| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
RAPID COMMUNICATION |
Department of Internal Medicine (C.T., P.Q.L.), Children’s Unit, Centre Hospitalier Etterbeek-Ixelles, B-1050 Brussels; Institut für Biochemie (S.W.S., S.H., M.W.), Universität Erlangen-Nürnberg, D-91054 Germany; and 3Department of Medical Genetics (J.P., M.J.A.), Hôpital Erasme and Laboratory of Medical Genetics, Université Libre de Bruxelles, B-1070 Brussels, Belgium
Address all correspondence and requests for reprints to: Marc J. Abramowicz, M.D., Ph.D., Genetics Department, Hôpital Erasme, Univerxité Libre de Bruxelles, 808 Route de Lennik, 1070 Brussels, Belgium. E-mail: marcabra{at}ulb.ac.be.
Abstract
Isolated hypoparathyroidism is an uncommon metabolic disorder characterized by hypocalcemia and hyperphosphatemia, with absent or low levels of PTH. It may present as an apparently sporadic disorder or may be transmitted in families as a genetic trait. Mutations of the calcium-sensing receptor gene and of the preproPTH gene have been reported in occasional cases, and a mutation of the parathyroid-specific transcription factor GCMB gene has been reported in one familial case. We report a second family with isolated hypoparathyroidism and a GCMB mutation. The patients were two siblings from asymptomatic, first-cousin parents, indicating autosomal recessive inheritance. The mutation consisted of the substitution of a glycine residue with a serine at position 63 (G63S) in the DNA-binding GCM domain of GCMB. Functional studies in transfected cells showed that the mutation caused loss of GCMB function, as it abolished transactivation capacity, despite normal subcellular localization, protein stability, and DNA-binding specificity. Contrary to the previously reported family, our patients displayed low but clearly detectable levels of PTH in plasma. This residual hormone secretion probably results from a very small residual activity of the G63S mutant GCMB.
CALCIUM CONCENTRATION IN serum is regulated by a complex endocrine pathway that orchestrates the intricate mechanisms of calcium reabsorption from renal glomerular filtrate, absorption from intestine, and mobilization from bone. Crucial to this regulation is PTH. PTH secretion by parathyroid gland cells is inversely correlated with the extracellular level of ionized calcium, whose tight feedback action is mediated through a specific calcium-sensing receptor (CaSR) abundantly expressed at the plasma membrane of parathyroid gland and renal medullary cells (1).
Hypoparathyroidism is characterized by hypocalcemia and hyperphosphatemia resulting from insufficient or defective PTH action (2). Hypoparathyroidism may be further classified as isolated or syndromic. Hypoparathyroidism is a constant, or frequent, feature of several genetic syndromes of abnormal development, among which the DiGeorge syndrome of abnormal thymic, parathyroid, and conotruncal heart development is relatively frequent (3). Several other genetic syndromes with hypoparathyroidism include the hypoparathyroidism, sensorineural deafness, renal anomaly (HDR) syndrome (4), the Kenny-Caffey syndrome (5), the Kearns-Sayre syndrome (6), the autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (7), and others (8). Isolated hypoparathyroidism has been reported as apparently sporadic, or as a familial trait with either autosomal dominant, autosomal recessive, or X chromosome-linked inheritance (9). Mutations in the preproPTH gene have been described in both autosomal dominant and autosomal recessive forms of familial isolated hypoparathyroidism (10). Some CaSR mutations may cause autosomal dominant hypocalcemic hypercalciuria that can mimic hypoparathyroidism (11). Analyses of the two latter genes fail, however, to identify mutations in many, if not most, cases of isolated hypoparathyroidism.
GCMB (referred to as Gcm2 and GCMb in mice) is a transcription factor specifically expressed in parathyroid cells (12). Gcm2/GCMb-deficient mice fail to develop parathyroid glands. Although some 30% die shortly after birth, the remaining are viable and fertile but display hypocalcemia and hyperphosphatemia and an inappropriately normal serum concentration of PTH. Residual PTH secretion in these mice originates from GCMa-expressing cells located in the thymus (13). One single family has been reported with a mutated GCMB and isolated hypoparathyroidism. Unlike the Gcm2/GCMb-deficient mice, the two affected relatives studied in the latter report had very low or undetectable levels of circulating PTH (14). We here report a newly ascertained family with isolated hypoparathyroidism and low but clearly detectable PTH levels associated with a loss-of-function mutation of the GCMB gene.
Subjects and Methods
Case Reports
The proband, a 5-yr-old boy, was the third child of consanguineous, first-cousin parents of Moroccan origin (Fig. 1A
). He was born at a gestational age of 40 wk after an uneventful pregnancy and delivery. At birth his weight and length were 3200 g and 51 cm, respectively, and no dysmorphia was observed. Weight and growth development were normal. His medical history was unremarkable until the age of 5.1 yr. At this time, he was hospitalized because of a 3-d history of episodic falls, associated with generalized stiffness, pallor, ocular revulsion, and loss of consciousness lasting less than 1 min. Physical and neurological examinations were unremarkable except for a bilateral Chvostek sign. Blood chemistry revealed a total serum calcium concentration of 1.22 mmol/liter (normal range, 2.2–2.6 mmol/liter) and a phosphorus concentration of 3.42 mmol/liter (normal range, 1.29–1.84 mmol/liter). The serum concentration of magnesium was 0.85 mmol/liter (normal range, 0.74–0.9 mmol/liter) and the total serum protein level was 68 g/liter. Serum concentration of 25-hydroxyvitamin D was 6 µg/liter (normal range, 10–60 µg/liter) and serum concentration of 1,25-dihydroxyvitamin D was 86.4 nmol/liter (normal range, 48–144 nmol/liter). The serum concentration of intact PTH (DiaSorin intact PTH immunoradiometric assay; DiaSorin, Antony, France) was 0.73 pmol/liter (normal range, 1–6.8 pmol/liter). Calciuria was 0.5 mmol/24 h. Ultrasound scanning of the neck showed a normal in-place thyroid and failed to detect parathyroid glands. A skeletal x-ray survey was normal. Ultrasound scanning of the kidneys and heart showed normal images. A brain computed tomography scan showed calcifications of the basal ganglia. Magnetic resonance imaging of the encephalon showed no further abnormality. On the electroencephalogram, a spike-and-wave pattern was observed.
|
). Chromosomal analysis by standard karyotyping of peripheral lymphocytes yielded a normal 46,XY result. Fluorescence in situ hybridization analysis for the DiGeorge critical region at chromosome 22q11 showed two normal hybridization signals. An 8-yr-old sister had presented with a cardiorespiratory arrest, associated with profound hypocalcemia, at the age of 10 months. Total serum calcium concentration was 1.25 mmol/liter, phosphorus concentration was 1.58 mmol/liter, and magnesium concentration was 0.53 mmol/liter. Investigations showed an inappropriately normal serum PTH (DiaSorin intact PTH immunoradiometric assay) level of 2.2 pmol/liter with normal 25-hydroxyvitamin D. Treatment with calcium supplementation and calcitriol was initiated. Compliance to treatment was poor, and the affected sister experienced repeated bouts of severe hypocalcemia contrasting with low but near-normal serum PTH levels and hypocalciuria. At the age of 5 yr, blood chemistry during an episode of hypocalcemia showed a 25-hydroxyvitamin D serum concentration of 5 µg/liter and a 1,25-hydroxyyvitamin D serum concentration of 43.2 nmol/liter. Epilepsy, progressive microcephaly, and severe neurological impairment followed, which were ascribed to the initial anoxic episode.
Blood and urine chemistry analyses in the proband’s 32-yr-old mother, 37-yr-old father, 7-month-old sister, and 9-yr-old brother produced entirely normal results, except for a low serum level of 25-hydroxyyvitamin D at 7 µg/liter in the father.
Peripheral blood was sampled with informed consent from all family members, and genomic DNA was extracted by the phenol-chloroform method (15). All procedures were performed according to the ethical guidelines of Hôpital Erasme.
Mutation analysis
Direct sequencing. Genomic DNA was isolated from the patient’s peripheral leukocytes. Specific primers were designed to amplify and sequence CaSR, preproPTH, and GCMB (primers’ sequences and PCR protocols are available upon request). PCR products were purified and sequenced using the Big Dye Terminator cycle sequencing kit v2 (Applied Biosystems, Foster City, CA), and sequencing products were analyzed on a 3100 Genetic Analyzer sequencing machine (Applied Biosystems). The in silico mutation search was performed using the SeqScape software version 2.0 (Applied Biosystems).
Denaturing HPLC (DHPLC) screening
A primer pair was designed to produce an amplimer containing the Gly63 codon and was used in PCR on the present family and on a series of 104 control subjects, including 51 subjects from the same ethnic background. DHPLC screening was done using a WAVE 3500HT +HSD system (Transgenomic, Cheshire, UK). Aliquots of PCR products were injected on DNASep HT column (Transgenomic). Gradients and column temperatures were used after computation by the Transgenomic Navigator software version 1.5.3, and protocols are available on request.
Functional analysis of the mutant
Plasmids. pCMV5-based expression plasmids for T7-epitope tagged full-length mouse GCMb and its amino-terminal region (GCMb180) have been described before (16, 17). Using these plasmids as a template, the glycine-to-serine mutation at position 63 was introduced into the GCM domain of mouse GCMb/Gcm2 using the QuikChange XL Site-Directed Mutagenesis (Stratagene, La Jolla, CA). All expression cassettes were verified by DNA sequencing.
Tissue culture, transfections, immunocytochemistry, and luciferase assays
HEK 293, HeLa, and DF-1 cells were maintained in DMEM containing 10% fetal calf serum (FCS). HEK 293 cells were transfected for preparation of extracts and metabolic labeling experiments with 10 µg of cytomegalovirus (CMV) expression plasmid per 100-mm plate using the calcium phosphate technique (16). For immunofluorescence studies, HeLa cells were seeded on glass slides and were transfected with 2 µg of CMV expression plasmid per 35-mm plate using Superfect reagent (QIAGEN, Valencia, CA) according to the manufacturer’s instructions. Twenty-four hours after transfection, HeLa cells were fixed in 4% paraformaldehyde, consecutively incubated with a monoclonal antibody directed against the T7-epitope (Novagen, Darmstadt, Germany) and Cy3-coupled goat antimouse antiserum, and mounted in moviol. Immunofluorescence was detected using a Leica (Wetzlar, Germany) inverted microscope (DMIRB) and was documented with a cooled MicroMax charge-coupled device camera (Princeton Instruments, Stanford, CA) and the IPLab spectrum and Adobe Photoshop software packages.
For luciferase assays, chicken DF-1 cells were transfected in 24 well plates with 0.5 µg of CMV expression plasmid and 0.5 µg of 6xgbs luciferase reporter (18) using Superfect reagent (QIAGEN) according to the manufacturer’s instructions. The total amount of plasmid was kept constant using empty CMV vector. Cells were harvested 48 h after transfection, and extracts were assayed for luciferase activity (19).
Metabolic labeling and immunoprecipitation
Forty-eight hours after transfection, HEK 293 cells on 100-mm plates were repeatedly washed with PBS and starved for 1 h in cysteine/methionine-free DMEM containing 1% FCS. Pulse labeling was performed for 1 h by addition of 35S-labeled cysteine and 35S-labeled methionine at a specific activity of 110 µCi/ml. After repeated washes, cells were placed back into regular DMEM supplemented with 10% FCS and were harvested after varying incubation periods in ice cold RIPA buffer (10 mM Tris-HCl, pH 7.6; 150 mM NaCl; 0.2% sodium-deoxycholate; 0.1% sodium dodecyl sulfate; 0.1% Nonidet P-40; 1 mM dithiothreitol; 10 µg/ml aprotinin; and 10 µg/ml leupeptin). For immunoprecipitation of epitope-tagged GCMb/Gcm2-based proteins, the resulting cell lysates were incubated under constant rotation at 4 C with monoclonal T7-tag antibody that was already coupled to bead-immobilized protein A. The precipitates were washed four times with RIPA buffer before subjecting them to SDS-PAGE. Gels were dried and exposed for autoradiography.
Western blots and EMSAs
HEK 293 cells were harvested for extract preparation 48 h after transfection as described (20). For Western blots, extracts were subjected to SDS-PAGE and transferred to nitrocellulose membranes. A monoclonal antibody directed against the T7-epitope (1:3000 dilution; Novagen) served as the primary antibody and horseradish-peroxidase-coupled protein A served as the secondary detection reagent using the enhanced chemiluminescence detection system.
For EMSAs, 0.5 ng 32P-labeled double-stranded oligonucleotide probes (for sequences see Fig. 2D) was incubated with HEK 293 cell extract for 20 min on ice in a 20-µl reaction mixture as described using poly(deoxyinosine-deoxycytidine) as unspecific competitor (20). Samples were loaded onto native 4% polyacrylamide gels and electrophoresed in 0.5x TBE (45 mM Tris/45 mM boric acid/1 mM EDTA, pH 8.3) at 120 V for 1.5 h. Gels were dried and exposed for autoradiography.
|
Mutation identified in GCMB
Direct sequencing of the whole coding region and intronic junctions of the CaSR and preproPTH genes in the proband failed to identify a disease-causing mutation. Several single-nucleotide polymorphisms (SNPs) were identified in the proband during the course of this search, including at least one heterozygous SNP in each of these genes (data not shown). Taken together with parental consanguinity, these results allowed us to reject linkage of the disease to both the CaSR and the preproPTH loci.
The phenotype of the mouse knockout (13) and the clinical picture in the previously reported family (14) indicated GCMB as a candidate gene for isolated hypoparathyroidism in the family. Analysis of three intragenic SNPs revealed that both parents shared a haplotype, heterozygous in both, and homozygous in both siblings although not homozygous in unaffected siblings (data not shown), consistent with linkage of the disease to GCMB. Direct sequencing of the whole coding region and of intronic junctions of GCMB revealed a G-to-A transition in exon 2 (Fig. 1C), predicting the substitution of a glycine residue with a serine at position 63 (G63S). No other change was found in the coding region or in intronic junctions. The mutation was homozygous in affected siblings, heterozygous in both parents, and either heterozygous or absent in the unaffected siblings (Fig. 1A).
We then searched for this exon 2 mutation by DHPLC in genomic DNAs from a series of 104 unrelated control subjects, including 51 from the same ethnic background, and the mutation was absent in all 208 chromosomes tested. Conversely, two heterozygous changes were found in our control population, each of them once only, predicting the substitution of an arginine with a cysteine residue at position 59 (R59Y) in a Moroccan subject and the substitution of an aspartate with an asparagine at position 107 (D107N) in a Western-European subject.
The predicted protein sequence of GCMB consists of 506 amino acid residues, with a DNA-binding GCM domain of approximately 150 residues in its N-terminal part. By comparison with the crystal structure of the murine GCMa bound to its octameric DNA binding site (21), the residue corresponding to Gly63 of GCMB is located in the DNA-binding region but is not conserved in all GCM domains, and it does not establish a direct contact with the cognate octameric DNA recognition element.
Functional activity of the mutant GCMB
To investigate the functional effect of the G63S mutation in GCMB, we introduced the mutation into the mouse GCMb/Gcm2 cDNA and produced mutant GCMb/Gcm2 protein in transfected cells. Both the wild-type as well as the mutant GCMb/Gcm2 protein showed the expected identical molecular mass of 56 kDa (Fig. 2B). and wild-type GCMb/Gcm2 almost exclusively localized to the nucleus (Fig. 2A). Similarly, the G63S mutant was confined to the nuclei of transfected cells in accordance with the fact that the nuclear localization signal of GCMb/Gcm2 (16) was not affected by the G63S mutation.
To analyze whether the substitution of glycine at position 63 by serine had any effect on protein stability, pulse-chase studies were performed on transiently transfected, metabolically labeled cells. As reported (17), wild-type GCMb/Gcm2 is a short-lived protein with a half-life of approximately 30 min (Fig. 2C). This turnover rate was not significantly altered by the G63S mutation, arguing that protein stability is not affected by the mutation. In agreement with this finding, both the wild-type and the mutant GCMb/Gcm2 protein accumulated to comparable steady-state levels in transfected cells (Fig. 2B).
Because the G63S substitution is situated within the GCM-domain, DNA-binding may be affected by the mutation. Therefore, EMSAs were performed with a set of double-stranded oligonucleotide probes that contained the preferred octameric recognition element 5'-ATGCGGGT-3' for all GCM proteins or single-nucleotide variations thereof (Fig. 2D). Whole-cell extracts from HEK 293 cells transfected with an expression plasmid for the wild-type or mutant DNA-binding domain of GCMb/Gcm2 served as protein source. Both proteins bound with high efficiency to the octameric recognition element (Fig. 2E). Changes at positions 2, 3, 6, or 7 of the octamer drastically reduced binding of both the wild-type and the mutant GCMb/Gcm2 protein to DNA. Changes at positions 1, 4, 5, or 8 had significantly less influence on the DNA-binding ability of wild-type or mutant GCMb/Gcm2. Thus, no altered binding specificity was detected for the G63S mutant relative to the wild-type protein, arguing that DNA-binding characteristics were not affected by the mutation.
With the G63S mutant being nuclear and exhibiting unaltered DNA-binding, it was possible to assess its transactivation capacity in transient transfections, in which expression of a luciferase reporter is under control of a GCM-dependent promoter. In the absence of known natural target gene promoters for mammalian GCM proteins, an artificial promoter was used with six GCM binding sites in tandem array (18). In DF-1 cells, wild-type GCMb/Gcm2 induced reporter gene expression approximately 150-fold, whereas GCMa led to a 450-fold induction (Fig. 2F). Although this result confirms that GCMb/Gcm2 is a weaker transactivator than the related GCMa (17), DF-1 cells exhibit the highest activation rates for GCMb/Gcm2 observed so far. When transient transfections were performed in the presence of the G63S mutant instead of wild-type GCMb/Gcm2, GCMb/Gcm2-dependent induction of reporter gene expression was reduced to a residual level of 5%, whereas the expression levels of the constructs were similar. Comparable results were also obtained in other cell lines (data not shown). Thus, at least on this artificial promoter, the G63S mutant exhibited an almost complete loss of its transactivation capacity.
Discussion
We identified a mutation of the GCMB gene in two siblings with isolated hypoparathyroidism who presented in childhood with severe, symptomatic hypocalcemia. The parents were first cousins and had normal parathyroid function, as well as two additional, asymptomatic siblings, consistent with autosomal recessive inheritance (Fig. 1A). The mutation (Fig. 1C), causing a glycine-to-serine substitution at codon 63 of GCMB (G63S), segregated with hypoparathyroidism in the family, as both patients were homozygous, both parents were heterozygous, and neither unaffected sibling was homozygous for the mutation. G63S is a rare mutation: it could not be found in a series of 104 unrelated control subjects, including 51 subjects from the same ethnic origin as the family described here.
G63S is a missense mutation located in the DNA-binding region of the GCM domain of GCMB. The affected glycine residue, however, is not conserved in all GCM domains. Indeed, a serine codon is found at the corresponding position of the murine GCMa (21). Our functional analysis of the mutation constructed in an expression vector unequivocally showed loss of transactivation function in cotransfection experiments using a GCM-responsive promoter fused to a luciferase reporter gene (Fig. 2F), despite unaltered protein expression as assessed by Western blot analysis (Fig. 2B), unaltered stability in pulse-chase labeling (Fig. 2C), unaltered nuclear localization (Fig. 2A), and unaltered DNA-binding specificity in gel-retardation assays (Fig. 2E).
GCMB is a transcription factor specifically expressed in the parathyroid gland (12). In mice, loss of GCMB leads to agenesis of the glands (13). In human, a large intragenic deletion of GCMB consistent with complete loss of function has also been associated with isolated hypoparathyroidism (14). To our knowledge, the present report is the second case of a GCMB defect in man.
Our patients had no detectable parathyroid glands on cervical ultrasound imaging. This negative result excludes significant parathyroid hyperplasia as observed in chronic hypocalcemia, e.g. from chronic renal failure, which argues for a defect of parathyroid embryogenesis and/or proliferation. We conclude that loss of GCMB transactivating function caused by the homozygous G63S mutation led to agenesis or hypoplasia of the parathyroid glands and, hence, to hypoparathyroidism. It must be noted, however, that GCMB function has not yet been studied in postnatal life, so this transcription factor might conceivably play a role, beyond embryogenesis, in PTH production. Loss of such a putative function might therefore provide an alternative, or complementary, explanation to our patients’ phenotype. Both our patients had normal levels of 1,25-dihydroxyvitamin D and low levels of 25-hydroxyvitamin D. This finding is unexpected in the frame of hypoparathyroidism but was also observed in the previously reported family (14).
GCMb/Gcm2-deficient mice and parathyroidectomized mice produce significant residual levels of PTH, secreted by a subset of cells located in the murine thymus. PTH production in these mice can be down-regulated by calcium infusion or serum calcium restoration by vitamin D treatment, consistent with the observation that these cells express the CaSR gene as well as the preproPTH gene (13). By contrast, the previously reported patients with a GCMB deletion had undetectable or very low serum levels of PTH (14). A recent study of human thymus tissue and intrathymic PTH-secreting adenomas provided additional evidence that the normal human thymus does not produce PTH (22), which the authors of the latter study interpreted as consistent with a different embryogenesis of the parathyroid glands in mouse and man. Indeed, rodents have two parathyroid glands, whereas humans have four (23). Contrary to the previously described family, our patients had either low but clearly detectable (proband) or inappropriately normal (affected sister) levels of serum PTH while being severely hypocalcemic. Furthermore, with treatment of hypocalcemia we observed down-regulation of serum PTH in the proband (Fig. 1B). However, the origin of PTH secretion, whether thymic or extrathymic, could not be assessed in our patients. Although very similar to the mouse model, the residual PTH secretion in our patients is likely to result from a very small (5%; see Results) residual activity observed in transfection studies with the G63S mutant GCMB. This minimal residual activity might have allowed some limited parathyroid gland development and/or some preservation of a putative function of GCMB in postnatal PTH production.
It is tempting to correlate the residual levels of PTH in our patients with their relatively later onset of symptoms as compared with the previously described family [5.1 yr of age in the proband and 10 months in the affected sister in the family described in this report, vs. 2 wk, 5 wk, and 2 months of age in the three affected relatives in the previously described family (14)]. This clinical difference probably reflects the very small residual activity of the G63S mutation. Alternatively, it might result from merely stochastic factors, from diet or other environmental factors, or from genetic modifiers not linked to the mutation.
In conclusion, we have identified a second family with isolated hypoparathyroidism and a loss-of-function mutation of the parathyroid-specific transcription factor GCMB. The low but clearly detectable plasma levels of PTH, although very similar to the mouse model of complete loss of GCMB function, likely results from a minimal residual activity of the mutant GCMB in our patients.
Acknowledgments
We thank S. Delbauve for expert technical assistance, C. Heinrichs for discussion, and G. Vassart for support.
Footnotes
This work was supported by Hôpital Erasme Grant No. 10.93.50.01 to M.J.A. and Deutschen Forschungsgemeinschaft (DFG) Center Grant SFB473 to S.H. Support from G. Vassart was not financial.
First Published Online February 22, 2005
Abbreviations: CaSR, Calcium-sensing receptor; CMV, cytomegalovirus; DHPLC, denaturing HPLC; FCS, fetal calf serum; single-nucleotide polymorphism.
Received December 13, 2004.
Accepted February 15, 2005.
References
This article has been cited by other articles:
 |
|
 |
|
  D. Shoback Hypoparathyroidism N. Engl. J. Med., July 24, 2008; 359(4): 391 - 403. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Maret, C. Ding, S. L. Kornfield, and M. A. Levine Analysis of the GCM2 Gene in Isolated Hypoparathyroidism: A Molecular and Biochemical Study J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1426 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. Suva PTH Expression, Not Always where You Think ... J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 396 - 397. [Full Text] [PDF]
|
|
|
|
|
|
J. N. VanHouten, N. Yu, D. Rimm, J. Dotto, A. Arnold, J. J. Wysolmerski, and R. Udelsman Hypercalcemia of Malignancy due to Ectopic Transactivation of the Parathyroid Hormone Gene J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 580 - 583. [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 |
