| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biochemistry, School of Medicine, Vanderbilt University, Nashville, Tennessee 37232
Address all correspondence and Requests for Reprints to: David E. Ong, Department of Biochemistry, Vanderbilt University, 23rd Avenue at Pierce, Nashville, Tennessee 37232. E-mail: david.e.ong{at}vanderbilt.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, ß, and 
and isoforms). These receptors bind to specific response elements in the promoter regions of target genes and work as ligand-inducible transcription enhancers and repressors (5). In addition to the nuclear receptors, a group of intracellular retinoid-binding proteins, cellular retinol-binding protein (CRBP) and CRBP (II) and cellular retinoic acid-binding protein (CRABP) and CRABP (II) are involved in the movement of retinoids in the cytoplasm, directing their ligands to particular sites for metabolism or function (6, 7, 8, 9, 10, 11). The current belief is that biosynthesis of RA is by two sequential oxidations, first producing retinal from retinol by either retinol dehydrogenases (RolDHs) of the short-chain alcohol dehydrogenase family or cytosolic alcohol dehydrogenases (ADHs). The retinal is then oxidized to RA by retinal dehydrogenases (RalDHs) of the ADH family 1 (also known as Aldh1a1–3). There are several forms of these enzymes in each of these dehydrogenase families that are expressed and regulated spatiotemporally in distinct manners. Recent genetic studies demonstrated that loss of these ADH family members increases the toxicity of very high levels of retinol, albeit levels far from levels expected to be encountered normally. ADH1 is expressed at very high levels in liver, intestine, and kidney, ADH3 is expressed ubiquitously, and both appear to provide some protection against toxic levels of administered retinol (12). ADH4 has been reported to be the most efficient enzyme for oxidation of retinol, but not if the retinol is bound to CRBP (13, 14). RolDHs of the SDR family have been demonstrated to use CRBP-retinol efficiently for the production of retinal (15, 16, 17) and is the family that we have examined here.
Genetic studies with RalDH II and RalDH III have confirmed them as involved in the production of RA for gene regulation (18, 19); RalDH I gene ablation renders mice more susceptible to the toxic effects of very high levels of retinol, as was seen for the ADH family (20, 21). A direct role of RalDH I in generation of RA for signaling has not yet been established. The several members of these families makes it essential to distinguish the specific and physiologically relevant enzymes catalyzing RA production in particular sites because it occurs in many different tissues, suggesting that there will be different factors that regulate its synthesis.
We have reported that administration of estrogen (E2) to prepubertal rats leads to a gain of ability of the uterus to synthesize RA, coincident with the appearance of CRABP (II) in the lumenal epithelium (22). The induction of CRABP (II) expression by E2 is direct (23). We also have described a retinol dehydrogenase, epithelial retinol dehydrogenase (eRolDH), which appears in the lumenal epithelium at this time (24). The human homolog of rat eRolDH has been extensively tested, establishing it as a true retinol dehydrogenase (17, 25). Others have established that pregnant mare serum gonadotropin treatment of the prepubertal mouse leads to the expression of retinal dehydrogenase II (RalDH II) in the uterus, with the message found predominately in stromal cells near the lumenal epithelium (26). Estrogen replacement therapy in humans also up-regulates expression of RalDH II as well as leading to the expression of genes known to be regulated by RA in the endometrium (27). These findings, taken together, suggest that E2 coordinately regulates uterine RA synthesis. To determine whether regulation of the expression of the enzymes necessary for RA synthesis is direct, as was true for CRABP (II) expression, we have examined the effects of E2 administration on message levels of the proposed enzymes and binding proteins in the uteri of ovariectomized rats. Our results indicate that E2 directly induced the expression of eRolDH and RalDH II while repressing expression of RalDH I, thus having direct control over RA synthesis. Immunolocalization of eRolDH, RalDH II, CRBP, and CRABP (II) in the uterus at stages of the estrous cycle further suggested that RA production is compartmentalized, with retinol oxidation occurring in the epithelium and subsequent oxidation of retinal to RA taking place in the underlying stromal cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ovariectomized rats were divided into six groups, each with at least three animals, and injected ip with either corn oil, puromycin (10 mg/rat), 17ß-estradiol in corn oil (E2, 10 µg/rat), puromycin plus E2, all-trans-retinoic acid in corn oil (RA, 500 µg/rat), or puromycin plus RA, respectively. All injection volumes were 0.1 ml. Puromycin was injected 30 min before injection of E2 or RA, for those experimental groups. After 4 h, uteri were harvested from the animals for RNA extraction. These studies were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with the oversight of the veterinarian of our local institutional animal care and use committee.
RNA extraction and RPA [ribonuclease (RNase) protection assays]
Total RNA was extracted from an individual rat uterus using TriZol reagent (Invitrogen Life Technologies, Inc., Rockville, MD) and quantified by spectrophotometry. The antisense riboprobes for rat CRABP(II), CRBP, eRolDH, RolDH2, and cyclophilin were transcribed using T7 polymerase; RalDH I and RalDH II by SP6 polymerase of the MAXIscript kit (Ambion Inc., Austin, TX) and -32UTP (10 Ci/ml; NEN Life Science Products, Boston, MA). The RNase protection assays were carried out using the RPA III kit (Ambion Inc.) according to user’s manual. Briefly, samples of total RNA were hybridized for 15–18 h at 50 C with excess radiolabeled antisense riboprobe (n = 3 individual samples/time point), and digested by RNase at 37 C for 30 min. The hybridized products were submitted to electrophoresis on 6% acrylamide gels containing 8% urea. Gels were exposed to BioMAX MR films (Kodak, Rochester, NY) with intensifying screens for up to 3 d. Loading variation between samples was standardized by including cyclophilin riboprobe in all hybridization reactions.
Preparation of immune reagents/immunohistochemistry
The peptide chosen to generate the antibody for rat RalDH II was CGGKGLGRKGFFIEP, synthesized by PeptidoGenic Research & Co., Inc. (Livermore, CA). The peptide was conjugated with carrier protein mcKLH (Pierce, Rockford, IL) and emulsified with Hunter’s TiterMax Gold adjuvant (CytRx Co., Norcross, GA). It was injected into male New Zealand White rabbits, with boosts at 5, 8, and 11 wk. The immune serum was collected at 12 wk, and the total IgG fraction of this antiserum was isolated by use of protein A column (Pierce), and then applied to recombinant rat RalDH II and SulfoLink Coupling Gel (Pierce) for isolation of the RalDH II specific IgG population.
The tissue samples used for immunolocalization were collected and fixed immediately in a freshly made solution composed of 20% isopropylalcohol, 4% paraformaldehyde, 2% trichloroacetic acid, and 2% zinc chloride at 4 C overnight, then placed in 70% ethanol. Paraffin embedding and sectioning were carried out by the Vanderbilt Histopathology Department (Nashville, TN). The slides were blocked for 30 min in Tris-buffered saline (pH 7.6) containing 0.1% Tween 20, 1% normal goat serum, and 3% crystalline BSA. This blocking agent was also used for dilution of the primary, secondary, and tertiary antibodies. The method for immunohistochemical localization has been previously described (24).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Retinol dehydrogenases examined in the ovariectomized rat
We first screened mRNA isolated from uteri obtained from normal rats, staged to be at estrus, for the known rat retinol dehydrogenases, types 1, 2, and 3 (28, 29, 30) and eRolDH by RT-PCRs specific for each message. Only eRolDH and RolDH2 were positive (results not shown). We then examined the effect of E2 or RA administration to ovariectomized animals on the expression of these two messages. When uteri were collected 4 h post administration, expression of eRolDH had increased an average of 6-fold, compared with control animals, but no effect was observed with RA (Fig. 1
, third panel). Prior treatment of animals with puromycin to block protein synthesis gave similar induction, consistent with the increased expression being by direct action of E2 on the eRolDH gene. The expression of RolDH2 was not significantly changed (Fig. 1, fourth panel). This suggested that eRolDH was the RolDH involved in E2 induction of RA synthesis. Values averaged for three animals are shown in the table (Fig. 1, bottom).
|
, fifth panel), and RalDH II, increased 9.3-fold (Fig. 1, sixth panel), when examined 4 h after administration. RA had no effect on expression of either mRNA. Again, pretreatment with puromycin did not alter the degree of change for either mRNA after E2 administration, indicating that both the repression of RalDH I expression and the stimulation of RalDH II expression was direct.
Retinoid binding proteins examined in the ovariectomized rat
We have previously shown that E2 induces CRABP (II) and RA induces CRBP expression in OVX rat uterus (23). They were analyzed in these samples as positive controls for the action of these two hormones (Fig. 1, first and second panels, respectively).
For the second part of this study, we examined expression in the normal female rat. Physiologically, the E2 level in rat plasma cyclically changes during the normal estrous cycle. To test whether the candidate genes, described above, were responding as expected in the intact animal, uteri from rats were collected at the stages of diestrus, estrus, and metestrus for RNA extraction and analysis by RPA. Additionally, tissue blocks were prepared for immunohistochemical determination of sites of expression. To provide a more complete picture of RA synthetic machinery, we repeated some previous work on the expression of the binding proteins CRBP and CRABP (II), this time analyzing expression by RPA analyses, as opposed to the less sensitive Northern analyses in the previous work (31).
Cellular binding proteins: CRABP(II) and CRBP during the estrous cycle
We have previously shown that CRABP (II) is directly induced by E2, but not by RA, in the uterus of the ovariectomized rat (23). During the normal cycle, there was little detectable CRABP (II) expression in normal rat uterus at metestrus; the expression at diestrus was low but increased at estrus to 71-fold greater than at diestrus (Fig. 2, first row).
|
, second panel).
Retinol dehydrogenases during the estrous cycle
The expression of eRolDH was strikingly high at estrus and metestrus, but below detectable level at diestrus (Fig. 2, third panel), consistent with the results from the ovariectomized rat. However, there was no significant difference between the level of RolDH2 expression in diestrus and estrus; at metestrus expression was decreased to about half of that in diestrus (Fig. 2, fourth panel).
Retinal dehydrogenases
RalDH I was most highly expressed in diestrus but was decreased at estrus to 0.1-fold and metestrus to 0.3-fold, compared with diestrus, consistent with the repression by E2 observed in the ovariectomized animal (Fig. 2, fifth panel). RalDH II expression was highest at estrus being 4-fold greater than that found in diestrus and metestrus (Fig. 2, sixth panel), consistent with the E2 induction observed above.
Immunolocalization of CRBP, CRABP (II), eRolDH, and RalDH II
To examine this system as completely as possible, we prepared sections of fixed uteri from rats at either estrus or diestrus for immunolocalization studies. Although we have shown previously the coordinate expression of CRBP, CRABP (II), and eRolDH proteins at estrus (24), here we add localization of RalDH II and use sequential or closely adjacent sections for analyses to give a more complete pattern of expressed proteins at a particular site in the rat uterus.
In sections from a uterus collected at diestrus, we saw immunohistochemical evidence for CRBP, in the stromal compartment (Fig. 3, IA), consistent with our previous observation. No expression of RalDH II protein was observed under these conditions (Fig. 3, ID) although we had detected the presence of its mRNA at this time. Consistent with our previous report, no expression of CRABP (II) or eRolDH was apparent at diestrus (24).
|
, IIA–C) as previously reported (24). However, RalDH II staining was observed in the stromal compartment, particularly strong in the stratum compactum region close to the lining epithelium (Fig. 3, IID). At higher magnification (Fig. 3, IIIA–D), the staining pattern indicated that the epithelial cells showed no detectable presence of RalDH II in the epithelium (Fig. 3, IIID) but strong staining of stromal cells underlying the lining epithelium. Staining for eRolDH was only detected in the luminal epithelial cells, with no staining evident in these stromal cells (Fig. 3, IIIC).
In another view, the stromal cells with the highest expression of RalDH II seem to be immediately next to the basement membrane that separates them from the epithelial cells (Fig. 4A). Interestingly, preparations of several sections showed a separation of stroma from the epithelium at different sites around the uterus. In all such cases, we observed that the stromal cells immediately adjacent to the epithelial cell layer remained with the epithelium, suggesting a tight association (Fig. 4B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous work had demonstrated that expression of both of these genes increases as a result of appropriate hormonal manipulation to induce estrus: eRolDH in rat (24) and RalDH II in mouse (26). In human, estrogen replacement therapy for 3 months induced expression of RalDH II in the endometrium (27). Here we extend those observations because the rapid induction by E2 in the ovariectomized rat indicates that this induction was by E2 acting directly on these two genes. We found no obvious ERE consensus sequence in the 5 kb of 5'-untranslated region before the initiation site of the rat eRolDH gene. However, the ERE for CRABP (II) also is not obvious. A potential ERE in the promoter of mouse RalDH II gene has been reported (34), and a similar conserved potential ERE binding site was found in the rat RalDH II gene by us (data not shown). The confirmation of direct induction will require further work.
We previously demonstrated that, at estrus in the rat, increased expression of CRABP (II), CRBP, and eRolDH protein occurred in the cells of the lining epithelium of the uterus (24). However, the message for RalDH II has been reported to be in the stromal cells in the mouse uterus (26), raising the question of whether two cell types are required for RA synthesis or that there might be a species difference in the site of uterine RA synthesis.
When we examined closely adjacent sections by immunohistochemistry for presence of these four proteins, prepared from uteri collected at estrus and diestrus from the normally cycling rat, we confirmed their absence at diestrus and their appearance at estrus. But clearly RalDH II was expressed in stromal cells, primarily in the stratum compactum, in contrast to the epithelial localization of the other proteins, with the strongest staining in the cells immediately underlying the epithelium. This was consistent with the localization of RalDH II mRNA reported for the mouse uterus (26). This localization strongly suggested to us that RA synthesis in the uterus requires both the epithelial cells, to convert retinol to retinal, and the underlying stromal cells, to convert retinal to RA. A requirement for two different cells to produce the active hormonal form from a precursor is not without precedent. In particular, ovarian synthesis of E2 itself involves conversion of androstenedione to testosterone in the thecal cells, then movement of testosterone to the adjacent granulosa cells where it is aromatized to E2.
This model must be reconciled with our previous report that the great majority of the synthesis of RA from retinol was observed for epithelial cells isolated and cultured from the uteri of prepubertal rats given E2 12 h before collection, with little synthesis observed for cultured stromal cells (22). The recovery of RA from epithelial cultures was at least 100-fold greater than from cultures of isolated stromal cells from the same pool of uteri. The lack of significant synthesis by stromal cells indicates that there is little ability to convert retinol to retinal to serve as substrate for RalDH II. The robust synthesis by the cultures of epithelial cells is quite possibly due to the fact that it very difficult to avoid contamination with stromal cells. The stromal cells just adjacent to the lining epithelial cells appeared to have the highest level of RalDH II (Fig. 4A). When the stromal and epithelial layers happened to separate (reason not known), those stromal cells closest to the lining epithelial cells stayed with the epithelium (Fig. 4B). This suggests that any stromal cell contamination of epithelial cells would be primarily of these cells, allowing both steps of RA synthesis to occur in the cultures.
We considered the possibility that there might be a previously unknown RalDH (a putative RalDH IV) that would have an epithelial localization. However, extensive work with different RT-PCR strategies with degenerate primers to highly conserved regions of this enzyme family, using a cDNA library prepared from mRNA isolated from rat uteri at estrus, failed to uncover a candidate. This library has been tested for a number of genes expressed at estrus and has been positive for all examined.
Therefore, the different locations of the two enzymes required for RA synthesis adds additional complexity to the regulation of RA synthesis and function, at least at this site. According to our results, we propose that de novo RA synthesis in the uterus (Fig. 5) is initiated in the epithelial cells where retinol is oxidized to retinal by eRolDH; the product retinal then diffuses or is transported by an unknown mechanism to the adjacent stromal cells, where RalDH II converts the retinal to RA. Presumably the RA then leaves the cell for directing differentiation of stem cells and, most likely, acts on gene expression in other cells as well. We are in the process of identifying specific genes regulated by E2-induced RA to clarify the sites of action of RA in the rat uterus.
|
|
Footnotes |
|---|
Abbreviations: ADH, Alcohol dehydrogenase; Aldh1, aldehyde dehydrogenase family 1; CRABP(II), cellular retinoic acid-binding protein(II); CRBP, cellular retinol-binding protein; E2, estrogen; ERE, estrogen response element; eRolDH, epithelial retinol dehydrogenase; RA, retinoic acid; RalDH, retinal dehydrogenases; RAR, retinoic acid receptor; RNase, ribonuclease; RolDH, retinol dehydrogenase; RPA, RNase protection assay.
Received April 21, 2004.
Accepted for publication June 7, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
thymocyte subset with homogeneous T-cell receptors to mucosal epithelia. Nature 343:754–757[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  K. Mitko, S. E Ulbrich, H. Wenigerkind, F. Sinowatz, H. Blum, E. Wolf, and S. Bauersachs Dynamic changes in messenger RNA profiles of bovine endometrium during the oestrous cycle Reproduction, February 1, 2008; 135(2): 225 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. K. Tan, J. Chen, H. Lehnert, R. Kennedy, and H. S. Randeva Raised Serum, Adipocyte, and Adipose Tissue Retinol-Binding Protein 4 in Overweight Women with Polycystic Ovary Syndrome: Effects of Gonadal and Adrenal Steroids J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2764 - 2772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Trasino, E. H. Harrison, and T. T. Y. Wang Androgen Regulation of Aldehyde Dehydrogenase 1A3 (ALDH1A3) in the Androgen-Responsive Human Prostate Cancer Cell Line LNCaP Experimental Biology and Medicine, June 1, 2007; 232(6): 762 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. D. Nadauld, D. N. Shelton, S. Chidester, H. J. Yost, and D. A. Jones The Zebrafish Retinol Dehydrogenase, rdh1l, Is Essential for Intestinal Development and Is Regulated by the Tumor Suppressor Adenomatous Polyposis Coli J. Biol. Chem., August 26, 2005; 280(34): 30490 - 30495. [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 |
