Genetic variants of mineral metabolism in health and disease

Abstract

Purpose of review

Disturbances in mineral metabolism are common among individuals with chronic kidney disease and have consistently been associated with cardiovascular and bone disease. The current review aims to describe the current knowledge of the genetic aspects of mineral metabolism disturbances and to suggest directions for future studies to uncover the cause and pathogenesis of chronic kidney disease – mineral bone disorder.

Recent findings

The most severe disorders of mineral metabolism are caused by highly penetrant, rare, single-gene disruptive mutations. More recently, genome-wide association studies (GWAS) have made an important contribution to our understanding of the genetic determinants of circulating levels of 25-hydroxyvitamin D, calcium, phosphorus, fibroblast growth factor-23, parathyroid hormone, fetuin-A and osteoprotegerin. Although the majority of these genes are known members of mineral homeostasis pathways, GWAS with larger sample sizes have enabled the discovery of many genes not known to be involved in the regulation of mineral metabolism.

Summary

GWAS have enabled remarkable developments in our ability to discover the genetic basis of mineral metabolism disturbances. Although we are far from using these findings to inform clinical practice, we are gaining understanding of novel biological mechanisms and providing insight into ethnic variation in these traits.

INTRODUCTION

Disturbances in mineral metabolism worsen with loss of kidney function, and are characterized by vitamin D deficiency, elevated levels of fibroblast growth factor-23, low calcitriol and secondary hyperparathyroidism [1–5]. These disturbances contribute to the burden of cardiovascular and bone disease both in chronic kidney disease (CKD) and in the general population [6–13].

A genetic basis for variability in circulating levels of mineral metabolism markers has long been suggested by rare Mendelian disorders with large effects on mineral homeostasis. More recently, large-scale genome-wide association studies (GWAS) have enabled the discovery of common variants with smaller contributions to systemic levels. Central elements of the pathophysiology of CKD mineral bone disorder remain incompletely understood. However, the elucidation of these loci has provided insights into the molecular pathogenesis of mineral metabolism disorders and may inform the development of new therapies, improved diagnosis, prevention and potentially prediction of risk and severity of complications.

The current review aims to describe the current knowledge of the genetic aspects of mineral metabolism disturbances, to suggest directions for future studies to delineate the cause and pathogenesis of CKD – mineral bone disorder, which could eventually pave the way for improvements in clinical management.

MONOGENIC DISORDERS OF MINERAL METABOLISM

Rare and strongly penetrant single-gene mutations cause the most severe disorders of mineral metabolism, recognizable clinical syndromes and monogenic disease. Traditionally, gene identification for monogenic disorders arises by studying affected kindreds for cosegregation with polymorphic genetic markers to define the chromosomal location, followed by DNA sequence analysis of genes located within the candidate region. More recently, this approach has been superseded by whole-exome and whole-genome sequence analysis of affected patients or kindreds [14,15]. Using these methods, the genetic cause of many monogenetic disorders leading to abnormal mineral metabolism has been described (Table 1). Some of these disorders are caused by loss-of-function or gain-of-function mutations affecting the same gene. For example mutations in the calcium-sensing receptor gene (CASR) cause familial hypocalciuric hypercalcemia whereas gain-of-function CASR mutations cause autosomal dominant hypocalcemia [16].

Table 1.

Monogenic disorders of mineral metabolism

Monogenic disorder	Genes	Ref.
Mode of inheritance: autosomal dominant
Autosomal dominant hypophosphatemic rickets	FGF23	[17]
FHH, types 1–3	CASR, GNAH, AP2S1	[18,19]
ADH, types 1–2	CASR, GNA11	[19]
Hypophosphatasia	TNSALP/ALPL	[20]
Vitamin D-dependent rickets, type 3	CYP3A4	[21]
Autosomal dominant hypoparathyroidism	PTH	[22]
Familial isolated hypoparathyroidism	GCM2	[23]
PHP1a	GNAS	[24]
PPHP	GNAS	[24]
PHP1b	GNAS	[24]
MEN1	MEN1	[25]
Kenny-Caffey syndrome, type 2	FAM111A	[26]
Mode of inheritance: autosomal recessive
Hypophosphatasia	TNSALP/ALPL	[20]
NSHPT	CASR	[18]
Vitamin D-dependent rickets, type 1A	CYP27B1	[27]
Vitamin D-dependent rickets, type 1B	CYP2R1	[28]
Vitamin D-dependent rickets, type 2A	VDR	[27]
Vitamin D-dependent rickets, type 2B	HNRNPC	[29]
Infantile hypercalcemia	CYP24A1	[30]
Autosomal recessive hypoparathyroidism	PTH	[22]
Autosomal recessive hypophosphatemic rickets	DMP1, ENPP1	[31]
Hereditary hypophosphatemic rickets with hypercalciuria	SLC34A3	[32,33]
Mode of inheritance: X-linked dominant
XLH rickets	PHEX	[34]

ADH, autosomal dominant hypocalcemia; FHH, familial hypocalciuric hypercalcemia; MEN1, multiple endocrine neoplasia 1 with hyperparathyroidism; NSHPT, neonatal severe hyperparathyroidism; PHP1a, pseudohypoparathyroidism, type 1a; PHP1b, pseudohypoparathyroidism, type 1b; PPHP, pseudopseudohypoparathyroidism; XLH, X-linked hypophosphatemic.

POLYGENIC DETERMINANTS OF MINERAL METABOLISM

Mineral metabolism disorders are complex traits produced by the interplay between susceptible genotypes and the provoking environmental factors such as kidney impairment or dietary factors. Direct effects of genetic variants, interactions among these genes, and between the genes and environment, as well as the epigenetic mechanisms regulating the gene expression might all contribute to the phenotype. GWAS have made an important contribution to our understanding of the genetic determinants of these traits and allow identification of the direct effects of genetic variants on mineral metabolism markers. Over the past decade, GWAS of circulating calcium, phosphorus, 25-hydroxyvitamin D [25(OH)D], parathyroid hormone (PTH), fibroblast growth factor-23 (FGF23), fetuin-A and osteoprotegerin have been conducted, shedding light on common genetic determinants of mineral metabolism markers (Table 2) [35–41]. Beyond uncovering novel biologic pathways, GWAS enable the construction of robust genetic instruments, explaining 3–16% of the variability in mineral metabolism traits. These instruments can be used as tools for downstream analysis such as polygenic risk scores [42] or Mendelian randomization [43].

Table 2.

Genome-wide association studies for markers of mineral metabolism

Marker	Sample size	Nearby genes	Variance explained	Population	Ref.
Phosphorus	16 264 (discovery) 5444 (replication)	ALPL, CSTA, RGS14, ENPP3, IP6K3, C12orf4	1.5%	European ancestry	[39]
Phosphorus	162255	SLC34A1, ENPP3 ETV6, LINC0182, PHEX, ALPL, IP6K3, FGF6, ALPL	0.9%	Japanese ancestry	[44]
Calcium	12 865	CASR	1.26%	European and Indian Asian ancestry	[45]
Calcium	20611	CASR	0.5%	European ancestry	[46]
Calcium	71 701	ITPR1, DOK7, PHTF2, STIM1, ATP2B1, PAQR5, ZFPM1, BCAS3, DGKD, CASR CYP24A1	1.1%	Japanese ancestry	[44]
Calcium	16 164 (discovery) 16 098 (replication)	CASR		European ancestry	[47]
Calcium	39400 (discovery) 21 875 (replication)	CYP24A1, GATA3, CARS, DGKD, GCKR KIAA0564		European ancestry	[38]
Vitamin D	229 (discovery) 1190 (replication)	DAB1		Hispanic	[48]
Vitamin D	673 (age 6) 1140 (age 14)	PDE3B, CYP2R1, GC, NPY		Western Australia	[49]
Vitamin D	4501 (discovery) 2221 (replication)	GC, NADSYN1, DHCR7, ACADSB, CYP2R1, C10orf88	2.8%	European ancestry	[50]
Vitamin D	1387 (discovery) 2151 (replication)	GC, CYP2R1, DAB1, FOXA2, SSTR4		Punjabi Sikh	[51]
Vitamin D	1829 (discovery) 1534 (replication)	GC, CYP2R1		European ancestry	[52]
Vitamin D	28 150	GC, KIF4B, DHCR7, ANO6, ARID2, HTR2A		Trans-ethnic	[53]
Vitamin D	42 274	CYP2R1		European ancestry	[54]
Vitamin D	79 366 (discovery) 40562 (replication)	GC, NADSYN1, DHCR7, CYP2R1, CYP24A1, SEC23A AMDHD1	7.5%	European ancestry	[37]
Vitamin D	16 125 (discovery) 9367 (replication) 8504 (replication)	GC, DHCR7, CYP2R1, CYP24A1	4%	European ancestry	[55]
Vitamin D	443 734 (meta-analysis)	GC, DHCR7, CYP2R1, CYP24A1 and 63 novel loci	16%	European ancestry	[56◼◼]
PTH	22 653 (discovery) 6502 (replication)	CYP24A1, RGS14, CLDN14, RTDR1, CASR	4.5%	European ancestry	[36]
FGF-23	16 624 (discovery) 4443 (replication)	CYP24A1, ABO, RGS14, LINC01506 LINC01229	3%	European and African ancestry	[35]
Fetuin-A	11174	AHSG	14%	European and African ancestry	[40]
Osteoprotegerin	10336	TNFRSF11B, COLEC10, MAL2, FJ40504, POLDIP2, TMEM97, MIR4723, VTN, IFT20, SARM1, TNFIP1, SLC46A1, TMEM199, SEBOX	11%	European and Asian ancestry	[41]

Mendelian randomization has emerged as a powerful study design that can provide evidence supporting or refuting causality by using the genetic determinants of a risk factor. Akin to a randomized trial, because of the random assortment and segregation of alleles at meiosis, genetic variation can be used as an effective tool to distinguish potentially causal from noncausal biomarkers. Under certain strong assumptions, single nucleotide polymorphisms associated with a risk factor through GWAS can be used as instruments to investigate the relationship between the risk factor and disease in an unconfounded manner. In addition, since genetic variants are assigned at conception and largely stable over the course of an individual’s life, Mendelian randomization studies have the correct temporal ordering to overcome the possibility of reverse causation. The increasing size and coverage of GWAS for mineral metabolites and the greater availability of summarized data on genetic associations are enabling the application of Mendelian randomization methods to gain understanding of mineral metabolism disturbances and their clinical consequences. Questions that can be addressed using conventional and more advanced Mendelian randomization methods include whether mineral metabolism biomarkers represent causal processes for complications, how kidney function impacts these processes and which biomarker, if any, may be the most promising interventional target.

Given the pleiotropic functions of vitamin D in maintaining human health, 25(OH)D has received a great deal of scientific attention, and accordingly, the genetic architecture of circulating 25(OH)D has been most extensively studied [37,50,53,55]. Four of the genes consistently reported through GWAS of 25(OH)D have a well recognized role in vitamin D metabolism. The group component gene, which encodes the vitamin D binding protein, is highly polymorphic, with isoforms differing strongly by race and by affinity for 25(OH)D [57,58]. CYP24A1 is critically important for maintaining serum 1,25(OH)₂D concentrations within a tight physiologic range and preventing vitamin D toxicity by catalyzing the conversion of 1,25(OH)₂D to 1,24,25(OH)₃D for subsequent excretion [59]. DHCR7 encodes 7-dehydrocholesterol reductase, which converts 7-dehydrocholesterol to cholesterol in skin, thereby removing the substrate from the synthetic pathway of 25(OH)D_3. CYP2R1 encodes another enzyme upstream of 25(OH)D, 25-hydroxylase, responsible for the hydroxylation of cholecalciferol and ergocalciferol in the liver. The latest (and largest) GWAS of 25(OH)D has now revealed 63 loci beyond those encoding known vitamin D metabolism enzymes, identifying new biological pathways influencing 25(OH)D. These novel loci are mostly related to lipid and skin keratinization phenotypes and warrant further functional investigation [56◼◼].

A similar pattern of discovery from GWAS has presented itself for other mineral metabolism markers. With larger sample sizes, the loci reported from GWAS move beyond those known to be proximally related to mineral metabolism to those with less-established roles. For example, in GWAS of circulating calcium, the CASR, a gene highly expressed in the parathyroid glands and kidneys, which influences PTH secretion, urinary Ca²⁺ excretion and skeletal development, was the first to be identified [45,46]. A recent GWAS in 71 701 Japanese individuals further identified 11 novel loci associated with serum calcium, including a newly discovered calcium sensor in the endoplasmic reticulum (STIM1) and genes with currently unknown functions (e.g. PAQR5 and PHTF2) [44]. Similarly, while variants within the CYP24A1 gene were unsurprising determinants of FGF23 levels, novel loci in ABO (the first discovered blood group system) and RGS14 (the regulator of G-protein signaling 14) were not. Adjacent to SLC34A1 (the NaPi2A sodium-phosphorus cotransporter), the RGS14 gene is of particular interest and ongoing investigation.

Recent evidence supports a role for RGS14, which is expressed in kidney proximal and distal tubule cells, in regulating PTH-sensitive phosphate transport [60◼].

Although mineral metabolism processes have traditionally been captured by established biomarkers (calcium, phosphate, vitamin D, FGF23 and PTH), novel ‘secondary pathways’ have emerged as prime candidates for causing complications. Through in-vitro and observational studies, investigators have come to appreciate the importance of fetuin-A and matrix-carboxyglutamic acid protein (matrix-Gla protein), which potentially inhibit ectopic calcification and promote bone mineralization [61–63], and osteoprotegerin, which enhances plaque calcification [64–67]. Circulating levels have been associated with arterial calcification and with mortality, in CKD and in the general population [68–70]. The actions in the vasculature of these bone-related factors are multifactorial, cell-mediated and dynamic, and still incompletely understood. For these biomarkers, the relative contribution of genetic variants to circulating concentrations is substantially higher than for traditional mineral metabolism markers (Table 2). This suggests that these biomarkers are regulated by fewer genes proximal to their metabolism, which may provide a more straightforward path to drug target identification.

Several identified genetic loci have been associated with more than one mineral metabolite. These pleiotropic sites suggest a shared genetic cause between traits. Genetic variants within CYP24A1 have been associated with circulating levels of calcium, 25(OH)D, PTH, FGF23 (Table 2) and estimated glomerular filtration rate [71]. Similarly, variants within RGS14 and CASR have also demonstrated pleiotropic effects on mineral metabolism markers. Distinguishing between direct and indirect associations of a single genetic variant on a given marker is difficult in a GWAS setting. Recent methods allowing for marginal analyses using summary statistics may allow for more thorough investigations of these associations [72].

Important differences in circulating levels of mineral metabolism markers and in the magnitude of associations of those markers with outcomes have been reported across race/ethnic groups [73–76]. Circulating 25(OH)D concentration varies strongly by race/ethnicity, being highest in white populations, intermediate in Hispanic populations, and lowest in African American populations [76,77]. As might be expected given differences in 25(OH)D, circulating concentrations of PTH are highest among African Americans [78]. However, neither the active form of vitamin D [1,25(OH)₂D] nor vitamin D binding protein concentrations are lower comparing African American to white race [79]. Genetic admixture analysis is one method that can be used to investigate the degree to which observed differences in a given trait are due to genetic (biologic) factors, rather than environmental factors, and these studies are currently underway [80].

GENETIC DETERMINANTS OF MINERAL METABOLISM MARKERS IN CHRONIC KIDNEY DISEASE

As we have discussed, large-scale GWAS have discovered common variants associated with mild differences in circulating levels mineral metabolism markers in the general population. It is possible that these effects may be more or less pronounced among individuals with overt mineral metabolism disturbances, such as those with CKD, in whom the biologic implications and potential treatment options are most relevant. This is the subject of ongoing work. As the development of kidney disease and mineral metabolism disturbances are strongly influenced by environmental factors, which can result in chromatin histone modifications and DNA methylation, it is likely that epigenetic mechanisms have a role in CKD complications.

CONCLUSION

It is apparent that a genetic and phenotypic continuum exists between rare monogenic disorders of mineral metabolism and genetic factors responsible for the control of homeostasis in the general population. Evidently, there is substantial overlap between the genes involved in rare monogenic diseases are those playing a role in common complex diseases and phenotypes (e.g. CASR, CYP24A1).

The use of GWAS has enabled remarkable developments in our ability to discover the genetic basis of complex traits, including mineral metabolism disturbances. Although we are far from using these findings to inform clinical practice, we are gaining understanding of novel biological mechanisms and providing insight into the ethnic variation in these traits. Hopefully the full potential of GWAS will be unlocked with larger sample sizes and technological and methodological advancements.

KEY POINTS.

• Rare and strongly penetrant single-gene mutations cause the most severe disorders of mineral metabolism, recognizable clinical syndromes and monogenic disease.

• GWAS have made an important contribution to our understanding of the genetic determinants of mineral metabolism.

• We are gaining understanding of novel biological mechanisms and providing insight into the ethnic variation in these traits.