Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jun 20.
Published in final edited form as: J Med Genet. 2011 May 27;48(7):438–443. doi: 10.1136/jmg.2010.085944

Replacement of the myotonic dystrophy type 1 CTG repeat with “non-CTG repeat” insertions in specific tissues

Michelle M Axford 1,2,#, Arturo López Castel 1,3,#, Masayuki Nakamori 4, Charles A Thornton 4, Christopher E Pearson 1,2,*
PMCID: PMC3379714  NIHMSID: NIHMS373246  PMID: 21622935

Abstract

Recently, curious mutations have been reported to occur within the (CTG)n repeat tract of the myotonic dystrophy type 1 (DM1) locus. For example the repeat, long-presumed to be a pure repeat sequence has now been revealed to often contain interruption motifs in a proportion of cases with expansions. Similarly, a few de novo somatic CTG expansions have been reported to arise from non-expanded DM1 alleles with 5–37 units, thought to be genetically stable. We have characterized a novel mutation configuration at the DM1 CTG repeat that arose as somatic mosaicism in a juvenile onset DM1 patient with a non-expanded allele of (CTG)12 and tissue-specific expansions ranging from (CTG)1100–6000. The mutation configuration replaced the CTG tract with a non-CTG repeat insertion of 43 or 60 nucleotides, precisely placed in the position of the CTG tract with proper flanking sequences. The inserts appeared to arise from a longer human sequence on chromosome 4q12, and may have arisen through DNA structure-mediated somatic inter-gene recombination or replication fork template-switching errors. De novo insertions were detected in cerebral cortex and skeletal muscle, but not in heart or liver. Repeat tracts with −1 or −2 CTG units were also detected in cerebellum, which may have arisen by contractions of the short (CTG)12 allele. This non-CTG configuration expands our understanding of the sequence variations that can arise at this hypermutable site.

Keywords: myotonic dystrophy, non-CTG repeat, de novo mutation, insertions trinucleotide repeats

Introduction

Myotonic dystrophy type 1 (DM1) is caused by the expansion of a CTG repeat located in the 3’-UTR of the DMPK gene on chromosome 19 (19q13.3). DM1 is inherited in an autosomal dominant fashion with disease-causing expansions of >50 up to 6550 repeats arising in one allele [1]. Expansion of the unstable repeat occurs through consecutive generations and in somatic tissues during the lifetime of an affected individual. However, the short non-expanded allele is genetically stable. The mechanism(s) involved in a non-expanded CTG repeat tract (5 to 35 units) reaching the status of a highly unstable sequence (>50 units) is poorly understood. Occasional expansion-biased mutations of non-expanded CTG alleles in DMPK gene have been reported [2,3,4], usually involving the high end of the normal size range (19–37 repeats), although in one recent case the normal allele that expanded was only 12 repeats expanding to >50 repeats [4]. Aside from repeats expansions that arise through additions of more CTG units, the DM1 CTG tract has been shown to contain interspersed non-CTG units [5,6,7]. Somatic mosacism of the HD CAG tract [simultaneous presence of (CAG)20, (CAG)37 and (CAG)47] has been reported [8]. These rare events in DM1 and Huntington’s disease (HD) have in some instances been detected in specific tissues [3,4,8]. However, it should be noted such events are rare and alternative explanations, such as contamination were not ruled out; hence can be described as ‘putative’ CTG/CAG changes. Here, we report the presence of aberrant sequences that replace the CTG tract at the DM1 locus. These non-CTG configurations arise as somatic events in specific tissues, and appear to be derived from a region of chromosome 4.

Methods

DNA extraction

UR-ADM9 DNA from different tissues was extracted as described in [1].

DNA amplification and electrophoresis

Amplification through the DMPK CTG repeat tract in UR-ADM9 samples was performed using PCR primers 409 (5'-GAAGGGTCCTTGTAGCCGGGAA-3') and 407 (5'-CAGAGCAGGCGTCATGCACA-3') at an annealing temperature of 67 °C for 2 minutes, over 30 cycles. 4% non-denaturing acrylamide gels and 1% agarose gels were run to resolve the PCR products.

Amplification using the primers designed based on the areas of homology found between the non-CTG insert in the skeletal muscle and cortex (3PC: 5'-CACCGGGTGGGTTACACC-3' and 5PSC: 5'-CTGGCCTTAGCCACGCCAC-3') was performed at an annealing temperature of 63 degrees for 45 seconds, over 30 cycles. Products were run on a 1% agarose gel.

Sequencing

Chosen PCR products were sent routinely to the sequencing facility of The Hospital for Sick Children (The Centre for Applied Genomics) after a step of gel excision/cleaning (QIAquick PCR purification kit, Qiagen) and cloning (TOPO-TA 4.0 cloning kit, Invitrogen). A minimum of 10 clones were performed for each PCR product analyzed. Each sequence was proofread against the electropherogram to avoid base misspelling.

DNA alignment and sequence location

Non-repetitive DNA sequences present in the repeat location were aligned against human, bacteria (E.coli) and TOPO vector DNAs using the Blast2Align tool available at www.ncbi.nlm.nih.gov.

DNA sequences found from the insert-specific primer PCRs were searched against the human genome using the BLAST tool at www.ncbi.nlm.nih.gov.

Methylation Status

The methylation status of the allele on which the non-CTG insert is located was carried out using SacII digestion followed by multiplex PCR (as described in [1]). 200ng of UR-ADM9 patient samples were digested with 10 units of SacII over night, then subjected to a multiplex PCR protocol. Multiplex amplification across the DMPK CTG repeat tract was performed using PCR primers 409 (5'-GAAGGGTCCTTGTAGCCGGGAA-3'), CTCFIa (5'-CTGCCAGTTCACAACCGCTCCGAG-3') and CTCFIIb (5'-AAAGCAAATTTCCCGAGTAAGCAGGC-3'). 4% non-denaturing acrylamide gels were used to resolve PCR products.

Results

During our analysis of CTG repeat length variations between DM1 patient tissues [1] we found that while characterizing the non-expanded allele by PCR several aberrant PCR products arose in several tissues of a DM1 patient. The aberrant PCR products, in addition to the main PCR product, arose after the amplification of the short DMPK allele of a 44 year-old-woman (UR-ADM9) diagnosed with a juvenile form of DM1 disease (Fig. 1). This individual had expanded alleles ranging from 1100–6000 repeats [1]. The appearance of unexpected PCR products suggests their formation in only a fraction of the cells from the tissues affected. The detection was achieved in cerebral cortex, skeletal muscle and cerebellum of the UR-ADM9 patient during the DMPK alleles repeat sizing in several tissues from up to 8 different DM1 affected individuals (Fig. 1A) [1]. The additional PCR fragments detected in the cortex and muscle were larger than the characterized (by sequencing) (CTG)12 repeat length short allele, and shorter than that in the cerebellum. Such mosacism in the shorter ranges was not detected in other tissues of the same individual nor in several tissues in other DM1 individuals (Fig. 1B).

Figure 1. Detection of extra bands after short DM1 allele PCR amplification in UR-ADM9 DM1 patient.

Figure 1

Open in a new tab

Products of PCR amplification (30 cycles) were analysed by (A) Non-denaturing acrylamide gel showing additional bands in cortex, skeletal muscle and cerebellum in UR-ADM9 patient. (B) No detection of extra bands in other DM1 patients from the same study.[6] *detection of extra bands (C) Agarose gel in UR-ADM9 cerebellum, cortex and skeletal muscle tissues to compare the appearance of additional bands vs. acrylamide gel methodology.

Initially several faint extra PCR amplified fragments were detected in cerebral cortex, skeletal muscle and cerebellum when resolved on non-denaturing acrylamide gels, which separates based upon both molecular weight and DNA secondary structure (Fig. 1A). The same PCR products resolved in agarose electrophoresis, relatively insensitive to structure (Fig. 1C) suggested that a single additional product was present in these tissues with the largest in the cortex and the smallest in the cerebellum, while the other bands on acrylamide gels were likely heteroduplexes of the two differently sized PCR products [9–11]. Sequence analysis of these additional PCR products from all three tissues (see Material and Methods) showed that 6 of 15 clones from cortex (40%) and 6 of 13 clones from skeletal muscle (46%) contained, as expected, larger sequences (a total of 63 and 46 base pairs, respectively) than the non-expanded repeat tract ((CTG)12 = 36 base pairs). Strikingly, the sequences of these variants were completely devoid of CTG repeats except for the first 5'-CTG unit (Fig. 2A). In the cerebral cortex the complete 5'-(CTG)-3' appeared to have been replaced with a sequence containing the first CTG followed by a 60 base 5'-CTG-repeat-free sequence (Fig. 2A). In the muscle the complete 5'-(CTG)-3' appeared to have been replaced with a sequence containing only the first CTG followed by a 43 base 5'-CTG-repeat-free sequence (Fig. 2A). Interestingly, a comparison between the muscle or cortex insertion sequences revealed high levels of sequence identity at the beginning and end: Complete alignment (100%) of the first 15 bases and strong alignment (81%) in the last 16 bases (13 of 16) (Fig. 2B). Inserts began with a CTG and ended with a TG, presumably belonging to the first and last CTG units. The flanking DMPK sequence also PCR amplified in these samples was perfectly aligned with the known gene sequence (GeneID 1760; see Fig. 2A) in all 28 clones. The rest of the clones showed only the pure repeat tract with 12 or 11 units, suggesting a mixing of DNA products due to poor electrophoretic resolution. None of the 12 clones derived from cerebellum DNA of the same individual (isolated and characterized in parallel) showed this obscure repeat configuration; as all contained a pure CTG tract, where 5 of 12 clones (~50%) contained (CTG)10 or (CTG)11; which may be contractions of −1 or −2 repeat units from the normal (CTG)12 allele. These short repeat length variants explain the presence of the extra PCR products derived from this tissue resolved on acrylamide (Fig. 2A). Sequence analysis of 10 clones derived in parallel from the main PCR product from the cerebellum showed 12 CTG repeats and the proper flanking sequence (data not shown).

Figure 2. Characterization by cloning approach of the DM1 repeat site in the short allele from the different UR-ADM9 tissues analyzed.

Figure 2

Open in a new tab

(A) Examples of clones with a normal 12 CTG repeat (36 bp) sample (main PCR band in the UR-ADM9 DM1 patient), a −2 CTG units sample (cerebellum), a 63 bp non-repeated sample (cerebral cortex), and a 46 bp non-repeated sample (skeletal muscle). (B) Homology degree between both non-repeated sequences found in cerebral cortex and skeletal muscle in UR-ADM9 DM1 patient.

We disfavor the possibility that these aberrant DM1 configurations arose by PCR contamination or artefactual events as it is unlikely that these would occur in all reactions from only specific tissues, but not others. Further argument against contamination is from assessing additional clones from the main PCR fragment amplified (short allele) in cerebral cortex (tissue showing the extra band) and from other UR-ADM9 tissues (not showing the extra band): All 25 clones analyzed showed only the presence of the (CTG)12 repeat tract after sequencing, confirming the additional extra PCR products as the only ones containing the non-repetitive sequences. We also performed a sequence database search for potential donor sources for the inserted sequences. Both inserted sequences were assessed using the Blast2Align tool available online (http://www.ncbi.nlm.nih.gov) and the alignments against different sources showed: (1) very low score alignments (<35%) with any known sequence in the human genome; (2) no alignments with the bacteria genome (E.coli) used to transform in the clones; and (3) no alignments with the cloning TOPO TA vector sequence. These results argue against a bacterial or cloning vector DNA contamination.

The absence of the inserted sequence in the human genome database may be due to it arising from a sequencing gap that has not been annotated [12,13,14,15] or possibly to a novel recombined sequence. Towards learning if the insert sequences are in fact part of the human genome we designed primers specific to the insert using the regions of homology (Fig. 2B) and these were able to PCR amplify products from the DNA of various patient tissues, confirming that this sequence is in fact of human origin (Fig. 3A). The size of the PCR products was approximately 800 bp; considerably longer than the non-CTG insert indicating that its source might not be a continuous sequence but harbors intervening sequences. Direct sequencing of these revealed a total sequence of 858 bp (Fig. 3B). Sequencing was carried out on both the clones of direct PCR products of skeletal muscle, cortex, and cerebellum, as well as the gel-purified and cloned PCR products, with three clones of each separate tissue having been sequenced. Each clone contained at least a partial sequence, with a BLAST search revealing that each sequence mapped to the same location within the human genome (chromosome 4q12) (Accession # NT_022853.15) [16–18]. The sequence contained motifs capable of forming unusual DNA structures, including Z-DNA (Fig. 3B, [19]), as well as containing a pol-α mutation hotspot site [20,21], which may suggest a tendency to be recombinogenic, thereby permitting it to be inserted into the DM1 CTG repeat tract region [15].

Figure 3. Characterization of the PCR products of the insert-specific designed primers.

Figure 3

Open in a new tab

(A)PCR amplification of UR-ADM9 skeletal muscle and cortex DNA with the primers designed from the homology between the non-CTG insert in both tissues revealed a product approximately 800 bp in length. Primers were imperfectly hybridized at the ends of the sequence. (B) Sequencing and a BLAST search determined that the PCR product came from chromosome 4q12. Red text indicates locations of purine/pyrimidine tracts able to form Z-DNA [19]. Bold green text shows a polymerase α mutation hotspot site [20,21].

Although close in size with the short repeat allele we initially cannot discard the expanded allele as the source of these aberrant products. Towards learning whether the non-CTG insertions arose on the expanded or non-expanded DM1 alleles we took advantage of known epigenetic marks specific to the expanded allele. Initially we considered using the adjacent markers that could discern DM1 alleles. Unfortunately, in patient ADM9 the upstream BpmI polymorphism in exon 10 [22] is homozygous and the downstream Cac8I polymorphism in exon 3 of the Six5 gene that is in complete linkage disequilibrium with the DM1 CTG expansion [23] is extremely far (~5kb) from the CTG repeat and rich in G+C content making PCR amplification impossible. However, we recently demonstrated the presence of CpG methylation on the expanded but not the non-expanded allele in many DM1 patient tissues [1]. In patient UR-ADM9 the cortex showed high levels of CpG methylation upstream of the CTG repeat spanning the adjacent methyl-sensitive SacII restriction site. However, the cerebellum was relatively free of methylation. We used the methyl-specific multiplex PCR assay that we had devised to assess the methylation status of the SacII site 42 bp upstream of the CTG tract (Fig. 4A). Forward and reverse primers were placed on opposite sides of the CTG repeat, and therefore were only capable of amplifying across the non-expanded CTG tracts, as well as the short non-CTG insertions. (No primer sets were able to amplify across the expanded allele due to the very large expansions they harbored; skeletal muscle [with (CTG) = 3700–4300], cortex [with (CTG) = 4000–5500], and cerebellum [with (CTG) = 1100–1500 and 3800–4600] of this individual [1]). Using two upstream primers, one that did and one that did not cover the SacII site, permitted determination of the methylation status of the SacII site in the non-expanded allele (Fig. 4A). PCR amplification revealed that in the absence of SacII digestion two major PCR products, both from the non-expanded allele, were evident. Also evident were the slower migrating products, representing the non-CTG insertions in the muscle and cortex DNAs and the CTG length stutter products in the cerebellum (Fig. 4B, see grey arrowhead). Successful SacII digestion prior to PCR amplification, an indication of a non-methylated template, considerably reduced the production of the PCR products that encompassed the SacII site. The loss of the non-expanded CTG allele, the non-CTG insertion and the CTG stutter PCR products supports the conclusion that the DNA templates from which these were amplified were not methylated at the SacII site.

Figure 4. Assessment of methylation allele-specificity.

Figure 4

Open in a new tab

(A) SacII methylation-sensitive digestion combined with multiplex PCR amplification was performed, using two upstream primers and one downstream primer to produce two PCR products from only the non-expanded CTG allele, that either did or did not encompass the methylation-sensitive SacII site present upstream of the CTG tract. The insert, or stutter PCR products of the skeletal muscle, cortex, or cerebellum, are visible as lighter grey bands in the schematic. (B) Following SacII digestion the elimination of only the PCR product encompassing the SacII site revealed that this site on the allele which the non-CTG insert is found was not methylated in either the cortex or the skeletal muscle. The same is true of the stutter found in the cerebellum sample.

Together these results could suggest that the non-CTG insertions arose on the non-expanded DM1 allele or that they arose on the subset of expanded alleles that was free of methylation or that some time following insertion the expansion-specific methylation was lost.

Discussion

Here we have observed a DM1 allele that has completely lost the CTG tract, retaining only the first CTG and the TG of the very last CTG unit. Previously, the complete loss of all CGG repeats at the FMR1 gene was reported, but this also incurred loss of flanking sequences, and appeared to have occurred during transmission [24]. In our DM1 sample the loss of the CTG tract, did not incur any changes in flanking sequences and arose as a somatic insertion mutation. Insertions into repetitive sequences have not been reported, however, SCA31 was recently reported to have arisen through the insertion of an unstable repetitive element [25]. It is noteworthy that there are several reports claiming of apparent DM1 allele length mosaicism, that have been interpreted to be the result of small CTG length alterations of the non-expanded allele [26]. However, sequence analysis of these presumed CTG length changes was not performed. Our findings suggest other reports of presumed CTG or CAG length changes [2–4,26–28] may actually be insertions at the repeat of non-CTG sequences, much as we described herein. Thus, the insertion mutations we report here may be more common than suspected.

It is curious that these non-CTG somatic insertions arose in some, but not all tissues. Previously, somatic CTG expansions were present at higher levels in the cardiac muscle than in peripheral blood mononuclear cells [4]. Tissue preference for somatic length heterogeneity at the DM1 repeat and the Huntington’s disease locus has been reported [8,27]. We previously reported for this same juvenile onset DM1 individual, a range of heterogenous CTG expansions in the heart [with (CTG) = 4100–5300 and 6000], liver [with (CTG) = 4200–4600], skeletal muscle [with (CTG) = 3700–4300], cortex [with (CTG) = 4000–5500] and cerebellum [with (CTG) = 1100–1500 and 3800–4600] of this individual [1]. The presence of non-repeat insertions in cortex and muscle, and limited CTG length variations of the short allele in cerebellum, with absence of variant short products in the heart and liver, does not offer any obvious pattern. While such variant DM1 configurations may be present at extremely low levels in other tissues, they are not detectable using the bulk PCR methods used herein. The significance, if any, of the tissue selections is unknown.

The frequency and timing of these aberrant insertions is unclear. We speculate that the unique nature of the mutant alleles makes it extremely unlikely that these mutations occurred more than once in the lifetime of the patient. Likewise, de novo mutations were detected in cerebral cortex and skeletal muscle, but not in heart or liver. Only two mutant alleles were detected, and although they must have occurred de novo at some point it remains unclear if they actually occurred in those tissues or in a precursor lineage at an earlier developmental stage. The level of mosaicism of such novel species has not been determined rigorously. However, considering that they are readily detectable using the bulk PCR methods applied herein we suggest that their occurrence is likely to be around the 10% of the cells in the cortex and muscle.

As to how the DM1 insertion mutations arose we can only speculate. Such events may have arisen on the expanded or the non-expanded alleles, through events involving only one or both alleles. Short size alterations may have arisen through insertions into the non-affected CTG allele, or through catastrophic contractions and insertions into the expanded allele. Previously, a germline contraction of the expanded DM1 allele to non-affected lengths was reported [29], a transmission event that appeared to arise by recombination between the two DM1 alleles. Similarly, somatic mosacism of the HD CAG tract was shown to arise by the contraction of the expanded (CAG)47 to (CAG)37 rather than by the expansion of the normal (CAG)20 allele [8]. While we are unable to definitively assign which allele incurred the non-CTG insertions our analyses, considering the facility of the expanded repeat to form slipped-DNA structures [30] and the absence of CTG-region mutations in non-DM1 individuals, suggest that these insertions arose on the CTG expanded allele. Recent studies in yeast have suggested that other expanded repetitive elements can facilitate gene conversion events and such events may be facilitated by unusual DNA structures, double- or single-strand breaks and the capturing of single-strand regions [31]. Another possible mechanism is a template-switching event, where a replication fork at the expanded CTG tract may arrest and through multiple microhomology-mediated template-switches with the chromosome 4 region (or other), may incur a series of insertions and aberrant nucleotide incorporations – resulting in a novel sequence at the DM1 CTG tract region [32,33]. Such obscure mutation configurations, although rare, have been observed in model systems and may involve unusual DNA structures [20,21,31–35]. The likely source of the DM1 insert sequence on chromosome 4 (Fig. 3C) is enriched in G+C content and can form unusual DNA structures, with this region having also been associated with sequence deletions in tumors [16–18]. As to whether such events can lead to the non-CTG insertions observed here remains to be determined.

In conclusion, we detected a CTG-free configuration at the DM1 CTG repeat location that completely replaced the CTG tract. These non-CTG configurations arise as somatic events in specific tissues, and appear to be derived from a region of chromosome 4. Strikingly, these de novo mutations did not contain the CTG repeat. The DM1 repeat tract has recently been described as presenting configurations distinct from the pure CTG trinucleotide motif. Interruptions by non-CTG repeat units like CCG and GGC through the DM1 CTG tract have been linked with altered CTG repeat stability [5–7]. However, like those interruptions, it is unknown if the non-repeat configuration observed herein is a product of instability of the larger pure tract. Interruptions in the DM1 tract have been suggested in one family to be associated with altered clinical presentation [7]. The impurity of the repeat will essentially deplete CUG repeats within the toxic DMPK transcript and may well affect its ability to bind and sequester RNA splicing factors such as MBNL1 which specifically binds CUG, as has been suggested [7]. Similarly, DMPK transcripts with a complete absence of CUG repeats, as in the alleles observed herein, would also be expected to have altered toxic effects (if any). While the clinical significance of these distinct configurations of the DM1 repeat tract awaits future analysis their existence advises against presuming a pure repeat when interpreting changes in nucleotide sizes of the DM1 CTG region.

Acknowledgements

We thank Drs. Tom Kunkel and Tom Petes for scientific discussions. This work was supported by the Canadian Institutes of Health Research (CIHR) (CEP), the Muscular Dystrophy Canada (CEP), by the University of Rochester Paul Wellstone Muscular Dystrophy Cooperative Research Center with support from the NIH (grant U54NS48843) (CEP & CAT), The Muscular Dystrophy Association USA (MDAUSA) (CAT), a Doctoral Training Award from the CIHR Collaborative Graduate Training Program in Molecular Medicine (MMA), and an Ontario Graduate Scholarship (MMA), a postdoctoral fellowship from The Hospital for Sick Children Research Training Centre (ALC), and a postdoctoral fellowships from the MDAUSA and the Cell Science Research Foundation and the Uehara Memorial Foundation (MN).

Footnotes

Copyright license statement

The authors declare no conflict of interest

License for Publication

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive license (or non exclusive for government employees) on a worldwide basis to the BMJ Publishing Group Ltd to permit this article (if accepted) to be published in Journal of Medical Genetics and any other BMJPGL products and sublicenses such use and exploit all subsidiary rights, as set out in our license (http://group.bmj.com/products/journals/instructions-for-authors/license-forms).

References

  • 1.Lopez Castel A, Nakamori M, Tome S, et al. Expanded CTG repeat demarcates a boundary for abnormal CpG methylation in myotonic dystrophy patient tissues. Hum Mol Genet. 2011;20(1):1–15. doi: 10.1093/hmg/ddq427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Martorell L, Gamez J, Cayuela ML, et al. Germline mutational dynamics in myotonic dystrophy type 1 males: allele length and age effects. Neurology. 2004;62(2):269–274. doi: 10.1212/wnl.62.2.269. [DOI] [PubMed] [Google Scholar]
  • 3.Meiner A, Thamm B, Strenge S, et al. Instability in the normal CTG repeat range at the myotonic dystrophy locus. J Med Genet. 1998;35(9):791. doi: 10.1136/jmg.35.9.791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Furutama D, Negoro N, Terasaki F, et al. Possible de novo CTG repeat expansion in the DMPK gene of a patient with cardiomyopathy. J Clin Neurosci. 17(3):408–409. doi: 10.1016/j.jocn.2009.06.010. [DOI] [PubMed] [Google Scholar]
  • 5.Leeflang EP, Arnheim N. A novel repeat structure at the myotonic dystrophy locus in a 37 repeat allele with unexpectedly high stability. Hum Mol Genet. 1995;4(1):135–136. doi: 10.1093/hmg/4.1.135. [DOI] [PubMed] [Google Scholar]
  • 6.Musova Z, Mazanec R, Krepelova A, et al. Highly unstable sequence interruptions of the CTG repeat in the myotonic dystrophy gene. Am J Med Genet A. 2009;149A(7):1365–1374. doi: 10.1002/ajmg.a.32987. [DOI] [PubMed] [Google Scholar]
  • 7.Braida C, Stefanatos RK, Adam B, et al. Variant CCG and GGC repeats within the CTG expansion dramatically modify mutational dynamics and likely contribute toward unusual symptoms in some myotonic dystrophy type 1 patients. Hum Mol Genet. 19(8):1399–1412. doi: 10.1093/hmg/ddq015. [DOI] [PubMed] [Google Scholar]
  • 8.Norremolle A, Hasholt L, Petersen CB, et al. Mosaicism of the CAG repeat sequence in the Huntington disease gene in a pair of monozygotic twins. Am J Med Genet A. 2004;130A(2):154–159. doi: 10.1002/ajmg.a.30128. [DOI] [PubMed] [Google Scholar]
  • 9.Olejniczak M, Krzyzosiak WJ. Genotyping of simple sequence repeats--factors implicated in shadow band generation revisited. Electrophoresis. 27:3724–3734. doi: 10.1002/elps.200600136. [DOI] [PubMed] [Google Scholar]
  • 10.Poirier J, Ohshima K, Pandolfo M. Heteroduplexes may confuse the interpretation of PCR-based molecular tests for the Friedreich ataxia GAA triplet repeat. Hum Mutat. 1999;13(4):328–330. doi: 10.1002/(SICI)1098-1004(1999)13:4<328::AID-HUMU10>3.0.CO;2-J. [DOI] [PubMed] [Google Scholar]
  • 11.Pearson CE, Tam M, Wang YH, et al. Slipped-strand DNAs formed by long (CAG)•(CTG) repeats: slipped-out repeats and slip-out junctions. Nucleic Acids Res. 2002;30(20):4534–4547. doi: 10.1093/nar/gkf572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bork P, Copley R. The draft sequences. Filling in the gaps. Nature. 2001;409(6822):818–820. doi: 10.1038/35057274. [DOI] [PubMed] [Google Scholar]
  • 13.Eichler EE. Segmental duplications: what's missing, misassigned, and misassembled--and should we care? Genome Res. 2001;11(5):653–656. doi: 10.1101/gr.188901. [DOI] [PubMed] [Google Scholar]
  • 14.Eichler EE, Clark RA, She X. An assessment of the sequence gaps: unfinished business in a finished human genome. Nat Rev Genet. 2004;5(5):345–354. doi: 10.1038/nrg1322. [DOI] [PubMed] [Google Scholar]
  • 15.Garber M, Zody MC, Arachchi HM, et al. Closing gaps in the human genome using sequencing by synthesis. Genome Biol. 2009;10(6):R60. doi: 10.1186/gb-2009-10-6-r60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cetin E, Cengiz B, Gunduz E, et al. Deletion mapping of chromosome 4q22-35 and identification of four frequently deleted regions in head and neck cancers. Neoplasma. 2008;55(4):299–304. [PubMed] [Google Scholar]
  • 17.Berg M, Agesen TH, Thiis-Evensen E, et al. Distinct high resolution genome profiles of early onset and late onset colorectal cancer integrated with gene expression data identify candidate susceptibility loci. Mol Cancer. 9:100. doi: 10.1186/1476-4598-9-100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Brosens RP, Belt EJ, Haan JC, et al. Deletion of chromosome 4q predicts outcome in stage II colon cancer patients. Anal Cell Pathol (Amst) 33(2):95–104. doi: 10.3233/ACP-CLO-2010-0531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Zhao J, Bacolla A, Wang G, et al. Non-B DNA structure-induced genetic instability and evolution. Cell Mol Life Sci. 67(1):43–62. doi: 10.1007/s00018-009-0131-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Christodoulou J, Craig HJ, Walker DC, et al. Deletion hotspot in the argininosuccinate lyase gene: association with topoisomerase II and DNA polymerase alpha sites. Hum Mutat. 2006;27(11):1065–1071. doi: 10.1002/humu.20352. [DOI] [PubMed] [Google Scholar]
  • 21.Kunkel TA. The mutational specificity of DNA polymerases-alpha and -gamma during in vitro DNA synthesis. J Biol Chem. 1985;260(23):12866–12874. [PubMed] [Google Scholar]
  • 22.Neville CE, Mahadevan MS, Barcelo JM, et al. High resolution genetic analysis suggests one ancestral predisposing haplotype for the origin of the myotonic dystrophy mutation. Hum Mol Genet. 1994;3(1):45–51. doi: 10.1093/hmg/3.1.45. [DOI] [PubMed] [Google Scholar]
  • 23.Thornton CA, Wymer JP, Simmons Z, et al. Expansion of the myotonic dystrophy CTG repeat reduces expression of the flanking DMAHP gene. Nat Genet. 1997;16(4):407–409. doi: 10.1038/ng0897-407. [DOI] [PubMed] [Google Scholar]
  • 24.Gronskov K, Hjalgrim H, Bjerager MO, et al. Deletion of all CGG repeats plus flanking sequences in FMR1 does not abolish gene expression. Am J Hum Genet. 1997;61(4):961–967. doi: 10.1086/514872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sato N, Amino T, Kobayashi K, et al. Spinocerebellar ataxia type 31 is associated with "inserted" penta-nucleotide repeats containing (TGGAA)n. Am J Hum Genet. 2009;85(5):544–557. doi: 10.1016/j.ajhg.2009.09.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Coolbaugh-Murphy MI, Xu J, Ramagli LS, et al. Microsatellite instability (MSI) increases with age in normal somatic cells. Mech Ageing Dev. 2005;126(10):1051–1059. doi: 10.1016/j.mad.2005.06.005. [DOI] [PubMed] [Google Scholar]
  • 27.Hasan Q, Mohan V, Ahuja YR. (CTG)n expansion at DMPK locus seen only in muscle tissue: a novel case. Indian J Exp Biol. 2004;42(9):937–940. [PubMed] [Google Scholar]
  • 28.Dow DJ, Rubinsztein DC, Yates JR, et al. Instability of normal (CTG)n alleles in the DM kinase gene. J Med Genet. 1997;34(10):871–873. doi: 10.1136/jmg.34.10.871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.O'Hoy KL, Tsilfidis C, Mahadevan MS, et al. Reduction in size of the myotonic dystrophy trinucleotide repeat mutation during transmission. Science. 1993;259(5096):809–812. doi: 10.1126/science.8094260. [DOI] [PubMed] [Google Scholar]
  • 30.Pearson CE, Sinden RR. Alternative structures in duplex DNA formed within the trinucleotide repeats of the myotonic dystrophy and fragile X loci. Biochemistry. 1996;35(15):5041–5053. doi: 10.1021/bi9601013. [DOI] [PubMed] [Google Scholar]
  • 31.Tang W, Dominska M, Greenwell PW, et al. Friedreich's Ataxia (GAA)(n)•(TTC)(n) Repeats Strongly Stimulate Mitotic Crossovers in Saccharomyces cerevisae. PLoS Genet. 7(1):e1001270. doi: 10.1371/journal.pgen.1001270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hicks WM, Kim M, Haber JE. Increased mutagenesis and unique mutation signature associated with mitotic gene conversion. Science. 2010;329:82–85. doi: 10.1126/science.1191125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Muers M. Mutation rate: DNA repair and indels boost errors. Nat Rev Genet. 2010;11:592. doi: 10.1038/nrg2848. [DOI] [PubMed] [Google Scholar]
  • 34.Kunkel TA, Soni A. Mutagenesis by transient misalignment. J Biol Chem. 1988;263(29):14784–14789. [PubMed] [Google Scholar]
  • 35.Kunkel TA, Hamatake RK, Motto-Fox J, et al. Fidelity of DNA polymerase I and the DNA polymerase I-DNA primase complex from Saccharomyces cerevisiae. Mol Cell Biol. 1989;9(10):4447–4458. doi: 10.1128/mcb.9.10.4447. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES