Dynamics of Imprinted DNA Methylation and Gene Transcription for ...

3 downloads 133 Views 2MB Size Report
Oct 9, 2013 - in oncofertility practice as an alternative to ovarian cortical tissue transplantation [2] has also attracted research attention. So far, complete in ...
BIOLOGY OF REPRODUCTION (2013) 89(6):130, 1–8 Published online before print 9 October 2013. DOI 10.1095/biolreprod.113.111641

Dynamics of Imprinted DNA Methylation and Gene Transcription for Imprinting Establishment in Mouse Oocytes in Relation to Culture Duration Variability1 Ellen Anckaert,2 Flor Sa´nchez, Katy Billooye, and Johan Smitz Follicle Biology Laboratory, Vrije Universiteit Brussel, Brussels, Belgium

Several studies have linked assisted reproductive technologies to aberrant imprinting. We previously showed that 12-day in vitro follicle culture supports normal imprinting establishment in mouse oocytes. The aim of the present study was to assess whether shortened in vitro follicle growth (8 days of culture compared with 12 days, as a model for human in vitro maturation) or preovulatory intrafollicular oocyte ‘‘aging’’ in culture (14 days of culture) leads to imprinting mutations in oocytes. Limiting-dilution bisulphite sequencing showed that shortened in vitro follicle growth (8 days) does not induce oocyte epimutations at the imprinted Snrpn and Mest genes. In contrast, extension of oocyte residence in large unluteinized follicles in vitro was associated with a low level (1 of 54 alleles) of epimutations for Mest but not for Snrpn. The latter condition may occur during controlled ovarian stimulation where the oocyte growth phase may be extended for several days. Furthermore, we studied the dynamics during follicle culture of transcript levels for genes previously shown to be essential for imprinting establishment in oocytes, including Dnmt3a, Dnmt3L, and Zfp57. Oocyte total mRNA levels during in vitro follicle culture showed the timely shutdown in transcription at the antral follicle stage, and total mRNA levels were comparable to those of in vivo grown equine chorionic gonadotropinstimulated oocytes. DNA methylation, follicle culture, genomic imprinting, oocyte maturation, total mRNA

INTRODUCTION Oocyte in vitro maturation (IVM) has been introduced in the fertility clinic as a patient-friendly and minimal stimulus alternative to conventional ovarian stimulation. Advantages of the technique are avoidance of ovarian hyperstimulation syndrome and reduced treatment cost and burden [1]. Follicle culture technology for human clinical applications in oncofertility practice as an alternative to ovarian cortical tissue transplantation [2] has also attracted research attention. So far, complete in vitro growth of follicles from the primordial or early preantral stages to mature developmentally competent oocytes has only been successful in the mouse model [3]. There is concern that in vitro culture of oocytes might interfere with genomic imprinting. Several investigations have 1 Supported by research funds WFWG 2009 and 2011 from Wetenschappelijk Fonds Willy Gepts and by funding from Cook Medical. 2 Correspondence: Ellen Anckaert, Follicle Biology Laboratory, Vrije Universiteit Brussel, Brussels, Belgium B-1090. E-mail: [email protected]

Received: 27 June 2013. First decision: 24 July 2013. Accepted: 2 October 2013. Ó 2013 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

1

Article 130

Downloaded from www.biolreprod.org.

indeed linked assisted reproductive technologies (ARTs) such as ovarian stimulation and in vitro embryo culture to aberrant imprinting in different species [4]. Furthermore, evidence has suggested an increased incidence of rare human imprinting disorders such as Beckwith-Wiedemann syndrome in children conceived after the use of ARTs [5–13]. The most important epigenetic mechanism regulating imprinted gene expression is DNA methylation: differentially methylated DNA patterns are present at maternal and paternal DNA regions that result in expression of one allele and inactivation of the opposite parental allele for imprinted genes [14, 15]. During oocyte growth and maturation, DNA methylation at imprinting control regions of imprinted genes should be correctly imposed on the maternal allele [16, 17], and DNA methylation maintenance factors such as DNA methyltransferase 1 (Dnmt1), Stella, zinc finger protein 57 (Zfp57), and methyl-cytosine phosphodiester bond guanine (CpG)-binding protein 3 (Mbd3) should be expressed and stored to allow these oocytes to give rise to healthy live offspring after fertilization [18–21]. The timely acquisition of correct imprinted DNA methylation patterns in oocytes (imprinting establishment) and the maintenance of genomic imprinting after fertilization are thus both required for normal embryonic development. Well-designed human studies are currently not available, but animal models provide reassuring data on imprinting establishment in cultured oocytes [22]. Our group has previously described normal imprinting establishment in mouse oocytes from in vitro follicle culture from the early preantral stage [23]. The study of various suboptimal culture conditions indicated that imprinting establishment in oocytes during culture is a rather robust process [23–25]. The latter analyses were done after culturing a precisely selected class of intact follicles measuring 110 to 130 lm for a period of 12 days, yielding maximal meiotic completion and evidence of healthy offspring [26]. An influence of variation in culture duration length, for mimicking IVM protocols and ‘‘intrafollicular’’ oocyte aging, is possible when setting up experiments with a narrow class of isolated follicles at the start and exposing them to different culture times. Shortening the culture time from 12 days to 8 days or extending it from 12 days to 14 days affects the blastocyst formation rate [27], oocyte meiosis [28], and the expression of key oocyte genes [29], but the effects on imprinting are currently unknown. The shortened (8 day) culture is a model for human IVM where cumulus-oocyte complexes (COCs) from small follicles are being retrieved and matured in vitro nearly 1 wk earlier than in conventional in vitro fertilization. We also aimed to study imprinted DNA methylation in oocytes from follicles grown for 2 days longer than normal: this is intrafollicular aging. The latter may occur during controlled ovarian stimulation where inhibition of the luteinizing hormone surge by the use of gonadotropin-releasing hormone analogues extends the growth phase for several days, which may lead

ABSTRACT

ANCKAERT ET AL.

essential medium (Invitrogen) supplemented with 5% heat-inactivated fetal bovine serum, 5 lg/ml of insulin, 5 lg/ml of transferrin, 5 ng/ml of selenium (ITS; Sigma), and 10 IU/L of recombinant follicle-stimulating hormone (rFSH) (Gonal-F; Serono Benelux) from the start of culture. Follicles were individually cultured in 75 ll of medium in 96-well flat-bottom plates (Costar) containing approximately 10 follicles per plate without oil overlay for 8, 12 (in vitro control), or 14 days to grow to antral follicles. Follicles were cultured in an incubator at 378C, 100% humidity, and 5% carbon dioxide in air. Part of the medium (30 ll) was refreshed at Days 4, 8, and 12. At the end of the 12-day culture period, an ovulatory stimulus was given with 1.2 IU/ml of recombinant human chorionic gonadotropin (r-hCG) (Ovitrelle; Serono Benelux) supplemented with 4 ng/ml of recombinant epidermal growth factor (r-EGF) (Roche). Cumulus-oocyte complexes containing MII oocytes are available 18 h after rhCG/r-EGF administration (Day 13). Sixteen independent follicle cultures were performed for the gene expression analysis. For the DNA methylation analysis, nine independent cultures were done.

to ovulation of overripe oocytes. The DNA methylation was studied at the imprinted Snrpn and Mest genes. These two genes were selected because 1) Mest is methylated during the final stages of oocyte growth and is therefore particularly relevant for shortened follicle culture time [16] and 2) aberrant methylation of both Mest and Snrpn after in vitro postovulatory oocyte aging has been described [30, 31]. The methylation studies were complemented with measurements of transcript levels for genes previously shown to be essential for imprinting establishment in oocytes. These include Dnmt3a, Dnmt3L, and Zfp57 [21, 32–34]. MATERIALS AND METHODS Animals This study was performed with F1 mice (C57BL/6J 3 CBA/Ca; Charles River), housed and bred according to national standards for animal care. Our study was approved by the Vrije Universiteit Brussel (project 01-395-1) ethical committee for animal experiments.

In Vitro Oocyte Collection Oocytes at the germinal vesicle stage (GV) were collected at Day 6 of follicle culture (preantral follicles), at Day 8 (early antral follicles), at Day 12 (antral follicles), and at Day 14; MII oocytes were collected at Day 13 of follicle culture. Oocytes were collected into refreshment medium and denuded by repeated pipetting with a fine-glass pipette and washed thoroughly in refreshment medium with special care to avoid cumulus cell contamination. For the expression and DNA methylation analysis, oocytes were pooled in groups of 10. Oocytes were transferred into cryovials (under controlled microscopic inspection), snap frozen in liquid nitrogen, and stored at 808C until the RNA or DNA extraction.

In Vitro Follicle Culture The experimental setup is shown in Figure 1. In vitro follicle culture was performed as described previously [26]. In brief, early preantral follicles with a diameter between 110 and 130 lm were mechanically isolated from the ovaries of 13-day-old to 14-day-old C57BL/6J 3 CBA/ca F1 mice in Leibovitz L15 medium (Invitrogen). The follicle culture medium consisted of a-minimal

2

Article 130

Downloaded from www.biolreprod.org.

FIG. 1. Schematic overview of the experimental setup. Top) Representative views are shown of the in vitro growth of an early preantral follicle in the following defined conditions: Day 1 to Day 4 (theca cells attach the preantral follicle to the bottom of the culture plate), Day 6 (the granulosa cells have broken through the basal membrane and have proliferated), Day 8 (the granulosa cells differentiate into a follicular wall and a cumulus-corona part), and Day 12 (a clear antral-like cavity has formed). Red: The GV and MII oocytes were collected for expression analysis. Blue: The GV oocytes were collected for DNA methylation analysis. 1The MII oocytes were collected at Day 13 after an hCG stimulus at Day 12. 2The GV oocytes were collected after in vitro follicle culture for 14 days without hCG stimulus (original magnification 3400 [Day 1], 3100 [Days 6–12], and 340 [Day 14]). Bottom) In vivo grown GV oocytes were collected from 25-day-old prepubertal mice 48 h after intraperitoneal injection with 2.5 IU of eCG. For collection of MII oocytes, prepubertal mice were stimulated by intraperitoneal injection with 2.5 IU of eCG, followed 48 h later by injection with 2.5 IU of hCG. The MII oocytes were collected 14 h after hCG administration.

IMPRINTING IN OOCYTES RELATED TO CULTURE DURATION TABLE 1. Primers used in nested PCR for Oct4, Snrpn, and Mest DNA methylation analysis. Gene

Primer

Sequence (5 0 –3 0 )*

Amplicon length (base pair)

Oct4

Outer forward Outer reverse Inner forward Inner reverse Outer forward Outer reverse Inner forward Inner reverse Outer forward Outer reverse Inner forward Inner reverse

TTGAGTGGGTTGTAAGGATAGG AAAAAATTTCACCTCTCCCTCC TGTAAAACGACGGCCAGTCCACTCACTCACCCACCCGTAGGGGTGAGAGGATTTTGAA CAGGAAACAGCTATGACCGGGTGGGAGGTGGGAGGGCCACCCTCTAACCTTAACCTCT TGGTAGTTGTTTTTTGGTAGGATAT CAAAAAATTACTCACCAATTCTCA TGTAAAACGACGGCCAGTCCACTCACTCACCCACCCTGGTAGTTGTTTTTTGGTAGGATAT CAGGAAACAGCTATGACCGGGTGGGAGGTGGGAGGGATCCACAAACCCAACTAACCT GTTTGTGTGAGGTTTGGAAT CCACACAAACATTTATTTCCTAC TGTAAAACGACGGCCAGTCCACTCACTCACCCACCCGTTTGTGTGAGGTTTGGAAT CAGGAAACAGCTATGACCGGGTGGGAGGTGGGAGGGTAACCCTCTCTCAAATACTAACATTTTCA

197

Snrpn

Mest

241

304

* The inner primers were tagged with an M13 tag (italic) and a stuffer (bold) sequence. [MII, after hCG administration]) and 2) in vivo (GV oocytes from eCGstimulated large antral follicles and superovulated MII oocytes from prepubertal mice).

In Vivo Oocyte Collection The GV oocytes from 25-day-old prepubertal mice were collected 48 h after intraperitoneal injection with 2.5 IU of equine chorionic gonadotropin (eCG) (Folligon; Intervet). Briefly, cumulus cell-enclosed oocytes (immature GV) were retrieved from ovaries by puncturing the largest antral follicles with insulin syringe needles. Only oocytes from compacted COCs were used. The COCs were washed, and cumulus cells were removed in Leibovitz L15 medium by repeated pipetting using fine-glass pipettes. For collection of mature oocytes, prepubertal mice were stimulated by an intraperitoneal injection with 2.5 IU of hCG (Chorulon; Intervet) 48 h after eCG administration. Fourteen hours after hCG injection, oviducts were removed, and COCs were gently released from the ampulla. The COCs were placed in medium containing 40 IU/ml of hyaluronidase to remove cumulus cells (by repeated pipetting) and then washed in Leibovitz L15 medium. Only oocytes that had extruded the polar body (MII oocytes) were collected. The GV and MII oocytes were frozen in pools of 10 oocytes, snap frozen in liquid nitrogen, and stored at 808C for further gene expression analysis.

cDNA Synthesis

Quantitative Real-Time PCR Specific primers for the following three genes of interest were designed with the Universal Probe Library (Roche): Dnmt3a (forward: CTTTGATGG GATTGCTACAGG; reverse: ACACCTCGGAGGCAATGTAG [fragment size, 80 base pair]), Dnmt3L (forward: AGAATGCTATGCGGGTGTG; reverse: CTCTTCTTCCTTTG [fragment size, 77 base pair]), and Zfp57 (forward: GAAACCTTCAAGAACCTGACATTT; reverse: CCTGTGCAAC TGGAGGACTT [fragment size, 110 base pair]). A LightCycler apparatus (LC480; Roche) was used for absolute quantification analysis with Probes master mix (Roche). Each reaction mixture consisted of 2 ll of cDNA and 8 ll of PCR mix containing 5 ll of Probes master mix, 1 ll of primer mix (600 nM), and 2 ll of nuclease-free water per sample. The PCR protocol used was 10 min at 958C, followed by 55 cycles at 958C for 10 sec and 608C for 30 sec. Triplicate reactions were performed for each sample. Using 10-fold serial dilutions of corresponding synthetic oligo (range, 1 to 106 copies), standard curves were constructed for each gene and were run for each reaction plate. Quantification of exogenous control was carried out using SYBR Green PCR master mix (Roche) and using the PCR protocol already described, followed by acquisition of the melting curve (958C for 5 sec and 608C for 1 min, with a continuous fluorescence measurement). For each sample, oocyte transcripts were normalized to the amount of exogenous control (Luciferase) and to the number of oocytes.

DNA Methylation Analysis Limiting-dilution bisulphite sequencing was used as described previously [35]. This technique allows a high number of independent alleles to be amplified in small DNA amounts from pools containing 10 oocytes without risk for amplification bias. Twelve to sixteen samples with 10 pooled GV oocytes from three independent follicle cultures (involving nine mice) were assessed for each condition (Days 8, 12, and 14). Briefly, DNA was extracted from the oocyte pools and bisulphite treated with the EZ DNA methylation direct kit (Zymo Research). Bisulphite-treated DNA was eluted in 10 ll of elution buffer, diluted to a final volume of 200 ll, and evenly distributed into 20 PCR tubes. Additionally, eight negative controls were added for each PCR reaction. Multiplex PCR for three genes (Oct4, Snrpn, and Mest) was performed, followed by gene-specific nested PCR with primers for bisulphite-modified DNA; FastStart Taq polymerase (Roche) was used in all PCR reactions. Oct4 was included as a nonimprinted marker gene for cumulus cell contamination showing methylated DNA alleles in cumulus cells; in contrast, Oct4 alleles are unmethylated in oocytes [35]. Primers used for Oct4 and for the imprinted Snrpn and Mest genes were designed with Pyrosequencing assay design software (Biotage) (Table 1). The inner primers were tagged with a stuffer sequence [36] to improve the direct sequencing of the CpG-poor DNA. The sequence quality was further improved by tagging the stuffer with an additional M13 tag and using the whole M13/stuffer sequence as a target for the sequencing primers [37]. Sequencing of the PCR fragments was performed on a Genetic Analyzer 310 (Applied Biosystems).

Statistical Analysis Significant differences in mRNA levels between days of in vitro culture were assessed by Kruskal-Wallis test, followed by Dunn test of multiple comparisons. Differences within in vivo conditions were assessed by MannWhitney U-test. Differences between in vitro and in vivo conditions were assessed by Kruskal-Wallis test, followed by Dunn test of multiple comparisons. Differences between global methylation levels of alleles in the three groups (Days 8, 12, and 14) were analyzed by Kruskal-Wallis test, followed by Dunn test of multiple comparisons. P , 0.05 was considered significant. Differences in the epimutation rate or in the CpG error rate were assessed by chi-square test or by Fisher exact test (when 2 3 2 tables contained ,5 samples), both with Bonferroni adjustment for comparison of multiple groups. P , 0.0083 (0.05/6) was considered significant.

RNA Extraction Total RNA was extracted from the oocyte pools using the RNeasy micro-kit (Qiagen) as described previously [29]. A deoxyribonuclease step was included to remove residual genomic DNA. To normalize for technical variation among samples, 10 pg of Luciferase mRNA (Promega) was added to each sample as an exogenous control. Extracted mRNA was eluted in 14 ll of ribonuclease-free water. For total mRNA, six samples of 10 pooled oocytes were extracted per condition. These included 1) during in vitro follicle culture under 10 IU/L of rFSH (at Days 6, 12, and 14 [GV, before hCG administration] and at Day 13

3

Article 130

Downloaded from www.biolreprod.org.

Synthesis of cDNA was performed using the iScript cDNA synthesis kit (Bio-Rad Laboratories). The reverse transcription mixture contained 10 ll of extracted RNA, 5 ll of nuclease-free water, 4 ll of reaction mixture, and 1 ll of reverse transcriptase per sample. Cycling conditions for reverse transcription were 5 min at 258C, 30 min at 428C, and 5 min at 858C. After the cDNA synthesis, samples were diluted to a final volume of 60 ll.

ANCKAERT ET AL. TABLE 2. Summary of DNA methylation results at Snrpn and Mest in GV oocytes after 8, 12, and 14 days of culture. Epimutation rate

CpG error rate (%)

Global CpG methylation level (%)

8 12 14

0/47 0/36 1/54

15/47 (31.9) 11/36 (30.6) 19/54 (35.2)

96.3 97.0 95.2–97.0*

8 12 14

0/55 0/55 0/48

19/55 (34.5)a 12/55 (21.8) 4/48 (8.3)a

97.3b 98.1 99.2b

Gene Mest Day Day Day Snrpn Day Day Day

* After exclusion of the single allele with the epimutation. a P ¼ 0.0018. b P , 0.01.

RESULTS DNA Methylation Analysis at Imprinting Control Regions of the Imprinted Genes Snrpn and Mest Respectively, 16, 12, and 16 oocyte pools (containing 10 GV oocytes per pool) were analyzed for Days 8, 12, and 14 of follicle culture; this resulted in the amplification of, respec-

FIG. 2. The DNA methylation of Oct4 in GV oocytes after 8 (left), 12 (center), or 14 (right) days of in vitro follicle culture. Each line represents an individual allele, and each circle represents a CpG site within the differentially methylated region analyzed; filled circles, methylated cytosines; open circles, unmethylated cytosines; and missing circles, CpG sites whose methylation status could not be determined.

4

Article 130

Downloaded from www.biolreprod.org.

tively, 49, 42, and 61 different alleles for Oct4 (nonimprinted marker gene for cumulus cell contamination); 55, 55, and 48 different alleles for Snrpn; and 47, 36, and 54 different alleles for Mest (Table 2). The high number of different alleles amplified confirms earlier reports of the increased efficiency of limiting-dilution bisulphite sequencing (including multiplex PCR) compared with classic bisulphite sequencing performed on oocyte pools [35]. Furthermore, by diluting the bisulphitetreated DNA so that each PCR well contains a maximum of one DNA molecule, amplification bias can be prevented. This is particularly relevant because amplification bias, mostly toward unmethylated DNA molecules, has been described [38]. The analysis of the nonimprinted marker gene (Oct4) for cumulus cell contamination showed that the oocyte collection protocol based on mechanical denudation of oocytes from cumulus cells (without the use of enzymes) and careful visual inspection represent a reliable technique because only 1 of 45 pools analyzed needed to be discarded due to the presence of a few methylated Oct4 alleles (indicating contamination by cumulus cell DNA). As shown in Figure 2, Oct4 alleles in the remaining 44 oocyte pools showed unmethylated patterns. Results of the DNA methylation analysis of the imprinted genes Snrpn and Mest are shown in Figures 3 and 4, respectively. An epimutation was defined as an abnormal (loss

IMPRINTING IN OOCYTES RELATED TO CULTURE DURATION

in Figure 5 (left). For the three genes, total mRNA levels were significantly higher at Day 6 than at Day 12 of culture (P , 0.05), consistent with a shutdown in transcription at the late antral follicle stage. After the ovulatory stimulus (MII oocytes at Day 13) and in preovulatory aged oocytes (Day 14), total mRNA levels did not change compared with Day 12.

of) methylation of the entire allele; in contrast, CpG errors were defined as one or a few unmethylated cytosines present on an allele. The numbers of epimutations and CpG errors in the three time conditions tested are summarized in Table 2. For Snrpn, no single epimutation was found in GV oocytes cultured for 12 days (Day 12) as described previously [23]. Shortened in vitro follicle culture (Day 8) and in vitro preovulatory oocyte aging (Day 14) also did not induce epimutations at Snrpn. For Mest, no epimutations were present for the Day 8 and Day 12 conditions. However, 1 of 54 alleles (1.9%) displayed an epimutation for the Day 14 condition, although this did not reach statistical significance compared with Day 12 or Day 8. In contrast to epimutations, alleles with CpG errors were frequent (range, 21.8%–34.5%) for both Snrpn and Mest at Day 8 and Day 12 (Table 2). A high percentage of alleles with CpG errors remained for Mest (35.2%) at Day 14, but a decrease was noted for Snrpn (8.3%) at Day 14, which reached statistical significance compared with Day 8 (P ¼ 0.0018). This is in line with the results for global DNA methylation levels that show no difference with culture time for Mest but demonstrate a slight (although significant) increase from Day 8 (97.3%) to Day 14 (99.2%) for Snrpn (P , 0.01).

Comparison of Total mRNA Expression Levels Between In Vivo and In Vitro Grown Oocytes Expression levels in GV oocytes grown in vitro up to Day 14, in MII oocytes from follicle culture (Day 13), and in eCGstimulated and eCG/hCG-stimulated in vivo grown oocytes from 25-day-old mice are shown in Figure 5 (left and right). There was no significant difference in total mRNA levels between Day 12 or Day 14 oocytes from follicle culture and eCG-stimulated in vivo grown oocytes for any of the genes. Likewise, total mRNA levels were similar in MII oocytes from follicle culture (Day 13) and in in vivo grown eCG/hCGstimulated MII oocytes. Similar to the in vitro condition, total mRNA levels of Dnmt3a, Dnmt3L, and Zfp57 did not change after the ovulatory stimulus in vivo. DISCUSSION

Total mRNA Expression Levels in Oocytes During In Vitro Follicle Culture

Imprinting Establishment at Snrpn and Mest We previously described normal imprinting establishment for Snrpn, Igf2r, Peg3, Mest, and H19 in mouse MII oocytes

Total mRNA expression levels for Dnmt3a, Dnmt3L, and Zfp57 in oocytes from culture Days 6, 12, 13, and 14 are shown 5

Article 130

Downloaded from www.biolreprod.org.

FIG. 3. The DNA methylation of Snrpn in GV oocytes after 8 (left), 12 (center), or 14 (right) days of in vitro follicle culture. Each line represents an individual allele, and each circle represents a CpG site within the differentially methylated region analyzed; filled circles, methylated cytosines; open circles, unmethylated cytosines; and missing circles, CpG sites whose methylation status could not be determined.

ANCKAERT ET AL.

These data show that shortening of the in vitro growth phase compared with the in vivo situation does not result in detectable epimutations. At Day 8 of culture, GV oocytes are in the (early) antral follicle phase and have acquired the imprinting establishment patterns expected from mouse studies with oocytes derived from in vivo grown antral follicles [16, 17]. These data provide reassurance for IVM programs in human fertility clinics where oocytes harvested from small antral follicles at Day 7 of the cycle are being abrogated by nearly 1 wk in their final growth phase. However, for preovulatory intrafollicular oocyte aging (Day 14) in culture, it was found that 1 of 54 Mest alleles showed an epimutation, although this did not reach statistical significance compared with Day 8 and Day 12. Besides epimutations, the number of alleles with one or a few CpG methylation errors was determined, although (in contrast to epimutations) there is no evidence that these CpG errors might affect gene expression after fertilization. The CpG errors were present in up to one third of Snrpn and Mest alleles at Day 8 and Day 12. At Day 14, the frequency of alleles with CpG errors remained unchanged at 35.2% for Mest but decreased significantly to 8.3% for Snrpn. This is in line with global DNA methylation levels showing no difference with culture time for Mest but a slight (although significant) increase from Day 8 (97.3%) to Day 14 (99.2%) for Snrpn. These CpG errors are likely without functional implications, but imprinted gene expression investigations in mouse blastocysts should be performed to assess whether these CpG errors have the potential to influence gene expression and, if so, to what (presumably low) level. A strength of the present study is the inclusion of a marker for cumulus cell contamination in the oocyte pools (Oct4) as

derived from a 12-day in vitro follicle culture system [23–25]. In this study, we wanted to assess the effect on oocyte imprinting establishment of shortened in vitro follicle culture (Day 8) compared with the control in vitro follicle system of 12 days (Day 12). The study of Day 8 is relevant as a model for human IVM where COCs from small antral follicles of 6 to 12 mm are used to be in vitro matured. Cultured mouse Day 8 oocytes have reached 95% of their final oocyte diameter, 90% have reached the surrounded nucleolus stage, most oocytes have reached transcriptional silencing, and 60% of these oocytes reach the MII stage after an hCG stimulus in vitro [27, 29]. Human GV oocytes collected for IVM from follicles with a maximal diameter less than 6 mm had a mean diameter of 109 lm, corresponding to 95% of the final oocyte diameter of GV oocytes from a follicle cohort of at least 15 mm in diameter [39]. Furthermore, 50% of human oocytes from follicles less than 6 mm reach the MII stage after IVM [39]. The effect of a prolonged culture of 14 days as a model for preovulatory (intrafollicular) oocyte ‘‘aging’’ was also studied. This condition may occur during ART treatments when ovarian stimulation may lead to having oocytes in transcriptional arrest in follicles exceeding the usual diameters at which the ovulation trigger is given. We found no epimutations (defined as a loss of methylation of the entire allele) for the control Day 12 condition as described previously [23]. Likewise, the Day 8 condition was not associated with Snrpn or Mest epimutations in oocytes, although a duration of 8 days of follicle culture was previously shown to result in lower oocyte maturation rates but a higher blastocyst/MII oocyte yield compared with 12 days of culture [27]. 6

Article 130

Downloaded from www.biolreprod.org.

FIG. 4. The DNA methylation of Mest in GV oocytes after 8 (left), 12 (center), or 14 (right) days of in vitro follicle culture. Each line represents an individual allele, and each circle represents a CpG site within the differentially methylated region analyzed; filled circles, methylated cytosines; open circles, unmethylated cytosines; and missing circles, CpG sites whose methylation status could not be determined.

IMPRINTING IN OOCYTES RELATED TO CULTURE DURATION

to cumulus cell DNA contamination despite repeated wash steps. This calls for caution in interpreting those investigations using enzymatic denudation of oocytes that omitted the inclusion of marker genes for cumulus cell contamination. Total mRNA Levels of Oocyte Genes for Imprinting Establishment In Vitro and In Vivo

FIG. 5. Relative expression levels (mean 6 SEM) for total mRNA in oocytes derived from in vitro follicle culture (at Days 6, 12–13, and 14; grey bars) and in vivo grown eCG-stimulated and eCG/hCG-stimulated oocytes from 25-day-old prepubertal mice (white bars). Different small letters indicate significant differences within in vitro follicle culture conditions; different capital letters indicate significant differences between eCG-stimulated and eCG/hCG-stimulated in vivo grown oocytes.

described previously [35], preventing any bias caused by cumulus cell alleles that might be wrongfully interpreted as abnormally methylated. It has been the experience in our laboratory that the use of enzymatic methods (hyaluronidase or Tyrode acidic solution) to denude human or mouse oocytes led 7

Article 130

Downloaded from www.biolreprod.org.

Besides analyzing imprinting establishment in oocytes, the effect of follicle culture on the expression of genes necessary for imprinting establishment (Dnmt3a, Dnmt3L, and Zfp57) in oocytes was examined. Total transcript levels for Zfp57, Dnmt3L, and Dnmt3a were decreasing in oocytes from Day 6 to Day 12 of culture, consistent with a general shutdown in transcription in oocytes at the late antral follicle stage. During the oocyte growth phase, RNAs and proteins are accumulated that are essential for development into the blastocyst stage. However, once GV oocytes in late antral follicles reach their maximal diameter, transcriptional activity is shut down [40, 41]. A decrease in transcript levels in fully grown GV oocytes has been described for developmentally important genes such as Gdf-9, Bmp-15, and Mater in the current mouse follicle culture system [29], suggesting that the cessation in transcriptional activity occurring before oocyte maturation remains unaltered in the present model. At culture Day 12, total transcript levels in fully grown GV oocytes were similar to those among in vivo grown oocytes from prepubertal mice, again suggesting normal regulation of transcript abundance in this model. Total transcript levels of Gdf-9, Mater, Npm-2, Zar1, and Fgf-8 in fully grown GV oocytes obtained at Day 12 in the same model also featured total transcript levels comparable to those of in vivo grown oocytes; nevertheless, Bmp-15 had lower transcript levels compared with those of in vivo grown oocytes from adult animals [29]. A study by Sa´nchez et al. [29] demonstrated that the duration of culture influenced transcript levels, with the lowest transcript levels occurring when extending culture to Day 14. For the genes tested in the present study, however, no decrease in Dnmt3a, Dnmt3L, and Zfp57 transcript levels was measured at Day 14 of culture. It is well established that during meiotic resumption many transcripts are degraded in the mouse oocyte [40, 41] and that the degradation of transcripts during meiotic maturation is a selective rather than a promiscuous process [42]. Dnmt3a, Dnmt3L, and Zfp57 expression levels did not change after giving the ovulatory trigger in vitro or in vivo. Using the same 12-day follicle culture system, it was previously described that total mRNA levels of Mater were significantly decreased after in vitro and in vivo (in 22-day-old prepubertal mice) maturation, while Gdf-9 and Bmp-15 levels remained stable both in vitro and in vivo [29]. In the present study, total transcript levels of the genes remained stable during meiotic progression in vivo. Our findings corroborate the data by Su et al. [42] showing that transcription degradation during meiotic maturation is a selective rather than a general process. Total transcript levels were similar in MII oocytes from follicle culture and in vivo grown MII oocytes from 25-day-old mice for all genes. Collectively, these results indicate that follicle culture induces no major alterations in oocyte total transcript levels for Zfp57, Dnmt3L, and Dnmt3a. In conclusion, abrogation of a significant part of the final follicle growth process in vitro (when oocytes have reached 95% of their final diameter) does not induce oocyte epimutations at the imprinted Snrpn and Mest genes. In contrast, extension of oocyte residence in large unluteinized follicles in vitro was associated with a low level (1 of 54 alleles) of epimutations for Mest but not for Snrpn. Oocyte total

ANCKAERT ET AL.

ACKNOWLEDGMENT The authors are grateful to Professor Thomas Haaf and Nady El Hajj from the Institute of Human Genetics, Wu¨rzburg University, Wu¨rzburg, Germany, for their help with the limiting-dilution bisulphite sequencing assay design. Sandra Wathlet is acknowledged for her help with the mRNA analysis.

REFERENCES 1. Smitz JEJ, Thompson JG, Gilchrist RB. The promise of in vitro maturation in assisted reproduction and fertility. Semin Reprod Med 2011; 29:24–37. 2. Xu M, Barrett SL, West-Farrell E, Kondapalli LA, Kiesewetter SE, Shea LD, Woodruff TK. In vitro grown human ovarian follicles from cancer patients support oocyte growth. Hum Reprod 2009; 24:2531–2540. 3. Smitz J, Dolmans MM, Donnez J, Fortune JE, Hovatta O, Jewgenow K, Picton HM, Plancha C, Shea LD, Stouffer RL, Telfer EE, Woodruff TK, et al. Current achievements and future research directions in ovarian tissue culture, in vitro follicle development and transplantation: implications for fertility preservation. Hum Reprod Update 2010; 16:395–414. 4. Denomme MM, Mann MR. Genomic imprints as a model for the analysis of epigenetic stability during assisted reproductive technologies. Reproduction 2012; 144:393–409. 5. DeBaun MR, Niemitz EL, Feinberg AP. Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 2003; 72:156–160. 6. Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, Le Bouc Y. In vitro fertilization may increase the risk of Beckwith-Wiedemann syndrome related to the abnormal imprinting of the KCN1OT gene. Am J Hum Genet 2003; 72:1338–1341. 7. Maher ER, Afnan M, Barratt CL. Epigenetic risks related to assisted reproductive technologies: epigenetics, imprinting, ART and icebergs? Hum Reprod 2003; 18:2508–2511. 8. Halliday J, Oke K, Breheny S, Algar E, Amor DJ. Beckwith-Wiedemann syndrome and IVF: a case-control study. Am J Hum Genet 2004; 75: 526–528. 9. Chang AS, Moley KH, Wangler M, Feinberg AP, Debaun MR. Association between Beckwith-Wiedemann syndrome and assisted reproductive technology: a case series of 19 patients. Fertil Steril 2005; 83:349–354. 10. Rossignol S, Steunou V, Chalas C, Kerjean A, Rigolet M, ViegasPequignot E, Jouannet P, Le Bouc Y, Gicquel C. The epigenetic imprinting defect of patients with Beckwith-Wiedemann syndrome born after assisted reproductive technology is not restricted to the 11p15 region. J Med Genet 2006; 43:902–907. 11. Sutcliffe AG, Peters CJ, Bowdin S, Temple K, Reardon W, Wilson L, Clayton-Smith J, Brueton LA, Bannister W, Maher ER. Assisted reproductive therapies and imprinting disorders: a preliminary British survey. Hum Reprod 2006; 21:1009–1011. 12. Bowdin S, Brueton L, Allen C, Harrison R, Kirby G, Maher ER, Afnan M, Kirkman-Brown J, Barratt C, Reardon W. A survey of assisted reproductive technology births and imprinting disorders. Hum Reprod 2007; 22:3237–3240. 13. Gomes MV, Gomes CC, Pinto W Jr, Ramos ES. Methylation pattern at the KvDMR in a child with Beckwith-Wiedemann syndrome conceived by ICSI. Am J Med Genet 2007; 143:625–629. 14. Reik W, Walter J. Genomic imprinting: parental influence on the genome. Nat Rev Genet 2001; 2:21–32. 15. Li E. Chromatin modification and epigenetic reprogramming in mammalian development. Nat Rev Genet 2002; 3:662–673. 16. Lucifero D, Mann MR, Bartolomei MS, Trasler JM. Gene-specific timing and epigenetic memory in oocyte imprinting. Hum Mol Genet 2004; 13: 839–849. 17. Hiura H, Obata Y, Komiyama J, Shirai M, Kono T. Oocyte growthdependent progression of maternal imprinting in mice. Genes Cells 2006; 11:353–361. 18. Howell CY, Bestor TH, Ding F, Latham KE, Mertineit C, Trasler JM, Chaillet JR. Genomic imprinting disrupted by a maternal effect mutation in the Dnmt1 gene. Cell 2001; 104:829–838.

8

Article 130

Downloaded from www.biolreprod.org.

19. Nakamura T, Arai Y, Umehara H, Masuhara M, Kimura T, Taniguchi H, Sekimoto T, Ikawa M, Yoneda Y, Okabe M, Tanaka S, Shiota K, et al. PGC7/Stella protects against DNA demethylation in early embryogenesis. Nat Cell Biol 2007; 9:64–71. 20. Reese KJ, Lin S, Verona RI, Schultz RM, Bartolomei MS. Maintenance of paternal methylation and repression of the imprinted H19 gene requires MBD3. PLoS Genet 2007; 3:e137. 21. Li X, Ito M, Zhou F, Youngson N, Zuo X, Leder P, Ferguson-Smith AC. A maternal-zygotic effect gene, Zfp57, maintains both maternal and paternal imprints. Dev Cell 2008; 15:547–557. 22. Anckaert E, De Rycke M, Smitz J. Culture of oocytes and risk of imprinting defects. Hum Reprod Update 2013; 19:52–66. 23. Anckaert E, Adriaenssens T, Romero S, Dremier S, Smitz J. Unaltered imprinting establishment of key imprinted genes in mouse oocytes after in vitro follicle culture under variable follicle-stimulating hormone exposure. Int J Dev Biol 2009; 53:541–548. 24. Anckaert E, Adriaenssens T, Romero S, Smitz J. Ammonium accumulation and use of mineral oil overlay do not alter imprinting establishment at three key imprinted genes in mouse oocytes grown and matured in a long-term follicle culture. Biol Reprod 2009; 81:666–673. 25. Anckaert E, Romero S, Adriaenssens T, Smitz J. Effects of low methyl donor levels in culture medium during mouse follicle culture on oocyte imprinting establishment. Biol Reprod 2010; 83:377–386. 26. Cortvrindt R, Smitz J. Follicle culture in reproductive toxicology: a tool for in-vitro testing of ovarian function. Hum Reprod Update 2002; 8:1–12. 27. Segers I, Adriaenssens T, Ozturk E, Smitz J. Acquisition and loss of oocyte meiotic and developmental competence during in vitro antral follicle growth in mouse. Fertil Steril 2010; 93:2695–2700. 28. Cortvrindt RG, Hu Y, Liu J, Smitz JE. Timed analysis of the nuclear maturation of oocytes in early preantral mouse follicle culture supplemented with recombinant gonadotropin. Fertil Steril 1998; 70:1114–1125. 29. Sa´nchez F, Adriaenssens T, Romero S, Smitz J. Quantification of oocytespecific transcripts in follicle-enclosed oocytes during antral development and maturation in vitro. Mol Hum Reprod 2009; 15:539–550. 30. Imamura T, Kerjean A, Heams T, Kupiec JJ, Thenevin C, Pa`ldi A. Dynamic CpG and non-CpG methylation of the Peg1/Mest gene in the mouse oocyte and preimplantation embryo. J Biol Chem 2005; 280: 20171–20175. 31. Liang XW, Zhu JQ, Miao YL, Liu JH, Wei L, Lu SS, Hou Y, Schatten H, Lu KH, Sun QY. Loss of methylation imprint of Snrpn in postovulatory aging mouse oocyte. Biochem Biophys Res Commun 2008; 371:16–21. 32. Bourc’his D, Xu GL, Lin CS, Bollman B, Bestor TH. Dnmt3L and the establishment of maternal genomic imprints. Science 2001; 294: 2536–2539. 33. Hata K, Okano M, Lei H, Li E. Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice. Development 2002; 129:1983–1993. 34. Kaneda M, Okano M, Hata K, Sado T, Tsujimoto N, Li E, Sasaki H. Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 2004; 429:900–903. 35. Trapphoff T, El Hajj N, Zechner U, Haaf T, Eichenlaub-Ritter U. DNA integrity, growth pattern, spindle formation, chromosomal constitution and imprinting patterns of mouse oocytes from vitrified pre-antral follicles. Hum Reprod 2010; 25:3025–3042. 36. Han W, Cauchi S, Herman JG, Spivack SD. DNA methylation mapping by tag-modified bisulfite genomic sequencing. Anal Biochem 2006; 355: 50–61. 37. Heinzmann J, Hansmann T, Herrmann D, Wrenzycki C, Zechner U, Haaf T, Niemann H. Epigenetic profile of developmentally important genes in bovine oocytes. Mol Reprod Dev 2011; 78:188–201. 38. Warnecke PM, Stirzaker C, Song J, Grunau C, Melki JR, Clark SJ. Identification and resolution of artifacts in bisulfite sequencing. Methods 2002; 27:101–107. 39. Guzman L, Ortega-Hrepich C, Albuz FK, Verheyen G, Devroey P, Smitz J, De Vos M. Developmental capacity of in vitro-matured human oocytes retrieved from polycystic ovary syndrome ovaries containing no follicles larger than 6 mm. Fertil Steril 2012; 98:503–507. 40. Bachvarova R. Gene expression during oogenesis and oocyte development in mammals. Dev Biol (N Y 1985) 1985; 1:453–524. 41. De La Fuente R, Viveiros MM, Burns KH, Adashi EY, Matzuk MM, Eppig JJ. Major chromatin remodeling in the germinal vesicle (GV) of mammalian oocytes is dispensable for global transcriptional silencing but required for centromeric heterochromatin function. Dev Biol 2004; 275: 447–458. 42. Su YQ, Sugiura K, Woo Y, Wigglesworth K, Kamdar S, Affourtit J, Eppig JJ. Selective degradation of transcripts during meiotic maturation of mouse oocytes. Dev Biol 2007; 302:104–117.

mRNA levels for genes involved in imprinting establishment during the in vitro follicle culture proved the timely shutdown in transcription at the antral follicle stage, and their total mRNA levels were comparable to those in vivo. Future experiments on blastocysts derived from in vitro cultured oocytes will be performed to determine whether oocyte culture might interfere with imprinting maintenance after fertilization.