| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (S.I., E.A.I., A.H.S., M.J.E.), Indiana University School of Medicine, Indianapolis, Indiana 46202; Department of Pediatrics (E.A.I.), Indiana University School of Medicine, Indianapolis, Indiana 46202; Division of Endocrinology (R.S., P.K.), Medical College of Wisconsin, Milwaukee, Wisconsin 53226; The Eye Institute (G.J.H.), Medical College of Wisconsin, Milwaukee, Wisconsin 53226; Endocrine-Diabetes Center (J.L.S.), St. Lukes Medical Center, Milwaukee, Wisconsin 53215; and Department of Medical and Molecular Genetics (M.J.E.), Indiana University School of Medicine, Indianapolis, Indiana 46202
Address all correspondence and requests for reprints to: Michael J. Econs, M.D., Department of Medicine, Indiana University School of Medicine, 541 North Clinical Drive, Clinical Building 459, Indianapolis, Indiana 46202-5121. E-mail: mecons{at}iupui.edu.
| Abstract |
|---|
|
|
|---|
-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 3 (GalNAc transferase 3 or GALNT3). Objective: The objective was to identify mutations in FGF23 or GALNT3 responsible for a mild TC phenotype by DNA sequencing and to determine serum FGF23 levels by ELISA.
Patients or Other Participants: The subject was a 25-yr-old Caucasian woman with eyelid calcifications and biochemical features of TC.
Results: Eyelid biopsy revealed superficial dermis calcifications. There was no history of metastatic calcifications, mineral homeostasis abnormalities, or renal dysfunction. Biochemistry revealed normal levels of calcium, creatinine, PTH, and 25-hydroxyvitamin D, with elevated phosphorous, TmP/GFR, and high normal 1,25-dihydroxyvitamin D levels. Intact FGF23 was undetectable (<3 pg/ml), whereas C-terminal FGF23 was elevated (698.2 RU/ml). Mutation detection revealed compound heterozygosity for two novel mutations in the glycosyl transferase domain of the GALNT3 gene.
Conclusion: Previously reported GALNT3 mutations in TC have been null mutations. This study shows that missense mutations affecting the glycosyl transferase domain of GalNAc transferase 3 also cause TC. Elevated C-terminal FGF23 fragments with undetectable intact FGF23 suggest that the mutant enzyme lacks the ability to glycosylate FGF23 and that glycosylation by GalNAc transferase 3 is necessary for secretion of functional full-length FGF23.
| Introduction |
|---|
|
|
|---|
Familial TC is caused by biallelic mutations in either the fibroblast growth factor 23 (FGF23) gene (1, 2, 3, 4) or the uridine diphosphate-N-acetyl-
-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 3 (GalNAc transferase 3 or GALNT3) gene (5, 6, 7, 8). FGF23 has been identified as a hormone-regulating renal tubular phosphate reabsorption and 1,25-dihydroxyvitamin D metabolism. Increased circulating levels of functional FGF23 are involved in the pathogenesis of renal phosphate wasting disorders, such as autosomal dominant hypophosphatemic rickets, X-linked hypophosphatemic rickets, and tumor-induced osteomalacia (9, 10). These disorders are characterized by low serum phosphate concentration due to decreased renal phosphate reabsorption and inappropriately low or normal 1,25-dihydroxyvitamin D and are, therefore, considered the clinical converse of TC. GALNT3 encodes a Golgi-associated biosynthetic enzyme that initiates mucin-type O-glycosylation of proteins. In addition to TC, GALNT3 mutations have been found in hyperostosis-hyperphosphatemia syndrome (HHS) (11), which is characterized by the presence of hyperphosphatemia and transient, localized bone lesions with cortical hyperostosis (12). Mutations previously identified in the GALNT3 gene have been null mutations affecting conserved splice sites or introducing a stop codon (5, 6, 7, 8).
Here we present a patient with eyelid calcifications and biochemical abnormalities suggestive of TC who was found to carry two novel missense mutations in the glycosyl transferase domain of the GALNT3 gene.
| Subjects and Methods |
|---|
|
|
|---|
Blood samples were collected from the patient with TC and healthy population-matched individuals. The study was approved by the Institutional Review Board of Indiana University-Purdue University Indianapolis. Written informed consent was obtained from each individual before his/her participation in the study. Serum biochemistry of the patient was analyzed at Aurora Clinical Laboratory (Milwaukee, WI) and Dynacare Laboratories (Milwaukee, WI). 25-Hydroxyvitamin D and 1,25-dihydroxyvitamin D levels were measured at Laboratory Corporation of America (Burlington, NC). The tubular maximum for phosphate reabsorption per deciliter of glomerular filtrate (TmP/GFR) was determined in the fasting state using the nomogram of Walton and Bijvoet (13).
Mutation analysis
Genomic DNA was extracted from blood obtained from the affected subject, using a QIAamp DNA Blood Mini Kit (QIAGEN Inc., Valencia, CA). All exons and their adjacent intronic sequences in the FGF23 and GALNT3 genes were amplified, using AmpliTaq DNA Polymerase (Applied Biosystems, Foster City, CA). Primer sequences and PCR conditions are available upon request. PCR products were electrophoresed in a 2% agarose gel and purified, using a DNA Gel Extraction Kit (QIAGEN Inc.). Approximately 100 ng of each PCR amplicon was directly sequenced from forward PCR primers, using Big-Dye Terminator Cycle Sequencing Kit and the ABI PRISM 3100 Genetic Analyzer (Applied Biosystems).
PCR-restriction fragment length polymorphism (RFLP) analysis
To determine whether the observed nucleotide changes in the GALNT3 gene were mutations or polymorphisms, we performed PCR-RFLP analysis in healthy Caucasian controls. The C-to-A transversion in exon 3 was amplified with forward 5'-TTTGAACGGCTAGGGTTGAC-3' and reverse 5'-CTTACAGTGAGCATCTAAAAATGTAAGC-3'. The mismatch nucleotide (underlined) was introduced in the reverse primer to create a restriction endonuclease HindIII site in the potential mutant allele. The C-to-A transversion in exon 5 was amplified with forward 5'-CTGGGAGTCGCTTCCTGAT-3' and reverse 5'-AGTCCTCCTGCAAAAGTGGAT-3'. The mismatch nucleotide (underlined) was introduced in the reverse primer to create a restriction endonuclease BstF5I site in the normal allele. The PCR amplicons were incubated with five units of HindIII at 37 C or BstF5I at 65 C (New England Biolabs, Beverly, MA). The digested PCR products were electrophoresed in a 3% agarose gel and visualized under the UV light.
Measurement of serum FGF23 levels
Serum FGF23 concentrations were determined using two different assays according to the manufacturers instructions. Full-length FGF23 concentration was measured using a FGF23 ELISA Kit (Kainos Laboratories Inc., Tokyo, Japan). FGF23 was also measured using the Human FGF-23 (C-Term) ELISA Kit (Immutopics International, San Clemente, CA), which detects both full-length and C-terminal fragments of FGF23.
| Results |
|---|
|
|
|---|
A 25-yr-old Caucasian woman was evaluated for a 3-yr history of "white dots" on her eyelids (Fig. 1
). A biopsy revealed calcifications within the superficial dermis. There was no history of other metastatic calcifications, abnormalities of mineral homeostasis, or renal dysfunction. She did not have a history of the typical painful bone swellings that are seen in patients with HHS. The past medical history was significant for stress fractures of her left tibia while involved in sporting activities in the fourth and seventh grades. She had a history of asthma and seasonal allergies. Her medications included an oral contraceptive, cetirizine, and inhaled albuterol. Other than the eyelid calcifications, the physical examination was normal. Whole body dual-energy x-ray absorptiometry imaging revealed no obvious sites of metastatic calcification. A radiological skeletal survey showed no evidence of ectopic calcification or hyperostosis. However, she was noted to have more calcified costal cartilage than typical for her age.
|
|
Mutation analysis
The affected individual has no mutations in the FGF23 gene. However, analysis of the GALNT3 sequence revealed that she is a compound heterozygote for two novel mutations in GALNT3: a C-to-A transversion in exon 3 (c.815C>A or g.8773C>A) and a C-to-A transversion in exon 5 (c.1076C>A or g.11839C>A) (Fig. 2
). Both transversions result in the change of an uncharged residue, threonine, to a positively charged residue, lysine (Thr272Lys and Thr359Lys, respectively).
|
Serum FGF23 levels
Intact FGF23 measured in an assay that only immunoreacts with full-length FGF23 was undetectable (<3.0 pg/ml) in serum of the affected individual, whereas C-terminal FGF23 measurement (assayed in another assay that detects both full-length and C-terminal fragments) was elevated (698.2 RU/ml). The control means ± SD determined in 118 healthy normophosphatemic control subjects in our laboratory are 29.7 ± 20.7 pg/ml for intact FGF23 and 72.9 ± 38.2 RU/ml for C-terminal FGF23 (14).
| Discussion |
|---|
|
|
|---|
TC usually manifests as soft tissue calcifications, most often around major joints, in the first two decades of life (15). Eyelid calcification has been previously described (3, 16, 17), but we found no prior reports of isolated eyelid calcification in TC. We cannot exclude the possibility that the condition may become more aggressive in this patient. Milder phenotypes seen in our patient would be expected if missense mutations did not completely abolish the enzymatic activity of encoded GalNAc transferase 3. However, although C-terminal FGF23 fragments were elevated, intact (functional) FGF23 was not detected in our patient. This result suggests that these missense mutations are likely functional null mutations as in other reported GALNT3 mutations (5, 6, 7, 8), and thus, residual enzymatic activity (if any) is unlikely to explain the milder phenotype. It should be noted that our patient was clearly not producing as much FGF23 in response to elevated serum phosphate because C-terminal fragments were lower than most previously described subjects with nonsense or splice-site mutations in GALNT3 (8). It is possible that this represents individual variation in this disorder, differences between the mutations described, or partial response to fluctuations in phosphate or 1,25-dihydroxyvitamin D in our patient.
Interestingly, the GALNT3 mutations are also responsible for another hyperphosphatemic disease, HHS (11). Narchi (18) reported the case of a girl who initially displayed a HHS feature and years later developed TC symptoms. There is another case of a young girl with both TC and cortical hyperostosis (19). Three children with diaphyseal bone lesions in addition to usual TC symptoms have also been reported (20). Therefore, mutations in GALNT3 result in subtle eyelid calcifications seen in our patient and in more severe phenotypes such as large calcific masses in typical TC, painful bone swelling in HHS, or likely a combination of the latter two. Because hyperphosphatemia is a common feature between HHS and TC, phenotypic manifestations may depend on how one responds to a primary defect in phosphate homeostasis (i.e. hyperphosphatemia). Furthermore, these observations suggest the presence of a modifier gene(s) or environmental factors that are responsible for phenotypic variability in this disease.
FGF23 is a phosphaturic hormone that is associated with various disorders of phosphate homeostasis, including autosomal dominant hypophosphatemic rickets and X-linked hypophosphatemic rickets. In the TC patients with FGF23 mutations, intact FGF23 levels in serum are low or undetectable, whereas C-terminal FGF23 fragments are highly elevated (1, 4). TC patients with GALNT3 mutations also had elevated C-terminal FGF23, presumably to compensate for increased phosphate and 1,25-dihydroxyvitamin D concentrations (8). However, intact FGF23 levels in patients with GALNT3 mutations have been unknown. The present study shows that mutations in GALNT3 also result in a pattern of serum FGF23 levels (undetectable intact; high C-terminal) similar to that observed in the patients with FGF23 mutations causing TC. These results suggest that hyperphosphatemia results from the inability of the cell to secrete active (intact) FGF23 in the setting of inactivating GALNT3 mutations.
Inactivating mutations in either FGF23 or GALNT3 result in familial TC. Furthermore, all reported FGF23 mutations causing TC have been either replacement of serine (1, 2, 4) or creation of threonine (3). As GalNAc transferase 3 transfers GalNAc from the sugar donor UDP-GalNAc to the hydroxyl group of a serine or threonine residue, the residues affected by the FGF23 mutations could be a potential substrate for the enzyme. In fact, the FGF23 protein is known to be O-glycosylated (21). Taken together with levels of intact and C-terminal FGF23 observed in patients with TC, these observations suggest that O-glycosylation by GalNAc transferase 3 might be necessary for stability and secretion of full-length FGF23. Strong in vitro evidence provided by a recent study confirmed that secretion of FGF23 indeed requires O-glycosylation (22).
In conclusion, we identified two novel missense mutations in GALNT3 resulting in coding changes in the catalytic domain of GalNAc transferase 3. The fact that GALNT3 mutations were found in a patient with a milder form of TC suggests that the disease may be more prevalent than once expected, with a range of potentially milder phenotypes. Dysfunction of GalNAc transferase 3 causes increased circulating C-terminal, but not intact, FGF23. Lack of active intact FGF23 results in the phenotype of increased tubular reabsorption of phosphate, hyperphosphatemia, and other manifestations of TC.
| Footnotes |
|---|
Disclosure statement: S.I., E.A.I., A.H.S., R.S., P.K., G.J.H., and M.J.E. have nothing to declare. J.L.S. received consulting fees, paid advisory board, and lecture fees from P&G and Merck.
First Published Online August 29, 2006
Abbreviations: FGF23, Fibroblast growth factor 23; GALNT3 (or GalNAc transferase 3), uridine diphosphate-N-acetyl-
-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 3; HHS, hyperostosis-hyperphosphatemia syndrome; RFLP, restriction fragment length polymorphism; TC, tumoral calcinosis; TmP/GFR, tubular maximum for phosphate reabsorption per deciliter of glomerular filtrate.
Received June 13, 2006.
Accepted August 18, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Liu and L. D. Quarles How Fibroblast Growth Factor 23 Works J. Am. Soc. Nephrol., June 1, 2007; 18(6): 1637 - 1647. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ichikawa, V. Guigonis, E. A. Imel, M. Courouble, S. Heissat, J. D. Henley, A. H. Sorenson, B. Petit, A. Lienhardt, and M. J. Econs Novel GALNT3 Mutations Causing Hyperostosis-Hyperphosphatemia Syndrome Result in Low Intact Fibroblast Growth Factor 23 Concentrations J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1943 - 1947. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |