help button home button Endocrine Society JCEM JCEM Call for Nominations for EIC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pauws, E.
Right arrow Articles by Ris-Stalpers, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pauws, E.
Right arrow Articles by Ris-Stalpers, C.
The Journal of Clinical Endocrinology & Metabolism Vol. 85, No. 5 1923-1927
Copyright © 2000 by The Endocrine Society

Original Studies

Serial Analysis of Gene Expression as a Tool to Assess the Human Thyroid Expression Profile and to Identify Novel Thyroidal Genes1

E. Pauws, J. C. Moreno, M. Tijssen, F. Baas, J. J. M. de Vijlder and C. Ris-Stalpers

Laboratory of Pediatric Endocrinology (E.P., J.C.M., M.T., J.J.M.d.V., C.R.-S.) and Neurozintuigen Laboratory (F.B.), Academic Medical Center, University of Amsterdam, 1100 DE Amsterdam, The Netherlands

Address correspondence and requests for reprints to: E. Pauws, Department of Pediatric Endocrinology, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: E.Pauws{at}AMC.UVA.NL


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Discussion
 References
 
The assessment of the expression profile of normal human thyroid tissue using serial analysis of gene expression (SAGE) generated a collection of 10,994 sequence transcripts (tags). Each tag represented a messenger RNA transcript, and, in total, 6099 different tags could be distinguished. The presence and abundance of thyroid-specific transcripts showed the overall expression profile to be from a normal thyroid cell. The expression level of several transcripts was confirmed on Northern blot. Seventy percent of tags could not be attributed to a known human gene and, therefore, possibly correspond to novel genes putatively involved in thyroid function. The tag sequence generated by the SAGE technique can be used to further characterize these novel genes. In this way, application of the SAGE technique to thyroid tissue gives insight in the expression profile of a normal thyroid gland and provides the information to characterize novel genes involved in thyroid pathology, such as congenital hypothyroidism and thyroid neoplasia.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Discussion
 References
 
THE FUNCTIONAL and biochemical features of a specific cell-type are determined by its particular profile of gene expression. Most studies on gene expression focus on one or more previously identified genes of interest. This approach greatly underestimates the complexity of molecular mechanisms. Serial analysis of gene expression (SAGE) is a recently developed technique that provides an expression profile, also called transcriptome. The transcriptome describes genes expressed, including their relative abundance in the tissue or cell-type studied (1). The SAGE technology is based on two main principles (2). First, a short sequence tag (10 bp) is generated, which contains sufficient information to specifically identify a messenger RNA (mRNA) transcript, provided that the sequence tag is derived from a defined location within this transcript. Second, the concatenation of many sequence tags into a large single DNA molecule facilitates high throughput sequencing. The transcriptome is generated by identifying the corresponding gene to each tag and determining the relative abundance of each individual tag. The sensitivity of the method is limited only by the total amount of tags analyzed that will influence the minimal expression level that can be detected. Data from a SAGE library can be used for several purposes. Comparison of SAGE profiles from various physiological or disease states provides insight into the molecular and cellular background of such events showing up- or down-regulation of certain transcripts (3, 4, 5). SAGE tags that have no matches to the current set of known human genes (NoMatch tags) can be used to identify the corresponding uncharacterized genes using the tag sequences generated by the SAGE technique. Compared to SAGE, other techniques quantify only a limited number of previously identified genes at a time (Northern blotting, RNase protection, RT-PCR) or do enable characterization of unidentified mRNA transcripts but do not show direct information about abundance (cDNA subtraction, differential display). Although several genes involved in thyroid development and function have been identified (6, 7, 8), more still remain to be elucidated since not all cDNAs corresponding to proteins known to be involved in thyroid hormonogenesis have been cloned. Cases of primary congenital hypothyroidism (CH) are known where the mutated gene is identified and linked to the patients phenotype, as is the case with the thyroid peroxidase gene (TPO), thyroglobulin (TG), iodide symporter (NIS), TSH receptor (TSH-R) (8, 9), and thyroid-transcription factors PAX8, TTF1, and TTF2 (10, 11, 12). Also, recently the Pendrin (PDS) gene coding for a chloride/iodide-transporter associated with Pendred syndrome was identified as a gene mutated in cases of congenital deafness associated with a mild type of thyroid organification defect (13). There are, however, still unresolved cases of CH where currently unidentified genes must be involved. The thyroid NADPH oxidase responsible for H2O2 generation (14) and the dehalogenase enzyme(s) are two of the most obvious candidates. In thyroid neoplasia little is known about diagnostic and/or prognostic factors concerning the genetic mechanism behind the pathology of the tumor. To address these fields of interest, we constructed and analyzed a SAGE library from a normal human thyroid gland as a starting point for the identification of novel thyroid-specific genes involved in thyroid disease. The genes most prominently expressed in thyroid and their relative abundance are studied, with special attention to genes involved in thyroid function.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Discussion
 References
 
Tissue and RNA extraction

Thyroid gland tissue was obtained from a single individual without thyroid pathology after resection at a routine autopsy. Tissue was immediately frozen in liquid nitrogen. Informed consent to use this material for scientific research was obtained. After homogenization total RNA was extracted using Trizol (Gibco BRL). mRNA was extracted using PolyA/T-Tract mRNA Isolating System III (Promega).

Construction of SAGE library

The library was constructed using 5 µg thyroid mRNA essentially following SAGE Protocol 1.0c by Velculescu et al. (2). Additional information, including graphical presentation of the SAGE technique, can be found at URL www.sagenet.org mRNA was converted to double-stranded complementary DNA (cDNA) using the cDNA Synthesis System kit (Life Technologies, Inc., Gaithersburg, MD) with a biotinylated oligo-dT18. cDNA was digested with NIaIII and 3' cDNA fragments were isolated using Streptavidin Dynabeads M-280 (Dynal, Oslo, Norway) and divided into two equal pools. Each pool was ligated to a different SAGE-linker and subsequently digested with BsmFI to release tags. Tags were then blunted using T4-DNA polymerase, and pools were combined and ligated to form ditags. Ditags were amplified using the attached linkers as priming sites, and PCR products were digested with NIaIII to release linkers from ditags. Ditags (20–24 bp) were isolated from polyacrylamide gel and concatenated by self-ligation. SAGE clones with an average length of 500 bp were cloned into the SphI site of pZero (Invitrogen, Groningen, The Netherlands). Ligation was transformed into Top10F' electrocompetent cells (Invitrogen), and clones with inserts were selected using colony-PCR with M13 vector-located primers.

Sequencing of SAGE library

SAGE clones were sequenced with the Dyenamic Direct cycle sequencing kit using the ET-T7 primer (Amersham Pharmacia, Uppsala, Sweden). Samples were run on a ABI377XL Automatic Sequencer (Perkin-Elmer Corp., Norwalk, CT) and analyzed using Sequence Analysis 3.0 software.

Tag abundance and identification

SAGE data was analyzed using specialized UNIX software USAGE1.5 developed in our institute for extraction of single tags from sequence data and subsequent identification on EMBL human gene database (February 1999). To further study tag identification and expression, NCBI/CGAP’s SAGEMAP program was used at URL www.ncbi.nlm.nih.gov/SAGE/.

Northern hybridizations

RNA gels were prepared using the glyoxal/NaPi method (15) electrophorizing 10 µg of total RNA. Capillary blotting was performed overnight in 20 x SSC, followed by ultraviolet cross-linking (1.2 J/m2) and baking (80 C). Hybridizations were performed following the Church and Gilbert (16) protocol at 65 C overnight, and blots were exposed for 16 h and analyzed using PhosphorImager 2.0 (PE Applied Biosystems, Foster City, CA).


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Discussion
 References
 
The sequencing of 10,994 tags from a human thyroid SAGE library resulted in an expression profile of 6099 unique mRNA transcripts. From these 6099 transcripts, the larger part, 4813 tags, was present only once (0.01% of total library), indicating that the bulk of expressed genes is present at a basic low level. Only 98 genes scored more than 10 tags from which 9 genes were expressed at a very high level (>50 tags), including the TG transcript (Table 1Go). The percentage of identification was larger (85–100%) in tags scored more than 10 than in tags scored 10 or less (25–63%). In total, 30% of tags could be attributed to a known gene transcript. In general, the high abundance class showed expression of several ribosomal and mitochondrial transcripts. The tags corresponding to TG and TPO mRNA are also present in the top 50 of highest expressed genes (Table 2Go). Furthermore, seven NoMatch tags are present putatively corresponding to novel yet unidentified transcripts. Further analysis of SAGE data showed that in the low range of expression the NoMatch tags become more prominent (Table 1Go). Starting with 80 NoMatch tags expressed five times, we screened GenBank databases intensively to try to identify these transcripts. Using the human Expressed Sequence Tag (EST) database about 20 NoMatch transcripts could be identified as a known gene. Other NoMatches could be excluded as artifacts from the linkers used in the construction of the library (linker 1-tag: TCCCTATTAA; and linker 2-tag: TCCCCGTACA). The remaining group of NoMatches could be considered an interesting group of sequences, possibly corresponding to novel genes. Some of the highest expressed NoMatches are listed in Table 2Go. Focusing more on genes important for thyroid function, we summarized in Table 3Go the SAGE expression data of some thyroid-specific genes. Apart from the presence of the expected tags corresponding to TG, TPO, PAX8, TSH-R, PDS, TTF1, iodothyronine deiodinase type 1 and type 2, we found two alternative tags probably corresponding to TG transcripts. Tags corresponding to expected transcripts like TTF2 or NIS could not be detected. The actual TG SAGE tag is located near the end of the TG mRNA spanning the polyadenylation site. Because the two additional tags were scored as a NoMatch with no homology to any known sequence (including ESTs), except TG, we ascribe these tags to be alternative TG SAGE tags as a result of alternative polyadenylation. The total expression level of TG when adding up all tags corresponding to TG comes to 289 tags or 2.6% of total mRNA pool. TPO scored 24 tags (0.26%), showing an expression level 10-fold lower than TG. The identification of the PDS transcripts could only be done after the presence of an internal polyA-stretch in the 3' untranslated region (3' UTR) was noticed. The actual PDS tag flanked the last NIaIII site before this internal polyA-stretch. The validity of the abundance data in our SAGE library was checked using Northern blots. RNA was isolated from several normal thyroid tissues, including the one used for the SAGE library; liver RNA was used as a control. TPO, NIS, and PDS, as well as ELF1{alpha} and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), were hybridized on this blot, and intensities were compared correcting for RNA loading with a 28S ribosomal probe. In Fig. 1Go, the results show that the relative abundance of TPO, ELF1{alpha}, GAPDH, and PDS are similar to that observed in the SAGE library. The expression in thyroid RNA TH4 (RNA used for the SAGE library) was similar to that in five other normal thyroids. Thyroid-specific mRNAs were absent in liver RNA. Because of the absence of a SAGE tag corresponding to the NIS transcript we checked for expression of NIS in the same Northern analysis. NIS was present in the thyroid RNA used for the SAGE library with an estimated expression comparable to TPO and GAPDH.


View this table:
[in this window]
[in a new window]
  		Table 1. Distribution of tags in SAGE library of human thyroid

 

View this table:
[in this window]
[in a new window]
  		Table 2. Top 50 of the most abundant transcripts in normal human thyroid SAGE library TH4

 

View this table:
[in this window]
[in a new window]
  		Table 3. List of transcripts involved in thyroid function in normal human thyroid SAGE library TH4

 


View larger version (74K):
[in this window]
[in a new window]
  		Figure 1. Expressed genes in thyroid SAGE library as determined by Northern blot. The expression of several (thyroid-specific) genes (right) in a group of normal thyroid RNAs is shown. First lane TH4 shows expression levels of RNA as used in the SAGE library. On the left, the SAGE abundance out of 10,994 total tags is indicated. The most right lane is liver control RNA.

 

   Discussion
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Discussion
References
 
The SAGE library data from a normal human thyroid gland constructed in this study show the expected expression profile of the cell type studied. The expression of several housekeeping genes and mitochondrial transcripts was similar to that observed in a recent SAGE study (17). We, as well as other laboratories, checked the validity of SAGE data extensively. Quantitatively we demonstrated that the relative abundance of genes inside the SAGE expression profile was similar to that seen in Northern blot hybridizations and that the expression in this particular normal thyroid was representative. Qualitatively we wanted to see if all genes expressed were present in the SAGE library, focusing on the absence of NIS and TTF2 in our library. The absence of NIS in our thyroid SAGE library can be explained due to the absence of the full-length NIS mRNA sequence in GenBank database. When comparing the 2.2-kb NIS cDNA sequence [accession no. D87920) with the published mRNA length (4.5 kb on Northern blot (18)], around 2 kb of 3' UTR downstream sequence is missing, including the polyadenylation signal and polyA-tail. Also, no 3'EST sequences corresponding to NIS mRNA could be found in human EST databases. The SAGE tag corresponding to this transcript is probably in this missing 2-kb 3' UTR sequence since the average distance from tag to polyA-tail is 256 bp. Table 2Go shows several NoMatch tags in the expected range of expression putatively belonging to the NIS mRNA transcript. Northern analysis of thyroid RNA used for the SAGE library showed NIS mRNA expression at an abundance level slightly lower than that of TPO. TTF2 could not be detected, although the full-length mRNA sequence is known. Because the abundance in the library of transcription factors TTF1 (3) and PAX8 (7) was relatively low, this can be due to the detection limit of this particular SAGE library. The difficulties in identifying PDS in the SAGE library arose from the fact that an internal polyA-stretch is present in the PDS 3' UTR. In the initial cDNA synthesis the internal polyA-stretch is probably used as a priming site, resulting in this alternative SAGE tag. The actual last NlaIII site and corresponding tag situated between the polyA-stretch and the polyA-tail could not be found in the SAGE library, indicating that the internal poly-A stretch is preferentially used in oligo-dT priming. Transcripts specific for C cells (calcitonin, ret), which are also present in thyroid gland tissue, could not be found probably because the amount of C cells in the thyroid is too low (<1%) (19) to pick up any mRNA in the total RNA pool from thyroid tissue. The presence of ferritin H as one of the highest expressed messengers (Table 2Go) in thyroid cells was described previously for FRTL-5 cells (20). Its abundance in the SAGE library confirms this finding. The sensitivity of the SAGE library shows that a single tag present in the thyroid SAGE library corresponded to an expression level of 0.01% (1 in 10,994 total tags). Assuming the SAGE expression profile represents the total mRNA pool of the thyroid cell and the total amount of mRNA molecules per cell is about 300,000 (21), we estimate that the detection level was limited to mRNAs with 30 mRNA copies per cell. A possibility to increase the sensitivity of a SAGE library is by sequencing more clones and analyzing more tags. Next to the transcripts corresponding to known genes, we are especially interested in the NoMatch tags, not corresponding to a known gene. Seventy percent of the total amount of different tags in the library is a NoMatch. This number is, however, somewhat overestimated because 85% of the NoMatches are in the group of tags scored once. In the high range of expression (tags >=5) the amount of NoMatch tags is considerably lower (80 of 281, 28%), indicating that transcripts expressed at a high level are better characterized than lower expressed transcripts (Table 1Go). In the large group of single tags we expect to have some sequence artifacts (3, 17) explaining also the high percentage of NoMatches in this group. The possibility of an artifact, such as found in the alternative Tg polyA sites and the linkers from the SAGE technique, has to be ruled out. Before using any NoMatch tag sequence in an experiment it is, to our view, imperative to screen extensively the possibility of a NoMatch tag corresponding to a known gene from which the 3' UTR is not sequenced. This can partly be circumvented by screening the human EST database. Via the screening of the EST database it is also possible to use an unidentified 3' EST corresponding to a NoMatch tag as a probe sequence in further experiments. SAGE has become a promising technique in molecular genetics, generating enormous amounts of immortalized data. The data generated from this SAGE library bring us closer to identifying the complete thyroid transcriptome. Because SAGE libraries are made in different scientific institutions around the world according to the same principle and protocol, it is possible to even compare interlaboratory data. Comparing libraries from normal tissues vs. pathological tissues generates differentially expressed tags corresponding to known and unknown genes possibly involved in the pathology of the disease. We intend to use this possibility in the future by making new SAGE libraries from thyroid disease tissues. The NoMatch tags that at the moment can not be identified to a known human gene can, in the future, when the Human Genome Project and similar projects finish characterizing the complete human genome, be screened again. In the meantime, these NoMatch sequences provide valuable information to start characterizing novel genes using other techniques. We conclude that the SAGE library from human thyroid offers an extensive expression profile of both previously identified and unidentified genes that can be used to elucidate novel genes involved in thyroid hormonogenesis.


   Acknowledgments
 
We thank A. H. C. van Kampen and M. van der Mee (Bioinformatics Laboratory, AMC) for their work in designing the SAGE analysis software USAGE used in this study.


   Footnotes
 
1 The Dr. Ludgardine Bouwman foundation and the Stichting Kindergeneeskundig Kankeronderzoek (SKK) financially supported this study. This study was supported in part by ESPE Research Fellowship, sponsored by NOVO Nordisk A/S. Back

Received October 20, 1999.

Revised December 7, 1999.

Accepted December 20, 1999.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Discussion
 References
 

  1. Bertelsen AH, Velculescu VE. 1998 High-throughput gene expression analysis using SAGE. Drugs Discovery Today. 3:152–159.[CrossRef]
  2. Velculescu VE, Zhang L, Vogelstein B, Kinzler KW. 1995 Serial analysis of gene expression. Science. 270:484–487.[Abstract/Free Full Text]
  3. Zhang L, Zhou W, Velculescu VE, et al. 1997 Gene Expression Profiles in Normal and Cancer cells. Science. 276:1268–1272.[Abstract/Free Full Text]
  4. Velculescu VE, Zhang L, Zhou W, et al. 1997 Characterization of the yeast transcriptome. Cell. 88:243–251.[CrossRef][Medline]
  5. Madden SL, Galella EA, Zhu J, Bertelsen AH, Beaudry GA. 1997 SAGE transcript profiles for p53-dependent growth regulation. Oncogene. 15:1079–1085.[CrossRef][Medline]
  6. Dai G, Levy O, Carasco N. 1996 Cloning and characterization of the thyroid iodide transporter. Nature. 379:458–460.[CrossRef][Medline]
  7. Scott DA, Wang R, Kreman TM, Sheffield VC, Karniski LP. 1999 The Pendred syndrome gene encodes a chloride-iodide transport protein. Nat Genet. 21:440–443.[CrossRef][Medline]
  8. Taurog A. 1991 Hormone synthesis. In: Braverman LE, Utiger RD, eds. The thyroid. Philadelphia: JB Lippincott; 51–98.
  9. Bikker H, Baas F, deVijlder JJM. 1997 Molecular analysis of mutated thyroid peroxidase detected in patients with total iodide organification defects. J Clin Endocrinol Metab. 82:649–653.[Abstract/Free Full Text]
  10. Clifton-Bligh RJ, Wentworth JM, Heinz P, et al. 1998 Mutation of the gene encoding human TTF-2 associated with thyroid agenesis, cleft palate and choanal atresia. Nat Gene. 19:399–401.[CrossRef][Medline]
  11. Macchia PE, Lapi P, Krude H, et al. 1998 PAX8 mutations associated with congenital hypothyroidism caused by thyroid dysgenesis. Nat Gene. 19:83–86.[CrossRef][Medline]
  12. DeVriendt K, VanHol C, Matthijs G, DeZegher F. 1998 Deletion of thyroid transcription factor-1 gene in an infant with neonatal thyroid dysfunction and respiratory failure. N Engl J Med. 338:1317–1318.[Free Full Text]
  13. VanHauwe P, Everett LA, Coucke P, et al. 1998 Two frequent missense mutations in Pendred syndrome. Hum Mol Gene. 7:1099–1104.[Abstract/Free Full Text]
  14. Leseney A-M, Deme D, Legue O, et al. 1999 Biochemical characterization of a Ca2+/NAD(P)H-dependent H2O2 generator in human thyroid tissue. Biochimie. 81:373–380.[Medline]
  15. Ausubel FM, Brent R, Kingston RE, et al. 1999 Chanda VB, ed. Current protocols in molecular biology. John Wiley & Sons, Inc.
  16. Church GM, Gilbert W. 1984 Genomic sequencing. Proc Natl Acad Sci USA. 81:1991–1995.[Abstract/Free Full Text]
  17. Welle S, Bhatt K, Thornton CA. 1999 Inventory of high-abundance mRNAs in skeletal muscle of normal men. Genome Res. 9:506–513.[Abstract/Free Full Text]
  18. Smanik PA, Liu Q, Furminger TL, et al. 1996 Cloning of the human iodide symporter. Biochem Biophy Res Commun. 226:339–345.[CrossRef][Medline]
  19. Kalina M, Pearse AGE. 1971 Ultrastructural localization of calcitonin in C-cells of dog thyroid; an immunocytochemical study. Histochemie. 26:1.[CrossRef][Medline]
  20. Cox F, Gestautas J, Rapoport B. 1988 Molecular cloning of cDNA corresponding to mRNA species whose steady state levels in the thyroid are enhanced by thyrotropin. Homology of one of these sequences with ferritin H. J Biol Chem. 263:7060–7067.[Abstract/Free Full Text]
  21. Hastie ND, Bishop JO. 1976 The expression of three abundance classes of messenger RNA in mouse tissues. Cell. 9:761–774.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
L. de Sanctis, A. Corrias, D. Romagnolo, T. DI Palma, A. Biava, G. Borgarello, P. Gianino, L. Silvestro, M. Zannini, and I. Dianzani
Familial PAX8 Small Deletion (c.989_992delACCC) Associated with Extreme Phenotype Variability
J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5669 - 5674.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
Y. Yano, N. Uematsu, T. Yashiro, H. Hara, E. Ueno, M. Miwa, G. Tsujimoto, Y. Aiyoshi, and K. Uchida
Gene Expression Profiling Identifies Platelet-Derived Growth Factor as a Diagnostic Molecular Marker for Papillary Thyroid Carcinoma
Clin. Cancer Res., March 15, 2004; 10(6): 2035 - 2043.
[Abstract] [Full Text] [PDF]

Home page
 Endocr. Rev.Home page
C. P. Leo, S. Y. Hsu, and A. J. W. Hsueh
Hormonal Genomics
Endocr. Rev., June 1, 2002; 23(3): 369 - 381.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
E. Pauws, A. H. C. van Kampen, S. A. R. van de Graaf, J. J. M. de Vijlder, and C. Ris-Stalpers
Heterogeneity in polyadenylation cleavage sites in mammalian mRNA sequences: implications for SAGE analysis
Nucleic Acids Res., April 15, 2001; 29(8): 1690 - 1694.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pauws, E.
Right arrow Articles by Ris-Stalpers, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pauws, E.
Right arrow Articles by Ris-Stalpers, C.

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