Expression of Enzymes Synthesizing (Aldehyde Dehydrogenase 1

0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 10 Printed in U.S.A.

Expression of Enzymes Synthesizing (Aldehyde Dehydrogenase 1 and Retinaldehyde Dehydrogenase 2) and Metabolizing (Cyp26) Retinoic Acid in the Mouse Female Reproductive System* ´ RIE FRAULOB, PASCAL DOLLE ´, JULIEN VERMOT, VALE KAREN NIEDERREITHER

AND

Institut de Geńe´tique et de Biologie Molećulaire et Cellulaire, Centre National de la Recherche Scientifique, Institut National de la Sante´ et de la Recherche Me´dicale, Universite´ Louis Pasteur, Colle`ge de France, BP 163, 67404 Illkirch Cedex, CU de Strasbourg, France ABSTRACT Vitamin A is required for female reproduction. Rodent uterine cells are able to synthesize retinoic acid (RA), the active vitamin A derivative, and express RA receptors. Here, we report that two RAsynthesizing enzymes [aldehyde dehydrogenase 1 (Aldh1) and retinaldehyde dehydrogenase 2 (Raldh2)] and a cytochrome P450 (Cyp26) that metabolizes vitamin A and RA into more polar metabolites exhibit dynamic expression patterns in the mouse uterus, both during the ovarian cycle and during early pregnancy. Aldh1 expression is up-regulated during diestrus and proestrus in the uterine glands, whereas Raldh2 is highly induced in the endometrial stroma in metestrus. Cyp26 expression, which is not detectable during the normal ovarian cycle, is strongly induced in the uterine luminal epithe-

V

ITAMIN A (RETINOL) and its active metabolite retinoic acid (RA) serve dual roles in the female reproductive tract, by controlling epithelial differentiation and allowing for reproductive fertility. In the case of vitamin A deficiency, the uterine epithelium forms regions of keratinized squamous metaplasia, indicating a role for RA in the maintenance of the simple columnar endocervical and uterine epithelia (1, 2). Ovariectomy seems to modify the phenotype and survival rate of vitamin-A-deficiency animals (2). In rodent species, estrogen-dependent keratinization of the vagina and cervical junction epithelium occurs normally during the estrus phase of the ovarian cycle (Ref. 3, and references therein). Physiological withdrawal of RA has been suggested to be required for this vaginal/cervical keratinization to occur (4, 5). Indeed, several lines of evidence suggest that endogenous RA production and signaling is hormonally regulated in the female genital tract. The uterine RA Received March 9, 2000. Address all correspondence and requests for reprints to: Pascal Dolle´ or Karen Niederreither, Institut de Geńe´tique et de Biologie Molećulaire et Cellulaire, Centre National de la Recherche Scientifique, Institut National de la Sante´ et de la Recherche Me´dicale, Universite´ Louis Pasteur, Colle`ge de France, BP 163, 67404 Illkirch Cedex, CU de Strasbourg, France. E-mail: [email protected] or [email protected]. * This work was supported by funds from Centre National de la Recherche Scientifique, Institut National de la Sante´ et de la Recherche Me´dicale, Colle`ge de France, the Hoˆpitaux Universitaires de Strasbourg, the Association pour la Recherche sur le Cancer, the Fondation pour la Recherche Me´dicale, and Bristol-Myers Squibb Co.

lium, 24 h after human CG hormonal administration. Raldh2 stromal expression also strongly responds to gonadotropin (PMSG and human CG) induction. Furthermore, Raldh2 expression can be hormonally induced in stromal cells of the vagina and cervix. All three enzymes exhibit differential expression profiles during early pregnancy. Aldh1 glandular expression is sharply induced at 2.5 gestational days, whereas Raldh2 stromal expression increases more steadily until the implantation phase. Cyp26 epithelial expression is strongly induced between 3.5– 4.5 gestational days, i.e. when the developing blastocysts colonize the uterine lumen. These data suggest a need for precise regulation of RA synthesis and/or metabolism, in both cycling and pregnant uterus. (Endocrinology 141: 3638 –3645, 2000)

content has been found to increase after administration of PMSG or estrogen (6). The expression of cellular retinolbinding protein (CRBP I), cellular RA-binding proteins (CRABP I and II), RA receptors (RARs), and retinoid X receptors also fluctuate in response to hormonal stimulation in the uterus, cervix, and vagina (3, 6 –10). However, the complex expression profiles of these proteins do not clearly indicate where and when retinoid-dependent signaling does occur. RA administration was found to reduce the estrogeninduced proliferation of rat uterine stromal and myometrial cells (11), suggesting that endogenous retinoids may protect against the appearance of female reproductive tract tumors. Because retinoids are being clinically tested for their efficacy in gynecological cancer treatment and chemoprevention (Ref. 12, and references therein), a better understanding of the tissue-specific profiles of RA production and their regulation by sex hormones may help to define cancer treatment strategies. Here, we analyzed the expression of two RA-producing and a retinoid-metabolizing enzyme within murine female reproductive tissues. Aldehyde dehydrogenase 1 (Aldh1) and retinaldehyde dehydrogenase 2 (Raldh2) are two members of class I aldehyde dehydrogenases, which act to produce RA from its precursor retinaldehyde (13, 14). Raldh2 is specifically expressed during early mouse embryogenesis at sites of RA production (15, 16), whereas Aldh1 is expressed in specific organs during later development and adult life (Ref. 17, and references therein). Targeted knockout of the mouse Raldh2 gene deprives

3638

RA-METABOLIZING ENZYMES IN FEMALE GENITAL TRACT

the embryo of RA, resulting in midgestational lethality (18). Cyp26 is a RA-inducible member of the cytochrome p450 (Cyp) family, which selectively transforms RA into polar metabolites (4-OH-, 18-OH-, 4-oxo-RA, and other unidentified products), which are considered as elimination forms (19, 20). However, it can also oxidize vitamin A into the biologically active metabolite 4-oxo retinol (21, 22). We found that all three enzymes exhibit distinct expression patterns, which fluctuate during the normal ovarian cycle and can be induced by gonadotropin hormonal stimulation. Moreover, these enzymes exhibit stageand tissue-specific differential expression patterns during early pregnancy, likely indicating a need for regulated RA production before and during embryo implantation. Materials and Methods The 129/Sv mouse strain was used for this study (except for the analysis of gestational stages, which was performed on CD1 mice, whose pregnancy rate is higher). Ovarian cycle stages were identified by analysis of vaginal smears and subsequent histological analysis (23). Tissues (ovary, oviduct, uterus, and vagina) were collected in the morning or evening of each of the 4 days of the cycle (proestrus, estrus, metestrus, and diestrus). For gonadotropin inductions, mice were killed 48 h after an ip injection of 5 IU PMSG, or at various time points (see Results) after a second ip injection of 2.5 IU human CG (hCG). For the analysis of early pregnancy stages, natural matings were performed, and mice were killed in the morning of each gestational day (designated as X.5 days post coitum (dpc), fertilization being assumed to take place during the night).

FIG. 1. Modulation of the Aldh1 and Raldh2 uterine expression patterns during the ovarian cycle. In situ hybridization was performed on series of adjacent sagittal uterine sections. A–C, Proestrus; D–F, estrus; G–I, metestrus; J–L, diestrus. The transcript patterns (Aldh1: B, E, H, and K; Raldh2: C, F, I, and L) are viewed under dark-field illumination, which shows the signal grain as white dots. A, D, G, and J show one of the corresponding bright-field views.

3639

Each type of analysis was performed on at least two independent animals. Animals were treated in accordance with the European directives (CEE 86/609). In situ hybridization analysis was performed with 35S-labeled antisense or sense RNA probes. The Raldh2 and Cyp26 complementary DNAs (cDNAs) have been previously described (13, 24), and the Aldh1 cDNA was cloned after PCR amplification from P19 cell RNA (our unpublished data). The in situ hybridization procedure was as described (25). Emulsion autoradiography exposure time was 3 weeks. As expected, control sense probes only gave uniform background labeling (data not shown).

Results

We first analyzed whether Aldh1, Raldh2, and Cyp26 are expressed in the female mouse genital tract during the normal ovarian cycle (Fig. 1). Both Aldh1 (Fig. 1, B, E, H, and K) and Raldh2 (Fig. 1, C, F, I, and L) exhibited cycling patterns of expression, although in distinct uterine cell populations (Fig. 2, A–D) and during distinct phases of the cycle. Aldh1 was expressed at high levels in the endometrial glands during proestrus (Figs. 1B; and 2, A and B). Expression was down-regulated in estrus (Fig. 1E), and reappeared at low levels in the uterine epithelium during metestrus (Fig. 1H), to increase again in the glandular epithelium during diestrus (Fig. 1K). Raldh2 expression, on the other hand, was restricted to of the endometrial stroma (Fig. 2, C and D). Raldh2 transcripts were rather weakly expressed in proestrus but

3640


FIG. 2. Cell type-specificity of Aldh1, Raldh2, and Cyp26 uterine expression. High-power views of the endometrium from sections hybridized with Aldh1 (A, B, proestrus), Raldh2 (C, D, proestrus), or Cyp26 (E, F, 24 h post-hCG) probes. Compare the histology (left) with the signal grain distribution on the corresponding dark-field views (right). ep, Luminal epithelium; gl, uterine glands; st, stroma.

exhibited a graded distribution with higher levels in cells underlying the epithelium (Figs. 1C; and 2, C and D). Raldh2 expression was almost completely down-regulated in estrus (Fig. 1F) but was strongly induced during the metestrus phase (Fig. 1I). Expression decreased again and was quite heterogeneous along the endometrial stroma at the diestrus stage (Fig. 1L). Cyp26 transcripts were not expressed at detectable levels in the cycling uterus (n ⫽ 18 mice, analyzed in the morning and evening of each day of the cycle; data not shown). Expression of these enzymes was also analyzed in the vagina, which undergoes major epithelial changes during the

Endo • 2000 Vol. 141 • No. 10

estrous cycle (23). Of the three enzymes studied, only Raldh2 exhibited regulated expression in the vagina. Expression was undetectable in proestrus and estrus but was detected in the lamina propria cells underlying the epithelium in metestrus and, at lower levels, in diestrus (data not shown; see next section). Aldh1 and Cyp26 transcripts were not detected in the vagina (data not shown). The hormonal regulation taking place during the estrous cycle can be artificially induced by injection of gonadotropin hormones. The well-defined protocol used for mice consists of an injection of PMSG, which has a follicle-stimulating activity, followed after 48 h by an injection of hCG, which induces ovulation after about 12 h and subsequent luteinization (Ref. 26, and references therein). We therefore analyzed uteri collected 48 h after PMSG injection (i.e. in the estrogenic proliferative phase) and 24 h, 3 days and 5 days after the hCG injection (i.e. during the progesterone secretory and pseudopregnant phase) (Fig. 3). As seen during the normal estrous cycle, Aldh1 (Fig. 3, B, F, J, and N) and Raldh2 (Fig. 3, C, G, K, and O) were expressed in the uterine glandular epithelium and the uterine stroma, respectively. After PMSG injection, Aldh1 was expressed at rather high levels in the uterine glands and at very low levels throughout the rest of the uterine epithelium (Fig. 3B). Glandular expression was lower 24 h and 3 days after hCG injection (Fig. 3, F and J), but it markedly increased 5 days after hCG (Fig. 3N). Raldh2 transcripts exhibited a different profile, given that these were strongly expressed in the stroma 48 h after PMSG (Fig. 3C) and decreased 24 h after hCG (Fig. 3G), to be maximally induced again 3 days after hCG (Fig. 3K). Expression was low and concentrated toward subepithelial cells at 5 days post hCG (Fig. 3O). Cyp26 transcripts were not detected after PMSG treatment (Fig. 3D) but were strongly induced in the uterine luminal epithelium 24 h after hCG injection (Figs. 2, E and F; and 3H). Expression levels were extremely low at 3 and 5 days post hCG (Fig. 3, L and P). This hormonal induction, which has no apparent counterpart during the normal estrous cycle, may reflect Cyp26 up-regulation that takes place during early pregnancy (see below). Raldh2 transcripts were also regulated in the vagina of hormonally-treated mice (Fig. 4). After PMSG stimulation, weak expression was detected in the lamina propria cells underlying the epithelium (Fig. 4B). Raldh2 expression was undetectable 24 h after hCG (Fig. 4D) but was strongly induced 3 days after hCG treatment (Fig. 4F). Two days later, Raldh2 expression disappeared from the vagina proper but persisted toward the cervical region (Fig. 4H). No expression of Aldh1 and Cyp26 was found in the vagina after hormonal treatments (data not shown). The uterine endometrium undergoes further changes during early pregnancy (Fig. 5) to become receptive for the implantation of the developing blastocysts, which takes place at 3.5– 4.5 dpc in the mouse. The most conspicuous alterations are a local apoptosis of the epithelium around the implanting blastocysts and a marked stromal cell proliferation or decidual reaction (Fig. 5, M and Q; and data not shown). Interestingly, all three RA-metabolizing enzymes exhibited distinct spatial and temporal profiles of expression during early mouse gestation. Aldh1 transcripts were almost


3641

FIG. 3. Gonadotropin hormonal modulation of Aldh1, Raldh2, and Cyp26 uterine expression patterns. In situ hybridization was performed on a series of adjacent sagittal sections of uteri collected 48 h after an ip injection of 5 IU PMSG (A–D) and 24 h (E–H), 3 days (I–L) or 5 days (M–P) after a second ip injection of 2.5 IU hCG. The bright-field views (A, E, I, and M) correspond to the Raldh2-hybridized sections.

undetectable on the first day of pregnancy (Fig. 5B; 0.5 dpc). However, they were induced, by 2.5 dpc, throughout the uterine epithelium and, more intensely, in the uterine glands (Fig. 5F). Expression levels decreased at 3.5– 4.5 dpc (Fig. 5, J and N) and exhibited a second increase in the uterine glands by 5.5– 6.5 dpc (Fig. 5R, and data not shown). Aldh1 expression was also detected in the decidual cells surrounding the implanting blastocysts at 4.5 dpc (Fig. 5N). Raldh2 expression was very weak and was restricted to the most superficial stromal cells by 0.5 dpc (Fig. 5C). Expression increased and extended throughout the deeper stroma at 2.5 and 3.5 dpc (Fig. 5, G and K). From 4.5– 6.5 dpc, Raldh2 transcripts became progressively excluded from the inner decidual cells (Fig. 5, O and S). Thus, at 6.5 dpc, strong expression was found in the peripheral stromal cells surrounding each conceptus (Fig. 5S). Cyp26 transcripts were not detected by 0.5 dpc (Fig. 5D). Expression appeared in the uterine epithelium by 2.5 dpc (Fig. 5H) and further increased during the next gestational days, to reach maximal intensity by 4.5 dpc (compare Fig. 5, H, L, and P). Strong expression persisted through-

out the uterine epithelium at 5.5– 6.5 dpc, both between and at the level of each implantation site (Fig. 5T, and data not shown). Differential expression of the RA-producing enzymes was also found in the mouse ovary, and oviduct. Aldh1 was expressed at high levels in ovarian interstitial cells, including those forming the theca of growing follicles (Fig. 6, A and B; note that dense signal grain appears in black in the bright-field view). Expression was also seen in corpus luteum cells (Fig. 6B). Weak Aldh1 expression was specifically detected in the ampullar region of the oviduct epithelium (Fig. 6B). Raldh2 was expressed at low levels throughout ovarian interstitial cells, whereas stronger expression was seen in the outermost cell layer (most likely in mesothelial cells of the tunica albuginea; Fig. 6C). Cyp26, which was not expressed at detectable levels in the ovary, was strongly expressed in the oviduct epithelium, except at the level of the ampulla (Fig. 6D). These ovarian and oviductal expression patterns did not significantly evolve during the estrous cycle or during early pregnancy (data not shown).

3642


FIG. 4. Gonadotropin hormonal induction of Raldh2 expression in the vaginal and cervical stromal cells. Sagittal sections of the vagina and cervix were analyzed 48 h after PMSG injection (A and B) and 24 h (C and D), 3 days (E and F), or 5 days (G and H) after subsequent hCG injection. The cervical region is oriented toward the right side.

Discussion

Previous studies have investigated retinoid levels in the female reproductive tract by direct spectroscopic analysis after HPLC (6, 27), analysis of the vitamin A-associated autofluorescence (Ref. 28, and references therein), or indirectly by detection of the various retinoid-binding proteins (6, 9, 10, 29). These data suggest that uterine retinoid levels vary with the menstrual cycle, in a manner that is regulated by sex hormones (notably estrogen). Studies on primary cell cultures indicated that the uterine luminal epithelial cells have the highest ability to synthesize RA, in comparison with stromal, myometrial, or glandular cells (6). This RA-synthesizing ability was enhanced after hormonal induction by PMSG or estrogen (6). Because the same treatments increase CRABP II expression in the uterine surface epithelial cells, the presence of this protein has been proposed as a hallmark of RA-producing or storing cells. However, because the knockout of the CRABP II gene does not overtly affect mouse embryonic development or postnatal functions (30), the sig-

Endo • 2000 Vol. 141 • No. 10

nificance of CRABP II expression, with respect to RA signaling, remains unclear. Here, we show that the two RA-producing enzymes Aldh1 and Raldh2 are expressed in the uterine glands and stromal cells, respectively, rather than in the luminal epithelium (Fig. 2). Although the RA-synthesizing ability of uterine epithelial cells (6) could be explained by expression of Aldh1, or another Raldh activity, the lack of RA synthesis by isolated stromal cells (6), which express Raldh2, seems more intriguing. An explanation could be that the stromal cells must be in their normal environment to express their RA-synthesizing ability (e.g. by responding to endocrine hormonal signals, or to paracrine signal(s) from the myometrium and/or epithelium). Disrupting the integrity of the endometrium may thus lead to rapid loss of RA synthesis by stromal cells. Monitoring the expression levels of Raldh2 during the process of stromal cell isolation (6) may help resolve this discrepancy. A previous study concluded that estrogen and progesterone hormones do not regulate the expression of RARs and retinoid X receptors in human endometrial stromal cells (31). Our results rather indicate a hormonal control at the level of ligand synthesis, via the induction of RAproducing enzymes during specific stages of the cycle. We found that both enzymes are essentially down-regulated during the estrus phase (Fig. 1), thus suggesting a transient lack of RA production. A drop in CRBP I levels has also been observed in the same phase, correlating with decreased retinol responsiveness (5). Raldh2 induction, which takes place in metestrus, is consistent with the proposed role of RA to promote differentiation of stromal and myometrial cells, after the estrogen-mediated phase of proliferation (11). Of the three enzymes studied, Raldh2 is the only one showing expression in the vagina and cervix (Fig. 4). Its induction in stromal (lamina propria) cells during metestrus corresponds to the step of sloughing of cornified cells and reduced proliferation in the epithelium. The action of estrogens is required to allow the proliferation and keratinization of the vaginal epithelium that take place, respectively, during the proestrus and estrus phases. Retinoids have been proposed to counteract the effects of estrogen and regulate the transition to a nonkeratinized epithelium, as suggested by: 1) the squamous metaplasia observed after vitamin A deficiency in rats (Ref. 32, and references therein); and 2) the fact that a single RA injection during the estrogenic phase of the cycle (late diestrus/proestrus) can efficiently inhibit keratinization of the vaginal epithelium (33). Raldh2 induced after the estrus phase may be responsible for such a stage-specific RA synthesis. Our data highlight the uterus as a region of high RA production, whereas the vagina and cervix express lower levels of RA-producing enzymes. Perhaps this transition from high to low RA production occurring in the cervix may mark this region as susceptible to the neoplastic effects of estrogens. Tissues subject to estrogen-induced proliferation, yet having a reduced capacity to synthesize RA, may be prone to this type of neoplastic conversion. Interestingly, reductions in the level of expression of the RA-inducible RAR␤ gene have been found by Northern blot analysis of cervical carcinoma cell lines, or in situ hybridization of tissue


3643

FIG. 5. Differential modulation of Aldh1, Raldh2, and Cyp26 uterine expression during early pregnancy. Adjacent sagittal uterine sections are shown at 0.5 (A–D), 2.5 (E–H), 3.5 (I–L), 4.5 (M–P), and 6.5 (Q–T) dpc. The bright-field views (A, E, I, M, and Q) correspond to the Aldh1-hybridized sections. de: Decidua; em: embryo; ep, luminal epithelium; st, stroma; gl, glandular epithelum.

specimens (34, 35). Thus, modulation of Raldh2 expression may be another avenue for cancer chemoprevention. This study also shows a differential regulation of RA-

metabolizing enzyme expression during early mouse gestation (Fig. 5). Aldh1 is sharply up-regulated in the uterine epithelium and glands shortly before implantation (2.5 dpc,

3644


Endo • 2000 Vol. 141 • No. 10

FIG. 6. Expression of Aldh1 (B), Raldh2 (C), and Cyp26 (D) in the ovary and oviduct. The bright-field view (A) corresponds to the section hybridized to the Aldh1 probe. gf, Growing follicles; OV, ovary; OD, oviduct; am, ampulla; af, preovulatory antral follicle; cl, corpus luteum.

Fig. 5F), whereas its down-regulation by 3.5 dpc coincides with the appearance of Cyp26 transcripts in the luminal epithelium. Raldh2 expression in the endometrial stroma also increases by 2.5 dpc (Fig. 5G) and remains high during and after implantation (Fig. 5, K and O). However, this expression is progressively down-regulated as stromal cells undergo the decidual reaction, because only the most peripheral (nondecidual) cells surrounding each implantation site continue to express the gene by 4.5– 6.5 dpc (Fig. 5S). A recent study has shown that the RA-synthesizing ability of rat endometrial cells significantly increases during early gestational days (27). This observation is consistent with the increase of Raldh2 expression during early mouse pregnancy. The same study reported that rat decidual cells, isolated at day 8 (roughly equivalent to day 6.5 in mouse), have RA-synthesizing activity (27). This could be explained if the decidual balls collected by these authors contained peripheral cells that still express Raldh2, or if stable Raldh2 protein persists in differentiating decidual cells even after silencing of the gene. The presence of CRBP I and CRABP II, as well as RAR messenger RNAs, in decidual cells further suggests that, during their differentiation, they have the ability to produce and/or respond to RA (28, 29). In this respect, it is interesting to note that RA treatment has a strong inhibitory effect on the differentiation of cultured human endometrial stromal cells toward decidual cells (36). Altogether, these various data strongly suggest that regulated RA synthesis plays a role in the control of stromal cell differentiation toward decidual cells and, therefore, in the process of embryonic implantation. Cyp26 expression becomes high in the luminal epithelium by the time blastocysts colonize the uterus and start to implant (3.5– 4.5 dpc; Fig. 5, L and P). Whether it may function there to produce biologically active metabolites, such as 4-oxo-retinol (22), or rather to inactivate RA to prevent its release in the uterine fluid and transfer to the embryo, remains to be investigated. Retinol-binding protein (RBP), on the other hand, represents one of the major proteins secreted in the uterine fluid before implantation (Ref. 28, and references therein), probably to deliver vitamin A as a source of retinoids for the early embryo.

The present study provides evidence for regulated patterns of RA synthesis and metabolism in the mouse uterus, both during the normal estrus cycle and during early pregnancy. The precise physiological function(s) of RA signaling in female reproduction remains to be established. Heterozygous Raldh2 knockout female mice, whose enzymatic activity should be reduced, seem to be normally fertile, whereas Raldh2 null mutant embryos are not viable. Thus, it will be important to develop strategies for tissue-specific knockouts in adult mice, to analyze the functions of these various enzymes in reproduction. Acknowledgments We are grateful to Prof. P. Chambon for his constant support and suggestions, J.-M. Garnier for the cloning of Aldh1 cDNA, S. Abu-Abed and M. Petkovich for providing the Cyp26 cDNA, and B. Schuhbaur for excellent technical assistance.

References 1. Wolbach SB, Howe PR 1925 Tissue changes following deprivation of fat soluble A. J Exp Med 42:753–777 2. Ponnamperuma RM, Kirchhof SM, Trifiletti L, DeLuca LM 1999 Ovariectomy increases squamous metaplasia of the uterine horns and suvival of SENCAR mice fed a vitamin A-deficient diet. Am J Clin Nutr 70:502–508 3. Celli G, Darwiche N, DeLuca LM 1996 Estrogen induces retinoid receptor expression in mouse cervical epithelia. Exp Cell Res 226:273–282 4. Tannous-Khuri L, Hillemanns P, Rajan N, Wright TC, Talmage DA 1994 Expression of cellular retinol- and cellular retinoic acid-binding proteins in the rat cervical epithelium is regulated by endocrine stimuli during normal squamous metaplasia. Am J Pathol 144:148 –159 5. Tannous-Khuri L, Talmage DA 1997 Decreased cellular retinol-binding protein expression coincides with the loss of retinol responsiveness in rat cervical epithelial cells. Exp Cell Res 230:38 – 44 6. Bucco RA, Zhen WL, Davis JT, Sierra-Rivera E, Osteen KG,Chaudhary AK, Ong DE 1997 Cellular retinoic acid-binding protein II presence in rat uterine epithelial cells correlates with their synthesis of retinoic acid. Biochemistry 36:4009 – 4014 7. Darwiche N, Celli G, DeLuca LM 1994 Specificity of retinoid receptor gene expression in mouse cervical epithelia. Endocrinology 134:2018 –2025 8. Zhuang YH, Ylikomi T, Lindfors M, Piippo S, Tuohimaa P 1994 Immunolocalization of retinoic acid receptors in rat, mouse and human ovary and uterus. J Steroid Biochem Mol Biol 48:61– 68 9. Bucco RA, Zhen WL, Wardlaw SA, Davis JT, Sierra-Rivera E, Osteen KG, Melner MH, Kakkad BP, Ong DE 1996 Regulation and localization of cellular retinol-binding protein, retinol-binding protein, cellular retinoic acid-binding protein (CRABP), and CRABP II in the uterus of the pseudopregnant rat. Endocrinology 137:3111–3122

RA-METABOLIZING ENZYMES IN FEMALE GENITAL TRACT 10. Wardlaw SA, Bucco RA, Zhen WL, Ong DE 1997 Variable expression of cellular retinol- and cellular retinoic acid-binding proteins in the rat uterus and ovary during the estrous cycle. Biol Reprod 56:125–132 11. Boettger-Tong HL, Stancel GM 1995 Retinoic acid inhibits estrogen-induced uterine stromal and myometrial cell proliferation. Endocrinology 136:2975– 2983 12. Piamsomboon S, Termrugruanglert W, Kudelka AP, Edwards CL, Freedman RS, Mountain CF, Delclos L, Aapro MS, DeCaro L, Bianca J, Verschraegen CF, Kavanagh JJ 1996 Prolonged stabilization of progressive squamous cell cancer of the cervix with interferon-alpha and 13-cis-retinoic acid: report of two cases and review of the literature. Anticancer Drugs 7:800 – 804 13. Zhao D, McCaffery P, Ivins KJ, Neve RL, Hogan P, Chin WW, Dra¨ger U 1996 Molecular identification of a major retinoic-acid synthesizing enzyme: a retinaldehyde-specific dehydrogenase. Eur J Biochem 240:15–22 14. Moore SA, Baker HM, Blythe TJ, Kitson KE, Kitson TM, Baker EN 1999 A structural explanation for the retinal specificity of class 1 ALDH enzymes. Adv Exp Med Biol 463:27–38 15. Niederreither K, McCaffery P, Dra¨ger UC, Chambon P, Dolle´ P 1996 Restricted expression and retinoic acid-induced down-regulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development. Mech Dev 62:67–78 16. Moss JB, Xavier-Neto J, Shapiro MD, Nayeem SM, McCaffery P, Dra¨ger UC, Rosenthal N 1999 Dynamic patterns of retinoic acid synthesis and response in the developing mammalian heart. Dev Biol 199:55–71 17. Bhat PV, Marcinkiewicz M, Li Y, Mader S 1998 Changing patterns of renal retinal dehydrogenase expression parallel nephron development in the rat. J Histochem Cytochem 46:1025–1032 18. Niederreither K, Subbarayan V, Dolle´ P, Chambon P 1999 Embryonic retinoic acid synthesis is essential for early mouse post-implantation development. Nat Genet 21:444 – 448 19. White JA, Beckett-Jones B, Guo YD, Dilworth FJ, Bonasoro J, Jons G, Petkovich M 1997 cDNA cloning of human retinoic acid-metabolizing enzyme (hP450RAI) identifies a novel family of cytochromes P450. J Biol Chem 25:18538 –18541 20. Fujii H, Sato T, Kaneko S, Gotoh OI, Fujii-Kuriyama Y, Osawa K, Kato S, Hamada H 1997 Metabolic inactivation of retinoic acid by a novel P450 differentially expressed in developing mouse embryos. EMBO J 16:4163– 4173 21. Achkar CC, Derguini F, Blumberg B, Langston A, Levin AA, Speck J, Evans RM, Bolado Jr J, Nakanishi K, Buck J 1996 4-oxoretinol, a new natural ligand and transactivator of the retinoic acid receptors. Proc Natl Acad Sci USA 93:4879 – 4884 22. Lane MA, Chen AC, Roman SD, Derguini F, Gudas LJ 1999 Removal of LIF (leukemia inhibitory factor) results in increased vitamin A retinol metabolism to 4-oxoretinol in embryonic stem cells. Proc Natl Acad Sci USA 96:13524 – 13529 23. Yuan YD, Carlson RG 1987 Structure, cyclic change and function, vagina and vulva, rat. In: Jones TC, Mohr U, Hunt RD (eds) Genital System, Monographs on Pathology of Laboratory Animals. Springer-Verlag, Berlin, pp 161–168

3645

24. Abu-Abed SS, Beckett BR, Chiba H, Chithalen JV, Jones G, Metzger D, Chambon P, Petkovich M 1998 Mouse P450RAI (Cyp26) expression and retinoic acid-inducible retinoic acid metabolism in F9 cells are regulated by retinoic acid receptor gamma and retinoid X receptor alpha. J Biol Chem 273:2409 –2415 25. Niederreither K, Dolle´ P 1998 In situ hybridization with 35S-labelled probes for retinoid receptors. In: Redfern C (ed) Methods in Molecular Biology: Retinoids. Humana Press, Totowa, NJ, pp 247–267 26. Hogan B, Beddington R, Costantini F, Lacy E 1994 Manipulating the Mouse Embryo, a Laboratory Manual, ed 2. Cold Spring Harbor Laboratory Press, New York, pp 130 –132 27. Zheng WL, Sierra-Rivera E, Luan J, Osteen KG, Ong DE 1999 Retinoic acid synthesis and expression of cellular retinol-binding protein and cellular retinoic acid-binding protein type II are concurrent with decidualization of rat uterine stromal cells. Endocrinology 141:802– 808 28. Schweigert FJ, Bonitz K, Siegling C, Buchholz I 1999 Distribution of vitamin A, retinol-binding protein, cellular retinoic acid-binding protein I, and retinoid X receptor ␤ in the porcine uterus during early gestation. Biol Reprod 61:906 –911 29. Zheng WL, Ong DE 1998 Spatial and temporal patterns of expression of cellular retinol-binding protein and cellular retinoic acid-binding proteins in rat uterus during early pregnancy. Biol Reprod 58:968 –970 30. Lampron C, Rochette-Egly C, Gorry P, Dolle´ P, Mark M, Lufkin T, LeMeur M, Chambon P 1995 Mice deficient in cellular retinoic acid-binding protein II (CRABPII) or in both CRABPI and CRABPII are essentially normal. Development 121:539 –548 31. Kumarendran MK, Matthews CJ, Levasseur MD, Prentice A, Thomas EJ, Redfern CP 1994 Oestrogen and progesterone do not regulate the expression of retinoic acid receptors and retinoid “X” receptors in human endometrial stromal cells in vitro. Hum Reprod 9:229 –234 32. Jetten AM, DeLuca LM, Nelson K, Schroeder W, Burlingame S, Fujimoto W 1996 Regulation of cornifin ␣ expression in the vaginal and uterine epithelium by estrogen and retinoic acid. Mol Cell Endocrinol 123:7–15 33. Chateau D, Boehm N 1996 Regulation of differentiation and keratin 10 expression by all-trans retinoic acid during the estrous cycle in the rat vaginal epithelium. Cell Tissue Res 284:373–381 34. Geisen C, Denk C, Gremm B, Baust C, Karger A, Bollag W, Schwarz E 1997 High-level expression of the retinoic acid receptor beta gene in normal cells of the uterine cervix is regulated by the retinoic acid receptor alpha and is abnormally down-regulated in cervical carcinoma cells. Cancer Res 15:1460 – 1467 35. Xu XC, Mitchell MF, Silva E, Jeten A, Lotan R 1999 Decreased expression of retinoic acid receptors, transforming growth factor beta, involucrin and cornifin in cervical intraepithelial neoplasia. Clin Cancer Res 5:1503–1508 36. Brar AK, Kessler CA, Meyer AJ, Cedars MI, Jikihara H 1996 Retinoic acid suppresses in vitro decidualization of human endometrial stromal cells. Mol Hum Reprod 2:185–193