| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Center for Molecular Medicine and Division of Endocrinology and Metabolism, University of Connecticut School of Medicine (S.B.B., T.T.B., A.A.), Farmington, Connecticut 06030-3101; Memorial Sloan-Kettering Cancer Center (N.P., R.S.K.C.), New York, New York 10021; University of California Medical Center (I.S., W.G.), Los Angeles, California 90095; University of Florence (M.L.B.), 50139 Florence, Italy; Hôpital Necker (T.B.D., P.U.), 75743 Paris, France; Hôpital St. Louis (E.S.), 75475 Paris, France; and University of Cincinnati (J.W.P), Cincinnati, Ohio 45267
Address all correspondence and requests for reprints to: Dr. Andrew Arnold, Center for Molecular Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, Connecticut 06030-3101.
| Abstract |
|---|
|
|
|---|
First, Southern blotting and/or PCR analyses of 29 tumor samples from 14 genetically informative patients revealed no allelic losses at the VDR locus. Next, direct DNA sequencing of all VDR splice junctions, associated intronic sequences, and virtually the entire VDR-coding region for all 59 tumors revealed no acquired mutations. Last, 37 tumor DNA samples were subjected to comparative genomic hybridization, and no chromosomal losses in the VDR region (12cen-q12) were observed.
These observations suggest that inactivating defects within the VDR gene do not commonly contribute to the primary pathogenesis of severe refractory hyperparathyroidism in uremia.
| Introduction |
|---|
|
|
|---|
The existence of a monoclonal tumor implies that acquired (somatic) mutation in genes controlling cell proliferation or programmed cell death must have conferred a selective growth advantage on a parathyroid cell sufficient to lead to its clonal expansion (3). Although defects in two such genes, cyclinD1/PRAD1 and MEN1, are found in subsets of common parathyroid adenomas (4, 5, 6, 7, 8), specific oncogenes or tumor suppressor genes contributing to monoclonal uremic hyperparathyroidism have not yet been identified.
The vitamin D system represents an attractive potential target for acquired genetic mutation in parathyroid cells of uremic patients because of calcitriols effects on parathyroid cell growth and function. Calcitriol acts via its nuclear receptor to suppress parathyroid cell growth and down- regulate PTH synthesis and secretion (9). Vitamin D and its receptor may also play a role in determining the set-point of calcium-regulated PTH secretion, which has been shown to be abnormal in severe uremic hyperparathyroidism (10). Thus, an acquired mutation that inactivates the function of the vitamin D receptor (VDR) could confer a proliferative advantage on a parathyroid cell. Furthermore, although decreased VDR density has been observed in the parathyroid glands of both rats and dogs with experimental uremia and in patients with severe uremic hyperparathyroidism (11, 12, 13, 14, 15), the finding of recurrent clonal defects in the VDR gene would provide solid evidence that such abnormalities in VDR expression are truly a primary contributing cause of the proliferative disturbance. Therefore, to assess the potential role of the VDR in the pathogenesis of monoclonal parathyroid tumors in chronic renal failure, we evaluated 59 surgically resected parathyroid tumors from uremic patients for acquired inactivating mutations in the vitamin D receptor gene.
| Subjects and Methods |
|---|
|
|
|---|
Parathyroid tumor tissue and paired normal peripheral blood leukocytes were obtained from 33 patients (15 men and 18 women) with an average age of 43.5 yr (range, 1963 yr) who underwent surgical treatment for refractory uremic hyperparathyroidism. Indications for parathyroidectomy included intractable pruritus, bone pain and/or fractures, soft tissue calcifications, hypercalcemia, marked hyperphosphatemia, x-ray signs of renal osteodystrophy, and/or other symptoms that could not be controlled with medical therapy. Serum calcium levels were available for 32 of 33 patients. Nineteen of these 32 were normocalcemic; 13 were hypercalcemic, with an average total serum calcium of 11.5 mg/dL (range, 10.613 mg/dL), and were considered to have tertiary hyperparathyroidism. Serum PTH levels were elevated in all patients from whom data were available (32 of 33) to an average 16-fold above the upper limit of normal (range, 3.8- to 35fold). The number of years on intermittent dialysis treatment ranged from 0.518, with an average of 10 yr (data available from 30 of 33 patients, 2 patients were on continuous ambulatory peritoneal dialysis, the remaining 28 were on hemodialysis). A single hypercellular parathyroid gland was available for study from 16 patients, 2 glands were available from each of 11 patients, 3 glands were available from each of 3 patients, 4 glands were available from each of 2 patients, and from 1 patient parathyroid tissue was taken from an enlarged forearm autograft. In total, 59 parathyroid samples were obtained. Seventeen tumors were classified as nodular hyperplasia, and 32 were classified as generalized hyperplasia (9 tumors were not categorized or data were unavailable). The mean gland weight (data available for 24 of 59 tumor samples) was 1177 mg (range, 205980 mg). All tumors were snap-frozen immediately after surgical resection and stored at -80 C. Genomic DNA was isolated from tumor and peripheral blood leukocytes using standard techniques of proteinase K digestion and phenol/chloroform extraction. All samples were obtained in accordance with the regulations of the institutional review board on human studies.
Analysis of allelic loss at the VDR locus
Restriction digestion of paired peripheral blood leukocytes and tumor genomic DNA was performed with TaqI and subjected to Southern blotting using the 2.1-kb EcoRI fragment of human VDR complementary DNA plasmid pH13 as a probe (American Type Culture Collection, Manassas, VA). The presence of the polymorphic TaqI site in codon 352 yields a band of 2.0 kb (and invariant bands of 12.2, 10.2, and 1.5 kb) on the autoradiogram, whereas the absence of the polymorphic site yields a 2.2-kb band (16). For PCR analysis of allelic loss, exon 9 of the VDR gene was PCR amplified as described by Morrison et al. (17), and a TaqI restriction digest was performed. Products were separated on agarose gel containing ethidium bromide and visualized under UV light. Samples homozygous for the presence of the polymorphic site (tt) yield bands of 290, 245 (invariant), and 205 bp, whereas samples homozygous for the absence of the polymorphic site (TT) yield bands of 495 and 245 bp (18). The heterozygous (Tt) pattern thus contains bands of 495, 290, 245, and 205 bp. Relative allele intensities were assessed to determine whether one VDR allele was selectively lost in the patients tumor DNA.
Direct DNA sequencing
All eight coding exons were PCR amplified using the following intronic primers, based upon the VDR gene organization (19) and genomic sequence obtained by one of us (J.W.P.): exon 2: 2F, 5'-AGGAATTCAGCTGGCCCTGGCACTGACTCTGCTCT-3'; 2R, 5'-CTGAATTCA-TGGAAACACCTTGCTTCTTCTCCCTC-3'; exon 3: 3F, 5'-GTGAATTCAGGGTGAGGAGCCGGAAGTTCAGTGAC-3'; 3R, 5'-GCGAATTC-CTTTCCCTGACTCCACTTCAGGCCCAA-3'; exons 4 and 5 (amplified and sequenced together): 4/5F, 5'-GAGAATTCTGAGAGCTGCTGCAGC-3'; 4/5Fa, 5'-CATTCTGACAGATGAGGAAGTGC-3'; 4/5R, 5'-CCGAATTCTATGGAGACTGAAGTCCTG-3'; exon 6: 6F, 5'-TGGAATTCCAGTCTGGCTCTGCTG-3'; 6R, 5'-TTGAATTCTTGTAGCTCAGTC-TAGGA-3'; exons 7 and 8 (amplified and sequenced together): 7/8F, 5'-GCGAATTCCGTTACTGGTAACCTGACCTCCTTC-3'; 7/8R, 5'-AATTCATACACCCCGCTCCCCACGTCCCTGAG-3'; and exon 9: 9F, 5'-TGGAATTCAGCAGTGAGGTGCCCAGCTGAG-3'; 9R, 5'-TCGA-ATTCTGAGGAGGGCTGCTGAGTAGCC-3'.
Cycle sequencing reactions were performed on column-purified PCR product using each of the above primers and Big Dye fluorescence-labeled deoxy-NTPs (PE Applied Biosystems, Foster City, CA). Each exon was sequenced in both the forward and reverse directions. Products were separated on a denaturing polyacrylamide/formamide gel and visualized using the ABI Prism 377 DNA Sequencer (PE Applied Biosystems, Foster City, CA).
In total, over 98% of the VDR-coding region was successfully amplified and sequenced. PCR amplification of exons 4 and 5 with the intronic primers 4/5F and 4/5R above was unsuccessful in our hands. However, we were able to amplify most of the desired region by synthesizing a primer falling at the 5'-end of the exon 4 coding sequence (designated 4/5Fa above) and using it with the original intronic reverse primer (4/5R above) falling 3' of exon 5. This yielded a PCR product that included the entire exon 5, the intron between exons 4 and 5, and all but 20 nucleotides of exon 4 (constituting <2% of the VDR-coding region).
Comparative genomic hybridization (CGH)
CGH is a recently developed technique capable of detecting large scale chromosomal gains and/or losses in tumors. Probe (tumor) DNA and normal reference DNA were labeled differentially with fluorescein 12-deoxyuridine 5-triphosphate and Texas Red 5-deoxyuridine 5-triphosphate, respectively. Equal amounts of each were suspended in hybridization mixture and competitively hybridized to normal human metaphase chromosomes. The chromosomes were counterstained with 4,6-diamino-2-phenylindole for their identification. The fluorescence intensities and staining pattern were visualized using a cooled charge-coupled device camera attached to a microscope and quantitated as previously described (20, 21).
| Results |
|---|
|
|
|---|
Allelic loss was examined in 29 tumors from 14 patients who were
heterozygous, or informative, for the TaqI RFLP found in VDR
codon 352 (exon 9). Twenty-two of these tumors were analyzed for
allelic loss by Southern blotting, and 7 were analyzed by PCR. No loss
of heterozygosity (LOH) at this location directly within the VDR gene
was found in any of the 29 informative tumors analyzed. Figure 1
shows two examples of patients
heterozygous for the TaqI polymorphism, but with no
tumor-specific LOH; one was analyzed by Southern blotting (Fig. 1a
),
and the other by PCR (Fig. 1b
).
|
The VDR genes coding region and splice junctions were sequenced
in all 59 tumor DNA samples, looking for subtle inactivating mutations.
No mutations of the VDR gene were observed. The sequence analysis did
identify a number of genetic polymorphisms. Twelve of 33 (36.4%)
patients were heterozygous for a previously described ATG
ACG start
codon polymorphism at the beginning of exon 2 (22, 23, 24). Eleven of 33
(33.3%) were homozygous for ACG, and the remaining 10 patients
(30.3%) were homozygous for ATG. At the TaqI polymorphism
in exon 9, codon 352, 16 of 33 (48%) patients were heterozygous for
the TTGA
TCGA polymorphism in complete concordance with our
restriction enzyme analyses. There were 2 patients whose tumors were
heterozygous by sequence but for whom LOH analysis could not be
performed because the matched DNA from peripheral blood leukocytes was
degraded and unsuitable for PCR amplification. In addition, several
putative, single nucleotide polymorphisms, which were either intronic
or silent mutations causing no change in the amino acid sequence, were
found in sequencing the tumor DNA samples. At base 57 of exon 2 (codon
19), 1 patient (3%) was heterozygous for an AAC
AAT polymorphism,
previously undescribed to the best of our knowledge (both encode Asn).
Two of 33 (6%) patients were heterozygous for a C
T polymorphism at
base 8 of the intron between exons 2 and 3, and 1 (3%) was homozygous
for the variant T at this position. An additional intronic polymorphism
was identified at base 75 between exons 7 and 8. Three patients (9%)
were heterozygous for a 1-base, G, insertion at this position. Finally,
at base 147 of exon 4 (codon 141), 1 patient (3%) was heterozygous for
a TCC
TCT polymorphism (both encode Cys). Representative confirmation
of the ability of our sequencing methodology to sensitively detect
heterozygous mutations is shown in Fig. 2
, which depicts the sequence of a tumor
heterozygous for the polymorphism found at the start codon in exon
2.
|
CGH analysis of 37 uremia-associated parathyroid tumors showed no chromosomal losses in the region of the VDR gene (12cen-q12; data not shown) (25). This is in agreement with our data, which showed no allelic loss by PCR and/or Southern blotting.
| Discussion |
|---|
|
|
|---|
A majority of chronic renal failure patients with severe refractory hyperparathyroidism harbor at least one monoclonal parathyroid tumor (1, 2), but the specific acquired genetic defects that confer the selective growth advantage implied by such clonality remain poorly understood. Defining these genetic mechanisms is pertinent because the clonal status of a tumor seems to be an important determinant of its clinical profile. Monoclonal growth could contribute to disease severity through critical enlargement of the parathyroid cell burden and/or an abnormal Ca2+-PTH set-point, both of which are seen in other types of monoclonal parathyroid tumors and are characteristics of refractory uremic hyperparathyroidism (10, 30, 31, 32, 33, 34, 35, 36). In fact, large parathyroid gland size has been associated with higher levels of PTH and more severe bone symptoms refractory to medical therapy with calcitriol (33, 34, 35), a nodular parathyroid growth pattern with a higher proliferative rate (37), and a significantly reduced density of both the VDR and the Ca2+-sensing receptor (38, 39) compared with generalized hyperplasia.
The VDR gene is a compelling candidate parathyroid tumor suppressor gene, because a parathyroid cell in which both VDR alleles become somatically inactivated would plausibly gain a selective growth advantage through impaired responsiveness to the antiproliferative influence of 1,25-dihydroxyvitamin D3 (40, 41, 42). Furthermore, recurrent VDR gene defects would both explain the observation of abnormal VDR expression and testify to its pathogenetic significance in driving the growth of medically refractory uremia-associated parathyroid tumors.
Additional evidence for a possible role of the VDR gene comes from observations that have suggested a role of the VDR gene polymorphism in controlling parathyroid function and growth in patients with primary or secondary uremic hyperparathyroidism (43, 44). However, this issue remains a matter of debate (45).
Therefore, to assess VDR gene inactivations potential role in the primary pathogenesis of refractory uremic hyperparathyroidism, we rigorously analyzed the VDR gene in 59 parathyroid tumors surgically resected from uremic patients. We used multiple methods that collectively were capable of finding both large (chromosomal and subchromosomal) scale and subtle DNA defects of the types that commonly contribute to inactivation of authentic tumor suppressors. Our observations indicate that such tumor-specific defects occur rarely if at all and make it unlikely that somatic mutation in the VDR gene plays an important role in triggering or driving clonal expansions in uremic parathyroid glands. Consistent with our results, Chudek et al. (46) found an extremely low frequency of LOH on 12q in a similar set of tumors (with such LOH excluding the VDR locus in one tumor); neither VDR mutation analysis nor CGH was performed in this study. Although our methods did not exclude some possible inactivating lesions, such as mutation of the VDR promoter or its hypermethylation, knowledge of established tumor suppressor genes makes it implausible that frequent clonal VDR inactivation would occur to the complete exclusion of generally common mechanisms such as coding region mutation and deletion. Decreased VDR expression might still be functionally significant in clonal uremic hyperparathyroidism, but at best would appear to be a downstream consequence, directly or indirectly, of more primary acquired genetic abnormalities or of other manifestations of the growth-disturbed cell. Thus, somatic changes in genes other than the VDR are probably responsible for the emergence of monoclonal parathyroid growth and for the development of a more severe clinical phenotype in uremia-associated hyperparathyroidism.
| Footnotes |
|---|
Received September 28, 1999.
Revised November 9, 1999.
Accepted November 17, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Hibi, F. Kambe, Y. Tominaga, Y. Mizuno, H. Kobayashi, K. Iwase, T. Imai, and H. Seo Up-Regulation of the Gene Encoding Protein Kinase A Type I{alpha} Regulatory Subunit in Nodular Hyperplasia of Parathyroid Glands in Patients with Chronic Renal Failure J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 563 - 568. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Fukagawa, J. J. Kazama, and T. Shigematsu Management of patients with advanced secondary hyperparathyroidism: the Japanese approach Nephrol. Dial. Transplant., September 1, 2002; 17(9): 1553 - 1557. [Full Text] [PDF] |
||||
![]() |
L. Canaff and G. N. Hendy Human Calcium-sensing Receptor Gene. VITAMIN D RESPONSE ELEMENTS IN PROMOTERS P1 AND P2 CONFER TRANSCRIPTIONAL RESPONSIVENESS TO 1,25-DIHYDROXYVITAMIN D J. Biol. Chem., August 9, 2002; 277(33): 30337 - 30350. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Imanishi, H. Tahara, N. Palanisamy, S. Spitalny, I. B. Salusky, W. Goodman, M. L. Brandi, T. B. Drueke, E. Sarfati, P. Urena, et al. Clonal Chromosomal Defects in the Molecular Pathogenesis of Refractory Hyperparathyroidism of Uremia J. Am. Soc. Nephrol., June 1, 2002; 13(6): 1490 - 1498. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Marx Hyperparathyroid and Hypoparathyroid Disorders N. Engl. J. Med., December 21, 2000; 343(25): 1863 - 1875. [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 |