Connecting the dots on fibroblast growth factor 23 and left ventricular hypertrophy

Abstract

Apart from FGF23, the work by Leifheit-Nestler et al. provides welcome new insights into what role α-Klotho may have in cardiomyocyte hypertrophy.

Left ventricular hypertrophy (LVH) is common in individuals with chronic kidney disease (CKD), particularly those with end-stage renal disease (ESRD) [1, 2], and is strongly associated with cardiovascular disease events and death in CKD patients [3]. These data suggest that slowing the progressive increase in left ventricular mass as kidney function declines is key to reducing disproportionately high rates of cardiovascular mortality in CKD patients. Historically, the main strategies for achieving this end have focused on treating hypertension and avoiding volume overload. However, the high prevalence of LVH in CKD cannot be completely explained by these factors alone [3], indicating that other risk factors unique to CKD play a role. Among these less traditional risk factors for LVH, fibroblast growth factor 23 (FGF23) has emerged as a potential new therapeutic target.

FGF23 is a hormone that helps to maintain phosphorus balance in the setting of diminished nephron capacity by increasing urinary phosphorus excretion and decreasing circulating 1,25-dihydroxyvitamin D concentrations [4]. Increases in circulating FGF23 concentrations are among the earliest (if not the earliest) manifestation of disturbed mineral metabolism in early CKD, often predating the rise of other markers of mineral metabolism such as parathyroid hormone and phosphorus [5]. While this physiological adaptation is critical for maintaining phosphorus balance in CKD, prolonged exposure to elevated FGF23 concentrations may have long-term maladaptive effects. Converging lines of evidence from animal and human studies suggest that elevated circulating FGF23 concentrations are causally related to the increased prevalence of LVH in CKD through direct inducement of cardiomyocyte hypertrophy via the phospholipase C (PLC) γ/calcineurin/nuclear factor of activated T-cells (NFAT) pathway [6, 7]. Recent findings further suggest that fibroblast growth factor receptor 4 (FGFR4) is the critical FGFR required for FGF23 to induce cardiomyocyte hypertrophy [7]. While exciting, evidence linking the findings in animal studies to humans has been lacking, making it difficult to determine to what extent the experimental data are relevant to humans with CKD. For these reasons, the data presented by Leifheit-Nestler and colleagues in the current issue of Nephrology Dialysis Transplantation represent an important first step in addressing this issue [8].

Leifheit-Nestler et al. analyzed myocardial autopsy samples from all individuals with childhood-onset ESRD who died while receiving renal replacement therapy in four European pediatric nephrology centers from 1980 to 2012 (cases, n= 24) and an equal number of age-, sex- and race-matched deceased controls free of heart or kidney disorders or abnormal anthropometric values at the time of death. Cardiovascular disease events accounted for 32% of the deaths for cases, which is consistent with the incidence of fatal cardiovascular events described in other large pediatric ESRD registries [9]. Similarly, the prevalence of LVH in cases (67%) is within the range of values reported in other pediatric ESRD studies [10, 11], suggesting that the cases in this study were broadly representative of the larger population of Caucasian pediatric ESRD patients, at least with respect to cardiac pathology. As expected, cases and controls were well matched by sex, age and race, and only differed in height, body weight and relative heart weight.

In agreement with two prior studies of adults with heart failure [12, 13], Leifheit-Nestler and colleagues found that the full-length (intact) FGF23 peptide was detectable in cardiac tissue lysates, with the majority of the expression occurring within cardiomyocytes. They also found that FGF23 mRNA expression was higher in myocardium of cases as compared with controls, which is somewhat in contrast to findings from a prior study which showed no difference in cardiac FGF23 mRNA expression in adults with end-stage heart disease (most of whom had kidney disease) as compared with normal controls [13]. Distinct from prior studies was the finding that FGF23 mRNA expression in cases with autopsy-proven LVH (LVH⁺) was higher than in cases without LVH (LVH⁻), and that higher expression of FGF23 in myocardial tissue correlated with higher cardiomyocyte cross-sectional area and higher cardiac B-type natriuretic peptide (BNP) expression. These findings suggest that the quantity of FGF23 expression in cardiac myocytes is at least in part determined by the degree of myocyte hypertrophy.

Another novel result of this study was the finding that, despite no differences in the expression of FGFR1 in myocardial tissue from cases and controls, FGFR4 mRNA expression was significantly higher in cases than controls. Additionally, among cases, FGFR4 expression was higher in those with LVH as compared with those without LVH, and positively correlated with cardiomyocyte cross-sectional area and FGF23 expression. Interestingly, the myocardial expression of the regulatory subunit of calcineurin and NFAT were higher in LVH⁺ cases than LVH⁻ cases, which is broadly consistent with the notion that the calcineurin-mediated NFAT pathway is upregulated in the setting of LVH [14]. Further, FGFR4, FGF23, calcineurin and NFAT expression were all lower in cases with functioning renal transplants than in cases who were receiving dialysis at the time of death.

The findings with respect to the expression of FGFR4, calcineurin and NFAT are noteworthy in that they suggest that all the machinery is present in human myocardial tissue to support a hypertrophic signaling cascade characterized by FGFR4-dependent up-regulation of the PLC γ/calcineurin/NFAT pathway as previously demonstrated in animal models [6, 7]. What of course is missing is empirical evidence that this pathway is functionally operative in humans. In the absence of these latter data, the results of Leifheit-Nestler et al. essentially serve as provocative pilot data supporting the need for future studies establishing the relevance of FGF23 and FGFR4 in the pathophysiology of LVH in humans. To do so will ultimately require human clinical trials showing that direct inhibition of this pathway reduces left ventricular mass in CKD. There are also some important questions not answered by the data as presented. For example, while it is shown that FGF23, FGFR4, calcineurin and NFAT expression were higher in LVH⁺ cases than LVH⁻ cases, we do not know whether this was independent of demographic and clinical differences between two groups. LVH⁺ cases were on average 8 years older and had received renal replacement therapy for a longer period of time (by ∼4 years) at the time of death than LVH⁻ cases. This begs the question of whether the differences in expression and protein levels between LVH⁺ and LVH⁻ cases were a function of age and length of renal replacement therapy as opposed to anything having to do with FGF23 and LVH. In addition, while it is shown that FGFR4, FGF23, calcineurin and NFAT expression were lower in cases with functioning renal transplants than in cases who were receiving dialysis at the time of death, no information was provided as to how many transplant patients had LVH, precluding us from determining to what extent differences in the prevalence of LVH may have accounted for these latter findings. Perhaps more importantly, the cross-sectional design of the study does not allow us to determine whether receiving a kidney transplant directly lowered expression of FGFR4, FGF23, calcineurin and NFAT in transplant patients. Other limitations of the study include the fact that it examined only pediatric patients—whether similar findings would have been observed in adults is unclear.

Apart from FGF23, the work by Leifheit-Nestler et al. also provides welcome new insights into what role α-Klotho may have in cardiomyocyte hypertrophy. As they found in their animal studies [6, 7], the membrane-found form of α-Klotho was not detected in myocardial lysates of cases or controls. However, the circulating form of α-Klotho was detectable in tissue samples from cases, with the quantity being lower in those with LVH than those without LVH and inversely associated with cardiomyocyte cross-sectional area. This provides the first evidence that lower quantities of circulating α-Klotho associate with greater hypertrophy of cardiomyocytes in humans, consistent with data showing that circulating α-Klotho protects against cardiac hypertrophy in animal models of CKD [15]. It will be important to determine whether circulating α-Klotho similarly associates with LVH in humans once a reliable assay for this protein becomes available [16].

These are exciting times for research related to the link between disorders of phosphorus metabolism and cardiovascular disease in CKD. The emergence of FGF23 and Klotho in the pathophysiology of LVH have provided new fronts in the battle against the seemingly inexorable rise in left ventricular mass in CKD patients. While the final picture is still fuzzy, the results of Leifheit-Nestler et al. help connect some of the dots of this increasingly complex puzzle. Moreover, these data support the importance of human clinical trials examining whether targeting elevated FGF23 can meaningfully improve cardiovascular outcomes in CKD patients.

CONFLICT OF INTEREST STATEMENT

The results of this study have not been published previously in whole or part.

(See related article by Leifheit-Nestler et al. Induction of cardiac FGF23/FGFR4 expression is associated with left ventricular hypertrophy in patients with chronic kidney disease. Nephrol Dial Transplant 2016; 31: 1088–1099)