Assisted reproductive technologies do not increase risk of abnormal methylation of PEG1/MEST in human early pregnancy loss

Transcription

1 Assisted reproductive technologies do not increase risk of abnormal methylation of PEG1/MEST in human early pregnancy loss Hai-Yan Zheng, M.D., Xiao-Yun Shi, M.D., M.Sc., Fang-Rong Wu, M.D., Ya-Qin Wu, M.D., Le-Le Wang, M.D., and Shi-Ling Chen, M.D., M.Sc. Center for Reproductive Medicine, Department of Gynecology and Obstetrics, Nanfang Hospital, Southern Medical University, Guangzhou, People s Republic of China Objective: To evaluate the epigenetic risk linked to assisted reproductive technology (ART) by analyzing the methylation patterns of imprinted PEG1 gene in aborted human chorionic villus. Design: Experimental research study. Setting: Research laboratory. Patient(s): Four patients groups were tested: spontaneous abortion after ART (n ¼ 44), multifetal reduction after ART (n ¼ 22), spontaneous abortion of natural pregnancies (n ¼ 45), and induced abortion of natural pregnancies (n ¼ 47). Intervention(s): Methylation patterns of PEG1 in the aborted chorionic villus were determined. Main Outcome Measure(s): The DNA methylation patterns were analyzed using pyrosequencing and bisulfite sequencing polymerase chain reaction. The percentage of methylation was compared in chorionic villus from the four experimental groups. Result(s): Regardless of conception method, the PEG1 methylation percentage in chorionic villus from spontaneous abortions was significantly higher than in villus from induced abortions and multifetal reduction. In the spontaneous abortions groups, the percent methylation of PEG1 was similar in the villus derived from ART and from natural pregnancies. The two fertilization methods (IVF and intracytoplasmic sperm injection) did not show significant differences either. However, receiver operating characteristic curve analysis revealed a significant positive correlation between PEG1 methylation percentage and rate of early spontaneous abortions. Conclusion(s): As some studies have suggested, imprinting errors of PEG1 may contribute to spontaneous abortion, but ART procedures might not increase the occurrence of aberrant PEG1 methylation patterns. (Fertil Steril Ò 2011;96:84 9. Ó2011 by American Society for Reproductive Medicine.) Key Words: Assisted reproductive technology, spontaneous abortion, imprinted gene, DNA methylation Increasing widespread availability of assisted reproductive technology (ART) makes it possible for many infertile couples to have their own genetic child. In developed countries, children born after ART account for 1% 3% of all births every year (1). Concerns about this technology causing congenital malformations, chromosomal aberrations, and developmental problems have been raised. Several reports suggested that ART might increase the risk of diseases linked to aberrant genomic imprinting (2, 3). Imprinting is a mechanism of gene regulation by which only one of the parental copies of a gene is expressed (4 6) and is subject to certain epigenetic modifications, DNA methylation at the CpG dinucleotides being the most thoroughly studied to date (7). Many imprinted genes are active in the placenta and play a key role in fetal resource acquisition (8, 9). Received November 15, 2010; revised April 1, 2011; accepted April 4, 2011; published online May 14, H.-Y.Z. has nothing to disclose. X.-Y.S. has nothing to disclose. F.-R.W. has nothing to disclose. Y.-Q.W. has nothing to disclose. L.-L.W. has nothing to disclose. S.-L.C. has nothing to disclose. H.-Y.Z., X.-Y.S., and F.-R.W. contributed equally to this work. Supported by a grant (2007CB948104) from the National Key Basic Research Development Plan of China (973 Program), Beijing. Reprint requests: Shi-Ling Chen, M.D., M.Sc., Center for Reproductive Medicine, Department of Gynecology and Obstetrics, Nanfang Hospital, Southern Medical University, Guangzhou , People s Republic of China ( chensl_92@163.com). The effect of ART on DNA methylation in human gametes and offspring conceived by ART has been demonstrated in many studies. Sato et al. (10) found that superovulation can lead to altered methylation status in aspired oocytes. A study of human preimplantation embryos conducted by Chen et al. (11) also found that in vitro culture might cause imprinting errors of H19 gene in embryos. In general, it is presumed that the children born as a result of assisted conception are not exposed to an a priori higher risk of DNA methylation defects and imprinting disorders (12, 13). However, a consensus opinion on the absolute incidence rate of such disorders has not yet been reached (1, 14, 15). Despite growing interest in ART, very little is known about the possible impact of such interventions on the rate of spontaneous abortions. Sporadic miscarriages constitute perhaps the most common medical problem for women of reproductive age; they account for 15% 20% of all pregnancies. Often described as an example of self-selection behavior demonstrating the survival of the fittest principle, they serve as a mechanism reducing the frequency of birth defects. It has been suggested that in ART pregnancies, the risk of spontaneous abortion might be slightly increased (16). Among many reasons for such miscarriages, gene imprinting defects are some of the possible causes (17). Assisted reproductive technology procedures involve intervention in several stages of conception and early development, from stimulation of gamete production to ex vivo culture of embryos. Any of these steps might disturb the 84 Fertility and Sterility â Vol. 96, No. 1, July /$36.00 Copyright ª2011 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 normal imprinting process (1). Some animal studies have demonstrated that in vitro culture of embryos can cause methylation defects in individual genes, which might affect subsequent embryonic development and differentiation (18). Hitherto, the methylation pattern of PEG1 (paternally expressed gene 1) gene in extraembryonic cells (chorionic tissue) from ART pregnancies has not been investigated. Human PEG1, also known as MEST (mesoderm-specific transcript) maps to 7q32 and is imprinted in fetal tissues, with monoallelic expression of the paternal allele (19, 20). The biological function of PEG1 is still uncertain. Because its protein is closely related to a/b hydrolase folding family, it is likely to be a hydrolase with a regulatory function, important in the fetal mesoderm s development (21). Loss of PEG1 function might contribute to the intrauterine and postnatal growth retardation phenotype associated with mupd7 (22 24). In this study, methylation-specific polymerase chain reaction (MS-PCR), pyrosequencing and bisulfite sequencing PCR were used to investigate the methylation patterns of PEG1 in human chorionic villus from ART and natural pregnancies. Our results might help to determine whether ART procedures are related to the abnormal methylation status associated with early pregnancy loss. MATERIALS AND METHODS Experimental Groups The study protocol was approved by the Institutional Ethics Committee of Nanfang Hospital, and informed consent was obtained from all patients. Chorionic villus samples were collected from women who underwent abortion procedures in the Department of Gynecology and Obstetrics at Nanfang Hospital from May 2008 to November These patients had no known anatomic or genetic abnormalities. All the samples were divided into four groups according to the source, as follows. [1] Spontaneous abortion after ART (SA-ART, n ¼ 44). In cases of spontaneous abortions, an intrauterine sac without fetal heartbeat was observed. [2] Spontaneous abortion of natural pregnancies (SA-N, n ¼ 45). [3] Multifetal reduction after ART (FR-ART, n ¼ 22), which was performed transvaginally by ultrasound. [4] Induced abortion of natural pregnancies (IA-N, n ¼ 47). An early pregnancy was confirmed by ultrasonography with a fetal heartbeat. We used the samples from patients who had undergone two types of ART: IVF and intracytoplasmic sperm injection (ICSI). The duration of pregnancy ranged from 6 to 10 weeks. Patient age (mean SD) was years (range, years). Chorionic villus were collected immediately after abortion or multifetal reduction, washed with physiological saline, and stored at 80 C. Genomic DNA was extracted using Genomic DNA Purification Kit (Promega). Bisulfite Treatment and MS-PCR Bisulfite treatment of genomic DNA was performed with the EpiTect Bisulfite Kit (Qiagen), and MS-PCR was carried out to distinguish methylated from unmethylated DNA (25, 26). The primer sequences used are given below (27). Maternal primer pair PEG1-Mfor: 5 0 -TAGTTGCGTTTCGTAAGC TAGTGTC-3 0 ( -Mfor standing for maternal forward). PEG1-Mrev: 5 0 -ACA CAATCCTCCGCTCGCCTA-3 0 ( -Mrev standing for maternal reverse). Paternal primer pair PEG1-Pfor: 5 0 -GTGGTAGTTGTGTTTTG TAAGTGTAGTGTT-3 0 ( -Pfor standing for paternal forward). PEG1-Prev: 5 0 -CACACAATCCTCCACTCACCTACA-3 0 ( -Prev standing for paternal reverse). Each sample was analyzed in two independent MS-PCR reactions. The 25-mL PCR reaction mix contained 2 PCR HotStart premix buffer (TaKaRa Bio) and 50 ng of bisulfite-modified DNA. We used 0.5 mm PEG1-Mfor and 0.5 mm PEG1-Mrev when amplifying the maternal imprint or 0.5 mm PEG1- Pfor and 0.5 mm PEG1-Prev in the paternal PCR. The reactions were carried out in a GeneAmp PCR system The PCR conditions were as follows: 5- minute denaturation at 95 C, followed by 35 cycles of 94 C, 60 C, and 72 C for 45 seconds each, and a final extension at 72 C for 5 minutes. Polymerase chain reaction products were separated by 2% agarose gel electrophoresis, and gel images were analyzed using Labworks image acquisition and analysis software (Ultra-violet Products). Polymerase chain reaction amplification gave 300-bp PCR products from the methylated maternal PEG1 allele and from the unmethylated paternal allele. All the MS-PCR reactions were performed in triplicate. Pyrosequencing The PEG1 gene was amplified by PCR using bisulfite-converted DNA as template; the 219-bp sequence encompasses 22 CpG sites (Gene Bank, Y10620: ). The primers used were as follows: forward primer F, 5 0 -TYG TTG TTG GTT AGT TTT GTA YGG TT-3 0 ; and reverse primer R, 5 0 -CCC AAA AAC AAC CCC AAC TC-3 0 (28). The 50-mL reaction mix contained 1 mm primers and 2 PCR HotStart premix buffer. Polymerase chain reaction was performed under the following conditions: starting at 95 C for 5 minutes, followed by 50 cycles of 94 C for 45 seconds, 60 C for 30 seconds, 72 C for 45 seconds, and a final extension at 72 C for 5 minutes. The sequencing primer for bisulfite pyrosequencing was designed using Pyrosequencing Assay Design Software (Qiagen): its sequence was 5 0 -GAG GGG GTG TGG TTG-3 0 (Supplemental Fig. 1, available online). Bisulfite pyrosequencing was performed using the PyroMark Q96 ID System (Qiagen), and pyrosequencing reactions were performed according to the manufacturer s instructions. The degree of methylation at each CpG site was determined by Allele Quantification software. All samples were analyzed in triplicate. Human sperm DNA and oocyte DNA were studied as completely unmethylated control (theoretic percent methylation, none) and completely methylated control (theoretic percent methylation, 100%), respectively. Cloning and Sequencing To confirm the pyrosequencing results, PCR products for samples of interest were gel purified using the QIAEX II Gel Extraction Kit (Qiagen) and cloned into the pgem-t vector (Promega). Approximately 20 clones of each individual sample were sequenced. Statistical Analysis The statistical analysis program SPSS 16.0 was used to analyze the methylation values of PEG1 in chorionic villus samples from the four experimental groups. Quantitative data were calculated as mean SD. Box plots were generated using the program s default parameters. The bottom and the top of the box indicate the 25th and 75th percentile, respectively. The T bars extend from the boxes to at most 1.5 times the height of the box; the samples not contained within the T bars are defined as outliers. To compare the number of outliers, we used the c 2 test or Fisher s exact test. The receiver operating characteristic (ROC) curve method was used to analyze the potential association between PEG1 methylation percentage and the incidence of early spontaneous abortion. RESULTS Analysis of DNA Methylation Patterns Using MS-PCR None of the 158 chorionic villus samples showed a complete loss of maternal or paternal alleles of PEG1. The 300-bp products from methylated maternal chromosome and those from unmethylated paternal chromosome have been detected in all samples. A representative example of the MS-PCR analysis results for the four groups is shown in Supplemental Figure 2. DNA Methylation and Early Spontaneous Abortion In the sperm DNA control and oocyte DNA control, the methylation value of four CpG sites ranged from 0 to 3.4% and from 92.4% to 97.7%, respectively. Bisulfite conversion efficiency values for those Fertility and Sterility â 85

3 two controls were 96.4% 100% and 92.5% 96.8%, confirming the expected reliability and stability of the method (Fig. 1). For all the examined chorionic villus samples we achieved bisulfite conversion efficiency between 92.9% and 100%. The methylation value for those samples was in the range of 42.0% 86.9%. Clear hypomethylation (<10%) or hypermethylation (>90%) was not detected. However, the PEG1 percentage methylation for the four experimental groups showed significant differences (P<.05). The percentage of methylation in the SA-N group and the SA-ART group was significantly higher than in the FR-ART group and the IA-N group (P<.05) (Table 1). Furthermore, the number of potentially abnormal methylation values did not differ significantly between the SA-ART and SA-N groups (P>.05): 2 of 44 (4.5%) and 1 of 45 (2.2%) methylation values were outliers in the two groups, respectively. There were no outliers in the FR-ART and IA-N groups (Fig. 1, Table 1). Among all the cases of ART, there were 46 conceptions achieved by IVF and 20 as a result of ICSI. Samples from spontaneous abortions in the IVF group account for 63.0% (29 of 46) and in ICSI group for 75.0% (15 of 20); the rates for the two fertilization method groups were not significantly different (P>.05), and neither were the percentage of methylation or the outlier rates (Table 1). In addition, ROC curve analysis showed a positive correlation between PEG1 methylation percentage in chorionic villus and rates of early spontaneous abortion: the higher the methylation percentage, the greater the chance of early miscarriage (P<.001) (Fig. 1). To confirm the pyrosequencing results, we carried out cloning and sequencing for five samples of interest. The results are listed in Table 2. As expected, the data showed hypermethylation for two samples from the SA-ART and SA-N groups, hypomethylation for one sample from SA-ART, and a differential methylation pattern (approximately half methylated and half unmethylated) for samples FIGURE 1 (A C) Pyrosequencing results of PEG1 in controls and chorionic villus. T% of two sites marked with an asterisk (*) represents the bisulfite conversion efficiency. (A) Unmethylated pattern in sperm DNA (0.85% methylation). (B) Methylated pattern in oocyte DNA (94.8% methylation). (C) Normal methylation profile in villus (53.5% methylation). (D) Box plots show the distribution of methylation values for the four experimental groups. Median values are shown as horizontal lines. Bottom of the box indicates the 25th percentile, top the 75th percentile. Outliers are shown as asterisks (cases 1, 41, and 129). (E) ROC curve analysis demonstrates a positive correlation between methylation percentage of PEG1 and early spontaneous abortion rate. Area under the curve was (95% confidence interval, ). 86 Zheng et al. Imprinting errors in early pregnancy loss Vol. 96, No. 1, July 2011

4 TABLE 1 Comparison of PEG1 percentage of methylation in four chorionic villus groups and between the two fertilization methods. All samples Samples from ART SA-ART (n [ 44) FR-ART (n [ 22) SA-N (n [ 45) IA-N (n [ 47) P value IVF (n [ 46) ICSI (n [ 20) P value Variable Fluctuation range (%) Methylation, mean SD (%) a b c d NS Outliers, % (n) 4.5 (2/44) 0 (0/22) 2.2 (1/45) 0 (0/47) NS 2.2 (1/46) 5.0 (1/20) NS e Note: NS ¼ no significant difference (P>.05). a,b,c,d Percentage of methylation multiple comparisons: P<.05 for a vs. d, c vs. d, a vs. b, and b vs. c. e Fisher exact test. from FR-ART and IA-N (Fig. 2 and Table 2). This was in agreement with the results of pyrosequencing. DISCUSSION In this study we found significant differences between methylation percentage of PEG1 in the samples from early pregnancies and those obtained from spontaneous abortions. We also observed a positive correlation between methylation percentage and the rate of early miscarriage. These results suggest that an abnormal methylation pattern of PEG1 may affect the stability of early pregnancy, but ART might not increase the occurrence of aberrant methylation of PEG1. It seems that normal gene imprinting is an important factor in early pregnancy; it has been demonstrated that many imprinted genes play a crucial role in embryo development and cell differentiation (29). Some studies reported that imprinted gene defects lead to early embryonic and fetal death, prolonged labor, and development of embryonic tumors (17, 30). Animal studies have demonstrated that the inactivation of PEG1 can cause retardation of embryo development and lower live birth rate in mice (24). Monoallelic expression of PEG1 in human oocyte and embryo occurs during the preimplantation stage (31). The maternal uniparental disomy of PEG1 can cause Silver-Russell syndrome, with its associated growth retardation and sporadic cases of malformation (20). In our study, we found that the methylation percentage of PEG1 was higher in spontaneous abortions than in early pregnancies, regardless of the conception method. Receiver operating characteristic curve analysis demonstrated that the higher the PEG1 methylation percentage, the greater the chance of early miscarriage. It is likely that the abnormal methylation of PEG1 is a contributory factor in early spontaneous abortions. There is growing concern that ART procedures might be associated with imprinting abnormalities in early pregnancy and increase the risk of early miscarriage (16, 32 34). However, while evaluating such associations, we should keep in mind the developmental time scale and highly dynamic character of epigenetic events taking place before and during early pregnancy. Germ cells have the ability to erase the imprinting marks during their development, and the re-establishment of imprinting in female germ cells (oocytes) is not completed until just before ovulation (35, 36). Assisted reproductive technology procedures might include manipulations such as stimulation of oocyte growth, retrieval of oocytes before ovulation, ICSI, in vitro culture of preimplantation embryos, and cryopreservation of either gametes or embryos. TABLE 2 Methylation percentage of samples determined by pyrosequencing and BSP. Sample no. Source Methylation Pyrosequencing (%) BSP (%) 1 SA-ART SA-ART FR-ART SA-N IA-N Note: BSP ¼ bisulfite sequencing PCR. Fertility and Sterility â 87

5 FIGURE 2 Bisulfite-PCR sequencing for cases 1, 41, 57, 129, and 136. Vertical bars signify CpG sites. The region between two inner arrows, including four CpG sites (*), represents the sequences assessed by pyrosequencing. Each row represents a unique methylation profile within the pool of clones sequenced, with the frequency of that methylation state shown to the right of each row. Closed and open circles represent methylated and unmethylated CpGs, respectively. Results are summarized in Table 2. These interventions often fall within the phase of reprogramming of gene imprinting, so we could assume that some ART procedures might affect the methylation process in gametes and embryos (1). Several animal studies demonstrated that in vitro culture can alter embryo epigenetics, causing widespread changes in imprinted gene expression or even loss of imprinting (37 39). At least one study reports that human preimplantation embryos cultured in vitro show imprinting errors in the H19 gene differentially methylated region (11). However, the embryos used in that study were of poor quality (unsuitable for transfer or freezing); this might suggest that methylation abnormalities might be caused by a self-selection process rather than by ART procedures. Our data showed a significant difference in the methylation percentage of PEG1 in samples from the first-trimester pregnancies compared with samples from spontaneous abortions. After two different fertilization methods, IVF and ICSI, the methylation percentage of PEG1 in chorionic villus from the two groups was not significantly different. We presume that the altered methylation of imprinted gene PEG1 predisposes to early spontaneous abortions, but ART itself might not necessarily increase the risk of abnormal methylation. Some animal studies showed that IVF and in vitro embryo culture result in abnormal methylation in mouse blastula (40, 41). However, most of those studies were carried out at embryo blastula stage; on day 10.5 of development in mice pregnant by IVF, embryos and placentas displayed a normal methylation profile (42). Furthermore, in children born after ART, a higher risk of DNA methylation defects and imprinting disorders was not observed (12, 13, 15). It is likely that not just one gene but a complex, coordinated network of imprinted genes is responsible for the embryo selection process. To summarize, our study investigated methylation patterns of PEG1 gene in chorionic villus from spontaneous abortions after ART, multifetal reduction after ART, spontaneous abortions of natural pregnancies, and induced abortions of natural pregnancies. Our data demonstrate that two ART procedures (IVF and ICSI) might not increase methylation errors of PEG1, which possibly affects embryo development in early pregnancy stages. However, our study is limited by the small sample size, and we cannot rule out the possibility that the observed methylation patterns represent normal variation during fetal development. Therefore, our results should be confirmed by studies involving a large sample size and multiple imprinted genes. Acknowledgments: The authors thank all members of the Research Center of Clinical Medicine of Nanfang Hospital for their technical support and valuable suggestions; and Dr. Hao Bi (Department of Biochemistry and Molecular Biology, Tongji Medicine College, Huazhong University of Science and Technology, China) for his assistance in the preparation of the manuscript. REFERENCES 1. Manipalviratn S, DeCherney A, Segars J. Imprinting disorders and assisted reproductive technology. Fertil Steril 2009;91: Klemetti R, Gissler M, Sevon T, Koivurova S, Ritvanen A, Hemminki E. Children born after assisted fertilization have an increased rate of major congenital anomalies. Fertil Steril 2005;84: Sutcliffe AG, Peters CJ, Bowdin S, Temple K, Reardon W, Wilson L, et al. Assisted reproductive therapies and imprinting disorders a preliminary British survey. Hum Reprod 2006;21: Barlow DP. Gametic imprinting in mammals. Science 1995;270: Constancia M, Pickard B, Kelsey G, Reik W. Imprinting mechanisms. Genome Res 1998;8: Hall JG. Genomic imprinting: review and relevance to human diseases. Am J Hum Genet 1990;46: Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003;33(Suppl): Miozzo M, Simoni G. The role of imprinted genes in fetal growth. Biol Neonate 2002;81: Constancia M, Kelsey G, Reik W. Resourceful imprinting. Nature 2004;432: Sato A, Otsu E, Negishi H, Utsunomiya T, Arima T. Aberrant DNA methylation of imprinted loci in superovulated oocytes. Hum Reprod 2007;22: Chen SL, Shi XY, Zheng HY, Wu FR, Luo C. Aberrant DNA methylation of imprinted H19 gene in human preimplantation embryos. Fertil Steril 2010; 94:2356 8, 2358.e1. 88 Zheng et al. Imprinting errors in early pregnancy loss Vol. 96, No. 1, July 2011

6 12. Tierling S, Souren NY, Gries J, Loporto C, Groth M, Lutsik P, et al. Assisted reproductive technologies do not enhance the variability of DNA methylation imprints in human. J Med Genet 2010;47: Manning M, Lissens W, Bonduelle M, Camus M, De Rijcke M, Liebaers I, et al. Study of DNAmethylation patterns at chromosome 15q11-q13 in children born after ICSI reveals no imprinting defects. Mol Hum Reprod 2000;6: Allen C, Reardon W. Assisted reproduction technology and defects of genomic imprinting. BJOG 2005;112: Bowdin S, Allen C, Kirby G, Brueton L, Afnan M, Barratt C, et al. A survey of assisted reproductive technology births and imprinting disorders. Hum Reprod 2007;22: Wang JX, Norman RJ, Wilcox AJ. Incidence of spontaneous abortion among pregnancies produced by assisted reproductive technology. Hum Reprod 2004;19: Ono R, Nakamura K, Inoue K, Naruse M, Usami T, Wakisaka-Saito N, et al. Deletion of Peg10, an imprinted gene acquired from a retrotransposon, causes early embryonic lethality. Nat Genet 2006;38: Khosla S, Dean W, Brown D, Reik W, Feil R. Culture of preimplantation mouse embryos affects fetal development and the expression of imprinted genes. Biol Reprod 2001;64: Nishita Y, Yoshida I, Sado T, Takagi N. Genomic imprinting and chromosomal localization of the human MEST gene. Genomics 1996;36: Kobayashi S, Kohda T, Miyoshi N, Kuroiwa Y, Aisaka K, Tsutsumi O, et al. Human PEG1/MEST, an imprinted gene on chromosome 7. Hum Mol Genet 1997;6: Kaneko-Ishino T, Kuroiwa Y, Miyoshi N, Kohda T, Suzuki R, Yokoyama M, et al. Peg1/Mest imprinted gene on chromosome 6 identified by cdna subtraction hybridization. Nat Genet 1995;11: Voss R, Ben-Simon E, Avital A, Godfrey S, Zlotogora J, Dagan J, et al. Isodisomy of chromosome 7 in a patient with cystic fibrosis: could uniparental disomy be common in humans? Am J Hum Genet 1989;45: Spence JE, Perciaccante RG, Greig GM, Willard HF, Ledbetter DH, Hejtmancik JF, et al. Uniparental disomy as a mechanism for human genetic disease. Am J Hum Genet 1988;42: Lefebvre L, Viville S, Barton SC, Ishino F, Keverne EB, Surani MA. Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat Genet 1998;20: Kubota T, Das S, Christian SL, Baylin SB, Herman JG, Ledbetter DH. Methylation-specific PCR simplifies imprinting analysis. Nat Genet 1997;16: Zeschnigk M, Lich C, Buiting K, Doerfler W, Horsthemke B. A single-tube PCR test for the diagnosis of Angelman and Prader-Willi syndrome based on allelic methylation differences at the SNRPN locus. Eur J Hum Genet 1997;5: Kagami M, Nagai T, Fukami M, Yamazawa K, Ogata T. Silver-Russell syndrome in a girl born after in vitro fertilization: partial hypermethylation at the differentially methylated region of PEG1/MEST. J Assist Reprod Genet 2007;24: Kerjean A, Dupont JM, Vasseur C, Le Tessier D, Cuisset L, Paldi A, et al. Establishment of the paternal methylation imprint of the human H19 and MEST/ PEG1 genes during spermatogenesis. Hum Mol Genet 2000;9: Falls JG, Pulford DJ, Wylie AA, Jirtle RL. Genomic imprinting: implications for human disease. Am J Pathol 1999;154: Kim SJ, Park SE, Lee C, Lee SY, Kim IH, An HJ, et al. Altered imprinting, promoter usage, and expression of insulin-like growth factor-ii gene in gestational trophoblastic diseases. Gynecol Oncol 2003;88: Salpekar A, Huntriss J, Bolton V, Monk M. The use of amplified cdna to investigate the expression of seven imprinted genes in human oocytes and preimplantation embryos. Mol Hum Reprod 2001;7: Schieve LA, Tatham L, Peterson HB, Toner J, Jeng G. Spontaneous abortion among pregnancies conceived using assisted reproductive technology in the United States. Obstet Gynecol 2003;101: Klipstein S, Regan M, Ryley DA, Goldman MB, Alper MM, Reindollar RH. One last chance for pregnancy: a review of 2,705 in vitro fertilization cycles initiated in women age 40 years and above. Fertil Steril 2005;84: Winter E, Wang J, Davies MJ, Norman R. Early pregnancy loss following assisted reproductive technology treatment. Hum Reprod 2002;17: Schaefer CB, Ooi SK, Bestor TH, Bourc His D. Epigenetic decisions in mammalian germ cells. Science 2007;316: Obata Y, Kaneko-Ishino T, Koide T, Takai Y, Ueda T, Domeki I, et al. Disruption of primary imprinting during oocyte growth leads to the modified expression of imprinted genes during embryogenesis. Development 1998;125: Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG, Carolan C, et al. Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 2001;27: Suzuki JJ, Therrien J, Filion F, Lefebvre R, Goff AK, Smith LC. In vitro culture and somatic cell nuclear transfer affect imprinting of SNRPN gene in preand post-implantation stages of development in cattle. BMC Dev Biol 2009;9: Mann MR, Lee SS, Doherty AS, Verona RI, Nolen LD, Schultz RM, et al. Selective loss of imprinting in the placenta following preimplantation development in culture. Development 2004;131: Doherty AS, Mann MR, Tremblay KD, Bartolomei MS, Schultz RM. Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 2000;62: Sasaki H, Ferguson-Smith AC, Shum AS, Barton SC, Surani MA. Temporal and spatial regulation of H19 imprinting in normal and uniparental mouse embryos. Development 1995;121: Fauque P, Ripoche MA, Tost J, Journot L, Gabory A, Busato F, et al. Modulation of imprinted gene network in placenta results in normal development of in vitro manipulated mouse embryos. Hum Mol Genet 2010;19: Fertility and Sterility â 89

7 SUPPLEMENTAL FIGURE 1 Location of PEG1 gene s sequencing primer. There are four methylated CpG sites and two conversion efficiency sites. T represents methylated CpG site, and C represents conversion efficiency site. 89.e1 Zheng et al. Imprinting errors in early pregnancy loss Vol. 96, No. 1, July 2011

8 SUPPLEMENTAL FIGURE 2 Representative example of MS-PCR analysis of PEG1 in chorionic villus from four experimental groups. Paternal-specific MS-PCR products (300 bp) are shown on the left, maternal-specific products (300 bp) on the right. Lane 1, spontaneous abortion after ART. Lane 2, multifetal reduction after ART. Lane 3, spontaneous abortion of natural pregnancy. Lane 4, induced abortion of natural pregnancy. Fertility and Sterility â 89.e2