Noncoding variations in Cyp24a1 gene are associated with Klotho-mediated aging phenotypes in different strains of mice.

In mutant mice, reduced levels of Klotho promoted high levels of active vitamin D in the serum. Genetic or dietary manipulations that diminished active vitamin D alleviated aging-related phenotypes caused by Klotho down-regulation. The hypomorphic Klotho [kl/kl] allele that decreases Klotho expression in C3H, BALB/c, 129, and C57BL/6 genetic backgrounds substantially increases 1,25(OH)2D3 levels in the sera of susceptible C3H, BALB/c, and 129, but not C57BL/6 mice. This may be attributed to increased basal expression of Cyp24a1 in C57BL/6 mice, which promotes inactivation of 1,25(OH)2D3. Decreased expression of Cyp24a1 in susceptible strains was associated with genetic alterations in noncoding regions of Cyp24a1 gene, which were strongly reminiscent of super-enhancers that regulate gene expression. These observations suggest that higher basal expression of an enzyme required for catabolizing vitamin D renders B6-kl/kl mice less susceptible to changes in Klotho expression, providing a plausible explanation for the lack of aging phenotypes on C57BL/6 strain.

INTRODUCTION

The Klotho gene encodes α‐Klotho, a protein with sequence homology to β‐glucuronidase enzymes (reviewed in Dalton, Xie, An, & Huang, 2017). The hypomorphic Klotho (kl/kl) mutation displayed premature aging phenotypes in C3H‐kl/kl and BALB/c‐kl/kl mice (Kuro‐o et al., 1997). Deletion of the Klotho gene in 129 genetic background has also been shown to result in severe aging‐related phenotypes (University of California, Davis, Mutant Mouse Regional Resource Centers). By contrast, the kl/kl mutation on the C57BL/6 genetic background displayed normal lifespan with reduced aging‐related phenotypes (Phelps, Pettan‐Brewer, Ladiges, & Yablonka‐Reuveni, 2013). These observations indicate that genetic differences between inbred mouse strains play a role in aging‐related outcomes of the kl/kl mutation. Klotho is primarily expressed in the kidney where it has been shown to regulate vitamin D metabolism (Kuro‐o et al., 1997). Decreased Klotho expression in C3H‐kl/kl mice correlates with increased serum levels of FGF‐23, phosphate, calcium, and vitamin D3 [1,25(OH)2D3] (Kuro‐o, 2006). Increased levels of active 1α,25‐dihydroxy vitamin D3 [1α,25(OH)2D3] were shown to be a major cause of the aging‐related phenotypes in aging‐susceptible strains of Klotho mutant mice (Leibrock, Voelkl, Kuro, Lang, & Lang, 2016). However, the basis for the difference in age‐related phenotypes in kl/kl mutant mice on different genetic backgrounds remains unknown. To address the difference in aging‐related phenotypes between susceptible [C3H, 129, and BALB/c] and nonsusceptible [C57BL/6] strains, we assayed changes in metabolic pathways that have been associated with Klotho function. In particular, we addressed the expression of Cyp24a1 and Cyp27b1 genes whose products are known to regulate the levels of vitamin D3 in the serum (Christakos, Ajibade, Dhawan, Fechner, & Mady, 2010).

RESULTS AND DISCUSSION

Despite significantly reduced Klotho expression (Figure 1a), B6‐kl/kl mice showed no significant increases in serum phosphate, calcium, and vitamin D3 levels (Figure 1b–d) on a standard rodent diet (Supporting Information Table S1). Accordingly, the increase of FGF‐23 levels following Klotho decrease (Figure 1e) was also less dramatic in B6‐kl/kl mice than in the susceptible strains (Leibrock et al., 2016). These data provide a plausible explanation for the lack of premature aging phenotypes in B6‐kl/kl mice.

Figure 1

Description of B6‐kl/kl mice. (a) Klotho mRNA expression in B6 and B6‐kl/kl kidney normalized to Gapdh. (b) phosphate, (c) calcium, (d) vitamin D3, and (e) FGF‐23 levels in serum of B6 and B6‐kl/kl mice at different ages. C57BL/6 (young n = 6, old n = 7) and B6‐kl/kl (young n = 8, old n = 7). (f) Expression of Klotho, Cyp27b1, and Cyp24a1 in kidney of indicated mouse strains. C57BL/6 (n = 11), C3H (n = 8), 129 (n = 6), and BALB/c (n = 2) mice. *p < 0.05–0.001

To further probe the absence of increased vitamin D levels and aging‐related phenotypes, we interrogated expression of Cyp27b1 and Cyp24a1, which are involved in controlling serum levels of active vitamin D3, 1,25(OH)2D3 (Christakos et al., 2010). To determine basal expression, we assayed mRNA for Cyp27b1 and Cyp24a1 in kidneys from C57BL/6, C3H, 129, and BALB/c mice (Figure 1f). Whereas, basal levels of Klotho and Cyp27b1 expression were comparable in all strains of mice, we found that basal expression of Cyp24a1 was significantly higher in kidneys of C57BL/6 mice compared to susceptible strains (Figure 1f). Because Klotho/FGF‐23 heterodimer has been implicated in regulating vitamin D metabolism via Cyp27b1 and Cyp24a1, these observations suggest that lower basal level of Cyp24a1 expression in susceptible strains may render those mice more reliant on Klotho/FGF‐23 and FGFR‐axis to maintain healthy levels of serum vitamin D3. We propose that high basal expression of Cyp24a1 makes C57BL/6 mice less sensitive to decreased Klotho expression and contributes to abrogation of Klotho‐dependent aging‐related phenotypes in B6‐kl/kl mice.

To uncover regulatory mechanisms that alter expression of Cyp24a1 in susceptible mouse strains compared to C57BL/6, we examined chromatin state and genetic variations near the Cyp24a1 gene. Kidney‐specific expression of Cyp24a1 was evident in RNA sequencing profiles (Figure 2a, tracks labeled RNA). The 20 kb region immediately 3′ of Cyp24a1 contained four prominent DNase 1 hypersensitive sites (DHS) specific to kidney samples and a fifth site that was shared with liver (Figure 2a, tracks labeled DHS). The known relationship between DHS and transcriptional enhancers suggested that this region may regulate Cyp24a1 expression. As another measure of regulatory potential, we examined epigenetic histone modifications associated with transcriptional enhancers. This entire region was selectively marked by enhancer‐associated histone H3 acetylated at lysine 27 (H3K27ac) and histone H3 mono‐methylated at lysine 4 (H3K4me1) only in kidney samples. The broad distribution of these marks was reminiscent of super‐enhancers (Whyte et al., 2013; Witte, O'Shea, & Vahedi, 2015). The close tissue‐specific correlation between expression of Cyp24a1 and presence of H3K27ac‐ and H3K4me1‐marked modified DHS suggests that this cluster of DHS regulates kidney‐specific expression of Cyp24a1 gene.

Figure 2

Genetics and epigenetics of the Cyp24a1 gene locus. (a) A portion of chromosome 2 containing the Cyp24a1 gene and flanking sequences, top row; RNA, RNA‐seq tracks from indicated tissues; H3K4me1, H3K27ac, and H3K4me3 refer to locations of these forms of modified histones obtained from chromatin immunoprecipitation and sequencing (ChIP‐seq) analysis of indicated tissues, middle rows; DHS: DNase 1 hypersensitive site analysis of Cyp24a1 flanking sequences in indicated tissues, bottom row. (b) Comparison of tracks for conserved noncoding sequences for indicated species with reference C57BL/6 genome. (c) Single nucleotide polymorphism (SNPs) and insertions and deletions (Indels) in various mouse strains compared to reference C57BL/6 genome. Regions of possible regulatory function based on epigenetic marks are indicated by highlighted areas in a–c

Sequences surrounding these regions of chromatin modifications and DHS were also found to be conserved in other mammalian species, including humans (Figure 2b, tracks labeled CNS [conserved noncoding sequence]), suggesting common regulatory pathway across species. Remarkably, the same region contained multiple genetic variations/deletions in the Klotho‐susceptible strains compared to C57BL/6 (Figure 2c, tracks labeled Indels, SNPs). We propose that these noncoding genetic variations in the putative Cyp24a1 super‐enhancer lead to reduced basal expression of Cyp24a1 in susceptible strains making them more reliant on Klotho for regulation of Cyp24a1 to maintain healthy levels of serum vitamin D3.

CONFLICT OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

JMS designed and supervised the study; AV and MS performed the experiments; AS and RS did the genetic analysis of kidney‐specific variations in Cyp24a1 gene. JMS and RS wrote the paper and all authors read and commented on the manuscript.

Click here for additional data file.

Click here for additional data file.