In addition to being a chloride channel, CFTR, may function as a regulator of other channels, including aquaporin water channels (channels that allow rapid flow of water across epithelial membranes) and epithelial sodium channels. Enhancement of aquaporin channel activity [31
] and suppression of epithelial sodium channel-mediated absorptive activity [32
] that results in net increases in electrolyte and fluid secretion may also contribute to the pathogenesis of OHSS.
To test this hypothesis, we developed the OHSS rat model by using several combinations of gonadotropins as previously described [33
]. There was a marked increase in fluid accumulation in the peritoneal cavity and pleural cavity (Figure B and C). Therefore, the increased body WT in the hyperstimulated rats was significantly different from those in the controls (Figure A). Enhanced CFTR and AQP1 expression in body cavity peritoneal epithelial cells was observed in the OHSS rats (Figure D). Thus, up-regulated CFTR and AQP1 expression in peritoneal epithelial cells is correlated with the fluid accumulation observed in the body cavity of OHSS animals. It has been well documented that OHSS patients have high plasma estrogen levels [26
]. It is possible that hyper stimulation of ovaries leads to high production and release of estrogen, which is responsible for the upregulation of CFTR and AQP1 observed. This may explain why OHSS may not occur or may become severe despite high estrogen levels in some individuals who could have CFTR mutations and defective CFTR function.
Estrogen levels had been monitored in the induced OHSS rats as well as in the controls at both estrous and diestrous stages of the estrous cycle. The results showed that estrogen levels increased over 6-fold in the OHSS rats compared with that in the controls (Figure A), which is consistent with other OHSS animal models [33
]. The relationship between elevation of estrogen alone and OHSS had been investigated by developing of OHSS symptoms in ovariectomized rats. It confirmed that hyper stimulation did not produce OHSS symptoms in these ovariectomized rats unless treated with estrogen, but not progesterone, for 5 days (Figure B). Furthermore, progesterone suppressed estrogen induced OHSS symptom as well as CFTR and AQP1 expression upregulation (Figure B and C). Exogenous estrogen induced WT increase as well as CFTR and AQP1 expression upregulation in ovariectomized rats, similar to those in the OHSS rat model without ovariectomy (Figure B and C). It should be noted that although estrogen drastically increased CFTR and AQP1 expression, progesterone completely suppressed both of them in ovariectomized rats. This is consistent with previous results obtained in ovary-intact animals showing upregulation and downregulation of CFTR by estrogen and progesterone, respectively [17
]. OHSS symptoms and CFTR upregulation in peritoneal epithelial cells were also observed in ovary-intact rats treated with high levels of estrogen, but not progesterone (Figure B and C). The serum level of exogenously administrated estrogen was similar to the estrogen levels in OHSS rats (Figure B and C). These results suggest that the ovarian sex hormone, estrogen, but not progesterone, is responsible for the upregulation of CFTR and AQP1 leading to OHSS. These results support the clinical observation that OHSS rarely develops in individuals undergoing ovarian hyper stimulation in the absence of high estrogen levels [26
To confirm that up regulation of CFTR indeed leads to enhancement of its channel activity, from where excessive epithelial fluid secretion derives, we compared functional CFTR activity between freshly isolated peritoneal epithelial cells of normal and OHSS rats by the Isc measurement [34
]. To exclude a possible contribution of Na+
absorption to the current, we used amiloride, an epithelial sodium channel blocker. The results revealed that the forskolin-induced Isc from the peritoneal epithelial cells of OHSS rats was substantially increased compared with that of untreated controls at the diestrous or estrous stage of the sexual cycle (CFTR is known to be minimally and maximally expressed, respectively). The cAMP-dependent Isc could be blocked by a specific CFTR inhibitor-172 (Figure A-D), which has been demonstrated to inhibit CFTR-mediated, cholera toxin-induced fluid secretion [35
]. It confirmed that the observed cAMP-activated Isc reflects CFTR channel activity. The effect of exogenously administrated estrogen and progesterone for 5 d on peritoneal CFTR channel activity was also examined. Isc measurements showed that forskolin-induced or CFTR-mediated Isc in estrogen-treated peritoneum was significantly higher than that of progesterone-treated (Figure A and B). Progesterone suppressed estrogen induced CFTR-mediated Isc in peritoneum (Figure C). Taken together, it appears that up regulation of CFTR and AQP1 due to elevated estrogen may be responsible for the excessive fluid transport and accumulation in OHSS. Therefore, it may be possible to alleviate OHSS symptoms through suppressing CFTR or AQP1 expression or through suppressing CFTR or AQP1 function. To validate this hypothesis, we first used progesterone, as it was shown to suppress CFTR expression in previous studies [17
] and in the present study as well. Indeed, progesterone treatment of OHSS rats also eliminated peritoneal and pleural fluid accumulation and significantly reduced CFTR and AQP1 protein expression (Figure A-D). The progesterone selective progesterone antagonist, mifepristone, was found to suppress the action of progesterone (Figure A-D). Interestingly, progesterone has been used in a few clinical situations for luteal support during ART, in which it was found to significantly lower the incidence of OHSS in IVF cycles. However no explanation for such an effect was published [28
]. In summary, progesterone may suppress the expression of CFTR and AQP1 to alleviate the symptom of OHSS.
To confirm the coactions of CFTR and AQP1 in the pathogenesis of OHSS, the cAMP-dependent Isc was used to measure CFTR channel activity affected by AQP1. The results showed that the cAMP-dependent Isc could also be reduced by an AQP1 inhibitor, HgCl2
(Figure A-C), which has been demonstrated to inhibit AQP1-mediated water transport [36
]. This confirmed that AQP1 can affect CFTR channel activity. Furthermore, treatment of PMSG-/hCG-induced OHSS rats injected with 300 μl CFTR antiserum resulted in the elimination of the body WT increase about 3.23 g, the peritoneal fluid accumulation about 2.69 ml and the pleural fluid accumulation about 0.58 ml. Injection of 300 μl AQP1 antiserum resulted in the elimination of the body WT increase about 3.23 g, the peritoneal fluid accumulation about 2.56 ml and the pleural fluid accumulation about 0.64 ml (Figure D-F). The ability to alleviate OHSS symptoms through interfering with CFTR or AQP1 function confirmed a critical role of the coactions of CFTR and AQP1 in OHSS pathogenesis. It also suggested that these treatments may be a preventive measure for OHSS during ART.
In this study, we demonstrated that up regulation of both CFTR and AQP1 by either excessive estrogen or ovarian hyper stimulation leads to the development of OHSS symptoms in rodents. Considering the observed high levels of estrogen, exceeding normal physiological levels, in patients undergoing ovarian hyper stimulation [25
] and regulation of CFTR and AQP1 expression by ovarian hormones in humans [14
], the present findings in OHSS rodent models are of strong clinical relevance, suggesting the coactions of CFTR and AQP1 in the pathogenesis of OHSS. OHSS appears to be the result of abnormally up regulating CFTR as well as AQP1. This leads to excessive fluid shift and accumulation in the peritoneal cavity and pleural cavity, which can be life threatening. It is because that fluid transport throughout the body due to water effusion following ion transportation.