UBC Faculty Research and Publications

Functional consequences of copy number variants in miscarriage Wen, Jiadi; Hanna, Courtney W; Martell, Sally; Leung, Peter C; Lewis, Suzanne M; Robinson, Wendy P; Stephenson, Mary D; Rajcan-Separovic, Evica Jan 31, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-13039_2015_Article_109.pdf [ 1.04MB ]
JSON: 52383-1.0223520.json
JSON-LD: 52383-1.0223520-ld.json
RDF/XML (Pretty): 52383-1.0223520-rdf.xml
RDF/JSON: 52383-1.0223520-rdf.json
Turtle: 52383-1.0223520-turtle.txt
N-Triples: 52383-1.0223520-rdf-ntriples.txt
Original Record: 52383-1.0223520-source.json
Full Text

Full Text

RESEARCHFunctional consequences oCKtioprfamsomal abnormalities has been recently revolutionized first line test for subjects with postnatal developmentalWen et al. Molecular Cytogenetics  (2015) 8:6 DOI 10.1186/s13039-015-0109-8in 1.6% of chromosomally normal cases. These CNVswere defined as >5 Mb or overlapping with critical2Child & Family Research Institute, Vancouver V5Z 4H4, CanadaFull list of author information is available at the end of the articlewith the development of chromosome microarray ana-lysis (CMA) which facilitates the detection of large scaleand subtle chromosomal microdeletions and microdupli-cations (DNA copy number variants or CNVs). CMA isdelay and congenital abnormalities [6,7]. It identifiesclinically relevant CNVs in 10-15% of chromosomallynormal cases, with the majority of CNVs undetectableby routine karyotyping. In contrast, array studies of mis-carriages are still rare, and no more than 3000 casesworldwide were reported [3]. Most of the miscarriagecases (2392) were described in the recent publication byLevy et al. [8], which reported putatively causative CNVs* Correspondence: eseparovic@cw.bc.ca†Equal contributors1Department of Pathology and Laboratory Medicine, University of BritishColumbia, Vancouver V6T 2B5, Canadaof reproductive genetics [4,5]. The detection of chromo-relevance to miscarriage by expression analysis of 14 genes integral to CNVs in available miscarriage chorionic villi.As familial CNVs could cause miscarriage due to imprinting effect, we investigated the allelic expression of one ofthe genes (TIMP2) previously suggested to be maternally expressed in placenta and involved in placental remodellingand embryo development.Results: Six out of fourteen genes had detectable expression in villi and for three genes the RNA and protein expressionwas altered due to maternal CNVs. These genes were integral to duplication on Xp22.2 (TRAPPC2 and OFD1) or disruptedby a duplication mapping to 17q25.3 (TIMP2). RNA and protein expression was increased for TRAPPC2 and OFD1 andreduced for TIMP2 in carrier miscarriages. The three genes have roles in processes important for pregnancy developmentsuch as extracellular matrix homeostasis (TIMP2 and TRAPPC2) and cilia function (OFD1). TIMP2 allelic expression was notaffected by the CNV in miscarriages in comparison to control elective terminations.Conclusion: We propose that functional studies of CNVs could help determine if and how the miscarriage CNVs affectthe expression of integral genes. In case of parental CNVs, assessment of the function of their integral genes in parentalreproductive tissues should be also considered in the future, especially if they affect processes relevant for pregnancydevelopment and support.Keywords: Miscarriage, Copy number variation, TIMP2, OFD1, TRAPPC2, Gene expressionBackgroundGenetic factors, such as single gene defects and chromo-somal abnormalities, are a common cause of miscarriage[1-3]. Their identification is important for informed re-productive decisions and counselling and is the key goalbased on the assessment of DNA directly obtained fromtissues and thus minimizes the negative impact of tissueculture artefacts and failure associated with traditionalcytogenetic analysis of miscarriages.Due to its many benefits, including improved reso-lution and diagnostic yield, CMA is now considered ain miscarriageJiadi Wen1,2†, Courtney W Hanna2,3†, Sally Martell2, PeterMary D Stephenson5 and Evica Rajcan-Separovic1,2*AbstractBackground: The presence of unique copy number variagenes have a role in maintaining early pregnancy. In ourstudied euploid miscarriages, which were predominantly© 2015 Wen et al.; licensee BioMed Central. ThCommons Attribution License (http://creativecreproduction in any medium, provided the orDedication waiver (http://creativecommons.orunless otherwise stated.Open Accessf copy number variantsLeung2,4, Suzanne ME Lewis3, Wendy P Robinson2,3,ns (CNVs) in miscarriages suggests that their integralevious work, we identified 19 unique CNVs in ~40% ofilial in origin. In our current work, we assessed theiris is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,Wen et al. Molecular Cytogenetics  (2015) 8:6 Page 2 of 9regions causing or predisposing to viable microdeletion/duplication syndromes.Interestingly, the majority of CNVs identified so far inmiscarriages with a normal karyotype were familial inorigin. CNVs inherited from normal parents are gener-ally considered less likely to be causative of an ab-normal phenotype, however, they can be pathogenicif they i) uncover a mutation within the intact allele ofthe developmental gene in the conceptus, ii) containgenes with variable expressivity or penetrance iii) involveimprinting (parent of origin dependent) genes. Inaddition, parental CNVs could still lead to miscarriage ifthey affect genes required for normal parental reproduct-ive function, including, for example, genes required forplacenta function, maternal preparation for and mainten-ance of pregnancy and genomic stability of the sperm.These processes are essential for successful pregnancyoutcome [9], but may not necessarily impair parents’overall health status and can demonstrate pathogeniccharacteristics only at the time of pregnancy. Recently,Nagirnaja et al. [10], suggested that a CNV from 5p13.3chromosomal region is enriched in women with recur-rent pregnancy loss (RPL) in comparison to fertilecontrols and could represent a risk factor for preg-nancy complications, as it contains genes predominantlyexpressed in placenta.We have recently performed CMA analysis of miscar-riages from couples with RPL and sporadic miscarriageswith embryoscopy findings [11,12]. Unique CNVs notreported in Database of Genomic Variants were notedin ~40% of miscarriages and were predominantly paren-tal in origin, making their interpretation more challen-ging. In order to further investigate the potential role ofthese CNVs in miscarriage we performed functional ana-lysis (RNA and protein expression) of genes integral tomiscarriage CNVs, using chorionic villi from miscar-riages from 6 families with RPL, reported in our previ-ous work. Our study shows changes in RNA/proteinexpression for 3/14 tested genes from maternal CNVs,which could be of relevance for miscarriage due to theirrole in processes important for growth of the conceptusand/or maternal preparation for and support of preg-nancy. Functional studies of miscarriage CNVs couldtherefore help identify miscarriage candidate genes, butshould be accompanied with functional studies of CNVsin reproductive tissues of carrier parents.ResultsExpression of genes integral to CNVs in miscarriagecell culturesThree genes, OFD1, TRAPPC2 and TIMP2 out of 14 se-lected for expression analysis had altered mRNA andprotein expression in cultured miscarriage chorionic villicells (Table 1 and Figures 1 and 2). For NDUFAF2, CHSY3and PRMT3 (Table 1) the expression in control pregnan-cies and miscarriages with CNVs was comparable. Theremaining genes assayed had either very low or no expres-sion in cultured chorionic villi from controls (PARK, LIPC,CTNNA3, EGFL6, GPM6B, RAB9A, POU6F2 and C7orf10)and were not assessed in miscarriages.OFD1 and TRAPPC2 are integral to a CNV on Xp22.2(duplication with breakpoints from 13415099–13745233)identified by Agilent 105 k array (Figure One in Rajcan-Separovic et al. [12]) in female 09–1 who had 6 mis-carriages (current paper, Additional file 1: Table S1). Two ofher 6 miscarriages had array analysis: male miscarriage09-3A had a normal array result, while female miscarriage09-3B inherited the maternal Xp22.2 CNV. These two mis-carriages were available for functional analysis. IncreasedRNA and protein expression for OFD1 and TRAPPC2 wasdetected in 09-3B (Figure 1), while the remainder of thegenes from the Xp22.2 CNV (EGFL6, GPM6B, andRAB9A) had low or undetectable expression in control ETcell cultures and were not evaluated in 09-3B (Table 1).Random chromosome X inactivation was identified in09-3B, while the mother’s chromosome X inactivationstatus was uninformative [12].The CNV disrupting the TIMP2 gene mapped to 17q25.3(duplication with breakpoints from 74,381,287 -74,466,887)and was detected in female 6–1 and in 4/5 availablemiscarriages, as reported previously (Figure Two inRajcan-Separovic et al. [12]). Cell cultures from chorionicvilli were available from four miscarriages (06-3A, C, Dwhich contained the CNV and 06-3E which did not). Thethree miscarriages with the TIMP2 CNV (06-3A, C and D)showed a ~50% decrease in mRNA and protein expressionin comparison to four control elective terminations (ET 15,17, 18 and 20). A ~2-fold increase of TIMP2 mRNA andprotein was noted in the fourth miscarriage (06-3E), whichdid not carry the TIMP2 CNV (Figure 2).TIMP2 Allelic expression analysisBased on previous reports suggesting preferential mater-nal expression of TIMP2 in placenta [13], we tested theparent-of-origin specific expression of TIMP2 in controlET and in miscarriages from female 06–1 to determine ifthe CNV affected the TIMP2 allelic expression. Monoal-lelic expression of the maternal allele was detected in 2of 3 informative miscarriages (06-3C with and 06-3Ewithout the CNV, Figure 3) while a close to biallelicexpression (~60%) was noted in the third miscarriage(06-3D) which was trisomic for chromosome 16 andcontained the TIMP2 CNV. Monoallelic expression ofthe maternal allele was also noted in two of the seven in-formative ET samples heterozygous for the polymorphicrs2277698 G/A SNP in exon 3 (out of 35 genotyped).The cells from the remaining 5 control and informativeETs had biparental expression.Table 1 CNV gene analysisSamples Miscarriage Locus Breakpoint CNV size(kb)Type of CNVand originGene picked for expression(type of CNV abnormality)mRNA expression in controlelective terminations/miscarriagethat carries the CNVProtein expressionMiscarriages fromreference Rajcan-Separovicet al. [12]03-3A 6q26 162126633-162271770 145 loss-pat PARK2 (loss involves part ofexon and intron)N/not attempted05-3A 5q12.1 60407026-60464658 58 loss-pat NDUFAF2 (loss involves exon 3) yes/no difference between CNVcarriers and controls06-3A,B,C and D(not in 3E)17q25.3 74381287-74466887 86 gain-mat TIMP2 (gain involves exon1,2) yes/decrease in CNV carriers decreased06-3E 15q22.1 56487120-56562873 76 loss-pat LIPC (loss of exon1) N/not attempted07-3A 10q21.3 67992425-68064617 72 loss-mat CTNNA3 (loss involves exon11) N/not attempted09-3B (not in 3A) Xp22.2 13415099-13745233 33 gain-mat EGFL6 (complete gain) N/not attemptedGPM6B (almost complete gain) N/not attemptedOFD1 (complete gain) yes/increase in CNV carrier increased in CNVcarrierRAB9A (complete gain) N/not attemptedTRAPPC2 (complete gain) yes/increase in CNV carrier increased in CNVcarrier10-3A 5q23.3 129388119-129441487 53 loss-pat CHSY3 (CSS3) (loss involes intron) yes/no difference between CNVcarrier and controls11p15.1 20442396-20559837 117 Gain-pat PRMT3 (gain involves exon9-11) yes/no difference between CNVcarrier and controls7p14.1 39470588-39647671 177 Gain-mat POU6F2 (gain involves last exon) N/not attemptedC7orf10 (complete gain) N/not attemptedN = expression in control villi tissue low or not detectable, the expression in miscarriage therefore not attempted.Wenetal.MolecularCytogenetics (2015) 8:6 Page3of9Wen et al. Molecular Cytogenetics  (2015) 8:6 Page 4 of 9DiscussionIn this study we explored further the miscarriage CNVswe identified in our previous work by testing the expres-sion levels of their integral genes in available miscarriagetissues with CNVs. Three genes integral to maternalCNVs had altered expression in miscarriages consistentwith genomic alternation caused by the CNV.TRAPPC2 and OFD1 had increased RNA and proteinexpression in miscarriage 9-3B with a gain of Xp22.2.TRAPPC2 has a role in procollagen transportation [14]while OFD1 regulates cilia function [15]. Collagens arethe main components of extracellular matrix (ECM) andprovide structural support for the tissues, but alsoplay important roles in cell growth, differentiation,adhesion, and migration [16,17]. The dysfunction ofOFD1, through abnormal ciliogenesis, results in defects insonic hedgehog (Shh) and canonical Wnt signaling path-ways [15,18-20] which are linked to abnormal implant-ation and embryonic development [21-23]. Although themis-expression of TRAPPC2 and OFD1 could potentiallyexplain miscarriage 9-3B, the Xp22.2 CNV was notFigure 1 mRNA and protein expression of OFD1 and TRAPPC2 in primcontrol samples from the cell cultures of elective terminations (n = 4). 09-3A(ii) values for (A) OFD1 and (B) TRAPPC2 expression were normalized to ththree repeats of each sample and the mean level of each sample is represesamples (as a group) and the two CNV carrier miscarriages, as two groups,present in miscarriage 09-3A, and therefore cannot be as-sociated with all the pregnancy losses of female 09–1. In-stead, the unifying cause of her miscarriages could be theeffect of the CNV on maternal OFD1 and TRAPPC2 geneexpression, particularly because of the reports of abnormaluterine collagen content resulting in pregnancy loss[24,25]. It would be of interest the explore the expressionof these genes in the decidua of female 9–1. Unfortunatelythis tissue was not available for analysis.TIMP2 is the third gene that showed altered expres-sion in miscarriages and is known to inhibit matrixmetaloproteases (MMPs), which degrade extra cellularmatrix (ECM), and have a critical role in tissue remodel-ling and angiogenesis in placenta/endometrium [26-29].Expression of TIMP2 was reduced in miscarriages withthe CNV, possibly because the CNV (duplication) disruptsthe 5′ end of the gene and causes structural alternationsin the genomic region of TIMP2. It has been previouslyobserved that the genomic position of regulatory elementsin the TIMP2 promoter (AP-1) affects the gene expression[30]. However, the CNV appears not to be the sole causeary culture of human chorionic villi. ET 15, 17, 18 and 20 indicateand B indicate miscarriages from female 09–1. mRNA (i) and proteine corresponding ß-actin mRNA levels. The results derived from at leastnted in the scatter graph. The difference between the four controlhas been evaluated by Student t-test (*, p < 0.05).Wen et al. Molecular Cytogenetics  (2015) 8:6 Page 5 of 9of recurrent miscarriage in this family as TIMP2 expres-sion was also altered (increased) in the miscarriage with-out the CNV. It is conceivable that, due to the CNV, theexpression of TIMP2 is also abnormal (reduced) in theendometrium/decidua of female 6–1 and could impairpregnancy development. This is of interest consideringthat decidual TIMP2 has a role in regulating trophoblastinvasion by modulating trophoblast MMP and TIMP2 ex-pression [31], and in particular inhibiting trophoblastTIMP2 expression. In keeping with the possibility of theeffect of the maternal and/or pregnancy CNV on TIMP2function are the pathology findings for six out of ten mis-carriages from female 6–1 which demonstrated morpho-logic abnormalities of the maternal vasculature (intimalhyperplasia of maternal vessels) and/or placenta (perivil-lous fibrin deposits) [12] (Additional file 1: Table S1).Familial CNVs are frequently hypothesised to cause anadverse outcome of the pregnancy if they contain orFigure 2 mRNA and protein expression of TIMP2 in primary culture ocultures from elective terminations (n = 4). 06-3A, C, D and E indicate miscaTIMP2 expression were normalized to the corresponding ß-actin mRNA levthe mean level of each sample is represented in the scatter graph. The diffmiscarriages, as two groups, has been evaluated by student t-test (*, p < 0.0disrupt imprinting genes. Previous reports suggestedthat TIMP2 is an imprinted gene with preferential ma-ternal expression in placenta. This was based on obser-vations of reduced expression in placenta from completemoles [13], and overexpression of the maternal copy in amouse model of RPL [32]. We were therefore interestedin finding out if the maternal CNV, disrupting theTIMP2 gene, affects its allelic expression in miscarriagesthat carry the CNV. Our parent of origin expressionanalysis demonstrated biallelic expression in most casesincluding one miscarriage with the CNV. However, it ap-pears that the regulation of TIMP2 allelic expression iscomplex, as it was preferentially maternal in 2/3 miscar-riages and 2/7 ETs. It is possible that the allelic expres-sion of this gene is affected by the degree of clonalityafter placental cell culture, the ratio of methylation ofCpG islands (Chernov, et al. [33]), or additional geneticpolymorphisms affecting gene expression on one orf human chorionic villi. ET15, 17, 18 and 20 indicate control cellrriages of the female 06–1 (n = 4). (A) mRNA and (B) protein values forels. The results derived from at least three repeats of each sample anderence between the four control samples and the three CNV carrier5).marks at the TIMP2 gene promoter region(s) in relationthetheuse ianWen et al. Molecular Cytogenetics  (2015) 8:6 Page 6 of 9to expression is needed to elucidate whether this gene isindeed imprinted and which modifications are importantfor regulation of gene expression.Conclusionthe other copy. Comprehensive analysis of epigeneticFigure 3 Allelic expression of TIMP2 gene. (A) schematic diagram ofand transcription start site (TSS). Important genomic features, includingfamily, and a polymorphic coding single nucleotide polymorphism (SNP)gene expression of the major allele for rs2277698 in exon 3 of TIMP2 genand miscarriages from recurrent miscarriage family 06. Miscarriage 06-3COverall, our findings underscore the need for additionalfunctional characterization of miscarriage CNVs todevelop an understanding of the effect of their integralgenes on pregnancy development. These studies can bechallenging due to lack of miscarriage cell cultures orRNA/protein for functional studies in miscarriages andlack of parental reproductive tissues in cases withparental CNVs. Collection of reproductive tissues (e.g.maternal decidua) for functional analysis would bedesirable in future miscarriage CNV analysis. As moremiscarriage associated CNVs and genes are identified,their individual and collective role in miscarriage willbecome more apparent.MethodsSubjectsControl pregnancy and miscarriage tissuesControl chorionic villus samples were obtained fromfirst trimester elective terminations (ET) of pregnancyfor social reasons by dilation and curettage (6–12 weeksof gestation). The control ET had no evidence of aneu-ploidy as determined by multiplex ligation dependentprobe amplification (MLPA) [34]. Four ET tissues wereused for tissue culture, as previously described [35].Thirty five uncultured ET tissues were assessed forallelic expression analysis of TIMP2 gene (see below).Chorionic villus cell cultures of 10 miscarriages from 6families (03,05,06,07,09 and 10 as described in Table 1 inthe current paper and in Rajcan-Separovic et al. [12])were available for RNA and protein expression analysis.Culture conditions were as for the ET tissues [35].Maternal contamination has been ruled out in all ETstructure of the TIMP2 gene region, including its 5 exons (grey bars)copy number variation (CNV) identified in the recurrent miscarriageed to assess TIMP2 allelic expression have been labelled. (B) Percentagen informative chorionic villi from 1st trimester elective terminations (ET)d 06-3D had the CNV while miscarriage 06-3E did not.and miscarriage cultures by examining microsatellitemarkers, using standard protocols [12,36]. The use ofparental blood samples, control and miscarriage tissueswas approved by the Committee for Ethical Reviewof Research involving Human Subjects, University ofBritish Columbia and Institutional Review Board of theUniversity of Chicago. All subjects gave written informedconsent for these studies.Expression analysisRNA, DNA and protein extractionTotal RNA, DNA and protein were extracted from thechorionic villus (CV) cell cultures simultaneously usingcommercially available kits (AllPrep DNA/RNA Minikit, Qiagen) according to the manufacturer’s instruc-tions. The purity and concentration of total RNA, DNAand protein present in each of these extracts were quanti-fied using a NanoDrop 1000 Spectrophotometer (ThermoScientific, Wilmington, USA).For allelic expression analysis, RNA and DNA wereextracted from ET tissue stored in RNA later. GenomicDNA (gDNA) was used to obtain fetal genotype andinformative (heterozygous) samples were used for theallelic expression analysis.Wen et al. Molecular Cytogenetics  (2015) 8:6 Page 7 of 9cDNA synthesisAliquots of the total RNA extracts (~500 ng) preparedfrom the CV cell cultures were subsequently reverse-transcribed into cDNA using GeneAmp Gold RNA PCRCore Kit (Applied Biosystems, Melbourne, Australia).High Capacity cDNA Reverse Transcription Kit (AppliedBiosystems, Melbourne, Australia) was used to generatecDNA from 35 ET tissues for TIMP2 allelic expressionanalysis.Real-time quantitative (q)PCRRNA expression analysis was performed in culturedchorionic villi from miscarriages with CNVs and fourETs. Fourteen genes integral to the miscarriage CNVswere selected, and bioinformatics tools and publiclyavailable human genome databases (Ensemble GenomeBrowser, UCSC) are used to select appropriate primers.The genes for expression analysis were selected based onthe availability of cells/RNA/protein from miscarriagesthat contained their copy number change and included:PARK2, NDUFAF2, TIMP2, LIPC, CTNNA3, EGFL6,GPM6B, OFD1, RAB9A, TRAPPC2, CHSY3, PRMT3,POU6F2, and C7orf10 (Table 1). The nucleotide se-quences for primers specific for the 14 genes or thehousekeeping gene β-actin were designed using PrimerExpress software (Perkin-Elmer Applied Biosystems)and purchased from Integrated DNA Technologies(IDT). Primer sequences for tested regions are listedin Additional file 1: Table S2. The first-strand cDNAgenerated from the chorionic villi cells served as a tem-plate for qRT-PCR using the ABI PRISM 7300 SequenceDetection System (Perkin-Elmer Applied Biosystems)equipped with a 96-well optical reaction plate for primersspecific for 14 selected genes and the housekeeping gene,β-actin. Real-time qPCR was performed as previously de-scribed [37].Western blot analysisWe detected changes in RNA expression for 3 genes inmiscarriages: OFD1 and TRAPPC2 from Xp22 CNV(present in female 09–1 and one out of two miscar-riages) and TIMP2 from 17q25 CNV (present in female6–1 and 4/5 miscarriages). Protein expression analysisfor these genes was performed using polyclonal anti-bodies directed against human OFD1 (Abcam, No.ab97861, ON, Canada), TRAPPC2 (generous gift fromDr De Matteis, Italy) and TIMP2 (NovusBiologicals, No.NB100-92000, Littleton, US). To standardize theamounts of protein loaded into each lane, the blots werereprobed with a monoclonal antibody directed againsthuman β-actin (Novus Biologicals, Littleton, US). TheECL Western Blotting system was used to detect theamount of each antibody bound to antigen and theresultant photographic films were analyzed by UVdensitometry (GE Healthcare Life Sciences, Pittsburgh,USA). The absorbance values obtained for OFD1,TRAPPC2 or TIMP2 were then normalized relative to thecorresponding β-actin absorbance value. The average ofOFD1, TRAPPC2 and TIMP2 protein expression were ob-tained from 3 independent replicates for each miscarriageand ET control samples.Allelic expression analysis of TIMP2 geneWe assessed the effect of the CNV on allelic expressionin miscarriages from female 6–1 rather than methylationsince the TIMP2 promoter is unmethylated in cervixand blood [38], human term placenta (unpublished data)and mouse placenta throughout pregnancy [39]. Threeheterozygous SNPs were identified within the exons of theTIMP2 gene using UCSC Genome Browser: rs7503726and rs2277698 in the 5′UTR and rs2277698 within exon 3.Genotyping assays were designed for all three SNPs, usingthe PSQ Assay Design Software; with successful targetedamplification for only the rs2277698 G/A SNP inexon 3 (heterozygosity = 0.241 ± 0.250) (Additional file 1:Table S3). Genotyping was done on the Pyromark MDmachine using the PyroGold SQA reagent kit (Qiagen,Hilden, Germany), using gDNA from placenta to obtainfetal genotypes. Genotyping was performed on DNA from35 ET samples and 5 miscarriages from female 6–1 withRPL, 4/5 carried the TIMP2 CNV. 7/35 ET and 3/5 testedmiscarriages were informative (heterozygous). Parental ori-gin of the rs2277698 alleles in ET and miscarriage sampleswas determined by assessing maternal genotypes in DNAextracted from the placental decidua or blood. The sequen-cing primer used for the gDNA assay (Additional file 1:Table S3), was then used to assess whether there was allelicbias in expression of TIMP2 in the cDNA, after cDNA-specific amplification. The relative percentage of expressedalleles was obtained using the allelic quantification settingon the Pyromark MD software, and averaging of two inde-pendent replicates for each sample.Statistical analysisAll statistical tests were performed using the VassarStats:Statistical Computation Web Site (Vassar College,Poughkeepsie, USA), R Statistical Software 2.12.0 (TheR Project for Statistical Computing, Auckland, NewZealand) or GraphPad Prism 4 computer software(GraphPad, San Diego, CA). Student’s t-test was usedto determine significant differences in transcript levelsbetween the four miscarriages from family 6 with avail-able cultures (3 that carry the TIMP2 CNV and one with-out it) and four control ETs. p < 0.05 was consideredsignificant. Significance between the variances of the twogroups was determined using F- test.The absorbance values obtained from the real-timeqPCR products and the photograph generated byWen et al. Molecular Cytogenetics  (2015) 8:6 Page 8 of 9Western blotting were subjected to statistical analysisusing GraphPad Prism 4 computer software (GraphPad,San Diego, CA). Statistical significance between the ab-sorbance values were assessed by the analysis of variance(ANOVA), with p < 0.05 considered significant. Compari-son of the means between patients and controls were de-termined using Dunnett’s test. The results are presentedas the mean ± S.E.M. from at least three independentexperiments.Additional fileAdditional file 1: Table S1. Pathology description of the recurrentmiscarriages with familial CNV. Table S2. Primer sequences for Real-timeqPCR. Table S3. Primers used to assess allelic expression of TIMP2.Competing interestsThe authors declare that they have no competing interest.Authors’ contributionsERS: Conception, design, interpretation of data, drafting the article andrevising it critically for important intellectual content. JW, CWH: designed andconducted experiments, analysed and interpreted data, drafted and revisedthe article. SM: conducted experiment, analysed and interpreted data. PCKL,SMEL, WPR, MS: revised the article for important intellectual content. Allauthors read and approve the final manuscript.AcknowledgementsThe study was funded by CIHR (grant MOP 106467) and MSFHR CareerScholar salary award to ERS. We appreciate the generous gift of the TRAPPC2antibody from Dr De Matteis (Telethon Institute of Genetics and Medicine,Via Pietro Castellino 111, Naples 80131, Italy).Author details1Department of Pathology and Laboratory Medicine, University of BritishColumbia, Vancouver V6T 2B5, Canada. 2Child & Family Research Institute,Vancouver V5Z 4H4, Canada. 3Department of Medical Genetics, University ofBritish Columbia, Vancouver V6T 1Z3, Canada. 4Department of Obstetrics andGynaecology, University of British Columbia, Vancouver V6Z 2 K5, Canada.5Department of Obstetrics and Gynecology, University of Illinois at Chicago,Chicago 60612, USA.Received: 13 December 2014 Accepted: 9 January 2015References1. Warren JE, Silver RM. Genetics of pregnancy loss. Clin Obstet Gynecol.2008;51(1):84–95.2. Group ECW. Genetic aspects of female reproduction. Hum Reprod Update.2008;14(4):293–307.3. van den Berg MM, van Maarle MC, van Wely M, Goddijn M. Genetics ofearly miscarriage. Biochim Biophys Acta. 2012;1822(12):1951–9.4. Stembalska A, Slezak R, Pesz K, Gil J, Sasiadek M. Prenatal diagnosis–principlesof diagnostic procedures and genetic counseling. Folia Histochem Cytobiol.2007;45 Suppl 1:S11–6.5. Stephenson M, Kutteh W. Evaluation and management of recurrent earlypregnancy loss. Clin Obstet Gynecol. 2007;50(1):132–45.6. Miller DT, Adam MP, Aradhya S, Biesecker LG, Brothman AR, Carter NP, et al.Consensus statement: chromosomal microarray is a first-tier clinical diagnostictest for individuals with developmental disabilities or congenital anomalies. AmJ Hum Genet. 2010;86(5):749–64.7. Kang JU, Koo SH. Evolving applications of microarray technology inpostnatal diagnosis (review). Int J Mol Med. 2012;30(2):223–8.8. Levy B, Sigurjonsson S, Pettersen B, Maisenbacher MK, Hall MP, Demko Z,et al. Genomic imbalance in products of conception: single-nucleotidepolymorphism chromosomal microarray analysis. Obstet Gynecol.2014;124(2 Pt 1):202–9.9. Larsen EC, Christiansen OB, Kolte AM, Macklon N. New insights intomechanisms behind miscarriage. BMC Med. 2013;11:154.10. Nagirnaja L, Palta P, Kasak L, Rull K, Christiansen OB, Nielsen HS, et al.Structural genomic variation as risk factor for idiopathic recurrentmiscarriage. Hum Mutat. 2014;35(8):972–82.11. Rajcan-Separovic E, Qiao Y, Tyson C, Harvard C, Fawcett C, Kalousek D, et al.Genomic changes detected by array CGH in human embryos withdevelopmental defects. Mol Hum Reprod. 2010;16(2):125–34.12. Rajcan-Separovic E, Diego-Alvarez D, Robinson WP, Tyson C, Qiao Y,Harvard C, et al. Identification of copy number variants in miscarriagesfrom couples with idiopathic recurrent pregnancy loss. Hum Reprod.2010;25(11):2913–22.13. Okamoto T, Niu R, Yamada S, Osawa M. Reduced expression of tissueinhibitor of metalloproteinase (TIMP)-2 in gestational trophoblastic diseases.Mol Hum Reprod. 2002;8(4):392–8.14. Venditti R, Scanu T, Santoro M, Di Tullio G, Spaar A, Gaibisso R, et al. Sedlincontrols the ER export of procollagen by regulating the Sar1 cycle. Science.2012;337(6102):1668–72.15. Ferrante MI, Zullo A, Barra A, Bimonte S, Messaddeq N, Studer M, et al. Oral-facial-digital type I protein is required for primary cilia formation and left-right axis specification. Nat Genet. 2006;38(1):112–7.16. Aplin AC, Fogel E, Zorzi P, Nicosia RF. The aortic ring model ofangiogenesis. Methods Enzymol. 2008;443:119–36.17. Heino J. The collagen family members as cell adhesion proteins. Bioessays.2007;29(10):1001–10.18. Hunkapiller J, Singla V, Seol A, Reiter JF. The ciliogenic protein oral-facial-digital 1 regulates the neuronal differentiation of embryonic stem cells.Stem Cells Dev. 2011;20(5):831–41.19. Cortellino S, Wang C, Wang B, Bassi MR, Caretti E, Champeval D, et al.Defective ciliogenesis, embryonic lethality and severe impairment of theSonic Hedgehog pathway caused by inactivation of the mouse complex Aintraflagellar transport gene Ift122/Wdr10, partially overlapping with theDNA repair gene Med1/Mbd4. Dev Biol. 2009;325(1):225–37.20. Ferrante MI, Romio L, Castro S, Collins JE, Goulding DA, Stemple DL, et al.Convergent extension movements and ciliary function are mediated byofd1, a zebrafish orthologue of the human oral-facial-digital type 1 syndromegene. Hum Mol Genet. 2009;18(2):289–303.21. Harman RM, Cowan RG, Ren Y, Quirk SM. Reduced signaling through thehedgehog pathway in the uterine stroma causes deferred implantation andembryonic loss. Reproduction. 2011;141(5):665–74.22. Chen Q, Zhang Y, Lu J, Wang Q, Wang S, Cao Y, et al. Embryo-uterine cross-talk during implantation: the role of Wnt signaling. Mol Hum Reprod.2009;15(4):215–21.23. Mohamed OA, Dufort D, Clarke HJ. Expression and estradiol regulation ofWnt genes in the mouse blastocyst identify a candidate pathway forembryo-maternal signaling at implantation. Biol Reprod. 2004;71(2):417–24.24. Diao H, Aplin JD, Xiao S, Chun J, Li Z, Chen S, et al. Altered spatiotemporalexpression of collagen types I, III, IV, and VI in Lpar3-deficient peri-implantation mouse uterus. Biol Reprod. 2011;84(2):255–65.25. Iwahashi M, Muragaki Y, Ooshima A, Nakano R. Decreased type IV collagenexpression by human decidual tissues in spontaneous abortion. J ClinEndocrinol Metab. 1996;81(8):2925–9.26. Murphy G. Tissue inhibitors of metalloproteinases. Genome Biol. 2011;12(11):233.27. Curry Jr TE, Osteen KG. The matrix metalloproteinase system: changes,regulation, and impact throughout the ovarian and uterine reproductivecycle. Endocr Rev. 2003;24(4):428–65.28. Wang H, Wen Y, Mooney S, Li H, Behr B, Polan ML. Matrix metalloproteinaseand tissue inhibitor of matrix metalloproteinase expression in humanpreimplantation embryos. Fertil Steril. 2003;80 Suppl 2:736–42.29. Brew K, Nagase H. The tissue inhibitors of metalloproteinases (TIMPs): anancient family with structural and functional diversity. Biochim Biophys Acta.2010;1803(1):55–71.30. Hammani K, Blakis A, Morsette D, Bowcock AM, Schmutte C, Henriet P, et al.Structure and characterization of the human tissue inhibitor ofmetalloproteinases-2 gene. J Biol Chem. 1996;271(41):25498–505.31. Godbole G, Suman P, Gupta SK, Modi D. Decidualized endometrial stromal cellderived factors promote trophoblast invasion. Fertil Steril. 2011;95(4):1278–83.32. Dixon ME, Chien EK, Osol G, Callas PW, Bonney EA. Failure of decidualarteriolar remodeling in the CBA/J x DBA/2 murine model of recurrentpregnancy loss is linked to increased expression of tissue inhibitor ofmetalloproteinase 2 (TIMP-2). Am J Obstet Gynecol. 2006;194(1):113–9.33. Chernov AV, Sounni NE, Remacle AG, Strongin AY. Epigenetic control of theinvasion-promoting MT1-MMP/MMP-2/TIMP-2 axis in cancer cells. J BiolChem. 2009;284(19):12727–34.34. Diego-Alvarez D, Ramos-Corrales C, Garcia-Hoyos M, Bustamante-Aragones A,Cantalapiedra D, Diaz-Recasens J, et al. Double trisomy in spontaneousmiscarriages: cytogenetic and molecular approach. Hum Reprod.2006;21(4):958–66.35. Graham CH, Lysiak JJ, McCrae KR, Lala PK. Localization of transforminggrowth factor-beta at the human fetal-maternal interface: role in trophoblastgrowth and differentiation. Biol Reprod. 1992;46(4):561–72.36. Tyson C, Harvard C, Locker R, Friedman JM, Langlois S, Lewis ME, et al.Submicroscopic deletions and duplications in individuals with intellectualdisability detected by array-CGH. Am J Med Genet. 2005;139(3):173–85.37. Wen J, Lopes F, Soares G, Farrell SA, Nelson C, Qiao Y, et al. Phenotypic andfunctional consequences of haploinsufficiency of genes from exocyst andretinoic acid pathway due to a recurrent microdeletion of 2p13.2. OrphanetJ Rare Dis. 2013;8:100.38. Ivanova T, Vinokurova S, Petrenko A, Eshilev E, Solovyova N, Kisseljov F, et al.Frequent hypermethylation of 5' flanking region of TIMP-2 gene in cervicalcancer. Int J Cancer. 2004;108(6):882–6.39. Kim HR, Han RX, Diao YF, Park CS, Jin DI. Epigenetic characterization of thePBEF and TIMP-2 genes in the developing placentae of normal mice. BMBRep. 2011;44(8):535–40.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionWen et al. Molecular Cytogenetics  (2015) 8:6 Page 9 of 9Submit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items