Synergistic role of nitric oxide and progesterone during the ...

33 downloads 0 Views 566KB Size Report
Decidual cells are believed to play a key role in providing nutrients to the embryo, and in controlling tropho- blast invasion (Parr and Parr, 1989; Loke and King, ...
Human Reproduction vol.14 no.2 pp.542–552, 1999

Synergistic role of nitric oxide and progesterone during the establishment of pregnancy in the rat*

Kristof Chwalisz1,4, Elke Winterhager2, Thomas Thienel2 and Robert E.Garfield3 1Research Laboratories of Schering AG, Mullerstr 170–178, 13342 ¨ Berlin, 2Institute of Anatomy, Medical School, University of Essen, Germany, 3University of Texas Medical Branch, Galveston, USA 4To

whom correspondence should be addressed

Successful pregnancy is strictly dependent on the trophoblast-decidual interaction and on an adequate blood supply to the implantation sites. Nitric oxide (NO) has been shown to play an important role during advanced gestation, although its role during early pregnancy is unclear. The aim of the present study in rats was to evaluate whether NO plays a role during the preimplantation [days 1–4 post coitum (p.c.)] and peri-implantation (days 6–8 p.c.) phases of pregnancy. The rats were treated with the non-specific nitric oxide synthase (NOS) inhibitor NG-nitro-L-arginine methyl ester (L-NAME), and the iNOS inhibitor aminoguanidine in the presence and absence of low-dose antiprogestin, onapristone, and evaluated on days 9 p.c. and 19 p.c., respectively. Before implantation, the treatments alone (L-NAME, aminoguanidine, onapristone) had little effect on pregnancy outcome. Conversely, aminoguanidine plus onapristone treatment completely prevented pregnancy, whereas L-NAME plus onapristone reduced the pregnancy rate to approximately 50%. In addition, both treatments drastically reduced decidualization. Oviductal flushing experiments revealed arrest of embryo development at around the 8-cell stage after aminoguanidine plus onapristone treatment on days 1–4 p.c. Similarly, treatment during the peri-implantation period with L-NAME, aminoguanidine, and onapristone each had only marginal effects on pregnancy. However, a combination of L-NAME and onapristone, and aminoguanidine plus onapristone prevented pregnancy in 71% and 42% of dams, respectively, as determined on day 19 p.c. These treatments also markedly inhibited the decidualization process. This study demonstrates synergistic effects of NOS inhibitors and an antiprogestin in preventing pregnancy. NOS, particularly the cytokine- and progesterone-inducible iNOS, may represent a new target for novel therapeutic agents capable of promoting or inhibiting pregnancy. Key words: antiprogestin/embryo development/implantation/ nitric oxide/progesterone

*This work was presented in part at the 44th and 45th Annual Meetings of the Society for Gynecological Investigation in San Diego, California (March, 1997), and in Atlanta, Georgia, USA (March, 1998), respectively.

542

Introduction Implantation of the embryo is a critical event in pregnancy. In humans, peri-implantation pregnancy loss may contribute to more than 20% of unexplained infertility. Approximately onethird of human pregnancies end in spontaneous abortions, twothirds of them occurring prior to the clinical detection of pregnancy (Wilcox et al., 1988). Successful implantation of blastocysts is strictly dependent on the synchrony between embryo development and the cascade of cellular and molecular events that occur in the endometrium during the peri-implantation period of the fertile cycle (Beier et al., 1994). In rodents during early pregnancy, the surrounding endometrial stroma undergoes a dramatic differentiation into the decidua, a specialized, well-vascularized tissue that encapsulates the developing embryo. Decidual cells are believed to play a key role in providing nutrients to the embryo, and in controlling trophoblast invasion (Parr and Parr, 1989; Loke and King, 1995). In rodents, the decidual cell reaction occurs in response to either blastocysts or artificial stimuli and is always preceded by an increase in endometrial vascular permeability which is normally initiated at the antimesometrial sites where blastocysts implant (Psychoyos, 1973). The exact factors which invoke decidualization of the endometrial stroma are still poorly understood. Cytotrophoblast invasion, encompassing dynamic changes in cell–cell and cellmatrix interactions, can be viewed as an inflammatory reaction. There is ample evidence indicating that prostaglandins mediate, at least in part, the changes in endometrial vascular permeability and subsequent decidualization in pregnant as well as in pseudo-pregnant animals. Hence, prostaglandin concentrations are elevated at implantation sites in areas of increased permeability (Kennedy, 1980; Evans and Kennedy, 1978), and prostaglandin inhibitors, such as indomethacin, delay or inhibit (i) implantation, (ii) the localized increase in endometrial vascular permeability, and (iii) artificially-induced decidualization in several non-primate mammals (Kennedy, 1977; Psychoyos et al., 1995). In addition, the targeted disruption of the cytokineinducible cyclooxygenase (COX) isoform COX-2, but not COX-1 (a constitutive isoform), impairs fertilization, implantation and decidualization in mice (Lim et al., 1997). In rhesus monkeys, however, post-ovulatory treatment with high-dose diclofenac, a non-specific COX inhibitor, had only a partial effect on pregnancy establishment (Nayak et al., 1997) suggesting that prostaglandins are not essential in primate implantation. It is, therefore, likely that other vasoactive agents are able to replace prostaglandins in order to support vascularization and the decidual reaction during implantation. Nitric oxide (NO) has come to prominence recently as a © European Society of Human Reproduction and Embryology

Nitric oxide and implantation

major mediator of numerous biological processes, including vascular functions and inflammation (Ignarro, 1989; Moncada and Higgs, 1993). NO is produced by nitric oxide synthesizing enzymes (NOS enzymes) which exist as three isoforms: ecNOS or endothelial NOS; iNOS or cytokine-inducible NOS, and nNOS or neural NOS (Nathan and Xie, 1994). NO was recently implicated as an important regulatory agent in various steps of female reproduction (reviewed by Chwalisz et al., 1996; Rosseli, 1997). During pregnancy NO plays a pivotal role in controlling uterine contractility, cervical ripening, and utero– feto–placental blood flow (Izumi et al., 1993; Sladek et al., 1993; Buhimschi et al., 1996; Liao et al., 1997). Our previous studies have shown that in rats the iNOS is the dominant NOS isoform in the pregnant uterus, cervix and placenta (Buhimschi et al, 1996; Purcell et al., 1997; Ali et al., 1997). In rats, the uterine, cervical and placental iNOS expression is gestationally regulated at protein and mRNA levels, with progesterone as the key physiological regulating agent (Buhimschi et al., 1996; Chwalisz et al., 1996; Ali et. al., 1997). The effects of progesterone on iNOS expression are tissue-dependent. In the uterus and placenta progesterone up-regulates iNOS expression and NO production, whilst exerting opposite effects in the cervix. The L-arginine-NO-cGMP system (NO system) is also present in the decidua in rats (Purcell et al., 1997; Sladek et al., 1998) and rabbits (Sladek et al., 1993). An increased expression of both iNOS and ecNOS was recently found in the mouse uterus during implantation (Purcell et al., 1998). In addition, increased expression of iNOS was described in human decidualized stromal cells (Telfer et al., 1997) and in the secretory endometrium (Tschugguel et al., 1998). Overall, these data suggest that (i) iNOS is the dominant NOS isoform in the endometrium, (ii) endometrial iNOS is directly or indirectly controlled by progesterone, and (iii) NO may play a role in the local control of endometrial functions. However, the precise function of NO during implantation has not yet been defined. Meanwhile, it is well established that iNOS-derived NO may invoke pro-inflammatory effects, including vasodilatation, oedema, tissue remodelling, and the mediation of cytokinedependent processes at the inflammation site (Evans et al., 1995; Salvemini et al., 1995). It is also well known that the proinflammatory cytokines are up-regulated in the uterus during implantation (Simo´n et al., 1993; Chard, 1995; Tazuke and Giudice, 1996). Studies with antiprogestins clearly show that progesterone is essential for endometrial receptivity and for the entire implantation process in all mammals investigated to date (reviewed by Chwalisz et al., 1997). However, it still remains unclear how exactly progesterone acts in the uterus during the pre- and peri-implantation stages of pregnancy. Because NO mediates certain progesterone effects in the myometrium, placenta and cervix, we hypothesize that NO and progesterone play a synergistic role during the establishment of pregnancy. To test this hypothesis, we performed a series of functional studies in rats using the NOS inhibitors NG-nitroL-arginine methyl ester (L-NAME, a non-specific NOS inhibitor) and aminoguanidine [a 20- to 30-fold selective inhibitor of iNOS (Misko et al., 1993)], and an antiprogestin (onapristone) prior to and during implantation. Our aims were: (i) to

Figure 1. Experimental design.

evaluate whether NO is involved in the preparation of the endometrium for implantation during the pre-receptive phase (days 1–4 post coitum); (ii) to determine whether it plays a role during the peri-implantation phase (days 6–8 p.c.); and (iii) to examine the interaction of NO with progesterone during implantation. Materials and methods Animals Early pregnant female Wistar rats (Schering, TZH, Berlin, Germany) were used for all experiments. The animals were kept under standard conditions. The light/dark-cycle was 14/10 h (light 6:30–20:30). The presence of sperm in the vaginal smear on the morning following mating was defined as day 1 post coitum (day 1 p.c.). Compounds and formulations L-NAME (Sigma-Aldrich Chemie, Munich, Germany) was dissolved in sterile 0.9% saline solution. Osmotic minipumps (Alza, Palo Alto, CA, USA; model 2ML1 with a pumping rate of 10 µl/h) were filled with vehicle or with L-NAME solution and implanted s.c. during ether anaesthesia. The drug-release was verified by opening the minipumps during autopsy to determine the remaining volume. Aminoguanidine (Sigma-Aldrich Chemie) was dissolved in water at pH 6.0 and administered orally (p.o.) in 1 ml. The specific progesterone antagonist onapristone (ZK 98 299; 11β-[4-(dimethylamino)phenyl]17α-hydroxy-17β-(3-hydroxypropyl)-13 α-estra-4,9-dien-3-one) was formulated in 0.2 ml benzylbenzoate 1 castor oil (1 1 4 vol/vol) and administered s.c. Onapristone was synthesized at Schering AG, Berlin, Germany. Experimental design Treatments with L-NAME (50 mg/rat/day s.c.) and aminoguanidine (120 mg/rat/day) alone as well as in combination with low-dose onapristone (0.3 mg/rat/day s.c.) were performed after randomization in accordance with two experimental protocols (Figure 1A, B). L-NAME at this dose produced hypertension and fetal growth restriction in late pregnant rats (Liao et al., 1997). The aminoguanidine dose used was based on the results of pilot dose-finding studies

543

K.Chwalisz et al.

of the synergistic effects of onapristone on fertility in rats (data not shown). The effects of NO inhibition and low-dose onapristone on implantation after treatment during the preimplantation phase (days 1–4 p.c; Figure 1A) Experiment 1: the effects of L-NAME in the presence and absence of onapristone on implantation after treatment on days 1–4 p.c. Osmotic minipumps containing L-NAME or 0.9% saline (vehicle) were implanted s.c. on day 1 p.c. and removed on day 4 p.c. Onapristone or the respective vehicle (controls) was administered (s.c.) once daily on days 1–4 p.c. (n 5 7/group). Group 1 (vehicle control) received the respective vehicles (saline-containing mini-pump and onapristone vehicle). Group 2 was treated with L-NAME and the onapristone vehicle. Group 3 received onapristone and the L-NAME vehicle (i.e. saline-containing mini-pump was implanted). Group 4 received a combination of L-NAME plus onapristone. Peripheral blood for progesterone and oestradiol radioimmunoassays was collected from the retro-orbital plexus during CO2 anaesthesia at 9–10 a.m. on days 1, 5, 7 and 9 p.c. During autopsy on day 9 p.c. the uteri were opened, photographed and macroscopically examined for implantation sites and resorptions. They were then removed and processed for histology. Experiment 2: the effects of aminoguanidine in the presence and absence of onapristone on implantation after treatment on days 1–4 p.c. This experiment was performed in accordance with the protocol described for experiment 1, using the same animal numbers per group. Group 1 (controls) was treated orally and s.c. with both vehicles. Group 2 received aminoguanidine orally and onapristone vehicle s.c. Group 3 was treated s.c. with onapristone and with the aminoguanidine vehicle orally. Group 4 was treated orally with aminoguanidine and s.c. with onapristone at the respective dose. Peripheral blood collections and autopsy (day 9 p.c.) were performed as described for experiment 1. Experiment 3: pregnancy outcome (on day 19 p.c). after treatment with L-NAME and aminoguanidine in the presence and absence of onapristone on days 1–4 p.c. Treatments were performed as described for experiments 1 and 2. However, in order to assess the effects of treatment on pregnancy outcome, the autopsy was performed on day 19 p.c., i.e. approximately 3 days before birth (term: days 21–22 p.c.). During autopsy, the number and condition of pups, fetal weight, placental weight and the number of resorptions per uterus were recorded. In addition, the empty uteri were stained with 10% ammonium sulphide in order to identify early resorptions. Experiment 4: the effects of combined aminoguanidine and onapristone treatment on tubal transport, preimplantation embryo development and endometrial receptivity after treatment on days 1–4 p.c. The rats (n 5 8/group) were treated with a combination of aminoguanidine and low-dose onapristone or the vehicles (controls) at 9 a.m. on days 1–4 p.c. The autopsy was performed between 1–2 p.m. on days 4 p.c. [groups 1 and 2 (controls)], 5 p.c. [groups 3 and 4 (controls)], and 6 p.c. [groups 5 and 6 (controls)]. During autopsy the uteri and oviducts were dissected and flushed with 0.9% saline which was subsequently examined for the presence or absence of preimplantation embryos using inverted phase contrast microscopy. The effects of NO inhibition and low-dose onapristone on implantation after treatment during the peri-implantation phase (days 5–8 p.c.; Figure 1B) Experiment 5: the effects of L-NAME alone or in combination with low-dose onapristone after treatment on days 5–8 p.c. Pregnant rats (n 5 8/group) were treated on days 5–8 p.c. using the same groups as described for experiment 1. Experiment 6: the effects of aminoguanidine alone and in combination

544

with low-dose onapristone after treatment on days 5–8 p.c. Rats were randomly allocated to four groups (n 5 7/group) and were treated on day 5–8 p.c. The same treatment groups as described for experiment 2 were used. Experiment 7: pregnancy outcome (on day 19 p.c). after treatment with L-NAME and aminoguanidine alone and in combination with low-dose onapristone on days 5–8 p.c. The animals were treated on days 5–8 p.c. analogously with experiment 3. The following groups were used (see Table II for numbers): group 1: vehicle control; group 2: onapristone alone; group 3: L-NAME alone; group 4: aminoguanidine alone; group 5: L-NAME plus onapristone; group 6: aminoguanidine plus onapristone. Morphology In experiments 2, 5 and 6, the uteri were fixed in 4% paraformaldehyde overnight and routinely embedded in paraffin. The uteri were then cut into serial sections, stained with haematoxylin/eosin and viewed on a Zeiss Axiophot® microscope. Radioimmunoassays Progesterone and oestradiol concentrations were measured in serum samples after ether extraction with a specific radioimmunoassay as previously described (Hasan and Caldwell, 1971). Statistical analysis For the analysis of hormonal data, the differences between experimental days 5, 7 and 9 in each individual group were calculated. The logarithmic ratios of the treatment groups were then compared with those of the control group on each day using Dunnett’s test at the significance level of α 5 0.05. The treatment effects on pregnancy rates were analysed using one-sided Fisher’s exact test. For the statistical analysis of treatment effects on implantation numbers, the Wilcoxon test for comparison between the groups was applied.

Results The effects of NO inhibition and low-dose onapristone after treatment during the preimplantation phase (days 1–4 p.c.) The effects on pregnancy rates and implantation numbers after treatment on days 1–4 p.c (experiments 1 and 2) Neither L-NAME nor aminoguanidine alone significantly affected pregnancy rates (Figure 2A and B) nor implantation numbers (Figure 2C and D) when administered before implantation. The implantation sites were normal in these groups (Figure 2; black bars). However, treatment with onapristone alone reduced the total number of implantation sites in both experiments by approximately 50–60% compared to controls (Figure 2C and D), the majority of them being macroscopically changed (smaller and in part haemorrhagic; presented as striped bars). In addition, the effects of each NOS inhibitor in combination with low-dose onapristone on both pregnancy rates and implantation numbers were quite dramatic. After L-NAME plus onapristone treatment, implantation sites were detectable in only two animals: six normal implantation sites in one animal and one very small implantation site in the second animal (Figure 2A and C). After aminoguanidine plus onapristone treatment, only two highly retarded and haemorrhagic implantation sites were observed in 1/7 animals (Figure 2B, D). Thus, the combination of onapristone and aminoguanidine completely prevented pregnancy.

Nitric oxide and implantation

Figure 2. The effects of the NOS inhibitors L-NAME (LN) and aminoguanidine (A) in the presence or absence of low-dose onapristone (ONA) on pregnancy rates (upper panel; A, B) and implantation number (lower panel; B, C) established on day 9 p.c. after treatment at preimplantation (day 1–4 p.c.). There were seven animals per group. The autopsy was performed on day 9 p.c. Upper panel (A, B) presents the percentage of pregnant animals per experimental group. Black bars demonstrate the percentage of animals exhibiting macroscopically normal implantation sites. Striped bars present the percentage of animals showing macroscopically changed implantation sites (haemorrhagic and growth retarded). Lower panel (C, D) shows the total number of implantation sites per group. Black bars present the number of normal implantation sites, whereas striped bars demonstrate the number of macroscopically changed implantation sites. Asterisks* mark the statistically different values (P , 0.05) of normal implantation sites (black bars) from the respective control group.

The effects on serum progesterone and oestradiol concentrations after treatment on days 1–4 p.c. (experiments 1 and 2) Figure 3 presents the results of progesterone measurements. Neither treatment before implantation (experiments 1 and 2) significantly influenced the serum progesterone (Figure 3) and oestradiol concentrations (data not shown) compared to the control animals. The effects on implantation chamber morphology after treatment on days 1–4 p.c. (experiments 1 and 2) For morphological evaluation, cross-sectional slides of the middle part of the implantation chambers of each experimental group were selected. Vehicle control groups The cross-sections of an implantation chamber of all controls (Figure 4A, B) on day 9 p. c. revealed a high degree of stromal decidualization surrounding the developing embryo. Decidualization was most developed at the antimesometrial side, whilst the decidual cells formed a special wall-like region surrounding the embryo, the primary

Figure 3. The effect of the NOS inhibitors L-NAME (LN) and aminoguanidine (A) in the presence or absence of low-dose onapristone (ONA) on serum progesterone concentrations after treatment at preimplantation (day 1–4 p.c.). Upper panel: L-NAME 6 onapristone treatment. Lower panel: Aminoguanidine 6 onapristone treatment. Blood samples were taken on days 1, 5, 7 and 9 p.c. C 5 vehicle control. The results are expressed as means 6 SD. For statistical analysis Dunnett’s test at the significance level of α 5 0.05 was used. There were no significant differences between the treatment groups and the control.

decidual zone. Higher magnification demonstrates the typical morphological features of the decidual cells (Figure 4B) with closely packed polygonal, partly multinuclear cells, separated only by the small extracellular matrix. L-NAME and onapristone alone Compared to controls, treatment with either L-NAME or onapristone alone before implantation produced relatively subtle changes in endometrial morphology. Each drug exerted virtually identical effects on decidual differentiation, thus both are described together. The differentiation process seemed retarded after L-NAME or onapristone alone. The decidual cells were more loosely connected (Figure 4D). Typically, the closure reaction in the mesometrial part remained incomplete, despite the degeneration of the epithelium leading to a kind of cleft connecting the antimesometrial chamber to the uterine lumen (Figure 4E and F). Aminoguanidine only treatment Aminoguanidine alone had virtually no effect on the implantation reaction which was indistinguishable from those treated with the vehicle (data not shown) on days 1–4 p.c. Treatment with L-NAME plus onapristone Dramatic changes 545

K.Chwalisz et al.

occurred after combined treatment with L-NAME and onapristone. The main morphological disturbances were: incomplete decidualization, incomplete closure of the uterine lumen, and hyperplasia of the uterine epithelium (Figure 4G). Only in one case could signs of stromal decidualization be observed (Figure 4H). In this implantation chamber, merely a compact primary decidual zone was formed without enclosing any remnants of embryonic tissues (Figure 4G). Treatment with aminoguanidine plus onapristone Surprisingly, no implantation reaction was found in any animal, the morphological appearance of the uteri being comparable to those of non-pregnant animals. The serial uterine sections of animals treated with this combination revealed neither the presence of decidualization nor any evidence of early resorptions. The effects on pregnancy outcome (on day 19 p.c.) of treatment on days 1–4 p.c. (experiment 3) The results are summarized in Table I. No pregnancies were found after aminoguanidine plus onapristone treatment. After a combined L-NAME plus low-dose onapristone treatment there was a 50% inhibition of fertility and an increase in early resorptions. Only seven live, but extremely growth-retarded, fetuses were found in this group. There were no inhibitory effects on pregnancy rates after either treatment alone; however, onapristone treatment alone significantly reduced the number and weight of living fetuses.

Figure 4A–H. Cross sections of an implantation chamber on day 9 p.c. after treatment with L-NAME (50 mg/rat/day) in the presence and absence of low-dose onapristone (0.3 mg/rat/day) before implantation (days 1–4 p.c.). Vehicle control group (A): the embryo proper (EM) is surrounded by compact, well-developed decidual tissue up to the myometrial layers. The decidua is subdivided into different regions: the primary decidual zone (PD), the mesometrial (M) and antimesometrial (AM) decidua. B: the polygonal decidual cells of the antimesometrial region are densely packed and the extracellular matrix is virtually missing. Some of the decidual cells contain several nuclei (arrow). L-NAME alone (C): overview of an implantation chamber treated with L-NAME during preimplantation. The decidua is well developed enclosing an appropriate differentiated embryo (EM). D: higher magnification of the antimesometrial side reveals more loosely-packed decidual cells compared to controls. Some spindle shaped cells can be observed (arrow). Onapristone alone (E): decidualization is reduced. Closure reaction is incomplete, the uterine lumen being still preserved (arrow). F: antimesometrial decidual tissue is less compact compared to both control and L-NAME-treated animals, and the cells are separated by matrix. L-NAME plus onapristone (G): a dramatic reduction in decidualization without embryonic tissue. H: a compact primary decidual zone (PD) at higher magnification. Stromal cells are not completely differentiated into decidual cells, as shown by small cells and large amounts of intracellular matrix. (AM, antimesometrial side; M, mesometrial side. Slides A, C, E, G are at the same magnification, i.e. represented by 400 µm bar in G. Slides B, D, F are at the same magnification and shown with a 30 µm bar in F. Bar in H represents 100 µm).

546

The effects of combined aminoguanidine and onapristone treatment on preimplantation embryo development, tubal transport and nidation after treatment on days 1–4 p.c. (experiment 4) The results of oviductal and uterine flushing on days 4 and 6 p.c. (first day after attachment) are presented in Figure 5. On day 4 p.c, the total number of 74 embryos at various developmental stages were recovered from the oviducts of the control animals. In contrast, only 25 pathologically changed embryos (small size, irregular plasma membranes and necrotic changes), mostly at the 8-cell stage, could be recovered from the oviducts of animals treated with aminoguanidine plus onapristone. On day 6 p.c., 77 well-developed implantation sites were recorded in control animals, whereas in animals treated with aminoguanidine plus onapristone, only small points of increased vascular reaction could be observed. There was no evidence of tubal transport time change (e.g. accelerated tubal transport) after treatment with aminoguanidine plus onapristone. The effects of NO inhibition and low-dose onapristone after treatment during the peri-implantation phase (days 5–8 p.c.) The effects on pregnancy rates and implantation numbers after treatments on days 5–8 p.c. (experiments 5 and 6) Neither treatment alone had any significant effect on pregnancy rate (Figure 6A and B), nor on the number of implantation sites (Figure 6C and D) on day 9 p.c. Almost all of the implantation sites were macroscopically unchanged (presented as black bars in Figure 6). However, in combination with low-dose onapristone, both L-NAME and aminoguanidine significantly reduced pregnancy rates and the total number of implantation sites. Moreover, in those animals which remained

Nitric oxide and implantation

Figure 5. The effects of combined aminoguanidine (A) plus onapristone (ONA) treatment on preimplantation embryo development on day 4 p.c. (left panel) and implantation rate on day 6 p.c. (right panel). There were eight animals per group. During autopsy the uteri and oviducts were dissected, flushed with 0.9% saline and examined for the presence or absence of preimplantation embryos using inverted phase contrast microscopy. White bars: normal preimplantation embryos, striped bars: abnormal preimplantation embryos, black bars: nidation points (N). Numbers 1, 2, 4, 8, 16 on the x-axis 5 cell stage of the preimplantation embryos, M: morula, B 5 blastocyst.

Table I. Outcome of pregnancy on day 19 p.c. after treatment with the NOS inhibitors NG-nitro-L-arginine methyl ester (L-NAME) and aminoguanidine (A) alone and in the presence and absence of low-dose onapristone (ONA) during the pre-receptive phase (days 1–4 p.c.; experiment 3). Asterisks mark the values which are significantly different from the vehicle control group (P , 0.05) Group

Treatment

Total dams (n)

Dams pregnant (n)

Implantation sites (n)

Fetal weight (g) (mean 6 SD)

1 2 3 4 5 6

Vehicle ONA L-NAME A L-NAME 1 ONA A 1 ONA

6 7 7 7 6 7

6 7 7 7 3* 0*

64 53 67 63 25* 0*

1.43 1.06 1.43 1.54 0.76 -

pregnant, most implantation sites were macroscopically abnormal (necrotic and haemorrhagic areas, and small size; presented as striped bars in Figure 6). In animals treated with L-NAME plus onapristone, 23 normal versus 62 macroscopically abnormal implantation sites were recorded (Figure 6A and C). Similarly, after aminoguanidine plus onapristone treatment, only eight macroscopically normal implantation sites were discovered (Figure 6B and D). The effects on serum concentrations of progesterone and oestradiol after treatment on days 5–8 p.c. Neither treatment in either experiment had any significant effect on serum progesterone concentrations on days 1, 5 and 7 p.c. (Figure 7). A slight, but statistically significant reduction in serum progesterone levels was observed only after L-NAME plus onapristone treatment on day 9 p.c. No statistically significant changes in serum oestradiol concentrations were observed with either treatment (data not shown). The effects on implantation chamber morphology after treatment on days 5–8 p.c. Onapristone alone A reduction in the extent of decidualization was observed in the implantation chambers. Except for the primary decidual zone, the decidua was less compact and the mesometrial decidualization was virtually missing. The tissue was composed of a mixture of either undifferentiated or less differentiated decidual cells that were separated by the extracellular matrix (Figure 8D). L-NAME and aminoguanidine alone The decidualization process was comparable to that of onapristone treatment although

6 6 6 6 6

0.16 0.22* 0.19 0.15 0.07*

Living fetuses (n)

Resorptions (n)

60 44 62 57 7* 0*

4 9 5 6 18* 0

less in extent (Figure 8A, B). Typical decidual cells dominated the antimesometrial decidua and substantial amounts of extracellular matrix were only observed in the mesometrial part of the implantation chamber. However, the embryos did not appear to be significantly altered, as also observed for those animals treated with onapristone alone. L-NAME plus onapristone This treatment produced pronounced changes in decidualization as well as in embryonic development. In the majority of implantation chambers examined no embryonic tissue could be found, except for some necrotic cells of embryonic origin. Decidualization was profoundly disturbed (Figure 8E, F). Aminoguanidine plus onapristone The effects were comparable to those exerted by L-NAME plus onapristone treatment, but slightly less pronounced (Figure 8G, H). As in the case of L-NAME plus onapristone there was a gross reduction in the extent of decidualization. The effects on pregnancy outcome on day 19 p.c. after treatment on days 5–8 p.c. (experiment 7) Treatment with each NOS inhibitor alone did not produce any significant effects on the pregnancy rate nor on the number of living fetuses. A slight reduction in the number of living fetuses was observed after onapristone alone. Interestingly, both L-NAME and onapristone significantly reduced fetal weights when administered alone. However, both combination regimens exerted significant inhibitory effects on pregnancy rates and the number of living fetuses. The effects of L-NAME 547

K.Chwalisz et al.

Figure 6. The effects of the NOS inhibitors L-NAME (LN) and aminoguanidine (A) in the presence or absence of low-dose onapristone (ONA) on pregnancy rates (upper panel; A, B) and implantation number (lower panel; B, C) established on day 9 p.c. after treatment at peri-implantation (days 5–8 p.c.). There were seven animals per group. Upper panel (A, B) presents the percentage of pregnant animals per experimental group. Black bars demonstrate the percentage of animals exhibiting macroscopically normal implantation sites. Striped bars present the percentage of animals showing macroscopically changed implantation sites (haemorrhagic and growth retarded). Lower panel (C, D) shows the total number of implantation sites per group. Black bars present the number of normal implantation sites, whereas striped bars demonstrate the number of macroscopically changed implantation sites. Asterisks* mark the statistically different values (P , 0.05) of normal implantation sites (black bars) from the respective control group.

plus onapristone were more pronounced than those of aminoguanidine plus onapristone (Table II). Discussion The results of our study in rats clearly demonstrate that NO inhibition with aminoguanidine and L-NAME substantially increased the inhibitory effects of the antiprogestin onapristone (i) on the establishment of pregnancy after treatment before implantation, (ii) on implantation and decidualization, and (iii) on the pregnancy outcome after treatment during the periimplantation phase. These effects were synergistic, since either treatment alone only marginally affected implantation and pregnancy outcome. Treatment with onapristone alone particularly before implantation induced macroscopic changes in implantation sites such as haemorrhage and size reduction during autopsy at day 9 p.c. (Figure 2). Interestingly, the majority of these embryos apparently recovered during the later stages of pregnancy except after combined treatments (Table I). Similar synergistic effects of NOS inhibitors and 548

Figure 7. The effect of the NOS inhibitors L-NAME (LN) and aminoguanidine (A) in the presence or absence of low-dose onapristone (ONA) on serum progesterone concentrations after treatment at peri-implantation (day 5–8 p.c.). Upper panel: L-NAME 6 onapristone treatment. Lower panel: aminoguanidine 6 treatment. Blood samples were collected on days 1, 5, 7 and 9 p.c. C 5 vehicle control. The results are expressed as means 6 SD. Asterisks* mark the statistically different values (P , 0.05) from respective control group on the respective day.

antiprogestins on pregnancy maintenance were observed in late pregnant rats (Yallampalli et al., 1996) and in guinea pigs (Chwalisz et al., 1996). Before implantation, each NOS inhibitor in combination with low-dose onapristone exerted a dramatic inhibitory effect on the establishment of pregnancy, aminoguanidine being more effective than L-NAME. The results of experiments employing oviductal flushing indicate that this treatment inhibited embryonic development thus causing the implantation defect. However, during the periimplantation phase, inhibition of the decidual reaction was the main treatment effect subsequently leading to the degeneration of the implantation sites. In both phases of pregnancy, neither treatment produced any significant changes in serum progesterone or oestradiol concentrations until day 7 p.c. indicating that the effects observed were direct, and not mediated by changes in ovarian hormone secretion. The results of our study are consistent with recent molecular studies on the expression of NOS isoforms in mice and rats. Western blot analysis of iNOS and ecNOS in conjunction with trace optical density reading revealed a significant elevation of iNOS and ecNOS expression in implantation sites versus intersite tissues on days 6, 7 and 8 of mouse pregnancy (Purcell et al., 1998). An increase in Ca21-independent and Ca21-

Nitric oxide and implantation

Figure 8A–H. Cross-sections of an implantation chamber on day 9 p. c. after peri-implantation (days 5–8 p.c.) treatment with L-NAME and onapristone each, and the combination of onapristone with either L-NAME or aminoguanidine. L-NAME alone (A): compared to controls, the decidualization process is reduced. The tissue shows, however, less compaction than after onapristone treatment alone. B: most of the decidual cells in the antimesometrial (Am) region seem to be well differentiated and intermingled with some spindle-shaped cells (arrows). Onapristone alone (C): this treatment reduced the extent of decidualization, the cells being loosely packed. D: higher magnification of the antimesometrial region shows small cells (arrow) separated by the extracellular matrix. L-NAME plus onapristone (E): a dramatic reduction in the extent of decidualization. Some necrotic embryo rudiments are detectable (arrow). Decidualization process seems to be incomplete. F: a higher magnification reveals spindle-like stromal cells with large intercellular spaces, some of the cells being necrotic (arrow). Aminoguanidine plus onapristone (G): an extensive retardation of decidualization, especially in the mesometrial (M) region, although the embryo is still present (arrow). H: at higher magnification, the decidual cells appear welldifferentiated and closely packed, virtually without any intracellular space. Slides A, C, E, G are at the same magnification, represented by 400 µm bar in G. Slides B, D, F, H are at the same magnification, represented by 30 µm bar in H.

dependent NOS activity was also described in the rat uterus on day 5 of pregnancy (Novaro et al., 1997). Our present study might also indicate that NO is not essential for the establishment of pregnancy, since NOS inhibitors did not prevent implantation nor terminate pregnancy when administered alone. The results of knock-out experiments in

mice indicate that the disruption of individual genes encoding iNOS (MacMicking et al., 1995), ecNOS (Huang et al., 1995), and bcNOS (Huang et al., 1993) did not significantly alter fertility or litter size. However, uterine NO production may be maintained by the remaining isoforms in animals lacking a single NOS isoform. The most likely explanation for the absence of any significant effects of NOS inhibitors on the establishment of pregnancy is the existence of redundant mediators such as prostaglandins (Kennedy, 1977, 1980) or histamine (Day and Hubbard, 1981) which may compensate for NO deficiency in both the preimplantation embryo and decidua. Another possible explanation is that the increased local NO production in both the preimplantation embryo and the decidua is not entirely inhibited by NOS inhibitors around the time of implantation, even when high doses are administered. Our findings from experiments employing oviduct flushing indicate that the inhibitory effects of aminoguanidine plus onapristone were due to the impairment of early embryo development, resulting in the inhibition of cleavage between the 8-cell and morula stages. Our data are consistent with the results of recent studies in mice indicating that preimplantation embryos express both iNOS and ecNOS, and can produce NO under in-vitro conditions (Gouge et al., 1998). In addition, this study shows that the non-specific NOS inhibitor L-NAME impairs embryogenesis. Overall, these data suggest that NO is required for normal embryonic development. Whether progesterone has a direct effect on preimplantation embryo development is still a matter of dispute. The growth rate and the implantation ability of preimplantation monkey embryos were not affected by treatment with RU 486 in vitro (Wolf et al., 1990). Similarly, blastocysts transferred from rats exposed to RU 486 displayed delayed implantation in the untreated recipients (Roblero and Croxatto, 1991). In addition, it was clearly demonstrated in rabbits that short-term treatment with various antiprogestins during the luteal phase leads to the transposition of the implantation window by antiprogestins acting primarily on the endometrium without any effects on the embryo (Beier et al., 1994). However, there are also studies suggesting that progesterone does play a role in preimplantation embryo development. Impairment of embryonic growth and development was reported after pre-ovulatory or early luteal phase treatment with RU 486 in rats and mice (Psychoyos and Prapas 1987; Roh et al., 1988; McRae, 1994), guinea pigs (Chwalisz et al., 1997), and monkeys (Ghosh et al., 1997). Implantation is subject to the interaction of the trophoblast cells with the decidua. The results of the present study strongly suggest that NO is involved in the decidual reaction since NOS inhibitors, in particular L-NAME, partially reduced the extent of decidualization even in the absence of onapristone after peri-implantation treatment. It is well established that cytokines (some of which are indispensable) play an important role during implantation (Chard, 1995; Tazuke and Giudice, 1996). IL-1 and its receptor are up-regulated in the human endometrium during the peri-implantation period (Simo´n et al., 1993), and are also present in preimplantation mouse embryos reaching maximal expression in hatching blastocysts (Cresol et al., 1997). The female osteopetrotic (op/op) mice, which 549

K.Chwalisz et al.

Table II. Outcome of pregnancy on day 19 p.c. after treatment with the NOS inhibitors NG-nitro-L-arginine methyl ester (L-NAME) and aminoguanidine (A) alone and in the presence and absence of low-dose onapristone (ONA) during the peri-implantation phase on days (days 5–8 p.c.; experiment 7). Asterisks mark the values which are significantly different from the vehicle control group (P , 0.05) Group

Treatment

Total dams (n)

Dams pregnant (n)

Implantation sites (n)

Fetal weight (g) (mean 6 SD)

Living fetuses (n)

1 2 3 4 5 6

Vehicle ONA L-NAME A L-NAME 1 ONA A 1 ONA

7 7 6 7 7 7

7 6 6 7 2* 4*

70 64 61 67 22* 54

62 52 54 58 19* 44

1.37 1.25 1.30 1.55 1.34 1.33

exhibit mutation of the colony-stimulating factor (CSF-1) gene, have markedly impaired fertility (Pollard et al., 1991). Also, female mice with a null mutation of the genes encoding leukaemia inhibitory factor (LIF) (Stewart et al., 1992), and interleukin-11 receptor alpha chain (IL-11Rα) (Robb et al., 1998) remain infertile because of implantation failure. Interestingly, both LIF and IL-11Rα knock out mice also exhibit a decidualization defect. iNOS can produce high levels of NO for prolonged periods of time (typically 4–6 h) in response to pro-inflammatory cytokines, including IL-1 and tumour necrosis factor α (TNFα) (Moncada and Higgs, 1993; Nathan and Xia-wen, 1994). Similarly, several types of CSF have been shown to up-regulate NO via iNOS induction in macrophages (Jorens et al., 1993). On the other hand, pro-inflammatory cytokines are known to up-regulate COX-2 at the inflammation site. Furthermore, it was recently found in various models of inflammation that NO is also a powerful inducer of COX-2 and elevates local PGE2 concentrations (Salvemini et al., 1995). The results of the present study suggest that NO acts as a local mediator of inflammatory changes during various stages of implantation. It may act on endometrial blood vessels and contribute to the vascular reaction. Since NO may directly regulate the activity of matrix metalloproteinases (MMP) (Trachtman et al., 1996), it may play a role in extracellular matrix changes occurring during the trophoblast invasion. Recent in-vitro (Behrendtsen et al., 1992) and in-vivo (Alexander et al., 1996) studies in mice show that embryo-activated MMPs and their inhibitors (TIMP-1, TIMP-2, and TIMP-3) play an important role during implantation. In addition, the treatment of early pregnant mice with a MMP inhibitor reduced the length and overall size of implantation sites and disturbed embryo orientation (Alexander et al., 1996). These effects are similar to those observed in the present study. To date an up-regulation of the IL-1 system during mice and human implantation has been linked to the attachment process in the early stages of implantation (Simo´n et al., 1993; Kruessel et al., 1997). Since IL-1β is produced by the embryo itself in humans (Simo´n et al., 1993; Kruessel et al., 1997), this cytokine may also act as an embryonic signal promoting implantation. It could in fact, via iNOS and COX-2 induction, initiate the vascular reaction, decidulization, and the remodelling of the extracellular matrix during trophoblast invasion. In all mammals during early pregnancy, an adequate uterine blood supply is essential for embryo development. In women, 550

6 6 6 6 6 6

0.09 0.01* 0.11* 0.19 0.01 0.10

Resorptions (n) 8 12 5 9 3 10

impaired blood flow to the uterus can jeopardize the establishment of pregnancy (Edwards, 1995). Furthermore, uterine blood flow, as measured by colour Doppler, was proposed as the physiological parameter to assess endometrial receptivity to blastocysts implantation following assisted reproductive treatments (Achiron et al., 1995; Friedler et al., 1996). In guinea pigs dilatation of uteroplacental arteries was observed when invading trophoblast cells coexpressing ecNOS and iNOS are present in the exravillous trophoblast (Nanaev et al., 1995) suggesting that NO mediates spiral arterial changes occurring during pregnancy. The present study demonstrates that both L-NAME and onapristone alone induced fetal growth retardation after treatment on days 6–8 p.c. (Table II). Interestingly, in rats during advanced pregnancy, the fetal growth retardation induced by L-NAME can be partially reversed by progesterone and some synthetic progestins, but not by oestrogen (Liao et al., 1997). The present results further support the concept of NO and progesterone being the key factors regulating the adaptation-related changes in uterine and placental blood vessels during pregnancy. The exact mechanism of the synergistic effects of onapristone and NOS inhibitors before implantation and in the periimplantation period remains, however, unclear. Hence, further studies are needed to determine whether these effects are simply due to a more powerful combined blockade of embryonic and decidual NO, or to more complex, perhaps independent effects of NOS inhibitors and onapristone. It is well established that the endometrium is generally very sensitive to antiprogestins that inhibit or modulate a number of progesterone-dependent genes even at very low doses (reviewed by Chwalisz et al., 1997). Hence, onapristone may inhibit other (redundant) pathways involved in embryogenesis, vascularization and decidualization, thereby contributing to the synergistic effect with NOS inhibitors. In conclusion, this study clearly demonstrates the synergistic inhibitory effects of NOS inhibitors and an antiprogestin in preventing pregnancy by inhibiting preimplantation embryo development and impairing decidualization. Furthermore, these findings suggest that NOS, in particular iNOS, may represent a new target for novel therapeutic agents capable of promoting or inhibiting implantation. Up-regulating uterine NO production with either the NO substrate L-arginine and NO donors alone or in combination with progesterone might have beneficial effects on pregnancy outcome during assisted conception. Such schedules may also have implications for the management

Nitric oxide and implantation

of early pregnancy disorders, including recurrent abortions. On the other hand, the effects of NOS inhibitors in combination with antiprogestins point to a novel method for controlling fertility, particularly by enhancing the efficacy of antiprogestins used for endometrial contraception, menstrual induction and post-coital contraception.

Acknowledgements We are grateful to Mrs B. Bragulla and Mrs P. Maulwurf for expert technical assistance in animal studies. We also thank Dr N. Benda for the statistical analyses, Mrs R. Ja¨ger for reading the manuscript, and T. Purcell for discussions on studies of NOS expression in rats and mice.

References Achiron, R., Levran, D., Sivan, E. et al. (1995) Endometrial blood flow response to hormone replacement therapy in women with premature ovarian failure: a transvaginal Doppler study. Fertil. Steril., 63, 5050–5554. Alexander, C.M., Hansell, E.J. Behrendtsen, O. et al. (1996) Expression and function of matrix metalloproteinases and their inhibitors at the maternalembryonic boundary during mouse embryo implantation. Development, 122, 1723–1736. Ali, M., Buhimschi, I., Chwalisz, K. and Garfield, R.E. (1997) Changes in expression of nitric oxide synthase isoforms in rat uterus and cervix during pregnancy and parturition. Mol. Hum. Reprod., 3, 995–1003 Behrendtsen, O., Alexander, C.M. and Werb, Z. (1992) Metalloproteinases mediate extracellular matrix degradation by cells from mouse blastocyst outgrowths. Development, 114, 447–456. Beier, H.M., Hegele-Hartung, C., Mootz, U. and Beier-Hellwig, K. (1994) Modification of endometrial cell biology using progesterone antagonists to manipulate the implantation window. Hum. Reprod., 9 (suppl. 1), 98–115. Buhimschi, I., Ali, M., Jain, V. et al. (1996) Differential regulation of nitric oxide in the uterus and cervix during pregnancy and labour. Hum. Reprod., 11, 1755–1766. Chard, T. (1995) Cytokines in implantation. Hum. Reprod. Update, 1, 385–396. Chwalisz, K., Buhimschi, I. and Garfield, R.E. (1996) Role of nitric oxide in obstetrics. Prenat. Neonat. Med., 1, 292–329. Chwalisz, K., Gemperlein, I., Puri, C.P. et al. (1997) Endometrial contraception with progesterone antagonists: an experimental approach. In Beier, H.M., Harper, M.J.K., Chwalisz, K. (eds) Ernst Schering Research Foundation Workshop 18. Springer Verlag, Berlin, pp 223–259. Day, S.K. and Hubbard, C.J. (1981) Role of histamine and cyclic nucleotides in implantation in the rabbit. Cell. Tiss. Res., 220, 549–554. Edwards, R.G. (1995) Clinical approaches to increasing uterine receptivity during human implantation. Hum. Reprod., 10 (suppl. 3), 60–67. Evans, C.A. and Kennedy, T.G. (1978) The importance of prostaglandin synthesis for the initiation of blastocyst implantation in the hamster. J. Reprod. Fertil., 54, 255–261. Evans, C.H., Stefanovic-Racic, M. and Lancaster, J. (1995) Nitric oxide and its role in orthopaedic disease. Clin. Orthop. Rel. Res., 312, 275–294. Friedler, S., Schenker, J.G., Herman, A. et al. (1996) The role of ultrasonography in the evaluation of endometrial receptivity following assisted reproductive treatments: a critical review. Hum. Reprod. Update, 2, 323–335. Ghosh, D., Kumar, P.G.L.L. and Sengupta, J. (1997) Early luteal phase administration of mifepristone inhibits preimplantation embryo development and viability in the rhesus monkeys. Hum. Reprod., 12, 575–582. Gouge, R.C., Marshburn, P., Gordon, B.E. et al. (1998) Nitric oxide as a regulator of embryonic development. Biol. Reprod., 58, 875–879. Hasan, S.H. and Caldwell, B.V. (1971) Radioimmunoassay of estradiol-17β using charcoal dextran suspension. In Kirkham, K.E. and Hunter W.M. Radioimmunoassay Methods. Churchill Livingstone, Edinburgh, pp. 334– 337. Huang, P.L., Dawson, T.D., Bredt, D.S. et al. (1993) Targeted disruption of the neuronal nitric oxide synthase gene. Cell, 75, 1273–1286. Huang, P.L., Huang, Z., Mashimo, H. et al. (1995) Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature, 377, 239–242. Ignarro, L.G. (1989) Biological actions and properties of endothelium-derived

nitric oxide formed and released from artery and vein. Circulation Res., 65, 1–21. Izumi, H., Yallampalli, C. and Garfield, R.E. (1993) Gestational changes in L-arginine-induced relaxation of pregnant rat and human myometrial smooth muscle. Am. J. Obstet. Gynecol., 169, 1327–1337. Jorens, P.G., van Overveld, F.J., Bult, H. et al. (1993) Muramyldipeptide and granulocyte-macrophage colony-stimulating factor enhance interferongamma-induced nitric oxide production by rat alveolar macrophages. Agents Actions, 38, 100–105. Kennedy, T.G. (1977) Evidence for a role for prostaglandins in the initiation of blastocyst implantation in the rat. Biol. Reprod., 16, 289–291. Kennedy, T.G. (1980) Timing of uterine sensitivity for the decidual cell reaction: role of prostaglandins. Biol. Reprod., 22, 519–525. Kruessel, J.K., Huang, H.Y., Wen, Y. et al. (1997) Different pattern of interleukin-1β-( IL-1β), interleukin-1 receptor antagonist-( IL-1ra) and interleukin-1 receptor type I-( IL-1R tI) mRNA-expression in single preimplantation mouse embryos at various developmental stages. J. Reprod. Immunol., 34, 103–120. Liao, Q.P., Buhimschi, I.A., Saade, G. et al. (1997) Regulation of vascular adaptation during pregnancy and post-partum: effects of nitric oxide inhibition and steroid hormones. Hum. Reprod., 12, 2777–2784. Lim, H., Paria, B.C., Das, S.K. et al. (1997) Multiple female reproductive failures in cyclooxygenase 2-deficient mice. Cell, 91, 197–208. Loke, Y.W. and King, A. (1995) Human implantation. Cell Biology and Immunology. Cambridge University Press, Cambridge. MacMicking, J.D., Nathan, C., Hom, G. et al. (1995) Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell, 81, 1273–1283. McRae, A.C. (1994) Effects of the anti-progestin and antiglucocorticoid steroid RU 486 on cell proliferation in oviductal embryos of lactating mice. J. Reprod. Fertil., 11, 307–313. Misko, T.P., Moore, W.M., Kasten, T.P. et al. (1993) Selective inhibition of the inducible nitric oxide synthase by aminoguanidine. Eur. J. Pharmacol., 233, 119–125. Moncada, S. and Higgs, E.A. (1993) The L-arginine-nitric oxide pathway. N. Engl. J. Med., 329, 2002–2012. Nanaev, A., Chwalisz, K., Frank, H-G. et al. (1995) Physiological dilation of uteroplacental arteries in the guinea pig depends upon nitric oxide synthase activity of extravillous trophoblast. Cell Tissue Res., 282, 407–421. Nathan, C. and Xie, Q.-N. (1994) Nitric oxide synthases: roles, tolls, and controls. Cell, 78, 915–918. Nayak, N.R., Ghosh, D., Lasley, L. et al. (1997) Anti-implantation activity of luteal phase mifepristone administration is not mimicked by prostaglandin synthesis inhibitor or prostaglandin analogue in the rhesus monkey. Contraception, 55, 103–114. Novaro, V., Gonza´lez, E., Jawerbaum, A. et al. (1997) Nitric oxide synthase regulation during embryonic implantation. Reprod. Fert. Dev., 9, 557–564. Parr, M.B. and Parr, E.L. (1989) The implantation reaction. In Wynn, R.M. and Jollie, W.P. (eds) Biology of the uterus, 2nd edn. Plenum, New York, pp. 233–267. Pollard, J.W., Hunt, J.S., Wiktor-Jedrzejak, W. et al. (1991) A pregnancy defect in the osteopetrotic (op/op) mouse demonstrates the requirement for CSF-1 in female fertility. Dev. Biol., 148, 273–283. Psychoyos, A. (1973) Hormonal control of ovoimplantation. Vitam. Horm., 31, 201–256. Psychoyos, A. and Prapas, I. (1987) Inhibition of egg development and implantation in rats after post-coital administration of the progesterone antagonist RU 486. J. Reprod. Fertil., 80, 487–491. Psychoyos, A., Nikas, G. and Gravanis, A. (1995) The role of prostaglandins in blastocyst implantation. Hum. Reprod. 10 (suppl. 2), 30–42. Purcell, T., Given, R., Chwalisz, K. et al. (1998) Inducible and endothelial nitric oxide synthases (iNOS and eNOS) exhibit continual increases in the uterus during the implantation and post-implantation periods. J. Soc. Gynecol. Invest., 5 (suppl.), abstract 314. Purcell, T.L., Buhimschi I.A., Given, R. et al. (1997) Inducible nitric oxide synthase (iNOS) is present in rat placenta at the fetal–maternal interface and decrease prior to labor. Mol. Hum. Reprod., 3, 485–491. Robb, L., Li, R., Hartley, L. et al. (1998). Infertility in female mice lacking the receptor for interleukin II is due to a defective uterine response to implantation. Nature Med., 4, 303–308. Roblero, L.S. and Croxatto, H.B. (1991) Effect of RU 486 on development and implantation of rat embryos. Mol. Reprod. Develop., 29, 342–346. Roh, S.I., Batten, B.E., Friedman, C.I. et al. (1988) The effects of progesterone

551

K.Chwalisz et al. antagonist RU 486 on mouse oocyte maturation, ovulation, fertilization and cleavage. Am J. Obstet. Gynecol., 159, 1584–1589. Rosselli, M. (1997) Nitric oxide and reproduction. Mol. Hum. Reprod., 3, 639–641. Salvemini, D., Misko, T.P., Masferrer, J. et al. (1995) Role of nitric oxide in the regulation of cyclooxygenase. In Moncada, S., Feelish, M. and Busse, R. (eds) Biology of Nitric Oxide 4. Enzymology, Biochemistry and Immunology. Portland Press, London. Simo´n, C., Piquette, G.N., Frances, A. and Polan, M.L. (1993) Localization of interleukin-1 receptor type I and interleukin-1β in human endometrium throughout the menstrual cycle. J. Clin. Endocrinol. Metab., 77, 549–555. Sladek, S.M., Regenstrin, A.C., Lykins, D. et al (1993) NO synthase activity in pregnant rabbit uterus decreases on the last day of pregnancy. Am. J. Obstet. Gynecol., 169, 1285–1291. Sladek, S.M., Kanbour-Shakir, A., Watkins, S. et al. (1998) Granulated metrial gland cells contain nitric oxide synthases during pregnancy in the rat. Placenta, 19, 55–65. Stewart, C.L., Kaspar, P., Brunet, L.J. et al. (1992) Blastocyst implantation depends on maternal expression of leukemia inhibitory factor. Nature, 359, 76–79. Tazuke, S.I. and Giudice, L.C. (1996) Growth factors and cytokines in endometrium, embryonic development, and meternal:embryonic interactions. Sem. Reprod. Endocrinol., 14, 231–245. Telfer, J.F., Irvine, G.A., Kohnen, G. et al. (1997) Expression of endothelial and inducible nitric oxide synthase in non-pregnant and decidalized human endometrium. Mol. Hum. Reprod., 3, 69–75. Trachtman, H., Futterweit, S., Garg, P. et al. (1996) Nitric oxide stimulates the activity of a 72-kDa neutral matrix metalloproteinase in cultured rat mesangial cells. Biochem. Biophys. Res. Comm., 218, 704–708. Tschugguel, W., Schneeberger, C., Unfriried, G. et al. (1998) Induction of inducible nitric oxide synthase expression in human secretory endometrium. Hum. Reprod., 13, 436–444. Yallampalli, C., Buhimschi, I., Chwalisz, K. et al. (1996) Preterm birth in rats produced by synergistic action of a nitric oxide inhibitor (NG-nitro-Larginine methyl ester) and an antiprogestin (onapristone). Am. J. Obstet. Gynecol, 175, 207–212. Wilcox, A.J., Weinberg, C.R., O’Connor, J.F. et al. (1988) Incidence of early pregnancy loss. N. Engl. J. Med., 319, 189–194. Wolf, J.P., Chilik, C.F., Dubois, C. et al. (1990) Tolerance of perinidatory primate embryo to RU 486 exposure in vitro and in vivo. Contraception, 41, 85–89. Received on July 23, 1998; accepted on November 5, 1998

552