| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Cis and trans effects of the myotonic dystrophy (DM) mutation in a cell culture model
Introduction
Results
The DM mutation has a cis effect on gene expression
RNA foci formation
The mutant DMPK 3[prime]-UTR mRNA acts in trans to inhibit myoblast fusion
Reversion of the myoblast fusion defect
Discussion
Cis effects of the DM mutation on gene expression
The mutant DMPK 3[prime]-UTR mRNA inhibits myoblast fusion
Materials And Methods
Reporter constructs
Cell culture
Transfection and analysis
Immunocytochemistry and RNA-FISH
Myoblast differentiation
Southern blotting
RNA analysis
Acknowledgements
References
Cis and trans effects of the myotonic dystrophy (DM) mutation in a cell culture model
Received June 11, 1999; Revised and Accepted July 26, 1999
The mutation causing myotonic dystrophy (DM) has been identified as a CTG expansion in the 3[prime]-untranslated region (3[prime]-UTR) of the DM protein kinase gene (DMPK), but the mechanism(s) of pathogenesis remain unknown. Studies using DM patient materials have often produced confusing results. Therefore, to study the effects of the DM mutation in a controlled environment, we have established a cell culture model system using C2C12 mouse myoblasts. By expressing chimeric reporter constructs containing a reporter gene fused to a human DMPK 3[prime]-UTR, we identified both cis and trans effects that are mediated by the DM mutation. Our data show that a mutant DMPK 3[prime]-UTR, with as few as 57 CTGs, had a negative cis effect on protein expression and resulted in the aggregation of reporter transcripts into discrete nuclear foci. We determined by deletion analysis that an expanded (CTG)n tract alone was sufficient to mediate these cis effects. Furthermore, in contrast to the normal DMPK 3[prime]-UTR mRNA, a mutant DMPK 3[prime]-UTR mRNA with (CUG)200 selectively inhibited myogenic differentiation of C2C12 myoblasts. Genetic analysis and the Cre-loxP system were used to clearly demonstrate that the myoblast fusion defect could be rescued by eliminating the expression of the mutant DMPK 3[prime]-UTR transcript. Characterization of spontaneous deletion events mapped the inhibitory effect to the (CTG)n expansion and/or the 3[prime] end of the DMPK 3[prime]-UTR. These results provide evidence that the DM mutation acts in cis to reduce protein production (consistent with DMPK haploinsufficiency) and in trans as a `riboregulator' to inhibit myogenesis.
INTRODUCTION
Myotonic dystrophy (DM) is an autosomal dominant neuromuscular disorder that affects 1 in 8000 individuals worldwide (1). DM is highly variable in clinical symptoms, symptom severity and age of onset. Classical adult onset DM develops during early adulthood and is characterized primarily by myotonia, muscle weakness and progressive muscle wasting, but can also include defects in other systems such as cardiac conduction disturbances, cataracts, premature frontal balding and testicular atrophy (1). Congenital myotonic dystrophy (CDM) is the most severe form of the disorder. CDM patients initially suffer from profound hypotonia, mental retardation and developmental abnormalities. In addition, they face a 25% neonatal mortality rate (1). Survivors go on to develop the adult form of DM, usually in their teenage years. Skeletal muscle biopsies from CDM patients display a marked immaturity of muscle fibers and an increased number of satellite cells, suggesting a defect in muscle maturation (2-5).
The DM mutation has been identified as an expanded CTG trinucleotide repeat tract in the 3[prime]-untranslated region (3[prime]-UTR) of the DM protein kinase gene (DMPK) (6-8). The number of CTG repeats ranges from five to 37 in unaffected individuals, whereas DM patients have from 50 to several thousand (6-8). The molecular mechanisms by which the CTG expansion causes DM are unknown. Various hypotheses including alterations in DMPK protein levels, chromatin effects on the expression of neighboring genes and altered RNA processing are all being actively pursued (for review see ref. 9). The effect of the DM mutation on DMPK remains controversial. Studies comparing DMPK expression between DM affected and unaffected patient samples have resulted in reports of increased (10), decreased (11-15) and unaltered (16) levels of DMPK mRNA and protein. Furthermore, transgenic mice expressing either increased or decreased amounts of DMPK fail to display the DM phenotype or only exhibit some aspects of DM (17-19). This suggests that changes in DMPK levels alone are insufficient to cause DM.
Many studies have looked at the effects of the DM mutation on DMPK mRNA. Some have used allele-specific quantitative RT-PCR assays to show that mutant DMPK alleles produce less mature mRNA than wild-type alleles (12,20), implicating a defect in the processing of mutant pre-mRNA. Northern blot and RNA fluorescence in situ hybridization (FISH) experiments have demonstrated that mutant DMPK transcripts are retained in the nuclei of DM patient cells and aggregate into distinct foci (21-23). Others have provided evidence that the processing of the normal DMPK transcript and insulin receptor mRNA is also affected in DM patients, leading the authors to hypothesize that the mutant DMPK mRNA acts in a trans-dominant manner to disrupt cellular RNA metabolism (20,24). The mutant DMPK transcript could exert these effects through aberrant gain-of-function interactions with factors involved in RNA metabolism. One such factor may be CUG-BP1, a 50 kDa heterogeneous nuclear ribonucleoprotein that binds the DMPK 3[prime]-UTR RNA and other CUG repeat tracts (25,26). It has been shown that CUG-BP1 regulates splicing of cardiac troponin T (cTNT) pre-mRNA, and that cTNT splicing is altered in DM muscle tissue and normal cells over-expressing CUG-containing mRNAs (27).
More recently, several groups have reported that over-expression of murine or human DMPK mRNA inhibits the terminal differentiation of mouse myoblasts in cell culture, an effect reminiscent of defective muscle differentiation observed in CDM patients (28-31). By over-expressing fragments of the DMPK 3[prime]-UTR in C2C12 mouse myoblasts, Sabourin et al. mapped the inhibitory effect to a 239 bp region 5[prime] of the CTG repeats (29). Two other reports suggested that either expression of a DMPK cDNA with (CTG)46 (30) or expressing a 500 bp CTG tract alone (31) can inhibit differentiation.
In this study we report the development of a new cell culture model system to investigate the effects of the DM mutation. We have used reporter constructs that consist of a reporter gene fused to either a normal or expanded human DMPK 3[prime]-UTR to provide a quantitative demonstration that the inclusion of the mutant DMPK 3[prime]-UTR has significant negative effects on translation (i.e. a cis effect). Furthermore, we have found that heterologous mRNAs containing the mutant DMPK 3[prime]-UTR are retained in the nuclei of mouse C2C12 myoblast cells and form distinct foci in an analogous manner to what is observed of mutant DMPK mRNAs in DM patient cells. In addition, we determined that the expression of a transcript with a CUG expansion alone is sufficient to cause the cis effect and nuclear foci formation. Lastly, through application of the Cre-loxP system, and the isolation and characterization of spontaneous deletion events, we provide unambiguous evidence that the expression of the mutant DMPK 3[prime]-UTR mRNA has a trans effect resulting in a disease relevant phenotype, namely delayed muscle differentiation. Taken together, the results from our cell culture model system demonstrate that the DM mutation exerts both cis effects on mRNA transcripts containing the mutation and trans effects mediated by the expression of the mutant DMPK 3[prime]-UTR mRNA that may contribute to DM pathogenesis.
RESULTS
The DM mutation has a cis effect on gene expression
To determine the effect of the DM mutation on gene expression, we took the 3[prime]-UTR sequence out of the context of the human DMPK gene and inserted it into the 3[prime]-UTR of the bacterial reporter genes chloramphenicol acetyltransferase (CAT) and lacZ (encoding [beta]-galactosidase) (Fig. 1A). A series of CAT reporter constructs, each containing a DMPK 3[prime]-UTR with a different number of CTG repeats [ranging from (CTG)5 to (CTG)200], were transiently transfected into C2C12 myoblasts. CAT protein production from cells transfected with CAT+(CTG)5 DMPK 3[prime]-UTR was 42% of the levels in cells transfected with a control vector (CAT gene alone). Transfection with constructs with an expanded DMPK 3[prime]-UTR [(CTG)57, (CTG)78 or (CTG)200] resulted in CAT protein levels between 13 and 16% of the control transfections (Fig. 1B). Thus, relative to a normal [(CTG)5] DMPK 3[prime]-UTR, a CTG expansion (57-200 CTGs) decreases CAT gene expression ~3-fold. Using lacZ as the reporter gene, we again observed an ~3-fold (2.8- versus 2.7-fold) decrease in expression from constructs with CTG expansions (data not shown). These results showed that a mutant DMPK 3[prime]-UTR had a significant negative cis effect on gene expression.
Figure 1. A CTG expansion has a negative cis effect on gene expression. (A) Structure of the reporter constructs transiently expressed in C2C12 to determine the effect of the DM mutation on protein expression. A CMV promoter drives expression of either lacZ or CAT fused to DMPK 3[prime]-UTR sequence. The 5[prime] region, 3[prime] region and (CTG)100-expansion-only fragments used to map the cis effect are shown. (B) Results from transfection experiments with CAT reporter constructs. Protein production from CAT constructs containing a DMPK 3[prime]-UTR with a CTG expansion, [(CTG)57, (CTG)78 or (CTG)200] was ~3-fold lower than a (CTG)5 DMPK 3[prime]-UTR [P = 0.0005 between (CTG)5 and (CTG)57 and P = 0.00003 between (CTG)5 and (CTG)200]. CAT constructs carrying a DMPK 3[prime]-UTR with the CTG repeats deleted [[Delta](CTG)] or just the 5[prime] region or 3[prime] region of the 3[prime]-UTR, expressed CAT levels greater than the (CTG)5 DMPK 3[prime]-UTR construct. A (CTG)100-expansion-only construct produced CAT expression levels similar to those produced by constructs with a mutant DMPK 3[prime]-UTR. The CAT expression level from each construct was normalized to [beta]-galactosidase to correct for transfection efficiency and is presented relative to the level produced by a CAT control vector, which was set at 100%.
In order to delineate the minimal DMPK 3[prime]-UTR sequences required to mediate the cis effect, we analyzed a series of deletions of the DMPK 3[prime]-UTR (Fig. 1A). We did not observe a negative cis effect on CAT expression with CAT constructs containing the 5[prime] region of the DMPK 3[prime]-UTR (the 225 bp 5[prime] of the CTG repeats), the 3[prime] region of the DMPK 3[prime]-UTR (the 484 bp 3[prime] of the CTG repeats) or a DMPK 3[prime]-UTR in which the CTG repeat tract had been deleted [[Delta](CTG)] (Fig. 1B). In contrast, a CAT reporter construct with 100 CTG repeats and minimal flanking sequences [(CTG)100 expansion only] expressed CAT levels similar to constructs with an intact mutant DMPK 3[prime]-UTR (Fig. 1B). Together these results showed that a CTG expansion alone is sufficient to cause a negative cis effect on gene expression.
RNA foci formation
Recently, it has been shown that mutant DMPK transcripts accumulate in the nuclei of DM patient cells and aggregate to form distinct foci (21-23). One possible explanation of the negative effect seen on reporter gene protein levels is that transcripts with increased CUG repeats are retained in the nucleus and so are not available for translation. To test whether reporter constructs with an expanded DMPK 3[prime]-UTR produce nuclear RNA foci in our cell culture system, we transiently transfected C2C12 cells with our reporter constructs. Transfected cells were identified by indirect immunofluorescent staining with an antibody against the reporter protein (CAT or [beta]-galactosidase), and then observed for nuclear RNA foci, as detected by RNA-FISH using a fluorescently labeled antisense (CAG)10 oligonucleotide probe. Reporter constructs (both CAT and lacZ) with a mutant DMPK 3[prime]-UTR containing [ge]57 CTG repeats produced nuclear foci, whereas constructs with a normal DMPK 3[prime]-UTR with 5 or 26 CTG repeats, and the control vector, did not (Fig. 2C-E). Several experiments were done to determine whether the foci were due to RNA or DNA. We found that: (i) the foci were resistant to DNase I; (ii) the foci were sensitive to RNase A; and (iii) the foci were not detected using a sense (CTG)10 probe (data not shown). Thus, the nuclear foci are aggregations of the reporter transcripts containing an expanded DMPK 3[prime]-UTR. These results reproduced the behavior of the mutant DMPK mRNA observed in DM patient cells (Fig. 2B), and demonstrated that foci formation can occur in murine cells.
Figure 2. Nuclear RNA foci formation. RNA-FISH analysis with a CY3 fluorescently labeled (CAG)10 oligonucleotide probe (red) detected RNA foci in the nuclei of fibroblasts from a DM patient (B) but not an unaffected individual (A). RNA foci were also detected in C2C12 myoblasts transiently transfected with CAT reporter constructs containing a mutant DMPK 3[prime]-UTR (57, 78 or 200 CTG repeats) (E) or a (CTG)100 expansion only (F). Foci were not detected in cells transfected with a construct containing a normal DMPK 3[prime]-UTR (D) or a control CAT vector (C). Indirect immunostaining with anti-CAT antibody identified transfected cells (green) and nuclei were counterstained with DAPI (blue).
Since the negative cis effect on gene expression mapped to the CTG repeats, we next tested whether a CTG expansion alone is also sufficient to mediate RNA foci formation. Using RNA-FISH analysis of cells transiently transfected with the (CTG)100 expansion only reporter construct, we found cells with nuclear RNA foci indistinguishable from cells transfected with a reporter construct carrying an entire DMPK 3[prime]-UTR with 57-200 CTG repeats (Fig. 2F). Therefore, a CTG expansion is the only sequence required for nuclear RNA foci formation.
The mutant DMPK 3[prime]-UTR mRNA acts in trans to inhibit myoblast fusion
To further study the effects of the DMPK 3[prime]-UTR in a disease relevant cell culture system, we have established stable C2C12 mouse myoblast clones that express a lacZ reporter construct containing a normal [(CTG)5], mutant [(CTG)200], or no DMPK 3[prime]-UTR. C2C12 myoblasts have been used extensively as a model system to study muscle cell biology and myogenic differentiation (32,33). The removal of growth factors and addition of low serum `differentiation media' will induce C2C12 myoblasts to exit the cell cycle and initiate terminal differentiation by fusing with one another to form elongated, multinucleated myotubes.
Stable C2C12 myoblast clones were established by selecting for neomycin resistance with G418 and screened for LacZ expression by X-gal staining. LacZ positive clones were analyzed by Southern and northern blotting to confirm that an intact DMPK 3[prime]-UTR was fused to the lacZ gene. Clones with a single copy integration of the lacZ reporter construct and which expressed comparable levels of RNA were chosen for further analysis. To test the effects of a normal and mutant DMPK 3[prime]-UTR on myoblast fusion in our stable C2C12 clones, the myoblasts were cultured in differentiation media for six days. Myoblasts expressing a (CTG)5 (normal) DMPK 3[prime]-UTR (lacZ+5-40) and a (CTG)200 (mutant) DMPK 3[prime]-UTR (lacZ+200-59) expressed LacZ, but only lacZ+5-40 myoblasts fused to form elongated myotubes comparable to untransfected C2C12 (Fig. 3A-C). Myoblast fusion was not significantly different between untransfected C2C12, lacZ+5-40 and lacZ-165 (a control clone expressing lacZ with no DMPK 3[prime]-UTR), but fusion in lacZ+200-59 was reduced to 6.8 ± 12% of the average level found in untransfected C2C12 (Fig. 4E). Importantly, X-gal staining of LacZ positive myotubes (Fig. 3B) indicated that expression of [beta]-galactosidase (protein or RNA) did not affect myoblast fusion.
Figure 3. The mutant DMPK 3[prime]-UTR mRNA inhibits myoblast fusion. Untransfected C2C12 (A) and C2C12 expressing either lacZ+(CTG)5 DMPK 3[prime]-UTR (B) or lacZ+(CTG)200 DMPK 3[prime]-UTR (C) were cultured in differentiation media for 6 days and then X-gal stained. Both lacZ+5-40 and lacZ+200-59 expressed LacZ (stained blue). However, only lacZ+5-40 and untransfected C2C12 myoblasts differentiated into myotubes [arrows in (A) and (B)]. RNA-FISH analysis detected nuclear RNA foci in lacZ+200-59 (F), but not in lacZ+5-40 (E) or untransfected myoblasts (D).
Figure 4. Rescue of the myoblast fusion defect. In contrast to untransfected C2C12 (A), lacZ+200-59 myoblasts (B) did not fuse into myotubes or initiate expression of myosin heavy chain (immunostained green with MY32 antibody) after 6 days in differentiation media. However, myoblasts derived from lacZ+200-59 that have undergone a Cre-mediated CMV promoter deletion (C), or a deletion event in the DMPK 3[prime]-UTR (D) regained the ability to differentiate. Nuclei were counterstained with DAPI (blue). (E) Quantitation of myoblast fusion showed that untransfected C2C12, lacZ-165 and lacZ+5-40 fused at similar levels, but fusion in lacZ+200-59 was reduced to 6.8% of untransfected C2C12. Spontaneous deletion events within the reporter construct (Cre-8, lacZ+200-59Rev) or a Cre-mediated deletion of the CMV promoter (Cre-19) resulted in significant restoration of fusion potential.
RNA-FISH analysis of the stable clones detected nuclear foci of RNA in clones expressing a mutant DMPK 3[prime]-UTR, but not in cells expressing a normal DMPK 3[prime]-UTR (Fig. 3E-F). These results were consistent with the results from transient transfection assays (Fig. 2). By northern blot analysis, we determined that lacZ+5-40, lacZ+200-59 and lacZ-165 expressed similar steady state levels of lacZ mRNA (normalized to [beta]-actin RNA levels) (Fig. 5). Therefore, at a comparable steady state mRNA level, the mutant DMPK 3[prime]-UTR mRNA, in contrast to the normal DMPK 3[prime]-UTR mRNA, selectively inhibited myoblast fusion and muscle differentiation.
Figure 5. Northern blot analysis of lacZ expression in stable C2C12 clones. A northern blot of nuclear (N) and cytoplasmic (C) RNA extracted from C2C12 clones was probed with lacZ, DMPK 3[prime]-UTR and [beta]-actin. (A) A lacZ probe detected a 6.5 kb, predominantly nuclear transcript in lacZ+200-59. A 4 kb transcript was detected in lacZ+200-59Rev clones corresponding to an internal deletion event that removed a region including the 3[prime] 444 bp of the DMPK 3[prime]-UTR and reduced the CTG tract to (CTG)5. lacZ mRNA is not detected in untransfected C2C12 or representative lacZ+200-59 derived clones with either a Cre mediated CMV promoter deletion (Cre-19) or the entire lacZ construct deleted (Cre-8). Transcripts of the expected sizes were detected in lacZ+5-40 and lacZ-165 myoblasts. A human DMPK 3[prime]-UTR probe hybridized to the same bands recognized by the lacZ probe in lacZ+200-59, lacZ+5-40 and lacZ+200-59Rev (data not shown). (B) The northern blot was stripped and re-probed with a [beta]-actin probe to serve as a loading control. (C) The level of steady state lacZ mRNA (nuclear + cytoplasmic) normalized to [beta]-actin. lacZ mRNA steady state levels are similar in lacZ+200-59, lacZ+200-59Rev, lacZ+5-40 and lacZ-165.
Reversion of the myoblast fusion defect
C2C12 myoblasts can spontaneously lose the ability to fuse into myotubes due to random genetic changes and selection through successive passage (33). Therefore, to ensure any phenotypic and molecular changes we detected were due solely to the presence of the relevant mRNA, the CMV promoter of each reporter construct was modified to include flanking loxP sites. The loxP sequences are substrates for Cre recombinase (34). Thus, expressing Cre in our stable clones should mediate a recombination event that deletes the CMV promoter and eliminates expression of the reporter construct. The strength of this system is that analysis within the same clone can be performed in the presence and absence of the relevant mRNA, enabling unambiguous attribution of any phenotypic changes to the expression of that mRNA.
We introduced the site-specific recombinase Cre into lacZ+200-59 myoblasts to mediate the deletion of the CMV promoter and silence expression of the reporter construct. Two plasmids, one expressing Cre and the other conferring resistance to hygromycin B were co-transfected. Twenty hygromycin resistant clones were screened for LacZ expression by X-gal staining and for a 1 kb deletion of the CMV promoter by PCR and Southern blotting. Sixteen clones were LacZ negative. Thirteen had deleted the entire construct, and three clones had undergone a selective Cre mediated deletion of the CMV promoter. Southern blot analysis confirmed that the reporter construct was otherwise intact in these three clones (Fig. 6). We then tested the ability of these 16 myoblast clones to differentiate into myotubes. Four of these clones failed to form myotubes, presumably through spontaneous loss of fusion potential or selection. However, 12 of these clones had regained the ability to fuse and form myotubes, including all three clones that had undergone the Cre mediated deletion of the CMV promoter (Fig. 4). This clearly demonstrated that the myoblast fusion defect was caused by the expression of the mutant DMPK 3[prime]-UTR mRNA and not by insertional mutagenesis. Northern blotting confirmed that these twelve clones did not express the lacZ-mutant DMPK 3[prime]-UTR transcript detected previously in lacZ+200-59 cells (Fig. 5A). Furthermore, as predicted by the northern blot results, RNA-FISH analysis confirmed that the nuclear foci of mRNA were eliminated (data not shown).
Figure 6. Southern blot analysis of lacZ+200-59 and revertant clones. Genomic DNA (10 mg) was digested with EcoRI and MfeI, separated through 0.8% agarose gel and transferred to a nylon membrane. (A) A lacZ probe detected a 4.2 kb fragment in lacZ+200-59 and lacZ+200-59Rev myoblasts indicating that the CMV promoter and lacZ gene were intact. In contrast, the 1 kb CMV promoter region had been deleted in clone Cre-19 and the entire reporter construct was lost in Cre-8. No signal was detected in untransfected C2C12. (B) The Southern blot was stripped and re-probed with a human DMPK 3[prime]-UTR probe. A 2.1 kb fragment detected in lacZ+200-59 and Cre-19 demonstrated that the 3[prime]-UTR is intact in both clones. The 4 kb band in lacZ+200-59Rev corresponded to the deletion of the EcoRI site found at the 3[prime] end of lacZ+200-59 and Cre-19. This deletion event was further characterized by sequencing RT-PCR products. (C-E) Diagrams representing the structure of the reporter construct integrated in lacZ+200-59 and derived clones as determined by Southern blotting, PCR and RT-PCR. Triangles indicate loxP sites.
Four out of the 20 clones screened were LacZ positive. Southern blot and PCR analysis of the reporter construct in these clones named `lacZ+200-59Rev' (all four turned out to be clonal) revealed that the CMV promoter remained intact, but a spontaneous deletion event had altered the 3[prime] region of the construct (Fig. 6). RT-PCR was used to amplify the 3[prime] end of the lacZ mRNA expressed in the lacZ+200-59 and the lacZ+200-59Rev cells. Both clones were polyadenylated at the same site, ~2 kb 3[prime] of the DMPK poly(A) signal. Direct sequencing of the RT-PCR product from the lacZ+200-59Rev showed that an internal fragment including the last 444 bp of the 3[prime] end of the DMPK 3[prime]-UTR had been deleted and the CTG repeat tract was reduced to (CTG)5. Further confirmation of these deletions was obtained by PCR amplification of genomic DNA spanning the deletions, and subsequent automated DNA sequence analysis of the PCR product.
Significantly, we found that these four LacZ positive myoblast clones had regained their fusion capacity and were able to form myotubes upon differentiation similar to untransfected C2C12 (Fig. 4D and E). Northern blot analysis confirmed that the lacZ transcript expressed in lacZ+200-59Rev is smaller than the transcript found in the parental lacZ+200-59 cells (Fig. 5A). In addition, the steady state levels of lacZ mRNA were not significantly different between lacZ+200-59Rev and lacZ+200-59 (Fig. 5), but the four revertant clones (now with five CTG repeats) expressed ~3-fold higher levels of [beta]-galactosidase. These results closely correlated with our transient transfection results (Fig. 1B).
DISCUSSION
Cis effects of the DM mutation on gene expression
Whether the DM mutation affects DMPK protein levels has been controversial. We have used our cell culture system to study the effect of a CTG expansion on protein expression from heterologous reporter constructs. We measured an ~3-fold decrease in protein produced between a reporter construct containing a normal and mutant DMPK 3[prime]-UTR. This was seen with constructs containing as little as (CTG)57, a repeat size that is associated with the mildest forms of DM. In addition, stable C2C12 clones expressing lacZ with a (CTG)200 DMPK 3[prime]-UTR (lacZ+200-59) produced nearly three times less [beta]-galactosidase than clones (lacZ+200-59Rev) in which a spontaneous deletion event reduced the CTG repeat tract to (CTG)5. This provides strong evidence that the results seen in the transient transfections (i.e. the cis effect) are not due to an artifact such as differential cellular uptake of plasmids with different (CTG)n. Together, these results clearly show that a CTG expansion has a negative cis effect on protein production.
Interestingly, even the normal DMPK 3[prime]-UTR has a negative effect on translation. Protein production from mRNA transcripts with the normal DMPK 3[prime]-UTR was ~42% of that from transcripts expressing only the reporter gene (Fig. 1B). Concurrent with these findings, northern blot analysis showed a significant proportion of the mRNA from the reporter construct in lacZ+5-40 is found in the nuclear fraction (Fig. 5). A similar degree of nuclear retention (~50%) of the normal DMPK transcript in human cells has been reported previously (22). Moreover, deletion of the CTG repeats, from an otherwise intact DMPK 3[prime]-UTR, restored protein expression levels to those seen with the reporter gene alone (Fig. 1B). These results indicate that the (CTG)n tract is the cause of the negative cis effect seen with a normal DMPK 3[prime]-UTR.
Several possible explanations exist for how a CTG expansion could reduce protein expression. The CTG tract could affect transcription, RNA processing or translational efficiency. In our stable clones (lacZ+5-40, lacZ+200-59, lacZ+200-59Rev and lacZ-165), there are no major differences in the steady state levels of the reporter gene transcript (Fig. 5). Moreover, in the lacZ+200-59 and the lacZ+200-59Rev cells, the same promoter in an identical genetic context is expressing the reporter gene. Using RNA-FISH analysis we found that a DMPK 3[prime]-UTR with a (CTG)57 expansion or greater, or a (CTG)100 expansion only, produced nuclear RNA foci. Furthermore, northern blot analysis showed that the lacZ transcript in lacZ+200-59 is found predominantly in the nucleus (Fig. 5). Our analysis indicates that an expanded (CTG)n tract alone is sufficient to mediate a reduction of protein expression through nuclear retention of the mRNA and foci formation.
These data support the hypothesis that nuclear retention of mutant DMPK mRNA decreases DMPK protein production and are consistent with reports of decreased DMPK levels in DM patients (11,14). Using DMPK antibodies, we too find clear evidence for reduced levels of DMPK in cells from DM patients (unpublished data). Recently, Berul et al. reported that Dmpk+/- knockout mice develop cardiac conduction disturbances similar to those seen in 70% of DM patients, indicating that decreased DMPK levels (haploinsufficiency) may be responsible for the cardiac phenotype in DM patients (19).
Although we observed nuclear foci in our cell culture system, some of the CUG expansion transcripts are obviously translated in order for us to be able to detect the reporter protein by ELISA. This observation indicates that: (i) mRNA retention is not complete up to 200 CUG repeats; (ii) transcripts are escaping the nucleus during mitosis (22); or (iii) the RNA retention mechanism is saturable. The strong viral promoter (CMV) used in our constructs could possibly lead to saturation of the retention machinery and allow some mRNA to exit the nucleus. However, under endogenous DMPK expression levels, retention of CUG expansion transcripts could remain unsaturated leading to more complete RNA retention. In addition, as opposed to our cell culture model using dividing C2C12 myoblasts, the tissues primarily affected in DM (heart, skeletal muscle, neurons) consist of post-mitotic cells, and thus mutant DMPK mRNA would continue to accumulate in the nuclei, without the potential for release into the cytoplasm during cell division. Both of these differences could in fact lead to a greater effect on protein expression from the mutant allele in DM tissues, than the 3-fold decrease that we measured in our model.
Studies using DM patient fibroblasts have found that mutant DMPK transcripts with (CUG)>400 are completely retained in the nucleus, but retention is incomplete with a (CUG)150 expansion (22) and not detected at all at (CUG)80 (35). In agreement, we do not detect RNA foci in nuclei of fibroblasts with a (CTG)50-80 DM mutation (data not shown). In our model system, however, we observe nuclear retention of transcripts containing 57 CUG repeats, indicating that (CUG)57 meets the sequence requirements for retaining transcripts in the nucleus. Importantly, expression of a reporter construct with a DMPK 3[prime]-UTR (CTG)26 does not result in nuclear foci, suggesting the threshold for foci formation is between (CUG)26 and (CUG)57, which is in agreement with the threshold for developing a DM phenotype. The difference in foci detection between patient fibroblasts and our model system is likely due to differential expression levels. Foci have been observed to be brighter and more numerous in patient myoblasts (where DMPK expression is higher) than in fibroblasts (22).
The mechanism by which transcripts containing a CUG expansion accumulate into nuclear foci is presently not known. It is possible that the DM mutation causes aberrant interactions between the mutant DMPK mRNA and RNA-binding proteins resulting in disruption of RNA processing and nuclear retention of mutant transcripts. Alternatively, a CUG expansion could alter the structure of the transcript such that it is not recognized as a substrate for export through the nuclear pore complex (36). However, it is clear from our model system using mouse myoblasts, that the molecular mechanisms involved in nuclear RNA foci aggregation are conserved between mice and humans.
The mutant DMPK 3[prime]-UTR mRNA inhibits myoblast fusion
Our results provide clear evidence that the expression of the mutant DMPK 3[prime]-UTR mRNA has deleterious effects on myoblast differentiation. We found that myoblasts expressing a (CTG)5 DMPK 3[prime]-UTR mRNA undergo normal myogenesis in culture, whereas equivalent expression of a (CTG)200 DMPK 3[prime]-UTR mRNA disrupted myoblast fusion. Furthermore, the identification of a spontaneous genetic event in lacZ+200-59 cells that restored myoblast fusion potential provides compelling evidence that the fusion defect is mediated by the (CTG)n repeat expansion, the deleted 3[prime] region, or both sequences.
The myoblast clones derived from the spontaneous deletion event (lacZ+200-59Rev) express an mRNA that contains the entire 5[prime] region of the DMPK 3[prime]-UTR. This observation is in direct contrast to a study by Sabourin et al. that showed over-expression of a normal DMPK 3[prime]-UTR, and specifically the 239 bp region 5[prime] of the CTG repeats, is sufficient to inhibit myogenesis (29). They also found that at least 4-10-fold over-expression of the mRNA above endogenous levels was required for the fusion defect. The differences between our results and theirs could be due to different expression levels of the DMPK 3[prime]-UTR mRNA. In our study we compared C2C12 clones that expressed either normal or mutant DMPK 3[prime]-UTR mRNA at relatively equal levels and found that only the mutant transcript affected myoblast differentiation. Thus, our results associate the fusion defect with the DM mutation, rather than an increase in DMPK mRNA expression.
How do these results relate to DM? Muscle maturation in adult DM patients apparently proceeds normally prior to the onset of symptoms. However, it has been noted by histology that there is a paucity of muscle regeneration in response to the ongoing muscle loss and dystrophy in DM-affected muscle tissue (1). This is in sharp contrast to the marked regenerative responses seen in other muscular dystrophies such as Duchenne's muscular dystrophy. It is possible that the blunted repair response may be a manifestation of impaired myogenesis in adult DM. In order for muscle regeneration to occur, satellite cells are normally recruited in a coordinated process that results in the synthesis of new myotubes and their incorporation into the existing muscle structure (37,38). The expression of the mutant DMPK 3[prime]-UTR mRNA and the resulting nuclear RNA foci in DM muscle tissue could have a detrimental impact on this process. The evidence for impaired myoblast fusion is far more compelling in CDM. Reports on skeletal muscle biopsies from CDM patients have repeatedly noted a marked immaturity of myofibers, persistence of small myotubes and an increased number of satellite cells, reflecting a defect in muscle maturation (2-5). The myoblast fusion defect we observe in our C2C12 model system is highly reminiscent of these clinical findings.
It is evident from our data that the mutant DMPK 3[prime]-UTR mRNA acts as a riboregulator. Several groups have shown in various model systems, including plants (39) and mammals (40-42), that certain untranslated mRNA molecules function as regulators (riboregulators) of growth and differentiation. In particular, riboregulators have been shown to affect muscle differentiation. The 3[prime]-UTR of skeletal muscle [alpha]-tropomyosin, a muscle-specific gene encoding a structural protein, has been reported to inhibit cell division and promote differentiation of differentiation-defective mutant myoblasts (43), suppress tumor formation (40) and induce transdifferentiation of embryonic fibroblasts into muscle cells (44). In contrast, the mutant DMPK 3[prime]-UTR mRNA inhibits myoblast differentiation.
The mutant DMPK 3[prime]-UTR mRNA could be exerting its trans effects through interactions with RNA-binding proteins such as CUG-BP or other proteins yet to be identified. The presence of the DM mutation could cause aberrant (increased, decreased or novel) interactions resulting in varied effects including altered stability, transport or processing of affected transcripts. Hoffman and colleagues (20) reported that the presence of the mutant DMPK transcript resulted in decreased levels of the wild-type DMPK mRNA in cells from DM patients. They subsequently reported a similar effect on the insulin receptor mRNA in DM muscle tissue (24). However, other studies found no evidence for an effect on the normal DMPK transcript in DM cells (22,35). Cooper et al. showed altered splicing of cardiac troponin T mRNAs in DM cells and that over-expression of an mRNA containing a large CUG tract, or CUG-BP could duplicate this effect (27). In cell culture, Sabourin et al. (29) reported decreased levels of myogenin mRNA in myoblasts over-expressing DMPK 3[prime]-UTR mRNA while another group (31) reported no change in myogenin transcripts. We are currently taking a systematic approach to identifying mRNAs affected in trans using candidate gene analysis and more global approaches to studying transcriptional differences including microchip arrays, by using the clones derived from our cell culture model.
In conclusion, we have identified both cis and trans effects of the mutant DMPK 3[prime]-UTR mRNA that may contribute to the pathogenesis of DM. This cell culture model provides support for the hypothesis that the DM mutation causes haploinsufficiency of DMPK through nuclear transcript retention and foci formation. Furthermore, it is clear that the mutant DMPK 3[prime]-UTR mRNA is functioning as a riboregulator to disrupt normal myoblast differentiation. The delayed myoblast fusion is strikingly similar to findings in congenital DM, and provides evidence to support an RNA mediated pathogenesis mechanism for some important aspects of DM. The generation and characterization of transgenic mice expressing the mutant DMPK 3[prime]-UTR mRNA outside the context of the endogenous DMPK transcript, and the identification of other genes affected in trans by the mutant DMPK 3[prime]-UTR mRNA, will be invaluable in evaluating this hypothesis.
MATERIALS AND METHODS
Reporter constructs
DMPK 3[prime]-UTR fragments containing (CTG)5 or (CTG)57 were PCR amplified using primer 033 (5[prime]-GCTTGGTACCTGAACC- CTAGAACTGTCT-3[prime]) or 063 (5[prime]-CCGAATTCTGAACCCT- AGAACTGTCT-3[prime]) and primer 057 (5[prime]-GAGCTCGAGGGGC- AGATGGAGGGC-3[prime]) from DM patient genomic DNA, and cloned as a KpnI-XhoI fragment into the pcDNA3/CAT vector (Invitrogen, Carlsbad, CA) and an EcoRI-XhoI fragment into pcDNA3.1/His/lacZ (Invitrogen). Reporter constructs containing (CTG)>57 were generated by in vivo expansion events in Escherichia coli DH5[alpha] (45).
To generate DMPK 3[prime]-UTR deletion constructs, the 225 bp 5[prime] of the CTG repeats were PCR amplified with primers 033 and 055 (5[prime]-AGCTGCAGGATCCCCGGCTAC-3[prime]) and cloned as a KpnI-BamHI fragment into pcDNA3/CAT, and the 484 bp 3[prime] of the repeats were amplified with primers 056 (5[prime]-CT- GCAGCTGGGGGGATCACAGA-3[prime]) and 057 and inserted as a BamHI-XhoI fragment. The 5[prime] and 3[prime] regions were also ligated together and cloned as a KpnI-XhoI fragment to generate a DMPK 3[prime]-UTR construct without CTG repeats [[Delta](CTG)]. A (CTG)100-expansion-only reporter construct was created by cloning an EagI (37 bp 5[prime] of the repeats) to DraII (40 bp 3[prime] of the repeats) fragment from a DMPK 3[prime]-UTR containing 100 CTGs into pcDNA3/CAT.
The CMV promoter of the reporter constructs used to establish stable C2C12 clones was flanked by loxP sequences. PCR amplification of pcDNA3.1/His/lacZ with primers 085 (5[prime]-GGGCAATTGATAACTTCGTATAGCATACATTATACGAAGTTATCATGAAGAATCTGCTT-3[prime]) and 089 (5[prime]-GGGAAGCTTTCAATAACTTCGTATAGCATACATTATACGAAGTTATAAGTTTAAACGCT-3[prime]) (loxP sites are underlined) generated a CMV promoter region with flanking loxP sites. This PCR fragment, digested with MfeI-HindIII, was used to replace the existing CMV promoter in pcDNA3.1/His/lacZ. All constructs were confirmed by automated DNA sequencing.
Cell culture
C2C12 myoblasts were maintained at subconfluency in Dulbecco's modified Eagle's medium (DMEM; Cellgro, Herndon, VA) supplemented with 10% cosmic calf serum (HyClone, Logan, UT). DM fibroblasts from patient GMO3755 with a (CTG)500 expansion (obtained from the NIGMS human genetic mutant cell repository) and non-DM IMR90 fibroblasts were grown in minimal essential media (Gibco BRL Life Technologies, Gaithersburg, MD) supplemented with 15% fetal bovine serum (HyClone) and 2 mM L-glutamine.
Transfection and analysis
C2C12 were transiently transfected in 60 mm dishes with 5 µg of target DNA (and 0.5 µg of a control plasmid to monitor transfection efficiency if necessary) using Lipofectamine PLUS reagent (Gibco BRL, Life Technologies) as recommended by the manufacturer. Seventy-two hours after transfection, cells were harvested for ELISAs or fixed for immunocytochemistry and RNA-FISH.
For ELISA, cytoplasmic proteins were extracted from cells collected in 40 mM Tris (pH 7.5), 1 mM EDTA, 150 mM NaCl by three quick freeze-thaw cycles. Protein concentrations were determined with a Bio-Rad protein assay (Bio-Rad, Hercules, CA) and CAT and [beta]-galactosidase protein levels were measured using ELISA kits (Boehringer Mannheim, Indianapolis, IN). The amount of CAT produced by a reporter construct was normalized to the amount of [beta]-galactosidase produced by a co-transfected lacZ plasmid to control for transfection efficiency. For each reporter construct, the measurements were repeated with at least six different transfections. Statistical significance was determined using a one-sided Wilcoxon Ranked Sum test.
To establish stable C2C12 clones, myoblasts were co-transfected (as above) with 2 µg of linear target DNA and 0.2 µg of a linear plasmid conferring either neomycin or hygromycin B resistance. Forty-eight hours after transfection, cells were transferred to media containing 0.8 mg/ml G418 (Cellgro) or 0.3 mg/ml hygromycin B (Sigma, St Louis, MO) to select for resistant clones. Ninety-six antibiotic resistant colonies were isolated and numbered accordingly for each construct, and screened for LacZ activity by X-gal staining. Cells were fixed for 2 min in PBS containing 0.625% glutaraldehyde (Sigma) and then incubated in 5 mM ferri/ferro cyanide, 0.4 mg/ml X-gal, 1 mM MgCl2 at 37°C for 3 h. A PCR assay using primers that flank the CMV promoter, 090 (5[prime]-CTCAGTACAATCTGCTCTGATG-3[prime]) and 088 (5[prime]-CGTACAGATCCCGACCCATTT-3[prime]), was used to screen for Cre-mediated CMV promoter deletions.
Immunocytochemistry and RNA-FISH
Cells grown on glass coverslips were washed with PBS, fixed in 4% paraformaldehyde for 15 min and stored in 70% ethanol at 4°C. After re-hydration in PBS for 10 min at room temperature, cells were incubated with primary antibody in PBS + 1% BSA for 1 h at 37°C. Following three washes in PBS for 5 min each, the cells were incubated with 7 pg/ml anti-mouse-FITC (Jackson Laboratories, West Grove, PA) at 37°C for 1 h. For RNA-FISH, cells were placed in 40% formamide, 2× SSC for 10 min and then incubated with a CY3 conjugated (CAG)10 oligonucleotide probe (Operon, Alameda, CA) in hybridization buffer as described by Taneja et al. (21). Cells were then washed and counterstained with 4,6-diamidino-2-phenylindole (DAPI) (Sigma) and mounted on slides with phenyldiamine in 90% glycerol.
Samples were visualized using an Olympus microscope with epifluorescence and images were captured with a cooled charge-coupled device camera (Princeton Instruments, Trenton, NJ) using IP Lab Spectrum software (Scanalytics, Fairfax, VA). Images from three wavelengths (CY3, DAPI, and FITC) were assembled and colored using Canvas software (Deneba Systems, Miami, FL).
Myoblast differentiation
To differentiate C2C12 myoblasts into myotubes, cells were grown to 70-90% confluency and then cultured in DMEM + 2% equine serum (HyClone) (differentiation medium) for 6 days. To quantify myoblast fusion of C2C12 clones, cells were fixed after 6 days of differentiation, stained with DAPI, and analyzed for myosin heavy chain expression by immunocytochemistry with MY32 antibody (Sigma). The numbers of myotubes (myosin heavy chain positive cells with >3 nuclei) in a 40× field were counted in 30 random fields. The average number of myotubes per field presented in Figure 4E represents data pooled from at least three experiments.
Southern blotting
DNA (10 µg) was digested and separated through 0.8% agarose gel and transferred to a nylon membrane (Hybond-N; Amersham Pharmacia Biotech, Little Chalfont, UK). DNA probes were labeled with [[alpha]-32P]dCTP using a Rediprime II random prime labeling kit (Amersham Pharmacia Biotech). Pre-hybridization and hybridization was performed in 50% formamide, 5× SSPE, 10× Denhardt's solution, 2% SDS, 0.1 mg/ml sheared salmon sperm DNA at 42°C. Membranes were washed in 2× SSC, 0.1% SDS at room temperature for 15 min, 0.1× SSC, 0.1% SDS at 55°C for 20 min and 0.1× SSC, 0.1% SDS at 65°C for 20 min. Membranes were then exposed to phosphor screens which were scanned on a STORM 860 phosphorimager (Molecular Dynamics, Sunnyvale, CA) and analyzed using ImageQuant software (Molecular Dynamics).
RNA analysis
RNA extracted from myoblasts was separated into nuclear and cytoplasmic fractions and purified through a 5.7 M CsCl density gradient essentially as described by Davis et al. (22). Cells were lysed in 10 mM Tris-HCl (pH 8.0), 0.14 M NaCl, 1.5 mM MgCl2 (lysis buffer), then passed through a 25 G needle 25 times and centrifuged at 2500 g for 15 min at 4°C to pellet nuclei. The supernatant (cytoplasmic fraction 1) was transferred to thiocyanate buffer [4 M guanidinium thiocyanate, 20 mM sodium acetate (pH 5.4), 0.5% sarkosyl, 0.1 mM DTT]. The pellet was resuspended in lysis buffer with 0.5% Tween-20 and 0.5% sodium deoxycholate, incubated on ice for 5 min and centrifuged at 2500 g for 15 min at 4°C. The nuclear pellet was subsequently lysed in thiocyanate buffer. The supernatant (cytoplasmic fraction 2) was pooled with cytoplasmic fraction 1. RNA from both fractions was pelleted by ultracentrifugation for 15 h through a 5.7 M CsCl cushion.
For northern blotting, 10 µg of nuclear RNA and the equivalent volume of cytoplasmic RNA were separated through 1% agarose, 2.2 M formaldehyde gel and transferred to nylon membrane (Hybond). Probe labeling, hybridization, washing conditions and analysis were performed as described above for Southern blotting.
RT-PCR analysis was performed using standard protocols. cDNA was synthesized in a 20 µl reaction utilizing 2 µg of total RNA (combined nuclear and cytoplasmic fractions). The RNA was denatured for 5 min at 95°C and cooled on ice. Reverse transcription was carried out at 37°C for 1 h after addition of 4 µl of 5× reverse transcription buffer (Gibco BRL), 2 µl of 0.1 M DTT, 4 µl of 2.5 mM dNTPs, 8 U RNasin (Promega, Madison, WI), 400 U MMLV reverse transcriptase (Gibco BRL) and 25 ng of a modified oligo(dT) primer: 5[prime]-GGGGATCC(T)18-3[prime]. From this, 1-2 µl cDNA reaction was used as a template for PCR (50 µl reactions) using the oligo(dT) primer and primer 113 (5[prime]-TCGCTACCATTACCAGTTGGT-3[prime]) at the 3[prime] end of lacZ. Resulting PCR products were gel purified and sequenced using an ABI 377 automated DNA sequencer.
ACKNOWLEDGEMENTS
We thank members of the Petrini laboratory and G. Tiscornia for helpful discussions. This work was supported in part by the Basil O'Connor Starter Scholar program research grant no. FY96-1193 from the March of Dimes Birth Defects Foundation, by the Muscular Dystrophy Association and by a grant to the University of Wisconsin Medical School under the Howard Hughes Medical Institute Faculty Development Program (M.S.M). J.D.A. was supported by NIH predoctoral training grant 5T32GM07133. This paper is no. 3541 from the University of Wisconsin-Madison Laboratory of Genetics.
REFERENCES
+To whom correspondence should be addressed. Tel: +1 608 265 8634; Fax: +1 608 262 2976; Email: msmahade{at}facstaff.wisc.edu
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Zhang, J. E. Lee, J. Wilusz, and C. J. Wilusz The RNA-binding Protein CUGBP1 Regulates Stability of Tumor Necrosis Factor mRNA in Muscle Cells: IMPLICATIONS FOR MYOTONIC DYSTROPHY J. Biol. Chem., August 15, 2008; 283(33): 22457 - 22463. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Osborne and C. A. Thornton RNA-dominant diseases Hum. Mol. Genet., October 15, 2006; 15(suppl_2): R162 - R169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Houseley, Z. Wang, G. J. R. Brock, J. Soloway, R. Artero, M. Perez-Alonso, K. M. C. O'Dell, and D. G. Monckton Myotonic dystrophy associated expanded CUG repeat muscleblind positive ribonuclear foci are not toxic to Drosophila Hum. Mol. Genet., March 15, 2005; 14(6): 873 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. A. Timchenko, R. Patel, P. Iakova, Z.-J. Cai, L. Quan, and L. T. Timchenko Overexpression of CUG Triplet Repeat-binding Protein, CUGBP1, in Mice Inhibits Myogenesis J. Biol. Chem., March 26, 2004; 279(13): 13129 - 13139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Storbeck, S. Drmanic, K. Daniel, J. D. Waring, F. R. Jirik, D. J. Parry, N. Ahmed, L. A. Sabourin, J.-E Ikeda, and R. G. Korneluk Inhibition of myogenesis in transgenic mice expressing the human DMPK 3'-UTR Hum. Mol. Genet., March 15, 2004; 13(6): 589 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kino, D. Mori, Y. Oma, Y. Takeshita, N. Sasagawa, and S. Ishiura Muscleblind protein, MBNL1/EXP, binds specifically to CHHG repeats Hum. Mol. Genet., March 1, 2004; 13(5): 495 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. Schulz, A. D. McIntosh, M. R. Kasten, B. Wieringa, and H. F. Epstein A Role for Myotonic Dystrophy Protein Kinase in Synaptic Plasticity J Neurophysiol, March 1, 2003; 89(3): 1177 - 1186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Amack, S. R. Reagan, and M. S. Mahadevan Mutant DMPK 3'-UTR transcripts disrupt C2C12 myogenic differentiation by compromising MyoD J. Cell Biol., November 7, 2002; 159(3): 419 - 429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Seznec, O. Agbulut, N. Sergeant, C. Savouret, A. Ghestem, N. Tabti, J.-C. Willer, L. Ourth, C. Duros, E. Brisson, et al. Mice transgenic for the human myotonic dystrophy region with expanded CTG repeats display muscular and brain abnormalities Hum. Mol. Genet., November 1, 2001; 10(23): 2717 - 2726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Timchenko, P. Iakova, Z.-J. Cai, J. R. Smith, and L. T. Timchenko Molecular Basis for Impaired Muscle Differentiation in Myotonic Dystrophy Mol. Cell. Biol., October 15, 2001; 21(20): 6927 - 6938. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Amack and M. S. Mahadevan The myotonic dystrophy expanded CUG repeat tract is necessary but not sufficient to disrupt C2C12 myoblast differentiation Hum. Mol. Genet., September 1, 2001; 10(18): 1879 - 1887. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Furling, L. Coiffier, V. Mouly, J.P. Barbet, J. Lacau St Guily, K. Taneja, G. Gourdon, C. Junien, and G.S. Butler-Browne Defective satellite cells in congenital myotonic dystrophy Hum. Mol. Genet., September 1, 2001; 10(19): 2079 - 2087. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mankodi, C. R. Urbinati, Q.-P. Yuan, R. T Moxley, V. Sansone, M. Krym, D. Henderson, M. Schalling, M. S. Swanson, and C. A. Thornton Muscleblind localizes to nuclear foci of aberrant RNA in myotonic dystrophy types 1 and 2 Hum. Mol. Genet., September 1, 2001; 10(19): 2165 - 2170. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Khajavi, A. M. Tari, N. B. Patel, K. Tsuji, D. R. Siwak, M. L. Meistrich, N. H. A. Terry, and T. Ashizawa 'Mitotic drive' of expanded CTG repeats in myotonic dystrophy type 1 (DM1) Hum. Mol. Genet., April 1, 2001; 10(8): 855 - 863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mankodi, E. Logigian, L. Callahan, C. McClain, R. White, D. Henderson, M. Krym, and C. A. Thornton Myotonic Dystrophy in Transgenic Mice Expressing an Expanded CUG Repeat Science, September 8, 2000; 289(5485): 1769 - 1772. [Abstract] [Full Text]
|
|
|
|
|
|
C. J. Cummings and H. Y. Zoghbi Fourteen and counting: unraveling trinucleotide repeat diseases Hum. Mol. Genet., April 1, 2000; 9(6): 909 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Good, Q. Chen, S. J. Warner, and D. C. Herring A Family of Human RNA-binding Proteins Related to the Drosophila Bruno Translational Regulator J. Biol. Chem., September 8, 2000; 275(37): 28583 - 28592. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||