| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Institut für Pharmakologie, Freie Universität Berlin (K.P., A.S., T.S.), D-14195 Berlin; Universitätsklinik und Poliklinik für Kinderheilkunde, Medizinische Fakultät (Charité) der Humboldt Universität zu Berlin (K.P., A.G., G.F.), D-10098 Berlin; Pädiatrische Nephrologie, Universitäts Kinderklinik Hamburg (K.T.), D-20246 Hamburg; and Klinik und Poliklinik für Kinder und Jugendmedizin (K.L.) and Institut für Humangenetik (M.H.), Ernst-Moritz-Arndt Universität, D-17487 Greifswald, Germany; and Department of Pediatrics, Kuopio University Hospital (J.J.), 70211 Kuopio, Finland
Address all correspondence and requests for reprints to: Dr. Torsten Schöneberg, Institut für Pharmakologie, Freie Universität Berlin, Thielallee 69–73, D-14195 Berlin, Germany. E-mail: schoberg{at}zedat.fu-berlin.de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Besides mutational analysis of the NDI patient and possible carriers, investigations include the functional characterization of the mutationally altered AVPR2 or AQP2 proteins. Mutations of the AVPR2 gene result in receptor malfunction at different levels, such as complete gene deletion (4, 5, 6), improper messenger ribonucleic acid splicing (7, 8), intracellular receptor retention (9, 10), or disturbances in receptor-ligand binding (10, 11). Such a large variety of naturally occurring mutations in the AVPR2 gene in conjunction with a detailed functional characterization offer a unique opportunity to study molecular mechanisms governing receptor function and dysfunction and may offer novel therapeutic approaches in the treatment of diseases caused by mutations in GPCRs.
Herein, we report 10 patients with NDI belonging to 8 families. Three novel missense mutations (I46K, F105V, I130F) and 4 recurrent mutations (D85N, R106C, R113W, Q225X) in the AVPR2 gene and a recurrent missense mutation (A147T) within the AQP2 gene were identified. In one family, the mother was found to be mosaic for the I130F mutation, indicating a de novo mutational event. Functional characterization of the 3 novel mutant AVPR2s demonstrated that the identified AVPR2 defects were responsible for the AVP resistance in our NDI patients. By performing cAMP accumulation assays and radioligand binding studies in parallel with enzyme-linked immunosorbent assay (ELISA) experiments we show that the 3 mutations interfere with proper receptor function at different stages. However, data from in vitro studies do not always reflect the clinical phenotype and have to be interpreted only in conjunction with clinical and laboratory values.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Nine male patients (patients 1–9) and one female patient
(patient 10) with congenital NDI belonging to eight families were
investigated (Fig. 1
and Table 1). The patients of families 1, 3, and
5–8, were of German descent. Patients of families 2 and 4, both of
Finnish descent, showed a mild NDI phenotype. The female NDI patient
(patient 10) was of Turkish ancestry and displayed mental retardation
and deafness in addition to the NDI symptoms. Most of the patients had
a history of fever, polyuria, polydipsia, periods of constipation, and
failure to thrive. The diagnosis of NDI was based on clinical symptoms
and lack of increase in urinary osmolality after administration of
1-desamino-8-D-arginine vasopressin (dDAVP;
available parameters are shown in Table 1). The parents of all families
but family 8 denied consanguinity. All chemical laboratory tests were
performed using standard clinical laboratory assays. Most plasma AVP
measurements were performed using an AVP RIA from Nichols Institute Diagnostics (Bad Nauheim, Germany).
|
|
Sample preparation and AVPR2 gene analysis
Genomic DNA was prepared from whole blood by conventional methods (QIAamp blood kit, QIAGEN, Hilden, Germany). The AVPR2 gene was amplified by PCR with 0.5 µg genomic DNA as template. Four overlapping primer sets were used to amplify the entire coding sequence including the two introns: V2–1/V2–2, 5'- TCCGCACATCACCTCCAGGCC-3'/5'-CCACTTCCTGGCTCCTAGCAGAG-3'; V2–3/V2–4, 5'-GTCTCTCCAGGCTGCCAATGAGTG-3'/5'-CAATCCAGGTGACAT-AGGTC-3'; V2–5/V2–6, 5'-CATCTTCGCCCAGCGCAACGT-3'/5'-CCACTAGAGGCAAGACACCCAACAGCTCC-3'; and V2–7/V2–8, 5'-CACGTCTTCATTGGCCACTTGTGC-3'/5'-ACTGGCATGAATCT-CCCG-GAAGAT-3'. For amplification of the four exons of the human AQP2 gene four primer pairs were used as previously described (3). All PCR reactions were carried out using the Expand High Fidelity system (Roche Molecular Biochemicals, Mannheim, Germany) under the following conditions (35 cycles): 1 min at 94 C, 1 min at 64 C, and 2 min at 68 C. PCR products were separated by 1.0% and 1.5% agarose gel electrophoreses. Phenol/chloroform-extracted PCR fragments were sequenced directly according to standard methodology by using an automated sequencer (PE Applied Biosystems, Foster City, CA).
Construction of mutant AVPR2 genes
All AVPR2 mutations (Fig. 2) were introduced into V2-R-pcDps
(12), a mammalian expression vector containing the entire coding
sequence of the human AVPR2, using a PCR-based site-directed
mutagenesis and restriction fragment replacement strategy. For
immunological detection of the various AVPR2 constructs, a stretch of
nucleotides coding for a nine-amino acid epitope (YPYDVPDYA) (13)
derived from the influenza virus hemagglutinin protein (HA tag) was
inserted after the initiating Met codon. The identities of the various
constructs and the correctness of all PCR-derived sequences were
confirmed by restriction analysis and direct DNA sequencing. To check
on the transfection efficiency and for control purposes in ELISA
studies, a mammalian expression plasmid for the green fluorescent
protein (GFP) was used (pEGFP-C1 vector, CLONTECH Laboratories, Inc., Palo Alto, CA).
|
COS-7 cells were grown in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 C in a humidified 7% CO2 incubator. For transient transfections of COS-7 cells, a calcium phosphate coprecipitation method (14) was applied. Thus, cells were split into 12-well plates (2 x 105 cells/well) and transfected with a total amount of 5 µg plasmid DNA/well.
After 48 h, cells were prelabeled with 2 µCi/mL [3H]adenine (31.7 Ci/mmol; NEN Life Science Products, Boston, MA) and incubated overnight. For the cAMP assay, transfected cells were washed once in serum-free DMEM containing 1 mmol/L 3-isobutyl-1-methylxanthine (Sigma, St. Louis, MO), followed by incubation in the presence of the indicated AVP (Sigma) concentrations for 1 h at 37 C. Reactions were terminated by aspiration of the medium and addition of 1 mL 5% trichloric acid. The cAMP content of cell extracts was determined by anion exchange chromatography as previously described (15).
For radioligand binding studies, cells were harvested 72 h after transfection (20 µg plasmid DNA/100-mm dish), and saturation binding assays were performed using membrane homogenates. Incubations were carried out for 1 h at 22 C in a 0.25-mL volume, with six different concentrations (1.25–100 nmol/L) of [3H]AVP (64 Ci/mmol; NEN Life Science Products). Nonspecific binding was defined as binding in the presence of 10 µmol/L AVP. Data from cAMP assays and radioligand binding studies were analyzed by a nonlinear curve-fitting procedure using the computer program GraphPad Prism version 2.0 (GraphPad Software, Inc., San Diego, CA).
Immunological studies
To estimate cell surface expression of receptors carrying an amino-terminal HA tag, we developed an indirect cellular ELISA (12), referred to as a surface ELISA. Briefly, COS-7 cells were seeded into 48-well plates, transfected, formaldehyde fixed without disrupting the cell membrane, and incubated with a biotin-labeled anti-HA monoclonal antibody (12CA5, Roche Molecular Biochemicals, Mannheim, Germany). Bound anti-HA antibody was then detected with the help of a peroxidase-labeled streptavidin conjugate (Sigma).
To further assess the amounts of full-length HA-tagged AVPR2s, a previously developed sandwich ELISA was used (16). In brief, 3 days after transfection (12 µg plasmid DNA/60-mm dish), COS-7 cells were harvested, and cell pellets were resuspended in 150 µL lysis buffer [10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1 mmol/L dithiothreitol, 1 mmol/L ethylenediamine tetraacetate, 1% deoxycholate, 1% Nonidet P-40, 0.2 mmol/L phenylmethylsulfonylfluoride, and 10 µg/mL aprotinin). Cell debris was removed by centrifugation, and supernatants were used for ELISAs. Microtiter plates were coated with a polyclonal rabbit antibody directed against a peptide corresponding to the carboxyl-terminal 29 amino acids of the human AVPR2 (provided by Dr. Paul Goldsmith, NIH; 5 µg/mL in PBS). After incubation at 4 C for 16 h, plates were blocked with 10% FBS in PBS. Then, cell lysates were applied and incubated at 37 C for 2 h. Plates were washed three times with PBS containing 0.05% Triton X-100 (PBS-T). Thereafter, the biotin-labeled monoclonal anti-HA antibody (12CA5; 1 µg/mL PBS-T) was added, and plates were incubated at 37 C for 2 h. Plates were washed with PBS-T and incubated with an 1:5000 dilution of peroxidase-conjugated streptavidin for 1 h at 37 C. After removal of excess unbound conjugate, H2O2 and o-phenylenediamine (2.5 mmol/L each in 0.1 mol/L phosphate-citrate buffer, pH 5.0) were added to serve as substrate and chromogen, respectively. After 15 min, the enzyme reaction (carried out at room temperature) was stopped by the addition of 1 mol/L H2SO4 containing 0.05 mol/L Na2SO3, and color development was measured bichromatically at 492 and 620 nm using an ELISA reader (Titertek Multiskan MCC/340, Labsystems, Helsinki, Finland).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Genomic DNA from all available family members were prepared and
subjected to PCR amplification of the coding sequence of the AVPR2 gene
including the two introns. Sequence analysis of the PCR fragments
revealed three novel (I46K, F105V, I130F) and four recurrent (D85N,
R106C, R113W, Q225X) mutations (Fig. 2
and Table 1). In the case of patient 10, DNA analysis showed wild-type
AVPR2 sequence. Therefore, mutational analysis was extended to the AQP2
gene, and a homozygote occurrence of an A147T mutation was found. This
missense mutation is recurrent and was functionally characterized in
Xenopus oocytes, identifying this mutation as the
NDI-causing mechanism (17).
In most families with X-linked NDI, mutational analysis has identified
the mothers as carriers for the distinct AVPR2 mutation. As shown in
Fig. 3A, sequence analysis of the mother
of patient 8 (family 6) revealed wild-type AVPR2 sequence. This result
was verified by repeated bidirectional sequencing. The I130F mutation
found in patient 8 is caused by an A
T transversion generating a new
AflIII site. For restriction analysis, a PCR fragment (1036
bp) was amplified using the primers (V2–1/V2–4). One
AflIII restriction site is present in the wild-type
sequence, and therefore, digestion with AflIII generates two
fragments (630 and 406 bp; internal control of enzyme function).
Mutational introduction of an additional restriction site for
AflIII at codon position 130 results in three fragments
(630, 245, and 161 bp) upon AflIII treatment. Restriction
analysis of the PCR fragment from the father’s genomic DNA showed the
predicted wild-type restriction pattern, and the presence of the
mutation was verified for the patient (see Fig. 3B). An unexpected
result was found for the patient’s mother. In addition to the normal
band pattern, AflIII digest revealed two faint bands
corresponding to the mutationally derived fragments found in the
patient. The band intensities were lower, as expected for a
heterozygous genotype (see Fig. 3B). To exclude contamination or
specific amplification artifacts, maternal genomic DNA was prepared
from an independently drawn blood sample and PCR/AflIII
restriction analysis was performed using a second primer pair
(V2–7/V2–8) spanning the mutated AVPR2 segment. After
AflIII digest, two faint bands (161 and 307 bp) appeared
additionally to the uncut PCR product (468 bp), verifying the presence
of a low amount of mutant genomic DNA (data not shown).
|
We introduced the novel mutations found in the patients into the
wild-type AVPR2 gene by PCR-based site-directed mutagenesis to study
their functional relevance. COS-7 cells were transiently transfected
with the wild-type and various mutant AVPR2s. Agonist stimulation of
COS-7 cells expressing the wild-type AVPR2 resulted in a 15-fold
increase in intracellular cAMP levels over basal values
(EC50, 0.3 nmol/L). By contrast, cells
transfected with the truncated mutant (Q225X) did not show any
detectable AVP-induced cAMP formation (Fig. 4 and Table 2). Agonist stimulation of F105V resulted
in a 14.6-fold increase in intracellular cAMP levels, but exhibited a
shift in concentration-response curve toward higher AVP concentrations
(EC50, 200 nmol/L) compared with the AVPR2 (Fig. 4 and Table 2). AVPRs carrying the I46K, R106C, and I130F mutations
showed a reduced potency (2- to 3-fold over basal) and efficiency
(EC50, 1.2–3.7 nmol/L) of cAMP generation (Fig. 4 and Table 2). The functional relevance of D85N and R113W has been
investigated in previous studies (18, 19), obliviating the need to
further characterize their relevance. To assess the mechanism causing
loss of receptor function, radioligand binding studies were performed.
Saturation binding experiments (up to 100 nmol/L
[3H]AVP) on membranes of COS-7 cells
transfected with the wild-type AVPR2 yielded a binding capacity
(Bmax) of 1 x 105
receptors/cell and a Kd of 1.7 nmol/L. Using
similar amounts of transfected plasmid DNA (20 µg/10-cm dish), all
mutant AVPR2s showed no detectable binding sites (I46K, R106C, I130F,
Q225X; Table 2). In the case of F105V, no saturation of
[3H]AVP-binding sites was achieved. Taking the
cAMP accumulation data (Emax and
EC50 values; Table 2) into account, a wild-type
AVPR2 comparable Bmax value, but an about
1000-fold shifted Kd value, can be expected for
F105V. However, the [3H]AVP concentrations
necessary to achieve complete saturation for F105V are not available
with the required specific activity. Saturation binding experiments on
extreme right-shifted AVPR2 mutants such as F105V are additionally
limited by a considerable increase in the nonspecific binding at higher
[3H]AVP concentrations.
|
|
). This assay allows the
quantification of receptor protein present in the plasma membrane
independently of the receptor’s ligand-binding properties (12). The
I46K, I130F, and Q225X mutants showed cell surface expression levels
below 15% of the wild-type AVPR2. In the case of R106C, cell surface
expression was reduced by 50%. Interestingly, ELISA experiments with
the F105V mutant demonstrated that this mutation did not influence
receptor trafficking into the plasma membrane in this in
vitro expression system (Fig. 5).
|
). The
truncated receptor protein (Q225X) found in patient 9 was not
detectable in the sandwich ELISA because of the absence of the
carboxyl-terminal receptor portion. It should be noted here that only
full-length AVPR2 constructs in which both epitopes, the amino-terminal
HA epitope tag and the carboxyl-terminal receptor sequence, are present
can be detected in the sandwich ELISA. |
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Functional characterization of mutations found in the AVPR2 gene is usually performed in transient in vitro expression systems such as COS-7 cells. In the majority of cases, data match the clinical phenotype. For example, previous in vitro studies of AVPR2 mutations (D85N, R113W) that were also found in our study have confirmed the partial (D85N) and complete (R113W) phenotype of the NDI (18, 19). Herein, we have identified three novel (I46K, F105V, I130F) and two recurrent (R106C, Q225X) AVPR2 mutations that have not yet been functionally characterized. Protein truncation (Q225X) and amino acid exchange (I46K, I130F) led to a complete loss or dramatic reduction of agonist-mediated cAMP formation. In most cases, loss of function was accompanied by a reduced receptor cell surface expression, as demonstrated by radioligand binding assays and ELISA studies. The functional data for I46K, I130F, and Q225X were in good agreement with a complete AVP resistance in the NDI patients harboring these mutations.
Functional characterization of mutant AVPR2s in transient
overexpression systems can often feign a reasonable receptor function,
whereas expression of the same mutant AVPR2 in stably transfected cell
lines discloses the molecular receptor defect (20). In contrast to
these observations, in vitro characterization of R106C
(family 4) revealed an almost complete loss of function, even though
the two patients displayed a mild phenotype of NDI with basal urinary
osmolalities as high as those found in the patients of family 2 after
dDAVP administration (D85N, partial NDI). Substitution of
R106 with cysteine resulted in a complex alteration of receptor
function, characterized by a significant reduction in cell surface
expression and signal transduction potency and efficacy. The arginine
residue at amino acid position 106 is not conserved within the
vasopressin/oxytocin receptor family (Fig. 6), suggesting a subordinated role in
direct participation in agonist binding. This assumption is supported
by the fact that R106, which is replaced by histidine in the wild-type
rat AVPR2, does not account for species differences in agonist and
antagonist binding (21). Substitutions of various amino acid residues
to cysteine (R181C, G185C, R202C, R203C, Y205C) within the first and
second extracellular loops have been identified as inactivating
mutations causing NDI (22). Therefore, it was suggested that the
presence of an additional cysteine in one of the extracellular loops
may offer alternatives in the formation of the extracellular disulfide
bond that is highly conserved among GPCRs (10). Further site-directed
mutagenesis studies have to be applied to address the question of
whether this hypothetical mechanism accounts for the receptor
dysfunction. Nevertheless, neither of the two boys (family 4) received
any medication or had ever had any symptoms of severe dehydration. For
some reason both patients are able to partially compensate for the
receptor dysfunction. As an agonist-independent "constitutive"
activity (basal receptor activity) of the V2-R was experimentally
excluded, the high AVP concentration found in patient 5 in conjunction
with the residual receptor function of R106C, as observed in in
vitro assays, may account in vivo for an
agonist-dependent constitutive receptor activation.
|
, the coevolutionary
process of the vasopressin/oxytocin receptors and their physiological
agonists has preserved the F105, implicating a pivotal role for this
residue in receptor function. Beside the well established model of
agonist-receptor contact sites within the transmembrane domain
core (23), recent site-directed mutagenesis studies and approaches
using chimeric oxytocin/vasopressin receptors highlight the importance
of extracellular residues in the determination of vasopressin receptor
affinity and specificity (21, 24, 25). Therefore, it is reasonable to
assume that this evolutionary conserved phenylalanine residue directly
interacts with a structure that has also been conserved within the
vasopressin-like peptides or participates in stabilizing an
intramolecular architecture necessary for high affinity agonist
binding. In the case of patient 10, no mutation was found within the coding region of the AVPR2 gene. Mutational analysis of the AQP2 gene revealed a homozygous presence of threonine at position A147. This AQP2 mutation has been identified and functionally characterized in a previous study (17). Expression of the A147T mutant AQP2 protein in Xenopus oocytes revealed only a small increase in water permeability, and immunoblots of oocyte lysates showed that the A147T mutant protein was less stable than wild-type AQP2. Additionally to the classical phenotype of NDI, the female patient presented with congenital deafness. Detailed clinical examination revealed an inner ear defect. The combination of congenital NDI and deafness appears to be rare and was only reported as Meniere-type hearing loss (26). It is well established that the kidney and the cochlea are linked by structural and anatomical characteristics as well as by the physiological mechanism of electrolytes and fluid regulation. The presence of AQP2 in the inner ear is discussed controversially. Using specific antibodies, AQP1 and -4 expression was found in the inner ear, whereas AQP2, -3, and -5 were not detectable (27). In guinea pigs, both AVPR2 and AQP2 were expressed in endolymphatic sac epithelium (28). As deafness is not a common symptom of X-linked and autosomal recessive NDI, additional factors may be involved in the hearing loss of our patient.
In conclusion, we have identified three novel and four recurrent mutations of the AVPR2 gene and one recurrent mutation of the AQP2 gene in eight independent families. Negative data from sequencing analysis of potential carriers have to be interpreted with caution, because the possibility of mosaicism cannot be excluded. Functional characterization of receptor mutations not only establishes the molecular basis of disease, but also reveals new aspects in structure/function-relationship of GPCR.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 13, 1999.
Revised December 14, 1999.
Accepted December 20, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Faerch, J. H. Christensen, S. Rittig, J.-O. Johansson, N. Gregersen, F. de Zegher, and T. J. Corydon Diverse vasopressin V2 receptor functionality underlying partial congenital nephrogenic diabetes insipidus Am J Physiol Renal Physiol, December 1, 2009; 297(6): F1518 - F1525. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lemaire, D. Chitayat, D. F. Geary, D. G. Bichet, and C. Licht A novel disease-causing mutation in AVPR2: Q96H NDT Plus, February 1, 2009; 2(1): 20 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Robben, N. V. A. M. Knoers, and P. M. T. Deen Cell biological aspects of the vasopressin type-2 receptor and aquaporin 2 water channel in nephrogenic diabetes insipidus. Am J Physiol Renal Physiol, August 1, 2006; 291(2): F257 - F270. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zaki, T. Schoneberg, T. Al Ajrawi, A. N. Al Said, K. Sangkuhl, and H. Rompler Nephrogenic diabetes insipidus, thiazide treatment and renal cell carcinoma Nephrol. Dial. Transplant., April 1, 2006; 21(4): 1082 - 1086. [Full Text] [PDF]
|
|
|
|
 |
|
 V. Bernier, J.-P. Morello, A. Zarruk, N. Debrand, A. Salahpour, M. Lonergan, M.-F. Arthus, A. Laperriere, R. Brouard, M. Bouvier, et al. Pharmacologic Chaperones as a Potential Treatment for X-Linked Nephrogenic Diabetes Insipidus J. Am. Soc. Nephrol., January 1, 2006; 17(1): 232 - 243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Robben, N. V. A. M. Knoers, and P. M. T. Deen Characterization of vasopressin V2 receptor mutants in nephrogenic diabetes insipidus in a polarized cell model Am J Physiol Renal Physiol, August 1, 2005; 289(2): F265 - F272. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Boccalandro, F. de Mattia, D.-C. Guo, L. Xue, P. Orlander, T. M. King, P. Gupta, P. M.T. Deen, V. R Lavis, and D. M. Milewicz Characterization of an Aquaporin-2 Water Channel Gene Mutation Causing Partial Nephrogenic Diabetes Insipidus in a Mexican Family: Evidence of Increased Frequency of the Mutation in the Town of Origin J. Am. Soc. Nephrol., May 1, 2004; 15(5): 1223 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Schulz, K. Sangkuhl, T. Lennert, M. Wigger, D. A. Price, A. Nuuja, A. Gruters, G. Schultz, and T. Schoneberg Aminoglycoside Pretreatment Partially Restores the Function of Truncated V2 Vasopressin Receptors Found in Patients with Nephrogenic Diabetes Insipidus J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 5247 - 5257. [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 |
