De novo copy number variations in cloned dogs from the same nuclear donor
Somatic mosaicism of copy number variants (CNVs) in human body organs and de novo CNV event in monozygotic twins suggest that de novo CNVs can occur during mitotic recombination. These de novo CNV events are important for understanding genetic background of evolution and diverse phenotypes. In this study, we explored de novo CNV event in cloned dogs with identical genetic background.
We analyzed CNVs in seven cloned dogs using the nuclear donor genome as reference by array-CGH, and identified five de novo CNVs in two of the seven clones. Genomic qPCR, dye-swap array-CGH analysis and B-allele profile analysis were used for their validation. Two larger de novo CNVs (5.2 Mb and 338 Kb) on chromosomes X and 19 in clone-3 were consistently validated by all three experiments. The other three smaller CNVs (sized from 36.1 to76.4 Kb) on chromosomes 2, 15 and 32 in clone-3 and clone-6 were verified by at least one of the three validations. In addition to the de novo CNVs, we identified a 37 Mb-sized copy neutral de novo loss of heterozygosity event on chromosome 2 in clone-6.
To our knowledge, this is the first report of de novo CNVs in the cloned dogs which were generated by somatic cell nuclear transfer technology. To study de novo genetic events in cloned animals can help understand formation mechanisms of genetic variants and their biological implications.
KeywordsCloned dog De novo copy number variation Mosaicism Loss of heterozygosity Somatic cell nuclear transfer
Copy number variant
Somatic cell nuclear transfer
Microarray-comparative genomic hybridization
Quantitative polymerase chain reaction
Single nucleotide polymorphism
Loss of heterozygosity
Kilo base pair
Log R ratio
Together with single nucleotide polymorphisms (SNPs), DNA structural variations generally termed copy number variations (CNVs) are the major part of genomic variations in diverse animals including dogs [1, 2]. CNVs are thought to contribute to phenotypic diversities including disease susceptibility and provide the important substrate for evolution [3, 4]. Although it has been widely believed that all the cells in one individual are genetically identical, this dogma is changing due to our understanding of genomic variants. Indeed, recent studies have demonstrated the somatic mosaicism of CNVs in body organs [5, 6], which suggests that de novo CNV events may commonly occur during mitotic recombination . This possibility was clearly demonstrated by Bruder et al’s report identifying de novo somatic CNV events in monozygotic (MZ) twins . They showed that the de novo post-twinning CNV frequency could be as high as 5% on a per-individual basis or 10% per twinning event. Also, Breckpot et al. reported CNVs in 1 out of 6 phenotypically discordant MZ twins . These results suggest that CNV analysis in phenotypically discordant MZ twins can be useful to identify genetic loci associated with various traits. However, the de novo rate of somatic CNV mosaicism events is still largely unknown. For the case of cloned dogs, there has been no report about the de novo somatic CNV mosaicism event itself.
Dogs (Canis familiaris) have been used as working animals due to their superior sniffing ability to any other species or machine. For example, a cancer-sniffing dog can detect the existence of colorectal cancer from stool samples with 97% sensitivity and 99% specificity . However, up to 70% of dogs that were bred from working dogs are not suitable for practical use because they lack trainability . To reduce the tremendous cost, effort, and time to train one working dog, dog-cloning technology has been applied for the propagation of elite working dogs .
In this study, we aimed to identify de novo somatic CNV events which may exist in cloned dogs with identical genetic backgrounds like in human MZ twins using microarray-based comparative genomic hybridization (array-CGH). Array-CGH is one of the most popular and useful tools to analyze genome-wide CNV profiles and a number of studies have identified canine CNVs using this technique [13, 14, 15, 16, 17]. We validated the CNVs by diverse analyses such as genomic qPCR, dye-swap array-CGH, and B-allele profile analysis. We examined the genome-wide de novo somatic CNV events for the seven cloned dogs generated in our previous study  and identified five de novo CNV events.
Validation of custom canine-array
Association with segmental duplications and repetitive sequences
To understand the mechanism of CNV formation, we assessed the association of CNVs with segmental duplications (SDs) and repetitive sequences. When we compared the 72 CNVs with the CNV data from Nicholas et al’s report, which systematically analyzed SDs and associated CNVs in canis familiaris , 48 (67%) overlapped with the SDs. Total length of the overlapped CNVs was about 6 Mb (~70% of the total length of the 72 CNVs). This result was coherent with Nicholas et al’s previous report demonstrating that 70% of CNVs were associated with SDs . Considering that SDs are often substrates of CNV formation via non allelelic homologous recombination (NAHR) , ours and previous results suggest that SD-associated NAHR might be the major mechanism of canine CNV generation. When we observed the genomic fraction (fraction of retroelements per kb) of long interspersed elements (LINE), short interspersed elements (SINE) and long terminal repeats (LTR) within (intraCNV) or in the vicinity of the CNVs (up to 10 Mb), LINE elements were relatively enriched the flanking regions (up to 10 kb away from CNVs), while relatively depleted LINEs/SINEs in intraCNV (Additional file 1: Figure S2). This result is coherent with the previous reports suggesting that repeated sequences may play a role in the formation of structural variation and genomic diversity [19, 20].
De novo CNVs identified in the cloned dogs
De novo CNVs in cloned dogs
Q, S, B
ARSE, ARSH, OBP,GYG2, ARSD, ARSF, MXRA5, PRKX, PRKY, NLGN4T, NLGN4X, HDHD1, STS, PNPLA4, AKL1, TAF9, APOO, SHROOM2
Q, S, B
B-allele profile of the de novo CNVs
Genomic qPCR validation of the de novo CNVs
In this study, we aimed to identify de novo post-cloning CNV events which may exist in cloned dogs like in human MZ twins. For this purpose, we designed a 720 K custom array by merging the NimbleGen 385 K dog array as a backbone and 335 K probes for known canine CNV regions . Using this platform, we identified five de novo CNVs in two of the seven cloned dogs from the same nuclear donor. Our findings support the occurrence of de novo post-cloning CNV events in mammals including dogs. Although de novo CNV events in phenotypically concordant MZ twins have been reported in humans , to the best of our knowledge, this is the first report of de novo CNVs in cloned animals including dogs generated by SCNT technology.
As mentioned above, the de novo CNVs were unevenly identified across the dog clones (2/7 clones); four CNVs were identified in one clone (clone-3) and one was in another clone (clone-6). Our result is similar with the previous reports observing the CNVs in human MZ twins. Breckpot et al. identified CNVs in 1 out of 6 phenotypically discordant MZ twins . In Ehli et al’s report with 50 concordant and discordant MZ twins, 18 de novo post-twining CNVs were detected in 14 out of 45 twin pairs . However, we cannot rule out the possibility of existence of more de novo CNVs because the array platform and defining algorithm we used in this study cannot be perfect. Also, it is possible that some genetic events may have happened in the clone-3 harboring 4 CNVs during its development or SCNT procedure. Further CNV studies with a larger number of cloned dogs will be required to understand this phenomenon.
When we screened CNVs between the donor Labrador retriever and a Boxer to validate the performance of our custom array, the number of CNVs was largely similar to the report by Nicholas et al. that compared CNVs using a 2.1 M array between nine breeds of modern domestic dogs including Labrador retriever . Most of the CNVs (90.3%) identified in our validation experiment overlapped with the previously reported CNVs [14, 15, 16, 17], and the overlap was significantly not random event (P < 0.001). All the nine randomly selected CNV loci and seven novel CNVs identified by array-CGH showed consistent copy number changes that were validated by qPCR. These results suggest the reliability of our custom array and CNV defining procedure. Although exploring the novel canine CNVs is not our aim in this study, this result suggests that a substantial number of canine CNVs that have yet to be identified.
To validate the five de novo CNVs in the cloned dogs, we performed three different verification experiments; genomic qPCR, dye-swap array-CGH analysis and B-allele profile analysis. Two larger de novo CNVs (5.2 Mb and 338 Kb) on chromosomes X and 19 in clone-3 were consistently validated by all three experiments. The other three CNVs on chromosomes 2, 15, and 32 (36.1 - 76.4 Kb) seem to be too small to be interpreted by B-allele pattern (Additional file 1: Figure S3). According to the technical note about CNV analysis algorithm for Illumina HumanHap240S array, which has a higher SNP density than the array used in this study (170 K), the minimum recommended window size for LOH score algorithm is 360 Kb . Of the three smaller-sized CNVs, two (a CNV on chromosome 2 in clone-6 and another one on chromosome 15 in clone-3) were verified by qPCR, but the CNV on the distal end of chromosome 32 was not verified. However, the CNV region on chromosome 32 was validated in dye-swap analysis, which suggests that the 76 Kb-sized CNV region on chromosome 32 might be real.
There are several genes in the 5.2 Mb-sized copy number gain CNV locus including the neuroligin 4 X-linked (NLGN4X), steroid sulfatase isozyme S (STS), and odorant-binding protein (OBP) genes. In human, genetic variants of the NLGN4X gene have been reported to be associated with autism spectrum disorder in Japanese and Han Chinese populations [25, 26, 27]. A CNV of the STS gene (duplication) has also been reported to be associated with intellectual disability in human . OBP is small soluble carrier protein that belongs to the family of lipocalins and that may be involved in the perception of pheromones . However, we did not observe any differences regarding the phenotypes related to these genes in the cloned dogs harboring the de novo CNVs in this study.
In addition to the de novo CNVs, we also detected a copy neutral de novo LOH event on chromosome 2. This 37 Mb-sized LOH event occurred in a cloned dog but was not present in the nuclear donor dog, indicating that the uniparental disomy (UPD) arose from a mitotic event in the cloned dog. It is well known that UPD is a genetic event associated with human diseases and mosaicism of aneuploidy can be associated with UPD . Our results suggest that SCNT could result in UPD-related phenotypic differences among clones.
There are several limitations in this study. First, since the sample size of this study is very small, the scope of our analysis could not go beyond describing de novo CNVs. Second, even though we identified de novo CNVs in cloned dogs, we could not analyze any association between the CNVs and phenotypic characteristics, because there have been no phenotypic differences among them so far. To overcome both limitations and draw biologically meaningful conclusions, we need to clone more dogs and to carefully observe the phenotypes of them, and apply more informative technologies such as whole genome sequencing.
In conclusion, we identified de novo CNV and LOH events in cloned dogs generated from one nuclear donor. To the best of our knowledge, this is the first report of de novo CNVs in cloned animals including dogs generated by SCNT technology. Since cloned animals share almost identical genetic background like MZ twins, to study their de novo genetic events can help understand formation mechanisms of genetic variants and their biological implications. Also, our results can be a useful tool to study the genetic basis of diverse disease/phenotype traits in dogs.
Source of dog DNA samples
Seven cloned Labrador retriever dogs named clone-1 to -7 were generated by somatic cell nuclear transfer (SCNT) using fibroblasts derived from a seven-year-old Labrador retriever, and the genetic homogeneity was validated by microsatellite analysis using nine markers (PEZ01, PEZ02, PEZ03, PEZ06, PEZ13, PEZ17, FH2079, FH2054, FH2010) in the previous study . In order to extract genomic DNA samples, blood was collected from the seven cloned dogs at three years of age and from the donor dog at ten years of age. Approximately 5 ml of blood was collected into anticoagulant tubes containing EDTA. Genomic DNA was extracted from whole-blood using a DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) and quantified by NanoDrop spectrophotometer. This study was performed with the approval of animal ethics committee of Seoul National University.
Whole-genome array-CGH experiment
We used a NimbleGen custom 720K canine whole-genome array platform (Roche NimbleGen, Penzberg, Germany), designed from the UCSC dog genome build of May 2005 (Broad/canFam2). This array is comprised of 385,000 probes evenly distributed throughout the unique sequence of the genome as a backbone and an additional ~330,000 probes which were designed based on previously reported dog CNV regions . The average probe spacing for our custom array platform was ~3.3 Kb. The array-CGH experiments were performed according to the manufacturer’s instructions. In brief, 1 μg genomic DNA from each cloned dog was labeled with Cy3-dCTP. The reference DNA was labeled with Cy5-dCTP. Each dye-labeled DNA was purified and mixed with hybridization reagents, applied on the array and incubated for 48 hours at 42°C in a MAUI hybridization machine (BioMicro, Salt Lake City, UT). After washing the slides, arrays were scanned using a GenePix 4000B scanner (Axon Instruments, Sunnyvale, CA). Image separation, signal intensity extraction, and normalization were performed using the NimbleScan software version 2.5 (Roche NimbleGen).
Detection of CNVs
The Rank-Segmentation CNV calling algorithm in NEXUS software v3.1 (BioDiscovery Inc., El Segundo, CA) was used to define CNVs of each sample. The parameters for defining CNVs were as follows: significance threshold = 1.0E-6; maximum contiguous probe spacing (Kbp) = 1000 Kb; minimum number of contiguous probes per CNV segment = 5; threshold of signal intensity ratio > 0.2 on the log2 scale for gains and < -0.2 on the log2 scale for losses.
Whole-genome SNP genotyping
Genome-wide SNP genotyping was conducted using the CanineHD SNP array, which contains 173,662 SNP markers (Illumina, Inc., San Diego, CA) according to manufacturer’s instructions. The average probe spacing is 13 Kb. The raw data consisted of the signal intensity (log R ratio: LRR) and allelic intensity (B allele frequency) that were obtained using GenomeStudio software (Illumina, Inc.). Based on the B allele frequency profile, we defined the CNVs according to the concept of integrating SNP and CNV genotyping . Samples that did not have a standard deviation (SD) of LRR < 0.24 were excluded for quality control. The SNPRank Segmentation algorithm in NEXUS software was used to define allele imbalance of each sample.
Genomic quantitative PCR analysis
Array-CGH results were experimentally validated by genomic qPCR in randomly selected CNV loci for boxer and all CNV loci for cloned dogs. As a diploid internal control, a genomic region on chr7:18946362 that showed no genomic alteration in the array-CGH data was used. Details including primer information for targets and the diploid control locus are available in Additional file 1: Table S3. Genomic qPCR was performed using the ViiA7™ real-time PCR system (Life Technologies, Carlsbad, CA) as describe elsewhere . In brief, a 10 μl of reaction mixture contained 10 ng of genomic DNA, THUNDERBIRD™ SYBR qPCR Mix (TOYOBO, Osaka, Japan), 1× ROX, and 6 pmole of each primer. Thermal cycling conditions consisted of one cycle of 1 min at 95°C followed by 40 cycles of 5 sec at 95°C, 10 sec at 61°C, and 20 sec at 72°C. All qPCR experiments were triplicate and the copy number of each target was defined as 2-ΔΔCT, where ΔCt is the difference in threshold cycles of the sample in question normalized against the reference region and expressed relative to the value obtained by calibrator DNA (individual/calibrator) as describe elsewhere . DNA from the nuclear donor dog was used as calibrator for all qPCR experiment.
Probability of identity
Probability of identity (PI) for multilocus microsatellite data was computed using GENECAP program  under the two different assumptions of Hardy-Weinberg equilibrium and the sibling PI as described previously [21, 22]. Where, PI indicates the probability of two individuals within the population sharing the same genotype. In GENECAP, multilocus PI values were computed by mutiplying all locus-specific PIs under the assumption that each locus is independent.
Significance of enriched CNV overlap with known dog CNV data
Statistical significance of enriched CNV overlap with known dog CNV data was evaluated by permutation test as described previously . In this test, original CNV overlap with database was calculated based on >1 bp overlap measure. In each iteration, CNVs were randomly reshuffled while maintaining their sizes and numbers within each chromosome. Based on these permuted CNVs, we calculated the overlap with the same database. After repeating this process 1000 times, we calculated P-value as the fraction of tests where the number of overlaps was greater than the original unpermuted CNV overlap.
Statistical analyses were performed using SPSS version 18.0 (SPSS Inc., Chicago, IL). P values < 0.05 were considered significant.
Availability of supporting data
The array-CGH data (NimbleGen custom 720 K canine whole-genome array) and the SNP genotyping data (Illumina CanineHD SNP array) from this study are submitted to the NCBI Gene Expression omnibus (http://www.ncbi.nlm.nih.gov/geo) under accession no GSE49092 and GSE49123, respectively. The following supporting data are available with the online version of this paper.
This study was supported by a grant from MRC (2012R1A5A2047939), Korea Healthcare Technology R&D Project (A092258), IPET (#311062-04-2-SB010) and RDA (PJ008975022013), Republic of Korea.
- 4.Lindblad-Toh K, Wade CM, Mikkelsen TS, Karlsson EK, Jaffe DB, Kamal M, Clamp M, Chang JL, Kulbokas EJ, Zody MC, Mauceli E, Xie X, Breen M, Wayne RK, Ostrander EA, Ponting CP, Galibert F, Smith DR, DeJong PJ, Kirkness E, Alvarez P, Biagi T, Brockman W, Butler J, Chin CW, Cook A, Cuff J, Daly MJ, DeCaprio D, Gnerre S, et al: Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature. 2005, 438: 803-819. 10.1038/nature04338.CrossRefPubMedGoogle Scholar
- 5.Piotrowski A, Bruder CE, Andersson R, Diaz de Ståhl T, Menzel U, Sandgren J, Poplawski A, von Tell D, Crasto C, Bogdan A, Bartoszewski R, Bebok Z, Krzyzanowski M, Jankowski Z, Partridge EC, Komorowski J, Dumanski JP: Somatic mosaicism for copy number variation in differentiated human tissues. Hum Mutat. 2008, 29: 1118-1124. 10.1002/humu.20815.CrossRefPubMedGoogle Scholar
- 7.Mkrtchyan H, Gross M, Hinreiner S, Polytiko A, Manvelyan M, Mrasek K, Kosyakova N, Ewers E, Nelle H, Liehr T, Bhatt S, Thoma K, Gebhart E, Wilhelm S, Fahsold R, Volleth M, Weise A: The human genome puzzle - the role of copy number variation in somatic mosaicism. Curr Genomics. 2010, 11: 426-431. 10.2174/138920210793176047.PubMedCentralCrossRefPubMedGoogle Scholar
- 8.Bruder CE, Piotrowski A, Gijsbers AA, Andersson R, Erickson S, Diaz de Ståhl T, Menzel U, Sandgren J, von Tell D, Poplawski A, Crowley M, Crasto C, Partridge EC, Tiwari H, Allison DB, Komorowski J, van Ommen GJ, Boomsma DI, Pedersen NL, den Dunnen JT, Wirdefeldt K, Dumanski JP: Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet. 2008, 82: 763-771. 10.1016/j.ajhg.2007.12.011.PubMedCentralCrossRefPubMedGoogle Scholar
- 18.Conrad DF, Pinto D, Redon R, Feuk L, Gokcumen O, Zhang Y, Aerts J, Andrews TD, Barnes C, Campbell P, Fitzgerald T, Hu M, Ihm CH, Kristiansson K, Macarthur DG, Macdonald JR, Onyiah I, Pang AW, Robson S, Stirrups K, Valsesia A, Walter K, Wei J, Tyler-Smith C, Carter NP, Lee C, Scherer SW, Hurles ME, Wellcome Trust Case Control Consortium: Origins and functional impact of copy number variation in the human genome. Nature. 2010, 464: 704-712. 10.1038/nature08516.PubMedCentralCrossRefPubMedGoogle Scholar
- 22.Wilberg MJ, Dreher BP: GENECAP: a program for analysis of multilocus genotype data for non-invasive sampling and capture-recapture population estimation. Mol Ecol Resour. 2004, 4: 783-785.Google Scholar
- 23.Ehli EA, Abdellaoui A, Hu Y, Hottenga JJ, Kattenberg M, van Beijsterveldt T, Bartels M, Althoff RR, Xiao X, Scheet P, de Geus EJ, Hudziak JJ, Boomsma DI, Davies GE: De novo and inherited CNVs in MZ twin pairs selected for discordance and concordance on Attention Problems. Eur J Hum Genet. 2012, 20: 1037-1043. 10.1038/ejhg.2012.49.PubMedCentralCrossRefPubMedGoogle Scholar
- 24.DNA Copy Number Analysis Algorithms. http://www.pasteur.fr/ip/portal/action/WebdriveActionEvent/oid/01s-00003f-00c,
- 25.Marshall CR, Noor A, Vincent JB, Lionel AC, Feuk L, Skaug J, Shago M, Moessner R, Pinto D, Ren Y, Thiruvahindrapduram B, Fiebig A, Schreiber S, Friedman J, Ketelaars CE, Vos YJ, Ficicioglu C, Kirkpatrick S, Nicolson R, Sloman L, Summers A, Gibbons CA, Teebi A, Chitayat D, Weksberg R, Thompson A, Vardy C, Crosbie V, Luscombe S, Baatjes R, et al: Structural variation of chromosomes in autism spectrum disorder. Am J Hum Genet. 2008, 82: 477-488. 10.1016/j.ajhg.2007.12.009.PubMedCentralCrossRefPubMedGoogle Scholar
- 28.Li F, Shen Y, Köhler U, Sharkey FH, Menon D, Coulleaux L, Malan V, Rio M, McMullan DJ, Cox H, Fagan KA, Gaunt L, Metcalfe K, Heinrich U, Hislop G, Maye U, Sutcliffe M, Wu BL, Thiel BD, Mulchandani S, Conlin LK, Spinner NB, Murphy KM, Batista DA: Interstitial microduplication of Xp22.31: Causative of intellectual disability or benign copy number variant?. Eur J Med Genet. 2010, 53: 93-99. 10.1016/j.ejmg.2010.01.004.CrossRefPubMedGoogle Scholar
- 30.Conlin LK, Thiel BD, Bonnemann CG, Medne L, Ernst LM, Zackai EH, Deardorff MA, Krantz ID, Hakonarson H, Spinner NB: Mechanisms of mosaicism, chimerism and uniparental disomy identified by single nucleotide polymorphism array analysis. Hum Mol Genet. 2010, 19: 1263-1275. 10.1093/hmg/ddq003.PubMedCentralCrossRefPubMedGoogle Scholar
- 32.Kim JH, Jung SH, Bae JS, Lee HS, Yim SH, Park SY, Bang SY, Hu HJ, Shin HD, Bae SC, Chung YJ: Deletion variants of RABGAP1L, 10q21.3, and C4 are associated with the risk of systemic lupus erythematosus in Korean women. Arthritis Rheum. 2013, 65: 1055-1063. 10.1002/art.37854.CrossRefPubMedGoogle Scholar
- 33.GENECAP Program. http://wilberglab.cbl.umces.edu/downloads.html,
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.