help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-0404
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
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 Umar, A.
Right arrow Articles by Brinkmann, A. O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Umar, A.
Right arrow Articles by Brinkmann, A. O.
Right arrowPubmed/NCBI databases
*Substance via MeSH
Endocrinology Vol. 144, No. 10 4637-4647
Copyright © 2003 by The Endocrine Society

Proteomic Profiling of Epididymis and Vas Deferens: Identification of Proteins Regulated during Rat Genital Tract Development

Arzu Umar, Marja P. Ooms, Theo M. Luider, J. Anton Grootegoed and Albert O. Brinkmann

Department of Reproduction and Development (A.U., M.P.O., A.G., A.O.B.), and Department of Neurology and Erasmus Center for Biomics (T.M.L.), Erasmus MC, Rotterdam 300 DR, The Netherlands

Address all correspondence and requests for reprints to: Arzu Umar, Erasmus MC, Department of Reproduction and Development, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. E-mail: a.umar{at}erasmusmc.nl.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Epididymis and vas deferens form part of the male internal genital tract and are dependent on androgens for their growth and development. To better understand the molecular action of androgens during male genital tract development, protein expression profiles were generated using two-dimensional gels, for rat epididymides and vasa deferentia isolated on embryonic days (E) 17–21. Proteins that were differentially expressed between E17 and E21 were cut from the gels, digested into tryptic peptides and analyzed on a matrix-assisted laser desorption/ionization time-of-flight mass spectrometer. Using this approach, 20 proteins could be identified that were regulated in time and were categorized into cytoskeletal proteins, nuclear proteins, transport proteins, chaperones, and enzymes (mainly glycolytic). Furthermore, epididymides and vasa deferentia isolated on E19 were cultured in vitro in the absence or presence of 10 nM of the synthetic androgen R1881, for 9, 24, and 48 h. Under these conditions, regulation and posttranslational modification were observed for glyceraldehyde 3-phosphate dehydrogenase, triosephosphate isomerase, heterogeneous nuclear ribonucleoprotein A2/B1 and heterogeneous nuclear ribonucleoprotein A3, similar to the observed changes in vivo. In addition, posttranslational modification of RhoGDI1 (also named RhoGDI{alpha}) was found in response to androgen. Androgen-induced posttranslational modification of RhoGDI1 and glycolytic enzymes may be an important functional link between signaling pathways and cytoskeletal rearrangements in control of growth and development of the male internal genital tract.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
ANDROGENS PLAY AN important role in regulation of male development and physiology, ranging from male sex differentiation during embryogenesis, several aspects of puberty and spermatogenesis, to adult male function (1, 2, 3). The principal androgen testosterone (T) is essential for the development of the internal sex organs derived from the Wolffian duct system: epididymis, vas deferens, and seminal vesicle. Dihydrotestosterone (DHT), the 5{alpha} reduced form of T, is involved in development of the prostate and external genitalia from the urogenital sinus and the tubercle (4, 5). Both T and DHT perform their actions via one and the same intracellular receptor, the androgen receptor (AR), but DHT is a more potent androgen. Binding of either T or DHT to the AR might be a mechanism to regulate distinct androgenic effects in target tissues (5). In humans, the differential action of T and DHT is illustrated in clinical syndromes such as androgen insensitivity (AIS) and 5{alpha}-reductase type2 deficiency. Complete AIS patients have a nonfunctional AR, which results in the absence of Wolffian duct and urogenital sinus derived structures, and feminized external genitalia (1). In contrast, patients with a 5{alpha}-reductase type2 deficiency have normally developed epididymides and vasa deferentia but have feminized external genitalia (6). Surprisingly, studies in mice deficient for steroid 5{alpha}-reductase 1 and 2, have shown that treatment with nonreducible T analogs was sufficient for secondary sex organ formation and growth (7). Therefore, it seems that in rodents, DHT acts largely as a signal amplifier (5). In experimental setups, the synthetic androgen R1881 (methyltrienolone) is often used because it is not further metabolized and exerts both T and DHT effects.

The AR is a member of the steroid/nuclear receptor superfamily, and functions as a transcription factor upon androgen binding (8). Structurally, the AR protein can be divided into separate domains, e.g. the N-terminal transcription activation domain, the DNA binding domain, and the C-terminal ligand binding domain. The ligand binding domain contains a strictly ligand-dependent transcription activation function and interacts with nuclear cofactors (9, 10). Upon ligand binding, the AR binds to specific genomic androgen response elements, thereby regulating transcription of specific genes.

AR expression in the developing male genital tract occurs in a strict temporal, and cranial to caudal fashion, which is first detected in mesenchymal cells and later in epithelial cells. Mesenchymal AR is expressed along the rat genital tract as early as embryonic d 14 (E14), whereas epithelial expression in the epididymis starts at E18 in the caput epididymis and after birth in the cauda epididymis and vas deferens (11). Thus, initiation of androgen-dependent differentiation of the Wolffian duct system into epididymis and vas deferens occurs before epithelial cells express a detectable level of AR protein. During development, mesenchymal cells are important androgen targets that elicit androgenic effects in epithelial cells via paracrine factors and mesenchymal-epithelial interactions (12, 13).

Morphologically, differentiation of the epididymis is characterized by growth and heavy coiling of the epithelial duct. In molecular and mechanistic terms, however, it is not completely understood which factors are involved in growth and differentiation of the Wolffian duct. We have recently described the effect of androgen on protein expression in a mouse fetal vas deferens (MFVD) cell line using a proteomics approach. Stimulation of MFVD cells with androgen resulted in posttranslational modification of two actin binding proteins, mElfin and CArG-binding factor A (CBF-A) (14). In the present investigation, we have applied proteomics to study and identify proteins that are involved in the androgen-dependent development of fetal rat epididymis and vas deferens, using freshly isolated or cultured tissues. Proteins regulated during fetal development were categorized as cytoskeletal proteins, nuclear proteins, transport proteins, chaperones, and glycolytic enzymes. Androgen regulation in vitro involved posttranslational modification of nuclear proteins and glycolytic enzymes, as well as RhoGDI1, a signaling molecule. Posttranslational modification of proteins by androgen stimulation may be a key regulatory event in genital tract development, suggesting a functional link between signaling pathways and cytoskeletal proteins.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals
Pregnant female Wistar rats (16–20 wk old) were purchased from Harlan Winkelmann GmbH (Bohren, Germany); the company determined the day of gestation by checking the presence of a vaginal plug in the morning after mating, which is referred to as E0. The rats were housed for 1–3 d in our in house facilities under standard animal housing conditions in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. This study was approved by the Dutch Experimental Animal Committee, Protocol No. 124-01-04.

Tissue preparation
Rats were killed by CO2 gas and cervical dislocation. Fetuses were removed from the uteri and placed in PBS on ice. Epididymides and vasa deferentia were isolated as a whole from male fetal rats on E17, E18, E19, E19.5, E20, and E21. Dissection was performed in PBS in a sterile hood. All tissues were either immediately frozen on dry ice or fixed in Bouin’s solution. In addition, tissues from fetuses of E19 were used for organ culture studies. Epididymides, together with vasa deferentia, were placed in a drop of medium on a Millicell-CM 0.4-µm culture plate insert (Millipore Corp., Bedford, MA) that was floating on top of 0.5 ml medium (DMEM/F12 + 2% vol/vol charcoal-stripped fetal calf serum, 10 µg/ml insulin, 10 µg/ml transferrin, and a mixture of penicillin (100 IU/ml), streptomycin (100 µg/ml), and fungizone (0.6 µg/ml), all from Sigma (St. Louis, MO) in a four-well plate (Nunc, Roskilde, Denmark) (15). The two epididymides and vasa deferentia from one male were always cultured in either the absence of hormone (0.1% vol/vol ethanol vehicle) or in the presence of 10 nM of the synthetic androgen R1881, methyltrienolone (NEN Life Science Products, Boston, MA), for 9, 24, or 48 h. For each culture condition, 25 tissues were collected. For the frozen material, 40 tissues of E17 and E18 each and 25 tissues of E19-E21 were collected.

Two-dimensional gel electrophoresis (2DE)
All collected tissues from one time point were pooled and lysed in 500 µl 2D lysis buffer [7 M urea, 2 M thiourea, 4% wt/vol 3-[3-(cholamidopropyl)-dimethyl-ammonio]-1-propane sulfonate, 40 mM Tris-HCl (pH 8.8), 40 mM dithiothreitol (DTT) and 0.5% vol/vol immobilized pH gradient (IPG) buffer (pH 3–10)] using the Sample Grinding kit (Amersham Biosciences, Piscataway, NJ). High purity PlusOne chemicals were from Amersham Biosciences. Insoluble proteins and cell debris were pelleted at 100,000 rpm for 10 min at 4 C, using the Optima TLX Tabletop Ultracentrifuge (Beckman, Palo Alto, CA). Protein concentration was determined using the RC DC Protein Assay kit (Bio-Rad, Hercules, CA) and either 75 or 100 µg protein from total tissue lysate was used for isoelectric focusing (IEF). IEF was performed in an IPGphor apparatus according to the instructions of the manufacturer (Amersham Biosciences) using 24-cm nonlinear IPG strips (pH 3–10) (Amersham Biosciences). The strips were actively rehydrated overnight in rehydration buffer (8 M urea, 2% wt/vol 3-[3-(cholamidopropyl)-dimethyl-ammonio]-1-propane sulfonate, 40 mM DTT, and 0.5% vol/vol IPG buffer) containing the sample applying a voltage of 30 V/h (Vh). After rehydration, the strips were run until 100 kVh was reached. Following IEF, the strips were equilibrated in sodium dodecyl sulfate (SDS) buffer [6 M urea, 2% wt/vol SDS, 30% vol/vol glycerol, 50 mM Tris-HCl (pH 8.8)] containing 65 mM DTT for 15 min and thereafter in SDS buffer containing 135 mM iodoacetamide, for 2 min. 2DE was performed on 10% vol/vol Duracryl gels (Genomic Solutions, Perkin-Elmer, Boston, MA) using the Ettan Dalt gel caster and electrophoresis device (Amersham Biosciences) at a constant power of 3 W for 30 min and 180 W for 3–4 h, depending on the amount of gels run simultaneously, at 25 C. Gels were fixed overnight in 40% vol/vol methanol, 5% vol/vol ortho-phosphoric acid, and stained with colloidal Coomassie using a Colloidal Blue staining kit (Invitrogen, Paisley, Scotland, UK). Images were scanned with the Bio-Rad GS800 densitometer and analyzed using the PDQuest software package (Bio-Rad).

Mass spectrometry
Peptide samples were prepared as described previously (14). Briefly, proteins of interest were manually excised from the gel, destained twice with 30% vol/vol acetonitrile (ACN) in 50 mM ammoniumhydrogen carbonate for 15 min, dried in a Speed Vac Plus (Savant, NY) for 30 min and enzymatically digested overnight using sequencing grade trypsin (Promega, Madison, WI). Peptides were eluted from the gel with 30% vol/vol ACN/0.1% vol/vol trifluoric acid, and 0.5 µl peptide sample was mixed with 4 volumes matrix solution (2 mg {alpha}-cyano hydroxycinnamic acid in 100% ACN), which was spotted on an AnchorChip target plate (Bruker Daltonik, Bremen, Germany). Mass spectra were generated on a Biflex III MALDI-TOF-MS (matrix-assisted laser desorption/ionization time-of-flight mass spectrometer) (Bruker Daltonik) and peptide fingerprints were analyzed using BioTools (Bruker Daltonik) and Mascot software (Matrix Science, London, UK) and the National Center for Biotechnology Information database.

Immunoprecipitation, SDS-PAGE, and Western immunoblotting
Immunoprecipitation was performed on 25 µg protein from total tissue lysate from rat E19 tissues cultured in the absence or presence of 10 nM R1881. Goat antirabbit-agarose beads were coupled to RhoGDI{alpha} antibodies for 1 h at room temperature, cell lysate in PBS was added and the mixture rotated for 2 h at 4 C. The agarose-antibody precipitate was washed with PBS and the pellet was resolved in 25 µl Laemmli sample buffer for SDS-PAGE. Total tissue lysate (12.5 µg protein) or immunoprecipitate were loaded on 10% vol/vol and 15% vol/vol acrylamide gels. Gels were run on the mini PROTEAN gel system (Bio-Rad) and proteins were blotted to a nitrocellulose membrane (Schleicher & Schuell, Keene, NH). Western immunoblotting was performed using polyclonal antibodies against {alpha}-fetoprotein (clone E19, 1:1000) and RhoGDI{alpha} (clone A20, 1:1000) both from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Proteins were visualized by chemiluminescence detection (Western Lightning, Perkin-Elmer, Boston, MA).

Histology and immunohistochemistry
Tissues were fixed in Bouin’s solution for 24 h and were then washed in 70% vol/vol ethanol for 24 h. The fixed tissues were first embedded in 2% wt/vol agar before they were embedded in paraffin. Paraffin sections were cut at 7 µm and stained with hematoxylin/eosin. For immunohistochemistry, paraffin sections from E17 and E21 were incubated with an {alpha}-fetoprotein antibody at a dilution of 1:500. Paraffin sections of E19 tissues cultured either in the absence or presence of R1881 for 48 h were used for immunohistochemistry with the RhoGDI{alpha} antibody (1:500). Antibody incubations were performed overnight at 4 C. Secondary antibodies were coupled to peroxidase using the Strept ABComplex/HRP (Dako A/S Denmark), visualized with DAP stain (Pierce, Rockford, IL) and counterstained with hematoxylin.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Analysis of growth-related proteins in developing epididymis and vas deferens
Epididymides and vasa deferentia from developing rats were isolated as a whole on E17, E18, E19, E19.5, E20, and E21. The developmental changes in these tissues are illustrated in Fig. 1Go. On E17, the epididymis is relatively small and underdeveloped. The epithelial duct consists of single layer of cuboidal cells with a small lumen, and no coiling of the duct is apparent (Fig. 1Go, A and C). On E21, the epididymis has grown much larger and the epithelial duct shows multiple coiling, which is apparent in the figure as multiple small cross-sections through the duct (Fig. 1DGo). It is obvious that marked tissue remodeling is needed to achieve such large changes in tissue structure, and this may require, and result in, differential expression and modification of a variety of proteins. To get more information about the spectrum and identity of proteins involved in growth-related tissue remodeling, we used a 2DE-driven proteomic approach, to identify proteins that are differentially regulated in time during this process of growth. Tissue lysates from the combined epididymides and vasa deferentia isolated on E17-E21 were used to separate proteins on 2D gels. All samples were run either in duplicate or triplicate. Protein expression profiles from E17-E21 were compared using the PDQuest software package.



View larger version (93K):
[in this window]
[in a new window]
  		FIG. 1. Rat epididymis and vas deferens on E17 and E21. Whole mount epididimis and vas deferens photographed on E17 (A) and E21 (B). On E17, the Wolffian duct organs have not fully developed yet, which is clearly visible by the presence of Müllerian duct remnants next to the vas deferens. On E21, the development of the Wolffian duct tissues is obvious by the gain of size and heavily coiling of the epithelial duct of the epididymis. C and D, Histological sections of the tissues from A (E17) and B (E21). Hematoxylin/eosin staining. MD, Müllerian duct; e, epithelium; m, mesenchyme; im, interstitial mesenchyme. Scale bar in all pictures, 500 µm.

 
Using MALDI-TOF-MS, we have identified 20 distinct proteins (numbered in Fig. 2Go) representing at least 30 protein isoforms (Table 1Go). Analysis of the Coomassie-stained 2D gels, showed up-regulation of 36 protein isoforms (Fig. 2Go, red circles) and down-regulation of 17 protein isoforms (Fig. 2Go, green circles), which were consistently regulated throughout all duplicate gels and statistically significant. Inconsistently regulated proteins spots (e.g. doublet of spots 10 o’clock of spots 9) were excluded from the analysis. Furthermore, for three proteins we observed a shift in isoelectric point (pI) or other changes in one or more of the protein isoforms during development (Fig. 2Go, blue circles). These changes most probably represent posttranslational protein modifications.



View larger version (113K):
[in this window]
[in a new window]
  		FIG. 2. 2DE gel analysis of total tissue lysates (100 µg protein) from epididymis/rat vas deferens at E17 and rat E21. Proteins were separated by IEF on pH 3–10 gel strips, and according to molecular mass (MM) on SDS-PAGE. Proteins that were down-regulated during development are encircled in green, up-regulated proteins are encircled in red, and the proteins that shifted in position are encircled in blue. The protein encircled in black did not change in expression. All proteins identified by MALDI-TOF-MS are numbered. Proteins that were identified to represent isoforms of one and the same protein, have the same number (2 4 7 9 10 13 ).

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. Identification of proteins by peptide mass fingerprinting1

 
Analyzing the gels in more detail revealed that temporal up- or down-regulation of the proteins occurred gradually. One of the proteins that was regulated in time was {alpha}-fetoprotein precursor, AFP (spot 1, Fig. 2Go). AFP occurred in the 2D gel as a 68-kDa multi-isoform protein and was down-regulated in time (Fig. 3AGo). However, two separate and fragmented isoforms of AFP were up-regulated in time. The up-regulation of the smaller AFP fragments was gradual but peaked at E19.5, after which the expression was going down again. The expression pattern of one of the AFP fragments is shown in Fig. 3DGo. Tropomyosin 5 was also up-regulated with a peak intensity at E19.5 (Fig. 3CGo). MALDI-TOF-MS revealed that fragmentation not only occurred for AFP but also for serum albumin precursor (SA), ß-actin, and {alpha}-tubulin. Fragmentation of these proteins has been reported in relation to particular physiological roles (16, 17, 18). All the SA fragments were up-regulated gradually in time, of which one is illustrated in Fig. 3BGo.



View larger version (90K):
[in this window]
[in a new window]
  		FIG. 3. Zoom-in regions of E17-E21 2DE gels. A, Down-regulation of one of the full length {alpha}-fetoprotein (AFP) precursor isoforms (spot 1 according to Fig. 2Go). B, Up-regulation of one of the serum albumin precursor fragments (spot 4 according to Fig. 2Go). C, Up-regulation of tropomyosin 5 (spot 10 according to Fig. 2Go). D, Up-regulation of an AFP fragmented isoform (spot 2 according to Fig. 2Go), with peak intensity at E19.5. All tissue lysates were run on duplicate gels and all gels were analyzed using the PDQuest software package. Corresponding arbitrary spot intensities of encircled (A) and boxed (B–D) proteins are represented as bars in the adjacent histograms. Proteins were selected to be up- or down-regulated at least 2-fold with a significance of P < 0.05, as calculated by PDQuest software.

 
The presence of transferrin (spot 5, Fig. 2Go), and the regulation of AFP (spots 1 and 2, Fig. 2Go) and SA fragments (spots 4, Fig. 2Go), raised the question whether the tissue lysates contained serum proteins from infiltrating capillaries. To address this question, immunohistochemistry for AFP was performed on E17 and E21 tissues (Fig. 4Go, A and B). The staining shows that there is no detectable AFP in the epithelial duct, nor in the mesenchymal cells directly surrounding the duct, but there is positive staining in the cytoplasm of interstitial mesenchymal cells on E17. On E21, AFP staining in the interstitial mesenchyme was more abundant and some staining was also observed in the mesenchyme directly surrounding the epithelial duct. The staining was uniform and not restricted to capillaries, indicating that the tissue lysates contain serum proteins originating from the (interstitial) mesenchymal cells. Furthermore, no clear-cut difference in subcellular localization of the AFP staining between E17 and E21 tissue sections was visible. Western blot analysis of the tissue lysates showed a slight decrease of the two main bands of AFP (68 kDa) and an increase of some of the fragments, in agreement with the data of the 2D gels (Fig. 4CGo).



View larger version (97K):
[in this window]
[in a new window]
  		FIG. 4. Immunohistochemistry for AFP of rat E17 (A) and E21 (B) epididymis sections. On E17, AFP stained positive in the interstitial mesenchyme. On E21, some staining was also visible in the mesenchyme directly surrounding the epithelial duct. Scale bar, 100 µm (A and B), and 50 µm (A-1, A-2, and B-1); e, epithelium; m, mesenchyme; im, interstitial mesenchyme. C, Western immunoblot for AFP. Total tissue lysate (12.5 µg protein) from rat E17 and E21 epididymis/vas deferens was separated on SDS-PAGE and immunoblotted. Note the down-regulation of the two 68-kDa AFP form and up-regulation of some of the fragments (AFPf).

 
Analysis of androgen-regulated proteins in cultured epididymis and vas deferens
During isolation of the epididymides and vasa deferentia from the different developmental stages, it became apparent that the largest morphological changes occurred between E19 and E20. Therefore, epididymides and vasa deferentia were isolated on E19 and cultured in either the absence or presence of synthetic androgen R1881, for 9, 24, and 48 h, to analyze regulation of protein expression induced by androgen. At the time points mentioned, photographs were taken and some tissues were used for histology. Culturing the tissues for 9 h in the absence or presence of R1881 did not result in macro- (Fig. 5AGo) and microscopical (Fig. 5BGo) changes. After 9 h of culture, the epithelial duct is clearly visible as a lighter part of the tissues, and shows some minor coiling starting in the epididymis (Fig. 5GoA-1, 2). However, 48 h of culture in the absence of androgen resulted in regression of the epithelial duct, as demonstrated by the absence of the lighter duct in the tissue (Fig. 5GoA-3). In histological sections, some remnants of the epithelial duct are still present (Fig. 5GoB-3). Culturing the tissues in the presence of androgen for 48 h resulted in stabilization of the epithelial duct, and in most cases in increased coiling of the duct as a sign of differentiation (Fig. 5GoA-4). In histological sections, this coiling and stabilization is visible as cross-sections at three different positions (Fig. 5GoB-4). Comparing the sections of Figs. 1BGo and 5GoB-4 illustrates that development of the epididymis in vitro during a culture period of 48 h, although much slower, occurs in a similar pattern as in vivo. Furthermore, we also noticed some swelling of the tissues in culture, irrespective of the presence or absence of androgen.



View larger version (86K):
[in this window]
[in a new window]
  		FIG. 5. Rat E19 vas deferens and epididymis tissues. A, Photographs of E19 tissues cultured in the absence (panels 1 and 3) or presence (panels 2 and 4) of 10 nM synthetic androgen R1881 for 9 h (panels 1 and 2) and 48 h (panels 2 and 4). Scale bar, 100 µm. B, Histological sections of E19 epididymis cultured in the absence (panels 1 and 3) or presence (panels 2 and 4) of 10 nM R1881 for 9 h (panels 1 and 2) and 48 h (panels 2 and 4). Scale bar, 100 µm. Hematoxylin/eosin staining. Epi, Epididymis; vas, vas deferens; e, epithelium; m, mesenchyme. Note that the epithelial duct regressed in the absence of androgen, whereas it differentiated in the presence of R1881.

 
Although morphological differences in the cultured tissues occurred visibly within 48 h (and to a lesser extent already visible after 24 h), we expected that most molecular changes, which eventually will lead to morphological differences, will precede these changes. Therefore, 2DE was performed on whole tissue lysates from organs cultured in the absence or presence of androgen and the gels were analyzed using PDQuest. At first instance, no major regulation in protein expression was observed between the different culture conditions, nor compared with the in vivo situation. However, a closer examination of the Coomassie-stained gels revealed that five protein spots were regulated upon androgen treatment. Four of these proteins showed regulation of one or more isoforms indicating posttranslational modification [RhoGDI, triosephosphate isomerase (TPI), glyceraldehyde 3-phosphate dehydrogenase (G3PD), and heterogeneous nuclear ribonucleoprotein (hnRNP) A2/B1; Fig. 6AGo spots a–c and e, and Table 1Go], and one protein was visibly up-regulated (hnRNPA3; Fig. 6AGo, spot d). Comparison of these results with Fig. 2Go and Table 1Go showed that TPI, G3PD, hnRNP A2/B1, and hnRNP A3 were also regulated in the in vivo situation.



View larger version (61K):
[in this window]
[in a new window]
  		FIG. 6. 2DE gels of tissue lysates (75 µg protein) from rat E19 epididymis/vas deferens. A, 2DE gels of tissues cultured for 9 h in the absence (-R) or presence (+R) of 10 nM synthetic androgen, R1881. B, Zoom-in region B of gels from tissues cultured for 0, 9, or 48 h in the absence or presence of 10 nM R1881. Regulation of a second RhoGDI1 isoform, spot a*. C, Zoom-in region C of gels from tissues cultures for 0, 9, or 48 h in the absence or presence of 10 nM R1881. Regulation of two triosephosphate isomerase (TPI) isoforms, spots b and b*. D, Arbitrary spot intensities of TPI isoforms b and b* during 48 h of tissue culture in the absence or presence of 10 nM R1881. In the absence of R1881, spot b is down-regulated and spot b* is up-regulated. Each point in the graph represents the mean (SD) intensity of spots from two separate gels.

 
RhoGDI1 was regulated by R1881 at the posttranslational level in vitro (Fig. 6BGo). One protein isoform of RhoGDI1 was identified in E19 tissues that were either directly frozen or cultured. However, culturing the tissues in the presence of androgen for 48 h resulted in the appearance of a second, more acidic, protein isoform that was also identified by MALDI-TOF-MS as RhoGDI1. The glycolytic enzyme TPI was regulated by R1881 in vitro as well as in vivo at the level of posttranslational modification (Fig. 6CGo). In the presence of androgen (E19, t = 0 and +10 nM R1881 for 9 h and 48 h), expression of a more basic isoform was relatively high, although the intensity decreased with increasing culture time. Androgen withdrawal resulted in a further decrease in the expression of the basic isoform, whereas a more acidic isoform appeared (Fig. 6Go, C and D). Apparently, androgens play a role in maintaining TPI in a more basic isoform.

RhoGDI1 is an interesting candidate protein to analyze in more depth because it was reported that this protein can increase transcriptional activity of the AR (19). We questioned whether addition of androgen resulted in a change in subcellular localization of RhoGDI1 and whether the second isoform represented a differential phosphorylated isoform because phosphorylation of RhoGDI1 has been reported (20). Immunohistochemical analysis of epididymis cultured in the absence of androgen for 48 h, confirmed the cytoplasmic localization that was already reported (21). As expected, no nuclear localization of RhoGDI1 was observed after androgen treatment (Fig. 7Go, A and C and B and D). Although the epithelial duct regresses in the absence of R1881, RhoGDI1 is still expressed in the cytoplasm of the remnants of the duct, and also in the surrounding mesenchyme. Western immunoblot analysis of tissue lysates showed an equal expression of RhoGDI under all culture conditions (Fig. 7EGo). Immunoprecipitation of RhoGDI1 from the lysates with a RhoGDI1 antibody and subsequent immunoblot analysis with a phospho-serine antibody confirmed that RhoGDI1 is a phosphoprotein (Fig. 7FGo).



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 7. RhoGDI1 expression. Immunohistochemistry for RhoGDI1 in rat E19 epididymis cultured in the absence (A and C) or presence (B and D) of 10 nM R1881 for 48 h. Scale bar, 100 µm (A and B), and 50 µm (C and D). E, RhoGDI1 Western immunoblot of E17 (lane 1), E19 (lanes 2–7), and E21 (lane 8) epididymis/vas deferens tissue lysates. RhoGDI1 protein expression was equal under all culture conditions. F, RhoGDI1 immunoprecipitation and phospho-serine Western immunoblot. Lane 1, E21 total lysate input; lane 2, E19 no hormone, 48 h; lane 3, E19 + 10 nM R1881, 48 h; lane 4, E21. Phosphorylated RhoGDI1 was detected after immunoprecipitation in lanes 2–4. In lane 1, several other phosphorylated proteins were detected on the total tissue lysate.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Androgen-controlled growth and differentiation of the epididymis and vas deferens involves regulation of specific proteins. It was our aim to identify proteins that are involved in growth and differentiation, and more specifically proteins that are regulated by androgen during this process. The dramatic change in morphology of the epididymis, and to a lesser extent of the vas deferens, during the last phase of the fetal period, was the starting point for the present study.

Identification of growth-related proteins
Using 2DE driven proteomics, 20 proteins were identified that are regulated during fetal growth and development of the rat epididymis and vas deferens. The proteins identified can be categorized into enzymes ({alpha}-enolase, TPI, G3PD, probable thioredoxin peroxidase 2), cytoskeletal proteins ({alpha}-tubulin, ß-actin, tropomyosin 5, F-actin capping protein {alpha}-2), transport proteins (AFP, SA, transferrin, fibrinogen ß/{gamma}), RNA binding proteins (hnRNP A2/B1/A3), and chaperones (Hsp-47 precursor). Furthermore, fragmented isoforms of SA, AFP, {alpha}-tubulin, and ß-actin were identified, which were up-regulated in time during genital tract development.

AFP is a specific fetal glycoprotein that is secreted by embryonic tissues and is involved in up- and down-regulation of cell growth (22), in immunosuppression, and in apoptosis (23). AFP belongs to the proteins encoded by the serum albumin multigene family, which present three structurally homologous domains I, II, and III (16). Fragmentation of AFP has been reported, with cleavage products of 32 and 38 kDa resulting from domains I and III. These products can be further fragmented into proteolytically stable isoforms of 23 and 26 kDa (17). A similar kind of fragmentation was described for SA, but with many more proteolytic cleavage products possible (24). These reported fragments of AFP correspond in size to the fragments that were identified in the present study. Furthermore, others have described that distinct AFP isoforms accumulate at different stages of fetal development (25, 26) and that AFP can interact with nuclear receptors (27). Thus, the up-regulation of AFP fragmented isoforms may be physiologically relevant for growth and differentiation of epididymis and vas deferens.

Glycolytic enzymes interact directly ({alpha}-enolase, G3PD), or indirectly (TPI), with structural proteins and complexes like actin and microtubules (28, 29). It was recently described for TPI that it binds to the actin-binding protein cofilin, which in turn binds to Na,K-ATPase upon phosphorylation by the {rho}-signaling pathway (30). TPI takes part in production of energy, which is transduced to the cytoskeleton that modulates cell function, proliferation, and differentiation (30). The cytoskeletal proteins actin and tubulin, and in particular specific cleavage of these proteins, have been implicated in several regulatory processes, including apoptosis (18) and other processes such as meiosis resumption in starfish oocytes (30) Our finding that fragments of ß-actin and {alpha}-tubulin accumulate taken together with the observation of posttranslational regulation of glycolytic enzymes could reflect the processes of vigorous tissue remodeling during the process of genital tract development.

Identification of androgen-regulated proteins during growth
Because it is known that androgen action is essential for growth and development of the male genital tract, it was our aim to identify androgen-regulated proteins that are involved in this process. Epididymides and vasa deferentia cultured in vitro in the absence and presence of androgen were subjected to 2DE, and the differentially expressed proteins identified were compared with the proteins identified in vivo. Glycolytic enzymes G3PD and TPI, and nuclear proteins hnRNP A2/B1/A3 were identified to be regulated by androgen, as was also observed in vivo. In addition, RhoGDI1 was identified as a protein that was regulated by androgen at the level of posttranslational modification. Although the regulation of these proteins occurs under the action of androgen, we cannot conclude that this is the result of a direct and/or indirect androgen action. Furthermore, we did not observe any obvious regulation of the AR protein in the 2D gels. However, the AR protein is of lower abundance and therefore not easily detectable in such a broad range 2D gel.

Rho guanine nucleotide dissociation inhibitor 1(RhoGDI1 or RhoGDI{alpha}), is a cytoplasmic protein originally identified as a negative regulator of RhoGTPases (21). RhoGTPases are molecular switches that cycle between an active membrane-associated GTP-bound state and an inactive cytoplasmic GDP-bound state. This switch is carefully controlled by exchange factors, activating proteins, and dissociation inhibitors (31). RhoA, Rac, and Cdc42 belong to the family of RhoGTPases, which regulate many signal transduction pathways, including those linked to the actin cytoskeleton, microtubule dynamics, vesicular transport dynamics, regulation of cell polarity, gene transcription, G1 cell cycle progression, and a variety of enzymatic activities (31, 32). Members of the RhoGDI family block GDP dissociation from RhoGTPases and control cytoplasmic localization of RhoGTPases. In addition, it was described that RhoGDI1 specifically increases the transcriptional activity of estrogen receptors (ER{alpha}, ERß), glucocorticoid receptor and AR (19). This activation is mediated via repression of RhoGTPases, which demonstrates that the {rho}-mediated signaling pathway is an important regulator of ERs, glucocorticoid receptor, and AR transcriptional activity (19). Although RhoGDI1 is ubiquitously expressed in all tissues, RhoGDI1 knockout mice show specific and progressive impairment of kidneys and reproductive organs probably as a result of the destruction of the actin cytoskeleton (33), meaning that the function of RhoGDI1 in the kidney and reproductive organs is nonredundant.

The testis of RhoGDI1 -/- mice reveals structural abnormalities, the number of germ cells is dramatically decreased and mature sperm cells are only rarely detected in seminiferous tubules and epididymides (33). RhoGDI1 -/- female mice have an intrinsic defect in their reproductive system, which is most evident in the postimplantation development of RhoGDI1 -/- embryos. For the epididymis, no functional developmental defect in the knockout mice was reported (33).

Bourmeyster et al. (20) have detected RhoGDI1 on a 2D gel as a single protein spot with a pI of 5.1, which represents the free, unbound form of RhoGDI. In the case were RhoGDI was associated with RhoA GTPase, they detected two RhoGDI protein spots with pI 4.6–5.1, which could be reduced to one protein spot after phosphatase treatment (20). Furthermore, they have shown that RhoA-RhoGDI1 association depends on RhoGDI1 phosphorylation. Thus, phosphorylation and dephosphorylation of RhoGDI1 determines the activation state of RhoA GTPase (20). In the present tissue culture experiments, 48 h of androgen treatment resulted in the appearance of a second, more acidic, RhoGDI1 isoform that could represent a phosphorylated isoform. If RhoGDI1 is indeed phosphorylated by androgens, this could lead to complex formation with Rho GPTase and repression of the Rho GTPase activity. Taken the fact that RhoGDI1 enhancement of AR transcriptional activity occurs via RhoGTPase repression (19), one could postulate that RhoGDI1 phosphorylation leads to enhanced AR transcription activation, and as a result regulation of RhoGDI1 activity by phosphorylation and dephosphorylation might function as a feedback loop in androgen signaling.

hnRNPs form a large family of proteins that are categorized on the basis of structural/functional motifs, and of which the A and/or B type are the most abundant hnRNP proteins (34). Besides a large number of posttranscriptional isoforms, the hnRNP A and/or B proteins also show extensive posttranslational modifications. For a specific subset of hnRNPs, hnRNP A2, hnRNP A3, and hnRNP-related DNA binding protein 40/CBF-A, an interaction with nuclear actin was described, and an interaction was suggested for cytoplasmic actin (34). In a previous study, we have identified CBF-A, which is a minor variant of hnRNP A2, to be regulated by androgen at the level of posttranslational modification in the mouse fetal vas deferens cell line MFVD (14). This coincides with the present finding that androgens regulate hnRNPs posttranslational modification in genital tract tissues.

Glycolytic enzymes, like TPI and G3PD, are dependent on androgen action for their activity, as was demonstrated by castration experiments. Enzymatic activity was found to be decreased in rat epididymis and monkey seminal vesicle after castration, which was restored after androgen replacement (35, 36, 37). Our present results suggest that androgenic regulation of glycolytic enzyme activity is controlled at the level of posttranslational modification.

In summary, our study shows that changes in protein expression profiles can be detected on 2D gels, when developing genital tract tissues are followed in time in vivo. Analyses of androgen-regulated proteins in epididymis and vas deferens tissues exposed to androgen, also revealed several changes in protein expression profiles when grown in vitro. In the present system, androgen action occurs mainly at the level of posttranslational modification. We hypothesize that during the process of androgen stimulation, which will lead to growth and differentiation of the male genital tract, posttranslational modification of glycolytic enzymes regulates their activity and their association with cytoskeletal proteins. Furthermore, androgen-induced posttranslational modification of RhoGDI1 and glycolytic enzymes may be an important functional link between signaling pathways and cytoskeletal rearrangements in control of growth and development of the male internal genital tract.


   Acknowledgments
 
The authors thank Albert Klooster for taking care of the animals.


   Footnotes
 
Abbreviations: ACN, Acetonitrile; AFP, {alpha}-fetoprotein; AIS, androgen insensitivity; AR, androgen receptor; 2D, two-dimensional; 2DE, 2D gel electrophoresis; DHT, dihydrotestosterone; DTT, dithiothreitol; E, embryonic day; ER, estrogen receptor; G3PD, glyceraldehyde 3-phosphate dehydrogenase; hnRNP, heterogeneous nuclear ribonucleoprotein; IEF, isoelectric focusing; IPG, immobilized pH gradient; MALDI-TOF-MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometer; MFVD, mouse fetal vas deferens; pI, isoelectric point; SA, serum albumin precursor; SDS, sodium dodecyl sulfate; T, testosterone; TPI, triosephosphate isomerase.

Received March 31, 2003.

Accepted for publication June 11, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Quigley CA, De Bellis A, Marschke KB, el-Awady MK, Wilson EM, French FS 1995 Androgen receptor defects: historical, clinical, and molecular perspectives. Endocr Rev 16:271–321[CrossRef][Medline]
  2. Hughes IA 2001 Minireview: sex differentiation. Endocrinology 142:3281–3287[Abstract/Free Full Text]
  3. McLachlan RI, O’Donnell L, Meachem SJ, Stanton PG, de Kretser DM, Pratis K, Robertson DM 2002 Identification of specific sites of hormonal regulation in spermatogenesis in rats, monkeys, and man. Recent Prog Horm Res 57:149–179[Abstract/Free Full Text]
  4. Jost A 1953 Problems of fetal endocrinology: the gonadal and hypophyseal hormones. Recent Prog Horm Res 8:379–418
  5. Wilson JD, Leihy MW, Shaw G, Renfree MB 2002 Androgen physiology: unsolved problems at the millennium. Mol Cell Endocrinol 198:1–5[CrossRef][Medline]
  6. Wilson JD, Griffin JE, Russell DW 1993 Steroid 5{alpha}-reductase 2 deficiency. Endocr Rev 14:577–593[CrossRef][Medline]
  7. Mahendroo MS, Cala KM, Hess DL, Russell DW 2001 Unexpected virilization in male mice lacking steroid 5{alpha}-reductase enzymes. Endocrinology 142:4652–4662[Abstract/Free Full Text]
  8. Gobinet J, Poujol N, Sultan C 2002 Molecular action of androgens. Mol Cell Endocrinol 198:15–24[CrossRef][Medline]
  9. Berrevoets CA, Doesburg P, Steketee K, Trapman J, Brinkmann AO 1998 Functional interactions of the AF-2 activation domain core region of the human androgen receptor with the amino-terminal domain and with the transcriptional coactivator TIF2 (transcriptional intermediary factor2). Mol Endocrinol 12:1172–1183[Abstract/Free Full Text]
  10. Berrevoets CA, Umar A, Brinkmann AO 2002 Antiandrogens: selective androgen receptor modulators. Mol Cell Endocrinol 198:97–103[CrossRef][Medline]
  11. Bentvelsen FM, Brinkmann AO, van der Schoot P, van der Linden JE, van der Kwast TH, Boersma WJ, Schroder FH, Nijman JM 1995 Developmental pattern and regulation by androgens of androgen receptor expression in the urogenital tract of the rat. Mol Cell Endocrinol 113:245–253[CrossRef][Medline]
  12. Cunha GR, Alarid ET, Turner T, Donjacour AA, Boutin EL, Foster BA 1992 Normal and abnormal development of the male urogenital tract. Role of androgens, mesenchymal-epithelial interactions, and growth factors. J Androl 13:465–475[Abstract/Free Full Text]
  13. Thomson AA 2001 Role of androgens and fibroblast growth factors in prostatic development. Reproduction 121:187–195[Abstract]
  14. Umar A, Luider TM, Berrevoets CA, Grootegoed JA, Brinkmann AO 2003 Proteomic analysis of androgen-regulated protein expression in mouse fetal vas deferens cells. Endocrinology 144:1147–1154[Abstract/Free Full Text]
  15. Emmen JM, McLuskey A, Adham IM, Engel W, Grootegoed JA, Brinkmann AO 2000 Hormonal control of gubernaculum development during testis descent: gubernaculum outgrowth in vitro requires both insulin-like factor and androgen. Endocrinology 141:4720–4727[Abstract/Free Full Text]
  16. Nishio H, Dugaiczyk A 1996 Complete structure of the human {alpha}-albumin gene, a new member of the serum albumin multigene family. Proc Natl Acad Sci USA 93:7557–7561[Abstract/Free Full Text]
  17. Dudich I, Tokhtamysheva N, Semenkova L, Dudich E, Hellman J, Korpela T 1999 Isolation and structural and functional characterization of two stable peptic fragments of human {alpha}-fetoprotein. Biochemistry 38:10406–10414[CrossRef][Medline]
  18. Asumendi A, Andollo N, Boyano MD, Hilario E, Perez-Yarza G, Atencia R, Arechaga J, Garcia-Sanz M 2000 The role of cleavage of cell structures during apoptosis. Cell Mol Biol 46:1–11
  19. Su LF, Knoblauch R, Garabedian MJ 2001 Rho GTPases as modulators of the estrogen receptor transcriptional response. J Biol Chem 276:3231–3237[Abstract/Free Full Text]
  20. Bourmeyster N, Vignais PV 1996 Phosphorylation of Rho GDI stabilizes the Rho A-Rho GDI complex in neutrophil cytosol. Biochem Biophys Res Commun 218:54–60[CrossRef][Medline]
  21. Fukumoto Y, Kaibuchi K, Hori Y, Fujioka H, Araki S, Ueda T, Kikuchi A, Takai Y 1990 Molecular cloning and characterization of a novel type of regulatory protein (GDI) for the rho proteins, ras p21-like small GTP-binding proteins. Oncogene 5:1321–1328[Medline]
  22. Dudich E, Semenkova L, Gorbatova E, Dudich I, Khromykh L, Tatulov E, Grechko G, Sukhikh G 1998 Growth-regulative activity of human {alpha}-fetoprotein for different types of tumor and normal cells. Tumour Biol 19:30–40
  23. Semeniuk DJ, Boismenu R, Tam J, Weissenhofer W, Murgita RA 1995 Evidence that immunosuppression is an intrinsic property of the {alpha}-fetoprotein molecule. Adv Exp Med Biol 383:255–269[Medline]
  24. Feldhoff RC, Ledden DJ 1983 Evidence for the spontaneous formation of interspecies hybrid molecules of human, rat and bovine serum albumins. Biochem Biophys Res Commun 114:20–27[CrossRef][Medline]
  25. Wong LT, Lu CY, Tinker DO, Hsia JC 1988 Application of high-performance liquid chromatography for the study of the microheterogeneity changes of mouse {alpha}-fetoprotein in fetal development. J Biochem Biophys Methods 15:267–272[CrossRef][Medline]
  26. Keel BA, Harms RL, Leal JA, Cho S 1990 Characterization of human {alpha} fetoprotein charge microheterogeneity during fetal development. Mol Reprod Dev 27:281–287[CrossRef][Medline]
  27. Dauphinee MJ, Mizejewski GJ 2002 Human {alpha}-fetoprotein contains potential heterodimerization motifs capable of interaction with nuclear receptors and transcription/growth factors. Med Hypotheses 58:453–461[CrossRef][Medline]
  28. Walsh JL, Keith TJ, Knull HR 1989 Glycolytic enzyme interactions with tubulin and microtubules. Biochim Biophys Acta 999:64–70[CrossRef][Medline]
  29. Knull HR, Walsh JL 1992 Association of glycolytic enzymes with the cytoskeleton. Curr Top Cell Regul 33:15–30[Medline]
  30. Jung J, Yoon T, Choi EC, Lee K 2002 Interaction of cofilin with triose-phosphate isomerase contributes glycolytic fuel for Na, K-ATPase via Rho-mediated signaling pathway. J Biol Chem 277:48931–48937[Abstract/Free Full Text]
  31. Etienne-Manneville S, Hall A 2002 Rho GTPases in cell biology. Nature 420:629–635[CrossRef][Medline]
  32. Hall A 1998 Rho GTPases and the actin cytoskeleton. Science 279:509–514[Abstract/Free Full Text]
  33. Togawa A, Miyoshi J, Ishizaki H, Tanaka M, Takakura A, Nishioka H, Yoshida H, Doi T, Mizoguchi A, Matsuura N, Niho Y, Nishimune Y, Nishikawa S, Takai Y 1999 Progressive impairment of kidneys and reproductive organs in mice lacking Rho GDI{alpha}. Oncogene 18:5373–5380[CrossRef][Medline]
  34. Percipalle P, Jonsson A, Nashchekin D, Karlsson C, Bergman T, Guialis A, Daneholt B 2002 Nuclear actin is associated with a specific subset of hnRNP A/B-type proteins. Nucleic Acids Res 30:1725–1734[Abstract/Free Full Text]
  35. Brooks DE 1976 Activity and androgenic control of glycolytic enzymes in the epididymis and epididymal spermatozoa of the rat. Biochem J 156:527–537[Medline]
  36. Brooks DE 1976 Control of glycolytic enzymes by androgens in the rat epididymis. J Endocrinol 71:355–365[Abstract/Free Full Text]
  37. Arunakaran J, Balasubramanian K, Srinivasan N, Aruldhas MM, Govindarajulu P 1993 Effects of prolactin and androgens on enzymes of carbohydrate metabolism in seminal vesicles of castrated mature bonnet monkeys, Macaca radiata. Biochem Mol Biol Int 31:211–218[Medline]



This article has been cited by other articles:


Home page
 EndocrinologyHome page
D. A. Pujianto, A. E. Damdimopoulos, P. Sipila, J. Jalkanen, I. Huhtaniemi, and M. Poutanen
Bfk, a Novel Member of the Bcl2 Gene Family, Is Highly Expressed in Principal Cells of the Mouse Epididymis and Demonstrates a Predominant Nuclear Localization
Endocrinology, July 1, 2007; 148(7): 3196 - 3204.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
G. Frenette, J. Girouard, and R. Sullivan
Comparison Between Epididymosomes Collected in the Intraluminal Compartment of the Bovine Caput and Cauda Epididymidis
Biol Reprod, December 1, 2006; 75(6): 885 - 890.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
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 Umar, A.
Right arrow Articles by Brinkmann, A. O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Umar, A.
Right arrow Articles by Brinkmann, A. O.
Right arrowPubmed/NCBI databases
*Substance via MeSH

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