- Email: [email protected]
Ovarian stimulation leads to a severe implantation defect in mice
Reproductive BioMedicine Online (2013) 27, 172– 175
www.sciencedirect.com www.rbmonline.com
SHORT COMMUNICATION
Ovarian stimulation leads to a severe implantation defect in mice Shaorong Deng
a,1
, Jing Xu
a,1
, Jianwu Zeng a, Linli Hu b, Yunxia Wu
a,*
a Department of Chinese Herbs, Tongji School of Pharmacy, Wuhan 430030, China; b Center of Reproductive Medicine, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
* Corresponding author. E-mail address: [email protected] (Y Wu). 1 These authors contributed equally to this paper.
Yunxia Wu obtained her MD in obstetrics and gynaecology in 2004 from the Tongji Medical College of Huazhong University of Science and Technology, Wuhan, China. She held post-doctoral positions in the Harvard Medical School. She is now an associate professor in the reproduction and infertility pharmacology centre of Tongji Medical College. Her main research interest is reproductive endocrinology with ovarian stimulation.
Abstract The aim of the present study was to evaluate whether ovarian stimulation could induce embryo implantation dysfunction
in mice and to explore the possible mechanisms involved. Ovarian stimulation was performed with intraperitoneal injections of 0, 2.5, 5.0, 7.5 and 10.0 IU pregnant mare serum gonadotrophin followed by the same dose of human chorionic gonadotrophin 48 h later. A dose-dependent implantation defect in stimulated mice was demonstrated, which can be mainly explained by premature luteolysis and secondary endometrial changes induced by an imbalance in oestradiol and progesterone. RBMOnline ª 2013, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: corpus luteum, decidua, embryo, implantation dysfunction, mouse, ovarian stimulation
Introduction In IVF–embryo transfer, gonadotrophins are commonly used for ovarian stimulation to increase the number of oocytes that can be retrieved for IVF, thus improving the overall chance for successful fertilization and pregnancy. Regardless of the important role of ovarian stimulation in IVF, it may itself have detrimental effects on implantation and pregnancy outcome. So this study sought to explore whether ovarian stimulation could induce embryo implantation dysfunction and to clarify the important factors responsible
for implantation dysfunction resulting from ovarian stimulation in mice.
Materials and methods The primary objective of the first experiment was to determine which doses of PMSG and HCG induced embryo implantation dysfunction. Female mice showing oestrus phase underwent ovarian stimulation with an intraperitoneal injection of 0, 2.5, 5, 7.5 or 10.0 IU pregnant mare serum gonadotrophin (PMSG; Ningbo Second Hormone Factory,
1472-6483/$ - see front matter ª 2013, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.rbmo.2013.03.018
Ovarian stimulation leads to a severe implantation defect in mice Table 1
173
Anti-implantation activity of different doses of PMSG/HCG in mice calculated on pregnancy day 8.
PMSG/HCG (IU)
No. of dams treated
No. of pregnant dams
Pregnancy rate (%)
Implanted embryos (mean ± SEM, range)
0 2.5 5 7.5 10
8 7 10 10 10
7 4 6 3 0
87.5 57.1 60.0 30.0a 0b
12.29 ± 2.49 (8–15) 15.75 ± 1.25 (14–17) 16.67 ± 10.65 (10–30) 16.00 ± 19.79 (2–30) 0
HCG = human chorionic gonadotrophin; PMSG = pregnant mare serum gonadotrophin. a P < 0.05 compared with naturally pregnant mice. bP < 0.001 compared with naturally pregnant mice.
Figure 1 External morphology of naturally pregnant mice (NP) (A–C) and stimulated mice (10.0 IU pregnant mare serum gonadotrophin and human chorionic gonadotrophin; D–F) implantation sites during peri- and post-implantation as visualized with a dissecting light microscope. Note the peri-implantation sites (A, D; white arrows), viable post-implantation sites in naturally pregnant mice (B, C; V and black arrows) and the resorbed post-implantation sites with relatively smaller size in stimulated mice (E, F; R and black arrows). C and F show at higher magnification the typically observed differences in the external appearance of viable implantation sites and resorbed implantation sites on pregnancy day (PD) 8. (G, H) Mean percentage ± SEM peri-implantation embryo loss on PD5 (G) and post-implantation resorption on PD8 (H). Bars = 5 mm.
174
S Deng et al.
China) at 18.00 hours followed by an additional injection of an equal dose of human chorionic gonadotrophin (HCG; Lizhu Pharmaceutical Factory, China) 48 h later. Presence of vaginal plug was considered as evidence of mating and the day was designated as pregnancy day (PD) 1. In the experiment designed to explore the possible mechanism of embryo implantation dysfunction, intraperitoneal injections of 10.0 IU PMSG/HCG were chosen. Under a dissecting microscope, the numbers of implantation sites (viable and resorbed) were counted on PD5- and PD8-pregnant mice uteri and the percentage of implantation embryo loss was calculated (Bindali and Kaliwal, 2002). From PD4 to PD8, blood samples of mice from each group were analysed for oestradiol and progesterone by radioimmunoassay, and the decidual weights of uteri were calculated. Then, uteri and ovaries of pregnant mice in each group were fixed in 4% formalin for 12 h and then embedded in paraffin for haematoxylin eosin staining and immunohistochemistry. The maximal diameter of the corpus luteum in the cortex of each ovary was measured (Goto et al., 1999). The expression of endometrial progesterone receptors was quantified using the H-scoring method (Sharpe-Timms et al., 2000). Analysis of variance one-way test was utilized to compare the nonparametric data. A chi-squared test was used to estimate the rate. The protocol of the study was approved by the Ethics Committee, Tongji Medical School of Huazhong University of Science and Technology.
Table 2
Results The pregnancy rates of the 2.5–10.0 IU PMSG/HCG groups decreased gradually, and complete inhibition was achieved in the 10.0 IU PMSG/HCG group (P < 0.001) (Table 1). Preimplantation loss in naturally pregnant and stimulated mice on PD5 were 22% and 54% (P = 0.002), and post-implantation loss were 18% and 100% (P < 0.001; Figure 1). As shown in Table 2, it is noteworthy that the maximum diameter of the corpus luteum decreased in stimulated ovaries and increased in normal ovaries during the implantation window. Additionally, the decidual weights of uteri decreased significantly (P = 0.011, 0.011 and <0.001 on PD5, PD7 and PD8, respectively). Oestradiol concentrations in stimulated mice remained marginally higher compared with the naturally pregnant mice, and serum progesterone concentrations were significantly higher in stimulated mice on PD5 and PD6 (P = 0.024 and 0.017, respectively). Further, progesterone receptors were significantly decreased from PD4 to PD6 compared with naturally pregnant mice (P = 0.024, 0.009 and <0.001 on PD4, PD5 and PD6, respectively).
Discussion This study revealed a dose-dependent defect in implantation after ovarian stimulation by PMSG and HCG. No
Summary of secondary findings on pregnancy days 4–8 in stimulated versus naturally pregnant mice (n = 8). Maximum diameter of corpus luteum (mm)
Decidual weight of uterus (mg)
Serum progesterone (ng/ml)
Endometrial progesterone receptor H-score
NP OS
1.23 ± 0.03 1.18 ± 0.05
121.0 ± 21.0 93.6 ± 15.6
7719.21 ± 4466.94 9062.86 ± 3890.96
193.04 ± 142.13 276.22 ± 327.16
1.98 ± 0.54 1.31 ± 0.12a
NP OS
1.25 ± 0.31 0.95 ± 0.04
139.3 ± 14.4 94.5 ± 13.3a
8353.54 ± 5788.91 12045.49 ± 4928.86
116.11 ± 55.49 551.44 ± 482.17a
2.59 ± 0.37 1.96 ± 0.19b
NP OS
1.36 ± 0.15 0.97 ± 0.17a
144.8 ± 20.5 98.8 ± 19.5
7420.63 ± 8359.38 8747.32 ± 5844.26
57.54 ± 67.11 398.17 ± 397.74a
2.59 ± 0.08 1.79 ± 0.22c
NP OS
1.31 ± 0.09 1.01 ± 0.15
178.6 ± 28.4 105.7 ± 13.5a
4458.82 ± 3323.96 6320.17 ± 3401.83
135.76 ± 153.97 191.82 ± 305.25
– –
NP OS
1.41 ± 0.19 0.93 ± 0.09c
210.6 ± 24.2 133.9 ± 27.1c
4894.52 ± 1668.77 6499.98 ± 3635.80
97.37 ± 79.03 229.39 ± 176.96
– –
Pregnancy day
Serum oestradiol (pg ml)
4
5
6
7
8
NP = naturally pregnant; OS = ovarian stimulation (10.0 IU pregnant mare serum gonadotrophin/human chorionic gonadotrophin). a P < 0.05 compared with naturally pregnant mice. b P < 0.01 compared with naturally pregnant mice. c P < 0.001 compared with naturally pregnant mice.
Ovarian stimulation leads to a severe implantation defect in mice implantations were observed with 10.0 IU PMSG/HCG injections. Other findings, including a progressive decline in the diameter of the corpus luteum, increase in vacuolization and invasion of the corpus luteum by stromal elements (data not shown) were similar to the morphological characteristics of structural luteolysis (Malven and Sawyer, 1966). This demise of the corpus luteum can lead to a relative decline in progesterone concentrations during the implantation window, which was observed in present study. An impaired endometrial decidualization and progesterone-receptor expression were involved in the consequent impairment of endometrial changes. The abnormal decidualization, including a loosely arranged stromal layer, enhanced inflammatory cell infiltration and hyperplastic luminal and hypertrophic glandular epithelia, is similar to that observed in progesterone-receptor knockout mice, and oestradiol is the primary proliferative stimulus of uterine epithelium and progesterone acting via its receptor plays a central role in regulating decidualization (Lydon et al., 1995). Additionally, progesterone is sufficient to sustain decidualization in the absence of exogenous oestradiol (Das et al., 2009); however, impaired progesterone-receptor endometrial expression in stimulated cycles was observed in the present study. All these observations indicate that the progesterone-controlled decidualization and expression of the progesterone receptor, which are both essential for implantation, were inhibited due to a disturbed balance between oestradiol and progesterone. In summary, this study revealed a dose-dependent implantation defect and embryo implantation dysfunction induced by ovarian stimulation with PMSG/HCG injection in mice, which can be mainly explained by premature luteolysis and secondary endometrial changes, including impaired endometrial decidualization and progesterone-receptor expression, induced by an imbalance between oestradiol and progesterone. This may provide valuable information for IVF treatment, the aim of which should be appropriate and personalized ovarian stimulation protocols for women in order to avoid iatrogenic implantation dysfunction.
175
Acknowledgements The authors are grateful to the National Natural Science Foundation of China (no. 30801538), the Hubei Population and Family Planning Commission (no. JS2012003) and the Central College Basic Scientific Research Business Special Fund (no. 2012QN011) for financial assistance.
References Bindali, B.B., Kaliwal, B.B., 2002. Anti-implantation effect of a carbamate fungicide mancozeb in albino mice. Ind. Health 40, 191–197. Das, A., Mantena, S.R., Kannan, A., Evans, D.B., Bagchi, M.K., Bagchi, I.C., 2009. De novo synthesis of estrogen in pregnant uterus is critical for stromal decidualization and angiogenesis. Proc. Natl. Acad. Sci. U. S. A. 106, 12542–12547. Goto, T., Endo, T., Henmi, H., Kitajima, Y., Kiya, T., Nishikawa, A., Manase, K., Sato, H., Kudo, R., 1999. Gonadotropin-releasing hormone agonist has the ability to induce increased matrix metalloproteinase (MMP)-2 and membrane type 1-MMP expression in corpora lutea, and structural luteolysis in rats. J. Endocrinol. 161, 393–402. Lydon, J.P., DeMayo, F.J., Funk, C.R., Mani, S.K., Hughes, A.R., Montgomery Jr., C.A., Shyamala, G., Conneely, O.M., O’Malley, B.W., 1995. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev. 9, 2266–2278. Malven, P.V., Sawyer, C.H., 1966. A luteolytic action of prolactin in hypophysectomized rats. Endocrinology 79, 268–274. Sharpe-Timms, K.L., Ricke, E.A., Piva, M., Horowitz, G.M., 2000. Differential expression and localization of de novo synthesized endometriotic haptoglobin in endometrium and endometriotic lesions. Hum. Reprod. 15, 2180–2185. Declaration: The authors report no financial or commercial conflicts of interest. Received 26 July 2012; refereed 28 March 2013; accepted 28 March 2013.
www.sciencedirect.com www.rbmonline.com
SHORT COMMUNICATION
Ovarian stimulation leads to a severe implantation defect in mice Shaorong Deng
a,1
, Jing Xu
a,1
, Jianwu Zeng a, Linli Hu b, Yunxia Wu
a,*
a Department of Chinese Herbs, Tongji School of Pharmacy, Wuhan 430030, China; b Center of Reproductive Medicine, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
* Corresponding author. E-mail address: [email protected] (Y Wu). 1 These authors contributed equally to this paper.
Yunxia Wu obtained her MD in obstetrics and gynaecology in 2004 from the Tongji Medical College of Huazhong University of Science and Technology, Wuhan, China. She held post-doctoral positions in the Harvard Medical School. She is now an associate professor in the reproduction and infertility pharmacology centre of Tongji Medical College. Her main research interest is reproductive endocrinology with ovarian stimulation.
Abstract The aim of the present study was to evaluate whether ovarian stimulation could induce embryo implantation dysfunction
in mice and to explore the possible mechanisms involved. Ovarian stimulation was performed with intraperitoneal injections of 0, 2.5, 5.0, 7.5 and 10.0 IU pregnant mare serum gonadotrophin followed by the same dose of human chorionic gonadotrophin 48 h later. A dose-dependent implantation defect in stimulated mice was demonstrated, which can be mainly explained by premature luteolysis and secondary endometrial changes induced by an imbalance in oestradiol and progesterone. RBMOnline ª 2013, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: corpus luteum, decidua, embryo, implantation dysfunction, mouse, ovarian stimulation
Introduction In IVF–embryo transfer, gonadotrophins are commonly used for ovarian stimulation to increase the number of oocytes that can be retrieved for IVF, thus improving the overall chance for successful fertilization and pregnancy. Regardless of the important role of ovarian stimulation in IVF, it may itself have detrimental effects on implantation and pregnancy outcome. So this study sought to explore whether ovarian stimulation could induce embryo implantation dysfunction and to clarify the important factors responsible
for implantation dysfunction resulting from ovarian stimulation in mice.
Materials and methods The primary objective of the first experiment was to determine which doses of PMSG and HCG induced embryo implantation dysfunction. Female mice showing oestrus phase underwent ovarian stimulation with an intraperitoneal injection of 0, 2.5, 5, 7.5 or 10.0 IU pregnant mare serum gonadotrophin (PMSG; Ningbo Second Hormone Factory,
1472-6483/$ - see front matter ª 2013, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.rbmo.2013.03.018
Ovarian stimulation leads to a severe implantation defect in mice Table 1
173
Anti-implantation activity of different doses of PMSG/HCG in mice calculated on pregnancy day 8.
PMSG/HCG (IU)
No. of dams treated
No. of pregnant dams
Pregnancy rate (%)
Implanted embryos (mean ± SEM, range)
0 2.5 5 7.5 10
8 7 10 10 10
7 4 6 3 0
87.5 57.1 60.0 30.0a 0b
12.29 ± 2.49 (8–15) 15.75 ± 1.25 (14–17) 16.67 ± 10.65 (10–30) 16.00 ± 19.79 (2–30) 0
HCG = human chorionic gonadotrophin; PMSG = pregnant mare serum gonadotrophin. a P < 0.05 compared with naturally pregnant mice. bP < 0.001 compared with naturally pregnant mice.
Figure 1 External morphology of naturally pregnant mice (NP) (A–C) and stimulated mice (10.0 IU pregnant mare serum gonadotrophin and human chorionic gonadotrophin; D–F) implantation sites during peri- and post-implantation as visualized with a dissecting light microscope. Note the peri-implantation sites (A, D; white arrows), viable post-implantation sites in naturally pregnant mice (B, C; V and black arrows) and the resorbed post-implantation sites with relatively smaller size in stimulated mice (E, F; R and black arrows). C and F show at higher magnification the typically observed differences in the external appearance of viable implantation sites and resorbed implantation sites on pregnancy day (PD) 8. (G, H) Mean percentage ± SEM peri-implantation embryo loss on PD5 (G) and post-implantation resorption on PD8 (H). Bars = 5 mm.
174
S Deng et al.
China) at 18.00 hours followed by an additional injection of an equal dose of human chorionic gonadotrophin (HCG; Lizhu Pharmaceutical Factory, China) 48 h later. Presence of vaginal plug was considered as evidence of mating and the day was designated as pregnancy day (PD) 1. In the experiment designed to explore the possible mechanism of embryo implantation dysfunction, intraperitoneal injections of 10.0 IU PMSG/HCG were chosen. Under a dissecting microscope, the numbers of implantation sites (viable and resorbed) were counted on PD5- and PD8-pregnant mice uteri and the percentage of implantation embryo loss was calculated (Bindali and Kaliwal, 2002). From PD4 to PD8, blood samples of mice from each group were analysed for oestradiol and progesterone by radioimmunoassay, and the decidual weights of uteri were calculated. Then, uteri and ovaries of pregnant mice in each group were fixed in 4% formalin for 12 h and then embedded in paraffin for haematoxylin eosin staining and immunohistochemistry. The maximal diameter of the corpus luteum in the cortex of each ovary was measured (Goto et al., 1999). The expression of endometrial progesterone receptors was quantified using the H-scoring method (Sharpe-Timms et al., 2000). Analysis of variance one-way test was utilized to compare the nonparametric data. A chi-squared test was used to estimate the rate. The protocol of the study was approved by the Ethics Committee, Tongji Medical School of Huazhong University of Science and Technology.
Table 2
Results The pregnancy rates of the 2.5–10.0 IU PMSG/HCG groups decreased gradually, and complete inhibition was achieved in the 10.0 IU PMSG/HCG group (P < 0.001) (Table 1). Preimplantation loss in naturally pregnant and stimulated mice on PD5 were 22% and 54% (P = 0.002), and post-implantation loss were 18% and 100% (P < 0.001; Figure 1). As shown in Table 2, it is noteworthy that the maximum diameter of the corpus luteum decreased in stimulated ovaries and increased in normal ovaries during the implantation window. Additionally, the decidual weights of uteri decreased significantly (P = 0.011, 0.011 and <0.001 on PD5, PD7 and PD8, respectively). Oestradiol concentrations in stimulated mice remained marginally higher compared with the naturally pregnant mice, and serum progesterone concentrations were significantly higher in stimulated mice on PD5 and PD6 (P = 0.024 and 0.017, respectively). Further, progesterone receptors were significantly decreased from PD4 to PD6 compared with naturally pregnant mice (P = 0.024, 0.009 and <0.001 on PD4, PD5 and PD6, respectively).
Discussion This study revealed a dose-dependent defect in implantation after ovarian stimulation by PMSG and HCG. No
Summary of secondary findings on pregnancy days 4–8 in stimulated versus naturally pregnant mice (n = 8). Maximum diameter of corpus luteum (mm)
Decidual weight of uterus (mg)
Serum progesterone (ng/ml)
Endometrial progesterone receptor H-score
NP OS
1.23 ± 0.03 1.18 ± 0.05
121.0 ± 21.0 93.6 ± 15.6
7719.21 ± 4466.94 9062.86 ± 3890.96
193.04 ± 142.13 276.22 ± 327.16
1.98 ± 0.54 1.31 ± 0.12a
NP OS
1.25 ± 0.31 0.95 ± 0.04
139.3 ± 14.4 94.5 ± 13.3a
8353.54 ± 5788.91 12045.49 ± 4928.86
116.11 ± 55.49 551.44 ± 482.17a
2.59 ± 0.37 1.96 ± 0.19b
NP OS
1.36 ± 0.15 0.97 ± 0.17a
144.8 ± 20.5 98.8 ± 19.5
7420.63 ± 8359.38 8747.32 ± 5844.26
57.54 ± 67.11 398.17 ± 397.74a
2.59 ± 0.08 1.79 ± 0.22c
NP OS
1.31 ± 0.09 1.01 ± 0.15
178.6 ± 28.4 105.7 ± 13.5a
4458.82 ± 3323.96 6320.17 ± 3401.83
135.76 ± 153.97 191.82 ± 305.25
– –
NP OS
1.41 ± 0.19 0.93 ± 0.09c
210.6 ± 24.2 133.9 ± 27.1c
4894.52 ± 1668.77 6499.98 ± 3635.80
97.37 ± 79.03 229.39 ± 176.96
– –
Pregnancy day
Serum oestradiol (pg ml)
4
5
6
7
8
NP = naturally pregnant; OS = ovarian stimulation (10.0 IU pregnant mare serum gonadotrophin/human chorionic gonadotrophin). a P < 0.05 compared with naturally pregnant mice. b P < 0.01 compared with naturally pregnant mice. c P < 0.001 compared with naturally pregnant mice.
Ovarian stimulation leads to a severe implantation defect in mice implantations were observed with 10.0 IU PMSG/HCG injections. Other findings, including a progressive decline in the diameter of the corpus luteum, increase in vacuolization and invasion of the corpus luteum by stromal elements (data not shown) were similar to the morphological characteristics of structural luteolysis (Malven and Sawyer, 1966). This demise of the corpus luteum can lead to a relative decline in progesterone concentrations during the implantation window, which was observed in present study. An impaired endometrial decidualization and progesterone-receptor expression were involved in the consequent impairment of endometrial changes. The abnormal decidualization, including a loosely arranged stromal layer, enhanced inflammatory cell infiltration and hyperplastic luminal and hypertrophic glandular epithelia, is similar to that observed in progesterone-receptor knockout mice, and oestradiol is the primary proliferative stimulus of uterine epithelium and progesterone acting via its receptor plays a central role in regulating decidualization (Lydon et al., 1995). Additionally, progesterone is sufficient to sustain decidualization in the absence of exogenous oestradiol (Das et al., 2009); however, impaired progesterone-receptor endometrial expression in stimulated cycles was observed in the present study. All these observations indicate that the progesterone-controlled decidualization and expression of the progesterone receptor, which are both essential for implantation, were inhibited due to a disturbed balance between oestradiol and progesterone. In summary, this study revealed a dose-dependent implantation defect and embryo implantation dysfunction induced by ovarian stimulation with PMSG/HCG injection in mice, which can be mainly explained by premature luteolysis and secondary endometrial changes, including impaired endometrial decidualization and progesterone-receptor expression, induced by an imbalance between oestradiol and progesterone. This may provide valuable information for IVF treatment, the aim of which should be appropriate and personalized ovarian stimulation protocols for women in order to avoid iatrogenic implantation dysfunction.
175
Acknowledgements The authors are grateful to the National Natural Science Foundation of China (no. 30801538), the Hubei Population and Family Planning Commission (no. JS2012003) and the Central College Basic Scientific Research Business Special Fund (no. 2012QN011) for financial assistance.
References Bindali, B.B., Kaliwal, B.B., 2002. Anti-implantation effect of a carbamate fungicide mancozeb in albino mice. Ind. Health 40, 191–197. Das, A., Mantena, S.R., Kannan, A., Evans, D.B., Bagchi, M.K., Bagchi, I.C., 2009. De novo synthesis of estrogen in pregnant uterus is critical for stromal decidualization and angiogenesis. Proc. Natl. Acad. Sci. U. S. A. 106, 12542–12547. Goto, T., Endo, T., Henmi, H., Kitajima, Y., Kiya, T., Nishikawa, A., Manase, K., Sato, H., Kudo, R., 1999. Gonadotropin-releasing hormone agonist has the ability to induce increased matrix metalloproteinase (MMP)-2 and membrane type 1-MMP expression in corpora lutea, and structural luteolysis in rats. J. Endocrinol. 161, 393–402. Lydon, J.P., DeMayo, F.J., Funk, C.R., Mani, S.K., Hughes, A.R., Montgomery Jr., C.A., Shyamala, G., Conneely, O.M., O’Malley, B.W., 1995. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev. 9, 2266–2278. Malven, P.V., Sawyer, C.H., 1966. A luteolytic action of prolactin in hypophysectomized rats. Endocrinology 79, 268–274. Sharpe-Timms, K.L., Ricke, E.A., Piva, M., Horowitz, G.M., 2000. Differential expression and localization of de novo synthesized endometriotic haptoglobin in endometrium and endometriotic lesions. Hum. Reprod. 15, 2180–2185. Declaration: The authors report no financial or commercial conflicts of interest. Received 26 July 2012; refereed 28 March 2013; accepted 28 March 2013.