated currents. Lower panel shows ET1-activated inward current (at −100 mV). Where indicated, purified recombinant soluble Klotho (sKl; 200 pM) was added and incubated for 2 hours before ruptured whole-cell recording. All values are means±SEMs (n=7–8 each). *P<0.05 versus sham; ^#P<0.05 between indicated groups.