Human type 2 17 beta-hydroxysteroid dehydrogenase mRNA and protein distribution in placental villi at mid and term pregnancy

Abstract

Background: During human pregnancy, the placental villi produces high amounts of estradiol. This steroid is secreted by the syncytium, which is directly in contact with maternal blood. Estradiol has to cross placental foetal vessels to reach foetal circulation. The enzyme 17beta-hydroxysteroid dehydrogenase type 2 (17beta-HSD2) was detected in placental endothelial cells of foetal vessels inside the villi. This enzyme catalyzes the conversion of estradiol to estrone, and of testosterone to androstenedione. It was proposed that estradiol level into foetal circulation could be regulated by 17beta-HSD2.

Methods: We obtained placentas from 10 to 26 6/7 weeks of pregnancy from women undergoing voluntary termination of pregnancy, term placentas were collected after normal spontaneous vaginal deliveries. We quantified 17beta-HSD2 mRNA levels in mid-gestation and term human placenta by RT-QPCR. We produced a new anti-17beta-HSD2 antibody to study its spatio-temporal expression by immunohistochemistry. We also compared steroid levels (testosterone, estrone and estradiol) and 17beta-HSD2 mRNA and protein levels between term placenta and endometrium.

Results: High 17beta-HSD2 mRNA and protein levels were found in both mid-gestation and term placentas. However, we showed that 17beta-HSD2 mRNA levels increase by 2.27 fold between mid-gestation and term. This period coincides with a transitional phase in the development of the villous vasculature. In mid-gestation placenta, high levels of 17beta-HSD2 were found in mesenchymal villi and immature intermediate villi, more precisely in endothelial cells of the stromal channel. At term, high levels of 17beta-HSD2 were found in the numerous sinusoidal capillaries of terminal villi. 17beta-HSD2 mRNA and protein levels in term placentas were respectively 25.4 fold and 30 to 60 fold higher than in the endometrium. Steroid levels were also significantly higher in term placenta than in the endometrium.

Conclusion: The spatial and temporal expression of 17beta-HSD2 in the placenta during pregnancy and the comparison of 17beta-HSD2 expression and steroid levels between placental villi and endometrium are compatible with a role in the modulation of active and inactive forms of estrogens. Our observations strongly support the hypothesis that 17beta-HSD2 acts as a barrier decreasing estradiol secretion rates in the foetal circulation.

Figure 1

Northern blot analysis of five samples of villous tissue collected at different positions relative to the insertion point of the umbilical cord using 20 μg of total RNA loaded in each lane. Blots were probed with 17βHSD2 and actin cDNAs. Two placentas obtained at term were studied. Autoradiograms for Northern blots were densitometrically scanned and no significant variation between samples was observed.

Figure 2

Expression of 17βHSD2 mRNA in human mid-pregnancy and term placentas. Levels of 17βHSD2 mRNA were determined by real-time quantitative PCR in 35 placental tissues collected at the indicated gestation time. For each sample, 17βHSD2 expression level was normalized by a factor generated from the expression levels, in the same sample, of the three most stable housekeeping genes studied (YWHAZ-HPRT1, and GAPDH), using the geNorm software (for details, see Methods). Means ± SD are presented.

Figure 3

Characterization of the anti-17βHSD2 antibody. Western blot analysis: Cytosol and microsomes extracted from human placental villi were resolved on a 12% SDS-PAGE and transferred on nitrocellulose membrane. Immunodetection was performed using purified rabbit anti-17βHSD2 antibody (Panel A), preimmune rabbit serum (Panel B), or antibody pre-adsorbed with an excess (10^-6M) of 17βHSD2 synthetic polypeptide (Panel C). Specificity of the anti-17βHSD2 antibody was also studied by two-dimensional gel electrophoresis (Panels D and E). Microsomal fraction of placental villi was separated by 2-dimensional isoelectric focusing-SDS-PAGE. Panel D, silver staining. Panel E, immunodetection with the purified anti-17βHSD2 antibody. Position of the circle on panel D corresponds to the location of the circle on panel E. The upper arrow corresponds to a pI value of approximately 8 to 9 and arrow on the right side corresponds to the expected molecular weight.

Figure 4

Localization of 17βHSD2 and KDR proteins in mid-gestation placentas. Tissue sections of human placentas obtained at 19 weeks (Panels A and B), 18 weeks (Panel C), 10 weeks (Panel D) or 20 weeks (Panels E-H) of pregnancy were immunostained with anti-17βHSD2 (Panels A, C-F), with pre-adsorbed antibody, as negative control (Panel B), or with anti-KDR antibody used as endothelial cell marker (Panels G and H). 17βHSD2 protein expression is observed in immature intermediate villi (IIV) (Panels A, C, E, F), in capillary endothelial (CE) cells and in cells forming the stromal channel (SC) (Panel C). Positive cells (most probably endothelial cells (EC)) were found in mesenchymal villi (MV). No signal was observed in trophoblastic sprout (TS), in syncytial layer (SL), and in decidua (DC) (Panel A). In stem villi (SV) (Panel A), EC of large blood vessels and those of paravascular capillary (PC) network are positive. Endothelial cells forming PC network in SV (Panel G) and those composing the wall of the stromal channel were immunostained by the anti-KDR (Panel H). Intervillous space (IS), where maternal blood circulates, is indicated.

Figure 5

Localization of 17βHSD2 protein in human term placentas. Tissue sections of term placentas were immunostained with anti-17βHSD2 (Panels A, C-E) or with pre-adsorbed antibody, as negative control (Panel B). Strong positive signals were observed in terminal villi (TV) (Panels A and C), more precisely in sinusoidal capillaries (Si). 17βHSD2 expression is observed in the paravascular capillary net (PC) as well as in endothelial cells (EC) forming the blood vessel walls of stem villi (SV) (Panels D and E). Some blood vessels are EC positive (+) and others EC negative (-).

Figure 6

Localization of 17βHSD2 mRNA in human mid-pregnancy and term placentas. In situ hybridization was performed with antisense (Panels A, C and D) and sense (negative control, Panel B) RNA probes on tissue sections of placentas obtained at 18 weeks of gestation (Panels A, B, and C), or at term (Panel D). Emulsion granules (positive signals) appear as white dots on darkfields. SV, stem villi; IIV, immature intermediate villi: SL, syncytial layer; PC, paravascular capillary net; EC+ and EC-, endothelial cells with, respectively, positive and negative signal with the 17βHSD2 antisense probe.

Figure 7

Quantitative evaluation of 17βHSD2 protein expression in immature intermediate villi (IIV) and stem villi (SV) of human mid-gestation placentas and from terminal villi (TV) and stem villi (SV) of human term placentas. Twenty-three IIV and 17 SV structures were analysed from 13 mid-gestation placentas, while 10 TV and 9 SV structures were analysed from 5 term placentas. Quantification of immunohistochemical stainings was performed using a Zeiss Axioskop 2 plus miscroscope linked to a digital camera (Spot Insight; Carsen Medical Scientific, Markham, ON, Canada) using Image Pro software analysis (Carsen Medical Scientific). *: p < 0.001.