The bovine IGF2 gene is differentially methylatedin oocyte and sperm DNA
Claudia Gebert a1 Christine Wrenzycki a Doris Herrmann a Daniela Groumlger b Richard Reinhardt bPetra Hajkova c Andrea Lucas-Hahn a Joseph Carnwath a Hans Lehrach b Heiner Niemann a
a Department of Biotechnology Institute for Animal Breeding Hoeltystrasse 10 D-31535 Neustadt-Mariensee Germanyb Max-Planck-Institute for Molecular Genetics Ihnestrasse 63-73 D-14195 Berlin-Dahlem Germany
c Wellcome TrustCancer Research UK Institute of Cancer and Developmental Biology and Department of Physiology University of CambridgeTennis Court Road Cambridge CB2 1QR UK
Received 13 December 2005 accepted 17 March 2006Available online 27 April 2006
Abstract
The insulin-like growth factor 2 gene (IGF2) encodes an essential growth factor and is imprinted in various mammalian species Differentiallymethylated regions (DMRs) are often located within CpG islands and are critically involved in the regulation of monoallelic Igf2 expression in themouse Only partial sequence information is available for the bovine IGF2 gene and no DMR has currently been identified The goal of this studywas to identify a DMR within the bovine IGF2 gene as a prerequisite for further studies on gene-specific methylation patterns duringpreimplantation development Here we describe the sequence analysis of a CpG-rich DNA fragment from the 5prime untranslated region spanningexons and introns 4 and 5 and the identification of a previously unknown DMR in exon 10 of the bovine IGF2 gene Bisulfite analysis revealedthat this DMR is differentially methylated in mature oocytes and sperm The identification of an intragenic DMR within a developmentallyimportant gene such as the bovine IGF2 gene provides a useful tool to evaluate the methylation patterns of embryos derived in vivo and in vitroOur study is the first report of a differentially methylated region in a bovine imprinted gene discovered by the analysis of female and malegametescopy 2006 Elsevier Inc All rights reserved
Keywords IGF2 Differentially methylated region Cattle
The insulin-like growth factor 2 (IGF2) gene encodes agrowth factor that plays a crucial role in tissue differentiationfetal growth [12] and placental development [3] It maps tochromosome 7 in the mouse [4] to chromosome 11 in humans[56] to chromosome 2 in pigs [78] to chromosome 21 in sheep[9] and to chromosome 29 in cattle [10] Molecular organizationand regulation of this gene are similar in these species In large
Sequence data from this article have been deposited with the EMBL DataLibrary under Accession No AJ890138 (exons and introns 4 and 5 of the bovineIGF2 gene) Corresponding author Fax +49 5034 871101E-mail address niemanntzvfalde (H Niemann)
1 Present address Laboratory of Mammalian Genes and DevelopmentBuilding 6B Room 2B206 NICHDNIH 9000 Rockville Pike BethesdaMD 20892 USA
0888-7543$ - see front matter copy 2006 Elsevier Inc All rights reserveddoi101016jygeno200603011
mammals the gene consists of 10 exons [11ndash14] The murineIgf2 gene differs from that and contains six functional exons andtwo nonfunctional ldquopseudo-exonsrdquo [4] Various IGF2 transcriptsincluding different exon sequences at the 5prime end but identicalcoding exons are produced by initiating expression from differentpromoters and a variety of IGF2 isoforms are subsequentlyformed by alternative splicing [41112] In total four differentpromoters were found at homologous chromosomal locations inthe IGF2 genes of human pig and ruminants This permits theproduction of unique isoforms of IGF2mRNA in each tissue andduring each stage of development [11121415]
The IGF2 gene has been found to be imprinted in all studiedspecies such as mouse [16] human [17ndash19] sheep [2021] pig[7] and cattle [22] Correct imprinting patterns are crucial fornormal mammalian development postnatal behavior and braindevelopment or function [23] Regulation of imprinted genes
223C Gebert et al Genomics 88 (2006) 222ndash229
correlates well with DNA methylation [2324] Mammaliangenomic DNA is methylated mainly at cytosinendashguanine (CpG)dinucleotides However CpG islands that represent regionswith a high proportion of CpGrsquos are mostly nonmethylatedexcept in the case of imprinted genes [25ndash28]
A characteristic feature of imprinted genes is the methylationof CpGrsquos on only one of the parental alleles The allele-specificmethylation is typical for differentially methylated regions(DMRs) and the methylation pattern is implicated in theregulation of imprinted gene expression [29ndash31] Parent-of-origin-specific DMR methylation is normally established duringgerm cell development and in the early embryo [23] Howeverrecent work indicates that allele-specific methylation can also beestablished in somatic tissues without going through the germline [32]
Three DMRs have been documented to be involved in theregulation of Igf2 expression in the mouse (Fig 3) DMR0 isplacenta-specific and located upstream of DMR1 in theintergenic region between the insulin (Ins) and the insulin-likegrowth factor 2 (Igf2) genes DMR0 is transcribed in bothdirections and both parental alleles are highly methylated [29]The paternally methylated DMR1 lies 3 kb upstream of the Igf2promoter P1 and represses maternal Igf2 transcription [3033]DMR2 is methylated on the paternal allele within the last exon(exon 6 in the mouse) of the Igf2 gene [34] and contains a 54-bpcore region that enhances Igf2 transcription [31]
Fig 1 Comparative representation of the IGF2 gene in different mammalian speciesand intragenic differentially methylated regions as DMR below a rectangleΨ1 andΨspecies
A first draft of the bovine genome has recently beenpublished (wwwncbinihgovGenbank) However precisesequence information for the bovine IGF2 gene is not yetavailable Our study focused on sequencing of critical portionsof the bovine IGF2 gene in which DMRs were expectedIdentification of these DMRs would facilitate ongoing studiesof the IGF2-specific methylation patterns during preimplanta-tion development Here we report the first study in which DNAmethylation of a bovine imprinted gene was analyzed to identifythe presence of an intragenic DMR by using DNA from maleand female germ cells as starting material
Results
Due to the lack of DNA sequence information for thebovine IGF2 gene in the GenBank database we used theovine IGF2 gene [12] to identify CpG-rich domains usingthe computer program CpGwin [35] This approach wasbased on the knowledge that DMRs are predominantlyfound within andor in the vicinity of CpG islands [3036]A DNA fragment within the 5prime untranslated region of theovine IGF2 gene was utilized to sequence the bovinecounterpart (Figs 1 and 2) The ovine sequence spanningexons 4 to 6 contained a high content of CpG dinucleotides(65ndash72) and fulfilled all the criteria of a CpG island [25]Two sets of primers (Table 1) were designed from the ovine
Numbered rectangles represent exons promoter regions are indicated as P1ndashP42 represent pseudo-exons 1 and 2 in the mouse These exons are active in other
Fig 2 The bovine IGF2 sequence of exons and introns 4 and 5 was obtained using primers for the first round of PCR that had been designed from the ovine IGF2 genesequence available in the GenBank database (U00664) Numbered rectangles represent exons primer pairs used for PCR amplification are indicated as horizontalarrowsWithin the ovine sequence the primer pair spanning exons 4 and 5was primer oIGF2-41 the primer pair spanning exons 5 and 6was primer oIGf2-51 To obtainthe complete sequence of intron 4 PCR amplification with a bovine-specific primer pair primer bIGF2-42 became necessary Introns 4 and 5 each contain a CpG island
224 C Gebert et al Genomics 88 (2006) 222ndash229
IGF2 gene encompassing introns 4 (primer pair oIGF2-4142) and 5 (primer pair oIGF2-51) Sequences of the twoindividually amplified PCR products resulted in a 1735-bpfragment from bovine kidney DNA (EMBL AJ890138) thatwas aligned against the ovine gene sequence (GenBankU00664 1041ndash3073 bp) (Fig 3) Bioinformatic analysisusing BLAST revealed this sequence representing exons 4and 5 and introns 4 and 5 of the bovine IGF2 gene Exons4 and 5 contained 96 and 98 and introns 4 and 5 95 and92 homologous sequence identity respectively to thecorresponding ovine sequences Alignment against humanand porcine exons and introns 4 and 4b revealed sequencehomologies of sim90 in these two species
A promoter region was identified within the bovine intron 5sequence by the computer program Neural Network PromoterPrediction [37] This region contains a TATA box (5prime-TATAA-
Fig 3 Representative gel photograph of PCR-amplified bovine IGF2 sequences Pri[12] to identify exons and introns 4 and 5 of the homologous bovine gene (lanes 1specific primer pairs (bIGF2) were cut out for sequencing and complete identification3 6 and 9
3prime) within intron 5 (1536ndash1586 bp) and a preceding basicrecognition sequence (5prime-GGGGCGGGGC-3prime) of the SP1transcription factor [38] The SP1 site (1459ndash1468 bp) islocated 67 bp upstream of the TATA box Furthermore aCCAAT box (5prime-CCATTGGCGCGGGC-3prime 1486ndash1499 bp)was identified between the SP1 site and the TATA box Theseelements are recognition sites for the transcriptional machineryand are also present in the homologous ovine intron 5 sequence(GenBank U00664 SP1 site 2769ndash2778 bp CCAAT box2794ndash2808 bp promoter sequence 2841ndash2890 bp [12]) and inthe human P3 promoter region [39]
To identify DMRs within the bovine IGF2 gene weanalyzed 236 bp of exon 4 and intron 4 (71ndash307 bp) and anupstream portion of intron 5 (915ndash1326 bp) of the sequenced1735-bp fragment (EMBL AJ890138) The size and location ofthe sequences analyzed by bisulfite sequencing were defined by
mers oIGF2 were designed from the sequenced and published ovine IGF2 gene2 7 8) The highest gel bands obtained from PCR amplification with bovine-of the bovine intron 4 sequence (lanes 4 5) Negative controls are shown in lanes
Table 1Primers used for PCR
Primer Primer sequence Annealing (degC) PCR supplement Accession No
oIGF2-51 F GTGAGCTCGGCCATTCAGGTAGGAT 62 5 M betain U00664R CGGGCGTTGAGGTAGACGAAGAGGA 40 cycles F2132ndash2156 R3106ndash3082
Bisulfite-41 F ATTGGGTATTGTTTTTAGTTTTT 491 F61ndash83 R438ndash458R TTATAATCTTTACACAAAACA 40 cycles
Bisulfite-42 F GTTTTTAGTTTTTTTTAAATTTG 474 F71ndash93 R288ndash307R AATAAATTTAAAAACCAAAC 35 cycles
Bisulfite-51 F TGGTTTTTTTAGATTTTTAAATGAT 51 F688ndash712 R465ndash490R AAAATAAAAACAAAAATCCTCAAAA 40 cycles
Bisulfite-52 F GTTTTATGTTTGGTTTTTAGT 502 F44ndash64 R434ndash454R TCTAAAATTACTATTCCAAAA 35 cycles
Bisulfite-101 F TGGGTAAGTTTTTTTAATATGATATT 496 X53553R TTTAAAACCAATTAATTTTATACATT 40 cycles F243ndash268 R672ndash697
Bisulfite-102 F TAATATGATATTTGGAAGTAGT 491 X53553R ACATTTTTAAAAATATTATTCT 35 cycles F257ndash278 R655ndash676
225C Gebert et al Genomics 88 (2006) 222ndash229
the specific primer positions excluding any CpGrsquos Thesequence of exon 10 had been published (GenBank X53553[40]) This is the last exon of the bovine IGF2 gene and it wasincluded in the bisulfite sequencing analysis because the lastexon of the murine Igf2 gene contains a well-characterizedDMR and an important enhancer element [3341] In the presentstudy (Figs 4 and 5) the methylation level of 27 CpGrsquos in exon4 and intron 4 was low both in DNA extracted from bovineoocytes (39 plusmn 07) and in sperm (19 plusmn 03) There was alsono difference in methylation between oocytes (28 plusmn 09) andsperm (39 plusmn 2) within intron 5 which contains the promoterregion (Fig 4)
By contrast a significant difference (P le 005) betweenDNA extracted from oocytes and sperm was detected in themethylation level of exon 10 This final exon of the bovineIGF2 gene was sixfold more methylated in sperm(99 plusmn 03) than in oocytes (16 plusmn 08 Fig 4) clearlyindicating the presence of an intragenic differentiallymethylated region The methylation difference included all27 CpGrsquos between nucleotides 257 and 676 bp of the codingregion (GenBank X53553) Importantly the 54-bp coreregion of the intragenic DMR2 in the mouse [3134](GenBank U71085 24416ndash24470 bp) is conserved withinthe homologous intragenic DMR of the bovine IGF2 gene(GenBank X53553 281ndash334 bp)
Discussion
The IGF2 gene is subject to imprinting in several mammalianspecies and was selected for this study because of its crucial rolein growth and development [1316] We sequenced putativelyimprinting-associated portions of the gene since only thesequence of exon 2 and the translated region of the bovineIGF2 gene were available in the GenBank database (AY23743X53553) Although the H19 DMR located in the intergenicregion downstream of the mouse Igf2 gene is known as the majorcis-acting element regulating expression of the two genes it was
excluded from the analysis Our approach sequencing parts ofthe bovine IGF2 gene with primer pairs specific to the ovineIGF2 gene was dependent on a high portion of homologoussequences between these two species [4243] This studyfocused on the identification of intragenic DMRs because weanticipated higher sequence conservation within the gene bodythan in an intergenic region
Analysis of the already published sequence of the ovineIGF2 gene [12] revealed a CpG island (data not shown) withinthe untranslated region based on the definition of Gardiner-Garden and Frommer [25] The presence of a CpG island withinthe ovine IGF2 gene was also confirmed in the homologousbovine 1735-bp fragment Homologies of sim90 were detectedfor this fragment of the bovine IGF2 gene with phylogeneticallydistant species such as human and pig Thus the molecularstructure of the IGF2 gene is highly conserved betweenmammals despite minor modifications such as the twopseudo-exons in the mouse [4] or the lack of exon 2 in sheep[12] A recently published study on promoter-specific expres-sion patterns in the bovine IGF2 gene confirms our results ofCpG-rich sequence identities and high sequence homologiesbetween different mammalian species [15] In this studypromoter 3 has been mapped downstream of exon 5 bycomparison of the corresponding sequence of the human andpig The analysis of promoter-specific elements presented inthis study verifies in detail the location of this promoter withinthe bovine IGF2 gene
Knowledge of the complex regulation of the IGF2 genestems predominantly from studies in the mouse The murineIgf2 gene is reciprocally imprinted to the downstream-locatedH19 gene and both genes share common enhancers [2344]Unlike the mouse gene-specific information on bovinemolecular structure and imprinting status has only recentlyemerged [152245ndash47] From the bovine insulin and IGFfamily the insulin-like growth factor 2 receptor (IGF2R) geneand the IGF2 gene were identified as imprinted genes [2245]Both studies used single-nucleotide polymorphisms (SNPs) to
Fig 4 Methylation patterns of exon and intron 4 intron 5 and exon 10 in in vitro-matured bovine oocytes () and frozenthawed sperm () Black boxes indicateexons each line represents one individual bacterial clone and each circle one single CpG dinucleotide Open circles show nonmethylated CpGrsquos and black circlesmethylated CpGrsquos
226 C Gebert et al Genomics 88 (2006) 222ndash229
discriminate between the two parental alleles In the presentstudy however we overcame the lack of known geneticmarkers such as SNPs for determination of differentiallymethylated regions within the bovine IGF2 gene by usingfemale and male gametes as the source for DNA A significantdifference in the methylation level was detected in the finalexon of the gene between the two gametes The extraordi-narily high methylation level of sperm DNA and the lowmethylation level in oocytes clearly indicate the presence of an
Fig 5 Representative gel photograph of PCR analyses using bisulfite-treated DNAisolated from in vitro-matured bovine oocytes and frozenthawed sperm PCR producinto Escherichia coli cells prior to sequencing (Lanes 1ndash3) Exon and intron 4 oocontrol (Lanes 7ndash9) Exon 10 oocytes sperm negative control
intragenic DMR and underline the validity of the presentmethodological approach Our finding supports the notion thatthe IGF2 gene is well conserved among mammals as thehomologous DMR had also been found in the last exon ofmouse and human in which the paternal allele is methylated aswell [3448] We propose that this DMR corresponds to theDMR2 in the mouse which becomes reprogrammed afterfertilization but paternal hypermethylation is restored laterduring development [49] In contrast to the mouse
to identify intragenic DMRs within the bovine IGF2 gene Genomic DNAwasts were cloned into the pGEMT-Easy vector system (Promega) and transformedcytes sperm negative control (Lanes 4ndash6) Intron 5 oocytes sperm negative
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
223C Gebert et al Genomics 88 (2006) 222ndash229
correlates well with DNA methylation [2324] Mammaliangenomic DNA is methylated mainly at cytosinendashguanine (CpG)dinucleotides However CpG islands that represent regionswith a high proportion of CpGrsquos are mostly nonmethylatedexcept in the case of imprinted genes [25ndash28]
A characteristic feature of imprinted genes is the methylationof CpGrsquos on only one of the parental alleles The allele-specificmethylation is typical for differentially methylated regions(DMRs) and the methylation pattern is implicated in theregulation of imprinted gene expression [29ndash31] Parent-of-origin-specific DMR methylation is normally established duringgerm cell development and in the early embryo [23] Howeverrecent work indicates that allele-specific methylation can also beestablished in somatic tissues without going through the germline [32]
Three DMRs have been documented to be involved in theregulation of Igf2 expression in the mouse (Fig 3) DMR0 isplacenta-specific and located upstream of DMR1 in theintergenic region between the insulin (Ins) and the insulin-likegrowth factor 2 (Igf2) genes DMR0 is transcribed in bothdirections and both parental alleles are highly methylated [29]The paternally methylated DMR1 lies 3 kb upstream of the Igf2promoter P1 and represses maternal Igf2 transcription [3033]DMR2 is methylated on the paternal allele within the last exon(exon 6 in the mouse) of the Igf2 gene [34] and contains a 54-bpcore region that enhances Igf2 transcription [31]
Fig 1 Comparative representation of the IGF2 gene in different mammalian speciesand intragenic differentially methylated regions as DMR below a rectangleΨ1 andΨspecies
A first draft of the bovine genome has recently beenpublished (wwwncbinihgovGenbank) However precisesequence information for the bovine IGF2 gene is not yetavailable Our study focused on sequencing of critical portionsof the bovine IGF2 gene in which DMRs were expectedIdentification of these DMRs would facilitate ongoing studiesof the IGF2-specific methylation patterns during preimplanta-tion development Here we report the first study in which DNAmethylation of a bovine imprinted gene was analyzed to identifythe presence of an intragenic DMR by using DNA from maleand female germ cells as starting material
Results
Due to the lack of DNA sequence information for thebovine IGF2 gene in the GenBank database we used theovine IGF2 gene [12] to identify CpG-rich domains usingthe computer program CpGwin [35] This approach wasbased on the knowledge that DMRs are predominantlyfound within andor in the vicinity of CpG islands [3036]A DNA fragment within the 5prime untranslated region of theovine IGF2 gene was utilized to sequence the bovinecounterpart (Figs 1 and 2) The ovine sequence spanningexons 4 to 6 contained a high content of CpG dinucleotides(65ndash72) and fulfilled all the criteria of a CpG island [25]Two sets of primers (Table 1) were designed from the ovine
Numbered rectangles represent exons promoter regions are indicated as P1ndashP42 represent pseudo-exons 1 and 2 in the mouse These exons are active in other
Fig 2 The bovine IGF2 sequence of exons and introns 4 and 5 was obtained using primers for the first round of PCR that had been designed from the ovine IGF2 genesequence available in the GenBank database (U00664) Numbered rectangles represent exons primer pairs used for PCR amplification are indicated as horizontalarrowsWithin the ovine sequence the primer pair spanning exons 4 and 5was primer oIGF2-41 the primer pair spanning exons 5 and 6was primer oIGf2-51 To obtainthe complete sequence of intron 4 PCR amplification with a bovine-specific primer pair primer bIGF2-42 became necessary Introns 4 and 5 each contain a CpG island
224 C Gebert et al Genomics 88 (2006) 222ndash229
IGF2 gene encompassing introns 4 (primer pair oIGF2-4142) and 5 (primer pair oIGF2-51) Sequences of the twoindividually amplified PCR products resulted in a 1735-bpfragment from bovine kidney DNA (EMBL AJ890138) thatwas aligned against the ovine gene sequence (GenBankU00664 1041ndash3073 bp) (Fig 3) Bioinformatic analysisusing BLAST revealed this sequence representing exons 4and 5 and introns 4 and 5 of the bovine IGF2 gene Exons4 and 5 contained 96 and 98 and introns 4 and 5 95 and92 homologous sequence identity respectively to thecorresponding ovine sequences Alignment against humanand porcine exons and introns 4 and 4b revealed sequencehomologies of sim90 in these two species
A promoter region was identified within the bovine intron 5sequence by the computer program Neural Network PromoterPrediction [37] This region contains a TATA box (5prime-TATAA-
Fig 3 Representative gel photograph of PCR-amplified bovine IGF2 sequences Pri[12] to identify exons and introns 4 and 5 of the homologous bovine gene (lanes 1specific primer pairs (bIGF2) were cut out for sequencing and complete identification3 6 and 9
3prime) within intron 5 (1536ndash1586 bp) and a preceding basicrecognition sequence (5prime-GGGGCGGGGC-3prime) of the SP1transcription factor [38] The SP1 site (1459ndash1468 bp) islocated 67 bp upstream of the TATA box Furthermore aCCAAT box (5prime-CCATTGGCGCGGGC-3prime 1486ndash1499 bp)was identified between the SP1 site and the TATA box Theseelements are recognition sites for the transcriptional machineryand are also present in the homologous ovine intron 5 sequence(GenBank U00664 SP1 site 2769ndash2778 bp CCAAT box2794ndash2808 bp promoter sequence 2841ndash2890 bp [12]) and inthe human P3 promoter region [39]
To identify DMRs within the bovine IGF2 gene weanalyzed 236 bp of exon 4 and intron 4 (71ndash307 bp) and anupstream portion of intron 5 (915ndash1326 bp) of the sequenced1735-bp fragment (EMBL AJ890138) The size and location ofthe sequences analyzed by bisulfite sequencing were defined by
mers oIGF2 were designed from the sequenced and published ovine IGF2 gene2 7 8) The highest gel bands obtained from PCR amplification with bovine-of the bovine intron 4 sequence (lanes 4 5) Negative controls are shown in lanes
Table 1Primers used for PCR
Primer Primer sequence Annealing (degC) PCR supplement Accession No
oIGF2-51 F GTGAGCTCGGCCATTCAGGTAGGAT 62 5 M betain U00664R CGGGCGTTGAGGTAGACGAAGAGGA 40 cycles F2132ndash2156 R3106ndash3082
Bisulfite-41 F ATTGGGTATTGTTTTTAGTTTTT 491 F61ndash83 R438ndash458R TTATAATCTTTACACAAAACA 40 cycles
Bisulfite-42 F GTTTTTAGTTTTTTTTAAATTTG 474 F71ndash93 R288ndash307R AATAAATTTAAAAACCAAAC 35 cycles
Bisulfite-51 F TGGTTTTTTTAGATTTTTAAATGAT 51 F688ndash712 R465ndash490R AAAATAAAAACAAAAATCCTCAAAA 40 cycles
Bisulfite-52 F GTTTTATGTTTGGTTTTTAGT 502 F44ndash64 R434ndash454R TCTAAAATTACTATTCCAAAA 35 cycles
Bisulfite-101 F TGGGTAAGTTTTTTTAATATGATATT 496 X53553R TTTAAAACCAATTAATTTTATACATT 40 cycles F243ndash268 R672ndash697
Bisulfite-102 F TAATATGATATTTGGAAGTAGT 491 X53553R ACATTTTTAAAAATATTATTCT 35 cycles F257ndash278 R655ndash676
225C Gebert et al Genomics 88 (2006) 222ndash229
the specific primer positions excluding any CpGrsquos Thesequence of exon 10 had been published (GenBank X53553[40]) This is the last exon of the bovine IGF2 gene and it wasincluded in the bisulfite sequencing analysis because the lastexon of the murine Igf2 gene contains a well-characterizedDMR and an important enhancer element [3341] In the presentstudy (Figs 4 and 5) the methylation level of 27 CpGrsquos in exon4 and intron 4 was low both in DNA extracted from bovineoocytes (39 plusmn 07) and in sperm (19 plusmn 03) There was alsono difference in methylation between oocytes (28 plusmn 09) andsperm (39 plusmn 2) within intron 5 which contains the promoterregion (Fig 4)
By contrast a significant difference (P le 005) betweenDNA extracted from oocytes and sperm was detected in themethylation level of exon 10 This final exon of the bovineIGF2 gene was sixfold more methylated in sperm(99 plusmn 03) than in oocytes (16 plusmn 08 Fig 4) clearlyindicating the presence of an intragenic differentiallymethylated region The methylation difference included all27 CpGrsquos between nucleotides 257 and 676 bp of the codingregion (GenBank X53553) Importantly the 54-bp coreregion of the intragenic DMR2 in the mouse [3134](GenBank U71085 24416ndash24470 bp) is conserved withinthe homologous intragenic DMR of the bovine IGF2 gene(GenBank X53553 281ndash334 bp)
Discussion
The IGF2 gene is subject to imprinting in several mammalianspecies and was selected for this study because of its crucial rolein growth and development [1316] We sequenced putativelyimprinting-associated portions of the gene since only thesequence of exon 2 and the translated region of the bovineIGF2 gene were available in the GenBank database (AY23743X53553) Although the H19 DMR located in the intergenicregion downstream of the mouse Igf2 gene is known as the majorcis-acting element regulating expression of the two genes it was
excluded from the analysis Our approach sequencing parts ofthe bovine IGF2 gene with primer pairs specific to the ovineIGF2 gene was dependent on a high portion of homologoussequences between these two species [4243] This studyfocused on the identification of intragenic DMRs because weanticipated higher sequence conservation within the gene bodythan in an intergenic region
Analysis of the already published sequence of the ovineIGF2 gene [12] revealed a CpG island (data not shown) withinthe untranslated region based on the definition of Gardiner-Garden and Frommer [25] The presence of a CpG island withinthe ovine IGF2 gene was also confirmed in the homologousbovine 1735-bp fragment Homologies of sim90 were detectedfor this fragment of the bovine IGF2 gene with phylogeneticallydistant species such as human and pig Thus the molecularstructure of the IGF2 gene is highly conserved betweenmammals despite minor modifications such as the twopseudo-exons in the mouse [4] or the lack of exon 2 in sheep[12] A recently published study on promoter-specific expres-sion patterns in the bovine IGF2 gene confirms our results ofCpG-rich sequence identities and high sequence homologiesbetween different mammalian species [15] In this studypromoter 3 has been mapped downstream of exon 5 bycomparison of the corresponding sequence of the human andpig The analysis of promoter-specific elements presented inthis study verifies in detail the location of this promoter withinthe bovine IGF2 gene
Knowledge of the complex regulation of the IGF2 genestems predominantly from studies in the mouse The murineIgf2 gene is reciprocally imprinted to the downstream-locatedH19 gene and both genes share common enhancers [2344]Unlike the mouse gene-specific information on bovinemolecular structure and imprinting status has only recentlyemerged [152245ndash47] From the bovine insulin and IGFfamily the insulin-like growth factor 2 receptor (IGF2R) geneand the IGF2 gene were identified as imprinted genes [2245]Both studies used single-nucleotide polymorphisms (SNPs) to
Fig 4 Methylation patterns of exon and intron 4 intron 5 and exon 10 in in vitro-matured bovine oocytes () and frozenthawed sperm () Black boxes indicateexons each line represents one individual bacterial clone and each circle one single CpG dinucleotide Open circles show nonmethylated CpGrsquos and black circlesmethylated CpGrsquos
226 C Gebert et al Genomics 88 (2006) 222ndash229
discriminate between the two parental alleles In the presentstudy however we overcame the lack of known geneticmarkers such as SNPs for determination of differentiallymethylated regions within the bovine IGF2 gene by usingfemale and male gametes as the source for DNA A significantdifference in the methylation level was detected in the finalexon of the gene between the two gametes The extraordi-narily high methylation level of sperm DNA and the lowmethylation level in oocytes clearly indicate the presence of an
Fig 5 Representative gel photograph of PCR analyses using bisulfite-treated DNAisolated from in vitro-matured bovine oocytes and frozenthawed sperm PCR producinto Escherichia coli cells prior to sequencing (Lanes 1ndash3) Exon and intron 4 oocontrol (Lanes 7ndash9) Exon 10 oocytes sperm negative control
intragenic DMR and underline the validity of the presentmethodological approach Our finding supports the notion thatthe IGF2 gene is well conserved among mammals as thehomologous DMR had also been found in the last exon ofmouse and human in which the paternal allele is methylated aswell [3448] We propose that this DMR corresponds to theDMR2 in the mouse which becomes reprogrammed afterfertilization but paternal hypermethylation is restored laterduring development [49] In contrast to the mouse
to identify intragenic DMRs within the bovine IGF2 gene Genomic DNAwasts were cloned into the pGEMT-Easy vector system (Promega) and transformedcytes sperm negative control (Lanes 4ndash6) Intron 5 oocytes sperm negative
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
Fig 2 The bovine IGF2 sequence of exons and introns 4 and 5 was obtained using primers for the first round of PCR that had been designed from the ovine IGF2 genesequence available in the GenBank database (U00664) Numbered rectangles represent exons primer pairs used for PCR amplification are indicated as horizontalarrowsWithin the ovine sequence the primer pair spanning exons 4 and 5was primer oIGF2-41 the primer pair spanning exons 5 and 6was primer oIGf2-51 To obtainthe complete sequence of intron 4 PCR amplification with a bovine-specific primer pair primer bIGF2-42 became necessary Introns 4 and 5 each contain a CpG island
224 C Gebert et al Genomics 88 (2006) 222ndash229
IGF2 gene encompassing introns 4 (primer pair oIGF2-4142) and 5 (primer pair oIGF2-51) Sequences of the twoindividually amplified PCR products resulted in a 1735-bpfragment from bovine kidney DNA (EMBL AJ890138) thatwas aligned against the ovine gene sequence (GenBankU00664 1041ndash3073 bp) (Fig 3) Bioinformatic analysisusing BLAST revealed this sequence representing exons 4and 5 and introns 4 and 5 of the bovine IGF2 gene Exons4 and 5 contained 96 and 98 and introns 4 and 5 95 and92 homologous sequence identity respectively to thecorresponding ovine sequences Alignment against humanand porcine exons and introns 4 and 4b revealed sequencehomologies of sim90 in these two species
A promoter region was identified within the bovine intron 5sequence by the computer program Neural Network PromoterPrediction [37] This region contains a TATA box (5prime-TATAA-
Fig 3 Representative gel photograph of PCR-amplified bovine IGF2 sequences Pri[12] to identify exons and introns 4 and 5 of the homologous bovine gene (lanes 1specific primer pairs (bIGF2) were cut out for sequencing and complete identification3 6 and 9
3prime) within intron 5 (1536ndash1586 bp) and a preceding basicrecognition sequence (5prime-GGGGCGGGGC-3prime) of the SP1transcription factor [38] The SP1 site (1459ndash1468 bp) islocated 67 bp upstream of the TATA box Furthermore aCCAAT box (5prime-CCATTGGCGCGGGC-3prime 1486ndash1499 bp)was identified between the SP1 site and the TATA box Theseelements are recognition sites for the transcriptional machineryand are also present in the homologous ovine intron 5 sequence(GenBank U00664 SP1 site 2769ndash2778 bp CCAAT box2794ndash2808 bp promoter sequence 2841ndash2890 bp [12]) and inthe human P3 promoter region [39]
To identify DMRs within the bovine IGF2 gene weanalyzed 236 bp of exon 4 and intron 4 (71ndash307 bp) and anupstream portion of intron 5 (915ndash1326 bp) of the sequenced1735-bp fragment (EMBL AJ890138) The size and location ofthe sequences analyzed by bisulfite sequencing were defined by
mers oIGF2 were designed from the sequenced and published ovine IGF2 gene2 7 8) The highest gel bands obtained from PCR amplification with bovine-of the bovine intron 4 sequence (lanes 4 5) Negative controls are shown in lanes
Table 1Primers used for PCR
Primer Primer sequence Annealing (degC) PCR supplement Accession No
oIGF2-51 F GTGAGCTCGGCCATTCAGGTAGGAT 62 5 M betain U00664R CGGGCGTTGAGGTAGACGAAGAGGA 40 cycles F2132ndash2156 R3106ndash3082
Bisulfite-41 F ATTGGGTATTGTTTTTAGTTTTT 491 F61ndash83 R438ndash458R TTATAATCTTTACACAAAACA 40 cycles
Bisulfite-42 F GTTTTTAGTTTTTTTTAAATTTG 474 F71ndash93 R288ndash307R AATAAATTTAAAAACCAAAC 35 cycles
Bisulfite-51 F TGGTTTTTTTAGATTTTTAAATGAT 51 F688ndash712 R465ndash490R AAAATAAAAACAAAAATCCTCAAAA 40 cycles
Bisulfite-52 F GTTTTATGTTTGGTTTTTAGT 502 F44ndash64 R434ndash454R TCTAAAATTACTATTCCAAAA 35 cycles
Bisulfite-101 F TGGGTAAGTTTTTTTAATATGATATT 496 X53553R TTTAAAACCAATTAATTTTATACATT 40 cycles F243ndash268 R672ndash697
Bisulfite-102 F TAATATGATATTTGGAAGTAGT 491 X53553R ACATTTTTAAAAATATTATTCT 35 cycles F257ndash278 R655ndash676
225C Gebert et al Genomics 88 (2006) 222ndash229
the specific primer positions excluding any CpGrsquos Thesequence of exon 10 had been published (GenBank X53553[40]) This is the last exon of the bovine IGF2 gene and it wasincluded in the bisulfite sequencing analysis because the lastexon of the murine Igf2 gene contains a well-characterizedDMR and an important enhancer element [3341] In the presentstudy (Figs 4 and 5) the methylation level of 27 CpGrsquos in exon4 and intron 4 was low both in DNA extracted from bovineoocytes (39 plusmn 07) and in sperm (19 plusmn 03) There was alsono difference in methylation between oocytes (28 plusmn 09) andsperm (39 plusmn 2) within intron 5 which contains the promoterregion (Fig 4)
By contrast a significant difference (P le 005) betweenDNA extracted from oocytes and sperm was detected in themethylation level of exon 10 This final exon of the bovineIGF2 gene was sixfold more methylated in sperm(99 plusmn 03) than in oocytes (16 plusmn 08 Fig 4) clearlyindicating the presence of an intragenic differentiallymethylated region The methylation difference included all27 CpGrsquos between nucleotides 257 and 676 bp of the codingregion (GenBank X53553) Importantly the 54-bp coreregion of the intragenic DMR2 in the mouse [3134](GenBank U71085 24416ndash24470 bp) is conserved withinthe homologous intragenic DMR of the bovine IGF2 gene(GenBank X53553 281ndash334 bp)
Discussion
The IGF2 gene is subject to imprinting in several mammalianspecies and was selected for this study because of its crucial rolein growth and development [1316] We sequenced putativelyimprinting-associated portions of the gene since only thesequence of exon 2 and the translated region of the bovineIGF2 gene were available in the GenBank database (AY23743X53553) Although the H19 DMR located in the intergenicregion downstream of the mouse Igf2 gene is known as the majorcis-acting element regulating expression of the two genes it was
excluded from the analysis Our approach sequencing parts ofthe bovine IGF2 gene with primer pairs specific to the ovineIGF2 gene was dependent on a high portion of homologoussequences between these two species [4243] This studyfocused on the identification of intragenic DMRs because weanticipated higher sequence conservation within the gene bodythan in an intergenic region
Analysis of the already published sequence of the ovineIGF2 gene [12] revealed a CpG island (data not shown) withinthe untranslated region based on the definition of Gardiner-Garden and Frommer [25] The presence of a CpG island withinthe ovine IGF2 gene was also confirmed in the homologousbovine 1735-bp fragment Homologies of sim90 were detectedfor this fragment of the bovine IGF2 gene with phylogeneticallydistant species such as human and pig Thus the molecularstructure of the IGF2 gene is highly conserved betweenmammals despite minor modifications such as the twopseudo-exons in the mouse [4] or the lack of exon 2 in sheep[12] A recently published study on promoter-specific expres-sion patterns in the bovine IGF2 gene confirms our results ofCpG-rich sequence identities and high sequence homologiesbetween different mammalian species [15] In this studypromoter 3 has been mapped downstream of exon 5 bycomparison of the corresponding sequence of the human andpig The analysis of promoter-specific elements presented inthis study verifies in detail the location of this promoter withinthe bovine IGF2 gene
Knowledge of the complex regulation of the IGF2 genestems predominantly from studies in the mouse The murineIgf2 gene is reciprocally imprinted to the downstream-locatedH19 gene and both genes share common enhancers [2344]Unlike the mouse gene-specific information on bovinemolecular structure and imprinting status has only recentlyemerged [152245ndash47] From the bovine insulin and IGFfamily the insulin-like growth factor 2 receptor (IGF2R) geneand the IGF2 gene were identified as imprinted genes [2245]Both studies used single-nucleotide polymorphisms (SNPs) to
Fig 4 Methylation patterns of exon and intron 4 intron 5 and exon 10 in in vitro-matured bovine oocytes () and frozenthawed sperm () Black boxes indicateexons each line represents one individual bacterial clone and each circle one single CpG dinucleotide Open circles show nonmethylated CpGrsquos and black circlesmethylated CpGrsquos
226 C Gebert et al Genomics 88 (2006) 222ndash229
discriminate between the two parental alleles In the presentstudy however we overcame the lack of known geneticmarkers such as SNPs for determination of differentiallymethylated regions within the bovine IGF2 gene by usingfemale and male gametes as the source for DNA A significantdifference in the methylation level was detected in the finalexon of the gene between the two gametes The extraordi-narily high methylation level of sperm DNA and the lowmethylation level in oocytes clearly indicate the presence of an
Fig 5 Representative gel photograph of PCR analyses using bisulfite-treated DNAisolated from in vitro-matured bovine oocytes and frozenthawed sperm PCR producinto Escherichia coli cells prior to sequencing (Lanes 1ndash3) Exon and intron 4 oocontrol (Lanes 7ndash9) Exon 10 oocytes sperm negative control
intragenic DMR and underline the validity of the presentmethodological approach Our finding supports the notion thatthe IGF2 gene is well conserved among mammals as thehomologous DMR had also been found in the last exon ofmouse and human in which the paternal allele is methylated aswell [3448] We propose that this DMR corresponds to theDMR2 in the mouse which becomes reprogrammed afterfertilization but paternal hypermethylation is restored laterduring development [49] In contrast to the mouse
to identify intragenic DMRs within the bovine IGF2 gene Genomic DNAwasts were cloned into the pGEMT-Easy vector system (Promega) and transformedcytes sperm negative control (Lanes 4ndash6) Intron 5 oocytes sperm negative
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
Table 1Primers used for PCR
Primer Primer sequence Annealing (degC) PCR supplement Accession No
oIGF2-51 F GTGAGCTCGGCCATTCAGGTAGGAT 62 5 M betain U00664R CGGGCGTTGAGGTAGACGAAGAGGA 40 cycles F2132ndash2156 R3106ndash3082
Bisulfite-41 F ATTGGGTATTGTTTTTAGTTTTT 491 F61ndash83 R438ndash458R TTATAATCTTTACACAAAACA 40 cycles
Bisulfite-42 F GTTTTTAGTTTTTTTTAAATTTG 474 F71ndash93 R288ndash307R AATAAATTTAAAAACCAAAC 35 cycles
Bisulfite-51 F TGGTTTTTTTAGATTTTTAAATGAT 51 F688ndash712 R465ndash490R AAAATAAAAACAAAAATCCTCAAAA 40 cycles
Bisulfite-52 F GTTTTATGTTTGGTTTTTAGT 502 F44ndash64 R434ndash454R TCTAAAATTACTATTCCAAAA 35 cycles
Bisulfite-101 F TGGGTAAGTTTTTTTAATATGATATT 496 X53553R TTTAAAACCAATTAATTTTATACATT 40 cycles F243ndash268 R672ndash697
Bisulfite-102 F TAATATGATATTTGGAAGTAGT 491 X53553R ACATTTTTAAAAATATTATTCT 35 cycles F257ndash278 R655ndash676
225C Gebert et al Genomics 88 (2006) 222ndash229
the specific primer positions excluding any CpGrsquos Thesequence of exon 10 had been published (GenBank X53553[40]) This is the last exon of the bovine IGF2 gene and it wasincluded in the bisulfite sequencing analysis because the lastexon of the murine Igf2 gene contains a well-characterizedDMR and an important enhancer element [3341] In the presentstudy (Figs 4 and 5) the methylation level of 27 CpGrsquos in exon4 and intron 4 was low both in DNA extracted from bovineoocytes (39 plusmn 07) and in sperm (19 plusmn 03) There was alsono difference in methylation between oocytes (28 plusmn 09) andsperm (39 plusmn 2) within intron 5 which contains the promoterregion (Fig 4)
By contrast a significant difference (P le 005) betweenDNA extracted from oocytes and sperm was detected in themethylation level of exon 10 This final exon of the bovineIGF2 gene was sixfold more methylated in sperm(99 plusmn 03) than in oocytes (16 plusmn 08 Fig 4) clearlyindicating the presence of an intragenic differentiallymethylated region The methylation difference included all27 CpGrsquos between nucleotides 257 and 676 bp of the codingregion (GenBank X53553) Importantly the 54-bp coreregion of the intragenic DMR2 in the mouse [3134](GenBank U71085 24416ndash24470 bp) is conserved withinthe homologous intragenic DMR of the bovine IGF2 gene(GenBank X53553 281ndash334 bp)
Discussion
The IGF2 gene is subject to imprinting in several mammalianspecies and was selected for this study because of its crucial rolein growth and development [1316] We sequenced putativelyimprinting-associated portions of the gene since only thesequence of exon 2 and the translated region of the bovineIGF2 gene were available in the GenBank database (AY23743X53553) Although the H19 DMR located in the intergenicregion downstream of the mouse Igf2 gene is known as the majorcis-acting element regulating expression of the two genes it was
excluded from the analysis Our approach sequencing parts ofthe bovine IGF2 gene with primer pairs specific to the ovineIGF2 gene was dependent on a high portion of homologoussequences between these two species [4243] This studyfocused on the identification of intragenic DMRs because weanticipated higher sequence conservation within the gene bodythan in an intergenic region
Analysis of the already published sequence of the ovineIGF2 gene [12] revealed a CpG island (data not shown) withinthe untranslated region based on the definition of Gardiner-Garden and Frommer [25] The presence of a CpG island withinthe ovine IGF2 gene was also confirmed in the homologousbovine 1735-bp fragment Homologies of sim90 were detectedfor this fragment of the bovine IGF2 gene with phylogeneticallydistant species such as human and pig Thus the molecularstructure of the IGF2 gene is highly conserved betweenmammals despite minor modifications such as the twopseudo-exons in the mouse [4] or the lack of exon 2 in sheep[12] A recently published study on promoter-specific expres-sion patterns in the bovine IGF2 gene confirms our results ofCpG-rich sequence identities and high sequence homologiesbetween different mammalian species [15] In this studypromoter 3 has been mapped downstream of exon 5 bycomparison of the corresponding sequence of the human andpig The analysis of promoter-specific elements presented inthis study verifies in detail the location of this promoter withinthe bovine IGF2 gene
Knowledge of the complex regulation of the IGF2 genestems predominantly from studies in the mouse The murineIgf2 gene is reciprocally imprinted to the downstream-locatedH19 gene and both genes share common enhancers [2344]Unlike the mouse gene-specific information on bovinemolecular structure and imprinting status has only recentlyemerged [152245ndash47] From the bovine insulin and IGFfamily the insulin-like growth factor 2 receptor (IGF2R) geneand the IGF2 gene were identified as imprinted genes [2245]Both studies used single-nucleotide polymorphisms (SNPs) to
Fig 4 Methylation patterns of exon and intron 4 intron 5 and exon 10 in in vitro-matured bovine oocytes () and frozenthawed sperm () Black boxes indicateexons each line represents one individual bacterial clone and each circle one single CpG dinucleotide Open circles show nonmethylated CpGrsquos and black circlesmethylated CpGrsquos
226 C Gebert et al Genomics 88 (2006) 222ndash229
discriminate between the two parental alleles In the presentstudy however we overcame the lack of known geneticmarkers such as SNPs for determination of differentiallymethylated regions within the bovine IGF2 gene by usingfemale and male gametes as the source for DNA A significantdifference in the methylation level was detected in the finalexon of the gene between the two gametes The extraordi-narily high methylation level of sperm DNA and the lowmethylation level in oocytes clearly indicate the presence of an
Fig 5 Representative gel photograph of PCR analyses using bisulfite-treated DNAisolated from in vitro-matured bovine oocytes and frozenthawed sperm PCR producinto Escherichia coli cells prior to sequencing (Lanes 1ndash3) Exon and intron 4 oocontrol (Lanes 7ndash9) Exon 10 oocytes sperm negative control
intragenic DMR and underline the validity of the presentmethodological approach Our finding supports the notion thatthe IGF2 gene is well conserved among mammals as thehomologous DMR had also been found in the last exon ofmouse and human in which the paternal allele is methylated aswell [3448] We propose that this DMR corresponds to theDMR2 in the mouse which becomes reprogrammed afterfertilization but paternal hypermethylation is restored laterduring development [49] In contrast to the mouse
to identify intragenic DMRs within the bovine IGF2 gene Genomic DNAwasts were cloned into the pGEMT-Easy vector system (Promega) and transformedcytes sperm negative control (Lanes 4ndash6) Intron 5 oocytes sperm negative
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
Fig 4 Methylation patterns of exon and intron 4 intron 5 and exon 10 in in vitro-matured bovine oocytes () and frozenthawed sperm () Black boxes indicateexons each line represents one individual bacterial clone and each circle one single CpG dinucleotide Open circles show nonmethylated CpGrsquos and black circlesmethylated CpGrsquos
226 C Gebert et al Genomics 88 (2006) 222ndash229
discriminate between the two parental alleles In the presentstudy however we overcame the lack of known geneticmarkers such as SNPs for determination of differentiallymethylated regions within the bovine IGF2 gene by usingfemale and male gametes as the source for DNA A significantdifference in the methylation level was detected in the finalexon of the gene between the two gametes The extraordi-narily high methylation level of sperm DNA and the lowmethylation level in oocytes clearly indicate the presence of an
Fig 5 Representative gel photograph of PCR analyses using bisulfite-treated DNAisolated from in vitro-matured bovine oocytes and frozenthawed sperm PCR producinto Escherichia coli cells prior to sequencing (Lanes 1ndash3) Exon and intron 4 oocontrol (Lanes 7ndash9) Exon 10 oocytes sperm negative control
intragenic DMR and underline the validity of the presentmethodological approach Our finding supports the notion thatthe IGF2 gene is well conserved among mammals as thehomologous DMR had also been found in the last exon ofmouse and human in which the paternal allele is methylated aswell [3448] We propose that this DMR corresponds to theDMR2 in the mouse which becomes reprogrammed afterfertilization but paternal hypermethylation is restored laterduring development [49] In contrast to the mouse
to identify intragenic DMRs within the bovine IGF2 gene Genomic DNAwasts were cloned into the pGEMT-Easy vector system (Promega) and transformedcytes sperm negative control (Lanes 4ndash6) Intron 5 oocytes sperm negative
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
227C Gebert et al Genomics 88 (2006) 222ndash229
remethylation of the bovine genome already starts at themorula stage [50] The intragenic DMR of the IGF2 gene cantherefore be used as a bovine marker DMR
The DMR2 was found to contain a 54-bp core regionresponsible for the enhancement of transcription from thepaternal allele in the mouse and was also determined in thehuman and porcine orthologues [143134] The core region isconserved in exon 10 of the bovine IGF2 gene suggesting thatregulatory elements of the IGF2 gene are similar amongdifferent mammalian species
The identification of this intragenic DMR within the bovineIGF2 gene provides a solid new basis for in-depth investiga-tions of bovine imprinting characteristics and its regulatorymechanisms Bovine in vitro-produced embryos are oftenaffected in their developmental capacity [5152] and aberrantgene expression of the bovine IGF2 IGF2R and H19 geneswas reported from calves derived after somatic nuclear transfer[53] Reports about developmental aberrations associated withthe continuously rising popularity of biotechnological methodsin both animals and humans demonstrate the need for diagnostictools to detect developmental abnormalities early in embryo-genesis The DMR identified in this study will provide such avaluable diagnostic tool to evaluate methylation patterns insingle bovine preimplantation embryos and be instrumental inidentifying putative developmental failures
Materials and methods
Collection of in vitro-matured bovine oocytes
Cumulus oocyte complexes were isolated from slaughterhouse ovaries byslicing and matured in vitro [5455] Briefly TCM 199 containing L-glutamine25 mM Hepes 22 μg pyruvate 22 μg NaHCO3 and 50 μg gentamicin wassupplemented with 10 IU eCG 5 IU hCG (Suigonan Intervet ToumlnisvorstGermany) and 01 BSA-FAF (SigmandashAldrich Chemie TaufkirchenGermany) In vitro maturation was performed at 39degC and 5 CO2 in ahumidified air atmosphere for 24 h Cumulus cells were removed from oocytesby incubation in 01 hyaluronidase in Ca2+- and Mg2+-free PBS Each oocytewas examined for extrusion of the first polar body Pools of 40 or 80 oocytes(metaphase II) were frozen in 2-ml tubes and stored in a minimum of PBSsupplemented with 01 PVA at minus80degC
DNA preparation
Genomic DNA was isolated from bovine kidney by lysis in 550 μl of abuffer consisting of 50 mM TrisndashHCl at pH 80 100 mM NaCl2 100 mMEDTA 1 SDS and treated with 70 μl of proteinase K (10 mgml) GenomicDNA from frozenthawed sperm was prepared in the same manner but 500 μlof lysis buffer was supplemented with 25 μl of Triton X-100 (MerckDarmstadt Germany) 21 μl of dithiothreitol (1 M) (SigmandashAldrich Chemie)and 40 μl of proteinase K (10 mgml) DNA precipitation was performed in asaturated sodium chloride solution with subsequent addition of 100 ethanol(Roth Hamburg Germany)
Isolation of genomic DNA from in vitro-matured oocytes was carried outusing the protocols of Hajkova et al [56] and Lopes et al [49] Briefly 40oocytes were boiled in water for 30 min then 100 μl lysis buffer (10 mMTrisndashHCl pH 8 10 mM EDTA pH 8 1 SDS (wv)) 20 μl proteinase K(10 mgml) and 1 μl yeast tRNA (11 mgml SigmandashAldrich Chemie) wereadded and the digestion proceeded overnight at 55degC in a thermal shakerGenomic DNA was isolated by phenolchloroform extraction using the PhaseLock Gel system (PLG-Heavy Eppendorf Hamburg Germany) Isolatedgenomic DNA was precipitated by the addition of 2 volumes of 100 ethanol
(Roth) 3 M sodium acetate (to reach a final concentration of 03 M) 1 μlyeast tRNA (11 mgml SigmandashAldrich Chemie) and cooling for 2 h atminus30degC The DNA pellet was washed twice in 500 μl 70 ethanol (Roth)followed by air-drying DNA was resuspended in 3 μl sterile water at 4degCovernight
PCR amplification and DNA sequencing
Primers designed from the sequence of the ovine IGF2 gene(GenBank U00664) were used to amplify bovine kidney DNA Twoindividual PCRs were performed one to amplify DNA fragmentsspanning exons 4 and 5 and the other to amplify exons 5 and 6Using 100 ng of bovine genomic kidney DNA PCR amplification wasperformed in 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen Karlsruhe Germany) 02 mM dNTPs(Amersham Biosciences Europe GmbH Freiburg Germany) 06 μMeach primer (oIGF2-4oIGF2-5) and primer specific PCR supplements(DMSO (Merck) or betain monohydrate (Fluca Taufkirchen Germany)Table 1) Taq polymerase (15 units Invitrogen) was added at 72degC after5 min of denaturation at 95degC Template DNA was amplified for40 PCR cycles Each PCR cycle consisted of a denaturation step at95degC for 20 s 30 s at the primer-specific annealing temperature and45 s of extension at 72degC The PCR was terminated by a finalextension step at 72degC for 10 min PCR products were separated by gelelectrophoresis and gel bands were cut out and purified using the GFXPCR DNA and Gel Band Purification Kit (Amersham BiosciencesEurope) prior to sequencing A bovine-specific primer pair (bIGF2-42)was designed from the sequence of the PCR product from the oIGF2-41primers These primers amplified a fragment within intron 4 of thebovine IGF2 gene
Bisulfite sequencing of oocyte and sperm DNA
Bisulfite sequencing was performed using the protocols of Hajkova etal [56] and Lopes et al [49] Briefly genomic DNA isolated from 40 invitro-matured oocytes and 16 ng of sperm DNA was digested with EcoRI(New England Biolabs Frankfurt aM Germany) Denatured DNA wasembedded in 7 μl of low-melting agarose (SeaPlaque GTGagarose 20 mgml Biozym Hess Oldendorf Germany) cooled to form agarose beadsand incubated in 25 M bisulfitendashhydroquinone solution pH 5 (Roth) for4 h at 50degC
First-round PCR was carried out in a final volume of 100 μl PCR conditionswere as follows 1times PCR buffer (20 mM TrisndashHCl pH 84 50 mM KCl2)15 mM MgCl2 (Invitrogen) 02 μM dNTPs (Amersham Biosciences Europe)and 06 μM each primer (Bisulfite-4 Bisulfite-5 Bisulfite-10 Table 1) After aninitial denaturation step at 95degC for 5 min 15 units Taq polymerase (5 UμlInvitrogen) was added at 72degC following 40 PCR cycles Nested PCR wasperformed by taking 2 μl of the first-round PCR products and repeating thereaction for 35 cycles PCR products were cloned into the pGEMT-Easy vectorsystem (Promega Mannheim Germany) and transformed into Escherichia colicells (XL10 Gold Stratagene Europe Amsterdam The Netherlands) andindividual clones were sequenced Only sequences derived from clones withge95 cytosine conversion were analyzed Percentages were calculated basedon the total number of cytosine molecules and the number of convertedcytosines within the fragment amplified The methylation patterns were used toidentify clones originating from different DNA templates
Statistical analysis
Within exon 4 intron 4 and exon 10 a total of 27 CpG dinucleotides wereanalyzed and within intron 5 a total of 38 CpG dinucleotides were analyzed bybisulfite sequencing Methylation patterns of each CpG were identified inmultiple individual clones from two to six independently amplified DNAtemplates The global methylation levels of each of the three fragments underinvestigation were calculated as the percentage of the total amount of
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
228 C Gebert et al Genomics 88 (2006) 222ndash229
methylation in all 27 and 38 CpGrsquos respectively Descriptive statistics and one-way ANOVAwere performed with the computer program SigmaStat 20 (JandelScientific San Rafael CA USA) Significant differences (P le 005) betweenthe groups were detected by multiple pair-wise comparisons using Dunnrsquosmethod
Acknowledgments
We are thankful to Wilfried A Kues for helpful discussionon the manuscript C Gebert was financially supported by theDepartment of Reproductive Medicine the Friends andSupporters of the University of Veterinary Medicine Hannover(Germany) the Institute for Animal Breeding (FAL) Neustadt-Mariensee (Germany) and DFG-grant Ni 25618-1
References
[1] TM DeChiara A Efstratiadis EJ Robertson A growth-deficiencyphenotype in heterozygous mice carrying an insulin-like growth factor IIgene disrupted by targeting Nature 345 (1990) 78ndash80
[2] A Efstratiadis Genetics of mouse growth Int J Dev Biol 42 (1998)955ndash976
[3] M Constacircncia M Hemberger J Hughes W Dean A Ferguson-SmithR Fundele et al Placental-specific IGF-II is a major modulator ofplacental and fetal growth Nature 417 (2002) 945ndash948
[4] P Rotwein LJ Hall Evolution of insulin-like growth factor IIcharacterization of the mouse IGF-II gene and identification of twopseudo-exons DNA Cell Biol 9 (1990) 725ndash735
[5] JE Brissenden A Ullrich U Franke Human chromosomal mapping ofgenes for insulin-like growth factors I and II and epidermal growth factorNature 310 (1984) 781ndash784
[6] JV Tricoli LB Rall J Scott GI Bell TB Shows Localization ofinsulin-like growth factor genes to human chromosomes 11 and 12 Nature310 (1984) 784ndash786
[7] C Nezer L Moreau B Brouwers W Coppieters J Detilleux RHanset et al An imprinted QTL with major effect on muscle mass andfat deposition maps to the IGF2 locus in pigs Nat Genet 21 (1999)155ndash156
[8] JT Jeon Ouml Carlborg A Toumlrnsten E Giuffra V Amarger P Chardonet al A paternally expressed QTL affecting skeletal and cardiac musclemass in pigs maps to the IGF2 locus Nat Genet 21 (1999) 157ndash158
[9] HA Ansari PD Pearce DWMaher TE Broad Regional assignment ofconserved reference loci anchors unassigned linkage and syntenic groupsto ovine chromosomes Genomics 24 (1994) 451ndash455
[10] JJ Goodall SM Schmutz Linkage mapping of IGF2 on cattlechromosome 29 Anim Genet 34 (2003) 313
[11] P De Pagter-Holthuizen M Jansen RA van der Kammen FM vanSchaik JS Sussenbach Differential expression of the human insulin-likegrowth factor II gene characterization of the IGF-II mRNAs and anmRNA encoding a putative IGF-II-associated protein Biochim BiophysActa 950 (1988) 282ndash295
[12] SM Ohlsen KA Lugenbeel EA Wong Characterization of the linkedovine insulin and insulin-like growth factor-II genes DNA Cell Biol 13(1994) 377ndash388
[13] R Mineo E Fichera SJ Liang Y Fujita-Yamaguchi Promoter usage forinsulin-like growth factor-II in cancerous and benign human breastprostate and bladder tissues and confirmation of a 10th exon BiochemBiophys Res Commun 268 (2000) 886ndash892
[14] V Amarger M Nguyen AS van Laere M Braunschweig C Nezer MGeorges et al Comparative sequence analysis of the INS-IGF2-H19 genecluster in pigs Mamm Genome 13 (2002) 388ndash398
[15] C Curchoe S Zhang Y Bin X Zhang L Yang D Feng et alPromoter-specific expression of the imprinted IGF2 gene in cattle (Bostaurus) Biol Reprod 73 (2005) 1275ndash1281
[16] TM DeChiara EJ Robertson A Efstratiadis Parental imprintingof the mouse insulin-like growth factor II gene Cell 64 (1991)849ndash859
[17] S Rainier LA Johnson CJ Dobry AJ Ping PE Grundy APFeinberg Relaxation of imprinted genes in human cancer Nature 362(1993) 747ndash749
[18] O Ogawa MR Eccles J Szeto LA McNoe K Yun MA Maw et alRelaxation of insulin-like growth factor II gene imprinting implicated inWilmsrsquo tumour Nature 362 (1993) 749ndash751
[19] R Ohlsson A Nystroem S Pfeifer-Ohlsson V Toehoenen F HedborgP Schofield et al IGF2 is parentally imprinted during humanembryogenesis and in the BeckwithndashWiedemann syndrome Nat Genet4 (1993) 94ndash97
[20] R Feil S Khosla P Cappai P Loi Genomic imprinting in ruminantsallele-specific gene expression in parthenogenetic sheep Mamm Genome9 (1998) 831ndash834
[21] RJ McLaren GW Montgomery Genomic imprinting of the insulin-likegrowth factor 2 gene in sheep Mamm Genome 10 (1999) 588ndash591
[22] SV Dindot PW Farin CE Farin J Romano S Walker C LongEpigenetic and genomic imprinting analysis in nuclear transfer derivedBos gaurusBos taurus hybrid fetuses Biol Reprod 71 (2004)470ndash478
[23] W Reik J Walter Genomic imprinting parental influence on the genomeNat Rev Genet 2 (2001) 21ndash32
[24] E Li TH Bestor R Jaenisch Targeted mutation of the DNAmethyltransferase gene results in embryonic lethality Cell 69 (1992)915ndash926
[25] M Gardiner-Garden M Frommer CpG islands in vertebrate genomesJ Mol Biol 196 (1987) 261ndash282
[26] F Larsen G Gundersen H Prydz Choice of enzymes for mapping basedon CpG islands in the human genome Genet Anal Tech Appl 9 (1992)80ndash85
[27] E Li C Beard R Jaenisch Role for DNA methylation in genomicimprinting Nature 366 (1993) 362ndash365
[28] D Takai PA Jones Comprehensive analysis of CpG islands in humanchromosomes 21 and 22 Proc Natl Acad Sci USA 99 (2002)3740ndash3745
[29] T Moore M Constacircncia M Zubair B Bailleul R Feil H Sasaki et alMultiple imprinted sense and antisense transcripts differential methyl-ation and tandem repeats in a putative imprinting control regionupstream of mouse Igf2 Proc Natl Acad Sci USA 94 (1997)12509ndash12514
[30] M Constacircncia W Dean S Lopes T Moore G Kelsey W Reik Deletionof a silencer element in Igf2 results in loss of imprinting independent ofH19 Nat Genet 26 (2000) 203ndash206
[31] A Murrell S Heeson L Bowden M Constacircncia W Dean GKelsey et al An intragenic methylated region in the imprinted Igf2gene augments transcription EMBO Rep 2 (2001) 1101ndash1106
[32] KY Park EA Sellars A Grinberg SP Huang K Pfeifer The H19differentially methylated region marks the parental origin of a heterologouslocus without gametic DNA methylation Mol Cell Biol 24 (2004)3588ndash3595
[33] H Sasaki PA Jones JR Chaillet AC Ferguson-Smith SC BartonW Reik et al Parental imprinting potentially active chromatin of therepressed maternal allele of the mouse insulin-like growth factor II(Igf2) gene Genes Dev 6 (1992) 1843ndash1856
[34] T Forneacute J Oswald W Dean JR Saam B Bailleul L Dandolo et alLoss of the maternal H19 gene induces changes in Igf2 methylation in bothcis and trans Proc Natl Acad Sci USA 94 (1997) 10243ndash10248
[35] R Anbazhagan JG Herman K Enika E Gabrielson Spreadsheet-basedprogram for the analysis of DNA Methylation Biotechniques 30 (2001)110ndash114
[36] K Shiota DNA methylation profiles of CpG islands for cellulardifferentiation and development in mammals Cytogenet Genome Res105 (2004) 325ndash334
[37] MG Reese Application of a time-delay neural network to promoterannotation in the Drosophila melanogaster genome Comput Chem 26(2001) 51ndash56
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
229C Gebert et al Genomics 88 (2006) 222ndash229
[38] JT Kadonaga KA Jones R Tjian Promoter-specific activation ofRNA polymerase II transcription by SP1 Trends Biochem 11 (1986)20ndash23
[39] AM Raizis MR Eccles AE Reeve Structural analysis of the humaninsulin-like growth factor-II P3 promoter Biochem J 289 (1993)133ndash139
[40] WM Brown KM Dziegielewska RC Foreman NR Saunders Thenucleotide and deduced amino acid sequences of insulin-like growth factorII cDNAs from adult bovine and fetal sheep liver Nucleic Acids Res 18(1990) 4614
[41] R Feil J Walter ND Allen W Reik Developmental control of allelicmethylation in the imprinted mouse Igf2 and H19 genes Development 120(1994) 2933ndash2943
[42] MS Bartolomei AL Webber ME Brunkow SM TilghmanEpigenetic mechanisms underlying the imprinting of the mouse H19gene Genes Dev 7 (1993) 1663ndash1673
[43] PA Leighton RS Ingram J Eggenschwiler A Efstratiadis SMTilghman Disruption of imprinting caused by deletion of the H19 generegion in mice Nature 375 (1995) 34ndash39
[44] RI Verona MRW Mann MS Bartolomei Genomic imprintingintricacies of epigenetic regulation in clusters Annu Rev Cell Dev Biol19 (2003) 237ndash259
[45] JK Killian CM Nolan AA Wylie T Li TH Vu AR Hoffman et alDivergent evolution in M6PIGF2R imprinting from the Jurassic to theQuaternary Hum Mol Genet 10 (2001) 1721ndash1728
[46] J Kim A Bergmann S Lucas R Stone L Stubbs Lineage-specificimprinting and evolution of the zinc-finger gene ZIM2 Genomics 84(2004) 47ndash58
[47] S Zhang C Kubota L Yang Y Zhang R Page M OrsquoNeill et alGenomic imprinting of H19 in naturally reproduced and cloned cattleBiol Reprod 71 (2004) 1540ndash1544
[48] W Reik KW Brown RE Slatter P Sartori M Elliott ER Maher
Allelic methylation of H19 and IGF2 in the BeckwithndashWiedemannsyndrome Hum Mol Genet 3 (1994) 1297ndash1301
[49] S Lopes A Lewis P Hajkova W Dean J Oswald T Forneacute et alEpigenetic modifications in an imprinting cluster are controlled by ahierarchy of DMRs suggesting long-range chromatin interactionsHum Mol Genet 12 (2003) 295ndash305
[50] W Dean F Santos M Stojkovic V Zakhartchenko J Walter E Wolfet al Conservation of methylation reprogramming in mammaliandevelopment aberrant reprogramming in cloned embryos Proc NatlAcad Sci USA 98 (2001) 13734ndash13738
[51] H Niemann C Wrenzycki Alterations of expression of developmentallyimportant genes in preimplantation bovine embryos by in vitro cultureconditions implications for subsequent development Theriogenology 53(2000) 21ndash34
[52] C Wrenzycki D Herrmann A Lucas-Hahn K Korsawe E Lemme HNiemann Messenger RNA expression patterns in bovine embryos derivedfrom in vitro procedures and their implications for development ReprodFertil Dev 17 (2005) 23ndash35
[53] L Yang P Chavatte-Palmer C Kubota M OrsquoNeill T Hoagland JPRenard Expression of imprinted genes is aberrant in deceased newborncloned calves and relatively normal in surviving adult clones MolReprod Dev 71 (2005) 431ndash438
[54] J Eckert H Niemann In vitro maturation fertilization and culture toblastocysts of bovine oocytes in protein-free media Theriogenology 43(1995) 1211ndash1225
[55] C Wrenzycki D Herrmann H Niemann Timing of blastocystexpansion affects spatial messenger RNA expression patterns of genesin bovine blastocysts produced in vitro Biol Reprod 68 (2003)2073ndash2080
[56] P Hajkova O El-Maarri S Engemann J Oswald A Olek J WalterDNA-methylation analysis by the bisulfite-assisted genomic sequencingmethod Methods Mol Biol 200 (2002) 143ndash154
