Skip to main content
PMC home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2008 Mar 25;93(6):2366–2374. doi: 10.1210/jc.2007-2813

Estrogen and Progesterone Metabolism in the Cervix during Pregnancy and Parturition

Stefan Andersson 1, Debra Minjarez 1, Nicole P Yost 1, R Ann Word 1
PMCID: PMC2435631  PMID: 18364378

Abstract

Context: Experimental and clinical studies in a variety of nonprimate species demonstrate that progesterone withdrawal leads to changes in gene expression that initiate parturition at term. Mice deficient in 5α-reductase type I fail to undergo cervical ripening at term despite the timely onset of luteolysis and progesterone withdrawal in blood.

Objective: Our objective was to test the hypothesis that estrogen and progesterone metabolism is regulated in cervical tissues during pregnancy, even in species in which parturition is not characterized by progesterone withdrawal in blood.

Design: Estradiol and progesterone metabolism was quantified in intact cervical tissues from nonpregnant and pregnant women at term before or after labor.

Setting: The study was conducted at a university hospital.

Patients: Tissues were obtained from five nonpregnant and 21 pregnant women (nine before labor and 12 in labor).

Main Outcome Measures: Enzyme activity measurements, Northern blot analysis, quantitative real-time RT-PCR, and immunohistochemistry were used to quantify steroid hormone metabolizing enzymes in cervical and myometrial tissues.

Results: During pregnancy, 17β-hydroxysteroid dehydrogenase type 2 was induced in glandular epithelial cells to catalyze the conversion of estradiol to estrone and stroma-derived 20α-hydroxyprogesterone to progesterone. During parturition, 17β-hydroxysteroid dehydrogenase type 2 was down-regulated in endocervical cells, thereby creating a microenvironment favorable for cervical ripening.

Conclusions: Together, the data indicate that cervical ripening during parturition involves localized regulation of estrogen and progesterone metabolism through a complex relationship between cervical epithelium and stroma, and that steroid hormone metabolism in cervical tissues from pregnant women is unique from that in mice.

Cervical ripening during parturition involves localized regulation of estrogen and progesterone metabolism through a complex relationship between cervical epithelium and stroma. Steroid hormone metabolism in the cervix of pregnant women is different from that in mice.

Cervical ripening is a process involving remodeling of the connective tissue of the cervix with a decrease in collagen and proteoglycan concentrations and an increase in water content compared with the nonpregnant cervix. These changes cause a rearrangement of the collagen fibrils and an increased susceptibility to degradation so that the tissue assumes the characteristics of a soft, easily distensible tissue that undergoes dilatation to allow delivery of the fetus. Studies from a number of species suggest that cervical ripening is mediated by prostaglandins and estrogens and progesterone, which ultimately result in increases in collagenase, elastase, metalloproteinases, and cytokine levels necessary for cervical remodeling (1). Despite intense research in this area, the key signaling mechanism involved in cervical ripening or cervical dilation during labor is not known.

Results of experiments conducted with mice deficient in steroid 5α-reductase type 1 indicate that progesterone withdrawal in blood alone is insufficient for normal parturition in mice (2). The studies further indicated that local tissue progesterone metabolism was necessary for successful cervical ripening under physiological conditions. In humans and nonhuman primates, there is no progesterone withdrawal in blood before the initiation of labor. Although there appears to be no alteration in progesterone metabolism, progesterone binding, or sequestration of progesterone in the myometrium to account for a progesterone withdrawal, recent studies indicate significant changes in the relative expression of progesterone receptor isoforms in the human myometrium during labor (3,4,5). However, it is not known whether these changes are sufficient to bring about uterine contractions of labor, or if the alterations in progesterone receptor isoforms precede labor (6). Nonetheless, antiprogestins are effective agents in inducing softening and dilatation of the cervix in women (7). Furthermore, estrogen stimulates collagen degradation, and progesterone blocks estrogen-induced collagenolysis in vitro (8,9). Estrogen- and progesterone-responsive genes are regulated in intrauterine tissues of pregnant women during parturition, with more pronounced changes in the cervix compared with the fundus or lower uterine segment (10). Thus, progesterone responsiveness seems important in the maintenance of cervical integrity during pregnancy, and estrogen appears to antagonize progesterone responses (11).

In this study we tested the hypothesis that localized estrogen and progesterone metabolism in cervical tissue is altered in cervical tissues from pregnant women before or after cervical ripening. We hypothesized that local changes in estrogen and progesterone metabolism in cervical tissues at term result in the biochemical processes of cervical ripening, even in species in which parturition is not characterized by progesterone withdrawal in blood. Here, we report expression of 17β-hydroxysteroid dehydrogenase (HSD) type 2 and 20α-HSDs in the human cervix. Data are presented to support the hypothesis that during pregnancy, expression of 17βHSD type 2 prevents ripening of the cervix by its oxidative 20αHSD activity, thereby maintaining elevated intracervical progesterone levels. During cervical ripening and dilation, 17βHSD type 2 is down-regulated, and, in the presence of reductive 20αHSD enzymes [mainly aldo-keto reductase (AKR) 1C1], results in progesterone inactivation, attenuation of progesterone action, and cervical ripening.

Subjects and Methods

Tissue acquisition

Cervical stroma, endocervical epithelium, lower uterine segment, and fundal myometrial tissues were obtained from pregnant women at cesarean-hysterectomy for various surgical indications (Table 1). Cervical tissues and fundal myometrium were obtained from premenopausal nonpregnant women at hysterectomy for benign gynecological indications. Women with cervical pathology or abnormal pap smears were excluded. All tissues were obtained under protocols approved by the institutional review board. All specimens were obtained immediately in the operating suite, placed on ice, and dissected within 30 min of surgery. In the case of pregnancy, hysterectomies were conducted within minutes of delivery of the placenta and fetal membranes. For enzyme activity measurements, intact tissues were finely minced in Hank’s Balanced Salt Solution and placed in DMEM culture media. Simultaneously, tissues were placed in RNAlater (Ambion, Inc., Austin, TX), and snap frozen in liquid nitrogen and stored at −80 C until RNA isolation. The implantation site was noted, and all specimens were obtained remote from identified pathology. In tissues obtained from women with placenta previa, the implantation site did not involve the cervix. The lower uterine segment was defined as the region between the internal cervical os and penetration of the uterine artery. Endocervical epithelial cells were removed from the endocervical canal using a Cytobrush (Medscand Inc., The Cooper Companies, Inc., Lake Forest, CA) to gently lift mucosa from the underlying stroma. Cervical stroma was obtained from the outer circumference of the cervix, avoiding endocervical glands. Squamous epithelium of the exocervix was not obtained. Samples for immunohistochemistry were fixed in 10% formalin.

Table 1.

Clinical characteristics of pregnant women from whom samples were obtained for steroid hormone metabolism studies

Not in labor Labor
No. of cesarean hysterectomies 7 5
Mean gestational age ± sem (wks) 37.0 ± 0.3 39.6 ± 0.90
Median parity (range) 3 (2–5) 3 (1–4)
Surgical indications 4 Repeat C/S; placenta previa/accreta 2 Repeat C/S; leiomyomas
1 Repeat C/S; failed tubal ligation 2 Repeat C/S placenta accreta
1 Primary C/S transverse lie; uterine atony 1 Breech; leiomyomas
1 Repeat C/S; hemorrhage from implantation site
Open in a new tab

C/S, Cesarean section. 

Materials

[1,2,6,7-3H]Progesterone (104 Ci/mmol), [2,4,6,7-3H]estradiol (98 Ci/mmol), and [α-32P]deoxycytidine triphosphate (3000 Ci/mmol) were purchased from Perkin-Elmer (Boston, MA). Steroid standards were from Steraloids (Newport, RI) (12). 4-MA, 17β-(N,N,-diethyl)carbamoyl-4-methyl-4-aza-5α-androstan-3-one was a kind gift from Merck Research Laboratories (Rahway, NJ).

Enzyme assays

Tissue minces (100 mg) were incubated in 500 μl DMEM media containing 10% fetal bovine serum and radiolabeled substrate (30 nm) for 2–24 h at 37 C in a shaking water bath. To determine, or prohibit, the formation of 5α-reduced progesterone metabolites, incubations with 10 μm of the steroid 5α-reductase types 1 and 2 specific inhibitor 4-MA were incubated in side-by-side reactions. The reaction was stopped with 10 volumes of dichloromethane, after by extraction; the organic phase was collected and evaporated to dryness under nitrogen. The residual material was dissolved in chloroform/methanol (2:1) and subjected to thin layer chromatography using Whatman Partisil LK5D chromatoplates (Clifton, NJ). The mobile phase was chloroform/ethyl acetate (3:1). The chromatoplates were scanned, and radioactivity was quantitated on a BioScan 200 Imaging System (Washington, DC) (13). Percent conversion of products was calculated as counts per minute in the 20α-hydroxyprogesterone peak divided by the sum of counts per minute in both the 20α-hydroxyprogesterone and progesterone peaks. The same method was used to quantify estrone production in experiments using estradiol as the substrate. The identities of the products were determined by comparison to the migration of known standards. Microscopical analysis of eight experiments (four before labor and four in labor) revealed that contamination of stromal cells in endocervical scrapings was minimal and not different among experiments.

Immunohistochemistry

Formalin-fixed, paraffin-embedded tissues were sectioned at 5 μm and mounted on slides together with tissue sections from positive and negative control sections (e.g. placenta, brain). Sections were stained with hematoxylin and eosin to confirm preservation of the histological structure and to validate normalcy. For immunostaining, slides were dried in a microwave oven, then deparaffinized in xylene and graded alcohols to distilled water, and 0.1 m citrate buffer (pH 6.0) epitope retrieval was performed in a pressure cooker for 9 min. Endogenous biotin activity was then blocked by placing the slides in dilute egg white solution for 15 min at room temperature, followed by a distilled water rinse and then incubation in fresh skim milk at room temperature for 15 min. Thereafter, slides were incubated with monoclonal antibody (mAb) (mAb-C2–12) directed against 17βHSD type 2 (14) for 30 min at 25 C using constant gentle orbital rotation. Tissue sections incubated with McCoy’s medium in place of the primary antibody served as negative controls. After rinsing three times in PBS, slides were incubated 30 min with a biotinylated antimouse IgG antibody (Scytek, Logan, UT) for 15 min, followed by a 10-min incubation in 0.3% H2O2 in PBS to quench endogenous peroxidase activity. After rinsing three times in PBS, the slides were incubated with horseradish peroxidase-conjugated streptavidin (Scytek) for 15 min at 25 C. The reaction product was developed by immersing slides in prepared diaminobenzidine solution (Research Genetics, Huntsville, AL) for 4 min at 32 C, then rinsed in water, and placed in 0.5% copper sulfate in PBS for 5 min at 25 C to enhance the appearance of chromogen. Finally, slides were rinsed in water, counterstained in hematoxylin, dehydrated in graded alcohols and xylene, and protected with a coverslip.

RNA preparation

Tissues were pulverized in liquid nitrogen and then homogenized in guanidinium isothiocyanate as described (10). Total RNA was isolated from cervical stroma, endocervical epithelium, lower uterine segment, and fundal myometrium, as previously described. Concentration of RNA was measured and purity confirmed by absorption spectroscopy at 260/280 nm.

RT

RT reactions were conducted with 2 μg total RNA in a reaction volume of 20 μl. Each reaction contained 10 mm dithiothreitol, 0.5 mm deoxynucleotide triphosphates, 0.15 μg/μl random primers, 40 U RNase inhibitor (no. 10777-019; Invitrogen Corp., Carlsbad, CA), and 200 U reverse transcriptase (no. 18064-014; Invitrogen). Reaction conditions were: 10 min at 23 C, 60 min at 42 C, followed by 15 min at 70 C.

Quantitative real-time PCR

To ensure accurate quantification of gene expression within a given tissue type and across various tissues, quantitative real-time PCR was used. Primer and probe sequences for amplifications were chosen using published cDNA sequences and the Primer Express program (Applied Biosystems, Foster City, CA) (Tables 2 and 3). TaqMan probes were used for detection of human AKR1C1-AKR1C4 amplicons. Gene expression was normalized to expression of 18S rRNA (no. 4310893E; Applied Biosystems). Cytokeratin 8 was used as a normalizer to control for possible stromal cell contamination. Cytokeratin expression was normalized to 18S to control for RNA input. Thereafter, AKR1C mRNA was expressed relative to cytokeratin 8/18S. There were no significant differences in cytokeratin 8 gene expression between the two groups, and data were virtually identical whether normalized to cytokeratin or 18S. All primer sets were tested to ensure that efficiency of amplification over a wide range of template concentrations was equivalent to that of 18S. PCR reactions were performed in the ABI Prism 7000 sequence detection system (Applied Biosystems). The RT product from 50 ng RNA was used as template, and 5 ng RNA was used for 18S amplification. Reaction volumes were 30 μl, containing 1× Master Mix (no. 4304437; Applied Biosystems). Primer concentrations were 900 nm. Cycling conditions were: 2 min at 50 C, followed by 10 min at 95 C; then 40 cycles of 15 sec at 95 C and 1 min at 60 C. Levels of mRNA were determined using the ΔΔCT method (Applied Biosystems) and expressed relative to an external calibrator present on each plate. For comparison of AKR1C1-AKR1C4 mRNA levels, expression of AKR1C1 was used as the calibrator, and levels of other AKR1C enzymes were expressed relative to that of AKR1C1 in the same tissues.

Table 2.

Expression of 20α-HSDs in the human cervix during pregnancy

Gene Term, NIL (n = 9) Term, IL (n = 10)
AKR1C1 1.0 ± 0.19 0.68 ± 0.41
AKR1C2 0.15 ± 0.07 0.02 ± 0.01a
AKR1C3 0.33 ± 0.15 0.12 ± 0.04a
AKR1C4
Open in a new tab

Real-time PCR with TaqMan probes specific for each AKR1C family member was conducted in endocervical scrapings from pregnant women before labor with an unripe cervix (Term, NIL) and after labor with a ripe cervix (Term, IL). Gene expression was normalized for RNA input using 18S and for epithelial cells using cytokeratin 8. Data represent mean ± sem and are expressed relative to that of AKR1C1 in pregnant women before labor. 

a

P ≤ 0.05. 

Table 3.

Sequences of primers and probes used for quantification of AKR1C1–4 gene expression

Gene GenBank accession no. Upstream primer Probe Downstream primer
AKR1C1 NM001353 258gcctgcagaggttcctaaaagtaa282 294caccaaattggcaattgaagctggct319 368ccaacctgctcctcattattgtataa343
AKR1C2 BC063574 388ggccgtcaaattggcaatag407 409agccgggttccaccatattgattctgc435 462acctgctcctcattattgtaaacatg437
AKR1C3 AF149416 851cgtgcaggtttttgagttcca871 903catctatggctttcatgtcctctgcagtcaa873 954ggtggctagcaaaacttatcactgtt930
AKR1C4 D26125 880ttctagatggtctaaacagaaattatcga908 913ttgtcatggattttcttatggaccatcctga943 981aacaccctctatgctaatattcatctga954
Open in a new tab

Northern analysis

Total RNA was size fractionated on a 1% agarose containing gel and transferred electrophoretically to a nylon membrane. The membrane was hybridized with radiolabeled full-length human cDNA probes of 17βHSD type 2 (12), steroid 5α-reductase type 1 (15), or steroid 5α-reductase type 2 (16) and subjected to autoradiography with intensifying screens for 1–3 d at −80 C.

Statistical analysis

Differences in gene expression between tissue specimens not-in-labor and in-labor were determined using the Student’s t test. Differences in gene expression among various regions within the uterus were determined using one-way ANOVA, with a significance level of P < 0.05. The Student-Newman-Keuls multiple comparisons procedure was performed when the P value was less than 0.05.

Results

Previous studies indicate that 5α-reductase type 1 is responsible for the irreversible conversion of progesterone to the inactive C21-steroid 5α-dihydroprogesterone in the mouse cervix, and that this activity is crucial for cervical ripening in mice (2). To test whether 5α-reductase enzyme is present in the human cervix, conversion of progesterone to 5α-dihydroprogesterone (and products thereof) in intact uterine and cervical tissues from pregnant women was determined (Fig. 1). Whereas progesterone was converted to 5α-reduced products in cervical stroma, metabolism of progesterone to 5α-reduced steroids was slow and inefficient in myometrial tissues or cervical epithelium. 5α-Reductase activity was similar in cervical stroma from nonpregnant and pregnant women before and after labor (Fig. 1).

Figure 1.

Figure 1

Open in a new tab

Conversion of progesterone to 5α-reduced products in the human uterus during pregnancy. Conversion of progesterone to 5α-reduced products was determined in cervical (A) and myometrial (B) tissues from nonpregnant women (NP) (n = 5), pregnant women before labor (Preg NIL) (n = 7), and pregnant women after the onset of labor (Preg IL) (n = 5). Cervical epithelium was obtained by scraping the endocervical canal, avoiding the underlying stroma or squamous epithelium. Cervical stroma was obtained remote from the endocervical canal. Myometrial tissues were obtained from the lower uterine segment (Lower Ut Seg) or fundus. Cervical and myometrial tissues were incubated with radiolabeled progesterone (30 nm) for 12 and 24 h, respectively. Each value represents the mean ± se of five to seven individual samples performed in duplicate. *, P ≤ 0.05 compared with cervical epithelium or myometrium at same gestational condition.

To confirm these observations and determine isozyme-specific expression of 5α-reductase in the human cervix, RNA was isolated from cervical stroma of nonpregnant and pregnant women before and after labor, and Northern blot analysis was conducted with radiolabeled probes specific for 5α-reductases types 1 and 2 (Fig. 2). A 2.5-kb mRNA species that hybridized to the 5α-reductase type 1 probe was present in all tissues with no significant differences among the tissues. In contrast, 5α-reductase type 2 was not expressed in the human cervix. Human prostate was used as a positive control for both mRNAs.

Figure 2.

Figure 2

Open in a new tab

Expression of steroid 5α-reductase types 1 and 2 mRNA in the human cervix. Total RNA (10 μg) was isolated from cervical stromal tissues obtained from a pool of nonpregnant premenopausal women (NP), pregnant women before labor (NIL), pregnant women in labor (IL), and human adult prostate (Pr). After transfer, the membranes were hybridized to cDNA probes specific for 5α-reductase type 1 (5αR1) or type 2 (5αR2), or glyceraldehyde 3 phosphate dehydrogenase (G3PDH). The molecular size in kilobase is noted. The experiment was repeated with three different samples with identical results.

During the course of these experiments, it was noted that progesterone was converted to the inactive C21-steroid 20α-hydroxyprogesterone in some tissues. Thus, rates of conversion of progesterone to 20α-hydroxyprogesterone in the presence of the 5α-reductase inhibitor 4-MA were determined in the human cervix and myometrium (Fig. 3). Conversion of progesterone to 20α-hydroxyprogesterone was completely absent in cervical epithelium from nonpregnant and pregnant women before labor. In contrast, conversion of progesterone to 20α-hydroxyprogesterone was increased dramatically in cervical epithelium from women in labor to levels one third that of decidua, a tissue known to be enriched in this enzyme activity (Fig. 3). Progesterone was also converted to 20α-hydroxyprogesterone in cervical stromal tissues but did not vary with pregnancy or parturition. However, in myometrium, progesterone metabolism to 20α-hydroxyprogesterone was low compared with cervical stroma, and did not vary before and after labor.

Figure 3.

Figure 3

Open in a new tab

Conversion of progesterone to 20α-hydroxyprogesterone in the human uterus during pregnancy. Conversion of progesterone to 20α-hydroxyprogesterone was determined in cervical (Panel A) and myometrial (Panel B) tissues from nonpregnant women (NP) (n = 5), pregnant women before labor (Preg NIL) (n = 7), and pregnant women after the onset of labor (Preg IL) (n = 5). Decidual tissues from three women before labor were used for comparison (dec). Cervical epithelium was obtained by scraping the endocervical canal, avoiding the underlying stroma or squamous epithelium. Cervical stroma was obtained remote from the endocervical canal. Myometrial tissues were obtained from the lower uterine segment (Lower Ut Seg) or fundus. Steroid 5α-reductase activity was inhibited by 4-MA (10 μm). Cervical and myometrial tissues were incubated with radiolabeled progesterone (30 nm) for 12 and 24 h, respectively. Each value represents the mean ± se of three to seven individual samples performed in duplicate. *, P ≤ 0.01 compared with cervical epithelium, stroma, or myometrium at any gestational condition.

Because the reductive 20αHSD activity is known to be catalyzed by four members of the human AKR family, AKR1C1-AKR1C4 (18,19), we examined the expression of AKR1C1–4 mRNAs by quantitative real-time PCR in the human cervix. Primer and TaqMan probe sequences designed to quantify expression of these genes are shown in Table 3. For these experiments, RNA from subjects listed in Table 1 was analyzed together with two additional women before labor (repeat cesarean section, placenta accreta) and five women in labor (two repeat cesarean section/placenta accreta, and three cesarean sections for failure to progress and uterine atony). AKR1C1 was the most abundant 20αHSD in the human endocervix, and expression of this isoform was similar in endocervical samples from women before and after labor (Table 2). Both AKR1C2 and 1C3 were expressed in the cervix, and, contrary to our hypothesis, mRNA levels of these two isoforms were decreased in the endocervix during labor. Although expressed in the fetal liver, AKR1C4 was not expressed in the cervix (Table 2). Human fetal liver was used as a positive control for all four isoforms, and a cDNA of AKR1C1 cloned from the human uterus was used as a positive control for AKR1C1 and negative control for the 1C2, 1C3, and 1C4 mRNAs.

During initial experiments to define the potential role of 5α-reductase type 1 in the human cervix during labor, we used radiolabeled testosterone as the substrate. Surprisingly, we found robust metabolism of testosterone to androstenedione in cervical epithelium from women before labor; no 5α-dihydrotestosterone formation could be detected. This activity has been described to be catalyzed by 17βHSD type 2 in vitro, which also catalyzes the conversion of estradiol to the inactive C18-steroid estrone, but conversely, converts 20α-hydroxyprogesterone to bioactive progesterone (12). These substrate specificities, together with the complete absence of 20α-hydroxyprogesterone formation in cervical epithelium from women before labor, led us to hypothesize that 17βHSD type 2 activity may be expressed in the human cervix and regulated during parturition. Thus, tissues were incubated with radiolabeled estradiol, and conversion of estradiol to estrone was assessed (Fig. 4A). Cervical epithelium, but not cervical stroma, possessed high oxidative 17βHSD activity by efficiently converting estradiol to estrone. In parallel experiments, conversion of estrone to estradiol could not be demonstrated (data not shown). Interestingly, although activity in endocervical epithelium was increased during pregnancy, enzyme activity was decreased significantly in tissues from women in labor compared with those before labor (Fig. 4A). These results strongly suggested that 17βHSD type 2 may be expressed in the cervical epithelium of pregnant women, and that the HSD17B2 gene is regulated during pregnancy and parturition. Northern blot analysis was performed, and revealed that 17βHSD type 2 mRNA is highly expressed in the human endocervix and secretory endometrium, but not in cervical stroma or myometrium (Fig. 4B). Consistent with enzyme activity measurements, 17βHSD type 2 mRNA levels were increased in the endocervix relative to cervical stroma. Furthermore, 17βHSD type 2 mRNA levels were decreased in endocervical tissues from women in labor (Fig. 4C).

Figure 4.

Figure 4

Open in a new tab

Differential regulation of 17β-HSD type 2 enzyme activity in endocervical epithelial and cervical stromal tissues during pregnancy and parturition. A, Conversion of estradiol (E2) to estrone (E1) was quantified in endocervical scrapings (Epithelium) and cervical stroma (Stroma) from nonpregnant women (NonP) (n = 5), pregnant women before labor (Preg NIL) (n = 7), and pregnant women in labor (Preg IL) (n = 5). Endocervical epithelial and stromal tissues were incubated with radiolabeled estradiol (30 nm) for 6 and 24 h, respectively. Each value represents the mean ± se of five to seven individual samples performed in duplicate. *, P ≤ 0.05 compared with pregnant women before labor, ANOVA. B, Expression of 17βHSD type 2 mRNA in human cervical tissues. Total RNA (10 μg/lane) from the endocervix (ECx), secretory endometrium (Em), cervical stroma (CStr), myometrium (Myo), and placenta (Pl) was subjected to blot hybridization on a nylon membrane as described in Subjects and Methods. Arrow indicates 17βHSD type 2 mRNA. Position to which mRNAs of known size migrated are shown on the right. Ethidium bromide-stained 28S rRNA subunits are presented for reference. C, Northern blot analysis of 17βHSD type 2 (17βHSD2) mRNA in human cervical tissues from pregnant women. Total RNA (10 μg/lane) isolated from endocervical and stromal tissues from pregnant women at term before labor (NIL) (lanes 1–3 and 7–9) and in labor (IL) (lanes 4–6 and 10–12) was hybridized with a radiolabeled cDNA probe for 17βHSD type 2 or glyceraldehyde 3 phosphate dehydrogenase (G3PDH). Endocervical tissue from a nonpregnant woman (NP) is shown for comparison.

We then determined the cell type-specific expression of 17βHSD type 2 in the human cervix by immunohistochemistry using the 17βHSD type 2 specific mAb-C2–12 antibody (Fig. 5). Consistently, strong immunoreactivity for 17βHSD type 2 was localized to the epithelial cells of the cervix (Fig 5, A–C). The histomorphology and immunoreactivity of 17βHSD type 2 in the nonpregnant cervix (Fig. 5A) compared with the cervix of pregnant women before labor (Fig. 5B) highlight the remarkable hyperplasia of endocervical epithelial cells during pregnancy. As a positive control, we incubated human term placenta with mAb-C2–12; strong immunostaining of 17βHSD type 2 was detected in the endothelial cells as expected (14) (Fig. 5E).

Figure 5.

Figure 5

Open in a new tab

Immunohistochemical localization of 17βHSD type 2 in the human cervixand placenta. Positive (brown) staining for 17βHSD type 2 is seen in the epithelial cells of the nonpregnant cervix (A) and cervix before labor (B and C). No staining is observed when the cervix is incubated without primary antibody (D). Positive 17βHSD type 2 staining is seen in the endothelial cells of the placenta (E). Magnification of A and B, 20×; C–E, 100×. Neg Ctl, Negative control.

Discussion

In this investigation we used intact tissues from the human cervix to demonstrate that estrogen and progesterone metabolism in the cervix of women is regulated during pregnancy and parturition, and is significantly different from that of mice. Intact tissues were used so that enzymes would be operative in their cellular and matrix environment using endogenous cofactors. In mice, expression of 5α-reductase type 1 in cervical epithelial cells is increased during cervical ripening and parturition, and significant growth of epithelial cells occurs during pregnancy (20) (Fig. 6). Results obtained from the current study indicate that, in women, although 5α-reductase type 1 is expressed in the cervix, the enzyme is stromal in origin, and is not regulated during pregnancy and parturition. Our working model of the pathways involved in the human cervix during pregnancy is presented in Fig. 6.

Figure 6.

Figure 6

Open in a new tab

Models of steroid hormone metabolism in mouse and human cervix during pregnancy. A, In the mouse, steroid 5α-reductase type 1 (5αR-1) is induced in cervical epithelial cells to inactivate progesterone locally. B, In the human cervix during pregnancy, circulating progesterone (P) and estradiol (E2) enter the cervical stroma, and are metabolized by 5α-reductase type 1 and 20α-HSDs (AKR1C1-C3) to 5α-dihydroprogesterone (5αDHP) and 20α-hydroxyprogesterone (20αP). Estradiol and 20α-hydroxyprogesterone are converted to estrone (E1) and progesterone, respectively, by high levels of 17βHSD type 2 (17βHSD2) in endocervical epithelial cells, resulting in inactivation of estradiol and the maintenance of progesterone levels in the surrounding stromal cells. Robust expression of 17βHSD type 2 in epithelial cells also counteracts 20α-reduction of progesterone by epithelial AKR1C enzymes. C, During cervical ripening and parturition, 17βHSD type 2 is decreased in endocervical epithelial cells. Conversion of estradiol to estrone and 20α-hydroxyprogesterone to progesterone is thereby compromised, resulting in increased estradiol and 20α-hydroxyprogesterone levels and inactivation of progesterone responses in the stroma.

17βHSD type 2 mRNA and enzyme activity are highly expressed in endocervical epithelial cells, but not cervical stromal cells. The size and number of endocervical glands are increased substantially during pregnancy. Because 17βHSD type 2 does not use progesterone as a substrate, but readily converts 20α-hydroxyprogesterone to progesterone and estradiol to estrone (12), the enzyme may play an important role in: 1) preservation of progesterone-dependent genes, and 2) inactivation of estrogen-dependent processes in the human cervix during pregnancy. Inactivation of estradiol to estrone may have more physiological significance for most of pregnancy, preventing the high levels of circulating estradiol from inducing genes that remodel the cervix. In contrast, in cervical stromal cells, progesterone is metabolized to 5α-dihydroprogesterone by steroid 5α-reductase, 20α-hydroxyprogesterone predominantly by AKR1C1, and to 5α-pregnan-3α,20α-diol by the combined actions of these two enzymes. Although we failed to find substantial up-regulation of AKR1C family members in the human cervix during parturition, we suggest that the loss of 17βHSD type 2 activity in endocervical cells at term, together with basal expression of progesterone metabolizing enzymes, may create a microenvironment favorable for increased expression of estrogen-dependent genes such as oxytocin receptors and cyclooxygenase-2 (10). Nevertheless, because IL-8 is suppressed by progesterone (21), increased local metabolism of progesterone would augment induction of IL-8, neutrophil chemotaxis, and subsequent remodeling of the cervical extracellular matrix by activated immune cells.

It is interesting to note that the length of the endocervical canal shortens during cervical effacement, and that cervical ripening and dilation begin near the internal cervical os (22). Thus, ripening of the cervix occurs initially in the area of the endocervical epithelial cells, which are enriched in 17βHSD type 2. The loss of the mucus plug, cervical effacement, and dilation may result in the loss of the number of cells expressing 17βHSD type 2. We speculate that the loss of epithelium, together with decreased expression of 17βHSD type 2 in endocervical epithelial cells, would result in the loss of progesterone maintenance and simultaneous increases in local estradiol concentrations. This hormonal milieu is favorable for increased expression of oxytocin receptors and cyclooxygenase-2 in the cervix. Furthermore, vascularization of the cervix is increased dramatically during pregnancy primarily through branches of the uterine artery that enter the cervix laterally, forming a rich capillary bed toward the endocervix. Venous effluent transports metabolites from the inner canal through the cervical stroma and into uterine veins. Decreases in epithelial 17βHSD type 2 would compromise conversion of stromal-derived 20α-hydroxyprogesterone to progesterone and would result in the increased delivery of estradiol to the surrounding stroma. A similar paracrine endocrine system has been described in fetal membranes during human pregnancy and parturition (23).

Endocervical epithelial cells are important in providing protection from invading microorganisms by secreting defensins and copious amounts of thick mucus (24). Furthermore, these cells mediate inflammatory responses and adaptive immune responses that limit bacterial proliferation through several mechanisms, including expression of Toll-like receptors that bind bacterial toxins. The cells are also enriched in enzymes involved in prostaglandin biosynthesis (10), contain high levels of cytokines and chemokines such as IL-8 (25), and secrete a number of potent protease inhibitors (26,27,28). Yoshimatsu et al. (29) used cervical sonography to detect the area of cervical glands throughout pregnancy. The area of cervical glands correlated with increased cervical length and palpable cervical rigidity. Incremental decreases in the cervical gland area after 31-wk gestation correlated with progressive increases in cervical softening and the loss of cervical rigidity. These findings, together with those of the current study, suggest that human cervical epithelial cells provide not only a protective barrier but also play an endocrine role in maintaining cervical competency during pregnancy.

Recent clinical trials indicate that progesterone supplementation has a profound impact on the incidence of preterm birth in women with a short cervix in the second trimester (30,31). The results of the current investigation provide a plausible mechanism by which progesterone may inhibit preterm birth in women with preterm cervical ripening. We suggest that progesterone supplementation may increase expression of 17βHSD type 2 in endocervical epithelial cells as it does in glandular endometrial cells (32), thereby resulting in inactivation of estradiol and maintenance of local progesterone concentrations in the cervix. Experiments are underway to test this hypothesis.

In summary, cervical ripening during parturition is associated with localized regulation of estrogen and progesterone metabolism through a complex relationship between cervical epithelium and stroma. Regulation of 17βHSD type 2 in endocervical cells at term may result in increased estradiol and attenuation of progesterone action, thereby creating a microenvironment favorable for cervical ripening. Further studies are indicated to understand the precise regulatory signaling pathways in endocervical epithelial cells and their interactions with the underlying stromal fibroblasts.

Acknowledgments

We thank the physicians and staff of Parkland Memorial Hospital, and Ms. Sheila Brandon and Ms. Valencia Hoffman for their valuable assistance in tissue procurement. We also thank Rodney Miller, M.D. (ProPath Laboratory, Inc., Dallas, TX), for his advice and technical expertise, as well as Mr. Jesse Smith and Mr. Patrick Keller for their expert technical assistance.

Footnotes

This work was supported by National Institutes of Health Grants HD11149 (to R.A.W.) and DK52167 (to S.A.).

Disclosure Statement: D.M., N.P.Y., and R.A.W. have nothing to declare. S.A. holds United States patent no. 5442262 for steroid 5α-reductase nucleic acid segments, recombinant vectors, and host cells.

First Published Online March 25, 2008

Abbreviations: AKR, Aldo-keto reductase; HSD, hydroxysteroid dehydrogenase; mAb, monoclonal antibody; 4-MA, 17β-(N,N,-diethyl)carbamoyl-4-methyl-4-aza-5α-androstan-3-one.

References

  1. Word RA, Li XH, Hnat M, Carrick K 2007 Dynamics of cervical remodeling during pregnancy and parturition: mechanisms and current concepts. Semin Reprod Med 25:69–79 [DOI] [PubMed] [Google Scholar]
  2. Mahendroo MS, Porter A, Russell DW, Word RA 1999 The parturition defect in steroid 5α-reductase type 1 knockout mice is due to impaired cervical ripening. Mol Endocrinol 13:981–992 [DOI] [PubMed] [Google Scholar]
  3. Mesiano S, Chan EC, Fitter JT, Kwek K, Yeo G, Smith R 2002 Progesterone withdrawal and estrogen activation in human parturition are coordinated by progesterone receptor A expression in the myometrium. J Clin Endocrinol Metab 87:2924–2930 [DOI] [PubMed] [Google Scholar]
  4. Madsen G, Macintyre DA, Mesiano S, Smith R 2007 Progesterone receptor or cytoskeletal protein? Reprod Sci 14:217–222 [DOI] [PubMed] [Google Scholar]
  5. Mesiano S, Welsh TN 2007 Steroid hormone control of myometrial contractility and parturition. Semin Cell Dev Biol 18:321–331 [DOI] [PubMed] [Google Scholar]
  6. Madsen G, Zakar T, Ku CY, Sanborn BM, Smith R, Mesiano S 2004 Prostaglandins differentially modulate progesterone receptor-A and -B expression in human myometrial cells: evidence for prostaglandin-induced functional progesterone withdrawal. J Clin Endocrinol Metab 89:1010–1013 [DOI] [PubMed] [Google Scholar]
  7. Chwalisz K, Hegele-Hartung C, Schulz R, Shi S-Q, Louton PT, Elger W 1991 Progesterone control of cervical ripening–experimental studies with the progesterone antagonists onapristone, lilopristone, and mifepristone. In: Leppert PC, Woessner Jr JF, eds. The extracellular matrix of the uterus, cervix, and fetal membranes, synthesis, degradation, and hormonal regulation. Ithaca, NY: Perinatology Press; 119–131 [Google Scholar]
  8. Rajabi M, Solomon S, Poole AR 1991 Hormonal regulation of interstitial collagenase in the uterine cervix of the pregnant guinea pig. Endocrinology 128:863–871 [DOI] [PubMed] [Google Scholar]
  9. Rajabi MR, Dodge GR, Solomon S, Poole AR 1991 Immunochemical and immunohistochemical evidence of estrogen-mediated collagenolysis as a mechanism of cervical dilatation in the guinea pig at parturition. Endocrinology 128:371–378 [DOI] [PubMed] [Google Scholar]
  10. Havelock JC, Keller P, Muleba N, Mayhew BA, Casey BM, Rainey WE, Word RA 2005 Human myometrial gene expression before and during parturition. Biol Reprod 72:707–719 [DOI] [PubMed] [Google Scholar]
  11. Hertelendy F, Zakar T 2004 Prostaglandins and the myometrium and cervix. Prostaglandins Leukot Essent Fatty Acids 70:207–222 [DOI] [PubMed] [Google Scholar]
  12. Wu L, Einstein M, Geissler WM, Chan HK, Elliston KO, Andersson S 1993 Expression cloning and characterization of human 17 beta-hydroxysteroid dehydrogenase type 2, a microsomal enzyme possessing 20 alpha-hydroxysteroid dehydrogenase activity. J Biol Chem 268:12964–12969 [PubMed] [Google Scholar]
  13. McKeever BM, Hawkins BK, Geissler WM, Wu L, Sheridan RP, Mosley RT, Andersson S 2002 Amino acid substitution of arginine 80 in 17beta-hydroxysteroid dehydrogenase type 3 and its effect on NADPH cofactor binding and oxidation/reduction kinetics. Biochim Biophys Acta 1601:29–37 [DOI] [PubMed] [Google Scholar]
  14. Moghrabi N, Head JR, Andersson S 1997 Cell type-specific expression of 17 β-hydroxysteroid dehydrogenase type 2 in human placenta and fetal liver. J Clin Endocrinol Metab 82:3872–3878 [DOI] [PubMed] [Google Scholar]
  15. Andersson S, Russell DW 1990 Structural and biochemical properties of cloned and expressed human and rat steroid 5 alpha-reductases. Proc Natl Acad Sci USA 87:3640–3644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Andersson S, Berman DM, Jenkins EP, Russell DW 1991 Deletion of steroid 5 alpha-reductase 2 gene in male pseudohermaphroditism. Nature 354:159–161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, Moore M, Palackal N, Ratnam K 2000 Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Biochem J 351(Pt 1):67–77 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Sharma KK, Lindqvist A, Zhou XJ, Auchus RJ, Penning TM, Andersson S 2006 Deoxycorticosterone inactivation by AKR1C3 in human mineralocorticoid target tissues. Mol Cell Endocrinol 248:79–86 [DOI] [PubMed] [Google Scholar]
  19. Minjarez D, Konda V, Word RA 2001 Regulation of uterine 5 alpha-reductase type 1 in mice. Biol Reprod 65:1378–1382 [DOI] [PubMed] [Google Scholar]
  20. Ito A, Imada K, Sato T, Kubo T, Matsushima K, Mori Y 1994 Suppression of interleukin 8 production by progesterone in rabbit uterine cervix. Biochem J 301(Pt 1):183–186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Cook CM, Ellwood DA 2000 The cervix as a predictor of preterm delivery in ‘at-risk’ women. Ultrasound Obstet Gynecol 15:109–113 [DOI] [PubMed] [Google Scholar]
  22. Mitchell BF, Wong S 1993 Changes in 17 beta,20 alpha-hydroxysteroid dehydrogenase activity supporting an increase in the estrogen/progesterone ratio of human fetal membranes at parturition. Am J Obstet Gynecol 168:1377–1385 [DOI] [PubMed] [Google Scholar]
  23. Baumann P, Romero R 1995 [Intra-amniotic infection, cytokines and premature labor]. Wien Klin Wochenschr 107:598–607 (German) [PubMed] [Google Scholar]
  24. Barclay CG, Brennand JE, Kelly RW, Calder AA 1993 Interleukin-8 production by the human cervix. Am J Obstet Gynecol 169:625–632 [DOI] [PubMed] [Google Scholar]
  25. Becher N, Hein M, Danielsen CC, Uldbjerg N 2004 Matrix metalloproteinases and their inhibitors in the cervical mucus plug at term of pregnancy. Am J Obstet Gynecol 191:1232–1239 [DOI] [PubMed] [Google Scholar]
  26. Moriyama A, Shimoya K, Ogata I, Kimura T, Nakamura T, Wada H, Ohashi K, Azuma C, Saji F, Murata Y 1999 Secretory leukocyte protease inhibitor (SLPI) concentrations in cervical mucus of women with normal menstrual cycle. Mol Hum Reprod 5:656–661 [DOI] [PubMed] [Google Scholar]
  27. Pfundt R, van Ruissen F, van Vlijmen-Willems IM, Alkemade HA, Zeeuwen PL, Jap PH, Dijkman H, Fransen J, Croes H, van Erp PE, Schalkwijk J 1996 Constitutive and inducible expression of SKALP/elafin provides anti-elastase defense in human epithelia. J Clin Invest 98:1389–1399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Yoshimatsu K, Sekiya T, Ishihara K, Fukami T, Otabe T, Araki T 2002 Detection of the cervical gland area in threatened preterm labor using transvaginal sonography in the assessment of cervical maturation and the outcome of pregnancy. Gynecol Obstet Invest 53:149–156 [DOI] [PubMed] [Google Scholar]
  29. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH 2007 Progesterone and the risk of preterm birth among women with a short cervix. N Engl J Med 357:462–469 [DOI] [PubMed] [Google Scholar]
  30. DeFranco EA, O'Brien JM, Adair CD, Lewis DF, Hall DR, Fusey S, Soma-Pillay P, Porter K, How H, Schakis R, Eller D, Trivedi Y, Vanburen G, Khandelwal M, Trofatter K, Vidyadhari D, Vijayaraghavan J, Weeks J, Dattel B, Newton E, Chazotte C, Valenzuela G, Calda P, Bsharat M, Creasy GW 2007 Vaginal progesterone is associated with a decrease in risk for early preterm birth and improved neonatal outcome in women with a short cervix: a secondary analysis from a randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol 30:697–705 [DOI] [PubMed] [Google Scholar]
  31. Tseng L, Gurpide E 1979 Stimulation of various 17 β- and 20 α-hydroxysteroid dehydrogenase activities by progestins in human endometrium. Endocrinology 104:1745–1748 [DOI] [PubMed] [Google Scholar]
  32. Casey ML, MacDonald PC, Anderson S 1994 17β-Hydroxysteroid dehydrogenase type 2: chromosomal assignment and progestin regulation of gene expression in human endometrium. J Clin Invest 94:2135–2141 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES