Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. 2004 Jan; 24(2): 629–637.
PMCID: PMC343816

MSH2-Dependent Germinal CTG Repeat Expansions Are Produced Continuously in Spermatogonia from DM1 Transgenic Mice


Myotonic dystrophy type 1 is a neuromuscular affection associated with the expansion of an unstable CTG repeat in the DM protein kinase gene. The disease is characterized by somatic tissue-specific mosaicism and very high intergenerational instability with a strong bias towards expansions. We used transgenic mice carrying more than 300 unstable CTG repeats within their large human genomic environment to investigate the dynamics of CTG repeat germinal mosaicism in males. Germinal mosaicism towards expansions was already present in spermatozoa at 7 weeks of age and continued to increase with age, suggesting that expansions are continuously produced throughout life. To determine the precise stage at which germinal expansions occur during spermatogenesis, we sorted and collected the different germ cell types produced during spermatogenesis from males of different ages and analyzed the CTG repeat mosaicism in each fraction. Strong mosaicisms towards expansions were already observed in spermatogonia before meiosis. In transgenic Msh2-deficient mice, germinal instability of the CTG repeats (only contractions) also occurs premeiotically. No significant difference in mosaicism was detected between spermatogonia and spermatozoa, arguing against continued expansions during postmeiotic stages. This indicates that germinal expansions are produced at the beginning of spermatogenesis, in spermatogonia, by a meiosis-independent mechanism involving MSH2.

Myotonic dystrophy type 1 (DM1) is associated with the expansion of a CTG trinucleotide repeat located in the 3′ untranslated region of the DM protein kinase (DMPK) gene at 19q13.3 (3, 7, 17, 27). In normal subjects, there are usually between 5 and 37 copies of this repeat, which remains stable following intergenerational transmissions. In DM1 patients, more than 50 CTG repeats are typically present, and the repeat is highly unstable and increases with each generation. The number of CTG repeats is positively correlated with the severity of symptoms and is negatively correlated with age at onset, resulting in an anticipation phenomenon that is particularly obvious in DM1 (19). The behavior of the CTG repeat between generations appears to depend on the sex of the transmitting parent. Paternal transmissions lead to larger expansions for <100 CTG repeats, whereas maternal transmissions lead to larger expansions when the CTG repeat tract contains >500 repeats in the transmitting parent (2, 24). In between, both paternal and maternal alleles expand. In addition to intergenerational instability, somatic CTG repeat-length mosaicism is also found in DM1 patients. Inter- and intratissue mosaicisms increasing with age are observed, with a strong bias towards expansions (1, 22).

DM1 is one of the growing group of diseases caused by dynamic mutations. This group currently comprises more than a dozen diseases, including Huntington's disease (HD), spinocerebellar ataxias, and fragile X syndrome, which are generally associated with CNG trinucleotide repeats (11, 46). The dynamics of the different trinucleotide repeats involved are very similar (intergenerational and somatic instability, bias towards expansions, CTG repeat number threshold for instability etc.), suggesting that the instability mechanisms are common to these diseases, especially those involving CTG or CAG tracts. Several hypotheses have been proposed to explain triplet repeat instability. These have been tested in bacteria, yeasts cell culture models, and transgenic mouse models. Expansions might occur during replication, through DNA polymerase pausing and/or slippage or FEN-1 endonuclease dysfunction in the CTG tract region (9, 20, 41-43, 45, 51). Furthermore, DNA repair mechanisms are clearly involved in instability, as the absence of MSH2, the main component of the mismatch repair pathway, leads in HD mice to the somatic stabilization of expanded repeats, showing that MSH2 is essential for the production of CAG expansions in tissues (29, 54a). Experiments using our DM300-328 transgenic mice, which carry >300 CTG repeats in their large (>45-kb) human genomic environment, revealed that the absence of MSH2 in this context does not stabilize the repeat but instead shifts the instability from expansions to contractions, not only in somatic tissues but also through generations (49). These results demonstrated that MSH2 is required for the formation of somatic and intergenerational CTG expansions. The CTG repeat is probably processed in a different way in the absence of MSH2, leading to intergenerational and somatic contractions. In a knock-in mouse model in which the 3′ part of the DMPK murine homolog is replaced by the corresponding human 3′ part of DMPK together with about 80 CTG repeats, van den Broek and collaborators observed a decrease of somatic CTG tract expansions in the absence of MSH3, one of the partners of MSH2 involved in mismatch repair and other mechanisms such as recombinational repair (54).

To determine the contribution of germinal mosaicism in the transmitting parent to intergenerational instability, several studies have been carried out on human patients with CAG repeat expansion diseases (HD, dentatorubropallidoluysian atrophy, spinocerebellar ataxia types 1 and 7, or fragile X syndrome) to look for correlations between the level of mosaicism in the sperm of the transmitting males and the size of the repeat inherited by their offspring. A high level of mosaicism was commonly observed in sperm, with this mosaicism overlapping the sizes inherited by the offspring, suggesting that intergenerational instability results mainly from germinal instability in the father (8, 12, 25, 34, 38, 52, 53). However, several studies, including one on fragile X fetuses and one on transmission in Friedreich ataxia families, suggested that intergenerational length changes could also involve a postzygotic instability event (13, 14, 37). In DM1 patients, a high level of CTG instability is observed in sperm from transmitting males, and this mosaicism more or less overlaps the CTG repeat sizes measured in blood from the offspring (22, 30).

A detailed analysis of CTG repeat-size variation between DM300-328 transgenic parents and their offspring carrying different MSH2 genotypes revealed that intergenerational instability probably results from the combination of two distinct events: germinal mosaicism in the transmitting parent and an MSH2-dependent instability event that takes place postzygotically, just after fertilization (49). Development analysis is required to elucidate the postzygotic instability event in our mice. Furthermore, the timing and dynamics of germinal and gametic instability in humans and in mouse models are still not fully understood. In HD mice carrying about 120 CAG repeats, male germinal expansions have been reported to be produced only in the latest stages of spermatogenesis (specifically during the transition from round spermatids to elongating spermatids) through a mechanism involving MSH2 (23, 28, 33). This is surprising, as the gene encoding MSH2 is not expressed during the final stages of mouse spermatogenesis and as no MSH2 protein can be detected in round spermatids in either men (5) or mice (47). In contrast, single-molecule DNA analysis of testicular germ cells from two HD patients suggested that expansions can occur before the end of the first meiotic division (55).

In several trinucleotide-repeat diseases and in our male transgenic mice, a positive correlation exists between the age of the transmitting parent and the repeat-length change in the offspring (48, 50). The evolution of parental gametic mosaicism throughout life may partly account for this association. A very small number of studies have measured repeat-size mosaicism in sperm from the same patient at different ages, but no clear increase was detected after 2 years in men with HD (25). However, the study periods may not be sufficiently long to detect changes in repeat-length mosaicism.

To elucidate the mechanism of germinal instability, we analyzed the status of CTG repeat instability in sperm at different ages to monitor the evolution of gametic mosaicism. By using the single-molecule PCR method, we demonstrated that a strong mosaicism towards expansions is present in spermatozoa, even very early during the mouse reproductive life. This mosaicism clearly increases with age, indicating that expansions are continuously produced through spermatogenesis in our mice. When assaying by normal or single-molecule PCR the levels of CTG repeat mosaicism in seven different germinal cell fractions sorted through a fluorescence-activated cell sorter (FACS), we found that mosaicism towards expansions was already present in spermatogonia. No differences were detected between mosaicisms in spermatogonia and spermatozoa. The presence of expansions in the first stages of spermatogenesis demonstrates that meiosis and postmeiotic mechanisms are probably not involved in germinal CTG repeat expansions. In Msh2−/− males, a majority of contractions were present in spermatogonia and no changes appeared in the following stages, indicating that MSH2 is necessary for the formation of germinal expansions in spermatogonia and that contraction events also occur before meiosis.


Transgenic mice.

We used the DM300-328 transgenic line carrying 45 kb of human genomic DNA cloned from a DM1 patient as described by Seznec et al. (50). These mice have been produced with B6D2/F1 fertilized mouse eggs and crossed over at least 15 generations with C57BL/6 mice. Transgenic status was assayed by PCR with the DMHR4 and DMHR5 primers (26). Tail DNA was amplified by PCR with primers 101 and 102, and CTG repeat size was measured after separating amplification products on a 4% acrylamide denaturing gel, as described previously (50).

Mouse dissection and blood and sperm DNA extraction.

Mice were killed and dissected at 7 weeks, 11 months, or 28 months of age, in accordance with the French veterinary laws. Blood was collected from the heart just after death, and DNA was extracted by the phenol-chloroform method (26). Semen was collected from both vas deferens, and DNA was carefully extracted from spermatozoa to avoid contamination by DNA from other cell types (49, 50).

FACS of different germinal cell types.

After dissection at 7 weeks or 11 months of age, both testes were collected and cleaned in 1× phosphate-buffered saline (PBS). After removal of the tunica albuginea, the seminiferous cords were unwound and rinsed in 1× PBS and then carefully minced to obtain the maximum amount of individual germ cells. After filtration through a 70-μm-pore-size cell strainer (Falcon), which retains all cellular aggregates, cells were counted and the cell solution was centrifuged for 10 min at 2,500 rpm (Jouan GR 422 instrument) at room temperature. The cell pellet was resuspended in cold 70% ethanol overnight at a final concentration of 5 × 106 cells/ml. The cell suspension was then centrifuged for 10 min at 2,500 rpm at room temperature and rinsed twice with fresh RPMI 1640 medium (GibcoBRL). RNA was removed by incubating for 30 min at room temperature with 500 μg of RNase A (Boehringer) per ml. Digitonin (10 μg/ml; Sigma) was then added and left for 2 min at 4°C in the dark to permeabilize the cells. Digitonin was removed by centrifugation at 2,500 rpm at room temperature for 10 min. The cell pellet was resuspended at a final concentration of 1.5 × 106 cells/ml in fresh RPMI 1640 with 200 nM MitoTracker GreenFM (Molecular Probes) for 30 min at 4°C in the dark. Propidium iodide (10 μg/ml; Sigma) was added 2 min before analyzing cell labeling with a FACSVantage cell sorter (Becton Dickinson). Cell doublets, apoptotic cells, and necrotic cells were discarded. The combination of the two labels allowed us to sort seven different germinal cell types (44). Each cell fraction was collected in fresh RPMI 1640 before DNA extraction. Cells from each fraction were mounted on microscope slides and analyzed with a fluorescence microscope (Leica) to assay the purity.

Extraction of DNA from FACS-sorted cells.

Cell fractions were centrifuged for 10 min at 10,000 rpm (Sigma 1K15 instrument) at room temperature, and the resulting pellet was resuspended in 0.2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate)-1% sodium dodecyl sulfate with 1 M β-mercaptoethanol for 1 h at room temperature. DNA was then extracted by the phenol-chloroform method.

Analysis of CTG repeat instability in the different germinal cell types.

The CTG repeat region of each fraction was amplified with primers 101 and 102. Mosaicism was assayed by separating the amplification products on 3.5% acrylamide denaturing gels as previously described (49, 50).

PCR and single-molecule PCR.

Blood, spermatozoa, and spermatogonia DNA was amplified by classical PCR and single-molecule PCR as previously described (35, 49).

Statistical analyses.

Statistical analyses were performed with the StatView software by using the Mann-Whitney test (SAS Institute, Inc.).


CTG repeat mosaicism in spermatozoa throughout the reproductive life span.

Transgenic DM300-328 males were dissected at 7 weeks, 11 months, and 28 months of age to monitor the CTG repeat-length mosaicism in the male germ line and to characterize the dynamics of germinal instability. We extracted DNA from blood and spermatozoa heads and used classical PCR and single-molecule PCR to determine the level of CTG repeat-length mosaicism. The CTG repeat-size mosaicism in blood and spermatozoa DNA was compared at each age and between the different ages to analyze their relative evolution. The CTG repeat remains relatively stable in the blood throughout life in our mice and in DM1 patients (49, 50), and thus the CTG repeat size determined from blood was used as a reference for CTG repeat-size mosaicism in other tissues. In mice, the appearance of the different germ cell types during development follows a time-dependent sequence. Spermatogonia differentiate from primordial germ cells and are the only cell type present at 7 days postpartum. While spermatogonia continue dividing, some of the germ cells enter meiosis and progress along Sertoli cells until the lumen of the seminiferous tubules. The first mature spermatozoa appear in the epididymis at 5.5 weeks after birth (31). Analysis of the mosaicism in spermatozoa of 7-week-old males allowed us to determine the influence of germ line development and the very early stages of spermatogenesis on CTG repeat mosaicism. In contrast, at 11 and 28 months of age, numerous rounds of spermatogenesis have already occurred and spermatogonia have already undergone numerous mitotic divisions.

Figure Figure1A1A shows typical results obtained after amplification of the CTG repeat in blood and spermatozoa at the three studied ages by classical PCR with primers flanking the repeat. To ensure that the basal repeat size in each male had no effect on the level of somatic mosaicism, we studied only males that had equivalent numbers of CTG repeats in tail DNA upon weaning. Although no variation in the size of the CTG repeat was detectable in blood DNA, a distinct mosaicism that increased with age was observed in DNA from spermatozoa.

FIG. 1.
Evolution of CTG repeat-length mosaicism in blood and sperm from transgenic mice throughout life. (A) CTG repeat-length mosaicism in blood (BL) and sperm (Sp) DNA from DM300-328 males at different ages, as measured by classical PCR. M, migration representation ...

Single-molecule PCR with diluted DNA was then used to define more precisely the heterogeneity in the CTG repeat sizes (Fig. (Fig.1B).1B). At 7 weeks, we could detect a very limited mosaicism in blood DNA, with very few variations around the major peak. In spermatozoa, the mosaicism was clearly detectable at this early stage, and CTG repeats already appeared to be longer than in blood (P < 0.0001), with a major peak 12 CTG repeats larger than in blood. At 11 months, the mosaicism in blood was still very weak. In contrast, the mosaicism in spermatozoa had increased significantly (P < 0.0001) and was clearly biased towards expansions, with a wider distribution of alleles of different sizes. The marked major peak disappeared, whereas large expansions could be observed (around +80 CTGs). At 28 months, some larger expansions could be observed in blood, but the mosaicism remained much more limited than in spermatozoa, in which the distribution of alleles of different lengths was much more widespread, with larger expansions reaching +140 CTGs. Thus, the CTG repeat-length mosaicism in spermatozoa increased considerably with age in our transgenic mice, with a strong bias towards expansions. Interestingly, in accordance with these results, we have previously found a positive correlation between the age of the transmitting male and the CTG repeat length inherited by the offspring (50). It should be noted that the CTG repeat lengths can vary slightly with age in blood. However, even at 28 months, most of the repeats were close to the initial size.

Dynamics of CTG repeat mosaicism throughout spermatogenesis.

The evolution of mosaicism throughout spermatogenesis may make it possible to measure the influence of particular steps of this process on instability, such as mitotic divisions, induced recombination, and meiosis. To determine the precise stage of spermatogenesis at which the germinal CTG repeat expansions are produced, we used two labels (one for DNA and one for mitochondria) and FACS to collect seven cellular fractions corresponding to the different stages of spermatogenesis. We then analyzed the CTG repeat-length mosaicism in each fraction, in mature spermatozoa, and also in blood as a control. We used the DNA intercalating agent propidium iodide to separate different germ cell types according to their DNA content: n (round and elongated spermatids), 2n (spermatogonia, preleptotene primary spermatocytes, and secondary spermatocytes), and 4n (leptotene/zygotene and pachytene primary spermatocytes). Figure Figure2A2A shows the DNA content labeling of germ cell samples from an 11-month-old transgenic male. Three major peaks were obtained, corresponding to the different DNA contents encountered during spermatogenesis. However, this labeling alone is not sufficient to separate germ cells from different types carrying the same DNA content, such as spermatogonia and secondary spermatocytes or round and elongated spermatids. We therefore also used a fluorescent label directed against a mitochondrial membrane protein to reflect the activity and repartition of mitochondria throughout spermatogenesis (44). According to the stage of spermatogenesis, germinal cells need more or less energy to perform specific processes and mitochondria have different activities. The combination of these two complementary labels enabled us to sort seven different germinal cell types (Fig. (Fig.2B):2B): spermatogonia (R5), preleptotene (R6), leptotene/zygotene (R9) and pachytene (R8) primary spermatocytes, secondary spermatocytes (R7), and round (R3) and elongated (R4) spermatids (44). We ensured the purity of each cell type (estimated to be over 90%) by analyzing slides from each collected fraction by fluorescence microscopy (Fig. (Fig.3).3). After extraction of DNA from each cell fraction and also from blood and from mature spermatozoa, the level of CTG repeat-length mosaicism in each sample was measured. Interestingly, at 7 weeks of age, CTG repeat sizes appeared slightly larger in spermatogonia than in blood, with no detectable difference between spermatogonia and spermatozoa (Fig. (Fig.4).4). At 11 months of age, we observed marked mosaicism in spermatogonia, with larger CTG repeat sizes than in blood. In the following stages of spermatogenesis, including spermatozoa, mosaicism appeared to be equivalent to that observed in spermatogonia. The apparently faster migration products for round spermatids was not observed on repeated gels. The mosaicism was less pronounced at 7 weeks than at 11 months, showing that instability in spermatogonia increases with age in our transgenic mice. Although the separated seminiferous tubules were carefully washed in 1× PBS, some remaining testis interstitium cells such as Leydig cells (2n) may have been sorted together with spermatogonia (44). In adult mice, these cells represent 3.8% of the testicular volume, whereas the seminiferous tubules occupy 89.3% (36). Therefore, the proportion of Leydig cells remaining in the preparation and sorted with the spermatogonia is negligible. In addition, we observed a similar mosaicism pattern in preleptotene primary spermatocytes, the stage immediately after spermatogonia in spermatogenesis, which should not suffer from any contamination by somatic cells through these labels. These observations are consistent with mosaicism presented by the spermatogonia fraction being representative of instability in spermatogonia.

FIG. 2.
Cellular labels used for the FACS of the different germinal cell types from transgenic mouse spermatogenesis. (A) The DNA content was measured by propidium iodide (PI) labeling. The horizontal axis represents fluorescence (arbitrary fluorescence units), ...
FIG. 3.
Verification of cell fraction purity. Fluorescent microscopy (propidium iodide) slides were prepared for each sorted cellular fraction. Representative cells are shown. Arrows indicate the progression of mouse spermatogenesis from spermatogonia (top left) ...
FIG. 4.
CTG repeat-length mosaicism in blood, in spermatozoa, and in the different cellular fractions in Msh2+/+ transgenic males at 7 weeks and 11 months of age and in an 11-month-old Msh2−/− transgenic male. Amplification products ...

To determine whether CTG repeat-size mosaicism is equivalent in spermatogonia and spermatozoa, we used single-molecule PCR to resolve the smears obtained for these two populations (Fig. (Fig.5).5). At 11 months of age, we observed a large CTG repeat-size mosaicism with a strong bias towards expansions in spermatogonia. Interestingly, in spermatozoa, the level of mosaicism was not statistically (by the Mann-Whitney test) different from that observed in spermatogonia. This suggests that in our mice, germinal expansions are already produced in spermatogonia (i.e., at the first stage of spermatogenesis) through a meiosis-independent mechanism.

FIG. 5.
Single-molecule distribution of alleles with different CTG repeat lengths in spermatogonia and spermatozoa from Msh2+/+ and Msh2−/− transgenic males at 11 months of age. The horizontal axis represents the CTG repeat size, ...

We previously studied the effect of DNA repair genes on CTG repeat instability, and we have showed that the absence of MSH2 in transgenic mice shifted the instability from expansions to contractions in tissues and over generations (49). To determine whether the two types of CTG repeat-length change (expansions in Msh2+/+ mice and contractions in Msh2−/− mice) are produced during the same time window, we separated the different germ line cell types from an Msh2−/− transgenic male at 11 months of age with a CTG repeat size in tail upon weaning similar to that of the Msh2+/+ male. Classical PCR amplification of the CTG repeat detected contractions in spermatogonia (Fig. (Fig.4).4). Like for the Msh2+/+ male, no significant difference could be noted between the levels of mosaicism in spermatogonia and in the subsequently cell types, including spermatozoa. This indicates that, similar to the expansion events in Msh2+/+ mice, the deletion events can occur some time prior to but not after meiosis. While expansions were age dependent, the age dependency of the deletions has yet to be determined. Single-molecule PCR analyses performed on spermatogonia and spermatozoa from this Msh2−/− male confirmed that there was no statistically significant difference between spermatogonia and spermatozoa in the distribution of alleles with different CTG repeat lengths (Fig. (Fig.5).5). This suggests that germinal expansions in Msh2+/+ males and germinal contractions in Msh2−/− males can both be produced in spermatogonia.


We investigated the evolution of the CTG repeat-length mosaicism in blood and sperm from transgenic males carrying >300 CTG repeats at different ages during their reproductive life span. Single-molecule PCR showed that expansions were already present in sperm at 7 weeks of age, just after male mice reach sexual maturity approximately 5.5 weeks after birth. Mosaicism in sperm increased with age, always with a strong bias towards expansions. The largest expansions were observed in sperm from the oldest males, which is consistent with the positive correlation between the age of the transmitting male and repeat sizes inherited by the offspring (50). In contrast, mosaicism in blood remained very weak at all stages, with only a few expansions being detected in the latest stages. At 7 weeks of age, the fact that the mean size of the CTG repeat was larger in spermatozoa than in blood can be explained by two nonexclusive hypotheses. The larger CTG repeat sizes observed in spermatozoa at this age may have been produced during germ line development or during the very early rounds of spermatogenesis, the first of which is completed at around 34 or 35 days after birth when the first mature spermatozoa appear in the epididymis (47). The fact that the germinal mosaicism increases with age suggests that expansions are continuously produced during spermatogenesis and that additive events are responsible for the high level of mosaicism observed in elderly males. In the male germ line, different types of repair occur depending on the stage of spermatogenesis. During germ line development and thereafter, DNA replication repair occurs in both stem and premeiotic cells (4). Furthermore, germinal cells entering meiosis are subjected to several molecular events, such as chromosome pairing and induced recombination, and to different types of DNA repair (32).

We also studied the timing of CTG repeat instability throughout spermatogenesis, to determine the precise stage at which expansions are produced, especially with regard to meiosis, i.e., before, during, or after meiosis. This information should help us to understand the mechanism(s) involved in the production of expansions. We studied CTG repeat instability in spermatogonia, which are renewed and which proliferate during spermatogenesis in adult males; in primary spermatocytes (preleptotene, leptotene/zygotene, and pachytene), in which several molecular events successively occur (induced recombination, chromosome pairing, and DNA synthesis); in secondary spermatocytes, which are subject to the first meiotic division; and in round and elongated spermatids, which undergo major morphological changes and chromatin compaction before finally becoming spermatozoa. In yeast, the instability of CAG/CTG repeats increases during meiosis, mainly due to deletions (10, 21). Once cells enter meiosis, they may undergo molecular mechanisms such as repair associated with homologous recombination, taking place after self-induction of double-strand breaks in an SPO11-dependent manner (21). However, our data show that in our transgenic mice, germinal expansions are produced in spermatogonia and the length of the CTG repeat does not change between spermatogonia and mature spermatozoa. No meiosis-specific mechanisms appear to be necessary for expansions to occur during spermatogenesis in our transgenic mice. Our data suggest that germinal instability results from a mechanism that occurs during spermatogonia, therefore excluding the repair of self-induced double-strand breaks and homologous recombination during meiosis. The increase in mosaicism with age in the DM300-328 males probably results from the accumulation of expansions over lifetime, due to spermatogonia undergoing mitotic divisions and/or the repeated action of DNA repair mechanisms on these cells.

We recently studied the influence of DNA repair on CTG repeat instability by crossing our transgenic mice with mice having knockouts of genes belonging to the different DNA repair pathways, including the Msh2 mismatch repair pathway. We did not observe any stabilization of the repeat in transgenic Msh2−/− mice, unlike in another transgenic mice model (23, 29). In contrast, in the absence of MSH2, instability was shifted from expansions to contractions, both across generations and in tissues including spermatozoa (49). It is still unclear whether expansions in Msh2+/+ mice and contractions in Msh2−/− mice are produced by the same mechanism, with the direction of the instability depending on the presence of MSH2, or whether a totally different mechanism processes the CTG repeat in the absence of MSH2, leading to contractions instead of expansions. We decided to determine whether expansions in Msh2+/+ males and contractions in Msh2−/− males are observed at the same stage of spermatogenesis, in order to obtain further insight into the mechanism(s) involved in both types of change in CTG repeat length. We therefore monitored the CTG repeat contractions throughout spermatogenesis in Msh2−/− males and observed that contractions were already present in spermatogonia, with a length mosaicism similar to that observed in spermatozoa. The role of two partners of MSH2 in somatic instability was recently investigated in knock-in mice carrying the last exons of the human DMPK gene with 80 CTGs (54). The absence of MSH3 leads to the loss of somatic expansions, suggesting that MSH3, like MSH2, is involved in the somatic expansions of CTG repeats in mice. During mouse spermatogenesis, Msh2 is usually expressed to higher levels in spermatogonia and early primary spermatocytes than in other germinal cell types or tissues, whereas the expression of Msh3 is very weak in spermatogonia but peaks in early primary spermatocytes (preleptotene and leptotene) (47). MSH2 is crucial for expansions in spermatogonia, and MSH3 is probably important in germinal instability given its involvement in somatic mosaicism. However, as the level of expression of Msh3 is very low in spermatogonia, the stage where expansions occur in our mice, it is important to determine whether the absence of MSH3, a potential partner of MSH2 in the mechanisms generating expansions, modifies germinal instability. The formation of contractions observed in the absence of MSH2 remains limited to spermatogonia in our mice, as no further deletions were observed in the following stages of spermatogenesis. We showed that contractions and MSH2-dependent expansions are probably produced at the same stage of spermatogenesis in DM300-328 mice. However, this does not imply that expansions and contractions result from the same repair pathways, and the proteins involved in these events remain to be identified.

A role for MSH2 in the generation of germinal expansions via gap repair was previously proposed by Kovtun and McMurray (23). In mice carrying about 120 CAG repeats, expansions during spermatogenesis seem to be limited to the latest stages of spermatogenesis, with no mosaicism detectable in the earlier stages, including spermatogonia, whereas instability appears to occur before meiosis in human HD patients (55). When Msh2 is missing in these transgenic mice, no expansions occur in spermatozoa and the CAG repeat is stabilized, suggesting that germinal expansions are produced in spermatozoa by an Msh2-dependent mechanism. After the completion of meiosis, postmeiotic repair might occur in spermatids to ensure the DNA integrity of future gametes. However, at this step, the expression levels of many repair genes and genes encoding enzymes such as MSH2, MSH3, or PMS2 dramatically decrease, eventually becoming undetectable in elongated spermatids (47). In contrast, in our transgenic mice, germinal expansions appear to be generated in spermatogonia, closely matching what was reported for human HD patients (55). In addition, the absence of MSH2 shifts instability from expansions to contractions. One of the first hypotheses to explain these differences in the dynamics and timing of trinucleotide repeat instability is the different genomic contexts surrounding the repeats in these models, as it has already been suggested that they affect the direction and range of repeat-length changes (6, 9, 18, 26a, 50, 56). Furthermore, the frequency and size of intergenerational repeat expansions were very high in our mice. The different technical approaches could also partly explain the differences. We collected spermatozoa from the vas deferens and then extracted DNA from spermatozoa heads rather than from epididymis sperm. We have previously observed that DNA from epididymal cells (where instability of the repeat is very high, especially in the cauda epidydimis [data not shown]) can contaminate the DNA extracted from sperm without a specific protocol. By combining mitochondrial and DNA labelings, we were able to distinguish seven cell types found during spermatogenesis, whereas only three can be distinguished when only DNA content and cell size are analyzed. The range of trinucleotide repeat-length changes also differs, as does the method used to measure them in the two cases. We used single-molecule PCR on a large set of DNA samples containing single genomes. This is a very powerful tool for resolving the mosaic smears obtained after classical PCR amplification of trinucleotide repeats (35). We were able to detect a broad range of expansions in spermatogonia and spermatozoa, from +1 to +150 CTGs with a mean of about +30 CTGs. These expansions were clearly distinguishable from the PCR stutter classically observed during triplet-repeat amplification, due to polymerase slippage. This overcomes the problem of small expansions and contractions when the DNA input is too great, as discussed by Zhang et al. (56).

Our results imply that CTG repeat-length mosaicism is already present in germinal cells prior to meiosis and that the mechanism responsible for generating expansions is meiosis independent, occurs continuously throughout life, and involves MSH2. This limits the number of molecular mechanisms that can generate expansions. Interestingly, in addition to their role in mismatch repair, MSH2 and MSH3 are also involved in the repair of double-strand breaks by homologous recombination (15). MSH2 and MSH3 together ensure the homology of the sequence used as a template for this type of repair and scan this template for mismatches, small insertions, and deletions. Therefore, this process does not seem to be involved in the production of expansions in our mice, as shown by our previous results with Rad52 and Rad54, both of which are major actors in this repair pathway but which have no effect on the frequency of CTG repeat expansions (49). MSH2 also participates in single-strand annealing (SSA), which is able to repair double-strand breaks but does not involve classical double-strand break repair proteins like Rad51 or Rad54 (40). It would be interesting to study the effects of the ERRC1 and ERCC4 proteins (encoded by the mammalian homologs of the Saccharomyces cerevisiae Rad10 and Rad1 genes, respectively) on CTG repeat-size mosaicism, as these proteins, together with MSH2, are crucial for the ability of SSA to correct DNA damage (39). As trinucleotide repeat tracts are length-dependent breaking sites in yeast (16), the repeated action of the SSA pathway on these particular sequences may be the factor leading to repeat instability in some tissues. In all cases, the occurrence of germinal expansion events in spermatogonia, which are mitotically dividing 2n germ cells, suggests that the mechanisms responsible for instability in germ line and somatic tissues are potentially identical.


We thank E. Lodato for caring for the mice.

This work was supported by grants from INSERM, the Association Française contre les Myopathies (AFM), and the Université René-Descartes Paris V. C.S. was supported by a grant from the Ministère Français de la Recherche et de la Technologie.


1. Anvret, M., G. Ahlberg, U. Grandell, B. Hedberg, K. Johnson, and L. Edstrom. 1993. Larger expansions of the CTG repeat in muscle compared to lymphocytes from patients with myotonic dystrophy. Hum. Mol. Genet. 2:1397-1400. [PubMed]
2. Ashizawa, T., M. Anvret, M. Baiget, J. M. Barcelo, H. Brunner, A. M. Cobo, B. Dallapiccola, R. G. Fenwick, Jr., U. Grandell, H. Harley, et al. 1994. Characteristics of intergenerational contractions of the CTG repeat in myotonic dystrophy. Am. J. Hum. Genet. 54:414-423. [PMC free article] [PubMed]
3. Aslanidis, C., G. Jansen, C. Amemiya, G. Shutler, M. Mahadevan, C. Tsilfidis, C. Chen, J. Alleman, N. G. Wormskamp, M. Vooijs, J. Buxton, K. Johnson, H. J. M. Smeets, G. G. Lennon, A. V. Carrano, R. G. Korneluk, B. Wieringa, and P. J. de Jond. 1992. Cloning of the essential myotonic dystrophy region and mapping of the putative defect. Nature 355:548-551. [PubMed]
4. Baarends, W. M., R. van der Laan, and J. A. Grootegoed. 2001. DNA repair mechanisms and gametogenesis. Reproduction 121:31-39. [PubMed]
5. Bocker, T., A. Barusevicius, T. Snowden, D. Rasio, S. Guerrette, D. Robbins, C. Schmidt, J. Burczak, C. M. Croce, T. Copeland, A. J. Kovatich, and R. Fishel. 1999. hMSH5: a human MutS homologue that forms a novel heterodimer with hMSH4 and is expressed during spermatogenesis. Cancer Res. 59:816-822. [PubMed]
6. Brock, G. J. R., H. A. Niall, and D. G. Monckton. 1999. Cis-acting modifiers of expanded CAG/CTG triplet repeat expandability: associations with flanking GC content and proximity to CpG islands. Hum. Mol. Genet. 8:1061-1067. [PubMed]
7. Brook, J. D., M. E. McCurrach, H. G. Harley, H. J. Buckler, D. Church, H. Aburatani, K. Hunter, V. P. Stanton, J. P. Thirion, T. Hudson, R. Sohn, B. Zemelman, R. G. Snell, S. A. Rundle, S. Crow, J. Davies, P. Shelbourne, J. Buxton, C. Jones, V. Juvonen, K. Johnson, P. S. Harper, D. J. Shaw, and D. E. Housman. 1992. Molecular basis of myotonic dystrophy: expansion of a trinucleotide (CTG) repeat at the 3′ end of a transcript encoding a protein kinase family member. Cell 68:799-808. [PubMed]
8. Chong, S. S., A. E. McCall, J. Cota, S. H. Subramony, H. T. Orr, M. R. Hughes, and H. Y. Zoghbi. 1995. Gametic and somatic tissue-specific heterogeneity of the expanded SCA1 CAG repeat in spinocerebellar ataxia type 1. Nat. Genet. 10:344-350. [PubMed]
9. Cleary, J. D., K. Nichol, Y. H. Wang, and C. E. Pearson. 2002. Evidence of cis-acting factors in replication-mediated trinucleotide repeat instability in primate cells. Nat. Genet. 31:37-46. [PubMed]
10. Cohen, H., D. D. Sears, D. Zenvirth, P. Hieter, and G. Simchen. 1999. Increased instability of human CTG repeat tracts on yeast artificial chromosomes during gametogenesis. Mol. Cell. Biol. 19:4153-4158. [PMC free article] [PubMed]
11. Cummings, C. J., and H. Y. Zoghbi. 2000. Fourteen and counting: unraveling trinucleotide repeat diseases. Hum. Mol. Genet. 9:909-916. [PubMed]
12. David, G., A. Durr, G. Stevanin, G. Cancel, N. Abbas, A. Benomar, S. Belal, A. S. Lebre, M. Abada-Bendib, D. Grid, M. Holmberg, M. Yahyaoui, F. Hentati, T. Chkili, Y. Agid, and A. Brice. 1998. Molecular and clinical correlations in autosomal dominant cerebellar ataxia with progressive macular dystrophy (SCA7). Hum. Mol. Genet. 7:165-170. [PubMed]
13. Delatycki, M. B., D. Paris, R. J. Gardner, K. Forshaw, G. A. Nicholson, N. Nassif, R. Williamson, and S. M. Forrest. 1998. Sperm DNA analysis in a Friedreich ataxia premutation carrier suggests both meiotic and mitotic expansion in the FRDA gene. J. Med. Genet. 35:713-716. [PMC free article] [PubMed]
14. De Michele, G., F. Cavalcanti, C. Criscuolo, L. Pianese, A. Monticelli, A. Filla, and S. Cocozza. 1998. Parental gender, age at birth and expansion length influence GAA repeat intergenerational instability in the X25 gene: pedigree studies and analysis of sperm from patients with Friedreich's ataxia. Hum. Mol. Genet. 7:1901-1906. [PubMed]
15. Evans, E., N. Sugawara, J. E. Haber, and E. Alani. 2000. The Saccharomyces cerevisiae Msh2 mismatch repair protein localizes to recombination intermediates in vivo. Mol. Cell 5:789-799. [PubMed]
16. Freudenreich, C. H., S. M. Kantrow, and V. A. Zakian. 1998. Expansion and length-dependent fragility of CTG repeats in yeast. Science 279:853-856. [PubMed]
17. Fu, Y. H., D. P. A. Kuhl, A. Pizzuti, J. R. G. Fenwick, J. King, S. Rajnarayan, P. W. Dunne, J. Dubel, G. A. Nasser, T. Ashizawa, P. de Jong, B. Wieringa, R. Korneluk, M. B. Perryman, H. F. Epstein, and C. Caskey. 1992. An unstable triplet repeat in a gene related to myotonic muscular dystrophy. Science 255:1256-1258. [PubMed]
18. Gourdon, G., P. Dessen, A. S. Lia, C. Junien, and H. Hoffman-Radvanyi. 1997. Intriguing association between disease associated unstable trinucleotide repeat and CpG island. Ann. Genet. 40:73-77. [PubMed]
19. Harper, P. S., H. G. Harley, W. Reardon, and D. J. Shaw. 1992. Anticipation in myotonic dystrophy: new light on an old problem. Am. J. Hum. Genet. 51:10-16. [PMC free article] [PubMed]
20. Henricksen, L. A., S. Tom, Y. Liu, and R. A. Bambara. 2000. Inhibition of flap endonuclease 1 by flap secondary structure and relevance to repeat sequence expansion. J. Biol. Chem. 275:16420-16427. [PubMed]
21. Jankowski, C., and D. K. Nag. 2002. Most meiotic CAG repeat tract-length alterations in yeast are SPO11 dependent. Mol. Genet. Gen. 267:64-70. [PubMed]
22. Jansen, G., P. Willems, M. Coerwinkel, W. Nillesen, H. Smeets, L. Vits, C. Howeler, H. Brunner, and B. Wieringa. 1994. Gonosomal mosaicism in myotonic dystrophy patients: involvement of mitotic events in (CTG)n repeat variation and selection against extreme expansion in sperm. Am. J. Hum. Genet. 54:575-585. [PMC free article] [PubMed]
23. Kovtun, I. V., and C. T. McMurray. 2001. Trinucleotide expansion in haploid germ cells by gap repair. Nat. Genet. 27:407-411. [PubMed]
24. Lavedan, C., H. Hofmann-Radvanyi, P. Shelbourne, J. P. Rabes, C. Duros, D. Savoy, I. Dehaupas, S. Luce, K. Johnson, and C. Junien. 1993. Myotonic dystrophy: size- and sex-dependent dynamics of CTG meiotic instability, and somatic mosaicism. Am. J. Hum. Genet. 52:875-883. [PMC free article] [PubMed]
25. Leeflang, E. P., S. Tavare, P. Marjoram, C. O. Neal, J. Srinidhi, H. MacFarlane, M. E. MacDonald, J. F. Gusella, M. de Young, N. S. Wexler, and N. Arnheim. 1999. Analysis of germline mutation spectra at the Huntington's disease locus supports a mitotic mutation mechanism. Hum. Mol. Genet. 8:173-183. [PubMed]
26. Lia, A., H. Seznec, H. Hofmann-Radvanyi, F. Radvanyi, C. Duros, C. Saquet, M. Blanche, C. Junien, and G. Gourdon. 1998. Somatic instability of the CTG repeat in mice transgenic for the myotonic dystrophy region is ade-dependent but not correlated to the relative intertissue transcription levels and proliferative capacities. Hum. Mol. Genet. 7:1285-1291. [PubMed]
26a. Libby, R. T., D. G. Monckton, Y.-H. Fu, R. A. Martinez, J. P. McAbney, R. Lau, D. D. Einum, K. Nichol, C. B. Ware, L. J. Ptacek, C. E. Pearson, and A. R. La Spada. 2003. Genomic context drives SCA7 CAG repeat instability, while expressed SCA7 cDNAs are intergenerationally and somatically stable in transgenic mice. Hum. Mol. Genet. 12:41-50. [PubMed]
27. Mahadevan, M., C. Tsilfidis, L. Sabourin, G. Shutler, C. Amemiya, G. Jansen, C. Neville, M. Narang, J. Barcelo, K. O'Hoy, S. Leblond, J. Earle-Macdonald, P. J. de Jong, B. Wieringa, and R. G. Korneluk. 1992. Myotonic dystrophy mutation: an unstable CTG repeat in the 3′ untranslated region of the gene. Science 255:1253-1255. [PubMed]
28. Mangiarini, L., K. Sathasivam, A. Mahal, R. Mott, M. Seller, and G. P. Bates. 1997. Instability of highly expanded CAG repeats in mice transgenic for the Huntington's disease mutation. Nat. Genet. 15:197-200. [PubMed]
29. Manley, K., T. L. Shirley, L. Flaherty, and A. Messer. 1999. Msh2 deficiency prevents in vivo somatic instability of the CAG repeat in Huntington disease transgenic mice. Nat. Genet. 23:471-473. [PubMed]
30. Martorell, L., D. G. Monckton, J. Gamez, and M. Baiget. 2000. Complex patterns of male germline instability and somatic mosaicism in myotonic dystrophy type 1. Eur. J. Hum. Genet. 8:423-430. [PubMed]
31. McLaren, A. 1998. Germ cells and germ cell transplantation. Int. J. Dev. Biol. 42:855-860. [PubMed]
32. McLaren, A. 2001. Mammalian germ cells: birth, sex, and immortality. Cell Struct. Funct. 26:119-122. [PubMed]
33. McMurray, C. T., and I. V. Kortun. 2003. Repair in haploid male germ cells occurs late in differentiation as chromatin is condensing. Chromosoma 111:505-508. [PubMed]
34. Monckton, D. G., M. L. Cayuela, F. K. Gould, G. J. Brock, R. Silva, and T. Ashizawa. 1999. Very large (CAG)(n) DNA repeat expansions in the sperm of two spinocerebellar ataxia type 7 males. Hum. Mol. Genet. 8:2473-2478. [PubMed]
35. Monckton, D. G., L. J. C. Wong, T. Ashizawa, and C. T. Caskey. 1995. Somatic mosaicism, germline expansions, germline reversions and intergenerational reductions in myotonic dystrophy males: small pool PCR analyses. Hum. Mol. Genet. 4:1-8. [PubMed]
36. Mori, H., D. Shimizu, R. Fukunishi, and A. K. Christensen. 1982. Morphometric analysis of testicular Leydig cells in normal adult mice. Anat. Rec. 204:333-339. [PubMed]
37. Moutou, C., M. C. Vincent, V. Biancalana, and J. L. Mandel. 1997. Transition from premutation to full mutation in fragile X syndrome is likely to be prezygotic. Hum. Mol. Genet. 6:971-979. [PubMed]
38. Nolin, S. L., G. E. Houck, Jr., A. D. Gargano, H. Blumstein, C. S. Dobkin, and W. T. Brown. 1999. FMR1 CGG-repeat instability in single sperm and lymphocytes of fragile-X premutation males. Am. J. Hum. Genet. 65:680-688. [PMC free article] [PubMed]
39. Pastink, A., J. C. Eeken, and P. H. Lohman. 2001. Genomic integrity and the repair of double-strand DNA breaks. Mutat. Res. 480-481:37-50. [PubMed]
40. Pastink, A., and P. H. Lohman. 1999. Repair and consequences of double-strand breaks in DNA. Mutat. Res. 428:141-156. [PubMed]
41. Pearson, C. E., and R. R. Sinden. 1998. Slipped strand DNA, dynamic, mutations, and human disease, p. 585-623. In R. D. Wells and S. T. Warren (ed.), Genetic instabilities and hereditary neurological diseases. Academic Press, San Diego, Calif.
42. Pearson, C. E., M. Tam, Y. H. Wang, S. E. Montgomery, A. C. Dar, J. D. Cleary, and K. Nichol. 2002. Slipped-strand DNAs formed by long (CAG)*(CTG) repeats: slipped-out repeats and slip-out junctions. Nucleic Acids Res. 30:4534-4547. [PMC free article] [PubMed]
43. Pelletier, R., M. M. Krasilnikova, G. M. Samadashwily, R. Lahue, and S. M. Mirkin. 2003. Replication and expansion of trinucleotide repeats in yeast. Mol. Cell. Biol. 23:1349-1357. [PMC free article] [PubMed]
44. Petit, J. M., M. H. Ratinaud, E. Cordelli, M. Spano, and R. Julien. 1995. Mouse testis cell sorting according to DNA and mitochondrial changes during spermatogenesis. Cytometry 19:304-312. [PubMed]
45. Petruska, J., M. J. Hartenstine, and M. F. Goodman. 1998. Analysis of strand slippage in DNA polymerase expansions of CAG/CTG triplet repeats associated with neurodegenerative disease. J. Biol. Chem. 273:5204-5210. [PubMed]
46. Ranum, L. P., and J. W. Day. 2002. Dominantly inherited, non-coding microsatellite expansion disorders. Curr. Opin. Genet. Dev. 12:266-271. [PubMed]
47. Richardson, L. L., C. Pedigo, and M. Ann Handel. 2000. Expression of deoxyribonucleic acid repair enzymes during spermatogenesis in mice. Biol. Reprod. 62:789-796. [PubMed]
48. Sato, T., M. Oyake, K. Nakamura, K. Nakao, Y. Fukusima, O. Onodera, S. Igarashi, H. Takano, K. Kikugawa, Y. Ishida, T. Shimohata, R. Koide, T. Ikeuchi, H. Tanaka, N. Futamura, R. Matsumura, T. Takayanagi, F. Tanaka, G. Sobue, O. Komure, M. Takahashi, A. Sano, Y. Ichikawa, J. Goto, I. Kanazawa, M. Katsuki, and S. Tsuji. 1999. Transgenic mice harboring a full-length human mutant DRPLA gene exhibit age-dependent intergenerational and somatic instabilities of CAG repeats comparable with those in DRPLA patients. Hum. Mol. Genet. 8:99-106. [PubMed]
49. Savouret, C., E. Brisson, J. Essers, R. Kanaar, A. Pastink, H. te Riele, C. Junien, and G. Gourdon. 2003. CTG repeat instability and size variation timing in DNA repair-deficient mice. EMBO J. 22:2264-2273. [PMC free article] [PubMed]
50. Seznec, H., A. Lia-Baldini, C. Duros, C. Fouquet, C. Lacroix, H. Hofmann-Radvanyi, C. Junien, and G. Gourdon. 2000. Transgenic mice carrying large human genomic sequences with expanded CTG repeat mimic closely the DM CTG repeat intergenerational and somatic instability. Hum. Mol. Genet. 9:1185-1194. [PubMed]
51. Spiro, C., R. Pelletier, M. L. Rolfsmeier, M. J. Dixon, R. S. Lahue, G. Gupta, M. S. Park, X. Chen, S. V. Mariappan, and C. T. McMurray. 1999. Inhibition of FEN-1 processing by DNA secondary structure at trinucleotide repeats. Mol. Cell 4:1079-1085. [PubMed]
52. Takiyama, Y., K. Sakoe, M. Amaike, M. Soutome, T. Ogawa, I. Nakano, and M. Nishizawa. 1999. Single sperm analysis of the CAG repeats in the gene for dentatorubral-pallidoluysian atrophy (DRPLA): the instability of the CAG repeats in the DRPLA gene is prominent among the CAG repeat diseases. Hum. Mol. Genet. 8:453-457. [PubMed]
53. Telenius, H., E. Almqvist, B. Kremer, N. Spence, F. Squitieri, K. Nichol, U. Grandell, E. Starr, C. Benjamin, I. Castaldo, et al. 1995. Somatic mosaicism in sperm is associated with intergenerational (CAG)n changes in Huntington disease. Hum. Mol. Genet. 4:189-195. [PubMed]
54. van den Broek, W. J., M. R. Nelen, D. G. Wansink, M. M. Coerwinkel, H. te Riele, P. J. Groenen, and B. Wieringa. 2002. Somatic expansion behaviour of the (CTG)n repeat in myotonic dystrophy knock-in mice is differentially affected by Msh3 and Msh6 mismatch-repair proteins. Hum. Mol. Genet. 11:191-198. [PubMed]
54a. Wheeler, V. C., L. A. Lebel, V. Vrbanar, A. Teed, H. te Riele, and M. E. MacDonald. 2003. Mismatch repair gene Msh2 modifies the timing of early disease in HdhQ11 striatum. Hum. Mol. Genet. 12:273-281. [PubMed]
55. Yoon, S. R., L. Dubeau, M. De Young, N. S. Wexler, and N. Arnheim. 2003. Huntington disease expansion mutations in humans can occur before meiosis is completed. Proc. Natl. Acad. Sci. USA 100:8834-8838. [PMC free article] [PubMed]
56. Zhang, Y., D. G. Monckton, M. J. Siciliano, T. H. Connor, and M. L. Meistrich. 2002. Age and insertion site dependence of repeat number instability of a human DM1 transgene in individual mouse sperm. Hum. Mol. Genet. 11:791-798. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...