Administration of a gonadotropin-releasing hormone agonist affects corpus luteum vascular stability and development and induces luteal apoptosis in a rat model of ovarian hyperstimulation syndrome

Abstract

Ovarian hyperstimulation syndrome (OHSS) is a complication of ovarian stimulation with gonadotropins followed by the administration of human chorionic gonadotropin (hCG) to trigger the final steps of oocyte maturation. Gonadotropin-releasing hormone (GnRH) analogs are thought to be effective in preventing this complication and a clinical trial has found a lower incidence of OHSS in patients treated with these molecules. Our aim was to analyze the in vivo effect of a GnRH-I agonist on corpus luteum development and regression, ANGPT-1, ANGPT-2 and Tie-2 protein expression and luteal blood vessel stabilization, the expression of the steroidogenic acute regulatory protein (StAR) and the cytochrome P450 side-chain cleavage enzyme (P450scc) and cell proliferation, in ovaries from an OHSS rat model. To this end immature female Sprague-Dawley rats were hyperstimulated and treated with a GnRH-I agonist from the start of pregnant mare serum gonadotropin (PMSG) administration until the day of hCG injection for 5 consecutive days. Blood and tissue samples were collected 48 h after hCG injection. Vascular endothelial growth factor VEGF levels were evaluated in the peritoneal fluid by ELISA. Serum progesterone and estradiol were measured by RIA. Histological features of sectioned ovaries were assessed in hematoxylin and eosin (H&E) stained slides. Luteal blood vessel stability, cell proliferation and apoptosis were assessed by immunohistochemistry for SMCA, PCNA, and TUNEL, respectively. P450scc, StAR, FLK-1, ANGPT-1, ANGPT-2, Tie-2 and PCNA protein levels were evaluated by Western blot from dissected corpora lutea (CL). The treatment with the GnRH-I agonist significantly decreased serum progesterone and estradiol levels as well as P450scc and StAR protein expression in the untreated OHSS group. In addition, the agonist significantly decreased the number of CL in the OHSS group, as compared with the untreated OHSS group. In the OHSS group, the area of periendothelial cells in the CL was larger than that of the control group. However, the treatment with the GnRH-I agonist significantly reduced the area of periendothelial cells in the CL in the OHSS group. The luteal levels of ANGPT-1 and its receptor Tie-2 significantly increased in the OHSS group when compared with the control group. Conversely, the administration of the GnRH-I agonist significantly decreased the levels of these factors in the CL from the OHSS group, as compared with the untreated OHSS group. In addition, the treatment with the GnRH-I agonist reduced the diameter of CL and decreased CL cell proliferation as compared with that observed in the untreated OHSS group. Finally, the GnRH-I agonist increased apoptosis in the CL from the OHSS group. In conclusion, these results show that GnRH-I agonist exerts diverse actions on the CL from a rat OHSS model. The decrease in P450scc, StAR, ANGPT-1 and Tie-2 expression, blood vessel stability and luteal proliferation leads to CL regression in the ovaries from OHSS rats. Moreover, our results suggest that the downregulation of ANGPT-1 and its receptor is a possible mechanism whereby GnRH-I agonists could prevent early OHSS.

Introduction

The ovarian hyperstimulation syndrome (OHSS) is an iatrogenic complication associated with ovarian stimulation for the treatment of infertility (Budev et al., 2005, Rizk et al., 1997). OHSS occurs in 5–10% of patients undergoing ovulation induction therapy, and the severe form takes place in 0.5–5.0% (Aboulghar and Mansour, 2003, Delvigne and Rozenberg, 2002). It is widely accepted that the main clinical components of this syndrome are marked enlargement of the ovaries, which contain luteal cysts, and hemorrhagic cysts along with the shifting of fluid to the third space, including the peritoneal cavity (Golan et al., 1989). Complications include renal failure, hypovolemic shock, thromboembolic episodes, and adult respiratory distress syndrome. Although the pathophysiology of OHSS has not been completely elucidated, it seems likely that increased capillary permeability, triggered by the release of vasoactive substances from the hyperstimulated ovaries under human chorionic gonadotropin (hCG) stimulation, might play a key role in the onset of OHSS.

Vascular endothelial growth factor (VEGFA), also referred to as vascular permeability factor, is a potential mediator in the development of OHSS because of its vasoactive properties (Senger et al., 1983). It has been found that the follicular fluid (FF) from women with severe OHSS presents high concentrations of VEGFA that lead to increased endothelial cell permeability and can be attenuated by VEGF antibody (Levin et al., 1998). High levels of VEGFA have also been detectable in sera (Lee et al., 1997) and ascites from OHSS patients (Chen et al., 1999). Several authors have shown that the source of VEGFA in OHSS patients seems to be the hyperstimulated ovary, since the VEGFA concentration in FF is 100 times higher than that in the serum (Geva and Jaffe, 2000, Krasnow et al., 1996, Lee et al., 1997). Accordingly, ovarian VEGFA, which originates mainly from ovarian follicular cells, is considered to play a major role in inducing OHSS. While the VEGF is the main initiator of angiogenesis, the formation and differentiation of a structurally and functionally mature vascular network probably requires the coordinated action of various factors. These include angiopoietins ANGPT-1 and ANGPT-2, which act via the tyrosine kinase receptor Tie-2 (Maisonpierre et al., 1997). Unlike VEGF, ANGPT1 is unable to stimulate endothelial cell proliferation (Davis et al., 1996) but is required for the recruitment of periendothelial cells that lead to the maturation and stabilization of newly developed capillaries (Maisonpierre et al., 1997, Suri et al., 1996). The ANGPT/Tie-2 system is expressed in ovarian follicles and corpora lutea (CL) of rodents, bovines and primates (Hayashi et al., 2003, Hazzard et al., 1999, Maisonpierre et al., 1997, Sugino et al., 2005, Wulff et al., 2001). To produce progesterone, not only high vascularization but also stabilization of blood vessels in the CL is necessary to provide luteal cells with the large amounts of cholesterol needed for progesterone synthesis. Hence, the blood vessels in the CL need to stabilize and mature to serve as functional blood vessels (Jain and Booth, 2003). It is worth mentioning that the mature CL is highly vascularized, with 50–70% of the tissue comprised of microvascular pericytes (periendothelial cells) and endothelial cells (Redmer et al., 2001, Reynolds et al., 2000). Pericytes are mesodermally derived cells that wrap around the outside of capillaries and are separated from endothelial cells by a basement membrane. Pericytes are of the vascular smooth muscle cell (SMC) lineage (Challier et al., 1995). Accordingly, ANGPTs act on vascular endothelial cells and contribute to blood vessel stabilization through interaction with perivascular cells (pericytes) (Thurston et al., 2000). However, to date, no reports have addressed the actions of GnRH-I agonists in blood vessel stability, or in ANGPT-1, ANGPT-2 and Tie-2 expression in the CL of a rat model of OHSS.

Chronic administration of GnRH-I agonists leads to pituitary desensitization and inhibition of gonadotropin and sex steroid levels. Besides their effects on the pituitary–gonadal axis, GnRH-I agonists have the potential to modulate the ovarian function through a direct effect on the ovary. Numerous studies have documented a functional ovarian GnRH-I and GnRH receptor system in rat and human ovaries (Hsueh and Schaeffer, 1985, Kang et al., 2001, Uemura et al., 1994).

We have previously demonstrated that the treatment with a GnRH-I agonist in prepubertal rats causes an increase in ovarian follicle DNA apoptotic fragmentation by interfering with the FSH, cAMP and/or growth factors pathways (Andreu et al., 1998, Parborell et al., 2001). Moreover, several studies performed in rats have demonstrated the antigonadal effect of GnRH analogs administered either in vivo or in vitro (Hazum and Nimrod, 1982, Muttukrishna et al., 1996, Sridaran et al., 1999, Yang et al., 2003). In addition, we have shown an apoptotic follicular effect of GnRH-I agonist in antral follicles, which correlates with an imbalance in the ratio of antiapoptotic:proapoptotic proteins (BCL-xL/BCL-xS) (Parborell et al., 2002, Parborell et al., 2005). These results indicate that GnRH-I interferes with follicular development through an increase in apoptotic events mediated by an imbalance among the BCL-2 family members. Moreover, Harwood et al. (1980) have reported that GnRH-I agonists affect the luteinization process of superovulated rats.

The effect of GnRH-I agonists on ovarian angiogenesis and their regulation is still largely unknown. By using a rat model of ovarian hyperstimulation syndrome (OHSS), Kitajima et al. have shown that the treatment with a GnRH-I agonist reduces ovarian vascular permeability and that this effect seems to be mediated by an increase in the expression of the tight junction protein claudin (Kitajima et al., 2004, Kitajima et al., 2006). Moreover, we have previously reported that a decline of the levels of VEGFA and its receptor Flk-1/KDR and ANGPT-1 by effect of GnRH-I agonist administration in prepubertal pregnant mare serum gonadotropin (PMSG)-treated rats is one of the mechanisms involved in the apoptosis of ovarian cells, and that this suggests an intraovarian role of an endogenous GnRH-like peptide in follicular development induced by gonadotropins (Parborell et al., 2008).

Presently, there is no cure for OHSS, and only preventive strategies are available. GnRH antagonists and/or GnRH agonists are now used for the treatment of OHSS (Dal Prato and Borini, 2005, Humaidan, 2009, Ragni et al., 2005). However, so far, no report has studied the in vivo effect of GnRH-I agonists on luteal development, luteal blood vessel stabilization, apoptosis and proliferation in the ovary from a rat model of OHSS. Therefore, the main purpose of this study was to investigate the in vivo effects of a GnRH-I agonist on steroidogenesis, peritoneal VEGF concentrations, FLK-1, ANGPT-1, ANGPT-2 and Tie-2 protein levels, luteal vascular stability, luteal development, apoptosis and proliferation in ovaries from an immature OHSS rat model stimulated by PMSG and hCG.