Genetic Ablation of Fgf23 or Klotho Does not Modulate Experimental Heart Hypertrophy Induced by Pressure Overload

Abstract

Left ventricular hypertrophy (LVH) ultimately leads to heart failure in conditions of increased cardiac pre- or afterload. The bone-derived phosphaturic and sodium-conserving hormone fibroblast growth factor-23 (FGF23) and its co-receptor Klotho have been implicated in the development of uremic LVH. Using transverse aortic constriction (TAC) in gene-targeted mouse models, we examine the role of Fgf23 and Klotho in cardiac hypertrophy and dysfunction induced by pressure overload. TAC profoundly increases serum intact Fgf23 due to increased cardiac and bony Fgf23 transcription and downregulation of Fgf23 cleavage. Aldosterone receptor blocker spironolactone normalizes serum intact Fgf23 levels after TAC by reducing bony Fgf23 transcription. Notably, genetic Fgf23 or Klotho deficiency does not influence TAC-induced hypertrophic remodelling, LV functional impairment, or LV fibrosis. Despite the profound, aldosterone-mediated increase in circulating intact Fgf23 after TAC, our data do not support an essential role of Fgf23 or Klotho in the pathophysiology of pressure overload-induced cardiac hypertrophy.

Figure 1

Pressure overload by transverse aortic constriction (TAC) up-regulates Fgf23 expression in WT mice. (a) Representative H&E-stained cardiac cross-sections, 4-weeks after sham or TAC surgery (scale bar: 1000 µm). (b) Heart/body weight ratio is significantly increased after TAC when constriction was performed with a 27G-needle (n = 6–7). (c) Serum intact Fgf23 levels (n = 4–5). (d) mRNA expression of Fgf23 in the heart (left ventricle and septum) and bone (lumbar vertebra L5) (n = 5–6), normalised to expression of ornithine decarboxylase antizyme (OAZ). (e) Cleaved serum Fgf23 calculated as C-terminal Fgf23 - intact Fgf23 and presented as ratio of cleaved Fgf23/intact Fgf23 (n = 6–8). (f) Cardiac mRNA expression of genes involved in Fgf23 processing. GalNt3: N-Acetylgalactosaminyl-transferase 3, Furin and Fam20C (n = 5–7). Data were obtained 4-week post-surgery. Values are mean ± SEM. *p < 0.05, **p < 0.01. If not otherwise specified, TAC was performed using a 27 G needle.

Figure 2

Effect of spironolactone treatment on FGF23 levels, morphological and functional parameters, 2 weeks after TAC. (a) Serum aldosterone levels 4-week post-surgery (n = 4–5). (b) Spironolactone (Spiro) effect on serum intact Fgf23 levels after TAC (n = 4–6). (c) Effect of spironolactone treatment on cardiac and bony Fgf23 mRNA expression (lumbar vertebra L5) after TAC (n = 3–7). (d–f) Spironolactone did not affect cardiac hypertrophy development measured as (d) heart/body weight ratio, (e) mean cross-sectional area of cardiomyocytes and (f) thickness of the left ventricular (LV) wall measured by echocardiography. (g) Reduced cardiac function after TAC measured as fractional shortening by echocardiography is not affected by spironolactone treatment (h) Spironolactone treatment does not prevent development of lung oedema in TAC mice. (i) Effect of spironolactone (Spiro) treatment on mean blood pressure in Sham and TAC, measured proximal to the ligation by intra-aortic pressure catheter (n = 4–6). Data in (b–i) were obtained after 2 weeks of daily gavage with vehicle (Veh) or spironolactone (Spiro). n = 4–7 if not otherwise specified; mean ± SEM, in a–c *p < 0.05, **p < 0.01; in d–i *p < 0.05 and ***p < 0.001 vs. sham control of the same treatment regimen.

Figure 3

VDR deficiency does not alter the TAC-induced increase in circulating Fgf23 and in cardiac Fgf23 protein expression. (a) Serum intact Fgf23 levels measured by ELISA (n = 4–7), and (b) Western blot analysis of Fgf23 protein expression in the left ventricle normalised to GAPDH expression (n = 3) in WT, VDR^Δ/Δ and Klotho ^−/−/VDR^Δ/Δ mice, 4 weeks after TAC surgery. Data are mean ± SEM, *p value < 0.05, **p < 0.01, ***p < 0.001.

Figure 4

Loss of Fgf23 or Klotho does not modulate afterload-induced cardiac hypertrophy. (a) Heart/body weight ratio and (b) lung/body weight ratio (n = 6–10). (c) Cardiac mRNA expression of brain natriuretic peptide (Bnp, n = 5–6). (d) Left: Representative FITC-labelled wheat germ agglutinin (WGA)-stained sections (Scale bar: 50 µm). Right: Quantification of mean cardiomyocyte size after FITC-WGA staining (n = 6–9). (e) Fractional shortening, left ventricular wall thickness and diastolic functional parameter (E/A ratio) measured by echocardiography 4-weeks post-surgery (n = 6–9). (g) Representative echocardiograms of M-mode and pulsed-wave Doppler analysis in sham VDR^Δ/Δ and in TAC VDR^Δ/Δ, Fgf23 ^−/−/VDR^Δ/Δ and Klotho ^−/−/VDR^Δ/Δ mice. Data in (a–e) presented as mean ± SEM for VDR^Δ/Δ, Fgf23 ^−/−/VDR^Δ/Δ and Klotho ^−/−/VDR^Δ/Δ mice after sham and TAC (27 G) surgery. *p < 0.05 vs. sham control of the same genotype.

Figure 5

Fgf23 and Klotho deficiency do not protect from afterload-induced cardiac fibrosis. (a) Cardiac relative mRNA expression of Col1α (n = 5–6). (b) Quantification of fibrosis after picrosirius red staining (n = 6–9). (c) Representative images of total collagen in cardiac sections after picrosirius red staining (Scale bar: 100 µm). (d–f) Effect of pressure overload and Fgf23 or Klotho deficiency on cardiac Fgf receptor mRNA expression (n = 5–6). Data in (a–f) are from VDR^Δ/Δ, Fgf23 ^−/−/VDR^Δ/Δ and Klotho ^−/−/VDR^Δ/Δ mice, 4-weeks after sham and TAC (27 G) surgery. Data are mean ± SEM, in (a,b) *p < 0.05 vs. sham control of the same genotype; in (a) #p < 0.05 vs. VDR^Δ/Δ and Fgf23 ^−/−/VDR^Δ/Δ mice, in (d–f) *p < 0.05.