221
Expression of corticotrophin-releasing hormone in the mouse uterus: participation in embryo implantation I Athanassakis1, V Farmakiotis1, I Aifantis1,2, A Gravanis3 and S Vassiliadis1 1
Department of Biology, University of Crete, Heraklion, Crete, Greece
2
Hoˆpital Necker, Unite INSERM 373, Paris, France
3
Department of Pharmacology, Medical School, University of Crete, Heraklion, Crete, Greece
(Requests for offprints should be addressed to I Athanassakis, Department of Biology, University of Crete, PO Box 2208, 714–09 Heraklion, Crete, Greece; E-mail:
[email protected])
Abstract The detection of corticotropin-releasing hormone (CRH) in the pregnant and non-pregnant uterus has driven research to determine the role of this 41 amino acid neuropeptide in the female reproductive system. As concentrations of CRH mRNA and its peptide product are greater in the implantation sites of the early pregnant uterus compared with the regions between implantation sites, CRH has been hypothesised to participate in blastocyst implantation. Using the mouse system as an experimental model, we studied the distribution of CRH in the uterus during the oestrus cycle and early gestational period, and now provide evidence for its involvement in embryo implantation using cell culture techniques. The percentage of CRH-positive uterine cells and the amount of CRH released during anoestrus, pro-oestrus and oestrus were determined by immunofluorescence and ELISA experiments respectively. The highest number of intra-
Introduction In addition to the hypothalamic origin of corticotropinreleasing hormone (CRH) (Antoni 1986), this hormone has been identified in the uterus of pregnant and nonpregnant women (Makrigiannakis et al. 1995a, Clifton et al. 1998). During pregnancy CRH is also detected in trophoblasts, amnion, chorion and decidua (Shibasaki et al. 1982, Grino et al. 1987, Sasaki et al. 1987, Petraglia et al. 1992). The functional role of this peptide in the uterus has been correlated with the presence of specific receptors; however, these show a different pattern of expression during the pregnant and non-pregnant state (Grammatopoulos et al. 1998). Thus, during pregnancy, four subtypes of the CRH receptor – 1, 1, 2 and the variant C – are expressed in the myometrium, whereas in the non-pregnant uterus only the subtypes 1 and 1 are detected in the myometrium.
cellularly CRH-positive cells was obtained during pro-oestrus, whereas the highest CRH concentration in uterine cell culture supernatants was detected during anoestrus. At early stages of gestation, CRH was detected in the endometrium on days 2, 3 and 4 of pregnancy and in the myometrium on days 3 and 4, whereas it was undetectable on day 5. The functional role of CRH during early gestation was evaluated by administering anti-CRH antibody to mice from day 3 to day 8 of pregnancy. This treatment resulted in implantation failure in 60% of the cases, in which implantation sites, although clearly present in the uterus, had failed to host an embryo. These results provide direct evidence about the involvement of CRH in murine embryo implantation and are in agreement with hypotheses postulated in humans. Journal of Endocrinology (1999) 163, 221–227
The distribution of CRH receptors in the pregnant and non-pregnant uterus indicates that CRH is an active peptide in the reproductive cycle and it has been postulated to participate in the processes of embryo implantation, endometrial vascularisation and labour by modulating myometrial contractibility (Makrigiannakis et al. 1977, Clifton et al. 1998). Specifically, in the early pregnant rat uterus, CRH mRNA and its end-product are found at higher levels in the implantation sites compared with uterine regions between implantation sites (Makrigiannakis et al. 1995b). Its involvement in the inflammatory process has also been demonstrated in vivo, where CRH immunoneutralisation discourages the development of an inflammatory response (Karalis et al. 1991). It has thus been postulated that endometrial CRH participates in the development of the inflammatory reaction during blastocyst anchorage to the uterine walls. This hypothesis is supported by the fact that prostaglandin E2, which is a major inducer of CRH expression in the
Journal of Endocrinology (1999) 163, 221–227 0022–0795/99/0163–221 1999 Society for Endocrinology Printed in Great Britain
Online version via http://www.endocrinology.org
222
I ATHANASSAKIS
and others ·
CRH in the mouse uterus before and during early pregnancy
placenta (Petraglia et al. 1989) and also stimulates the expression of CRH gene in transfected human endometrial cells (Makrigiannakis et al. 1996), is produced by blastocysts (Lewis 1987) to facilitate implantation (Chida & Mettler 1989). Furthermore, the vasodilatory effects of CRH (Hermus et al. 1987) can be considered equally important to embryo implantation, as secure vascular permeability surrounding the blastocyst anchorage site has to be developed. The present report defines the distribution of CRH production during anoestrus, pro-oestrus, oestrus and early pregnancy states in mice, and provides evidence about the participation of this peptide in the process of embryo implantation, by showing that administration of anti-CRH antibody during the early stages of pregnancy inhibits attachment of the embryo to the uterine implantation sites, thus demonstrating its direct involvement in pregnancy success. Materials and Methods Animals and tissue isolation BALB/c mice were housed in the Animal Facility of the Univesity of Crete (Department of Biology), in rooms with controlled light cycles (12 h light : 12 h darkness; lights on at 0600 h). Uteri that were non-pregnant after several days of vaginal examination were collected from sexually mature female mice (6–8 weeks old) during the three phases of the oestrous cycle (anoestrus–dioestrus, pro-oestus, and oestrus). Pregnant uteri were collected on days 2–5 of gestation. For experiments concerning the pre-implantation stages (days 2 and 3 of pregnancy), female mice were superovulated by receiving i.p. 5 IU pregnant mare serum (PMS, Sigma, St Louis, MO, USA) at 1400 h and, 46 h later, 5 IU human chorionic gonadotropin (Sigma). Five hours after the last injection, the female mice were individually caged with proven male breeders and examined for the presence of vaginal plug the following morning, which was considered as day 0 of pregnancy. Before the uteri were included in the study, the presence of embryos in the oviduct (day 2 of pregnancy) and uterus (days 3 of pregnancy) was verified. For the experiments concerning the collection of uteri on days 4 and 5 of pregnancy, female mice were only checked for oestrus, and were individually caged overnight with proven male breeders. In other experiments, female mice were mated to vasectomised males and, on the second day of pregnancy, they were transplanted with blastocysts isolated from superovulated females on the third day of pregnancy. Antibodies Rabbit anti-CRH polyclonal antibody (Penisula Labs, Baltimore, MD, USA) was used in a concentration of Journal of Endocrinology (1999) 163, 221–227
1 : 250 (containing 1 µg/ml specific anti-CRH antibody) for immunofluorescence and immunoperoxidase experiments, 1 : 1000 for ELISA experiments and in dilutions of 1 : 100, 3 : 50 000 and 1 : 50 000 for in vivo experiments. The antiserum cross-reacts with human, rat and mouse CRH, and has minimal affinity for ovine CRH or human ACTH, luteinizing hormone-releasing hormone (LHRH) and arginine vasopressin (AVP). Rabbit pre-immune serum was used as negative control in our experiments. In vivo animal manipulation Pregnant mice were injected i.p. from day 3 to day 8 of gestation with 100 µl different concentrations (1:100, 3:50 000, 1:50 000) of anti-CRH antibody or preimmune rabbit serum. Animals were killed on day 9 of pregnancy and the percentages of implanted embryos compared with empty implantation sites and resorptions were estimated. Cell cultures Uteri were put in single-cell suspension by cutting the tissue into small pieces using fine scissors and passing it through a syringe with an 18·5 gauge needle. The resulting cell suspension was washed three times and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, USA) at a concentration of 1 106 cells/ml in six-well plates (Nunc, Kampstrup, Denmark) for 2–5 days. Culture supernatants were collected, centrifuged and used in ELISA experiments, and at the same time, uterine cells were submitted to immunofluorescence staining. Enzyme-linked immunoassays Supernatant collected from days 2–4 of uterine cell cultures was used at the dilution of 1 : 2 in carbonate buffer pH 9·6 and coated onto 96-well flat bottom plates (Sarstedt, Newton, NC, USA), incubated overnight at 4 C and washed four times in 5% Tween-20. The remaining protein-free sites on the plate were blocked by 2% PBS–BSA solution after an incubation of 2 h at room temperature. After the plates had been washed four times, 100 µl test antibody diluted in 0·1% PBS–BSA were added and the system incubated for 1 h at room temperature. Extensive washing of the plate was followed by the addition of 100 µl goat anti-mouse IgG coupled to horseradish peroxidase (1 : 1000 dilution, Sigma) and incubation for 1 h at room temperature, in the dark. Finally, the reaction was developed by adding 100 µl/well tetramethyl benzidine–H2O2 (Sigma) for 20 min. The enzymatic reaction was stopped with 50 µl H2SO4 (4 N). Optical density (OD) was measured at 450 nm using a Titertec ELISA photometer (Digiscan, ASYS Hitech GmbH, Engendorf, Austria). Each experiment was repeated
CRH in the mouse uterus before and during early pregnancy ·
I ATHANASSAKIS
and others 223
at least four times. The results are expressed as the percent of OD increase over background (OD values of culture medium) ... calculated from four or more experiments. Indirect immunofluorescence Immunofluorescent staining of intracellular CRH content was performed as described by Sander et al. (1991). Uterine cells were scraped off the culture plates and fixed with ice-cold paraformaldehyde (4% in PBS) for 5 min. After washing, the cells were incubated for 30 min at room temperature with the first antibody diluted in Hanks’s balanced salt solution (HBSS)–saponin solution (HBSS: Gibco; 0·01 M HEPES: Gibco; 0·1% saponin: Sigma). After being washed in PBS–saponin, the cells were incubated with fluorescein isothiocyanate-conjugated rabbit anti-rat IgG antibody for 30 min at room temperature. The cells were washed, fixed in 25% glycerol, mounted on slides, and examined for cytoplasmic staining using a Leitz fluorescence microscope. Immunoperoxidase staining Frozen uteri were mounted onto the cryostat using M-1 embedding matrix (UK Lipshaw, Surrey, UK) and sections 6–8 µm thick were cut using a Leitz cryostat (Model 1720). The sections were collected on preheated slides, fixed in acetone for 10 min, and kept at 20 C until required for use. For the immunoperoxidase staining, we followed the technique described by Willingham (1990). The sections were then counterstained with Gill’s haematoxylin, dehydrated, mounted, and examined using a light microscope. Results CRH production was examined during the three phases of the oestrous cycle – including anoestrus (dioestrus), prooestrus and oestrus – by intracellular immunofluorescence staining and ELISA on uterine cultures and their supernatants respectively after 3 days of culture (Fig. 1). The greatest percentage of uterine cells with intracellular CRH was detected during pro-oestrous (415%) as assessed by immunofluorescence (Fig. 1A). However, the greatest concentration of CRH in culture supernatants of uterine cells detected by ELISA was obtained during anoestrus (403% over background), which corresponds to 0·25 pM as determined by comparison with a standard CRH control curve (Fig. 1B). As CRH is believed to be part of the inflammatory reaction taking place during implantation, we then investigated whether CRH could also be detected during early stages of pregnancy. We therefore isolated frozen sections of uteri during days 2, 3, 4 and 5 of pregnancy and
Figure 1 Detection of (A) intracellular and (B) secreted CRH activity by immunofluorescence and ELISA respectively, in cultured uterine cells during anoestrus (An), pro-oestrus (Pro) and oestrus (O) after 3 days of culture. The results represent the mean of four experiments and are expressed as the net percentage of positive cells (background values that varied from 3 to 7% have been subtracted) S.E. and the percent of OD increase over background (OD values obtained in the absence of culture supernatants) S.E., respectively.
determined the presence of CRH by immunoperoxidase staining (Fig. 2). CRH was detected on days 2 and 3 in the endometrium and on day 4 in the endometrium/ myometrium, whereas on the fifth day it was undetectable – results that indeed show the involvement of CRH in the embryo implantation reaction. In order to demonstrate the functional role of CRH in implantation that occurs on day 3·5 of gestation, antiCRH antibody was administered to pregnant mice from day 3 to day 8 of pregnancy and, 24 h after the last injection (day 9 of pregnancy), the number of implanted embryos compared with empty implantation sites was determined. To avoid possible neutralisation of hypothalamic CRH, which could disturb the nervous system, we determined the minimal antibody dilution giving a positive reaction with 1 nM CRH, considering that such a concentration would be expected to be found in the uterine lumen, as 1 106 uterine cells release 0·25 pM (see above). Therefore, the dilutions of the anti-CRH used were 1 : 50 000 (minimally reacting dilution), 3 : 50 000 (three times the minimally reacting dilution) and 1 : 100 (antibody given in excess). This treatment resulted in 60, 23 and 55% successfully implanted embryos, compared with 100% in non-treated mice (Table 1). Despite the observed inhibition of implantation by anti-CRH, the embryos that succeeded in implanting were normal and the feto–placental weight ranged from 334 to 438 mg, which did not significantly differ from that in control mice. In all mice tested, only one abortion was observed, indicating that anti-CRH does not affect embryo viability. The effect of anti-CRH can be seen in Fig. 3, in which a control day-9 pregnant uterus (Fig. 3A) is compared with a homogeneously haematosed nonpregnant uterus (Fig. 3C) and with the uterus of an anti-CRH-treated mouse (3 : 50 000 dilution; Fig. 3B) in Journal of Endocrinology (1999) 163, 221–227
224
I ATHANASSAKIS
and others ·
CRH in the mouse uterus before and during early pregnancy
Figure 2 Localisation of CRH in uterine tissue frozen sections on days 2 to 5 of pregnancy by immunoperoxidase staining. Control sections were stained with preimmune serum and only background reaction is seen. Test sections were reacted with anti-CRH rabbit serum in a dilution of 1 : 250. The endometrium (E) and myometrium (M) areas, and some areas of positive staining (arrows) are marked. Original magnification 10. Journal of Endocrinology (1999) 163, 221–227
CRH in the mouse uterus before and during early pregnancy ·
I ATHANASSAKIS
and others 225
Table 1 Effect of administration of anti-CRH to pregnant mice. Mice were treated with different doses of anti-CRH from day 3 to day 8 of pregnancy and killed on day 9
Anti-CRH dilution Control 1 : 50 000 3 : 50 000 1 : 100
Mice (No.)
Implantation sites (No.)
Non-implanted embryos (No.)†
Implanted embryos (No.)
7 9 6 6
42 53 34 38
0 21 26 17
42 32 8 21
†Calculated by subtracting the number of implanted embryos from the number of implantation sites.
which a uterus with 5 implantation sites carrying no embryos and one abortion is presented. The empty implantation sites (seen as small spots) are the most striking effect of the anti-CRH treatment during pregnancy, detectable in all mice tested. To confirm these observations, pseudopregnant mice were transplanted, in a left uterine site, with blastocysts bathed in anti-CRH immune serum (1 : 100 dilution) and, in a right uterine site, with blastocysts bathed in control preimmune serum (1 : 100 dilution). In all 10 mice tested, embryo implantation and development were seen only in the right site, whereas the left site contained only empty implantation sites. Discussion The presence of CRH in the uterine tissues and placenta has been demonstrated in humans (Makrigiannakis et al. 1977, 1995a, Shibasaki et al. 1982), baboons (Davies et al. 1996) and sheep ( Jones et al. 1989), in which, in view of its ability to modulate myometrial contractibility, it has been postulated to be involved in embryo implantation and labour. The present study has localised CRH in the non-pregnant and early pregnant uterus in mice and demonstrated its active role during ovum implantation. In non-pregnant uteri, the percentage of CRHreleasing cells was initially assayed by immunofluorescence, and a maximal number of uterine cells showing intracellular CRH activity was detected during prooestrus, whereas minimal numbers of CRH-positive cells were obtained during oestrus – a decrease that can be correlated with the suppressive activity exerted by oestrogens on the CRH promoter (Makrigiannakis et al. 1996). When the CRH activity in the corresponding culture supernatants was determined by ELISA, maximal CRH concentration was obtained during anoestrus. The observed difference between intracellular and extracellular CRH indicates that fine regulatory mechanisms dictate the release of specific amounts of CRH in the nonpregnant uterus. In a previous study, Grammatopoulos
et al. (1998) showed that human myometrium expressed different CRH receptor subtypes during pregnant and non-pregnant states. The existence of four different subtypes of CRH receptors in the uterus indicates that even a small amount of CRH can be very effective when bound to a high-affinity receptor. Taken together, these results indicate the complexity of CRH regulation within the uterine tissue, and underline the important role of this molecule during the reproductive cycle. In order to localise CRH activity in the early pregnant uterus in mice, frozen sections from day 2 to day 5 of pregnancy were submitted to immunoperoxidase staining. CRH was detected in the endometrium on days 2, 3 and 4 of pregnancy and in the myometrium on day 4, whereas no CRH activity was detected on the fifth day of gestation. As blastocyst implantation in mice takes place on the third day of pregnancy, the presence of CRH in the endometrium during this period indicates that it might be involved in the reaction to implantation, as it has also been postulated by other investigators (Makrigiannakis et al. 1995b). In an attempt to estimate the involvement of CRH in blastocyst implantation, pregnant mice were treated with anti-CRH antibody from day 3 to day 8 of pregnancy, and the number of implanted embryos compared with empty implantation sites was estimated. Although such treatment cannot distinguish between uterine and hypothalamic CRH, the results showed that the administration of anti-CRH significantly descreased the implantation rate, reaching an inhibition of 77% when the antibody was used in a 3 : 50 000 dilution. Furthermore, using embryo transfer techniques, blastocysts in control preimmune serum or in the presence of anti-CRH antibody were transferred to right and left uterine sites of pseudopregnant females, respectively. Embryo implantation and growth was seen only in the right site of the uterus, where embryos were bathed in control serum. The results thus indicate that CRH has a direct role in the process of implantation; however, the mechanisms controlling its action remain to be determined. Inflammatory mediators are provided to the implantation site, not only by the endometrium (production of interleukins-1 and -6), but also by the implanting blastocyst which secretes, among Journal of Endocrinology (1999) 163, 221–227
226
I ATHANASSAKIS
and others ·
CRH in the mouse uterus before and during early pregnancy
References
Figure 3 Uteri of (A) control pregnant mice, (B) CRH-treated pregnant mice, and (C) non-pregnant mice. Arrows indicate the empty implantation sites in the CRH-treated pregnant uterus (B).
other factors, interleukin-1 and prostaglandin E2 (Lewis 1987, Makrigiannakis et al. 1995b). In support of the results presented here, it has previously been shown (Makrigiannakis et al. 1996) that prostaglandin E2 increases the activity of the CRH promoter in a dosedependent manner, which could have a positive effect, not only in the process of blastocyst implantation, but also in vasodilatation and myometrial contractibility (Zoumakis et al. 1997), thus ensuring the successful progression of pregnancy. Journal of Endocrinology (1999) 163, 221–227
Antoni FA 1986 Hypothalamic control of adrenocorticotropin secretion: advances since the discovery of 41-residue corticotropinreleasing factor. Endocrine Reviews 7 351–362. Chida S & Mettler L 1989 Effetcs of indomethacin, prostaglandin E2 and prostaglandin F2 alpha on mouse blastocyst attachment and trophoblastic outgrowth in vitro. Prostaglandins 37 411– 416. Clifton VL, Tefler JF, Thompson AJ, Cameron IT, Teoh TG, Lye SJ & Challis JR 1998 Corticotropin-releasing hormone and proopiomelanocortin-derived peptides are present in human myometrium. Journal of Clinical Endocrinology and Metabolism 83 3716–3721. Davies WC, Albrecht ED & Pepe CJ 1996 Hypothalamic corticotropin-releasing hormone expression in the baboon fetus at mid- and late gestation. Biology of Reproduction 55 559–566. Grammatopoulos D, Dai Y, Chen J, Karteris E, Papadopoulou N, Easton AJ & Hillhouse EW 1998 Human corticotropin-releasing hormone receptor: differences in subtype expression between pregnant and nonpregnant myometria. Journal of Clinical Endocrinology and Metabolism 83 2539–2544. Grino M, Chrousos G & Margioris A 1987 The corticotropinreleasing hormone gene is expressed in human placenta. Biochemical and Biophysical Research Communications 148 1208–1214. Hermus AR, Pieters GF, Willesmen JJ, Ross HA, Smals AG & Benraad D 1987 Hypotensive effects of ovine and human CRF in man. European Journal of Clinical Pharmacology 31 531–537. Jones CT, Gu W & Parer JT 1989 Production of corticotropin releasing hormone by sheep placenta in vivo. Journal of Developmental Physiology 11 97–101. Karalis K, Sano H, Redwine J, Listwark S, Wilder R & Chrousos G 1991 Autocrine or paracrine inflammatory actions of corticotropinreleasing hormone in vivo. Science 254 421–423. Lewis GS 1987 Prostaglandin secretion by the blastocyst. Journal of Reproduction and Fertility 37 261–267. Makrigiannakis A, Psychoyos A, Zoumakis E, Margioris AN, Stournaras C & Gravanis A 1977 Endometrial corticotropinreleasing hormone: expression, regulation, and protein physiological implications. Annals of the New York Academy of Sciences 816 116–128. Makrigiannakis A, Zoumakis E, Margioris AN, Theodoropoulos P, Stournaras C & Gravanis A 1995a The corticotropin-releasing hormone (CRH) in normal and tumoral epithelial cells of human endometrium. Journal of Clinical Endocrinology and Metabolism 80 185–189. Makrigiannakis A, Margioris AN, Le Goascogne C, Zoumakis E, Nikas G, Stournaras C, Psychoyos A & Gravanis A 1995b Corticotropin-releasing hormone (CRH) is expressed at the implantation sites of early pregnant rat uterus. Life Sciences 57 1869–1875. Makrigiannakis A, Zoumakis E, Margioris AN, Stournaras C, Chroussos GP & Gravanis A 1996 Regulation of the promoter of the human corticotropin-releasing hormone gene in transfected human endometrial cells. Neuroendocrinology 64 85–92. Petraglia F, Sawchenko PE, Rivier J & Vale W 1989 Evidence for local stimulation of ACTH secretion by corticotropin-releasing hormone in human placenta. Nature 328 717–719. Petraglia F, Tabanelli S, Galassi C, Garuti GC, Mancini A, Genazzani A & Gurpide E 1992 Human decidua and in vitro desidualized endometrial stromal cells at term contain immunoreactive CRF and CRF mRNA. Journal of Clinical Endocrinology and Metabolism 74 1427–1431. Sander B, Andersson J & Andersson U 1991 Assessment of cytokines by immunofluorescence and the paraformaldehyde–saponin procedure. Immunological Reviews 119 65–93.
CRH in the mouse uterus before and during early pregnancy · Sasaki A, Shinkawa O, Margioris AN, Liotta A, Sato S, Murakami O, Go M, Shimizu Y, Hanew K & Yoshinaga K 1987 Immunoreactive corticotropin-releasing hormone in human plasma during pregnancy and parturition. Journal of Clinical Endocrinology and Metabolism 64 1054–1059. Shibasaki T, Odagiri E, Shizume K & Ling N 1982 Corticotropinreleasing factor-like activity in human placental extracts. Journal of Clinical Endocrinology and Metabolism 55 384–386. Willingham MC 1990 Immunocytochemical methods: useful and informative tools for screening hybridomas and evaluating antigen expression. Focus 2 62–67.
I ATHANASSAKIS
and others 227
Zoumakis E, Margioris AN, Makrigiannakis A, Stournaras C & Gravanis A 1997 Human endometrium as a neuroendocrine tissue: expression, regulation and biological roles of endometrial corticotropin-releasing hormone (CRH) and opiod peptides. Journal of Endocrinological Investigation 20 158–167.
Received 22 February 1998 Revised manuscript received 17 May 1999 Accepted 8 June 1999
Journal of Endocrinology (1999) 163, 221–227
