| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Neuroendocrine Unit, Departments of Medicine and Neurosurgery (B.S.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Anne Klibanski, M.D., Neuroendocrine Unit, Massachusetts General Hospital, 55 Fruit Street, BUL457B, Boston, Massachusetts 02114-2696.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Numerous studies describing the cloning of the activin family of cell surface type I and type II receptors have revealed a novel class of receptor complexes with transmembrane serine/threonine (Ser/Thr) kinase activity and varying affinities for activin and/or TGFß (7). A defined series of noncovalent molecular interactions between type I and type II receptors give rise to active signaling receptor complexes (10). Human type I receptors, Alk1–5 (activin-like receptor kinases), primarily function as intracellular signal transducers, whereas binding specificity for TGFß or activin resides in the type II receptors, TßRII (TGFß) or ActRIIA and ActRIIB (activin) (10). Therefore, receptor complexes comprised of both receptor classes are required for functional activin or TGFß signaling. For both activin and TGFß signaling, type I receptor is activated by ligand-bound type II receptor cross-phosphorylation at intracellular serine and threonine residues. This activated complex then stimulates specific intracellular signaling cascades via type I receptor-specific Ser/Thr kinase activity.(10) Although receptors for TGFß-related cytokines such as activin have been identified in human neoplastic pituitary tissue, their potential role as signal transducers in tumor pathogenesis remains unknown. To investigate whether activin/TGFß cell surface receptors may act as tumor suppressors in human pituitary tumors, we used single stranded conformational polymorphism (SSCP) analysis targeting the intracellular kinase domain regions of the human Alk1–5 type I receptors to scan for inactivating somatic mutations in human pituitary adenomas. SSCP analysis of the entire coding region of two activin and a single TGFß type II receptors was also performed in 64 human pituitary adenoma specimens to investigate whether these neoplasms harbor mutations in these receptors.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Histologically confirmed pituitary tumors were obtained from 64
patients who underwent transsphenoidal surgery and were investigated
for activin/TGFß receptor mutations. Patients (n = 8) with
GH-secreting macroadenomas (diameter, 
1 cm) ranged in age from 26–78
yr (median, 36 yr), and all had clinical and biochemical evidence of
acromegaly, with elevated serum levels of GH and IGF-I. Somatotroph
adenomas underwent immunohistochemical staining for GH, PRL, and
-subunit. All somatotroph tumors exhibited strong
immunohistochemical staining for GH, and tumors 4, 6, and 7 also showed
extensive PRL staining. Somatotroph adenoma 2 exhibited extensive
immunohistochemical staining for free glycoprotein hormone -subunit.
Patients with macroprolactinomas (n = 9) ranged in age from 17–60
yr (median, 35 yr), and all had elevated serum PRL levels and
immunohistochemical staining consistent with the diagnosis. Patients
with corticotroph tumors (n = 8; four microadenomas and four
macroadenomas) ranged in age from 20–54 yr (median, 46 yr). All had
dexamethasone suppression testing and petrosal catheterization results
consistent with Cushing’s disease. All corticotroph adenoma tissue
samples exhibited positive immunohistochemical staining for ACTH,
whereas only corticotroph tumor 3 exhibited scattered staining for PRL.
All GH-, PRL-, and ACTH-secreting tumors were negative for LHß mRNA
expression in RT-PCR assays, whereas all control normal pituitaries
were positive for LHß mRNA. These data indicate that these surgical
specimens were not contaminated with significant numbers of normal
pituitary cells (data not shown). Serum gonadotropins and free
-subunit levels were within the normal range for all 39 patients
with clinically nonfunctioning pituitary tumors, a tumor type that
consists largely of neoplastic gonadotroph cells. None of the patients
had clinical evidence of GH or ACTH hypersecretion, and serum PRL was
less than 100 µg/L. Patients ranged in age from 31–79 yr (median, 56
yr). Immunocytochemical staining was performed on pituitary tumor
tissue from each patient using specific antibodies for LHß, TSHß,
FSHß, -subunit, GH, PRL, and ACTH. Immunostaining for one or more
glycoprotein hormone subunits was positive in five of the
nonfunctioning tumors. All nonfunctioning adenomas were negative for
Pit-1 mRNA expression in RT-PCR assays, whereas the control normal
pituitary complementary DNA (cDNA) library was positive for Pit-1 mRNA.
These data indicate that these surgical specimens were not contaminated
with significant numbers of normal pituitary cells and were highly
likely to be of gonadotropic origin (data not shown). The control
normal pituitary tissues used in these studies were obtained 3–6 h
postmortem and snap-frozen in liquid nitrogen (National Disease
Research Interchange, Philadelphia, PA).
Pituitary adenomas were obtained in phosphate-buffered saline after transsphenoidal surgery and were frozen in liquid nitrogen. Normal human pituitary samples were obtained at autopsy. Total RNA from adenomatous and normal tissue was obtained using the guanidinium isothiocyanate/phenol/chloroform extraction technique, followed by enzymatic digestion of genomic DNA with 1 U RQ1 deoxyribonuclease/µg total nucleic acid at 37 C for 1 h (Promega Corp., Madison, WI). Total RNA was then reextracted with phenol/chloroform, ethanol precipitated, and quantitated by UV spectrophotometry. The normal human pituitary cDNA library (CLONTECH Laboratories, Inc., San Diego, CA) represents pooled mRNA from nine men and women, aged 15–83 yr, from tissue obtained 1–3 h postmortem.
RT-PCR of activin/TGFß receptors
For RNA samples, 1 µg total RNA was reverse transcribed in 50 mmol/L Tris-HCl (pH 8.3), 5 mmol/L KCl, 5 mmol/L MgCl2, 5 mmol/L dithiothreitol, 0.25 mmol/L spermidine, 200 mmol/L deoxy (d)-NTPs using avian myeloblastosis virus (AMV) reverse transcriptase (12 U/reaction), and random hexamers (1 µg/reaction) as first strand cDNA primers (Promega Corp.). RT reactions were carried out at 25 C for 10 min (random hexamer annealing), followed by a 10-min elongation step at 42 C, and AMV reverse transcriptase heat inactivation at 99 C for 5 min.
PCR oligonucleotide primers for activin receptors were tested for their ability to amplify 100 ng human leukocyte genomic DNA to rule out contaminating DNA as a source of signal in RT-PCR reactions. With the exception of ALK3, all primer pairs failed to amplify genomic DNA, indicating that they spanned at least one intron and were not effective in amplifying potential pseudogenes. Control reactions without AMV reverse transcriptase were also carried out with each RNA sample to exclude genomic DNA contamination as a source of amplified signal. All tumors and normal tissues were negative for receptor signal in reactions without RT (data not shown). All PCR oligonucleotide primers were compared to GenBank sequence libraries to assure specificity. The primers used for PCR/SSCP analysis are listed along with the 5'-position of each nucleotide based on the National Center for Biotechnology Information accession number of each cloned receptor. + and - indicate sense and antisense strands, respectively. The following primers were used: ActRII (11) (M93415): 1a [+/nucleotide (nt) 173], aat ggg agc tgc tgc aaa gtt; 1b (-/nt 461), atc ttt ttt ttc tgt cgg gac; 2a (+/nt 407), ttg gct gga tga tat caa ctg; 2b (-/nt 630), tcc ccg caa tta aca taa gtg; 3a (+/nt 570), aag cca ccc tat tac aac atc; 3b (-/nt 775), ccc ctt gct ttc act tct aat; 4a (+/nt 715), ccc cac ctt ctc cat tac tag; 4b (-/nt 929), tgc acc aat gaa ctg taa tat; 5a (+/nt 858), tgg caa aat gaa tac gaa gtc ta; 5b (-/nt 1064), agc cat ggt ttc tgc aat atg ac; 6a (+/nt 1009), tta agg cta atg tgg tct ctt g; 6b (-/nt 1243), tgg gta tcg cct gca gac tt; 7a (+/nt 1196), tgg gtt ggc ctt aaa att tga; 7b (-/nt 1469), aac aac ttc ctg cat gtc ttc; 8a (+/nt 1427), aat tgg cca gca tcc atc tct; 8b (-/nt 1704), att ctt tgg gag gaa agt caa; ActRIIB (12) (X77533): 1a (+/nt 227), aag ggc tgc tgg cta gat gac; 1b (-/nt 444), ccg atg ggc agc agt gag tag; 2a (+/nt 330), gca acg agc gct tca ctc att; 2b (-/nt 536), gcc cag ggt cct cat gga tgt; 3a (+/nt 465), tgc tgg cct ttt gga tgt acc; 3b (-/nt 680), tct gcc acg act gct tgt c; 4a (+/nt 621), agg ccc agc tca tga atg act; 4b (-/nt 877), agg cct cgt gac atc gtc tct; 5a (+/nt 852), cat gga acg aac tgt gtc atg ta; 5b (-/nt 1121), ctc cct cga gca cct cag gag; 6a (+/nt 1178), ggt agg cac gag acg gta ca; 6b (-/nt 1321), ggt ggg cct cat ctt ctt gt; 7a (+/nt 1237), gct gcc ctt tga gga aga gat; 7b (-/nt 1477), gtc cga ggt agt gcc gtt gac; TBRII (13) (M85079): 1a (+/nt 356), ggg cct gtg gcc gct gca cat; 1b (-/nt 701), ggc ttt ttt ttt tcc ttc ata; 2a (+/nt 652), tcc cct acc atg act tta ttc; 2b (-/nt 927), gcc ggt ttc cca ggt tga act; 3a (+/nt 902), ccg cgt taa ccg gca gca gaa; 3b (-/nt 1182), tgt ctt cca aga ggc ata ctc; 4a (+/nt 1167), atc ttt ccc tat gag gag tat; 4b (-/nt 1428), ggg cct ccc aca tgg agt gtg; 5a (+/nt 1404), ggg att gct cac ctc cac agt; 5b (-/nt 1715), act tct ccc act gca tta cag; 6a (+/nt 1692), gtg ctc tgg gaa atg aca tct; 6b (-/nt 2041), tgc cca gcc tgc ccc ata aga; Alk1 (14) (Z22533): 1a (+/nt 173), ggg gac ctc ctg gac agt gac; 1b (-/nt 461), ccc gga acc agg act gtt cat; 2a (+/nt 173), tct cag gcc tag ctc aga tga; 2b (-/nt 461) cag ccc cct gca ggc aga aag; Alk2-(14) (Z22534): 1a (+/nt 173), gac gtg gag tat ggc act atc; 1b (-/nt 461), gtt tcc ctg aac cat gac ttc; 2a (+/nt 173), tac cca aca gat ggt tct cag; 2b (-/nt 461), aat ttg tcg agg gaa tta tca; Alk3-(14) (Z22535): 1a (+/nt 173), aac agg atg aag cat tta ttc; 1b (-/nt 461), tgg ctt ctt cag tgg taa aga; 2a (+/nt 173), tcg gtg gaa cag tga tga atg; 2b (-/nt 461), cat gcc atg ggt aaa aac agt; Alk4-(14) (Z22536): 1a (+/nt 173), acg ctc cag gat ctt gtc tac; 1b (-/nt 461), gca tga ccg tct ggt ata tct; 2a (+/nt 173), ctg cca tat tac gac tta gtg; 2b (-/nt 461), gag gga gca gtt aga tct tca; Alk5 (15) (L11695): 1a (+/nt 173), gat cgc cct ttt att tca gag; 1b (-/nt 461), cac gaa cgt tct tct cta gag; 2a (+/nt 173), ggc caa ata tcc caa aca gat; 2b (-/nt 461), gag ttc agg caa agc tgt aga.
All PCR amplifications used 50 ng first strand cDNA from a single RT
reaction. To control for potential nonspecific RNA degradation in
pituitary tumor and normal pituitary RNA preparations, samples were
tested for the presence of glyceraldehyde-3-phosphate dehydrogenase
mRNA by PCR, and all were positive. For each receptor studied, all
pituitary tumors and normal pituitary samples then were amplified
simultaneously for each receptor using a common PCR reaction mixture to
ensure that any differences in receptor amplification between samples
were not due to variability in PCR reaction conditions. PCR was
carried out in 50 mmol/L KCl, 10 mmol/L Tris-HCl (pH 9.0), 3.5 mmol/L
MgCl2, 0.1% Triton X-100, 40 mmol/L dNTPs, and 0.125 U
TaqI polymerase (Promega Corp.) in a final
volume of 25 µL. To control for extraneous contaminating genomic DNA
or cDNA in experimental reagents, a tube containing the PCR reaction
mixture with no template was included in each receptor amplification
experiment and was negative for all experiments. PCR products were
visualized by incorporation of [-32P]dCTP (100
nCi/reaction) in the PCR reactions. PCR primers (12.5 pmol) were
used for each reaction, and amplifications were carried out in a
thermocycler (MJ Research, Inc., Watertown, MA). All
reactions were amplified for 30 cycles (1 min at 94 C, 1 min at
optimized annealing temperature, and 1 min, 15 sec at 72 C). Optimal
annealing temperatures and thermal stabilities for each primer set were
calculated using Oligo software (National Biosciences, Minneapolis,
MN).
SSCP and sequence analysis
Receptor-specific primers were end labeled with
[
-32P]ATP followed by PCR amplification of tumor first
strand cDNA at the appropriate annealing temperature. After PCR
amplification, products were fractionated on 6% acrylamide, 0.16%
bisacrylamide (Protogel, National Diagnostics, Atlanta, GA) with 8%
glycerol under nondenaturing conditions at 15 watts. Samples that
displayed mobility shifts were ligated into pGEM-T plasmid vectors
(Promega Corp.), and dideoxy sequencing of both sense and
antisense strands using more than six transformants was performed with
Sequenase (U.S. Biochemical Corp., Cleveland, OH). Base
changes were confirmed when at least two plasmids had the same point
mutation and were analyzed a second time by RT-PCR/SSCP from native
tumor RNA. Candidate mutations were also analyzed from patient
leukocyte DNA to evaluate the tumor specificity of candidate point
mutations detected in tumor samples.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Figure 1
displays the targeted SSCP
analysis used for mutational scans of type I receptors in human
pituitary adenomas. The gray areas indicate the region of
each receptor amplified by RT-PCR and analyzed by SSCP gel
electrophoresis. The GS domain of Alk1 through Alk5 was investigated
for potential point mutations; however, all tumor specimens were
negative for somatic gene alterations in this region. Kinase subdomains
X–XI were also examined by RT-PCR/SSCP. One germline point mutation
was found in the Alk1 gene at residue 482. An in-frame C to T
transversion at the second codon position of residue 482 alters the
translation of the Alk1 receptor, changing an alanine to a valine. This
A482V mutation is shown in Fig. 2. In
addition, it creates a novel HhaI site, which was also used
as an independent marker to screen tumor and leukocyte DNA samples for
the presence of the described mutation (data not shown). Both the
patient’s pituitary adenoma specimen as well as control leukocyte DNA
were heterozygous for the observed A482V mutation. This finding was
independently confirmed by a second full RT-PCR/SSCP study to discount
the potential for PCR error during amplification.
|
|
Full-length RT-PCR/SSCP was performed on ActRIIA, ActRIIB, and
TßRII to investigate potential somatic mutations in these type II
receptors in human pituitary tumors. Both ActRIIB and TßRII were
negative for any mutations in all pituitary tumors studied. However, a
prevalent silent polymorphism at P117 of ActRIIA was detected in 45%
of pituitary tumor and control leukocyte DNA studied (Fig. 3). A nucleotide G to A change in the
third "wobble" codon position of proline residue 117 results in a
conservative base change that fails to alter the ActRIIA protein. The
detected nucleotide substitution eliminates a MspI
restriction enzyme site that was used to screen for the polymorphism in
a large number of patient specimens. MspI restriction
analysis of representative tumors studied by SSCP analysis is shown in
Fig. 3C. No other somatic mutation or polymorphism was detected in any
other region of any of the type II receptors studied by
RT-PCR/SSCP.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our SSCP analysis of the activin/TGFß receptor superfamily demonstrates that mutations in signaling domains of those molecules are rare in human pituitary adenomas. Our mutational survey revealed a novel silent polymorphism (i.e. a base substitution that, due to the redundancy of the genetic code, fails to result in an amino acid change during protein translation) that was detected in 45% of tumors at codon P117 of the ActRIIA gene. Although the P117 polymorphism does not result in an altered ActRIIA protein, this prevalent base change was used as a positive control for SSCP and served to validate the sensitivity of our mutational screen. One patient with a gonadotroph tumor had a confirmed A482V germline mutation in the Alk1 gene within kinase subdomains X–XI. This did not represent a pituitary tumor-specific somatic mutation, and therefore, the pathogenetic implications of this mutation are unclear. No other mutations were detected in any tumor studied. These data suggest that somatic mutations within these intracellular kinase regions of type I/type II receptors are rare in human pituitary tumors.
No mutations were detected in the human type I activin receptor Alk4 (ActRIB), the key receptor that modulates activin signaling (25). Functional complementation studies in mink lung epithelial (Mv1Lu) cell lines that lack type I or type II receptor expression have demonstrated that specific heteromeric complexes mediate intracellular signaling in response to activin and TGFß. The type I activin receptor Alk4 is unique in its ability to up-regulate an activin-responsive reporter gene construct (3TPlux) when c-expressed with ActRII in Mv1Lu cells treated with activin (26, 27). In this system, coexpression of TßRII and Alk4 did not confer responsiveness when cells were treated with TGFß, confirming that this type I receptor is activin specific. Similarly, the TGFß specificity of Alk5 has been demonstrated, whereas Alk3 and Alk6 have been shown to be bone morphogenetic protein-2/4-specific type I receptors (28, 29), Alk1 does not confer signaling to any known TGFß-related ligand; however, it has been implicated in hereditary hemorrhagic telangectasia type 2 (30).
We targeted our SSCP mutational screen to specific regions of the intracellular kinase domain of the type I receptor family based on detailed functional studies that have delineated critical domains for type II receptor interactions as well as activation of downstream signaling pathways. A highly conserved glycine/serine-rich juxtamembrane region, termed the GS domain, was found to be critical for downstream activation of the 3TPlux reporter in Mv1Lu cells (25). Site-specific mutagenesis studies have demonstrated that alterations in the GS domain and adjacent residues can render Alk4 either signaling deficient or constitutively active depending on the exact location and nature of the amino acid substitution. For example, substitutions of nonpolar alanine residues for critical serines in the GS domain render Alk4 signaling deficient, whereas glutamic acid substitutions in the same region exhibit wild-type activity with respect to both basal and activin-stimulated levels of reporter 3TPlux activity (25). Similar mutational analysis of threonine residues immediately carboxyl to the GS domain showed that although alanine substitutions failed to support activin-induced luciferase activity, threonine to glutamic acid at residue 206 yielded a constitutively active Alk4. Together, these data strongly support the hypothesis that the GS domain, although lacking inherent kinase activity, is important for downstream signaling by type I receptors.
In addition to the juxtamembrane GS domain of type I receptors, the intracellular kinase region is critical for proper activin signaling. Phylogenetic mapping of conserved Ser/Thr kinase catalytic domains predicts that the cytoplasmic domain of Alk4 consists of 11 kinase subdomains that are critical for modulating receptor function and intracellular signaling (31, 32). The carboxyl kinase catalytic domain, the region with maximum conservation, resides within subdomains VI–XI. A missense mutation of Pro525 to Leu in human TßRII type II receptor disrupts transphosphorylation of TßR-I and subsequent downstream signaling by type I receptor. In addition, a truncated form of type II receptor lacking kinase subdomains X and XI has been shown to abolish either TGFß- or activin-induced signaling in transfected cell lines (32). Interestingly, kinase subdomain XI is highly variable among Ser/Thr kinase receptor subtypes and, given its demonstrated functional role in transphosphorylation by type II receptors, may play a critical role in kinase substrate specificity (31). Because subdomains VIII–XI are critical for Ser/Thr kinase activity as well as intracellular substrate specificity, they are hypothesized to play a critical role in type I receptor signaling in response to activin.
The finding of a germline mutation within the Alk1-coding region in a patient with a gonadotroph tumor, specifically within the highly conserved kinase X and XI subdomains, was confirmed by both SSCP and sequencing of tumor and control leukocyte DNA. Alk1 mutations have been implicated in hereditary hemorrhagic telangectasia type 2, or Osler-Rendu-Weber syndrome, an autosomal dominant vascular dysplasia (30). A review of clinical data and family history from this patient failed to indicate any clinical symptoms of Osler-Rendu-Weber syndrome. However, the A482V mutation detected by our analysis does not correspond to any of the three documented mutations in kindreds with this heritable disease (30). Therefore, the pathogenetic consequences of A482V of Alk1 are unknown at this time.
Given the paucity of mutations in the activin/TGFß receptor superfamily in human pituitary adenomas, it is possible that the observed tumor-nonresponsive phenotype resides in dysregulation or disruption of downstream intracellular signaling mediators. Potential candidates include the intracellular Smad proteins that mediate TGFß-type receptor signaling in response to activin, TGFß, and bone morphogenetic proteins, among other ligands. In the activin pathway, ligand binding to activin receptors leads to phosphorylation of Smad-2–4, which homo/heterodimerize and accumulate in the nucleus (33). There they form a transcriptional complex with forkhead activin signal transducer-1 and trans-activate activin-responsive genes, leading to their transcription (34). Several homologs of Smads have been well characterized and are involved in diverse processes, including activin signaling and early embryonic development (35, 36). Moreover, Smad-4, also known as "deleted in prostate cancer" (DPC4), is a putative tumor-suppressor protein (37). Future mutational analyses of this pathway in human pituitary tumors will therefore include the Smad effectors of downstream signaling and their potential role in the pathogenesis of human pituitary adenomas.
|
Footnotes |
|---|
2 Current address: Harvard Institutes of Medicine, Room 946, 77 Louis
Pasteur Avenue, Boston, Massachusetts 02114.
Received July 16, 1998.
Revised January 14, 1999.
Accepted February 17, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S A Abdalla and M Letarte Hereditary haemorrhagic telangiectasia: current views on genetics and mechanisms of disease J. Med. Genet., February 1, 2006; 43(2): 97 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. van den Driesche, C. L. Mummery, and C. J.J. Westermann Hereditary hemorrhagic telangiectasia: an update on transforming growth factor {beta} signaling in vasculogenesis and angiogenesis Cardiovasc Res, April 1, 2003; 58(1): 20 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chang, C. W. Brown, and M. M. Matzuk Genetic Analysis of the Mammalian Transforming Growth Factor-{beta} Superfamily Endocr. Rev., December 1, 2002; 23(6): 787 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C Olivieri, E Mira, G Delu, F Pagella, A Zambelli, L Malvezzi, E Buscarini, and C Danesino Identification of 13 new mutations in the ACVRL1 gene in a group of 52 unselected Italian patients affected by hereditary haemorrhagic telangiectasia J. Med. Genet., July 1, 2002; 39(7): e39 - 39. [Full Text] [PDF]
|
|
|
|
|
|
G. P. Risbridger, J. F. Schmitt, and D. M. Robertson Activins and Inhibins in Endocrine and Other Tumors Endocr. Rev., December 1, 2001; 22(6): 836 - 858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. C. Danila, W. J. Inder, X. Zhang, J. M. Alexander, B. Swearingen, E. T. Hedley-Whyte, and A. Klibanski Activin Effects on Neoplastic Proliferation of Human Pituitary Tumors J. Clin. Endocrinol. Metab., March 1, 2000; 85(3): 1009 - 1015. [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 |
