| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Comments |
Department of Medicine, University College London Medical School, Middlesex Hospital, London W1N 8AA, United Kingdom
Address correspondence and reprint requests to: Dr. S. M. Farrow, Department of Medicine, University College London Medical School, Middlesex Hospital, Mortimer Street, London, United Kingdom W1N 8AA. e-mail: s.farrow@ucl.ac.uk
Abstract
Hereditary vitamin D resistant rickets has been associated with a number of mutations within the DNA and ligand binding domains of vitamin D receptors (VDR). The aim of our study was to identify and characterize the causative mutations in three kindreds with this condition.
Resistance to 1,25(OH)2D3 was confirmed in cultured skin fibroblasts in which there was no induction of 24-hydroxylase activity; binding of 1,25(OH)2D3 to VDR was undetectable in patients 1 and 2, but normal in patients 3 and 4. The coding region of the VDR gene was sequenced to seek mutations. A mutation in the VDR gene of patient 1 resulted in a STOP codon, patient 2 showed a 56 bp deletion leading to frameshift and premature termination of VDR; a point mutation of A to C lying within the hormone-binding domain was shown for patients 3 and 4, who were siblings. Transactivation studies confirmed that these were functional mutations. Gel shift assays using nuclear extract from patient 3 demonstrated that the mutation that altered a conserved amino acid (glutamine-259) known to be involved in heterodimerization with other nuclear receptors affected protein:protein interactions.
1,25-DIHYDROXYVITAMIN D3 (1, 25(OH)2D3) acts through intracellular vitamin D receptors (VDR), which are ligand-dependent transcription factors, and binds to specific DNA sequences within target genes termed vitamin D response elements (VDRE). Human VDR is a 50 kDa protein, is a member of the nuclear receptor superfamily, and like other steroid hormone receptors such as, estrogen (ER), glucocorticoid (GR), thyroid hormone (TR), and retinoic acid (RAR) receptors, has a distinct domain organization. 1,25(OH)2D3 binds to a high affinity ligand binding domain separated from a well-conserved DNA binding domain by the hinge region. Lying within the DNA binding domain is a cysteine-rich region of 6070 amino acids that form two zinc finger motifs and are involved in specific interactions with DNA bases (1).
VDRE are composed of two hexameric half-sites usually arranged as a direct repeat (2). Although there is some evidence that VDR, like ER and GR, bind as homodimers to their respective response elements (3, 4, 5), VDR usually forms heterodimers with RAR (6), retinoid x receptors (RXR) (7, 8, 9), and possibly TR (10). Specific protein:protein interactions utilizing residues lying within the hormone-binding domain of nuclear receptors are thought to facilitate such heterodimerization. This subdomain has been termed the E1 region (11) and is well conserved throughout members of the receptor superfamily, and in VDR comprises residues 244263 (12).
Mutations within the VDR gene have been associated with a condition of end-organ resistance to 1,25(OH)2D3, hereditary vitamin D resistant rickets (HVDRR) (13, 14, 15). This is an autosomal recessive disorder characterized by high circulating levels of 1,25(OH)2D3, hypocalcemia, secondary hyperparathyroidism, early onset of rickets, and in approximately 50% of cases, alopecia (16). Mutations have been reported throughout the coding region of the VDR gene (17, 18, 19, 20, 21, 22, 23, 24, 25), and recently we have identified a novel intronic mutation in a 5' splice donor site of the VDR gene that causes exon skipping and deletion of the second zinc finger of the DNA binding domain (26). The aim of our present study was to identify the functional mutation in the VDR gene of three kindreds with HVDRR, to derive further information concerning the genesis of vitamin D resistance.
Materials and Methods
Patients
Four patients diagnosed with HVDRR with alopecia were studied. All were hypocalcemic and had raised PTH, alkaline phosphatase, and 1,25(OH)2D3 levels. Patient 1 (B.G.), male, presented at the age of 2 yr from a small village in Greece and had no family history of rickets or consanguinity. He was treated with One-Alpha (alfacalcidol, 0.6 µg/kg) and oral calcium with little effect and still had severe rickets at the age of 11. Patient 2 (A.H.) was a 3-yr-old German boy born of consanguineous parents who was diagnosed at the age of 7 months, while patients 3 (A.J.) and 4 (U.A.) were brother and sister, with parents of Indian origin who were first cousins. The brother was diagnosed at the age of 15 months and the girl at 2.5 yr old. Neither child responded to One-Alpha (20 µg/kg), but both subsequently received iv calcium with complete healing of the rickets and correction of the biochemical abnormalities.
Functional analysis of VDR
Skin fibroblasts were cultured and whole cell nuclear association assays were undertaken to determine the ability of 1,25(OH)2D3-VDR complexes to associate with nuclei in both patient and control skin fibroblasts, as described previously (27). The response of these cells to 1,25(OH)2D3 was assessed by measurement of 24-hydroxylase activity with high-performance thin-layer chromatography using the method of Thierry-Palmer and Gray (28).
Amplification of RNA and DNA
Total RNA was prepared from patient and control skin fibroblasts using the single step method (29). The coding region of the VDR gene was amplified in two overlapping segments using the reverse transcription-polymerase chain reaction (RT-PCR) with primers previously described (22) and conditions as described by Hawa et al (26). Products were cloned and sequenced using the ABI prism ready reaction dyedeoxy terminator sequencing kit and a 373 automated DNA sequencer (Applied Biosystems, Foster City, CA). Subsequently, the full-length coding region was amplified for transactivation studies. Following reverse transcription, primer 5'-GAGCACCCCTGGGCT (0.15 µmol/L) was added and complementary DNA amplified as above at an annealing temperature of 58 C, followed by a nested PCR step with primers 5'-CCTGCCCCCTGCTCCTTC and 5'-CCCAGGCACCGCACAGGC. The products were ligated into pSVK3 (Pharmacia LKB Biotechnology, Uppsala, Sweden).
Exon 7 was amplified from genomic DNA isolated from patient 2 and control fibroblasts as described (26) using primers 5'-CATGATGGACTCGTCCAGCTTC and 5'-CCTGGTATCATCTTAGCAAAGCC (annealing temperature 63 C, 2 mmol/L MgCl2).
Transactivation assays
VDR deficient CV-1 cells were grown to 5060% confluency in Dulbeccos modified Eagles medium supplemented with 10% (vol/vol) fetal calf serum, penicillin (100 Units/mL)/streptomycin (100 µg/mL), and 2 mmol/L L-glutamine (all reagents Gibco BRL Life Technologies, Paisley, UK). Cells were cotransfected with patient or control VDR coding region in pSVK3 (1 µg) and a pGL3-basic luciferase vector (Promega Corporation, Madison, WI) containing the human osteocalcin gene fragment (-834 to +1 bp) (1 µg) using 30 µL lipofectin (Gibco BRL Life Technologies) per plate (60 mm) for 6 h. A plasmid (1 µg) expressing ß-galactosidase (Promega) was used as an internal control for transfection efficiency. Cells were incubated for 48 h in the presence or absence of 1,25(OH)2D3 (10-7 mol/L), then cytosolic luciferase activity was determined (Promega assay reagents).
Gel shift assays
Nuclear extract was prepared from patient 3 fibroblasts using the method of Dignam et al. (30). The human osteocalcin gene fragment -576 to -426 bp, which contains the VDRE (31), was radiolabeled with T4 polynucleotide kinase (Boehringer Mannheim, Lewes, East Sussex, United Kingdom) and incubated (1.1 x 107 cpm/ug) with nuclear extract (10 µg) in 10 mmol/L Tris-HCl (pH 7.6) containing 100 mmol/L KCl and 10 µg poly(dI.dC) in a final volume of 50 µL at 20 C for 30 min. Reaction mixtures were separated on 4% nondenaturing polyacrylamide gels and protein:DNA complexes visualized by autoradiography.
In competition studies, 5-, 10-, and 20-fold molar excesses of unlabeled osteocalcin fragment -576 to -426 bp were added to the reaction mixture. Monoclonal antibodies directed against VDR (5 µL) (MAB1360, Chemicon International, Harrow, United Kingdom) or RXR (5 µL) (4RX-1D12, CNRS, INSERM, Universite Louis Pasteur, Strasbourg, France) and nonspecific rat immunoglobulin monoclonal antibodies (5 µL) (Sigma Chemicals, Poole, Dorset, United Kingdom) were used in the incubations.
Results
Functional characterization of VDR
There was no induction of 24-hydroxylase activity by 1,25(OH)2D3 (10-8 mol/L) in fibroblasts from all four patients (compared with control cells, which produced 640 fmoles/h/mg protein 24,25 dihydroxyvitamin D3). In whole cell nuclear association assays, receptors were undetectable in fibroblasts from patients 1 and 2, whereas receptor numbers in patients 3 and 4 fibroblasts were comparable to that in control cells (3300 and 4300 VDR/mg protein, respectively) and ligand bound with similar kDa values (2.82.95 x 10-10 mol/L).
Transactivation assays
To ascertain whether the mutations found in these patients were
functional, constructs containing the human osteocalcin VDRE in a
luciferase reporter gene were cotransfected into CV-1 cells expressing
patient or control VDR. Similar levels of luciferase activity were
observed for both control and patients in the absence of
1,25(OH)2D3 (Fig. 1
, open bars), but on
stimulation with 1,25(OH)2D3, a 6.5-fold
increase in luciferase activity was seen with control VDR (lane 2,
hatched bars), whereas there was less than 2-fold induction
with VDR from patients 1, 2, and 3. The apparent small increase in
activity seen in all the samples is likely to be the result of the
presence of endogenous wild type receptors in CV-1 cells, which are not
completely devoid of VDR.
|
To avoid artefacts, independent RT-PCR reactions (2, 3, 4) were performed and multiple clones were sequenced in each case.
Patient 1.
Following RT-PCR, analysis of the products obtained
from the 5' and 3' parts of the coding region revealed fragments of the
expected sizes of 720 bp and 646 bp, respectively. Sequencing the 5'
part of the coding region revealed a single base change in exon 4
[using the exon numbering system of Hawa et al (26)],
which codes for the second zinc finger of the DNA binding domain at
nucleotide position 217 of C to T (CGA to GA, [Table 1
using the numbering system of Baker
et al (32)]. This led to the generation of a STOP codon
replacing arginine at residue 73 and was the only mutation found within
the coding regions in the eleven clones sequenced from this
patient.
|
Patients 3 and 4.
In patient 3, a base change (CAG
to CCG, Table 1
) was detected at nucleotide position 776 in
patient 3. This falls in exon 8, which encodes the hormone binding
domain and alters the amino acid residue at position 259 from glutamine
to proline. This was the only mutation observed in ten clones
sequenced, and furthermore, was confirmed in his sibling. DNA from the
parents of patients 3 and 4 was used to amplify exon 8 for sequencing,
which demonstrated that they were both heterozygous for the
mutation.
Gel shift assays
As the mutation seen in patients 3 and 4 changed an amino acid
within the E1 subdomain involved in heterodimerization, the ability of
VDR prepared from cells from patient 3 and from the control to bind the
osteocalcin fragment containing the VDRE was assessed in gel shift
assays. In the absence of nuclear extract, no DNA:protein complexes
were seen (Fig. 2
; lane 1). while
addition of nuclear extract from control cells gave two complexes,
termed A and B (lane 2). Although both complexes were detected when
protein extract from patient 3 was used, complex B formation was
greatly reduced whereas that of complex A was enhanced (lane 3).
|
|
In this paper, we describe three mutations in the VDR gene associated with HVDRR. In patient 1, arginine-73 was mutated to a STOP codon in the second zinc finger motif, which would produce truncated VDR unable to bind to DNA. This change has previously been reported in a Moroccan patient by Weise et al (20); however, we have been unable to show an ancestral link between the two patients, and since Hughes et al (17) have also reported a mutation at this residue (to glutamine) in another patient, it seems likely that these are independent mutations in a particularly sensitive site.
In patient 2, part of exon 7 was deleted causing a frameshift and
premature STOP codon. In this case, the base change detected would
generate a cryptic 5' donor splice site within exon 7 (see Table 1
).
Specific sequences within introns and exons are essential for correct
RNA splicing, and it is likely that this upstream site within the exon
is recognized by the spliceosome complex in preference to the normal
splice site at the exon/intron boundary [for a review of RNA splicing
see Padgett et al (33)]. Such missplicing would be expected
to lead to the excision of the 3' end of exon 7 together with the
intron. This is not the first such exonic mutation reported that
affects splicing; other examples include a C to T mutation within the
acetoacetyl-coenzyme A thiolase gene that alters the secondary
structure of the pre-mRNA and, because of a misrecognition by the
splicesome assembly, results in skipping of an entire exon (34), while
in the human hypoxanthine-guanine phosphoribosyltransferase gene, a G
to T mutation within exon 3 generates a cryptic donor site leading to
skipping of this exon (35).
Interestingly, sequence analysis of the VDR gene in patient 2 revealed the presence of two initiation codons (ATG), as originally reported in the sequence (32). This was not seen in any of the other patients or control VDR that we have examined and has not been reported by a number of authors (18, 22, 24). However, recent work has suggested that the first ATG is a polymorphism, and initiation of translation from the first ATG produces VDR that bind ligand with equal affinity to that of VDR produced from the second ATG, but which apparently has reduced capability for target gene transactivation (36).
Our studies have demonstrated that cells from patients 3 and 4 contained a normal number of VDR with normal ligand binding; however, these cells were resistant to 1,25(OH)2D3, and there was little transactivation of reporter genes. In these patients, the mutation alters a residue (Gln-259), which is well conserved in nuclear receptors and which has been shown to be important in VDR heterodimerization, such that a substitution by glycine had no effect on ligand binding, but VDR-RXR-VDRE complex formation was impaired (12). Recently, Whitfield et al (25) have reported similar cases of HVDRR in which there were mutations in the ligand-binding domain which disrupted heterodimerization. Interestingly, the effect could be overcome in one patient by pharmacological doses of a vitamin D derivative, while the transactivation by the second mutant receptor could be restored by excess RXR with hormone.
From these data, it is clear that substitution of Gln-259 (a
hydrophilic amino acid) by proline (a hydrophobic amino acid) in
patients 3 and 4 would alter VDR conformation and so disrupt
heterodimerization and receptor function. Therefore, we investigated
VDR-VDRE interactions, which revealed two DNA:protein complexes using
nuclear extract from control cells (A and B, see Fig. 2
), both of which
contained VDR and RXR. However, although both complexes were evident
using patient sample, there was reduced formation of complex B
formation, with an enhancement of complex A. The DNA fragment used was
large and it is likely that, in addition to RXR and VDR, another as yet
unidentified protein(s) is present within complex A that requires the
presence of RXR and VDR in the complex. It is possible that the
mutation preferentially increases the affinity for the unidentified
protein(s) and favors formation of complex A. This is of particular
interest if this protein is a repressor protein similar to that
described as binding to thyroid hormone receptor (37), in which case
preferential formation of a complex containing this protein may lead to
lack of gene activation and HVDRR. Other possible candidate proteins
are general transcription factors such as TFIIB, which binds to the
carboxyl terminal portion of VDR (38), or Fos-Jun protein complexes,
which interact with VDR to form a heterotrimeric protein complex on the
human osteocalcin gene (39).
Thus in summary, we have identified three mutations in the VDR gene: one giving a premature STOP codon, another creating an exonic cryptic 5' splice site causing aberrant RNA processing, and a third that alters VDR heterodimerisation; all of these mutations are associated with HVDRR.
Acknowledgments
We would like to thank Dr. Talia Kakourou and Prof. Polyxeni Nicolaidou, of the First Department of Paediatrics, Athens University, Greece, for the donation of patient 1 fibroblasts; Dr. D. Schnabel of the Virchow Klinikum, Kinderklinik, Berlin, Germany, for the donation of patient 2 fibroblasts; Dr. M. A. Sills of Huddersfield Royal Infirmary, Huddersfield, England, for the donation of patients 3 and 4 fibroblasts and parental blood, and the Medical Research Council, United Kingdom, for their support.
Received January 24, 1997.
Revised May 20, 1997.
Accepted June 16, 1997.
References
This article has been cited by other articles:
![]() |
Y. Fang, J. B. J. van Meurs, F. Rivadeneira, N. M. van Schoor, J. P. T. van Leeuwen, P. Lips, H. A. P. Pols, and A. G. Uitterlinden Vitamin D Receptor Gene Haplotype Is Associated with Body Height and Bone Size J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1491 - 1501. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. Malloy, R. Xu, L. Peng, S. Peleg, A. Al-Ashwal, and D. Feldman Hereditary 1,25-Dihydroxyvitamin D Resistant Rickets due to a Mutation Causing Multiple Defects in Vitamin D Receptor Function Endocrinology, November 1, 2004; 145(11): 5106 - 5114. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. Malloy, J. W. Pike, and D. Feldman The Vitamin D Receptor and the Syndrome of Hereditary 1,25-Dihydroxyvitamin D-Resistant Rickets Endocr. Rev., April 1, 1999; 20(2): 156 - 188. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |