A unique human chorionic gonadotropin antagonist suppresses ovarian hyperstimulation syndrome in rats

Abstract

Ovarian hyperstimulation syndrome (OHSS) is a complication of in vitro fertilization associated with physiological changes after hCG administration to induce final oocyte maturation. It presents as widespread increases in vascular permeability and, in rare cases, results in cycle cancellation, multi-organ dysfunction, and pregnancy termination. These physiological changes are due primarily to activation of the vascular endothelial growth factor (VEGF) system in response to exogenous human chorionic gonadotropin (hCG). An hCG antagonist (hCG-Ant) could attenuate these effects by competitively binding to the LH/CG receptor, thereby blocking LH activity in vivo. We expressed a form of hCG that lacks three of its four N-linked glycosylation sites and tested its efficacy as an antagonist. The hCG-Ant binds the LH receptor with an affinity similar to native hCG and inhibits cAMP response in vitro. In a rat model for ovarian stimulation, hCG-Ant dramatically reduces ovulation and steroid hormone production. In a well-established rat OHSS model, vascular permeability and vascular endothelial growth factor (VEGF) expression are dramatically reduced after hCG-Ant treatment. Finally, hCG-Ant does not appear to alter blastocyst development when given after hCG in mice. These studies demonstrate that removing specific glycosylation sites on native hCG can produce an hCG-Ant that is capable of binding without activating the LH receptor and blocking the actions of hCG. Thus hCG-Ant will be investigated as a potential therapy for OHSS.

Figure 1

Gel electrophoresis of the hCG analogs. Differences in molecular mass reflect alterations in carbohydrate content. Apparent molecular mass in kilodaltons is indicated in the left margin. A, Reduced silver stain 10% SDS-PAGE; B, reduced Western blot of 10% SDS-PAGE visualized with an anti-hCG-α antibody R116.

Figure 2

In vitro bioassays of hCG analogs. The x-axis correlates to increasing concentrations of each protein. Each purified protein was evaluated for its ability to bind and activate the LH receptor. A, u-hCG and y-hCG-Ant competitive binding to the LH receptor; B, adenylate cyclase activation by the hCG analogs was determined by using CHO cells expressing the LH/CG receptor. Stimulation of cells was conducted at 37 C for 15 min, and total cAMP was measured by RIA.

Figure 3

Vascular permeability: effects of OHSS treatment on vascular permeability in hyperstimulated rats. Control animals received PMSG and saline. Hyperstimulated animals received PMSG and hCG to induce OHSS. In addition, the PMSG-treated animals received hCG-Ant alone or in combination with hCG (24 h later) to assess improvement in vascular permeability. The following dosages were administered: 0.1 ml saline (control) or 30 IU hCG, hCG-Ant, or hCG then hCG-Ant. Vascular permeability, assessed as micrograms of extravasated EB dye per gram animal weight, was measured 48 h after hCG injection. Note that after treatment with hCG-Ant, a significant decrease in vascular permeability was observed. Bars depict sem in each group. *, P < 0.001, comparison of OHSS treatment groups (hCG-Ant and hCG then hCG-Ant) to control group.

Figure 4

Western blot analysis of ovarian VEGF and VEGFR2 after hyperstimulation with PMSG: effect of hCG-Ant treatment alone or after hCG on VEGF and VEGFR2 protein expression in OHSS rats. Protein bands correspond to apparent mass in kilodaltons as indicated in the right margin for each Western blot. Black arrows indicate the expected band size corresponding to the mature, functionally active forms of the VEGF (26-kDa) and VEGFR2 (220-kDa) proteins. Please note the absence of VEGF and VEGFR2 after treatment with hCG-Ant alone or after hCG. A, Anti-VEGF antibody, 24 h after treatment: hCG (presence of 26-kDa band) and hCG-Ant (absence of 26-kDa band); B, anti-VEGF antibody, 48 h after treatment: hCG (presence of 26-kDa band) and hCG then hCG-Ant (absence of 26-kDa band); C, anti-VEGFR2 antibody, 24 h after treatment: hCG (220-kDa band) and hCG-Ant (negligible 220-kDa band).

Figure 5

Oocytes retrieved and embryos produced in C57BL/6J mice after various treatments: hCG, hCG-Ant, and hCG then hCG then hCG-Ant. A, Graph of mean number of oocytes retrieved per C57BL/6J female mouse. *, P < 0.01 compared with hCG. No symbol above the bar indicates results were not statistically different from hCG. Error bars represent ±sd. B, Graph of the mean number of embryos per developmental stage per 40 ovulated oocytes in C57BL/6J mice. *, P < 0.01 for OHSS treatment group compared with hCG and hCG-Ant. The hCG then hCG-Ant group produced the most numbers of viable blastocysts, whereas giving hCG-Ant alone dramatically reduced blastocyst formation.