**Monogenic Phosphate Balance Disorders**

Helge Raeder, Silje Rafaelsen and Robert Bjerknes *Department of Clinical Medicine, University of Bergen,* 

*Norway* 

## **1. Introduction**

414 Contemporary Aspects of Endocrinology

Porter RH, Cox BG, Heaney D, Hostetter TH, Stinebaugh BJ, Suki WN. (1978) Treatment of

Puig-Domingo M, Díaz G, Nicolau J, Fernández C, Rueda S, Halperin I. (2008) Successful

subcutaneous infusion of teriparatide. Eur J Endocrinol. Nov;159(5):653-7. Riccardi D, Gamba G. (1999) The many roles of the calcium-sensing receptor in health and

Sato M, Ma YL, Hock JM, Westmore MS, Vahle J, Villanueva A , Turner CH. 2002 Skeletal

Shiohara M, Shiozawa R, Kurata K, Matsuura H, Arai F, Yasuda T, Koike K. Effect of

Shoback D. (2008) Clinical practice. Hypoparathyroidism.N Engl J Med. Jul 24;359(4):391-

Sthephen J. 2000 Hyperparathyroid and hypoparathyroid disorders. N Engl J Med.

Torregrosa NM, Rodríguez JM, Llorente S, Balsalobre MD, Rios A, Jimeno L, Parrilla P. 2005

Valero RA, Senovilla L, Núñez L, Villalobos C. The role of mitochondrial potential in control

Winer KK, Yanovski JA, Cutler GB Jr. (1996) Synthetic human parathyroid hormone 1-34 vs

Winer KK, Yanovski JA, Sarani B, Cutler GB Jr.(1998) A randomized, cross-over trial of

Winer KK, Sinaii N, Peterson D, Sainz B Jr, Cutler GB Jr. 2008 Effects of once versus twice-

Winer KK, Sinaii N, Reynolds J, Peterson D, Dowdy K, Cutler GB Jr.(2010) Long-term

hypoparathyroidism. J Clin Endocrinol Metab. 1998 Oct;83(10):3480-6. Winer KK, Ko CW, Reynolds JC, Dowdy K, Keil M, Peterson D, Gerber LH, McGarvey C,

calcium. J Clin Endocrinol Metab. 2003 Sep;88(9):4214-20

calcium. J Clin Endocrinol Metab. 2010 Jun;95(6):2680-8.

Clin Endocrinol Metab. Sep;93(9):3389-95.

patient: parathyroid allotransplantation. Thyroid. 15(11):1299-302.

Rosen CJ. 2011 Clinical practice. Vitamin D insufficiency. N Engl J Med. 20;364(3):248- Sanda S, Schlingmann KP, Newfield RS. 2008 Autosomal dominant hypoparathyroidism

disease. Arch Med Res. Nov-Dec;30(6):436-48.

Endocrinol Metab. Apr;21(4):385-91.

81.

304–313.

403.

J.53(6):797-802.

343(25):1863-75.

276(8):631-6.

(September 2008) pp 259-69.

hypoparathyroid patients with chlorthalidone. N Engl J Med. Mar 16;298(11):577-

treatment of vitamin D unresponsive hypoparathyroidism with multipulse

with severe hypomagnesemia and hypocalcemia, successfully treated with recombinant PTH and continuous subcutaneous magnesium infusion. J Pediatr

efficacy with parathyroid hormone in rats was not entirely beneficial with longterm treatment. Journal of Pharmacological and Experimental Therapeutics 302

parathyroid hormone administration in a patient with severe hypoparathyroidism caused by gain-of-function mutation of calcium-sensing receptor. Endocr

Definitive treatment for persistent hypoparathyroidism in a kidney transplant

of calcium signals involved in cell proliferation. Cell Calcium. Vol. 44, nº 3

calcitriol and calcium in the treatment of hypoparathyroidism. JAMA. Aug

once-daily versus twice-daily parathyroid hormone 1-34 in treatment of

Cutler GB Jr. (2003) Long-term treatment of hypoparathyroidism: a randomized controlled study comparing parathyroid hormone-(1-34) versus calcitriol and

daily parathyroid hormone 1-34 therapy in children with hypoparathyroidism. J

treatment of 12 children with chronic hypoparathyroidism: a randomized trial comparing synthetic human parathyroid hormone 1-34 versus calcitriol and The last decade has seen that several of the dominant and recessive forms of hypo- and hyperphosphatemic bone disease have received their molecular explanation. This has led to new insight into the pathophysiology of hypo- and hyperphosphatemic bone disease, as well as the understanding of a bone-kidney axis which operates integrated and in parallel with the classical parathyroid-kidney axis in the regulation of phosphorus content in the body. In addition, it has led to the recognition of a Janus face of some of the involved genes, showing both hyper- and hypofunction, dependent on the nature of the mutation. In this book chapter, we will present an update on the emerging insight of monogenic hypo- and hyperphosphatemic disorders.

## **2. Genetic mechanisms and pathophysiology**

Hypophosphatemia may lead to bone or dental disease resulting from decreased mineralization (calcification) of bone or dental matrix or osteoid. The simultaneous blood calcium levels will also influence the degree of mineralization. Hypophosphatemia leads to rickets in children or to osteomalacia in adults. In many of the hypophosphatemic conditions, there is also an impairment of renal activation of vitamin D, further aggravating disease. The mineralization of teeth can also be affected, and there are clinical forms where bone affection is minimal and the dental disorders dominate.

Hyperphosphatemia may lead to increased mineralization of both bone and non-bone tissues (ectopic calcification) due to an increase in the body content of phosphorus. This results in tumoral calcinosis with calcification of muscles, skin and vessels. The monogenic forms affect the renal handling of phosphorus by various mechanisms resulting from inactivation or activation of the involved genes.

With the advancement of genetic insight and the subsequent possibility to study subjects with mutations and a wide range of phenotypes, a broader phenotypic pattern is recognized. Consequently, we suggest the more appropriate terms of monogenic hypophosphatemia and monogenic hyperphosphatemia for these disorders, and that the specific disorders should be classified according to the affected gene, e.g. *PHEX*hypophosphatemia and *FGF23*-hyperphosphatemia (Table 1).

We will now provide an overview of the genes directly implicated in monogenic phosphate balance disorders. Please refer to textbooks for a discussion of genes indirectly affecting phosphate balance (i.e. genes leading to defective parathyroid gland development or disrupted PTH receptor function).

Monogenic Phosphate Balance Disorders 417

integral membrane zinc-dependent endopeptidase family. The gene is expressed in a wide variety of tissues including the kidney with a higher expression in mature osteoblasts and odontoblasts. The substrate for the gene product is not known, but the pathogenesis seems to involve phosphate regulating humoral factors, phosphatonins, where the fibroblast growth factor-23 (FGFR-23) is central (Jonsson, Zahradnik et al. 2003; Juppner 2007; Bastepe and Juppner 2008). (See section 2.11 for a discussion on the physiological and pathophysiological mechanisms involved.) The protein is also believed to be involved in bone and dentin mineralization. Both the whole-body and bone-specific (osteocalcinpromoted inactivation) knockout mouse model of PHEX as well as the spontaneous Hyp mouse model display increased bone production, increased levels of serum FGF23, decreased kidney membrane NPT2 and osteomalacia (Yuan, Takaiwa et al. 2008). Cell studies indicate mechanistic defects both during protein processing in the endoplasmic reticulum and cell membrane (Sabbagh, Boileau et al. 2001) and as abrogated catalytic

There are several mutations associated with PHEX-hypophosphatemia (see the PHEX mutation database: http://www.phexdb.mcgill.ca/) and most of the mutations are located in the region encoding the extracellular domain, but there are also examples of pathological mutations in the 5'UTR (Dixon, Christie et al. 1998) and 3'UTR (Ichikawa, Traxler et al. 2008)

Fig. 1. *PHEX* gene structure and the corresponding encoded regions. Adapted from

Debray et al. 2009). The expressivity varies (Brame, White et al. 2004).

There is no clear genotype-phenotype correlation (Holm, Nelson et al. 2001). There is a slight dominance of familiar mutations (showing co-segregation with disease in a pedigree) to *de novo* mutations (sporadic) reported in literature (Holm, Nelson et al. 2001). The penetrance is high, although there are examples of non-penetrance (Gaucher, Walrant-

The *DMP1* gene (dentin matrix acidic phosphoprotein 1; MIM\* 600980) gene consists of 6 exons on chromosome 4q21, and was first implicated in a phosphate balance disorder in 2006 Lorenz-Depiereux, Bastepe et al. 2006). *DMP1* is highly expressed in osteocytes, and is a member of the 'SIBLING' (small integrin binding ligand n-linked glycoprotein) family of non-collagenous extracellular matrix proteins involved in bone mineralization (Huq, Cross

activity (Sabbagh, Boileau et al. 2003).

(Sabbagh, Boileau et al. 2003).

of the gene.

**2.2 DMP1** 


Table 1. Overview of genes directly implicated in monogenic phosphate balance disorders AR, autosomal recessive; AD autosomal dominant, HHRH, Hereditary hypophosphatemic rickets with hypercalciuria.

## **2.1 PHEX**

The *PHEX* (Phosphate-regulating endopeptidase homolog, XB; MIM\* 300550) gene consists of 22 exons (Sabbagh, Boileau et al. 2003) and was positionally cloned in 1995 (HYP Consortium 1995). This gene is encoding a transmembrane protein and belongs to the type II

hypophosphatemic rickets *PHEX*-hypophosphatemia

rickets *DMP1*-hypophosphatemia

rickets *FGF23-* hypophosphatemia

with hypophosphatemia *FGFR1-* hypophosphatemia

and hyperparathyroidism *KL-*hypophosphatemia

tumoral calcinosis *KL-*hyperphosphatemia

rickets, type 2 *ENPP1-*hypophosphatemia

*SLC34A1-*

*SLC9A3R1-*

hypophosphatemia

hypophosphatemia

hypophosphatemia

*GALNT3*-

hyperphosphatemia

*FGF23-* hyperphosphatemia

**Gene Classical disease name Suggested new name** 

hyperphosphatemic tumoral calcinosis

hyperphosphatemic tumoral calcinosis

Hypophosphatemic rickets

Hyperphosphatemic

Hypophosphatemic nefrolithiasis/ osteoporosis

Hypophosphatemic nephrolithiasis/ osteoporosis

Table 1. Overview of genes directly implicated in monogenic phosphate balance disorders AR, autosomal recessive; AD autosomal dominant, HHRH, Hereditary hypophosphatemic

The *PHEX* (Phosphate-regulating endopeptidase homolog, XB; MIM\* 300550) gene consists of 22 exons (Sabbagh, Boileau et al. 2003) and was positionally cloned in 1995 (HYP Consortium 1995). This gene is encoding a transmembrane protein and belongs to the type II

*SLC34A3* HHRH *SLC34A1-*

AD hypophosphatemic

*PHEX* X-linked dominant

*GALNT3*

*FGF23*

*KL*

*SLC34A1*

*SLC9A3R1*

**2.1 PHEX** 

rickets with hypercalciuria.

*DMP1* AR hypophosphatemic

*FGFR1* Osteogophonic dysplasia

*ENPP1* AR hypophosphatemic

Familial

Familial

integral membrane zinc-dependent endopeptidase family. The gene is expressed in a wide variety of tissues including the kidney with a higher expression in mature osteoblasts and odontoblasts. The substrate for the gene product is not known, but the pathogenesis seems to involve phosphate regulating humoral factors, phosphatonins, where the fibroblast growth factor-23 (FGFR-23) is central (Jonsson, Zahradnik et al. 2003; Juppner 2007; Bastepe and Juppner 2008). (See section 2.11 for a discussion on the physiological and pathophysiological mechanisms involved.) The protein is also believed to be involved in bone and dentin mineralization. Both the whole-body and bone-specific (osteocalcinpromoted inactivation) knockout mouse model of PHEX as well as the spontaneous Hyp mouse model display increased bone production, increased levels of serum FGF23, decreased kidney membrane NPT2 and osteomalacia (Yuan, Takaiwa et al. 2008). Cell studies indicate mechanistic defects both during protein processing in the endoplasmic reticulum and cell membrane (Sabbagh, Boileau et al. 2001) and as abrogated catalytic activity (Sabbagh, Boileau et al. 2003).

There are several mutations associated with PHEX-hypophosphatemia (see the PHEX mutation database: http://www.phexdb.mcgill.ca/) and most of the mutations are located in the region encoding the extracellular domain, but there are also examples of pathological mutations in the 5'UTR (Dixon, Christie et al. 1998) and 3'UTR (Ichikawa, Traxler et al. 2008) of the gene.

There is no clear genotype-phenotype correlation (Holm, Nelson et al. 2001). There is a slight dominance of familiar mutations (showing co-segregation with disease in a pedigree) to *de novo* mutations (sporadic) reported in literature (Holm, Nelson et al. 2001). The penetrance is high, although there are examples of non-penetrance (Gaucher, Walrant-Debray et al. 2009). The expressivity varies (Brame, White et al. 2004).

## **2.2 DMP1**

The *DMP1* gene (dentin matrix acidic phosphoprotein 1; MIM\* 600980) gene consists of 6 exons on chromosome 4q21, and was first implicated in a phosphate balance disorder in 2006 Lorenz-Depiereux, Bastepe et al. 2006). *DMP1* is highly expressed in osteocytes, and is a member of the 'SIBLING' (small integrin binding ligand n-linked glycoprotein) family of non-collagenous extracellular matrix proteins involved in bone mineralization (Huq, Cross

Monogenic Phosphate Balance Disorders 419

The *FGF23* gene (MIM\*605380) on chromosome 12 is composed of 3 exons, and encodes a member of the fibroblast growth factor family. The protein product, FGF23, acts via its receptor FGFR1 (fibroblast growth factor receptor 1, see 2.5), but is also dependent on the co-receptor Klotho (-Klotho) to exert its functions (see below). Furthermore, FGF23 belongs to the FGF19 family where the two other family members, FGF19 and FGF21, (also binding to FGFR1) are dependent on ß -Klotho to exert their functions, illustrating the role of a coreceptor to ensure tissue specificity and function (Kurosu, Choi et al. 2007). FGF23 exerts its physiological effects on the kidney by the downregulation of the C*YP27B1* gene leading to a loss of compensatory increase in 1,25(OH)2vitamin D levels, and by the endocytosis of the type IIa and IIc Na/phosphate (Pi) cotransporters (Npt2a and Npt2c) from the renal

Heterozygous activating mutations in the cleavage site RXXR motif of exon 3 of *FGF23*, leads to stabilization and decreased degradation of the FGF23. The clinical phenotype is

Homozygous inactivating missense mutations in *FGF23* lead to hyperphosphatemic familial tumoral calcinosis, due to decreased renal excretion of phosphate and increased renal αhydroxylation of vitamin D (Benet-Pages, Orlik et al. 2005; Ichikawa, Baujat et al. 2010).

The Fibroblast growth factor receptor 1 gene *FGFR1* (MIM\*136315) located on chromosome 8p11 encodes a protein member of the FGFR (1-4) family, where the members are all receptor tyrosine kinases. FGFR1-3 are implicated in skeletal development, and various mutations in the corresponding genes are responsible for a number of skeletal dysplastic syndromes (Passos-Bueno, Wilcox et al. 1999). There are several subclasses of FGFRs, depending on the number of immunoglobulin-like loops and splicing differences in the third loop. FGFR1C combines with Klotho (*KL*) to become the functional receptor for FGF23

Mutations in the *FGFR1C* lead to constitutive activation of the receptor and subsequent downregulation of the expression of the sodium-phosphate co-transporters NaPiIIa and NaPiIIc, as well as the downregulation of the C*YP27B1* gene leading to a loss of compensatory increase in 1,25(OH)2vitamin D levels (Shimada, Hasegawa et al. 2004).

*Klotho* (*KL*) (MIM\*604824) is located to chromosome 13q12, comprises 5 exons, and encodes the protein Klotho (also known as α-Klotho), which in mice is considered a hormone with anti-aging properties (Kurosu, Yamamoto et al. 2005). *KL* knockout mice will go through a rapid aging process, and have decreased insulin secretion and increased insulin sensitivity (Kuro-o, Matsumura et al. 1997), while overexpression of *KL* leads to a prolonged life span in mice (Kurosu, Yamamoto et al. 2005). In addition, Klotho has been associated with disturbances of phosphate metabolism, as it is an obligate co-receptor for the binding of FGF23 to FGFR1C. In humans, there are two *KL* transcripts; one encoding a membrane bound protein and one encoding a secreted protein. Human *KL* is expressed mainly in the kidney, and the secreted variant seems to dominate (Matsumura, Aizawa et al. 1998). Recent findings from mouse studies suggest that Klotho has endocrine, paracrine and autocrine

autosomal dominant hypophosphatemic rickets (Econs, McEnery et al. 1997; 2000).

**2.4 FGF23** 

**2.5 FGFR1** 

**2.6** *KL* 

proximal tubular brush border membrane.

(Urakawa, Yamazaki et al. 2006).

effects independent of FGF23 (Hu, Shi et al. 2010).

et al. 2005). The *DMP1* knockout model displays rickets and osteomalacia with isolated renal phosphate wasting associated with elevated FGF23 levels and normocalciuria (Feng, Ward et al. 2006). In humans, homozygous or compound heterozygous mutations in *DMP1* leads to hypophosphatemic rickets with elevated FGF23, isolated phosphate wasting, and no evidence of hypercalciuria. The exact relation between DMP1 and FGF23 levels is not known, but in vitro studies have shown that vitamin D increases the expression of both (Farrow, Davis et al. 2009).

There are only a few reported mutations in the literature (Feng, Ward et al. 2006; Lorenz-Depiereux, Bastepe et al. 2006; Farrow, Davis et al. 2009; Koshida, Yamaguchi et al. 2010; Makitie, Pereira et al. 2010; Turan, Aydin et al. 2010), making *DMP1* mutations a rare cause of hypophosphatemic rickets (Gaucher, Walrant-Debray et al. 2009).

Fig. 2. *DMP1* gene structure and the corresponding encoded regions. Based on (Huq, Cross et al. 2005).

## **2.3 GALNT3**

The O-glycosylation of serine and threonine residues on many glycoproteins depend on enzymatic catalyzation of the reaction UDP-GalNAc + polypeptide-(Ser/Thr)-OH to GalNAc-alpha-O-Ser/Thr-polypeptide + UDP. GalNAcT3 is one of 24 members in the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase protein family involved in this process.

GalNAcT3 is encoded by the *GALNT3* gene (MIM \*601756) on chromosome 2q24-q31, which contains 10 exons. GalNAcT3 is thought to protect FGF23 from proteolysis (Kato, Jeanneau et al. 2006) by *O*-glycosylation, and a deactivating mutation in *GALNT3* will thus lead to increased breakdown of FGF23. Mutations in *GALNT3* were the first to be associated with familial tumoral calcinosis (FTC) (Topaz, Shurman et al. 2004), and are also seen in the closely related disease, the hyperphosphatemic hyperostosis syndrome (HHS). These are the only diseases known to be caused by mutations in the family of UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Although the process of O-glycosylation is important in many tissues, mutations in *GALNT3* lead to a very restricted phenotype with hyperphosphatemia, periarticular calcifications and hyperostosis. This is thought to be explained by functional redundancy of this protein family. In addition to the effects on bone and renal phosphate handling caused by altered FGF23 metabolism, mutations in *GALNT3* are also thought to have direct effect in the process of ectopic calcification in extraosseous tissues (Chefetz and Sprecher 2009).

## **2.4 FGF23**

418 Contemporary Aspects of Endocrinology

et al. 2005). The *DMP1* knockout model displays rickets and osteomalacia with isolated renal phosphate wasting associated with elevated FGF23 levels and normocalciuria (Feng, Ward et al. 2006). In humans, homozygous or compound heterozygous mutations in *DMP1* leads to hypophosphatemic rickets with elevated FGF23, isolated phosphate wasting, and no evidence of hypercalciuria. The exact relation between DMP1 and FGF23 levels is not known, but in vitro studies have shown that vitamin D increases the expression of both

There are only a few reported mutations in the literature (Feng, Ward et al. 2006; Lorenz-Depiereux, Bastepe et al. 2006; Farrow, Davis et al. 2009; Koshida, Yamaguchi et al. 2010; Makitie, Pereira et al. 2010; Turan, Aydin et al. 2010), making *DMP1* mutations a rare cause

Fig. 2. *DMP1* gene structure and the corresponding encoded regions. Based on (Huq, Cross

The O-glycosylation of serine and threonine residues on many glycoproteins depend on enzymatic catalyzation of the reaction UDP-GalNAc + polypeptide-(Ser/Thr)-OH to GalNAc-alpha-O-Ser/Thr-polypeptide + UDP. GalNAcT3 is one of 24 members in the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase protein family involved in this

GalNAcT3 is encoded by the *GALNT3* gene (MIM \*601756) on chromosome 2q24-q31, which contains 10 exons. GalNAcT3 is thought to protect FGF23 from proteolysis (Kato, Jeanneau et al. 2006) by *O*-glycosylation, and a deactivating mutation in *GALNT3* will thus lead to increased breakdown of FGF23. Mutations in *GALNT3* were the first to be associated with familial tumoral calcinosis (FTC) (Topaz, Shurman et al. 2004), and are also seen in the closely related disease, the hyperphosphatemic hyperostosis syndrome (HHS). These are the only diseases known to be caused by mutations in the family of UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Although the process of O-glycosylation is important in many tissues, mutations in *GALNT3* lead to a very restricted phenotype with hyperphosphatemia, periarticular calcifications and hyperostosis. This is thought to be explained by functional redundancy of this protein family. In addition to the effects on bone and renal phosphate handling caused by altered FGF23 metabolism, mutations in *GALNT3* are also thought to have direct effect in the process of ectopic calcification in extraosseous

of hypophosphatemic rickets (Gaucher, Walrant-Debray et al. 2009).

(Farrow, Davis et al. 2009).

et al. 2005).

**2.3 GALNT3** 

process.

tissues (Chefetz and Sprecher 2009).

The *FGF23* gene (MIM\*605380) on chromosome 12 is composed of 3 exons, and encodes a member of the fibroblast growth factor family. The protein product, FGF23, acts via its receptor FGFR1 (fibroblast growth factor receptor 1, see 2.5), but is also dependent on the co-receptor Klotho (-Klotho) to exert its functions (see below). Furthermore, FGF23 belongs to the FGF19 family where the two other family members, FGF19 and FGF21, (also binding to FGFR1) are dependent on ß -Klotho to exert their functions, illustrating the role of a coreceptor to ensure tissue specificity and function (Kurosu, Choi et al. 2007). FGF23 exerts its physiological effects on the kidney by the downregulation of the C*YP27B1* gene leading to a loss of compensatory increase in 1,25(OH)2vitamin D levels, and by the endocytosis of the type IIa and IIc Na/phosphate (Pi) cotransporters (Npt2a and Npt2c) from the renal proximal tubular brush border membrane.

Heterozygous activating mutations in the cleavage site RXXR motif of exon 3 of *FGF23*, leads to stabilization and decreased degradation of the FGF23. The clinical phenotype is autosomal dominant hypophosphatemic rickets (Econs, McEnery et al. 1997; 2000).

Homozygous inactivating missense mutations in *FGF23* lead to hyperphosphatemic familial tumoral calcinosis, due to decreased renal excretion of phosphate and increased renal αhydroxylation of vitamin D (Benet-Pages, Orlik et al. 2005; Ichikawa, Baujat et al. 2010).

## **2.5 FGFR1**

The Fibroblast growth factor receptor 1 gene *FGFR1* (MIM\*136315) located on chromosome 8p11 encodes a protein member of the FGFR (1-4) family, where the members are all receptor tyrosine kinases. FGFR1-3 are implicated in skeletal development, and various mutations in the corresponding genes are responsible for a number of skeletal dysplastic syndromes (Passos-Bueno, Wilcox et al. 1999). There are several subclasses of FGFRs, depending on the number of immunoglobulin-like loops and splicing differences in the third loop. FGFR1C combines with Klotho (*KL*) to become the functional receptor for FGF23 (Urakawa, Yamazaki et al. 2006).

Mutations in the *FGFR1C* lead to constitutive activation of the receptor and subsequent downregulation of the expression of the sodium-phosphate co-transporters NaPiIIa and NaPiIIc, as well as the downregulation of the C*YP27B1* gene leading to a loss of compensatory increase in 1,25(OH)2vitamin D levels (Shimada, Hasegawa et al. 2004).

## **2.6** *KL*

*Klotho* (*KL*) (MIM\*604824) is located to chromosome 13q12, comprises 5 exons, and encodes the protein Klotho (also known as α-Klotho), which in mice is considered a hormone with anti-aging properties (Kurosu, Yamamoto et al. 2005). *KL* knockout mice will go through a rapid aging process, and have decreased insulin secretion and increased insulin sensitivity (Kuro-o, Matsumura et al. 1997), while overexpression of *KL* leads to a prolonged life span in mice (Kurosu, Yamamoto et al. 2005). In addition, Klotho has been associated with disturbances of phosphate metabolism, as it is an obligate co-receptor for the binding of FGF23 to FGFR1C. In humans, there are two *KL* transcripts; one encoding a membrane bound protein and one encoding a secreted protein. Human *KL* is expressed mainly in the kidney, and the secreted variant seems to dominate (Matsumura, Aizawa et al. 1998). Recent findings from mouse studies suggest that Klotho has endocrine, paracrine and autocrine effects independent of FGF23 (Hu, Shi et al. 2010).

Monogenic Phosphate Balance Disorders 421

The *ENPP1* (ectonucleotide pyrophosphatase/phosphodiesterase 1) (MIM\*173335) gene on chromosome 6q22-q23 comprises 23 exones and encodes a type II transmembrane glycoprotein ectoenzyme responsible for the generation of inorganic pyrophosphate (PPi). PPi is an inhibitor of hydroxyapatite crystal growth, and also suppress chondrogenesis. In mice, *ENPP1* is expressed in plasma cells, on hepatocytes, renal tubules, salivary duct epithelium, epididymis, capillary endothelium in the brain, and chondrocytes (Harahap and Goding 1988). In man it has been shown that *ENPP1* is expressed in liver, cartilage and bone, and is thought to regulate physiological mineralization processes and pathological

Homozygous mutations in *ENPP1* are known to cause generalized arterial calcifications of infancy (GACI) (Rutsch, Vaingankar et al. 2001; Rutsch, Ruf et al. 2003). Recently, homozygous mutations in *ENPP1* have been shown to cause autosomal recessive hypophosphataemic rickets (Levy-Litan, Hershkovitz et al. 2010). In some families, identical mutations cause GACI in some family members and hypophosphatemic rickets in other family members (Lorenz-Depiereux, Schnabel et al. 2010). Prolonged survival in GACI has been observed in subjects who have simultaneously displayed renal phosphate loss (Rutsch,

Mutations in *ENPP1* have also been associated with susceptibility to insulin resistance and

**2.11 An integrated model for the physiological and pathophysiological mechanisms in** 

Figure 3 shows the integrated physiological and pathophysiological mechanisms in the renal phosphate regulation. The parathyroid-renal axis has been the traditional model explaining how PTH stimulates the renal tubular cells to phosphaturia as a negative feedback loop response to elevated phosphate levels (Figure 3A). In this model PTH acts via its receptor to block the sodium-phosphate co-transporters NaPiIIa and NaPiIIc encoded by the *SLC34A1* and *SLC34A3* genes, respectively. In addition, PTH stimulates the C*YP27B1* gene leading to a compensatory increase in 1,25(OH)2vitamin D levels as a negative feedback loop to reduced serum levels of 1,25(OH)2vitamin D and calcium. There is, however, also a PTH-independent pathway where hormonal substances from bone, phosphatonins, stimulate the renal tubular cells to phosphaturia in a negative feedback response to elevated serum phosphate and 1,25(OH)2vitamin D levels. Recent emerging insight has laid the foundation for this model of a bone-kidney axis (Quarles 2003), where fibroblast growth factor 23 (FGF23) seems to be the central phosphatonin inhibiting phosphate reabsorption and hence inducing phosphaturia (Figure 3B). In contrast to PTH, FGF23 inhibits C*YP27B1* gene leading to an absent compensatory increase in 1,25(OH)2vitamin D levels, recognized by clinicians as inappropriate normal 1,25(OH)2vitamin D levels. In the normal state *PHEX* and *DMP1* gene products seem to inhibit FGF23 production, whereas the *GALNT3* gene product seems to stimulate FGF23 production. By interfering with the bone-kidney axis, increased FGF23 levels seem to play a central role in the pathogenesis of *PHEX*-hypophosphatemia (Jonsson, Zahradnik et al. 2003) (Figure 3C) and potentially also *DMP1*-hypophosphatemia (Lorenz-Depiereux, Bastepe et al. 2006; Turan, Aydin et al. 2010) and *FGF23-* hypophosphatemia (Imel, Hui et al. 2007), but the mechanisms are still poorly known (Strom and Juppner 2008). It is also poorly known how increased FGF23 levels in *FGF23*- hyperphosphatemia and *GALNT3*-

**2.10 ENPP1** 

Boyer et al. 2008).

chondrocalcinosis (Huang, Rosenbach et al. 1994).

obesity (Goldfine, Maddux et al. 2008).

**the renal phosphate regulation** 

Inactivating mutations will lead to familial hyperphosphatemic tumoral calcinosis, similar to the phenotypes seen in *GALNT3*-hyperphosphatemia and *FGF23*-hyperphosphatemia (Ichikawa, Imel et al. 2007).

There is also one report of an activating translocation of the *KL* gene, leading to hypophosphatemic rickets with a phenotype similar to *PHEX*-hypophosphatemia but with additional distinctive dysmorphic features of the head (Brownstein, Adler et al. 2008).

## **2.7 SLC34A1**

The solute carrier 34 (*SLC34*) gene family includes the three genes *SLC34A1*, *SLC34A2* and *SLC34A3*, all encoding sodium/phosphate cotransporters. *SLC34A2* encodes the intestinal NaPi-IIb, and will not be further discussed. *SLC34A1* and *SLC34A3* encode the two renal sodium/phosphate cotransporters, and the latter is described in 2.8 section.

The *SLC34A1* (MIM\*182309) gene is expressed in the renal proximal tubule, and encodes the type IIa Na/Pi cotransporter (NaPi-IIa), which plays a central role in renal phosphate handling in various animal models. The expression of NaPi-IIa in the brush border membrane is regulated at the post translational level, by endocytosis and lysosomal degradation or microtubular recruitment (Tenenhouse 2005). Both PTH and FGF23 lead to increased endocytosis of NAPi-IIa, and thus decreased reabsorption of Pi from filtered urine, whereas hypophosphatemia and 1,25 dihydroxyvitamin D stimulate phosphate reabsorption (Tenenhouse 2005). There also seems to be a directly regulating effect of dietary Pi on Na/Pi cotransport in proximal tubules, and the existence of an intestinal-renal axis for phosphate regulation has been proposed [review: (Biber, Hernando et al. 2009)]. NaPi-IIa double knockout mice have hypophosphatemia, phosphaturia, elevated 1,25 dihydroxyvitamin D with resulting hypercalcemia, hypercalcuria and nephrocalcinosis/nephrolithiasis (Beck, Karaplis et al. 1998). This phenotype resembles hereditary hypophosphatemic rickets with hypercalcuria (HHRH) seen in humans, which interestingly is not caused by mutations in *SLCA34A1*, but rather by mutations in *SLC34A3* (NaPi-IIc) (see 2.8).

In man, a few cases have been described of heterozygous mutations in *SLC34A1*, leading to a syndrome of hypophosphatemia, osteoporosis and nephrolithiasis (Prie, Huart et al. 2002).

## **2.8 SLC34A3**

The human *SLC34A3* (MIM\*609826) gene, consists of 13 exons on chromosome 9q34, and homozygous mutations in this gene lead to hereditary hypophosphatemic rickets with hypercalciuria (HHRH) (Bergwitz, Roslin et al. 2006; Lorenz-Depiereux, Benet-Pages et al. 2006). The phenotype of HHRH resembles that of NaPi-IIa knockout mice, but the patients also display rickets or osteomalacia. In animal models the type IIc Na/Pi cotransporter (NaPi-IIc) has been shown to play a more minor role in proximal tubular phosphate resorption than NaPi-IIa. The opposite might be the case in man (Amatschek, Haller et al. 2010).

## **2.9 SLC9A3R1**

The *SLC9A3R1* (MIM\*604990) gene on chromosome 17 encodes the protein NHERF1 (sodium/hydrogen exchanger regulatory factor 1), which plays a part in maintaining the cytoskeleton in polarized cells with microvilli, such as renal tubular cells. Three different mutations in *SLC9A3R1* have recently been identified in 7 subjects with hypophosphatemia due to phosphaturia, nephrolithiasis and osteoporosis (Karim, Gerard et al. 2008).

## **2.10 ENPP1**

420 Contemporary Aspects of Endocrinology

Inactivating mutations will lead to familial hyperphosphatemic tumoral calcinosis, similar to the phenotypes seen in *GALNT3*-hyperphosphatemia and *FGF23*-hyperphosphatemia

There is also one report of an activating translocation of the *KL* gene, leading to hypophosphatemic rickets with a phenotype similar to *PHEX*-hypophosphatemia but with additional distinctive dysmorphic features of the head (Brownstein, Adler et al. 2008).

The solute carrier 34 (*SLC34*) gene family includes the three genes *SLC34A1*, *SLC34A2* and *SLC34A3*, all encoding sodium/phosphate cotransporters. *SLC34A2* encodes the intestinal NaPi-IIb, and will not be further discussed. *SLC34A1* and *SLC34A3* encode the two renal

The *SLC34A1* (MIM\*182309) gene is expressed in the renal proximal tubule, and encodes the type IIa Na/Pi cotransporter (NaPi-IIa), which plays a central role in renal phosphate handling in various animal models. The expression of NaPi-IIa in the brush border membrane is regulated at the post translational level, by endocytosis and lysosomal degradation or microtubular recruitment (Tenenhouse 2005). Both PTH and FGF23 lead to increased endocytosis of NAPi-IIa, and thus decreased reabsorption of Pi from filtered urine, whereas hypophosphatemia and 1,25 dihydroxyvitamin D stimulate phosphate reabsorption (Tenenhouse 2005). There also seems to be a directly regulating effect of dietary Pi on Na/Pi cotransport in proximal tubules, and the existence of an intestinal-renal axis for phosphate regulation has been proposed [review: (Biber, Hernando et al. 2009)]. NaPi-IIa double knockout mice have hypophosphatemia, phosphaturia, elevated 1,25 dihydroxyvitamin D with resulting hypercalcemia, hypercalcuria and nephrocalcinosis/nephrolithiasis (Beck, Karaplis et al. 1998). This phenotype resembles hereditary hypophosphatemic rickets with hypercalcuria (HHRH) seen in humans, which interestingly is not caused by mutations in *SLCA34A1*, but rather by mutations in *SLC34A3*

In man, a few cases have been described of heterozygous mutations in *SLC34A1*, leading to a syndrome of hypophosphatemia, osteoporosis and nephrolithiasis (Prie, Huart et al. 2002).

The human *SLC34A3* (MIM\*609826) gene, consists of 13 exons on chromosome 9q34, and homozygous mutations in this gene lead to hereditary hypophosphatemic rickets with hypercalciuria (HHRH) (Bergwitz, Roslin et al. 2006; Lorenz-Depiereux, Benet-Pages et al. 2006). The phenotype of HHRH resembles that of NaPi-IIa knockout mice, but the patients also display rickets or osteomalacia. In animal models the type IIc Na/Pi cotransporter (NaPi-IIc) has been shown to play a more minor role in proximal tubular phosphate resorption than

The *SLC9A3R1* (MIM\*604990) gene on chromosome 17 encodes the protein NHERF1 (sodium/hydrogen exchanger regulatory factor 1), which plays a part in maintaining the cytoskeleton in polarized cells with microvilli, such as renal tubular cells. Three different mutations in *SLC9A3R1* have recently been identified in 7 subjects with hypophosphatemia

NaPi-IIa. The opposite might be the case in man (Amatschek, Haller et al. 2010).

due to phosphaturia, nephrolithiasis and osteoporosis (Karim, Gerard et al. 2008).

sodium/phosphate cotransporters, and the latter is described in 2.8 section.

(Ichikawa, Imel et al. 2007).

**2.7 SLC34A1** 

(NaPi-IIc) (see 2.8).

**2.8 SLC34A3** 

**2.9 SLC9A3R1** 

The *ENPP1* (ectonucleotide pyrophosphatase/phosphodiesterase 1) (MIM\*173335) gene on chromosome 6q22-q23 comprises 23 exones and encodes a type II transmembrane glycoprotein ectoenzyme responsible for the generation of inorganic pyrophosphate (PPi). PPi is an inhibitor of hydroxyapatite crystal growth, and also suppress chondrogenesis. In mice, *ENPP1* is expressed in plasma cells, on hepatocytes, renal tubules, salivary duct epithelium, epididymis, capillary endothelium in the brain, and chondrocytes (Harahap and Goding 1988). In man it has been shown that *ENPP1* is expressed in liver, cartilage and bone, and is thought to regulate physiological mineralization processes and pathological chondrocalcinosis (Huang, Rosenbach et al. 1994).

Homozygous mutations in *ENPP1* are known to cause generalized arterial calcifications of infancy (GACI) (Rutsch, Vaingankar et al. 2001; Rutsch, Ruf et al. 2003). Recently, homozygous mutations in *ENPP1* have been shown to cause autosomal recessive hypophosphataemic rickets (Levy-Litan, Hershkovitz et al. 2010). In some families, identical mutations cause GACI in some family members and hypophosphatemic rickets in other family members (Lorenz-Depiereux, Schnabel et al. 2010). Prolonged survival in GACI has been observed in subjects who have simultaneously displayed renal phosphate loss (Rutsch, Boyer et al. 2008).

Mutations in *ENPP1* have also been associated with susceptibility to insulin resistance and obesity (Goldfine, Maddux et al. 2008).

## **2.11 An integrated model for the physiological and pathophysiological mechanisms in the renal phosphate regulation**

Figure 3 shows the integrated physiological and pathophysiological mechanisms in the renal phosphate regulation. The parathyroid-renal axis has been the traditional model explaining how PTH stimulates the renal tubular cells to phosphaturia as a negative feedback loop response to elevated phosphate levels (Figure 3A). In this model PTH acts via its receptor to block the sodium-phosphate co-transporters NaPiIIa and NaPiIIc encoded by the *SLC34A1* and *SLC34A3* genes, respectively. In addition, PTH stimulates the C*YP27B1* gene leading to a compensatory increase in 1,25(OH)2vitamin D levels as a negative feedback loop to reduced serum levels of 1,25(OH)2vitamin D and calcium. There is, however, also a PTH-independent pathway where hormonal substances from bone, phosphatonins, stimulate the renal tubular cells to phosphaturia in a negative feedback response to elevated serum phosphate and 1,25(OH)2vitamin D levels. Recent emerging insight has laid the foundation for this model of a bone-kidney axis (Quarles 2003), where fibroblast growth factor 23 (FGF23) seems to be the central phosphatonin inhibiting phosphate reabsorption and hence inducing phosphaturia (Figure 3B). In contrast to PTH, FGF23 inhibits C*YP27B1* gene leading to an absent compensatory increase in 1,25(OH)2vitamin D levels, recognized by clinicians as inappropriate normal 1,25(OH)2vitamin D levels. In the normal state *PHEX* and *DMP1* gene products seem to inhibit FGF23 production, whereas the *GALNT3* gene product seems to stimulate FGF23 production. By interfering with the bone-kidney axis, increased FGF23 levels seem to play a central role in the pathogenesis of *PHEX*-hypophosphatemia (Jonsson, Zahradnik et al. 2003) (Figure 3C) and potentially also *DMP1*-hypophosphatemia (Lorenz-Depiereux, Bastepe et al. 2006; Turan, Aydin et al. 2010) and *FGF23-* hypophosphatemia (Imel, Hui et al. 2007), but the mechanisms are still poorly known (Strom and Juppner 2008). It is also poorly known how increased FGF23 levels in *FGF23*- hyperphosphatemia and *GALNT3*-

Monogenic Phosphate Balance Disorders 423

whereas PTH possibly stimulates FGF23 (Figure 3D). Mutations in genes encoding the sodium-phosphate co-transporters such as *SLC34A1*, *SLC34A3* and *SLC9A3R1* lead to increased phosphaturia but since the 1,25(OH)2vitamin D activation is unaffected, there is a normal compensatory increase in 1,25(OH)2vitamin D levels. Whether gene mutations lead to hypophosphatemia or hyperphosphatemia is dependent on the location of the gene product in the pathways outlined above and whether the mutation is activating or

The diagnosis of monogenic hypo- or hyperphosphatemia requires the demonstration of affected phosphate balance in patients in which acquired causes of phosphate disturbance have been excluded. A family history of rickets, kidney stones, soft tissue calcification, bone deformities or recurrent fractures, as well as an indication of monogenic inheritance pattern is usually found, unless the patient seems to represent a sporadic case. In the case of hypophosphatemia, there is typically low plasma phosphate, low renal tubular reabsorption of phosphate (% TRP) and tubular threshold maximum for phosphate for glomerular filtration rate (TmP/GFR), raised alkaline phosphatase, normal PTH, and inappropriate and normal 25(OH) and 1,25(OH)2 vitamin D levels. Moreover, the urinary calcium excretion is normal, whereas X-ray changes may demonstrate rickets or osteomalacia. FGF23 levels are typically high, either due to overproduction or under-catabolism, and in children with rickets, the combined evaluation of FGF23 and PTH leads differential diagnosis in the direction of impaired phosphate homeostasis (high FGF23 and normal PTH) or altered metabolism of vitamin D, calcium or magnesium (low FGF23 and high PTH) (Alon 2010). In the case of hyperphosphatemia, there is usually high plasma phosphate, an inappropriate normal % TRP and TmP/GFR, a low or normal PTH and normal renal function. In some cases, the clinical picture and inheritance pattern will suggest a specific genetic diagnosis, and, in addition, the blood FGF23 levels and hypercalciuria may differentiate between different genetic disorders of phosphate balance, although the clinical role of blood FGF23

*PHEX*-hypophosphatemia (X-linked dominant) is usually a progressive disorder with a typical onset at the age when the child starts to walk. The most common clinical manifestations include genu varus, radiological rickets, short stature, bone pain, dental abscesses and calcification of tendons, ligaments and joint capsules with boys being more severely affected than girls and a wide variation between families (Econs, Samsa et al. 1994; Carpenter 1997; Bastepe and Juppner 2008). Some patients may even have craniosynosteosis and spinal stenosis. Many patients suffer from long lasting dental problems, particularly tooth decay and recurrent spontaneous dental abscesses that occur in the absence of a history of trauma or dental decay. Histological findings include high pulp horns, globular dentin, and defects of dentin and enamel. The primary teeth are most commonly affected, as the mineralization process starts in utero. Permanent teeth develop after birth, and adequate treatment improve development in some cases (Batra, Tejani et al. 2006). In children with rickets, a low serum phosphorus level, combined with high serum alkaline phosphatase and normal serum calcium is typical (Carpenter 1997). Urinary leakage of phosphate is demonstrated by low % TRP and TmP/GFR, whereas urinary calcium is normal. The PTH levels are usually normal or slightly

inactivating the affected gene.

**3. Diagnostic considerations** 

levels is at present not fully elucidated.

**3.1 PHEX** 

hyperphosphatemia explain the opposite condition of hyperphosphatemia (Topaz, Shurman et al. 2004; Benet-Pages, Orlik et al. 2005). A current model postulates that mutations in *PHEX* lead to increased FGF23 production by cancelled *PHEX*-mediated inhibition of FGF23 production (Figure 3 C).

Fig. 3. Physiological and pathophysiological conditions in the phosphate regulation. For sake of clarity, only gene names are depicted and not the corresponding gene products. Adapted from (Bastepe and Juppner 2008; Strom and Juppner 2008).

Both the parathyroid-renal axis and the bone-kidney axis seem to be negative feedback loops where increased serum phosphate levels compared to a biological set value leads to phosphaturia. These two axes are different with respect to 1,25(OH)2vitamin D: Whereas low 1,25(OH)2vitamin D levels stimulating 1,25(OH)2vitamin D activation is the major regulation in the parathyroid-renal axis, high 1,25(OH)2vitamin D levels inhibiting 1,25(OH)2vitamin D activation is the major regulation in the bone-kidney axis. Recent work also points to interactions between these feedback loops where FGF23 inhibits PTH,

hyperphosphatemia explain the opposite condition of hyperphosphatemia (Topaz, Shurman et al. 2004; Benet-Pages, Orlik et al. 2005). A current model postulates that mutations in *PHEX* lead to increased FGF23 production by cancelled *PHEX*-mediated inhibition of FGF23

Fig. 3. Physiological and pathophysiological conditions in the phosphate regulation. For sake of clarity, only gene names are depicted and not the corresponding gene products.

Both the parathyroid-renal axis and the bone-kidney axis seem to be negative feedback loops where increased serum phosphate levels compared to a biological set value leads to phosphaturia. These two axes are different with respect to 1,25(OH)2vitamin D: Whereas low 1,25(OH)2vitamin D levels stimulating 1,25(OH)2vitamin D activation is the major regulation in the parathyroid-renal axis, high 1,25(OH)2vitamin D levels inhibiting 1,25(OH)2vitamin D activation is the major regulation in the bone-kidney axis. Recent work also points to interactions between these feedback loops where FGF23 inhibits PTH,

Adapted from (Bastepe and Juppner 2008; Strom and Juppner 2008).

production (Figure 3 C).

whereas PTH possibly stimulates FGF23 (Figure 3D). Mutations in genes encoding the sodium-phosphate co-transporters such as *SLC34A1*, *SLC34A3* and *SLC9A3R1* lead to increased phosphaturia but since the 1,25(OH)2vitamin D activation is unaffected, there is a normal compensatory increase in 1,25(OH)2vitamin D levels. Whether gene mutations lead to hypophosphatemia or hyperphosphatemia is dependent on the location of the gene product in the pathways outlined above and whether the mutation is activating or inactivating the affected gene.

## **3. Diagnostic considerations**

The diagnosis of monogenic hypo- or hyperphosphatemia requires the demonstration of affected phosphate balance in patients in which acquired causes of phosphate disturbance have been excluded. A family history of rickets, kidney stones, soft tissue calcification, bone deformities or recurrent fractures, as well as an indication of monogenic inheritance pattern is usually found, unless the patient seems to represent a sporadic case. In the case of hypophosphatemia, there is typically low plasma phosphate, low renal tubular reabsorption of phosphate (% TRP) and tubular threshold maximum for phosphate for glomerular filtration rate (TmP/GFR), raised alkaline phosphatase, normal PTH, and inappropriate and normal 25(OH) and 1,25(OH)2 vitamin D levels. Moreover, the urinary calcium excretion is normal, whereas X-ray changes may demonstrate rickets or osteomalacia. FGF23 levels are typically high, either due to overproduction or under-catabolism, and in children with rickets, the combined evaluation of FGF23 and PTH leads differential diagnosis in the direction of impaired phosphate homeostasis (high FGF23 and normal PTH) or altered metabolism of vitamin D, calcium or magnesium (low FGF23 and high PTH) (Alon 2010). In the case of hyperphosphatemia, there is usually high plasma phosphate, an inappropriate normal % TRP and TmP/GFR, a low or normal PTH and normal renal function. In some cases, the clinical picture and inheritance pattern will suggest a specific genetic diagnosis, and, in addition, the blood FGF23 levels and hypercalciuria may differentiate between different genetic disorders of phosphate balance, although the clinical role of blood FGF23 levels is at present not fully elucidated.

## **3.1 PHEX**

*PHEX*-hypophosphatemia (X-linked dominant) is usually a progressive disorder with a typical onset at the age when the child starts to walk. The most common clinical manifestations include genu varus, radiological rickets, short stature, bone pain, dental abscesses and calcification of tendons, ligaments and joint capsules with boys being more severely affected than girls and a wide variation between families (Econs, Samsa et al. 1994; Carpenter 1997; Bastepe and Juppner 2008). Some patients may even have craniosynosteosis and spinal stenosis. Many patients suffer from long lasting dental problems, particularly tooth decay and recurrent spontaneous dental abscesses that occur in the absence of a history of trauma or dental decay. Histological findings include high pulp horns, globular dentin, and defects of dentin and enamel. The primary teeth are most commonly affected, as the mineralization process starts in utero. Permanent teeth develop after birth, and adequate treatment improve development in some cases (Batra, Tejani et al. 2006). In children with rickets, a low serum phosphorus level, combined with high serum alkaline phosphatase and normal serum calcium is typical (Carpenter 1997). Urinary leakage of phosphate is demonstrated by low % TRP and TmP/GFR, whereas urinary calcium is normal. The PTH levels are usually normal or slightly

Monogenic Phosphate Balance Disorders 425

abnormalities leading to dwarfism and facial abnormalities similar to achondroplasia. There is often failure of tooth eruption, and mandibular malformations. Patients may also have

To date only one case of *KL*-hypophosphatemia has been described in the literature (Brownstein, Adler et al. 2008). A 1-year old girl suffered from poor linear growth and increasing head size. She had clinical and radiological signs of rickets, hypophosphatemia, renal phosphate wasting and elevated levels of parathyroid hormone and alkaline phosphatase. A balanced translocation between chromosomes 9 and 13 was detected (t(9,13)(q21.13;q13.1)). This translocation had led to upregulation of *KL*transcription. After a few years she demonstrated dysmorphic features of the face, and also an Arnold-Chiari 1 malformation (Brownstein, Adler et al. 2008). Dental affection has

*KL*-hyperphosphatemia has also been described in only one report (Ichikawa, Imel et al. 2007). A 13 year old girl presented with severe calcifications in soft tissues and in the vasculature, including the dura and the carotid arteries. In addition to hyperphosphatemia and hypercalcemia, she presented with hyperparathyroidism and elevated levels of FGF23. She had no signs of premature aging, which is seen in *KL* knockout mice. Dental affection

In *SLC34A1*- and *SLC34A3*-hypophosphatemia, there is hypophosphatemic rickets with hypercalciuria without other tubular defects (Tieder, Modai et al. 1985). The inheritance pattern is autosomal recessive. Since there is normal renal activation of vitamin D (in contrast to *PHEX*-hypophosphatemia and *DMP1*-hypophosphatemia), hypophosphatemia leads to a normal compensatory increase in 1,25(OH)2vitamin D levels and increased

A total of 7 cases of SLC9A3R1-hypophosphatemia (hypophosphatemia, nephrolithiasis/osteoporosis) have been described to date (Karim, Gerard et al. 2008). All patients were adults, and had either nephrolithiasis and/or bone demineralization combined with hypophosphatemia and hyperphosphaturia. 1,25 (OH)2 vitamin D levels were either elevated or in the upper normal range. Dental affection has not been described.

*ENPP1*-hypophosphatemia (autosomal recessive) has a variable age at onset and a variable phenotype including Generalized Arterial Calcification of Infancy (GACI). Also there seems to be phenotypic variation within the same family among affected subjects carrying the same mutation. Whereas the classic presentation is that of severe arterial calcification leading to death in infancy, some patients have renal phosphate wasting and hypophosphatemia. This phosphate loss seems to attenuate the tendency of arterial calcifications, and is associated with prolonged survival (Lorenz-Depiereux, Schnabel et al.

various degrees of craniosynostosis (White, Cabral et al. 2005).

**3.6** *KL* 

not been described.

has not been described.

**3.8 SLC9A3R1** 

**3.9 ENPP1** 

2010).

**3.7 SLC34A1 and SLC34A3** 

absorption of calcium and phosphate from the gut.

elevated, even before the onset of therapy. The 25(OH)vitamin D is normal, and there is no compensatory increase in 1,25(OH)2vitamin D levels due to defective renal activation of vitamin D, and, hence, no hypercalciuria. The FGF23 levels are increased (Jonsson, Zahradnik et al. 2003), and since the lower extremities are more severely affected than the other parts of the skeleton, radiographs of the knees and ankles will demonstrate the extent of rickets. The diagnosis of *PHEX*-hypophosphatemia is confirmed by genetic analysis.

## **3.2 DMP1**

*DMP1*-hypophosphatemia (autosomal recessive) is usually a progressive disorder with a typical onset at the age when the child starts to walk. The condition is rarer than *PHEX*hypophosphatemia, but is phenotypically quite similar to *PHEX*- and *FGF23* hypophosphatemie. There is no compensatory increase in 1,25(OH)2vitamin D levels due to defective renal activation of vitamin D and hence no hypercalciuria. The circulation levels of FGF23 are increased (Feng, Ward et al. 2006). The degree of skeletal abnormalities varies between families (Makitie, Pereira et al. 2010). Some patients also have dental affection, with hypomineralization, enlarged pulp chambers, and decrease in the dentin and enamel layers, which can cause dental abscesses and loss of teeth (Koshida, Yamaguchi et al. 2010; Turan, Aydin et al. 2010).

## **3.3 GALNT3**

*GALNT3*-hyperphosphatemia is the result of biallelic mutation in the *GALNT3*-gene, and leads to typical tumoral calcinosis (TC) (Topaz, Shurman et al. 2004) or hyperostosishyperphosphatemia syndrome (HHS). There are several mutations in the *GALNT3* gene, and the same mutation can lead to TC in some patients and HHS in other (Ichikawa, Baujat et al. 2010). TC is characterized by ectopic calcifications in soft tissues and around large joints, recognized clinically as palpable masses and/or on radiography. Calcifications may also be found in the retina, in blood vessels, as testicular microlithiasis, and there might be dental abnormalities. HHC is characterized by hyperostosis of long bones, seen radiographically as cortical hyperostosis, diaphysitis and periosteal apposition. The biochemical findings in TC and HHC are similar, with elevated serum phosphate levels, increased or normal 1,25(OH)2vitamin D levels. The levels of serum calcium and parathyroid hormone are normal. Some authors suggest that TC and HHS are clinical variants of the same disease (Ichikawa, Baujat et al. 2010).

## **3.4 FGF23**

*FGF23*-hypophosphatemia (autosomal dominant) shows a variable age at onset of disease.

The expression of disease varies, and some children may have fracture tendency without skeletal deformities, whereas other children may have only temporary renal phosphate loss (Econs and McEnery 1997). Tooth abscesses and loss also occurs (Imel, Hui et al. 2007).

*FGF23*-hyperphosphatemia (autosomal recessive) shows typical tumoral calcinosis , or more rarely the hyperostosis-hyperphosphatemia syndrome (Benet-Pages, Orlik et al. 2005).

## **3.5 FGFR1**

*FGFR1R*-hypophosphatemia is characterized by osteoglophonic dysplasia and can be associated with hypophosphatemia (Farrow, Davis et al. 2006). Clinical features are skeletal abnormalities leading to dwarfism and facial abnormalities similar to achondroplasia. There is often failure of tooth eruption, and mandibular malformations. Patients may also have various degrees of craniosynostosis (White, Cabral et al. 2005).

## **3.6** *KL*

424 Contemporary Aspects of Endocrinology

elevated, even before the onset of therapy. The 25(OH)vitamin D is normal, and there is no compensatory increase in 1,25(OH)2vitamin D levels due to defective renal activation of vitamin D, and, hence, no hypercalciuria. The FGF23 levels are increased (Jonsson, Zahradnik et al. 2003), and since the lower extremities are more severely affected than the other parts of the skeleton, radiographs of the knees and ankles will demonstrate the extent of rickets. The

*DMP1*-hypophosphatemia (autosomal recessive) is usually a progressive disorder with a typical onset at the age when the child starts to walk. The condition is rarer than *PHEX*hypophosphatemia, but is phenotypically quite similar to *PHEX*- and *FGF23* hypophosphatemie. There is no compensatory increase in 1,25(OH)2vitamin D levels due to defective renal activation of vitamin D and hence no hypercalciuria. The circulation levels of FGF23 are increased (Feng, Ward et al. 2006). The degree of skeletal abnormalities varies between families (Makitie, Pereira et al. 2010). Some patients also have dental affection, with hypomineralization, enlarged pulp chambers, and decrease in the dentin and enamel layers, which can cause dental abscesses and loss of teeth (Koshida, Yamaguchi et al. 2010; Turan,

*GALNT3*-hyperphosphatemia is the result of biallelic mutation in the *GALNT3*-gene, and leads to typical tumoral calcinosis (TC) (Topaz, Shurman et al. 2004) or hyperostosishyperphosphatemia syndrome (HHS). There are several mutations in the *GALNT3* gene, and the same mutation can lead to TC in some patients and HHS in other (Ichikawa, Baujat et al. 2010). TC is characterized by ectopic calcifications in soft tissues and around large joints, recognized clinically as palpable masses and/or on radiography. Calcifications may also be found in the retina, in blood vessels, as testicular microlithiasis, and there might be dental abnormalities. HHC is characterized by hyperostosis of long bones, seen radiographically as cortical hyperostosis, diaphysitis and periosteal apposition. The biochemical findings in TC and HHC are similar, with elevated serum phosphate levels, increased or normal 1,25(OH)2vitamin D levels. The levels of serum calcium and parathyroid hormone are normal. Some authors suggest that TC and HHS are clinical

*FGF23*-hypophosphatemia (autosomal dominant) shows a variable age at onset of

The expression of disease varies, and some children may have fracture tendency without skeletal deformities, whereas other children may have only temporary renal phosphate loss (Econs and McEnery 1997). Tooth abscesses and loss also occurs (Imel, Hui et al. 2007). *FGF23*-hyperphosphatemia (autosomal recessive) shows typical tumoral calcinosis , or more rarely the hyperostosis-hyperphosphatemia syndrome (Benet-Pages, Orlik et al. 2005).

*FGFR1R*-hypophosphatemia is characterized by osteoglophonic dysplasia and can be associated with hypophosphatemia (Farrow, Davis et al. 2006). Clinical features are skeletal

diagnosis of *PHEX*-hypophosphatemia is confirmed by genetic analysis.

variants of the same disease (Ichikawa, Baujat et al. 2010).

**3.2 DMP1** 

Aydin et al. 2010).

**3.3 GALNT3** 

**3.4 FGF23** 

**3.5 FGFR1** 

disease.

To date only one case of *KL*-hypophosphatemia has been described in the literature (Brownstein, Adler et al. 2008). A 1-year old girl suffered from poor linear growth and increasing head size. She had clinical and radiological signs of rickets, hypophosphatemia, renal phosphate wasting and elevated levels of parathyroid hormone and alkaline phosphatase. A balanced translocation between chromosomes 9 and 13 was detected (t(9,13)(q21.13;q13.1)). This translocation had led to upregulation of *KL*transcription. After a few years she demonstrated dysmorphic features of the face, and also an Arnold-Chiari 1 malformation (Brownstein, Adler et al. 2008). Dental affection has not been described.

*KL*-hyperphosphatemia has also been described in only one report (Ichikawa, Imel et al. 2007). A 13 year old girl presented with severe calcifications in soft tissues and in the vasculature, including the dura and the carotid arteries. In addition to hyperphosphatemia and hypercalcemia, she presented with hyperparathyroidism and elevated levels of FGF23. She had no signs of premature aging, which is seen in *KL* knockout mice. Dental affection has not been described.

## **3.7 SLC34A1 and SLC34A3**

In *SLC34A1*- and *SLC34A3*-hypophosphatemia, there is hypophosphatemic rickets with hypercalciuria without other tubular defects (Tieder, Modai et al. 1985). The inheritance pattern is autosomal recessive. Since there is normal renal activation of vitamin D (in contrast to *PHEX*-hypophosphatemia and *DMP1*-hypophosphatemia), hypophosphatemia leads to a normal compensatory increase in 1,25(OH)2vitamin D levels and increased absorption of calcium and phosphate from the gut.

## **3.8 SLC9A3R1**

A total of 7 cases of SLC9A3R1-hypophosphatemia (hypophosphatemia, nephrolithiasis/osteoporosis) have been described to date (Karim, Gerard et al. 2008). All patients were adults, and had either nephrolithiasis and/or bone demineralization combined with hypophosphatemia and hyperphosphaturia. 1,25 (OH)2 vitamin D levels were either elevated or in the upper normal range. Dental affection has not been described.

## **3.9 ENPP1**

*ENPP1*-hypophosphatemia (autosomal recessive) has a variable age at onset and a variable phenotype including Generalized Arterial Calcification of Infancy (GACI). Also there seems to be phenotypic variation within the same family among affected subjects carrying the same mutation. Whereas the classic presentation is that of severe arterial calcification leading to death in infancy, some patients have renal phosphate wasting and hypophosphatemia. This phosphate loss seems to attenuate the tendency of arterial calcifications, and is associated with prolonged survival (Lorenz-Depiereux, Schnabel et al. 2010).

Monogenic Phosphate Balance Disorders 427

Fig. 4. An example of ambiguous inheritance pattern. Note that in this case, the pattern is compatible both with an X-linked disorder (i.e. *PHEX*-hypophosphatemia) and an AR

Monogenic phosphate balance disorders warrant genetic counseling, because of the known inheritance pattern and the high penetrance. This is also the case for novel gene variants where it is necessary to establish evidence for causality based on co-segregation studies and prediction tools (such as Polyphen http://genetics.bwh.harvard.edu/pph/)). Predictive genetic testing is less straightforward, and the legal regulations vary in different countries. Communicating genetic information can be difficult and it is important to take into account how well the individual understands both genetics in general and the disorder itself and the consequences of potentially diagnosing other family subjects. The basic fact that there is a 25% or 50% probability for a child to carry the family's mutation should be conveyed to the parents. In addition, the probability for the development of the disorder in the presence of a mutation (i.e. the penetrance) is not always 100%. The variable and in some cases unpredictable age of onset of some of the disorders should also be discussed. By increasing knowledge of the clinical spectrum of mutations, novel expected manifestations need to be discussed with the patient. A system for follow-up is required for children without a phenotype but with affected family members and where parents still request that their child be tested. This follow-up may include periodic testing for hypophosphatemia, with a

We have established a national database of patients with hypophospahtemic bone disorder in order to study phenotype-genotype correlation in this disease and to be able to explore novel pathophysiologic pathways based on insight obtained from studies of families with no previously known genetic cause of monogenic hypophosphatemic bone disorder. We believe that a new classification of disease based on genetic etiology instead of clinical criteria may facilitate the finding of new phenotypes since it will facilitate the study of unobserved phenotypes, both in the patients and in their presumably unaffected relatives carrying mutations. In addition, it is possible that emerging new treatment options may vary based on

disorder (i.e. *DMP1*-hypophosphatemia).

frequency dependent on age and the suspected condition.

**5. Research perspectives** 

## **4. Management principles**

## **4.1 Hypophosphatemia**

Hypophosphatemic rickets is in childhood usually treated with elementary phosphorus at doses preferentially between 30 and 60 (100) mg/kg bodyweight and 24 hours, usually divided by 4-6 doses, whereas the deficient 1,25(OH)2vitamin D production is treated with active vitamin D, e.g. alphacalcidol or calcitriol in doses of 20 to 70 ng/kg bodyweight and 24 hours, usually divided by 2 doses. It should, however, be emphasized that the dosage ranges for both phosphate and active vitamin D are wide, dependent on the severity of the disease, the compliance and the occurrence of complications. In *SLC34A1*- and *SLC34A3* hypophosphatemia, activation of vitamin D is normal, and, consequently, there is no need for treatment with vitamin D.

It is important to adjust the drug doses individually and bear in mind that insufficient doses of elementary phosphorus and vitamin D may fail to prevent or correct skeletal deformities (rickets, osteomalacia) and can lead to growth retardation. On the other hand, excessive doses may lead to nephrocalcinosis (high phosphate doses), as well as hypercalciuria and hypercalcemia (high vitamin D levels). Secondary (and even tertiary) hyperparathyroidism is seen in patients with insufficient doses of vitamin D or excessive doses of phosphorus. We recommend aiming at normal levels of PTH, which in severe cases may be obtained by adding the calcimimetic drug cinacalcet to the treatment (Raeder, Bjerknes et al. 2008).

Close monitoring is necessary to balance the effects of phosphorus supplement and active vitamin D. Growth, serum calcium, phosphorus, alkaline phosphatase, PTH, as well as urinary calcium/creatinine ratio should be determined every 3-6 months, and X-rays of ankles, knees and wrist should be taken yearly. Renal ultrasound should be obtained yearly to assess nephrocalcinosis.

Supplementary treatment with growth hormone is currently not recommended for the growth retardation caused by hypophosphatemia (Huiming and Chaomin 2005), but may be warranted in selected cases. Corrective osteotomies are seldom necessary in childhood, and it should always be deferred until the rickets has healed. Future therapeutic possibilities may include direct targeting of blood FGF23 levels.

## **4.2 Hyperphosphatemia**

Patients with hyperphosphatemia due to monogenic phosphate balance disorders, i.e. *GALNT3*-hyperphosphatemia, *FGF23*-hyperphosphatemia and *KL*-hyperphosphatemia, develop ectopic and vascular calcifications. Combined use of intestinal phosphate binders and the carbonic anhydrase inhibitor acetazolamide has been reported to lower serum phosphorus levels and reduce tumoral masses in some patients (Garringer, Fisher et al. 2006). However, other reports suggest that neither medical nor surgical treatment seems to be effective in controlling ectopic calcifications in these conditions (Carmichael, Bynum et al. 2009). Future therapeutic possibilities may include direct targeting of blood FGF23 levels.

## **4.3 Genetic diagnostics and predictive testing**

Identification of a specific mutation has important therapeutic and prognostic implications and tailored follow-up as outlined above. Distinction between *PHEX*-hypophosphatemia and *DMP1*-hypophosphatemia can be done clinically based on the inheritance pattern, but in some cases there is an ambiguous inheritance pattern (Figure 4) and a genetic test will resolve this ambiguity.

Hypophosphatemic rickets is in childhood usually treated with elementary phosphorus at doses preferentially between 30 and 60 (100) mg/kg bodyweight and 24 hours, usually divided by 4-6 doses, whereas the deficient 1,25(OH)2vitamin D production is treated with active vitamin D, e.g. alphacalcidol or calcitriol in doses of 20 to 70 ng/kg bodyweight and 24 hours, usually divided by 2 doses. It should, however, be emphasized that the dosage ranges for both phosphate and active vitamin D are wide, dependent on the severity of the disease, the compliance and the occurrence of complications. In *SLC34A1*- and *SLC34A3* hypophosphatemia, activation of vitamin D is normal, and, consequently, there is no need

It is important to adjust the drug doses individually and bear in mind that insufficient doses of elementary phosphorus and vitamin D may fail to prevent or correct skeletal deformities (rickets, osteomalacia) and can lead to growth retardation. On the other hand, excessive doses may lead to nephrocalcinosis (high phosphate doses), as well as hypercalciuria and hypercalcemia (high vitamin D levels). Secondary (and even tertiary) hyperparathyroidism is seen in patients with insufficient doses of vitamin D or excessive doses of phosphorus. We recommend aiming at normal levels of PTH, which in severe cases may be obtained by adding the calcimimetic drug cinacalcet to the treatment (Raeder, Bjerknes et al. 2008). Close monitoring is necessary to balance the effects of phosphorus supplement and active vitamin D. Growth, serum calcium, phosphorus, alkaline phosphatase, PTH, as well as urinary calcium/creatinine ratio should be determined every 3-6 months, and X-rays of ankles, knees and wrist should be taken yearly. Renal ultrasound should be obtained yearly

Supplementary treatment with growth hormone is currently not recommended for the growth retardation caused by hypophosphatemia (Huiming and Chaomin 2005), but may be warranted in selected cases. Corrective osteotomies are seldom necessary in childhood, and it should always be deferred until the rickets has healed. Future therapeutic possibilities

Patients with hyperphosphatemia due to monogenic phosphate balance disorders, i.e. *GALNT3*-hyperphosphatemia, *FGF23*-hyperphosphatemia and *KL*-hyperphosphatemia, develop ectopic and vascular calcifications. Combined use of intestinal phosphate binders and the carbonic anhydrase inhibitor acetazolamide has been reported to lower serum phosphorus levels and reduce tumoral masses in some patients (Garringer, Fisher et al. 2006). However, other reports suggest that neither medical nor surgical treatment seems to be effective in controlling ectopic calcifications in these conditions (Carmichael, Bynum et al. 2009). Future therapeutic possibilities may include direct targeting of blood FGF23 levels.

Identification of a specific mutation has important therapeutic and prognostic implications and tailored follow-up as outlined above. Distinction between *PHEX*-hypophosphatemia and *DMP1*-hypophosphatemia can be done clinically based on the inheritance pattern, but in some cases there is an ambiguous inheritance pattern (Figure 4) and a genetic test will

**4. Management principles** 

for treatment with vitamin D.

to assess nephrocalcinosis.

**4.2 Hyperphosphatemia** 

resolve this ambiguity.

may include direct targeting of blood FGF23 levels.

**4.3 Genetic diagnostics and predictive testing** 

**4.1 Hypophosphatemia** 

Fig. 4. An example of ambiguous inheritance pattern. Note that in this case, the pattern is compatible both with an X-linked disorder (i.e. *PHEX*-hypophosphatemia) and an AR disorder (i.e. *DMP1*-hypophosphatemia).

Monogenic phosphate balance disorders warrant genetic counseling, because of the known inheritance pattern and the high penetrance. This is also the case for novel gene variants where it is necessary to establish evidence for causality based on co-segregation studies and prediction tools (such as Polyphen http://genetics.bwh.harvard.edu/pph/)). Predictive genetic testing is less straightforward, and the legal regulations vary in different countries. Communicating genetic information can be difficult and it is important to take into account how well the individual understands both genetics in general and the disorder itself and the consequences of potentially diagnosing other family subjects. The basic fact that there is a 25% or 50% probability for a child to carry the family's mutation should be conveyed to the parents. In addition, the probability for the development of the disorder in the presence of a mutation (i.e. the penetrance) is not always 100%. The variable and in some cases unpredictable age of onset of some of the disorders should also be discussed. By increasing knowledge of the clinical spectrum of mutations, novel expected manifestations need to be discussed with the patient. A system for follow-up is required for children without a phenotype but with affected family members and where parents still request that their child be tested. This follow-up may include periodic testing for hypophosphatemia, with a frequency dependent on age and the suspected condition.

## **5. Research perspectives**

We have established a national database of patients with hypophospahtemic bone disorder in order to study phenotype-genotype correlation in this disease and to be able to explore novel pathophysiologic pathways based on insight obtained from studies of families with no previously known genetic cause of monogenic hypophosphatemic bone disorder. We believe that a new classification of disease based on genetic etiology instead of clinical criteria may facilitate the finding of new phenotypes since it will facilitate the study of unobserved phenotypes, both in the patients and in their presumably unaffected relatives carrying mutations. In addition, it is possible that emerging new treatment options may vary based on

Monogenic Phosphate Balance Disorders 429

Benet-Pages, A., P. Orlik, et al. (2005). "An FGF23 missense mutation causes familial tumoral calcinosis with hyperphosphatemia." Hum Mol Genet 14(3): 385-390. Bergwitz, C., N. M. Roslin, et al. (2006). "SLC34A3 mutations in patients with hereditary

Biber, J., N. Hernando, et al. (2009). "Regulation of phosphate transport in proximal tubules."

Brame, L. A., K. E. White, et al. (2004). "Renal phosphate wasting disorders: clinical features

Brownstein, C. A., F. Adler, et al. (2008). "A translocation causing increased alpha-klotho

Carmichael, K. D., J. A. Bynum, et al. (2009). "Familial tumoral calcinosis: a forty-year

Carpenter, T. O. (1997). "New perspectives on the biology and treatment of X-linked hypophosphatemic rickets." Pediatric clinics of North America 44(2): 443-466. Chefetz, I. and E. Sprecher (2009). "Familial tumoral calcinosis and the role of O-

Dixon, P. H., P. T. Christie, et al. (1998). "Mutational analysis of PHEX gene in X-linked

Econs, M. J. and P. T. McEnery (1997). "Autosomal dominant hypophosphatemic

Econs, M. J., G. P. Samsa, et al. (1994). "X-Linked hypophosphatemic rickets: a disease often

Farrow, E. G., S. I. Davis, et al. (2006). "Extended mutational analyses of FGFR1 in

Farrow, E. G., S. I. Davis, et al. (2009). "Molecular analysis of DMP1 mutants causing autosomal recessive hypophosphatemic rickets." Bone 44(2): 287-294. Feng, J. Q., L. M. Ward, et al. (2006). "Loss of DMP1 causes rickets and osteomalacia and

Garringer, H. J., C. Fisher, et al. (2006). "The role of mutant UDP-N-acetyl-alpha-D-

Gaucher, C., O. Walrant-Debray, et al. (2009). "PHEX analysis in 118 pedigrees reveals new genetic clues in hypophosphatemic rickets." Hum Genet 125(4): 401-411. Gaucher, C., O. Walrant-Debray, et al. (2009). "PHEX analysis in 118 pedigrees reveals new genetic clues in hypophosphatemic rickets." Human genetics 125(4): 401-411.

hypophosphatemia." J Clin Endocrinol Metab 83(10): 3615-3623.

linked to chromosome 12p13." J Clin Invest 100(11): 2653-2657.

unknown to affected patients." Bone and mineral 24(1): 17-24.

endocrinology and metabolism 91(10): 4037-4042.

and pathogenesis." Seminars in nephrology 24(1): 39-47.

Hum Genet 78(2): 179-192.

Pflugers Arch 458(1): 39-52.

Acad Sci U S A 105(9): 3455-3460.

91(3): 664-671.

1792(9): 847-852.

537-539.

hypophosphatemic rickets with hypercalciuria predict a key role for the sodiumphosphate cotransporter NaPi-IIc in maintaining phosphate homeostasis." Am J

level results in hypophosphatemic rickets and hyperparathyroidism." Proc Natl

follow-up on one family." The Journal of bone and joint surgery. American volume

glycosylation in the maintenance of phosphate homeostasis." Biochim Biophys Acta

rickets/osteomalacia: clinical characterization of a novel renal phosphate-wasting disorder." The Journal of clinical endocrinology and metabolism 82(2): 674-681. Econs, M. J., P. T. McEnery, et al. (1997). "Autosomal dominant hypophosphatemic rickets is

osteoglophonic dysplasia." American journal of medical genetics. Part A 140(5):

identifies a role for osteocytes in mineral metabolism." Nat Genet 38(11): 1310-1315.

galactosamine-polypeptide N-acetylgalactosaminyltransferase 3 in regulating serum intact fibroblast growth factor 23 and matrix extracellular phosphoglycoprotein in heritable tumoral calcinosis." The Journal of clinical

the genetic diagnosis which warrants studies of associations between gene variants and therapeutic effects. Future studies of monogenic phosphate balance disorders will probably continue to include genomewide studies of families with genetically unexplained phosphate balance disorders. Animal and cell studies will probably also continue to contribute to the understanding of disease mechanism, and, in particular, the use of induced pluripotent stem cells (iPS) seems to be a promising new tool in the mechanistic and therapeutic studies (Rosenzweig 2010) as well as the use of small molecule screens in the search for new therapeutic options in monogenic disease (Shaw, Blodgett et al. 2011) .

## **6. Conclusion**

As we have discussed in this book chapter, several of the dominant and recessive forms of hypo- and hyperphosphatemic bone disease have received their molecular explanation leading to new insight into the pathophysiology of hypo- and hyperphosphatemic bone disease. The major advancement in pathophysiological understanding has come from the understanding of a bone-kidney axis where the central bone phosphatonin FGF23 acts on FGFR1-receptors in the kidneys to promote phosphaturia and from the understanding of all the factors converging on this axis. In fact, this axis ties together the known monogenic forms of renal phosphate disorders. In addition, the understanding of the genetics and pathophysiology of these disorders has led to the recognition of the two faces of some of the involved genes, showing both hyper- and hypofunction, dependent on the nature of the mutation, which is in particular the case for mutations affecting the *KL* and *FGF23* genes.

We recommend the use of a genetic-oriented classification instead of the traditional diseaseoriented classification since we believe that this will facilitate a broader understanding of the phenotype of monogenic phosphate balance disorders. Whereas increased molecular understanding has led to a more precise diagnosis, it has not yet led to new established treatment. We believe, however, that the molecular understanding will indeed facilitate the development of new treatment options with the use of the powerful tools including iPS cells and small molecular screens.

## **7. References**


the genetic diagnosis which warrants studies of associations between gene variants and therapeutic effects. Future studies of monogenic phosphate balance disorders will probably continue to include genomewide studies of families with genetically unexplained phosphate balance disorders. Animal and cell studies will probably also continue to contribute to the understanding of disease mechanism, and, in particular, the use of induced pluripotent stem cells (iPS) seems to be a promising new tool in the mechanistic and therapeutic studies (Rosenzweig 2010) as well as the use of small molecule screens in the search for new

As we have discussed in this book chapter, several of the dominant and recessive forms of hypo- and hyperphosphatemic bone disease have received their molecular explanation leading to new insight into the pathophysiology of hypo- and hyperphosphatemic bone disease. The major advancement in pathophysiological understanding has come from the understanding of a bone-kidney axis where the central bone phosphatonin FGF23 acts on FGFR1-receptors in the kidneys to promote phosphaturia and from the understanding of all the factors converging on this axis. In fact, this axis ties together the known monogenic forms of renal phosphate disorders. In addition, the understanding of the genetics and pathophysiology of these disorders has led to the recognition of the two faces of some of the involved genes, showing both hyper- and hypofunction, dependent on the nature of the mutation, which is in particular the case for mutations affecting the *KL* and *FGF23* genes. We recommend the use of a genetic-oriented classification instead of the traditional diseaseoriented classification since we believe that this will facilitate a broader understanding of the phenotype of monogenic phosphate balance disorders. Whereas increased molecular understanding has led to a more precise diagnosis, it has not yet led to new established treatment. We believe, however, that the molecular understanding will indeed facilitate the development of new treatment options with the use of the powerful tools including iPS cells

ADHR Consortium (2000). "Autosomal dominant hypophosphataemic rickets is associated

Alon, U. S. (2010). "Clinical practice : Fibroblast growth factor (FGF)23: a new hormone."

Amatschek, S., M. Haller, et al. (2010). "Renal phosphate handling in human--what can we learn from hereditary hypophosphataemias?" Eur J Clin Invest 40(6): 552-560. Bastepe, M. and H. Juppner (2008). "Inherited hypophosphatemic disorders in children and

Batra, P., Z. Tejani, et al. (2006). "X-linked hypophosphatemia: dental and histologic

Beck, L., A. C. Karaplis, et al. (1998). "Targeted inactivation of Npt2 in mice leads to severe

the evolving mechanisms of phosphate regulation." Rev Endocr Metab Disord 9(2):

renal phosphate wasting, hypercalciuria, and skeletal abnormalities." Proc Natl

with mutations in FGF23." Nat Genet 26(3): 345-348.

European journal of pediatrics.

findings." J Can Dent Assoc 72(1): 69-72.

Acad Sci U S A 95(9): 5372-5377.

therapeutic options in monogenic disease (Shaw, Blodgett et al. 2011) .

**6. Conclusion** 

and small molecular screens.

171-180.

**7. References** 


Monogenic Phosphate Balance Disorders 431

Koshida, R., H. Yamaguchi, et al. (2010). "A novel nonsense mutation in the DMP1 gene in a

Kuro-o, M., Y. Matsumura, et al. (1997). "Mutation of the mouse klotho gene leads to a

Kurosu, H., M. Choi, et al. (2007). "Tissue-specific expression of betaKlotho and fibroblast

Kurosu, H., M. Yamamoto, et al. (2005). "Suppression of aging in mice by the hormone

Levy-Litan, V., E. Hershkovitz, et al. (2010). "Autosomal-recessive hypophosphatemic

Lorenz-Depiereux, B., M. Bastepe, et al. (2006). "DMP1 mutations in autosomal recessive

Lorenz-Depiereux, B., A. Benet-Pages, et al. (2006). "Hereditary hypophosphatemic rickets

Lorenz-Depiereux, B., D. Schnabel, et al. (2010). "Loss-of-function ENPP1 mutations cause

Makitie, O., R. C. Pereira, et al. (2010). "Long-term clinical outcome and carrier phenotype in

Matsumura, Y., H. Aizawa, et al. (1998). "Identification of the human klotho gene and its

Passos-Bueno, M. R., W. R. Wilcox, et al. (1999). "Clinical spectrum of fibroblast growth

Prie, D., V. Huart, et al. (2002). "Nephrolithiasis and osteoporosis associated with

Quarles, L. D. (2003). "Evidence for a bone-kidney axis regulating phosphate homeostasis." J

Raeder, H., N. Shaw, et al. (2008). "A case of X-linked hypophosphatemic rickets (XLH):

Rosenzweig, A. (2010). "Illuminating the potential of pluripotent stem cells." The New

Rutsch, F., P. Boyer, et al. (2008). "Hypophosphatemia, hyperphosphaturia, and

FGF21." The Journal of biological chemistry 282(37): 26687-26695.

syndrome resembling ageing." Nature 390(6655): 45-51.

Klotho." Science 309(5742): 1829-1833.

homeostasis." Nat Genet 38(11): 1248-1250.

gene SLC34A3." Am J Hum Genet 78(2): 193-201.

hypophosphatemic rickets." Am J Hum Genet 86(2): 267-272.

factor receptor mutations." Hum Mutat 14(2): 115-125.

cotransporter." N Engl J Med 347(13): 983-991.

England journal of medicine 363(15): 1471-1472.

infantile arterial calcification." Nat Genet 34(4): 379-381.

Metab 28(5): 585-590.

Genet 86(2): 273-278.

Miner Res 25(10): 2165-2174.

Res Commun 242(3): 626-630.

Clin Invest 112(5): 642-646.

1):101-105.

Japanese family with autosomal recessive hypophosphatemic rickets." J Bone Miner

growth factor (FGF) receptor isoforms determines metabolic activity of FGF19 and

rickets is associated with an inactivation mutation in the ENPP1 gene." Am J Hum

hypophosphatemia implicate a bone matrix protein in the regulation of phosphate

with hypercalciuria is caused by mutations in the sodium-phosphate cotransporter

both generalized arterial calcification of infancy and autosomal-recessive

autosomal recessive hypophosphatemia caused by a novel DMP1 mutation." J Bone

two transcripts encoding membrane and secreted klotho protein." Biochem Biophys

hypophosphatemia caused by mutations in the type 2a sodium-phosphate

complications and the therapeutic use of cinacalcet." Eur J Endocrinol. 159 (Suppl.

bisphosphonate treatment are associated with survival beyond infancy in generalized arterial calcification of infancy." Circ Cardiovasc Genet 1(2): 133-140. Rutsch, F., N. Ruf, et al. (2003). "Mutations in ENPP1 are associated with 'idiopathic'


Goldfine, I. D., B. A. Maddux, et al. (2008). "The role of membrane glycoprotein plasma cell

Harahap, A. R. and J. W. Goding (1988). "Distribution of the murine plasma cell antigen PC-

Holm, I. A., A. E. Nelson, et al. (2001). "Mutational analysis and genotype-phenotype

Hu, M. C., M. Shi, et al. (2010). "Klotho: a novel phosphaturic substance acting as an autocrine enzyme in the renal proximal tubule." FASEB J 24(9): 3438-3450. Huang, R., M. Rosenbach, et al. (1994). "Expression of the murine plasma cell nucleotide

Huiming, Y. and W. Chaomin (2005). "Recombinant growth hormone therapy for X-linked

Huq, N. L., K. J. Cross, et al. (2005). "A review of protein structure and gene organisation for

encoded on human chromosome 4." Archives of oral biology 50(7): 599-609. HYP Consortium. A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nature Genetics 1995;11:130-6 Ichikawa, S., G. Baujat, et al. (2010). "Clinical variability of familial tumoral calcinosis caused by novel GALNT3 mutations." Am J Med Genet A 152A(4): 896-903. Ichikawa, S., E. A. Imel, et al. (2007). "A homozygous missense mutation in human KLOTHO causes severe tumoral calcinosis." J Clin Invest 117(9): 2684-2691. Ichikawa, S., E. A. Imel, et al. (2007). "A homozygous missense mutation in human

1 in non-lymphoid tissues." J Immunol 141(7): 2317-2320.

of clinical endocrinology and metabolism 86(8): 3889-3899.

beta." J Clin Invest 94(2): 560-567.

75.

CD004447.

318-319.

antigen 1/ectonucleotide pyrophosphatase phosphodiesterase 1 in the pathogenesis of insulin resistance and related abnormalities." Endocr Rev 29(1): 62-

correlation of the PHEX gene in X-linked hypophosphatemic rickets." The Journal

pyrophosphohydrolase PC-1 is shared by human liver, bone, and cartilage cells. Regulation of PC-1 expression in osteosarcoma cells by transforming growth factor-

hypophosphatemia in children." Cochrane database of systematic reviews(1):

proteins associated with mineralised tissue and calcium phosphate stabilisation

KLOTHO causes severe tumoral calcinosis." J Musculoskelet Neuronal Interact 7(4):

Therapeutic apheresis and dialysis : official peer-reviewed journal of the International Society for Apheresis, the Japanese Society for Apheresis, the

calcinosis. Secretion of fibroblast growth factor 23 requires O-glycosylation." The

Ichikawa, S., E. A. Traxler, et al. (2008). "Mutational survey of the PHEX gene in patients

Imel, E. A., S. L. Hui, et al. (2007). "FGF23 concentrations vary with disease status in autosomal dominant hypophosphatemic rickets." J Bone Miner Res 22(4): 520-526. Jonsson, K. B., R. Zahradnik, et al. (2003). "Fibroblast growth factor 23 in oncogenic osteomalacia and X-linked hypophosphatemia." N Engl J Med 348(17): 1656-1663. Juppner, H. (2007). "Novel regulators of phosphate homeostasis and bone metabolism."

Karim, Z., B. Gerard, et al. (2008). "NHERF1 mutations and responsiveness of renal

Kato, K., C. Jeanneau, et al. (2006). "Polypeptide GalNAc-transferase T3 and familial tumoral

with X-linked hypophosphatemic rickets." Bone 43(4): 663-666.

Japanese Society for Dialysis Therapy 11 Suppl 1: S3-22.

parathyroid hormone." N Engl J Med 359(11): 1128-1135.

Journal of biological chemistry 281(27): 18370-18377.


**19** 

*USA* 

Benjamin U. Nwosu

*Worcester, Massachusetts,* 

*University of Massachusetts Medical School* 

**Pseudohypoparathyroidism in Children** 

Albright hereditary osteodystrophy (AHO) is a genetic syndrome characterized by a distinctive set of developmental and skeletal defects that may easily be misdiagnosed as exogenous obesity in children. There are very few publications detailing the comprehensive management of children and adolescents with this disorder. This chapter provides a comprehensive discussion of the various aspects of this disorder. At the end, the reader should be able to: (1) List the clinical features of Albright hereditary osteodystrophy, (2) Identify the genetic and molecular abnormalities of AHO, (3) List the clinical features of pseudohypoparathyroidism type 1a (PHP 1a), (4) Describe the management of children and

The molecular basis for AHO is a heterozygous mutation of the gene that encodes the Gstimulatory subunit (Gsα) of guanine nucleotide-binding protein—the *GNAS* gene—that is located at chromosome 20q13.2. This type of mutation leads to a loss of expression or function of the Gsα which impairs the transmission of stimulatory signals to adenylate cyclase, limiting cyclic AMP generation necessary for hormone action(Lietman, 2008). The *GNAS* gene is subject to imprinting. Patients with AHO who have *GNAS* mutations on maternally inherited alleles manifest resistance to multiple hormones, such as parathyroid hormone (PTH), thyroid stimulating hormone (TSH), gonadotropins, growth-hormonereleasing hormone (GHRH), and glucagon(Brickman, 1986; Weinstein, 2004). These defects lead to PHP 1a. On the other hand, patients with AHO who have *GNAS* mutations on paternally inherited alleles have only the phenotypic features of AHO without hormonal resistance, a condition termed pseudopseudohypoparathyroidism (PPHP) (Lietman, 2008). PHP 1b is an autosomal dominant disorder that is associated with the presence of hormone resistance that is limited to PTH target organs, normal Gsα activity, and the absence of features of AHO(Levine, 1983). PHP 1c is associated with features of AHO, resistance to multiple hormones, and normal GNAS activity while PHP type 2 is associated with renal

resistance to PTH action and the absence of AHO phenotype(Levine, 2000).

More than 50 different loss-of-function mutations of *GNAS* have been reported in more than 70 affected individuals. Pohlenz *et al*.(Pohlenz, 2003) have reported a missense mutation, which results in the amino-acid substitution (Lys338Asn) in codon 338 of exon 12 of the *GNAS* gene associated with congenital hypothyroidism in AHO, though they did

**1. Introduction** 

adolescents with PHP 1a.

**2. Genetics of Albright hereditary osteodystrophy** 

