| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Gene (GNAS1) in Albright Hereditary Osteodystrophy1
Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (D.Y., S.Y., L.S.W.), Bethesda, Maryland 20892; Children’s Hospital, University of Wurzburg (V.S.), D-97080 Wurzburg, Germany; Children’s Hospital, University of Lubeck (K.K.), D-23538 Lubeck, Germany; and the Department of Pediatrics, University of New Mexico (C.L.C.), Albuquerque, New Mexico 87131
Address all correspondence and requests for reprints to: Dr. Lee S. Weinstein, Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Building 10, Room 8C101, Bethesda, Maryland 20892-1752. E-mail: leew{at}amb.niddk.nih.gov
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
, the -subunit of the G
protein that couples receptors to adenylyl cyclase stimulation, and in
a number of cases heterozygous loss of function mutations within the
Gs gene (GNAS1) have been identified.
Using PCR with the attachment of a high melting domain (GC-clamp) and
temperature gradient gel electrophoresis, two novel heterozygous
frameshift mutations within GNAS1 were found in two AHO
kindreds. In one kindred all affected members (both PHP Ia and PPHP)
had a heterozygous 2-bp deletion in exon 8, whereas in the second
kindred a heterozygous 2-bp deletion in exon 4 was identified in all
affected members examined. In both cases the frameshift encoded a
premature termination codon several codons downstream of the deletion.
In the latter kindred affected members were previously shown to have
decreased levels of GNAS1 messenger ribonucleic acid
expression. These results further underscore the genetic heterogeneity
of AHO and provides further evidence that PHP Ia and PPHP are two
clinical presentations of a common genetic defect. Serial measurements
of thyroid function in members of kindred 1 indicate that TSH
resistance progresses with age and becomes more evident after the first
year of life. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Heterotrimeric G proteins, a family of proteins that transmit signals
from heptahelical receptors to activate intracellular effectors, each
consist of -, ß-, and 
-subunits, each the product of separate
genes (5, 6, 7). Each G protein is defined by its specific -subunit,
which confers specificity to both receptor and effector interactions.
Steady state levels of Gs messenger ribonucleic acid
(mRNA) (8, 9) and protein (10) are reduced in the affected members of
most, but not all, AHO kindreds. The human Gs gene
(GNAS1) contains 13 exons encoding Gs (11)
and is located at 20q13 (12). Several heterozygous loss of function
mutations within GNAS1 have been identified in AHO patients
(with PHP Ia and PPHP). Except for a 4-bp deletion in exon 7, which has
been found in multiple independent kindreds (13), each AHO kindred
appears to have a unique GNAS1 mutation. In contrast,
activating GNAS1 mutations are present in patients with
McCune-Albright syndrome (14) and in a subset of GH-secreting pituitary
tumors and thyroid neoplasms (15). Using PCR and temperature gradient
gel electrophoresis (TGGE), we identified two novel GNAS1
frameshift mutations in two independent AHO kindreds. In each kindred,
the mutation was identified in all affected family members who were
tested (both PHP Ia and PPHP). Serial thyroid tests in affected members
of one kindred indicate that TSH resistance increases progressively
after the first year of life.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Kindred 1 (Fig. 1
and Table 1)
Proband (III-1).
The female proband was the product of a
normal term delivery; she weighed 7 lb, 2 oz at birth, with a length of
19 in. At 2 days of age, the newborn thyroid screen was slightly
abnormal, with a T4 of 7.4 µg/dL (normal, 8–24) and TSH
below 20 µU/mL (normal, <20). At age 2 months a repeat screen
was normal [T4, 7.4 (normal, 6–18); TSH, <20]. By 19
months she was obese, with global developmental delays; thyroid test
results included T4 of 4.9, TSH of 4.0, T3
resin uptake of 26.6%, and free T4 index of 1.3. These
results were interpreted as suggestive of secondary hypothyroidism. One
month later (20 months) thyroid tests revealed T4 of 5.8
(normal, 6.8–13.5), TSH of 5.1 (normal, 0.5–4.8), and free
T4 of 0.8 (normal, 0.8–2.3 for pubertal children). A
thyroid scan revealed a normal appearing thyroid with an uptake of 7%
at 6 h, which was low. During a TRH stimulation test, baseline TSH
was 8.1, and peak TSH was 30. She was begun on thyroid hormone
replacement, and all subsequent thyroid tests have remained normal. At
2 yr, 4 months of age, the patient was severely obese, and chemistries
revealed calcium of 10.0 mg/dL, phosphorus of 7.7 mg/dL, and PTH [by
immunoradiometric assay (IRMA)] of 145 pg/mL (Nichols Institute Diagnostics, San Juan Capistrano, CA; normal, 10–65). At 4 yr,
10 months of age, she became mildly hypocalcemic (calcium, 7.4;
phosphorus, 8.6) and was started on oral calcium and calcitriol. At 4
yr, 4 months of age, she was noted to be obese with weight above the
98th percentile and on exam was noted to have a rounded face and
brachydachtyly of the fifth proximal phalanges bilaterally. She also
has receptive speech deficits. Based upon the endocrine and clinical
manifestations, a presumptive diagnosis of PHP Ia was made.
|
|
Patient III-3.
This male patient was the product of a normal
term delivery; he weighed 6 lb, 19 oz at birth, with a length of 19 in.
Initial newborn thyroid screen at 1 day of age was abnormal, with
T4 of 3.4 and TSH of 34.6. At age 18 days,
T4 was 7.6, and TSH was 18.6. Over the next 6
months thyroid functions normalized. At 2 months,
T4 was 6.88 µg/dL (normal, 4.50–12.00), and
TSH was 5.94 µIU/mL (normal, 0.46–4.98); at 6 months,
T4 was 6.48, and TSH was 4.67. At 9 months,
T4 decreased to 4.76, TSH increased to 7.34, and
thyroid hormone replacement was begun. By this time he was obese, with
a length at the second percentile and delay in gross motor development.
Calcium levels were normal at 18 days, 2 months, 6 months, and 9 months
(9.0, 10.1, 9.2, and 10.1, respectively; normal, 8.4–10.2 mg/dL).
Simultaneous phosphorus levels were 6.3, 5.6, 5.4, and 6.1,
respectively (normal, 4–7 mg/dL). A low phosphate diet was instituted
at 20 months, when calcium was 9.7 and phosphorus was 7.4. At 4 yr, 9
months of age, PTH resistance was evident, with a serum PTH of 634 (by
IRMA) and mild hypocalcemia (calcium, 8.4; lower limit of normal, 8.8).
This patient has PHP Ia based upon the clinical presentation and
evidence of hypothyroidism and PTH resistance.
Parents II-1 and II-2.
The patients’ mother (II-1) is 5
ft tall, and their father (II-2) is 6 ft tall. Both have normal build
and intelligence and normal serum calcium, phosphorus, and PTH. The
mother has large palpable sc calcium deposits as well as sc
calcifications seen on hand x-rays and mammography. Based upon this she
is presumed to have PPHP. The maternal grandfather (I-2) is reportedly
5 ft, 4 in. tall and obese, with calcium deposits around his knees and
with broad feet and short toes. He is receiving thyroid hormone
replacement, although the nature of his thyroid disease is unknown, and
studies of mineral metabolism are unavailable. Based upon this
information we presume that he has AHO, although it is unclear whether
he has PHP Ia or PPHP. The maternal grandmother (I-1) is unaffected. A
maternal aunt (not shown) reportedly is 5 ft tall, weighs 160 lb, has
broad feet with short toes, and has sc calcifications noted on a
mammogram, making it likely that she also has AHO. She was unavailable
for genetic analysis.
Kindred (Fig. 3.) The clinical and biochemical data for
members of this German kindred have been previously reported (16, 17).
Patients III-1 (PPHP), III-4 (PPHP), IV-1 (unaffected), and IV-2 (PHP
Ia) correspond to patients with the same numbering as reported in Ref.
16 . Patient IV-3 has been previously reported as patient IV-1 in Ref.
17 . This patient has been reported as having PHP Ia based upon a low
cAMP response to administered PTH analogue and a slightly exaggerated
TSH response to TRH (17). However, up until the age of 5 yr, 5 months,
serum calcium, phosphorus, PTH, free T4, and free
T3 have remained normal, whereas TSH has risen slightly to
4.8 (normal, 0.3–4). Therefore, the diagnosis of PHP Ia vs. PPHP is
not yet clearly established. Patient IV-4, the younger sister of
patient IV-3, developed osteoma cutis on the trunk and extremities in
the first months of life. At the age of 2 yr, a modified
Ellsworth-Howard test was performed (18). After injection of human
PTH-(1–34) (Parathar, Rorer Pharmaceuticals, Collegeville, PA)
plasma and urinary cAMP responses were blunted (plasma cAMP: baseline,
27.2 nmol/L; after 5 min, 35.8; after 10 min, 31.2; normal, >60 nmol/L
after 5 or 10 min; urinary cAMP: baseline, 6.3 nmol/mg creatinine;
after 60 min, 20 nmol/mg creatinine; normal, >60). However, up until 4
yr 1 month of age, serum calcium, phosphorus, alkaline phosphatase,
PTH, free T4, free T3, and TSH were
all normal. Therefore, the diagnosis of PHP Ia vs. PPHP in
this patient is not clearly established. Gs mRNA
expression was previously shown to be decreased in four affected
members (three PPHP and one PHP Ia) of this kindred (16). Informed
consent was obtained from all individuals examined in this study.
|
DNA was isolated from blood, and the coding regions of GNAS1 were amplified in 100-µL PCR reactions containing deoxynucleotide triphosphates (200 µmol/L each), upstream and downstream oligonucleotide primers (0.5 µmol/L each), 0.01% (wt/vol) gelatin, 50 mmol/L KCl, 10 mmol/L Tris-HCl (pH 8.3), 1.5 mmol/L MgCl2, and 2.5 U Taq polymerase (Perkin Elmer Corp./Cetus, Emeryville, CA). To amplify genomic fragments containing exon 4, the primers were 5'-GCTACAAGACAGACAGCACAAG-3' (upstream primer) and 5'-GCGGCCGCCCGTCCCGCCGCCCCCGCCCCGCCGCGGCCGCCCAGTACTCCT-AACTGACATGG-3' (downstream primer, GC clamp is underlined). Primer sequences used to amplify exon 8 have been previously reported (14). Samples were denatured at 94 C for 5 min, followed by 30 cycles of annealing (58 C for 45 s), extension (72 C for 1 min), and denaturation (94 C for 1 min) and one final cycle with a 3-min extension. PCR amplification was confirmed by running products in 5% nondenaturing acrylamide gels, and then the PCR products were analyzed by TGGE as previously described (19), using a commercial apparatus (Diagen, Dusseldorf, Germany). PCR products were heated to 95 C for 3 min, immediately placed on ice for 20 min, and then left at room temperature for several hours. Samples were placed in sample buffer [final concentration, 20 mmol/L MOPS-NaOH (pH 8.0), 5 mmol/L EDTA, 0.1% bromophenol blue, and 2% glycerol], and 4 µl of each PCR product were analyzed in 5.5% acrylamide (70:1, acrylamide-bisacrylamide) gels containing 20 mmol/L MOPS-NaOH (pH 8.0), 1 mmol/L ethylenediamine tetraacetate, and 2% glycerol with a running buffer of 20 mmol/L MOPS-NaOH (pH 8.0), and 1 mmol/L ethylenediamine tetraacetate. To analyze exon 4 fragments samples were run in a gradient of 30–65 C for a total of 63 min with the wells located 12 cm from the bottom of the gradient, and then the gels were silver stained. To analyze exon 8 fragments, samples were run in a gradient of 30–65 C for a total of 70 min with the wells located 11 cm from the bottom of the gradient. To confirm the presence or absence of the exon 8 mutation in family members of kindred 1, samples were denatured, renatured, and electrophoresed on 6% acrylamide gel, and the gel was silver stained. For sequencing, 5 µL of the original PCR products were reamplified using the same conditions, except that only one primer was included to generate single stranded template. The reaction products were concentrated by filtration using Centricon-100 (Amicon, Beverly, MA) and sequenced with the opposite PCR primer using Sequenase (U.S. Biochemical Corp., Cleveland, OH).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
).
As heteroduplexes containing small heterozygous deletions often migrate
more slowly on nondenaturing acrylamide gels, amplified PCR products
from members of kindred 1 were separated on 6% acrylamide gels to
determine which family members have the mutation (Fig. 1
). Two extra slower migrating bands were
present in amplified DNA from affected family members (I-2, II-1, and
III-1,2,3), but not in those from unaffected family members (I-1 and
II-2) or a normal control. Therefore, in this kindred a heterozygous
2-bp deletion in exon 8 of GNAS1 is associated with the
development of AHO (both PPHP and PHP Ia).
|
). Mutations were not identified in
affected patients within exons 2 and 3 or exons 5–13. Direct sequence
of exon 4 PCR products from affected family members revealed a
heterozygous 2-bp deletion within exon 4, which results in a premature
stop codon within the mutant allele (Fig. 4).
Therefore, in this kindred a heterozygous 2-bp deletion in exon 4 of
GNAS1 is associated with the development of AHO (both PPHP
and PHP Ia).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
protein is produced, it would be
missing most of the guanine nucleotide-binding pocket as well as
regions that are critical for receptor and effector interaction.
Moreover, a frameshift mutation in exon 10, which is downstream of this
frameshift, has been shown to be associated with AHO (20). In both
kindreds, identical mutations were present in both PPHP and PHP Ia
patients, providing further evidence that PHP Ia and PPHP are two
distinct clinical presentations resulting from a common genetic
defect.
The molecular mechanisms by which a heterozygous null mutation could
lead to multihormone resistance in some AHO patients but not in others
has remained unclear. A retrospective analysis of published cases
revealed that maternal transmission of AHO leads to offspring with PHP
Ia, whereas paternal transmission leads to offspring with PPHP (21),
suggesting that GNAS1 might be an imprinted gene. Genomic
imprinting is an epigenetic phenomenon affecting a small number of
autosomal genes by which one allele (paternal or maternal) has partial
or total loss of expression (22). If the GNAS1 paternal
allele is imprinted (poorly expressed) in hormone target tissues, then
maternal transmission of a GNAS1 mutation would lead to
markedly decreased Gs expression (and hormone
resistance) due to imprinting of the paternal allele and mutation of
the maternal allele. In contrast, paternal transmission of a
GNAS1 mutation would have little effect on Gs
expression, since the paternal allele is normally silent due to an
imprint. Observations in mice with heterozygous null mutations of the
homologous gene (Gnas) show that the Gnas
paternal allele is imprinted in renal proximal tubules (the major renal
site of PTH action) and that PTH resistance results from maternal, but
not paternal, transmission of the Gnas null mutation
(23).
The observations in kindred 1 are consistent with the imprinting model, as paternal transmission of a GNAS1 mutation from patient I-2 to II-1 resulted in the expression of PPHP in patient II-1, whereas maternal transmission from patient II-1 to patients III-1, -2, and -3 resulted in the expression of PHP Ia in all three offspring. In kindred 2, paternal transmission of a GNAS1 mutation from generation II resulted in three offspring affected by PPHP in generation III (patients III-1 and -4 and a third female; see Ref. 17), whereas maternal transmission from patient III-4 resulted in PHP Ia in patient IV-2, all consistent with the imprinting model. Paternal transmission of the GNAS1 mutation from patient III-1 resulted in two affected offspring who had a blunted cAMP response to administered PTH, and one has been previously reported as having normocalcemic PHP Ia (17). However, both offspring continue to have normal serum calcium, phosphorus, PTH, TSH, and thyroid hormone levels, so that the diagnosis of PPHP vs. PHP Ia in these patients is not well defined.
It remains to be rigorously proven that the GNAS1 transcript
encoding Gs is imprinted in a tissue-specific manner,
probably because tissues in which Gs is likely to be
imprinted (such as the renal proximal tubules) have not been examined
(24). However, two alternative GNAS1 transcripts formed by
splicing of alternative upstream exons to exon 2 have been shown to be
oppositely imprinted in humans. Transcripts encoding XLs, a
Gs isoform with a long amino-terminal extension
localized to Golgi, are expressed only from the paternal allele (25).
The mutations described in this paper should disrupt XLs expression
only when present in the paternal allele; therefore, XLs would be
unaffected in PHP Ia patients and absent in PPHP patients. Given that
PPHP patients are generally more mildly affected and have no phenotypic
manifestations not also present in PHP Ia patients, it seems unlikely
that XLs plays a critical role in human development or physiology.
It is also possible that a protein similar to XLs is able to
compensate for its absence. Another alternative transcript encoding
NESP55, a chromogranin-like neurosecretory protein, is expressed only
from the maternal allele (26). Unlike XLs, the NESP55 coding region
is entirely within the upstream exon with exons 2–13 located within
the 3'-untranslated region. As the mutations described in this paper
disrupt mRNA expression, they probably disrupt NESP55 expression in PHP
Ia patients (when present in the maternal allele), but should have no
effect on NESP55 expression in PPHP patients (when present on the
paternal allele). However two lines of evidence make it unlikely that
loss of NESP55 plays a major role in the pathogenesis of PHP Ia: 1) the
major phenotypic difference between PHP Ia and PPHP, namely
multihormone resistance, is almost certainly caused by defects in
Gs, which is required for hormone action; and 2)
missense mutations in GNAS1 that should have no effect on
NESP55 expression lead to a phenotype of similar severity (27, 28).
Consistent with previous observations (29, 30, 31), the development of PTH
resistance in patients III-1 and -2 of kindred 1 was progressive over
the first years of life, with hypocalcemia being preceded by
hyperphosphatemia and elevations in circulating PTH. In all three
affected siblings, the newborn thyroid screen was reported as abnormal,
with elevated TSH in patients III-2 and -3. This is consistent with
previous reports that patients with PHP Ia present with abnormal
thyroid tests at birth (32, 33, 34). Interestingly, subsequent thyroid
studies within the first year (including TSH) were normal in all three
siblings, and evidence of hypothyroidism with elevated TSH redeveloped
from 9–20 months after birth. This may indicate that in these PHP Ia
patients, TSH resistance, like PTH resistance, progressed over time
within the first 1–2 yr of life. Immediately after birth, there is
normally a transient surge in TSH levels, presumed to be a response to
the rapid decrease in extracorporeal temperature and believed to be
important for the induction of nonshivering thermogenesis. The
exaggerated surge in TSH levels at birth present in these patients may
indicate that within the first year of life, PHP Ia patients have
subtle TSH resistance (perhaps due to partial Gs
deficiency) that can only be detected during maximal stimulation (the
TSH surge during parturition). Although TSH levels at birth may well
discriminate patients with PPHP vs. PHP Ia, the observations
in these three patients suggest that PHP Ia patients may have normal
thyroid tests and TSH in the first year of life, and therefore,
continued monitoring of thyroid status in early childhood is warranted
in AHO patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received February 18, 1999.
Revised May 14, 1999.
Accepted May 25, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-subunit of the guanine nucleotide-binding protein
Gs as the molecular basis for Albright hereditary osteodystrophy. Proc
Natl Acad Sci USA. 85:617–621.-subunit of the stimulatory G-protein of adenylyl cyclase in
patients with Albright’s hereditary osteodystrophy. J Clin
Endocrinol Metab. 71:1208–1214. gene. Proc Natl Acad
Sci USA. 85:2081–2085. subunit gene (GNAS1) to the
distal long arm of chromosome 20 using a polymorphism detected by
denaturing gradient gel electrophoresis. Genomics. 9:782–783.[CrossRef][Medline]
gene (GNAS1) in
patients with Albright hereditary osteodystrophy. Hum Mol Genet. 4:2001–2002.-subunit gene in Albright hereditary
osteodystrophy detected by denaturing gradient gel electrophoresis. Proc Natl Acad Sci USA. 87:8287–8290.-subunit (Gs) knockout mice is due to
tissue-specific imprinting of the Gs gene. Proc Natl
Acad Sci USA. 95:8715–8720. in a patient with pseudohypoparathyroidism. Mol
Endocrinol. 11:1718–1727.-contact region of Gs impairs receptor
stimulation. J Biol Chem. 271:19653–19655.This article has been cited by other articles:
 |
|
 |
|
  A.-S. Balavoine, M. Ladsous, F.-L. Velayoudom, V. Vlaeminck, C. Cardot-Bauters, M. d'Herbomez, and J.-L. Wemeau Hypothyroidism in patients with pseudohypoparathyroidism type Ia: clinical evidence of resistance to TSH and TRH Eur. J. Endocrinol., October 1, 2008; 159(4): 431 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Plagge, G. Kelsey, and E. L Germain-Lee Physiological functions of the imprinted Gnas locus and its protein variants G{alpha}s and XL{alpha}s in human and mouse J. Endocrinol., February 1, 2008; 196(2): 193 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. N. Long, S. McGuire, M. A. Levine, L. S. Weinstein, and E. L. Germain-Lee Body Mass Index Differences in Pseudohypoparathyroidism Type 1a Versus Pseudopseudohypoparathyroidism May Implicate Paternal Imprinting of G{alpha}s in the Development of Human Obesity J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 1073 - 1079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S M Park and V K K Chatterjee Genetics of congenital hypothyroidism J. Med. Genet., May 1, 2005; 42(5): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. G Riepe, W. Ahrens, N. Krone, R. Folster-Holst, J. Brasch, W. G Sippell, O. Hiort, and C.-J. Partsch Early manifestation of calcinosis cutis in pseudohypoparathyroidism type Ia associated with a novel mutation in the GNAS gene Eur. J. Endocrinol., April 1, 2005; 152(4): 515 - 519. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Linglart, J. C. Carel, M. Garabedian, T. Le, E. Mallet, and M. L. Kottler GNAS1 Lesions in Pseudohypoparathyroidism Ia and Ic: Genotype Phenotype Relationship and Evidence of the Maternal Transmission of the Hormonal Resistance J. Clin. Endocrinol. Metab., January 1, 2002; 87(1): 189 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. S. Weinstein, S. Yu, D. R. Warner, and J. Liu Endocrine Manifestations of Stimulatory G Protein {alpha}-Subunit Mutations and the Role of Genomic Imprinting Endocr. Rev., October 1, 2001; 22(5): 675 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Ahrens, O. Hiort, P. Staedt, T. Kirschner, C. Marschke, and K. Kruse Analysis of the GNAS1 Gene in Albright's Hereditary Osteodystrophy J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4630 - 4634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Mantovani, R. Romoli, G. Weber, V. Brunelli, E. De Menis, S. Beccio, P. Beck-Peccoz, and A. Spada Mutational Analysis of GNAS1 in Patients with Pseudohypoparathyroidism: Identification of Two Novel Mutations J. Clin. Endocrinol. Metab., November 1, 2000; 85(11): 4243 - 4248. [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 |
