Myotonic Dystrophy 1

A number sign (#) is used with this entry because myotonic dystrophy-1 (DM1) is caused by a heterozygous trinucleotide repeat expansion (CTG)n in the 3-prime untranslated region of the dystrophia myotonica protein kinase gene (DMPK; 605377) on chromosome 19q13.

A repeat length exceeding 50 CTG repeats is pathogenic (Musova et al., 2009).

Description

Myotonic dystrophy is an autosomal dominant disorder characterized mainly by myotonia, muscular dystrophy, cataracts, hypogonadism, frontal balding, and ECG changes. The genetic defect in DM1 results from an amplified trinucleotide repeat in the 3-prime untranslated region of a protein kinase gene. Disease severity varies with the number of repeats: normal individuals have 5 to 37 repeats, mildly affected persons have 50 to 150 repeats, patients with classic DM have 100 to 1,000 repeats, and those with congenital onset can have more than 2,000 repeats. The disorder shows genetic anticipation, with expansion of the repeat number dependent on the sex of the transmitting parent. Alleles of 40 to 80 repeats are usually expanded when transmitted by males, whereas only alleles longer than 80 repeats tend to expand in maternal transmissions. Repeat contraction events occur 4.2 to 6.4% of the time (Musova et al., 2009).

Genetic Heterogeneity of Myotonic Dystrophy

See also myotonic dystrophy-2 (DM2; 602668), which is caused by mutation in the ZNF9 gene (116955) on chromosome 3q21.

Clinical Features

ADULT-ONSET MYOTONIC DYSTROPHY

In adult-onset DM1, symptoms typically become evident in middle life, but signs can be detectable in the second decade. Bundey et al. (1970) found that the most useful method for identifying subclinical cases is slit-lamp examination for lens changes, followed by electromyography for myotonic discharges, and then by measurement of immunoglobulins.

Harper (1989) provided a monograph on myotonic dystrophy that has been updated regularly.

Unlike the other muscular dystrophies, DM initially involves the distal muscles of the extremities and only later affects the proximal musculature. In addition, there is early involvement of the muscles of the head and neck. Involvement of the extraocular muscles produces ptosis, weakness of eyelid closure, and limitation of extraocular movements. Atrophy of masseters, sternocleidomastoids, and the temporalis muscle produces a characteristic haggard appearance. Bosma and Brodie (1969) demonstrated both myotonia and weakness in patients with swallowing and speech disability. Myotonia, delayed muscular relaxation following contraction, is most frequently apparent in the tongue, forearm, and hand. Myotonia is rarely as severe as in myotonia congenita and tends to be less apparent as weakness progresses.

Many of the muscle biopsy changes are nonspecific. Most commonly there are central nuclei and ring fibers. Necrosis, regeneration, and increase of collagen are never as severe as in Duchenne muscular dystrophy. In 70% of patients there is hypotrophy of type I muscle fibers; less commonly there are markedly atrophic fibers (Casanova and Jerusalem, 1979). In many cases there are target fibers, suggesting neurogenic dysfunction, but intramuscular nerves appear histologically normal (Drachman and Fambrough, 1976). Ultrastructural studies show dilatation of T tubules or sarcoplasmic reticulum, whose contents may be unusually dense (Milhaud et al., 1964). In some cases the surface membrane may be irregular, with reduplication of basal lamina.

Neurologic Features

From a series of neurophysiologic investigations of 24 patients with myotonic dystrophy, Jamal et al. (1986) concluded that there was unequivocal evidence of widespread nervous system dysfunction. In many patients there was significant involvement of peripheral large diameter motor and sensory fibers and of small diameter sensory fibers peripherally and/or centrally. The authors stated that 'the concept of myotonic dystrophy as a pure myopathy can no longer be sustained.' This conclusion is supported by the findings in the family reported by Spaans et al. (1986). Thirteen members of a large family presented with a hereditary motor and sensory neuropathy in a dominant pedigree pattern. The mean motor conduction velocities for the median and peroneal nerves in the affected individuals were 62% and 56%, respectively, of those of the unaffected relatives. Eight of the 13 affected members also showed more or less prominent signs of myotonic dystrophy. There was no case of myotonic dystrophy alone.

Turnpenny et al. (1994) found that IQ in myotonic dystrophy declined as the age of onset of signs and symptoms decreased and as the size of the CTG expansion increased. The correlation appeared to be more linear with age of onset. Censori et al. (1994) carried out a prospective case-control study of 25 patients with myotonic dystrophy using magnetic resonance imaging (MRI) of the brain. They found that 84% of myotonic dystrophy patients showed white matter hyperintense lesions, compared with 16% of controls. Most of these lesions involved all cerebral lobes without hemispheric prevalence, but 28% of the myotonic dystrophy patients also showed particular white matter hyperintense lesions at their temporal poles. Myotonic patients also showed significantly more cortical atrophy than did controls. However, there was no relationship between atrophy or white matter hyperintense lesions and age, disease duration, or neuropsychologic impairment. Damian et al. (1994) found that amplification of the CTG repeat in leukocytes strongly correlated with cognitive test deficits when the expansion length exceeded over 1,000 trinucleotides. MRI lesions were associated with impaired psychometric performance, but the MRI findings of subcortical white matter lesions correlated only very weakly with the molecular findings.

Miaux et al. (1997) found that 9 (70%) of 13 patients with a mild form of adult myotonic dystrophy had T2-weighted signal abnormalities on brain MRI. Four patients (30%) had lesions greater than 1 cm in diameter. Lesions were symmetric, occurred in the subcortical white matter, and showed a predilection for the temporal lobe. There was some evidence of cerebral atrophy in the patients overall but no difference in IQ between patients and controls. There was no correlation between number of pathologic CTG repeats and white matter lesions, and there was no correlation between intellectual impairment and white matter lesions, except in 1 patient who had a difficult birth and temporal lobe epilepsy. Three patients had marked thickening of the skull, which was associated with ossification of the falx in 2.

Donahue et al. (2009) reported a 56-year-old woman with a 10-year history of myotonic dystrophy who presented with progressive lower extremity weakness. Brain MRI showed multiple discrete and confluent areas of abnormal signal intensity throughout the subcortical white matter with predominant involvement of the frontal and anterior temporal lobes. There was also diffuse thickening of the skull with ossification of the falx. Donahue et al. (2009) noted the similarity of the white matter findings with those observed in CADASIL (125310), but noted that skull abnormalities are not seen in CADASIL.

In a study of 21 patients with myotonic dystrophy, Akiguchi et al. (1999) found that MRI results indicated progressive brain atrophy. Magnetic resonance spectroscopy demonstrated a significant reduction of the neuronal marker N-acetylaspartate, even in young patients in whom imaging studies were still equivocal.

Delaporte (1998) found that 15 DM patients with no or minimal muscle weakness demonstrated a homogeneous personality profile characterized by avoidant, obsessive-compulsive, passive-aggressive, and schizotypic traits. Fourteen healthy control individuals and 12 patients with a mild form of muscle disease did not show the same trait homogeneity. Delaporte (1998) concluded that the personality disorders were not attributable to the adjustment to a disabling condition, but rather were primary manifestations of the genetic mutation.

Modoni et al. (2004) performed detailed neuropsychologic testing of 70 patients with DM1, including 10 with congenital onset and 60 with juvenile-adult onset, who were subdivided into 4 genotypic subgroups according to number of repeat expansion. Patients with congenital onset (CTG repeats greater than 1,000) obtained the lowest scores in verbal attainment, frontal and executive functions, and general intelligence, consistent with mental retardation. Patients with 50 to 150 repeats showed age-dependent impairment in memory, frontal lobe, and temporal lobe function. Patients with 151 to 1,000 repeats showed defects only in frontal and executive tasks. Although there was a correlation between number of repeats and degree of muscle involvement for all patients, there was not a significant correlation between number of repeats and cognitive impairment, except for the congenital group.

Sergeant et al. (2001) stated that neurofibrillary tangles (NFT), as described in patients with Alzheimer disease (AD; 104300), had been described in the neocortex and subcortical regions of patients with DM1. NFTs derive from pathologic aggregation of hyperphosphorylated tau (MAPT; 157140) proteins. By neuropathologic examination, Sergeant et al. (2001) identified hippocampal NFTs in 4 of 5 patients with DM1 ranging in age from 42 to 64 years. Three patients had clinical evidence of cognitive impairment or mental retardation. In some of the patients, other brain regions also had NFTs. Biochemical characterization showed overexpression of tau protein isoforms lacking exons 2 and 3, suggesting that the DMPK mutation disrupts normal MAPT isoform expression and alters the maturation of MAPT pre-mRNA. Maurage et al. (2005) identified biochemically similar NFTs in multiple brain regions of a patient with DM2; however, the patient with DM2 was mentally normal, demonstrated no cognitive decline, and died at age 71 years from a bilateral renal thrombosis.

Cardiac Features

Hawley et al. (1983) suggested that the tendency to have heart block or arrhythmia with myotonic dystrophy is a familial characteristic. The implication was that there may be 2 forms of myotonic dystrophy. They studied 18 families and found heart block in 4.

In a single large kindred, Tokgozoglu et al. (1995) compared the cardiac findings in 25 patients with myotonic dystrophy with age-matched normal family members. They found that the patients were more likely to have conduction abnormality (52% vs 9%), mitral valve prolapse (32% vs 9%), and wall motion abnormality (25% vs 0%). Left ventricular ejection fractions and stroke volume were reduced compared with normals. Using multivariate analysis, the number of CTG repeats (range, 69 to 1367; normal, less than 38) was the strongest predictor of abnormalities in wall motion and EKG conduction. Patients with more extensive neurologic findings had a higher incidence of wall motion and/or EKG conduction abnormalities. The authors also found that the relation of mitral valve prolapse to the size of the CTG repeat was of borderline significance.

Cardiac involvement is well described in adults with myotonic dystrophy. Bu'Lock et al. (1999) undertook detailed cardiac assessment in 12 children and young adults with congenital myotonic dystrophy using control data from 137 healthy children and young adults. All patients were in sinus rhythm with a normal P wave axis. Three had first-degree heart block and 4 had a borderline P-R interval (200 ms). Four others had more complex conduction abnormalities. Three patients had mitral valve prolapse. Eleven of the 12 patients had abnormalities of 1 or more parameter of left ventricular diastolic filling. None of these patients were symptomatic. The authors commented that the prognostic implications of these findings were unclear; however, they concluded that echocardiographic assessment of left ventricular diastolic function may be a useful adjunct to electrocardiographic monitoring of patients with congenital myotonic dystrophy.

Antonini et al. (2000) performed a prospective study of 50 DM1 patients without known cardiac disease at the time of enrollment. Nineteen patients developed major cardiac abnormalities during the 56-month study. No correlation was found between CTG length and frequency of EKG abnormality or type of arrhythmia. CTG length was inversely correlated with age at onset of EKG abnormality.

Bassez et al. (2004) reported 11 DM1 patients under the age of 18 years who had severe cardiac involvement. Two patients died suddenly, 1 patient had cardiac arrest with successful resuscitation, and 1 asymptomatic 13-year-old girl presented with recurrent presyncope. Rhythm disturbances included atrial flutter in 4, ventricular tachycardia in 4, and atrial fibrillation in 1. Five patients had atrioventricular block necessitating pacemaker implantation. Six of 11 patients (55%) experienced arrhythmic events with vigorous exercise. Genetic analysis detected between 235 and 1,200 CTG repeats in all patients. No cardiac involvement was detected before age 10 years. Bassez et al. (2004) concluded that patients with congenital or childhood forms of DM1 may present with cardiac abnormalities and that exercise testing is a necessary evaluation in these patients.

Groh et al. (2008) found that 96 of 406 patients with genetically confirmed DM1 had severe ECG abnormalities, and that these patients were older, had more CTG repeats, and had more severe muscular impairment compared to those without ECG abnormalities. After a mean follow-up period of 5.7 years, 69 patients who did not have ECG abnormalities at the start of the study had developed ECG abnormalities and 81 patients died. There were 27 sudden deaths, 32 deaths from progressive neuromuscular respiratory failure, 5 nonsudden deaths from cardiac causes, and 17 deaths from other causes. The major cause of death in the cohort was respiratory failure associated with progressive muscular weakness. A severe ECG abnormality and a clinical diagnosis of atrial tachyarrhythmia conferred relative risks for sudden death of 3.30 and 5.18, respectively.

CONGENITAL MYOTONIC DYSTROPHY

Harper (1975) observed that in a small proportion of cases, myotonic dystrophy may be congenital with neonatal hypotonia, motor and mental retardation, and facial diplegia. With rare exception, it is the mother who transmits the disease. Diagnosis can be difficult if the family history is not known because muscle wasting may not be apparent, and cataracts and clinical myotonia are absent, although the latter is sometimes detectable by electromyography. Fried et al. (1975) observed that infants with neonatal myotonic dystrophy (almost always the mother is affected) have thin ribs. Talipes at birth, together with hydramnios and reduced fetal movements during pregnancy, is frequent. Respiratory difficulties are frequent and are often fatal. Those that survive the neonatal period initially follow a static course, eventually learning to walk but with significant mental retardation in 60 to 70% of cases. By age 10 they develop myotonia and in adulthood develop the additional complications described for the adult-onset disease. Roig et al. (1994) reported long-term follow-up of 18 patients diagnosed with congenital myotonic dystrophy. Three of the 18 had died, and 5 were lost to follow-up. The remaining 10 had IQs of less than 65. Universal findings were language delay, hypotonia, and delayed motor development. There was no difficulty with routine immunizations nor were there anesthetic complications observed in any of the 7 patients who underwent surgery.

Rudnik-Schoneborn et al. (1998) reviewed the obstetric histories of 26 women with myotonic dystrophy who had a total of 67 gestations, comparing gestations with affected and unaffected fetuses. Of the 56 infants carried to term, 29 had or most likely had inherited the gene for DM from their affected mothers; 18 of the 29 (61%) were affected by the congenital form of DM. Perinatal loss rate was 11% and associated with congenital DM. Preterm labor was a major problem in gestations with DM-fetuses (55 vs 20%), as was polyhydramnios (21% vs none). While forceps deliveries or vacuum extractions were required in 21% of deliveries with DM-fetuses and only 5% of unaffected fetuses, the frequency of cesarean sections were similar in the 2 groups. Obstetric problems were inversely correlated with age at onset of maternal DM, while no effect of age at delivery or birth order on gestational outcome was seen.

Stratton and Patterson (1993) established the molecular diagnosis of myotonic dystrophy in a fetus shown to have bilateral effusions and scalp and upper torso edema by ultrasound examination at 30 weeks' gestation. Polyhydramnios was also present. Thus, nonimmune hydrops fetalis is a manifestation of congenital myotonic dystrophy. The mother had previously unsuspected myotonic dystrophy, but she did show grasp myotonia. Her brother had a confirmed diagnosis. The DM gene showed marked expansion in her fetus. Stratton and Patterson (1993) found reports of 15 other cases of nonimmune hydrops fetalis associated with congenital myotonic dystrophy. (Robin et al. (1994) described nonimmune hydrops fetalis in association with severely impaired fetal movement, giving support to the notion that fetal hypomobility is a cause of this disorder. The hydropic infant stopped moving 8 weeks before delivery and did not move postnatally. Autopsy revealed extensive CNS destruction of unknown cause.)

Other Features

Diabetes mellitus occurs in 5% of cases, frequently with hypersecretion of insulin (Barbosa et al., 1974). There is impaired responsiveness to follicle stimulating hormone with hypogonadism (Sagel et al., 1975), often impairment of adrenal androgens, and occasional thyroid dysfunction, but pituitary function is usually intact (Lee and Hughes, 1964). Di Chiro and Caughey (1960) reviewed radiographic findings in the skull in 18 cases. In 17, 'hyperostotic' changes in the vault were found, the sex distribution being equal. In 8 cases with hypogonadism, the hyperostosis was most advanced.

Excessive catabolism of IgG contributes to low circulating levels of IgG (Wochner et al., 1966).

Schwindt et al. (1969) claimed that 25 to 50% of patients have abdominal symptoms due to cholelithiasis. Brunner et al. (1992) described 4 DM patients with recurrent intestinal pseudoobstruction. In 1 patient it preceded significant muscle weakness by 15 years. Conservative measures usually were effective. Improved intestinal function was noted in 1 patient treated with the prokinetic agent cisapride. A partial sigmoid resection was performed in 3 patients with dolichomegacolon. Two of the patients were sibs. Brunner et al. (1992) pointed out that there are many reports of familial occurrence of specific complications of DM: cardiac conduction disturbances, focal myocarditis, mitral valve prolapse, pilomatrixomas, polyneuropathy, normal pressure hydrocephalus, and dilatation of the urinary tract. Familial idiopathic intestinal pseudoobstruction occurs as an intestinal myopathy (155310) or in a neuronal form (243180); it occurs also in Duchenne muscular dystrophy (310200).

Ciafaloni et al. (2008) found that 17 of 38 patients with DM1 reported excessive daytime sleepiness. Thirteen of these 17 patients underwent sleep studies, and 7 of them showed reduced sleep latency, sleep-onset REM, or both. However, CSF levels of hypocretin (HCRT; 602358), which is implicated in the pathogenesis of narcolepsy (161400), were normal in all 38 DM1 patients.

Biochemical Features

In the cytoplasm of cultured skin fibroblasts Swift and Finegold (1969) found an abnormally large amount of material with the staining properties of acid mucopolysaccharides. Because of the similarity of platelet actomyosin ('thrombosthenin') to that of muscle, Bousser et al. (1975) studied platelets in myotonic dystrophy. Although they found a normal pattern of aggregation in response to adenosine diphosphate and collagen, aggregation occurred with exceedingly low levels of adrenalin. A growing body of evidence was interpreted as indicating a generalized defect of cell membranes in myotonic dystrophy (Butterfield et al., 1974; Roses et al., 1975).

Using antisera developed against synthetic DM-PK peptide antigens for biochemical and histochemical studies, van der Ven et al. (1993) found lower levels of immunoreactive DM-kinase protein of 53 kD in skeletal and cardiac muscle extracts of DM patients than in normal controls. Immunohistochemical staining revealed that DM-PK is localized predominantly at sites of neuromuscular and myotendinous junctions of human and rodent skeletal muscles. The protein could also be demonstrated in the neuromuscular junctions of muscular tissues of adult and congenital cases of DM, with no gross changes in structural organization.

By quantitative RT-PCR and by radioimmunoassay using antisera developed against both synthetic peptides and purified myotonin-protein kinase (Mt-PK) protein expressed in E. coli, Fu et al. (1993) demonstrated that decreased levels of the mRNA and protein expression are associated with the adult form of myotonic dystrophy. From this they suggested that the autosomal dominant nature of the disease is due to an Mt-PK dosage deficiency and that means of elevating Mt-PK level or activity should be explored for therapeutic intervention in adult patients.

Inheritance

This disorder segregates as an autosomal dominant with greatly variable penetrance. Many obligatory gene carriers are asymptomatic. With only rare exception, it is the mother who transmits the disease in cases of congenital myotonic dystrophy. Patients born of affected mothers are more severely affected than those born of affected fathers (Harper and Dyken, 1972). In Japan, Tanaka et al. (1981) also noted the maternal effect in age of onset and severity, and thought that a chemical factor, deoxycholic acid, is responsible for the effect.

Ott et al. (1990) described DNA marker-based genetic counseling in a family with an affected mother and 3 children, each by a different partner. Two of the children were affected. In the third child, myotonic dystrophy could be excluded in the presymptomatic period. In genetic counseling, the recommended risk estimate that any heterozygous woman with myotonic dystrophy will have a congenitally affected child is 3 to 9%. However, after having such an offspring, a DM mother's risk increases to 20 to 37% (Koch et al., 1991). Koch et al. (1991) concluded that the clinical status of the mother at the time of pregnancy and delivery had an important influence on the outcome in the infant. Only women with multisystem effects of the disorder had a congenitally affected child. No heterozygous woman with polychromatic lens changes as the only finding had a congenitally affected child. For classically affected women with systemic manifestations, risk figures that approach the occurrence risk given to mothers with previously born congenitally affected children seemed appropriate. The findings of this study supported the earlier proposal that maternal metabolites acting on a heterozygous offspring account for the congenital involvement. Neither genomic imprinting nor mitochondrial inheritance could explain the correlation between the clinical status of heterozygous mothers and that of their children.

Contrary to the findings and conclusions of Koch et al. (1991), Goodship et al. (1992) described a family in which a 53-year-old woman had no symptoms of myotonic dystrophy, a normal electromyogram, and only dot polychromatic lens opacities on slit-lamp examination. She had, however, given birth 30 years before to a child with congenital myotonic dystrophy. Furthermore, she had a son and daughter with adult onset of symptoms of myotonic dystrophy and another daughter who after normal developmental milestones had early adult onset of symptoms and who gave birth to an offspring with congenital myotonic dystrophy.

Ives et al. (1989) described possible homozygosity for the DM gene. The possible homozygotes were more severely affected than heterozygotes. For a variety of reasons the authors had found it difficult to obtain molecular proof of homozygosity. On the other hand, Cobo et al. (1993) studied a consanguineous French Canadian family in which 2 sisters were homozygous for the 'at risk' haplotype but were asymptomatic and showed no evidence of DM on extensive clinical examination. Both sisters possessed 2 alleles with repeat sizes normally seen in minimally affected patients. Both parents were affected. Martorell et al. (1996) described 3 unrelated homozygous myotonic dystrophy patients. One patient had the classic form of myotonic dystrophy and the other 2 were mildly affected. A remarkable feature was the mildness of the phenotype in the homozygous patients; one, for example, had late-onset cataract as the only manifestation. With the observations of Cobo et al. (1993), this led Zlotogora (1997) to conclude that in myotonic dystrophy, homozygotes do not differ from heterozygotes and that, like Huntington disease (HD; 143100), DM is a 'true dominant.'

Zuhlke et al. (2007) reported 2 additional unrelated cases of homozygous myotonic dystrophy, both products of incestuous unions. Both patients had a severe, congenital phenotype and expanded alleles (330/770 repeats in one patient and 200/1,200 repeats in the other).

On the possibility that mitochondrial genetic modifying factors might be responsible for DM, Thyagarajan et al. (1991) completely sequenced the mitochondrial genome in 2 patients with congenital DM. Comparison of the 2 sequences with control data failed to reveal any specific nucleotide or length variant. After isolation of the gene mutant in myotonic dystrophy and identification of its gene product as a serine-threonine kinase, Jansen et al. (1993) tested for evidence of imprinting of either the paternal or the maternal alleles in both human and mouse tissues. No evidence of imprinting was found involving the expression of the DM kinase gene.

Jansen et al. (1994) used the term gonosomal mosaicism to refer to combined somatic and germline mosaicism which they demonstrated in DM. Studies of variation in the (CTG)n repeat in sperm and body cells of the same individual were demonstrated. The rather frequent observation of offspring with triplet repeat length larger than that found in sperm suggested that intergenerational length changes in the unstable (CTG)n repeat occur during early embryonic mitotic divisions. The initial size of the (CTG)n repeat, the overall number of cell divisions involved in tissue formation, and a specific selection process in spermatogenesis may all influence variation in repeat size.

Carey et al. (1994) examined meiotic drive and segregation distortion at the DM locus. The study was undertaken because the haplotype analysis of DM chromosomes had detected a very limited pool of founder chromosomes (Harley et al., 1992; Mahadevan et al., 1992), raising the question of how a disease that usually decreases reproductive fitness within a few generations has been maintained in the population over hundreds of generations. Carey et al. (1994) found that healthy individuals heterozygous for DM alleles in the normal size range preferentially passed on alleles of more than 19 CTG repeats to their offspring. They suggested that this phenomenon may act to replenish a reservoir of potential DM mutations and that this distortion of the transmission ratio may offer an example of meiotic drive in humans. This segregation distortion may act as a mechanism to maintain alleles in the population that lie at the larger end of the normal range in the trinucleotide repeat disorders. It was unclear whether the segregation distortion was a direct consequence of the CTG repeat number or whether the preferential transmission of the larger allele was due to linkage to segregation distorting loci on the same chromosome.

Martorell et al. (2001) studied the frequency and germline stability of DMPK (605377) alleles in an effort to understand the constant population incidence of the disease despite its low reproductive fitness. The authors analyzed the DMPK CTG repeat length in more than 3,500 individuals from 700 Spanish families. A trimodal distribution of CTG repeat lengths in the normal population was observed: 5 repeats, 9-18 repeats, and 19-37 repeats. Five-repeat alleles and 9- to 18-repeat alleles were stably inherited. The third mode, 19-37 repeats, was skewed toward increasing allele length with frequent de novo expansions. The authors also analyzed alleles with repeat lengths of 38-54 repeats, or 'premutation' alleles. Individuals with premutation alleles were asymptomatic. Premutation alleles were found to be very unstable and liable to frequent large expansions in the male germline, with expansion observed in 25 of 25 transmissions. Sperm from a premutation carrier demonstrated a range of diverse alleles positively skewed toward expansion. Martorell et al. (2001) concluded that the incidence of DM1 is likely maintained in the population by expansion of alleles within the normal range to the premutation range and subsequently into the disease-manifesting range in successive generations.

Leeflang et al. (1996) directly analyzed meiotic segregation and the question of meiotic drive at the DM locus using single-sperm typing. They studied samples of single sperm from 3 individuals heterozygous at the DM locus, each with one allele larger and one allele smaller than the 19 CTG repeats. To guard against the possible problem of differential PCR amplification rates based on the lengths of the alleles, the sperm were also typed at another closely linked marker whose allele size was unrelated to the allele size of the DM locus: D19S207 in 2 donors and D19S112 in the third. Using statistical models specifically designed to study single-sperm segregation data, they found no evidence of meiotic segregation distortion. This suggested to Leeflang et al. (1996) that any greater amount of segregation distortion at the DM locus must result from events following sperm ejaculation.

Magee and Hughes (1998) studied 44 sibships with myotonic dystrophy. When the transmitting parent was male, 58.3% of the offspring were affected, and when the transmitting parent was female, 68.7% were affected. Overall, the DM expansion was transmitted in 63% of cases. Magee and Hughes (1998) concluded that DM expansion tends to be transmitted preferentially.

Nakagawa et al. (1994) described 2 sisters with congenital myotonic dystrophy born to a normal mother and an affected father. The sisters had symptoms from birth. The age of onset of DM in the father was 39 years. Analysis of the CTG trinucleotide expansion in this family showed increase in the repeat length with increasing severity, with the smallest expansion in the grandfather and the largest expansion in the younger of the 2 affected sisters. The observation refutes the hypothesis that congenital DM is exclusively of maternal origin.

Bergoffen et al. (1994) observed inheritance from a mildly affected father. This family illustrated that the congenital form can occur without intrauterine or other maternal factors operating. Nakagawa et al. (1993) also reported a case of congenital myotonic dystrophy inherited from the father. De Die-Smulders et al. (1997) reported a further case of congenital myotonic dystrophy inherited from the father. The patient was a 23-year-old, mentally retarded male suffering from severe muscular weakness who presented with respiratory and feeding difficulties at birth. His 2 sibs suffered from childhood-onset DM, whereas their father had adult onset of DM at around 30 years of age. De Die-Smulders et al. (1997) reviewed 6 other cases of paternal transmission of congenital DM and found that the fathers of these children showed, on average, shorter CTG repeats and hence less severe clinical symptoms than the mothers of children with congenital DM. The authors concluded that paternal transmission of congenital DM preferentially occurs with onset of DM past 30 years of age in the father.

Zunz et al. (2004) examined whether myotonic dystrophy exhibits the phenomenon of preferential transmission of the larger mutated alleles that had been described in other trinucleotide repeat disorders. They cited several reports (e.g., Carey et al., 1994; Leeflang et al., 1996; Magee and Hughes, 1998) indicating that the frequency of transmission of the mutated alleles is higher than 50%, a finding contrary to mendelian laws of segregation. However, these studies were based on data from the analysis of pedigrees with ascertainment bias. Zunz et al. (2004) determined the frequency of transmission of mutated alleles using data from prenatal molecular studies, which were not subject to ascertainment bias. Eighty-three fetuses were examined. Thirty of 62 mothers (48.38%) and 8 of 21 fathers (38.09%) transmitted the mutated allele, giving an overall transmission rate of 45.78%. Zunz et al. (2004) found no evidence of statistically significant deviation of the frequency of transmission of the mutated alleles from the 50% expected in autosomal dominant disorders. Unlike previous studies, the study of Zunz et al. (2004) excluded preferential transmission in myotonic dystrophy, a finding they concluded might be attributable to the lack of correction for ascertainment bias in previous studies and to the use of prenatal data in their study.

Zeesman et al. (2002) reported a child with congenital DM and 1,800 CTG repeats born to an asymptomatic father with 65 repeats and compared the case to 4 previously reported cases. They noted that polyhydramnios was present in most cases and that all fathers whose status was known had small repeat sizes and/or were asymptomatic at the time of their child's birth.

In a study of mitochondrial DNA from 35 patients with congenital myotonic dystrophy, Poulton et al. (1995) could find no evidence that mutations in mtDNA are involved in the pathogenesis of congenital myotonic dystrophy. Associated mitochondrial mutations might help account for the maternal inheritance pattern and the early onset of the congenital form.

Mapping

The linkage of secretor (Se; 182100) and myotonic dystrophy was suspected by Mohr (1954) when he was doing the studies that demonstrated the first autosomal linkage in humans, that between secretor and Lutheran blood group (Lu; 111200). Mohr (1954) failed to establish fully the DM linkage because of the relative insensitivity of the sib-pair method of linkage analysis he was using (Smith, 1986). Renwick et al. (1971) confirmed the linkage. The Lu-Se-DM linkage group and the Km (Inv)-Jk-Co linkage group were tentatively tied together by a family with myotonic dystrophy reported by Larsen et al. (1979, 1980). From study of a single large kindred, Larsen et al. (1979) suggested that Km and Jk are linked to myotonic dystrophy. An order of Km, Jk, Lu, Se, and DM was suggested. No recombination in 7 informative meioses occurred between Km and Jk, none in 5 between Se and DM, 3 out of 10 between Jk and Se, and 3 in 12 between Jk and DM.

Eiberg et al. (1981, 1983) concluded that C3 (120700), Le (111100), myotonic dystrophy, secretor, and Lutheran are linked. Since fibroblast C3 had been assigned to chromosome 19, the finding indicated that myotonic dystrophy is on chromosome 19, providing serum C3 (polymorphism of which was used in the above linkage studies) is under the same genetic control (or at least syntenic genetic control) as fibroblast C3.

Cook (1981) had found positive lod scores for serum C3 and peptidase D (613230), a chromosome 19 locus. Linkage of peptidase D to myotonic dystrophy (O'Brien et al., 1983) proved the assignment of the Lutheran-secretor linkage group to chromosome 19 and provided regional assignment. Using an RFLP related to a C3 probe, Davies et al. (1983) found evidence of linkage with myotonic dystrophy. Laberge et al. (1985) found a lod score of 4.574 at a recombination fraction of 0.12 for linkage of DM and APOE (107741) in French Canadians (males and females combined). Meredith et al. (1985) found close linkage (maximum lod = 7.8 at 4% recombination) of DM to APOC2 (608083). APOE and APOC2 are known to be closely linked.

Brook et al. (1985) concluded that the DM locus is probably in the 19p13.2-19cen segment. Friedrich et al. (1987) quoted studies of somatic cell hybrids carrying various fragments of chromosome 19 that provide unambiguous proof for location of the PEPD gene on 19q, thus corroborating the assignment of DM to that region. The hereditary motor and sensory neuropathy in the family described by Jamal et al. (1986) showed segregation with genetic markers known to be linked to myotonic dystrophy on chromosome 19. Spaans et al. (1986) raised the question of whether the disorder might be caused by an allele of the 'common' DM gene or alternatively by 2 closely linked genes on chromosome 19.

Shaw et al. (1986) reviewed gene mapping of chromosome 19 with particular reference to myotonic dystrophy. Suppression of recombination near the centromere and the large male-female differences in recombination are 'complications' of linkage mapping of the DM locus and use of linkage markers in genetic counseling. Shaw et al. (1986) concluded from linkage studies that myotonic dystrophy is located in the region of the centromere of chromosome 19.

Roses et al. (1986) described RFLPs at the D19S19 locus, which is linked to DM (maximum lod = 11.04 at theta = 0.0). Bartlett et al. (1987) reported that the genomic clone called LDR152 (D19S19) is tightly linked to DM; the maximum lod score was 15.4 at a recombination fraction = 0.0 (95% confidence limits 0.0-0.03). Using 2 RFLPs of the APOC2 gene, Pericak-Vance et al. (1986) demonstrated tight linkage to myotonic dystrophy; the maximum lod score was 16.29 at a recombination fraction of 0.02.

In 3 large kindreds, Friedrich et al. (1987) did linkage studies using RFLPs related to the C3 gene and the chromosome 19 centromeric heteromorphism as genetic markers. Three-point linkage analysis excluded DM from the 19cen-C3 segment and strongly supported its assignment to the proximal long arm of chromosome 19.

Harper (1986) demonstrated 2 to 5% recombination between myotonic dystrophy and APOC2, leading him to the conclusion that myotonic dystrophy may be just onto 19q or very close to the centromere on 19p. Bird et al. (1987) concluded that the APOC2 gene is very closely linked to the DM locus and proposed that APOC2 markers may be used for prenatal diagnosis of myotonic dystrophy because the loci are closely linked. Smeets et al. (1988) used synthetic oligonucleotides to discriminate between E3 and E4 alleles of APOE. The relevant segment of the APOE gene was enzymatically amplified and linkage with DM tested. A maximum lod score of 7.47 at a recombination frequency of 0.047 was found (male theta = female theta). No recombination (maximum lod score = 5.61 at theta = 0.0) was found between APOE and APOC2. Further analysis of the relationship of the human APOC2 gene to myotonic dystrophy was provided by MacKenzie et al. (1989), who reported a linkage study utilizing 6 RFLPs in 50 families with myotonic dystrophy. They observed significant linkage disequilibrium between the DM locus and APOC2 alleles. The maximum lod score was 17.869 at a theta of 0.04.

Bender et al. (1989) found no evidence of linkage with any of 35 serologic and biochemical markers. Brunner et al. (1989) concluded that the DM and CKMM loci are distal to the APOC2-APOE gene cluster; the orientation of DM and muscle-type creatine kinase (CKMM; 123310) was undetermined.

Johnson et al. (1989) presented evidence that DM is distal to the apolipoprotein cluster. Yamaoka et al. (1990) found a maximum lod score of 28.41 at theta = 0.01 for the linkage between CKMM and DM. They concluded, furthermore, that CKMM is on the same side and closer to DM than APOC2. Walsh et al. (1990) found a peak lod score of 9.29 at 2 cM for linkage of DM to APOC1 (107710) and a lod score of 8.55 at 4 cM for linkage of DM to CYP2A (122720). A maximum lod score of 9.09 at theta = 0.05 was observed for the linkage of APOC1 to CYP2A. CYP2A appeared to be proximal to DM, CKMM, and APOC2.

Smeets et al. (1989), Davies et al. (1989), Roses et al. (1989), Brunner et al. (1989), Harley et al. (1989), Brook et al. (1989), and Miki et al. (1989) presented linkage data for markers surrounding the myotonic dystrophy locus on human chromosome 19. Smeets et al. (1989) and Davies et al. (1989) also presented physical maps of the region derived from pulsed field gel electrophoresis analysis.

In a study of 65 myotonic dystrophy families from Canada and the Netherlands, Brunner et al. (1989) obtained a maximum lod score of 22.8 at a recombination frequency of 0.03 for linkage to CKMM. MacKenzie et al. (1990) ruled out a defect of the RYR1 gene (180901) as the cause of myotonic dystrophy; the 2 loci showed an interval of about 10 cM (maximum lod = 4.8). The order of loci was found to be 19cen--RYR1--APOC2--CKMM--DM--qter.

Bailly et al. (1991) excluded mutation of the CKMM gene as the cause of this disorder. CKMM cDNA was isolated from the skeletal muscle of an individual with DM. Sequencing of the CKMM cDNA from the chromosome 19 carrying the DM gene showed 2 novel polymorphisms but no translationally significant mutation.

Harley et al. (1991) concluded that the DM gene lies in region 19q13.2-q13.3 and that the closest proximal markers are APOC2 and CKM, approximately 3 cM and 2 cM from DM, respectively, in the order cen--APOC2--CKMM--DM. Ten of 12 polymorphic markers on 19q were shown to be proximal to the DM gene; the 2 that were distal to DM, PRKCG (176980) and D19S22, were approximately 25 cM and 15 cM, respectively, removed from DM.

Brunner et al. (1991) restudied the family reported by Spaans et al. (1986), ruled out linkage to chromosome 17 markers, thus excluding the gene (601097) associated with Charcot-Marie-Tooth disease, type Ia (118220), and demonstrated linkage to DNA markers from the APOC2 locus on chromosome 19. All affected individuals had inherited a unique APOC2 haplotype that was not found in their clinically and electrophysiologically normal sibs. In this family, a moderately severe neuropathy appeared to be the only clinical sign of myotonic dystrophy for many years. The results were consistent with either an unusual neuropathic mutation in the DM gene or involvement of 2 closely linked genes.

Linkage studies by Cobo et al. (1992) established the D19S63 marker as useful for prenatal and presymptomatic diagnosis and, as the closest marker to DM, in isolating the gene.

Molecular Genetics

Identification of an Expanded Triplet Repeat

Harley et al. (1992) isolated a human genomic clone that detected novel restriction fragments specific to persons with myotonic dystrophy. A 2-allele EcoRI polymorphism was seen in normal persons, but in most affected individuals one of the normal alleles was replaced by a larger fragment, which varied in length both between unrelated affected individuals and within families. The unstable nature of this region was thought to explain the characteristic variation in severity and age at onset of the disease.

From a region of chromosome 19 flanked by 2 tightly linked markers, ERCC1 (126380) proximally and D19S51 distally, Buxton et al. (1992) isolated an expressed sequence that detected a DNA fragment that was larger in affected persons than in normal sibs or unaffected controls.

Aslanidis et al. (1992) cloned the essential region between the above mentioned markers in a 700-kb contig formed by overlapping cosmids and yeast artificial chromosomes (YACs). The central part of the contig bridged an area of about 350 kb between 2 flanking crossover borders. This segment, which presumably contained the DM gene, was extensively characterized. Two genomic probes and 2 homologous cDNA probes were situated within approximately 10 kb of genomic DNA and detected an unstable genomic segment in myotonic dystrophy patients. The length variation in this segment showed similarities to the instability seen in the fragile X locus (300624). The authors proposed that the length variation was compatible with a direct role in the pathogenesis of myotonic dystrophy.

Using positional cloning strategies, Brook et al. (1992) identified a CTG triplet repeat that is larger in myotonic dystrophy patients than in unaffected individuals. This sequence is highly variable in the normal population. Unaffected individuals have between 5 and 27 copies. Myotonic dystrophy patients who are minimally affected have at least 50 repeats, while more severely affected patients have expansion of the repeat-containing segment up to several kilobase pairs.

Tsilfidis et al. (1992) found a correlation between the length of the CTG trinucleotide repeat and the occurrence of severe congenital myotonic dystrophy. Furthermore, mothers of congenital DM individuals had higher than average CTG repeat lengths.

Shelbourne et al. (1993) described a probe that allowed direct identification of the myotonic dystrophy mutation in 108 of 112 unrelated patients. In 3 families for whom the clinical and genetic data obtained with linked probes were ambiguous, the specific probe identified persons at risk and demonstrated that a possible sporadic case of myotonic dystrophy was, in fact, familial. In 1 family, the size of the unstable myotonic dystrophy-specific fragment decreased on transmission to offspring who remained asymptomatic, which was an example of the reverse of anticipation.

Thornton et al. (1994) reported the clinical findings, muscle pathology, and genetic data on 3 individuals from 2 families with myotonic dystrophy in whom no trinucleotide repeat expansion was detected. The diagnosis of DM was based on involvement of the lens, cardiac conduction system, skin, and testes, in association with muscle weakness and myotonia. The diagnosis was supported by an autosomal dominant pedigree pattern and by features of muscle histopathology consistent with DM. This may be a situation like that of the fragile X syndrome in which rare affected individuals lack a trinucleotide repeat expansion and instead have deletions or point mutations.

Martorell et al. (1995) determined the CTG repeat length in 23 DM patients with varying clinical severity and various sizes of repeat amplification. They confirmed the findings of previous studies that there was no strong correlation between repeat length and clinical symptoms but found that the repeat length in peripheral blood cells of patients increased over a 5-year period, indicating continuing mitotic instability of the repeat throughout life. The degree of expansion correlated with the initial repeat size, and 50% of the patients with continuing expansion showed clinical progression of their disease symptoms over the 5-year study period.

Junghans et al. (2001) hypothesized that the diversity of phenotype in myotonic dystrophy may be due to the fact that the DM CTG repeat induces long-range cis chromosomal effects that suppress diverse genes on chromosome 19, resulting in manifest multisystem abnormalities in the clinical disorder. One of the features discussed in detail was hypercatabolism of immunoglobulin G in myotonic dystrophy and the possible significance of the FCGRT gene (601437) to the DM locus.

Using triplet-primed PCR (TP-PCR) of both DNA strands followed by direct sequencing, Musova et al. (2009) identified interruptions within expanded DM1 CTG repeats in almost 5% (3 of 63) of Czech DM1 families and in 2 of 2 intermediate alleles. None of 261 normal Czech alleles tested carried interruptions. The expanded alleles contained either regular runs of a (CCGCTG)n hexamer or showed a much higher complexity; they were always located at the 3-prime end of the repeat. The number and location of the interruptions were very unstable within families and subject to substantial change during transmission. However, 4 of 5 transmissions of the interrupted expanded allele in 1 family were accompanied by repeat contraction, suggesting that the interruptions render the DMPK CTG repeat more stable or could even predispose it to contractions. Overall, the contribution of the interrupted alleles to the phenotype was uncertain. Musova et al. (2009) suggested that the occurrence of interruptions may be missed by routine testing using PCR or Southern blotting.

Anticipation

Buxton et al. (1992) found that the size of the fragment varied between affected sibs and also increased through generations in parallel with increasing severity of the disease. They reported a family in which persons in the first 2 generations had mild symptoms and a CTG repeat unit of approximately 60 repeats, whereas persons in the third and fourth generations had severe symptoms and a dramatic expansion in allele size--a demonstration of the physical basis of anticipation in myotonic dystrophy. Mahadevan et al. (1992) found an expansion of the CTG repeat region in the 3-prime untranslated region of the DM candidate gene in 253 of 258 (98%) persons with DM. They likewise observed that an increase in the severity of the disease in successive generations was accompanied by an increase in the number of trinucleotide repeats. Thus, 'anticipation' (progressively earlier onset and greater severity of symptoms), long a puzzling feature of DM, has an explanation and physical documentation in the progressive 'worsening' of the mutation. Buxton et al. (1992) postulated that this represented an unstable DNA sequence responsible for DM.

Tsilfidis et al. (1992) also examined the amount of intergenerational amplification in DM mother/offspring pairs. The average increase in the pairs with congenital DM was not statistically greater than that shown by noncongenital DM pairs. It was noteworthy, however, that whereas 9 of 42 cases (21%) showed no intergenerational amplification between mother and noncongenital offspring, all mother/congenital offspring pairs showed intergenerational amplification. In another analysis, they found that the intergenerational CTG repeat length increase was the same whether the father or the mother contributed the DM allele to the offspring.

Fu et al. (1992) reported that in the case of severe congenital DM, the paternal triplet repeat allele was inherited unaltered, while the maternal, DM-associated allele was unstable. They suggested that the mutational mechanism leading to DM is triplet repeat amplification, similar to that occurring in the fragile X syndrome. The genomic repeat is p(AGC)n. Richards and Sutherland (1992), therefore, referred to the trinucleotide repeat as p(AGC)n/p(CTG)n. They pointed out that this is the same repeat sequence found in the androgen receptor gene (313700) and amplified in Kennedy disease (313200), although transcription in the latter disorder is from the opposite strand of DNA. Richards and Sutherland (1992) indicated that the instability of the DM element extends beyond meiotic instability in affected pedigrees to mitotic instability, manifest as somatic variation--a smear of bands evident in some affected persons. Progression of somatic CTG repeat length heterogeneity in the blood cells of myotonic dystrophy patients was documented by Martorell et al. (1998). They studied repeat length changes over time intervals of 1 to 7 years in 111 myotonic dystrophy patients with varying clinical severity and CTG repeat sizes. There was a correlation between the progression of size heterogeneity over time and the initial CTG repeat size.

The expansion of a CTG trinucleotide repeat, which represents the myotonic dystrophy mutation, is in complete linkage disequilibrium in both Caucasian (Harley et al., 1991) and Japanese (Yamagata et al., 1992) patients with a 2-allele insertion/deletion polymorphism located 5 kb upstream from the repeat, suggesting a single origin of the mutation. This finding was unexpected for a dominant disease that in its severe form diminishes or abolishes reproductive fitness. Such diseases are usually characterized by a high level of new mutations that compensate for the loss of abnormal alleles due to the decreased fitness. It was therefore suggested that DM could be due to recurrent mutations occurring on the background of a predisposing allelic form of the normal gene. Imbert et al. (1993) studied the association of CTG repeat alleles in a normal population to alleles of the insertion/deletion polymorphism and of a (CA)n repeat marker 90 kb from the DM mutation. The results strongly suggested that the initial predisposing event(s) consisted of a transition from a (CTG)-5 allele to an allele with 19 to 30 repeats. The heterogeneous class of (CTG)-19-30 alleles, which was found to have an overall frequency of about 10%, may constitute a reservoir for recurrent DM mutations.

Krahe et al. (1995) reported results in a Nigerian (Yoruba) DM family, the only indigenous sub-Saharan DM case reported to that time, that caused them to reassess the hypotheses that (1) the predisposition for (CTG)n instability resulted from a founder effect that occurred only once or a few times in human evolution; and (2) elements within the disease haplotype may predispose the (CTG)n repeat to instability. (A single haplotype composed of 9 alleles within and flanking the DM locus over a physical distance of 30 kb had been shown to be in complete linkage disequilibrium with DM.) All affected members of the Nigerian family had an expanded (CTG)n repeat in one allele of the DM gene. However, unlike all other DM populations studied to that time, disassociation of the (CTG)n repeat expansion from other alleles of the putative predisposing haplotype was found. Krahe et al. (1995) concluded that in this family, the expanded (CTG)n repeat was the result of an independent mutational event. This weakens the hypothesis that a single ancestral haplotype predisposes to repeat expansion.

Yamagata et al. (1996) studied linkage disequilibrium between CTG repeats and an Alu insertion/deletion polymorphism in the DMPK gene (605377) in 102 Japanese families, of which 93 were affected with DM. All of the affected chromosomes were in complete linkage disequilibrium with the Alu insertion allele. A strikingly similar pattern of linkage disequilibrium observed in European populations suggested a common origin of the DM mutation in the Japanese and European populations. The authors speculated that this mutation arose in a common Eurasian ancestor after the first separation of the African and the non-African populations, in light of the fact that the family reported by Krahe et al. (1995) did not show linkage disequilibrium with the Alu insertion/deletion polymorphism. Presumably, the mutation in that family represented a less-ancient event than the Eurasian mutation, accounting for the fact that DM is extremely rare in African populations.

Harley et al. (1993) demonstrated in 439 individuals affected with myotonic dystrophy from 101 kindreds that the size of the unstable CTG repeat detected in nearly all cases was related both to age at onset of the disorder and to the severity of the phenotype. The largest repeat sizes, 1.5 to 6.0 kb, were seen in patients with congenital myotonic dystrophy, while the minimally affected patients had repeat sizes of less than 0.5 kb. Only 4 of 182 parent-child pairs showed a definite decrease in repeat size in the offspring; almost all showed that the offspring had an earlier age of onset and a larger repeat size than their parents. Increase in repeat size from parent to child was similar for both paternal and maternal transmissions when the increase was expressed as a proportion of the parental repeat size. Analysis of congenitally affected cases showed not only that they had on the average the largest repeat sizes, but also that their mothers had larger mean repeat sizes, supporting previous suggestions that a maternal effect is involved.

Brunner et al. (1993) examined the kinetics of triplet expansion by analyzing repeat length in offspring of 38 carriers with small mutations (less than 100 CTG trinucleotides). Repeat lengths greater than 100 were more common in offspring of male transmitters than in offspring of female transmitters. They suggested that selection against sperm with extreme amplifications may be required to explain maternal inheritance of congenital myotonic dystrophy.

Sutherland and Richards (1992) editorialized on the legitimization of anticipation. According to Harper et al. (1992), 'The history of the scientific study of anticipation is...to a remarkable degree, the history of myotonic dystrophy.' In the second decade of this century, several observers noticed that ancestors of myotonic dystrophy patients had cataracts but no muscular symptoms themselves.

Brunner et al. (1993) and others observed the opposite of anticipation, namely, reverse mutation. They observed 2 families in which an affected father transmitted a normal allele to an offspring; in each case, an expanded CTG trinucleotide repeat decreased in size to the normal range. This was the first report of spontaneous correction of a deleterious mutation upon transmission to unaffected offspring in humans. Abeliovich et al. (1993) likewise observed what they referred to as 'negative expansion': a family in which the affected father had a 3.0-kb expansion of the DM unstable region, and a fetus inherited the mutated gene but with an expansion of only 0.5 kb. See review by Brook (1993). Ashizawa et al. (1994), who referred to the phenomenon as contraction rather than negative expansion, showed that it occurred in 6.4% of 1,489 DM offspring. Approximately one-half of these cases showed clinical anticipation despite the reduced CTG repeat size in the offspring. The most striking examples were 2 cases in which anticipation resulted in congenital DM in the offspring with contractions of the CTG repeat. They did not observe a single case in which the age at onset of DM in the symptomatic offspring was later than the age at onset in the parent, although Harley et al. (1993) reported 3 such cases.

Lavedan et al. (1993) found differently sized repeats in various DM tissues from the same individual, which may explain why the size of the mutation observed in lymphocytes does not necessarily correlate with the severity and nature of symptoms. With CTG sequences of more than 0.5 kb, Lavedan et al. (1993) observed that intergenerational variation was greater through female meioses, whereas a tendency to compression was observed almost exclusively in male meioses. For CTG sequences under 0.5 kb, a positive correlation was observed between the size of the repeat and the intergenerational enlargement for both male and female meioses. Anvret et al. (1993) found in 8 patients with myotonic dystrophy that the length of the CTG repeat expansion was greater in DNA isolated from muscle than in DNA isolated from lymphocytes. Dubel et al. (1992) found heterogeneity in the size of amplification in affected identical twins.

A family with myotonic dystrophy described by de Jong (1955) was restudied by de Die-Smulders et al. (1994) from the point of view of the long-term effects of anticipation. They defined clinical anticipation as the cascade of mild, adult, childhood, or congenital disease in successive generations. Such clinical anticipation appeared to be a relentless process occurring in all affected branches of the 5-generation family studied. The transition from the mild to the adult type was associated with transmission through a male parent. Stable transmission of the asymptomatic/mild phenotype showed a female transmission bias. Gene loss in the patients in this family was complete, owing to infertility of the male patients with adult-onset disease and the fact that mentally retarded patients did not procreate. Of the 46 at-risk subjects in the 2 youngest generations, only 1 was found to have a full mutation. This is the only subject who may transmit the gene to the sixth generation. No protomutation carriers were found in the fourth and fifth generations. Therefore, it seemed highly probable that the DM gene would be eliminated from this pedigree within 1 generation.

Simmons et al. (1998) demonstrated relatively stable transmission of a (CTG)60 repeat allele through 3 generations of a large DM family; only 3 members, all offspring of male carriers, had expansions in the clinically significant range.

Barcelo et al. (1994) insisted that there must be a maternal 'additive' factor involved in congenital DM. Their findings suggested that while a high number of repeats seem to be a necessary condition for congenital DM, this alone is not sufficient to explain its exclusive maternal inheritance. This was most clearly reflected in the fact that in their study group, approximately one-quarter of DM cases inherited from affected fathers had repeat numbers equal to or greater than those found in the congenital DM cases with the lowest number of repeats (approximately 700 repeats).

Novelli et al. (1995) provided additional evidence that size of repeat was insufficient to explain the severity. Two affected mothers with similar numbers of repeats gave birth to offspring with discordant phenotypes. Childhood and congenital myotonic dystrophy affected the son and the daughter of one sister, with CTG triplet repeats in lymphocytes of 700 and 1,100, respectively. In contrast, the affected son of the other sister had onset mild myotonic dystrophy at age 14 years, despite having 1,400 CTG triplets detected in lymphocytes.

Hamshere et al. (1999) found that in patients with CTG expansions of greater than 1.2 kb, there was no significant correlation between the age of onset of symptoms and the size of their repeat. Regression analysis predicted that the absolute size of the CTG repeat may not be a good indicator of the expected age of onset of symptoms when the size of the repeat is 0.4 kb or greater.

Khajavi et al. (2001) investigated the mechanism of expansion bias by cloning single lymphoblastoid cells from DM1 patients and normal subjects. In all DM1 cell lines, the expanded CTG repeat alleles gradually shifted toward further expansion by 'step-wise' mutations. Of 29 cell lines, 8 yielded a rapidly proliferating mutant with a gain of large repeat size that became the major allele population, eventually replacing the progenitor allele population. By mixing cell lines with different repeat expansions, the authors found that cells with larger CTG repeat expansion had a growth advantage over those with smaller expansions in culture. This growth advantage was attributable to increased cell proliferation mediated by Erk1 (601795) and Erk2 (176948) activation, which is negatively regulated by p21(WAF1) (116899). The authors designated this phenomenon 'mitotic drive,' which they suggested is a novel mechanism that can explain the expansion bias of DM1 CTG repeat instability at the tissue level, on a basis independent of the DNA-based expansion models. Since the life spans of the DM1 cells were significantly shorter than normal cell lines, the authors hypothesized that DM1 cells drive themselves to extinction through a process related to increased proliferation.

Puymirat et al. (2009) reported 2 unrelated French families in which paternal transmission of an expanded CTG repeat resulting in contraction of the repeat in the offspring. In 1 family, 2 affected brothers with 500 and 630 repeats, respectively, transmitted the alleles to their 4 offspring, who had between 260 and 360 repeats. Three of the 4 young adult offspring were asymptomatic. In the second family, the transmitting father had 500 repeats and his 4 asymptomatic young adult children all had 250 repeats. The findings suggested that a paternal factor acts to prevent CTG repeat expansion in DM1.

Genotype/Phenotype Correlations

Arsenault et al. (2006) examined 102 patients with DM1 carrying small CTG repeat expansions in the DMPK gene. Most patients with 50 to 99 repeats were asymptomatic except for cataracts. Patients with 100 to 200 repeats were significantly more likely to have myotonia, weakness, excessive daytime sleepiness, and myotonic discharges on EMG.

Barbe et al. (2017) examined DMPK CAG repeat expansion length and CpG methylation status surrounding the repeat in peripheral blood samples from 59 patients with classic DM1 and 20 patients with congenital DM1 (CDM1), as well as 7 chorionic villous samples (CVS), 1 fetal skin sample, 1 sperm sample, and 4 human embryonic stem cell (hESC) lines carrying a DM1 mutation and the corresponding blood DNAs. There was a significant correlation between congenital DM1 and increased methylation both upstream and downstream of the repeat (19 of 20 samples showed this; p = 7.05 x 10(-12)). The repeat size in congenital DM1 ranged from 1,100 to 4,700. Most non-CDM1 individuals were devoid of methylation, although a few showed downstream methylation. Only 2 non-CDM1 individuals showed upstream methylation; both had maternally-derived childhood-onset. Among CVS and hESC lines, there was a correlation between maternal inheritance and increased methylation. In contrast, paternally-derived samples never showed upstream methylation. CTG tract length did not strictly correlate with CDM1 or methylation. Barbe et al. (2017) concluded that methylation patterns flanking the CTG repeat are stronger indicators of congenital DM1 than repeat size, and that DMPK methylation may account for the maternal bias for CDM1 transmission, larger maternal CTG expansions, age of onset, and clinical continuum.

Pathogenesis

CTG-Expansion Effects on Chromosome Structure

The mechanism by which the expanded trinucleotide repeat in the 3-prime untranslated region of the DMPK gene (605377) leads to the clinical features is unclear. The DM region of chromosome 19 is gene rich, and it is possible that the repeat expansion may lead to dysfunction of a number of transcription units in the vicinity, perhaps as a consequence of chromatin disruption. Boucher et al. (1995) searched for genes associated with a CpG island at the 3-prime end of DMPK. Sequencing of the region showed that the island extends over 3.5 kb and is interrupted by the (CTG)n repeat. Comparison of genomic sequences downstream (centromeric) of the repeat in human and mouse identified regions of significant homology. This led to the identification of the gene which Boucher et al. (1995) called 'DM locus-associated homeodomain protein' (DMAHP; 600963). They found that this protein is expressed in a number of human tissues, including skeletal muscle, heart, and brain.

Harris et al. (1996) reviewed the molecular genetics of DM. They noted that published results on the effect of the trinucleotide repeat in the 3-prime end of DMPK on the gene's transcription have been contradictory. There were reports that DMPK expression is increased at the transcriptional level and reports that transcription is decreased. They noted also that the complexity of clinical manifestations in myotonic dystrophy and the results of animal studies suggest that other genes may be involved in this disease. Harris et al. (1996) reviewed results of studies on mice in which DMPK had been homozygously deleted (Jansen et al., 1996), and studies in which a DMPK transgene had been introduced to produce overexpression (Reddy et al., 1996). Harris et al. (1996) concluded that the animal studies ruled out haploinsufficiency of DMPK or overexpression of DMPK as the only contributing factor in DM. Harris et al. (1996) postulated that other genes may be involved. They proposed that the gene encoding DM locus-associated homeodomain protein (DMAHP), which lies immediately downstream of the repeat, may play a role in DM.

Roberts et al. (1997) used material from a DM homozygote who had expansion of CTG repeats on both alleles to study pathogenetic mechanisms in myotonic dystrophy.

Otten and Tapscott (1995) demonstrated that a nuclease-hypersensitive site is positioned adjacent to the CTG repeat at the wildtype DM locus and that large expansions of the repeat eliminated the hypersensitive site, converting the region surrounding the repeats to a more condensed chromatin structure. As nuclease-hypersensitive sites often coincide with gene regulatory regions, the decreased accessibility of transcription factors to this region in the expanded allele might affect local gene expression. Therefore Klesert et al. (1997) sought to determine whether this hypersensitive site contained regulatory elements that would enhance transcription in fibroblasts or skeletal muscle cells, 2 cell types in which the site was known to be present. They found that the hypersensitive site contains an enhancer element that regulates transcription of the adjacent DMAHP homeobox gene. Analysis of DMAHP expression in the cells of DM patients with loss of the hypersensitive site revealed a 2- to 4-fold reduction in steady-state DMAHP transcript levels relative to wildtype controls. Thus the results demonstrated that CTG-repeat expansions can suppress local gene expression and implicate DMAHP in DM pathogenesis. Along the same line, Thornton et al. (1997) showed that DMAHP expression in myoblasts, muscle, and myocardium was reduced by the DM mutation in cis, and the magnitude of this effect depended on the extent of the CTG repeat expansion. These observations supported the hypothesis that DMAHP participates in the pathophysiology of DM.

Sarkar et al. (1998) described a bacterial system that recapitulates the striking bimodal pattern of CTG amplification. Incremental expansions predominated in CTG tracts smaller than Okazaki fragment size, while saltatory expansions increased in repeat tracts larger than or equal to Okazaki fragment size. CTG amplification required loss of SbcC, a protein that modulates cleavage of single-stranded DNA and degradation of duplex DNA from double-strand breaks. These results suggested to Sarkar et al. (1998) that noncanonical single strand-containing secondary structures in Okazaki fragments and/or double-strand breaks in repeat tracts are intermediates in CTG amplification.

Saveliev et al. (2003) demonstrated that the relatively short triplet repeat expansions found in myotonic dystrophy and Friedreich ataxia (see 229300) confer variegation of expression on a linked transgene in mice. Silencing was correlated with a decrease in promoter accessibility and was enhanced by the classic position effect variegation (PEV) modifier heterochromatin protein-1 (HP1; 604478). Notably, triplet repeat-associated variegation was not restricted to classic heterochromatic regions but occurred irrespective of chromosomal location. Because the phenomenon described shares important features with PEV, Saveliev et al. (2003) suggested that the mechanisms underlying heterochromatin-mediated silencing might have a role in gene regulation at many sites throughout the mammalian genome and may modulate the extent of gene silencing and hence severity in several triplet-repeat diseases.

Using methylation-sensitive restriction enzymes, Steinbach et al. (1998) characterized the methylation pattern on the 5-prime side of the CTG repeat in the DMPK gene of normal individuals and of patients with myotonic dystrophy who showed expansions of the repetitive sequence. The gene segment analyzed corresponded to the restriction fragment carrying exons 11 to 15. There was constitutive methylation in intron 12 at restriction sites that were localized 1,159 to 1,232-bp upstream of the CTG repeat, whereas most, if not all, of the other restriction sites in this region were unmethylated, in normal individuals and most of the patients. In a number of young and severely affected patients, however, complete methylation of these restriction sites was found in the mutated allele. In most of these patients, the onset of the disease was congenital. Preliminary in vivo footprinting data gave evidence for protein-DNA contact in normal genes at an Sp1 consensus binding site upstream of the CTG repeat and for a significant reduction of this interaction in cells with a hypermethylated DMPK gene. The findings suggested that hypermethylation may be another genetic factor causally related to earlier onset and more severe manifestations of myotonic dystrophy.

An expansion of a CTG repeat at the DM1 locus causes myotonic dystrophy by altering the expression of 2 adjacent genes, DMPK and SIX5 (600963) and through a toxic effect of the repeat-containing RNA. Filippova et al. (2001) identified 2 CTCF (604167) binding sites that flank the CTG repeat and form an insulator element between DMPK and SIX5. Methylation of these sites prevents binding of CTCF, indicating that the DM1 locus methylation in congenital DM would disrupt insulator function. Furthermore, CTCF binding sites were associated with CTG/CAG repeats at several other loci. Filippova et al. (2001) suggested a general role for CTG/CAG repeats as components of insulator elements at multiple sites in the human genome.

In contrast to the findings of Steinbach et al. (1998), Spits et al. (2010) found no correlation between increased methylation of CpG sites upstream of the CTG repeat and CTG expansion size or disease severity in samples from 22 DM1 patients with expansions ranging from 180 to 2,800 repeats. The authors studied 8 CpG sites, including the previously studied SacII, HpaII, and HhaI endonuclease sites. The HhaI and HpaII sites were found to be constitutively unmethylated in all samples, including wildtype, whereas the SacII site showed differential methylation, but it did not correlate with expanded repeat or disease severity.

CTG-Expansion Effects on RNA

Timchenko et al. (1996) identified a novel hnRNP gene whose product, NAB50 (601074), binds to the CUG repeat region of the DM kinase mRNA. Since myotonic dystrophy is caused by a CTG expansion in the 3-prime untranslated region of the DM gene, one model of DM pathogenesis suggests that RNAs from the expanded allele create a gain-of-function mutation by the inappropriate binding of proteins to the CUG repeats. Philips et al. (1998) presented data indicating that the conserved heterogeneous nuclear ribonuclear protein CUG-binding protein (CUGBP; 601074) may mediate the transdominant effect of the RNA. CUGBP was found to bind to the human cardiac troponin T (TNNT2; 191045) pre-messenger RNA and regulate its alternative splicing. Splicing of cardiac troponin T was disrupted in DM striated muscle and in normal cells expressing transcripts that contain CUG repeats. Altered expression of genes regulated posttranscriptionally by CUGBP, therefore, may contribute to DM pathogenesis. Philips et al. (1998) predicted that processing (e.g., splicing) of transcripts from muscle-specific genes is disrupted in DM.

Tiscornia and Mahadevan (2000) identified 4 RNA-splicing factors that bind to 2 short regions 3-prime of the (CUG)n of the DMPK (605377) mRNA: HNRNPC (164020), U2 auxiliary factor (see U2AF1; 191317), polypyrimidine tract-binding protein (PTB; 600693), and PTB-associated splicing factor (PSF; 605199). They also identified a novel 3-prime DMPK exon that results in an mRNA lacking the repeats. In contrast to (CUG)n-containing mRNAs, the novel isoform is not retained in the nucleus in DM cells, resulting in imbalances in relative levels of cytoplasmic DMPK mRNA isoforms and a dominant effect of the mutation on DMPK.

To study the effects of the DM mutation in a controlled environment, Amack et al. (1999) established a cell culture model system using mouse myoblasts. By expressing chimeric reporter constructs containing a reporter gene fused to a human DMPK 3-prime-untranslated region (3-prime-UTR), they identified both cis and trans effects that were mediated by the DM mutation. They found 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. They determined by deletion analysis that an expanded (CTG)n tract alone was sufficient to mediate these cis effects. Moreover, 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 the mouse myoblasts. The myoblast fusion defect could be rescued by eliminating the expression of the mutant DMPK 3-prime-UTR transcript. These results provided 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.

Evidence supports a model in which nuclear accumulation of RNA from the expanded allele contributes to pathogenesis through a trans-dominant effect of CUG-repeat RNA on RNA processing by altering the function of CUG-binding proteins (Timchenko, 1999; Miller et al., 2000). One CUG-binding protein, CUGBP, is a member of the CELF family of RNA-processing factors that regulate alternative splicing (Ladd et al., 2001). Savkur et al. (2001) demonstrated that alternative splicing of the insulin receptor (INSR; 147670) pre-mRNA is aberrantly regulated in DM1 skeletal muscle tissue, resulting in predominant expression of the lower-signaling nonmuscle isoform, IR-A, which lacks exon 11. IR-A predominates in DM1 skeletal muscle cultures, which exhibit a decreased metabolic response to insulin relative to cultures from normal controls. Steady-state levels of CUGBP are increased in DM1 skeletal muscle; overexpression of CUGBP in normal cells induces a switch to IR-A. The CUGBP protein mediates this switch through an intronic element located upstream of the alternatively spliced exon 11, and specifically binds with this element in vitro. These results supported a model in which increased expression of a splicing regulator contributes to insulin resistance in DM1 by affecting alternative splicing of INSR pre-mRNA. Alternative splicing of cardiac troponin T (TNNT2; 191045), a demonstrated target of CUGBP regulation, is altered in DM1 heart tissues and skeletal muscle cultures (Philips et al., 1998). The aberrant regulation of cardiac troponin T alternative splicing in DM1 cells requires the intronic binding site for CUGBP, demonstrating that the aberrant regulation is mediated by an abnormal activity of CUGBP or other CELF proteins.

Amack and Mahadevan (2001) showed that DMPK transcripts containing expanded CUG tracts can form both nuclear and cytoplasmic RNA foci. However, transcripts containing neither a CUG expansion alone nor a CUG expansion plus the distal region of the DMPK 3-prime UTR RNA affected C2C12 myogenesis. This implies that RNA foci formation and perturbation of any RNA binding factors involved in this process are not sufficient to block myoblast differentiation. RNA analysis of myogenic markers revealed that mutant DMPK 3-prime UTR mRNA significantly impeded upregulation of the differentiation factors myogenin (159980) and p21 (116899).

Sergeant et al. (2001) showed that the pattern of MAPT isoforms aggregated in DM1 brain lesions was distinct, consisting mainly of the shortest human tau isoform. Reduced expression of tau isoforms containing exon 2 was observed at both the mRNA and protein levels. Large expanded CTG repeats were detected and showed marked somatic heterogeneity between DM1 cases and in cortical brain regions analyzed. The authors suggested a relationship between the CTG repeat expansion and the alteration of tau expression.

Mankodi et al. (2001) investigated the possibility that DM2 (602668) is caused by expansion of a CTG repeat or related sequence. Analysis of DNA by repeat expansion detection methods and RNA by ribonuclease protection did not show an expanded CTG or CUG repeat in DM2. However, hybridization of muscle sections with fluorescence-labeled CAG-repeat oligonucleotides showed nuclear foci in DM2 similar to those seen in DM1. Nuclear foci were present in all patients with symptomatic DM1 (n = 9) or DM2 (n = 9), but not in any disease controls or healthy subjects (n = 23). The foci were not seen with CUG- or GUC-repeat probes. Foci in DM2 were distinguished from DM1 by lower stability of the probe-target duplex, suggesting that a sequence related to the DM1 CUG expansion may accumulate in the DM2 nucleus. Muscleblind proteins (see 606516), which interact with expanded CUG repeats in vitro, localized to the nuclear foci in both DM1 and DM2. The authors proposed that nuclear accumulation of mutant RNA is pathogenic in DM1, a similar disease process may occur in DM2, and muscleblind may play a role in the pathogenesis of both disorders.

In DM, expression of RNAs that contain expanded CUG or CCUG repeats is associated with degeneration and repetitive action potentials (myotonia) in skeletal muscle. Using skeletal muscle from a transgenic mouse model of DM, Mankodi et al. (2002) showed that expression of expanded CUG repeats reduces the transmembrane chloride conductance to levels well below those expected to cause myotonia. The expanded CUG repeats trigger aberrant splicing of pre-mRNA for CLC1 (CLCN1; 118425), the main chloride channel in muscle, resulting in loss of CLC1 protein from the surface membrane. Mankodi et al. (2002) identified a similar defect in CLC1 splicing and expression in human DM1 and DM2. They proposed that a transdominant effect of mutant RNA on RNA processing leads to chloride channelopathy and membrane hyperexcitability in DM.

Charlet-B et al. (2002) demonstrated loss of CLC1 mRNA and protein in DM1 skeletal muscle tissue due to aberrant splicing of the CLC1 pre-mRNA. They showed that the splicing regulator, CUGBP, which is elevated in DM1 striated muscle, binds to the CLC1 pre-mRNA, and that overexpression of CUGBP in normal cells reproduces the aberrant pattern of CLC1 splicing observed in DM1 skeletal muscle. Charlet-B et al. (2002) proposed that disruption of alternative splicing regulation causes a predominant pathologic feature of DM1.

Ebralidze et al. (2004) showed that DMPK mutant RNA binds and sequesters transcription factors, with up to 90% depletion of selected transcription factors from active chromatin. Diverse genes are consequently reduced in expression, including the ion transporter CLC1, which has been implicated in myotonia. When transcription factor specificity protein-1 (SP1; 189906) was overexpressed in DM1-affected cells, low levels of mRNA for CLC1 were restored to normal. The authors concluded that transcription factor leaching from chromatin by mutant RNA provides a potentially unifying pathomechanistic explanation for this disease.

The myotubularin-related 1 gene (MTMR1; 300171) belongs to a highly conserved family of eukaryotic phosphatases. Buj-Bello et al. (2002) identified 3 coding exons in the MTMR1 intron 2 that are conserved between mouse and human, are alternatively spliced, and give rise to 6 mRNA isoforms. One of the transcripts is muscle specific, is induced during myogenesis, and represents the major isoform in adult skeletal muscle. The authors found a striking reduction in the level of the muscle-specific isoform and the appearance of an abnormal MTMR1 transcript in differentiated congenital DM1 muscle cells in culture as well as in skeletal muscle from congenital DM1 patients. The authors hypothesized that MTMR1 may play a role in muscle formation, and may represent another target for abnormal mRNA splicing in myotonic dystrophy.

Jiang et al. (2004) found that in postmortem DM1 brain tissue, mutant DMPK transcripts were widely expressed in cortical and subcortical neurons. The mutant transcripts accumulated in discrete foci within neuronal nuclei. Proteins in the muscleblind (see MBNL1, 606516) family were recruited into the RNA foci and depleted elsewhere in the nucleoplasm. In parallel, a subset of neuronal pre-mRNAs showed abnormal regulation of alternative splicing. The authors suggested that CNS impairment in DM1 may result from a deleterious gain of function by mutant DMPK mRNA.

Kimura et al. (2005) investigated the alternative splicing of mRNAs of 2 major proteins of the sarcoplasmic reticulum, the ryanodine receptor-1 (RYR1; 180901) and sarcoplasmic/endoplasmic reticulum Ca(2+)-transporting ATPases SERCA1 (ATP2A1; 108730) or SERCA2 (ATP2A2; 108740), in skeletal muscle from DM1 patients. The fetal variants, ASI(-) of RYR1, which lacks residues 3481 to 3485, and SERCA1b, which differs at the C-terminal end, were significantly increased in DM1 skeletal muscle and a transgenic mouse model of DM1 (HAS-LR). In addition, a novel variant of SERCA2 was significantly decreased in DM1 patients. The total amount of mRNA for RYR1, SERCA1, and SERCA2 in DM1 and the expression levels of their proteins in HAS-LR mice were not significantly different. However, heterologous expression of ASI(-) in cultured cells showed decreased affinity for ryanodine but similar calcium dependency, and decreased channel activity in single-channel recording when compared with wildtype RYR1. In support of this, RYR1-knockout myotubes expressing ASI(-) exhibited a decreased incidence of calcium oscillations during caffeine exposure compared with that observed for myotubes expressing wildtype RYR1. Kimura et al. (2005) suggested that aberrant splicing of RYR1 and SERCA1 mRNAs may contribute to impaired calcium homeostasis in DM1 muscle.

Hino et al. (2007) identified motifs downstream of exon 22 of the SERCA1 gene that serve as MBNL1-binding motifs and positively regulate SERCA1 exon 22 splicing. Overexpression of the CUG repeat expansion of DMPK mRNA resulted in the exclusion of exon 22 of SERCA1. These results suggested that sequestration of MBNL1 into the CUG repeat expansion of DMPK mRNA caused the splicing defect and exclusion of SERCA1 exon 22. The expression of this aberrantly spliced SERCA1 could affect the regulation of calcium concentration of sarcoplasmic reticulum in DM1 patients.

Using a reversible transgenic mouse model of RNA toxicity in DM1, Yadava et al. (2008) showed that overexpression of a normal human DMPK 3-prime UTR with only (CUG)5 resulted in cardiac conduction defects, increased expression of Nkx2.5 (NKX2E; 600584), and profound disturbances in connexin-40 (GJA5; 121013) and connexin-43 (GJA1; 121014). Overexpression of the DMPK 3-prime UTR in mouse skeletal muscle also induced transcriptional activation of Nkx2.5 and its targets. Human DM1 muscle, but not normal human muscle, showed similar aberrant expression of NKX2.5 and its targets. In mice, the effects on Nkx2.5 and its targets were reversed by silencing toxic RNA expression. Furthermore, haploinsufficiency of Nkx2.5 in Nkx2.5 +/- mice had a cardioprotective effect against defects induced by DMPK 3-prime UTR. Yadava et al. (2008) concluded that NKX2.5 is a modifier of DM1-associated RNA toxicity in heart.

Using RT-PCR to study skeletal and cardiac muscle from patients with DM1, Nakamori et al. (2008) observed splicing abnormalities in the alpha-dystrobrevin gene (DTNA; 601239), which is part of the skeletal muscle dystrophin (DMD; 300377)-glycoprotein complex. Protein analysis showed that 1 of the abnormally spliced DTNA isoforms localized to the sarcolemma of DM1 muscle and caused enhanced recruitment of alpha-syntrophin (SNTA1; 601017) to the sarcolemma. Nakamori et al. (2008) postulated that these changes may interfere with signaling in DM1 muscle cells.

Botta et al. (2008) found that the DMPK CTG repeat expansion size correlated with splicing defects observed in muscle samples from 12 patients with DM1, with particular attention to the developmentally regulated genes INSR, TNNC1 (191040), CLCN1, and MBNL1. There was also a correlation between increased expansion size and the number of ribonuclear foci, which represented nuclear retention of untranslated DMPK transcripts. There was no relationship between expression levels of the DMPK transcript and repeat expansion size.

Fugier et al. (2011) demonstrated that alternative splicing of the BIN1 gene (601248) was disrupted in muscle cells derived from patients with DM1 and DM2. Exon 11 of BIN1 mRNA was skipped, and the amount of skipped mRNA correlated with disease severity. This splicing misregulation was associated with sequestration of the splicing regulator MBNL1 due to pathogenic expanded CUG or CCUG repeats. Expression of BIN1 without exon 11 resulted in little or no T tubule formation in cultured muscle cells, since this splice variant lacks a phosphatidylinositol 5-phosphate-binding site necessary for membrane-tubulating activities. Skeletal muscle biopsies from patients with DM1 showed disorganized BIN1 localization and irregular T tubule networks. Promotion of the skipping of Bin1 exon 11 in mouse skeletal muscle resulted in abnormal T tubules and decreased muscle strength, although muscle integrity was maintained. There was also decreased expression of Cacna1s (114208), which plays a role in the excitation-contraction coupling process. The findings suggested a link between abnormal BIN1 expression and muscle weakness in myotonic dystrophy.

Tang et al. (2012) observed altered splicing of the calcium channel subunit CAV1.1 (CACNA1S) in muscle of patients with DM1 and DM2 compared with normal adult muscle and muscle of patients with facioscapulohumeral muscular dystrophy (FSHD; see 158900). A significant fraction of CAV1.1 transcripts in DM1 and DM2 muscle showed skipping of exon 29, which represents a fetal splicing pattern. Forced exclusion of exon 29 in normal mouse skeletal muscle altered channel gating properties and increased current density and peak electrically evoked calcium transient magnitude. Downregulation of Mbnl1 in mouse cardiac muscle or overexpression of Cugbp1 in mouse tibialis anterior muscle enhanced skipping of exon 29, suggesting that these splicing factors may be involved in the CAV1.1 splicing defect in myotonic dystrophy.

Rinaldi et al. (2012) found downregulation and aberrant splicing of the MYH14 (608568) gene in muscle biopsies from 12 patients with DM1 compared to 7 controls. DM1 patients had increased amounts of the alternatively spliced MYH14 isoform NMHCII-C0 that lacks 8 amino acids in exon 6 close to the ATP binding loop; this isoform has decreased actin-activated MgATPase activity compared to the isoform with the 8 amino acids (NMHCII-C1). The amount of aberrantly spliced MYH14 was proportionate to the DMPK CTG expansion grade. However, MYH14 retained normal subcellular localization in DM1 patient muscle, albeit at lower amounts than in controls. Minigene assays indicated that levels of MBNL1 positively regulated the inclusion of MYH14 exon 6, suggesting that the DMPK expansion interferes with MBNL1 processing of MYH14 pre-mRNA. Rinaldi et al. (2012) suggested that alterations in expression of the MYH14 gene may contribute to the pathogenesis of DM1 and may underlie the occasional observation of sensorineural hearing loss in DM1 patients.

Jain and Vale (2017) showed that repeat expansions create templates for multivalent basepairing, which causes purified RNA to undergo a sol-gel transition in vitro at a similar critical repeat number as observed in Huntington disease (143100), spinocerebellar ataxia (e.g., 164400), myotonic dystrophy, and FTDALS1 (105550). In human cells, RNA foci form by phase separation of the repeat-containing RNA and can be dissolved by agents that disrupt RNA gelation in vitro. Jain and Vale (2017) concluded that, analogous to protein aggregation disorders, their results suggested that the sequence-specific gelation of RNAs could be a contributing factor to neurologic disease.

CTG-Expansion Effects on Cell Function

Furling et al. (2001) developed an in vitro cell culture system which displayed several defects previously described for congenital myotonic dystrophy (CDM) muscle in vivo. Satellite cells are quiescent muscle cells which retain the ability to become myogenic precursor cells (myoblasts). Human satellite cells were isolated from the quadriceps muscles of 3 CDM fetuses with different clinical severity. By Southern blot analysis, all 3 cultures were found to have approximately 2,300 CTG repeats. This CTG expansion was found to progressively increase during the proliferative life span, confirming instability of this triplet in skeletal muscle cells. The CDM myoblasts and myotubes showed abnormal retention of mutant RNA in nuclear foci. The proliferative capacity of the CDM myoblasts was reduced and a delay in fusion, differentiation, and maturation was observed in the CDM cultures compared with unaffected myoblast cultures. The clinical severity and delayed maturation observed in the CDM fetuses were closely reflected by the phenotypic modifications observed in vitro. The authors concluded that satellite cells are defective in CDM and may be implicated in the delay in maturation and muscle atrophy that has been described in CDM fetuses.

Diagnosis

In classic adult-onset cases, clinical diagnosis is straightforward with demonstration of progressive distal and bulbar dystrophy in the presence of myotonia, with frontal balding, and cataracts. Confirmatory evidence is provided by demonstration of depressed IgG and elevated CPK in the serum. Clinical diagnosis can be difficult in mild cases, where cataracts may be the only manifestation (Bundey et al., 1970).

In studies of an extensively affected Labrador kindred, Webb et al. (1978) concluded that lens opacities are not a reliable diagnostic sign. Many younger affected persons, including one in his 20s, did not have lens opacities despite clear muscular involvement. On the other hand, Ashizawa et al. (1992) concluded that bilateral iridescent and posterior cortical lens opacities are highly specific for DM and are useful for establishing the clinical diagnosis. The sensitivity of these 2 features was found to be 46.7% and 50.0%, respectively, in their series, while their specificities were 100% in both cases.

Direct analysis of the size of the CTG repeat by Southern blotting permits DNA diagnosis. Normal individuals have 5 to 37 CTG repeats, whereas patients have from more than 50 to several thousand CTG repeats in peripheral leukocytes (see review by Pizzuti et al., 1993).

Reardon et al. (1992) described a 5-year experience in providing presymptomatic and prenatal molecular diagnostic services based on the linkage principle using closely linked markers in 161 families. Only 10 analyses out of 235 proved uninformative, but a further 5 requests (1.9%) could not be reported because of uncertainty in clinical status. Seven of 81 (8.6%) patients considered to be at low risk on clinical grounds were found to be at high risk of carrying the gene. Reardon et al. (1992) emphasized that careful clinical examination and appropriate investigations of nonmolecular nature remain a cornerstone of diagnosis.

Clinical Management

Lightweight ankle-foot orthoses are useful for foot drop, as are specially designed utensils for hand weakness. Weakness of respiratory muscles may require postural drainage and nocturnal respiratory support in advanced ages. Heart failure and aspiration pneumonia secondary to impaired esophageal motility should be considered. Nocturnal hypoventilation may contribute to a hypersomnia distinct from narcolepsy (161400), and should be evaluated with sleep studies.

Prolongation of the PR interval can progress to heart block, requiring placement of a pacemaker. Periodic EKGs and avoidance of drugs such as procainamide and quinine (Griggs et al., 1975) are recommended.

Myotonia is rarely a major clinical concern. Those patients with significant stiffness benefit most from avoiding cold and by doing warm-up exercises. In selected patients, dilantin, quinidine, procainamide, myxilitene, diamox, and other drugs reduce myotonia modestly.

Periodic ophthalmoscopy is needed to assess posterior capsular cataracts, which may require extraction if vision is impaired significantly--rarely before the third or fourth decade. If tarsorrhaphy is undertaken for repair of ptosis, care must be taken not to overcorrect lest failure of eyelid closure lead to corneal abrasion.

Dysfunction of sex hormones does not cause infertility. Obstetric difficulties are common. Hypomotility of the intestinal tract is not infrequent but usually does not require treatment. Dysphagia is usually manageable with conservative dietary measures. Schwindt et al. (1969) claimed that 25 to 50% of patients have abdominal symptoms due to cholelithiasis. Brunner et al. (1992) described 4 DM patients with recurrent intestinal pseudoobstruction. In 1 patient it preceded significant muscle weakness by 15 years. Conservative measures usually were effective. Improved intestinal function was noted in 1 patient treated with the prokinetic agent cisapride. A partial sigmoid resection was performed in 3 patients with dolichomegacolon.

Keller et al. (1998) stated that respiratory insufficiency at birth was the most critical factor for the survival of patients with congenital myotonic dystrophy. They reported 2 premature infants with congenital myotonic dystrophy requiring prolonged ventilatory support who were successfully weaned using nasal continuous positive airway pressure.

In 15 patients with genetically confirmed DM1, Logigian et al. (2004) used a device to measure the relaxation time of the first dorsal interosseus muscle after ulnar nerve twitch and tetanic stimulation. Compared to controls, tetanic and twitch relaxation time was longer in patients, mainly due to delay in the terminal (measured as 50 to 5% peak force), rather than the initial (90 to 50% peak force), phase of relaxation. The delay in relaxation was much greater in tetanic than single-twitch recordings, and both were positively correlated with leukocyte DMPK (605377) CTG repeat length, suggesting a triplet repeat toxic dosage effect. Logigian et al. (2004) suggested that quantitative analysis of muscle myotonia may be used to follow the natural history of the disease and to assess response to therapeutic intervention.

Orngreen et al. (2005) found that 12 patients with myotonic dystrophy responded well to a 12-week program of aerobic training on a cycle ergometer. The patients increased maximal oxygen uptake by 14% and maximal workload by 12%. There was an increase in muscle fiber diameter without an increase in serum creatine kinase. The authors concluded that aerobic training is safe and effective for improving fitness in myotonic dystrophy patients.

The expanded (CTG)n tract in the 3-prime untranslated region (UTR) of the DMPK gene results in nuclear entrapment of the 'toxic' mutant RNA and interacting RNA-binding proteins such as MBNL1 (606516) in ribonuclear inclusions. It had been suggested that therapy aimed at eliminating the toxin would be beneficial. Timchenko (2006) commented on the study of Mahadevan et al. (2006) in transgenic mice showing that normalizing the number of CUG repeat-containing DMPK transcripts reversed the myotonia and cardiac conduction defects in the mouse model. Developing an approach to reduce CUG repeats might be a viable therapeutic strategy. An alternative approach would be to learn how to control CUGBP1 (601074) RNA-binding activity in order to reduce its toxicity. The results of Mahadevan et al. (2006) represented the first in vivo proof of principle for a therapeutic strategy for treatment of myotonic dystrophy by ablating or silencing expression of the toxic RNA molecules.

Wheeler et al. (2009) used a transgenic mouse model to show that derangements of myotonic dystrophy are reversed by a morpholino antisense oligonucleotide, CAG25, that binds to CUG(exp) RNA and blocks its interaction with MBNL1, a CUG(exp)-binding protein. CAG25 disperses nuclear foci of CUG(exp) RNA and reduces the overall burden of this toxic RNA. As MBNL1 is released from sequestration, the defect of alternative splicing regulation is corrected, thereby restoring ion channel function. Wheeler et al. (2009) concluded that their findings suggested an alternative use of antisense methods, to inhibit deleterious interactions of proteins with pathogenic RNAs.

Mulders et al. (2009) identified a CAG(7) antisense oligonucleotide that silenced mutant DMPK RNA expression and reduced the number of ribonuclear aggregates in a (CUG)n-length-dependent manner in both mouse and human DM1 cells. Direct administration of this oligonucleotide in muscle of DM1 mice in vivo caused a significant reduction in the level of toxic (CUG)n RNA and showed a normalizing effect on aberrant pre-mRNA splicing. The data demonstrated proof of principle for therapeutic use of simple sequence antisense oligonucleotides in DM1 and potentially other unstable microsatellite diseases.

Logigian et al. (2010) found that treatment of DM1 patients with mexiletine resulted in a significant reduction in grip relaxation time without major side effects or EKG conduction abnormalities. The study involved 2 parts, each with 20 patients taking 150 or 200 mg 3 times daily, respectively, over 7 weeks. Mexiletine is a lidocaine analog that acts as a sodium-channel blocker in skeletal and cardiac muscle.

Population Genetics

The overall prevalence of DM1 is estimated to be 1 in 8,000 (Musova et al., 2009).

In the Saguenay region of the province of Quebec, the prevalence of myotonic dystrophy is about 1 in 475; about 600 cases are known in a population of 285,000. Mathieu et al. (1990) estimated that the prevalence of myotonic dystrophy in the Saguenay-Lac-Saint-Jean region of Quebec province is 30 to 60 times higher than the prevalence in most other regions of the world. They identified 746 patients (673 still alive) distributed in 88 families in this region, and traced all patients to a couple who settled in New France in 1657. De Braekeleer (1991) estimated the prevalence of myotonic dystrophy in the French Canadian population in the Saguenay-Lac-Saint-Jean region of Quebec province at more than 1/514, as contrasted with the estimate of 1/25,000 for European populations generally. Dao et al. (1992) found no differences in fertility in myotonic dystrophy individuals in the Saguenay-Lac-Saint-Jean region in a case-control study of 373 affected persons who married between 1855 and 1971.

Bouchard et al. (1988) reviewed the genetic demography of the disorder. They were unable to demonstrate the selective disadvantage of the DM gene. Ashizawa and Epstein (1991) claimed that DM among ethnic Africans, especially in central and southern Africa, as well as in Cantonese, Thai, and probably Oceanians, has a low prevalence. In their survey they used Duchenne muscular dystrophy as a control and found that it had an incidence similar to that in western nations. They suggested that the findings are consistent with the evolution and migration of the human species from Africa. Novelli et al. (1994) found a low frequency of the 'at risk' CTG alleles (n = repeat number less than 19), postulated to be the basis of the expanded repeats causing myotonic dystrophy, in Albanians, Egyptians, and Italians, whereas they did not detect alleles of this sort in any chromosomes of the Bamilekes, a Bantu-speaking people from central and southern Cameroon. They interpreted the findings as consistent with the low frequency reported by Ashizawa and Epstein (1991) and provided a molecular basis supporting a north Eurasian origin of the DM mutation.

Harley et al. (1991) found linkage disequilibrium between DM and the D19S63 marker, the first demonstration of this phenomenon in a heterogeneous DM population. The results suggested that at least 58% of DM patients in the British population, as well as those in a French Canadian population, are descended from the same ancestral DM mutation. The result was considered entirely consistent with previous population studies which indicated a very low mutation rate in DM (Harper, 1989). (Harley et al. (1992) stated that no case of mutation had been proven.) The DM mutation in the French Canadian population (Mathieu et al., 1990) appears to have been introduced into Quebec province by one of the original founders over 300 years ago and may have originated in northern Europe before the spread of this population to the British Isles. The remaining 42% of DM chromosomes may include some that have the same mutation (which has become associated with different D19S63 alleles through recombination) together with one or more other DM mutations. Although linkage disequilibrium with other closely linked markers--APOC2 (608083), CKM (123310), and BCL3 (109560)--was not observed in the Welsh population, strong disequilibrium was observed in the French Canadian population.

Goldman et al. (1995) studied the association between normal alleles at the CTG repeat in 2 nearby polymorphisms in the myotonin protein kinase gene in South African Negroids, a population in which myotonic dystrophy had not been described. They found a significantly different CTG allelic distribution from that in Caucasoids and Japanese: CTG repeat lengths greater than 19 were very rare. The striking linkage disequilibrium between specific alleles at the Alu insertion/deletion polymorphism, the HinfI polymorphism of intron 9, and the CTG repeat polymorphism seen in Caucasoids in Europe and Canada was also found in the South African Negroid population. Goldman et al. (1995), however, found numerous haplotypes not previously described in Europeans. Thus it seemed likely that only a small number of these 'African' chromosomes were present in the progenitors of all non-African peoples. The data provided support for the 'out of Africa' model for the origin of modern humans and suggested that the rare ancestral DM mutation event may have occurred after the migration from Africa, thus accounting for the absence of DM in sub-Saharan Negroid peoples. Goldman et al. (1996) reported molecular evidence for a DM founder effect in South African families. DM haplotype I was found in the South African DM population and rarely in the non-DM population. Goldman et al. (1996) noted that both the geographic distribution of families with DM (occurrence primarily in Afrikaans-speaking families who originated in the Northern Transvaal) and a previous genealogic study by Lotz and van der Meyden (1985) also suggested a founder effect as the likely explanation for the high prevalence of DM. Lotz and van der Meyden (1985) found no single case of DM in an indigenous Negroid or Khoisan person from southern Africa, despite a survey representing a population of more than 30 million (Ashizawa and Epstein, 1991).

Harley et al. (1992) found that a second polymorphism near the triplet repeat was in almost complete linkage disequilibrium with myotonic dystrophy, strongly supporting these earlier results (Harley et al., 1991) that indicated that most cases are descended from one original mutation. Cobo et al. (1992) found that DM and D19S63 showed linkage disequilibrium in the Spanish population also. They studied 33 Spanish families from 5 different geographic regions.

Passos-Bueno et al. (1995) found a relatively low frequency of DM families of black racial background in Brazil. Three of 41 DM families were of that ancestry in the city of Sao Paulo in which 40% of the population was black. The authors thought that bias in ascertainment could not be the explanation.

In 72 French families, Lavedan et al. (1994) found that 100% of chromosomes with the DM mutation carried an intragenic 1-kb insertion. They also detected significant linkage disequilibrium between the DM locus and D19S63 for which allelic frequencies were different from other European populations. The results were consistent with the hypothesis that the CTG expansion occurred on one or a few ancestral chromosomes carrying the large 1-kb insertion allele.

Goldman et al. (1996) studied the CTG trinucleotide repeat in the DMK gene by PCR analysis in 246 unrelated South African Bantu-speaking Negroids, 116 San and 27 Pygmies. The size and distribution of the CTG repeat were determined and showed that the alleles ranged in length from 5 to 22 repeats. The most common CTG repeat was 5 (25% of chromosomes) in the South African Negroids but 11 (27% of chromosomes) in the San population, and 12 (22% of chromosomes) in the Pygmies. The South African Bantu-speaking Negroids and San thus had significantly larger repeat length alleles than do Caucasoid and Japanese populations. Again, Goldman et al. (1996) concluded that the occurrence of fewer large CTG repeats in the normal range accounts for the absence of DM from Southern African Negroids and suggests that the rare DM mutation event postulated to have occurred on a specific chromosomal haplotype took place originated after the migration of humans from Africa.

Deka et al. (1996) analyzed the CTG repeat length and the neighboring Alu insertion/deletion (+/-) polymorphism in DNA samples from 16 ethnically and geographically diverse human populations. They found that the CTG repeat length is variable in human populations. Although the (CTG)5 repeat is the most common allele in most populations, it was absent among Costa Ricans and New Guinea highlanders. They detected a (CTG)4 repeat allele, the smallest CTG known, in an American Samoan individual. Alleles with 19 or more CTG repeats were the most frequent in Europeans, followed by the populations of Asian origin, and are absent or rare in Africans. To understand the evolution of CTG repeats, Deka et al. (1996) used haplotype data from the CTG repeat and Alu(+/-) locus. The results were consistent with previous studies and showed that among individuals of Caucasian and Japanese origin the association of the Alu(+) allele with CTG repeats of 5 and at least 19 is complete, whereas the Alu(-) allele is associated with (CTG)11-16 repeats. However, these associations are not exclusive in non-Caucasian populations. Most significantly, Deka et al. (1996) detected the (CTG)5 repeat allele on an Alu(-) background in several populations including native Africans. As no (CTG)5 repeat allele on an Alu(-) background had been observed hitherto, they proposed that the Alu(-) allele arose on a (CTG)11-13 background. They suggested further that the most parsimonious evolutionary model is (1) that (CTG)5-Alu(+) is the ancestral haplotype; (2) that (CTG)5-Alu(-) arose from a (CTG)5-Alu(+) chromosome later in evolution; and (3) that expansion of CTG alleles occurred from (CTG)5 alleles on both Alu(+) and Alu(-) backgrounds.

Tishkoff et al. (1998) studied the origin of myotonic dystrophy mutations by analyzing haplotypes consisting of the (CTG)n repeat, as well as several flanking markers at the myotonic dystrophy locus, in normal individuals from 25 human populations (5 African, 2 Middle Eastern, 3 European, 6 East Asian, 3 Pacific/Australo-Melanesian, and 6 Amerindian) and in 5 nonhuman primate species. They found that non-African populations had a subset of haplotype diversity present in Africa, as well as a shared pattern of allelic association. (CTG)18-35 alleles (large normal) were observed only in northeastern African and non-African populations and exhibited strong linkage disequilibrium with 3 markers flanking the (CTG)n repeat. The pattern of haplotype diversity and linkage disequilibrium observed supported a recent African-origin model of modern human evolution and suggested that the original mutational event that gave rise to DM-causing alleles arose in a population ancestral to non-Africans before migration of modern humans out of Africa.

Neville et al. (1994) performed a high-resolution genetic analysis of the DM locus using PCR-based assays of 9 polymorphisms immediately flanking the DM repeat. With the exception of the case reported from Africa by Krahe et al. (1995), all cases of DM in the world appear to share a single haplotype that contains putative at-risk CTG alleles, i.e., alleles with 19 to 30 CTG repeats that may serve as a reservoir for recurrent mutations to unstable alleles with 30 to 50 repeats (Imbert et al., 1993). Yamagata et al. (1998) found 6 different haplotypes in the Japanese population and determined that DM alleles were always haplotype A (in the nomenclature of Neville et al., 1994), the same as in Caucasians. In both Caucasian and Japanese populations, a multistep process of triplet repeat expansion originated by expansion of an ancestral n = 5 repeats to n = 19 to 37 copies. A similar multistep model has been suggested for Friedreich ataxia (229300).

Pan et al. (2001) described a low frequency (1.4%) of CTG repeats (larger than 18 repeats) in the Taiwanese population, predicting a low prevalence of DM1. As in Caucasian and Japanese populations, all of the Taiwanese DM1 chromosomes examined were exclusively associated with the Alu insertion and 7 additional single base polymorphic markers (haplotype A). The findings suggested that the Taiwanese, and maybe all non-African, DM1 chromosomes may have originated from a pool of large-sized normal alleles with haplotype A, which was generated after the migration out of Africa.

Siciliano et al. (2001) calculated the DM prevalence rates in Padua (northeast Italy) and in 4 provinces in northwest Tuscany (central Italy) using molecular genetic testing. A minimum prevalence rate of 9.31 x 10(-5) persons was found, consistent with epidemiologic rates worldwide, and more than 2 times the size of those of 2 previous studies conducted in the same areas during the era before molecular genetic testing. The results underlined the importance of direct genetic diagnosis of DM, especially in detecting mildly affected patients.

In a comprehensive epidemiologic survey among Jews living in Israel, Segel et al. (2003) found that the average prevalence of DM was 15.7 per 100,000 (1 case in 6,369), with intercommunity variations: Ashkenazi Jews had the lowest rate (1 case in 17,544) as compared to those in Sephardi/Oriental Jews and Yemeni Jews (1 case in 5,000 and 1 case in 2,114, respectively). The rate of unrelated DM sibships per million persons of each community was used as an estimate of the transition rate from stable to unstable DMPK-(CTG)n alleles assuming that each transition is a beginning of a new DM sibship. This study indicated that the difference in the incidence of DM is a result of higher mutation rate in the non-Ashkenazi Jews as compared to the rate in the Ashkenazi Jews. The intragenic haplotype of the DM alleles was the same as that in DM patients in many populations worldwide; however, 2 markers closely linked to DM, D19S207 and D19S112, were in linkage disequilibrium with the DM mutation in patients of Yemeni and Moroccan (the largest subgroup of the Sephardi Jews) extractions but not in the Ashkenazi patients. This observation indicated a common ancestral origin for the DM premutation in patients of the same ethnic origin. Segel et al. (2003) concluded that the difference in DM prevalence among the Jewish communities is a consequence of founder premutations in the non-Ashkenazi Jewish communities.

Yotova et al. (2005) used SNP and microsatellite markers to characterize a 2.05-Mb DNA segment spanning the DM1-expansion site in 50 DM1 families from northeastern Quebec. The results suggested the existence of 3 basic haplotype families, A, B, and C, with A being the most common. By analyzing proportions of recombinant haplotypes, Yotova et al. (2005) estimated that haplotype A was the 'driver' founder effect, with an age of 9 generations, consistent with the settlement of Charlevoix at the turn of the 17th century and subsequent colonization of Saguenay-Lac-Saint-Jean. The minor haplotypes B and C were likely introduced independently.

Medica et al. (2007) found that 4 (1.46%) of 274 unrelated adults with cataract, but no evidence or family history of DM1, carried a 'protomutation' in the DMPK gene ranging between 52 and 81 CTG repeats. The authors hypothesized that these patients with protomutations represented a source of full expansion mutation, which could be responsible for maintaining DM1 mutations in a population. Stable transmission to an unaffected offspring was observed in 1 individual with a protomutation. Three of the patients were from the Croatian region of Istria, which has a high prevalence of DM1.

Acton et al. (2007) reported 2 African American brothers from Alabama who had DM1, both with CTG repeats of 5/639; their father was reportedly affected and had CTG repeats of 5/60. Other unaffected family members had CTG repeats of 5 to 14. Another unrelated African American patient from Alabama had CTG repeats of 27/191. Among 161 African American controls from Alabama, the authors observed 18 CTG alleles from 5 to 28 repeats. A comparison with other ethnic groups showed that the African American individuals from Alabama had more CTG repeats than some African black populations, but fewer than European white or Japanese populations. These data suggested that the risk for DM1 in American blacks is intermediate between that of African blacks and whites of European descent.

Suominen et al. (2011) found 2 DM1 mutations among 4,520 Finnish control individuals and no DM1 mutations among 988 Finnish patients with a neuromuscular disorder. One of the expanded DM1 mutations had 80 repeats, but the size of the other expansion could not be determined. Overall, the DM1 mutation frequency was estimated to be 1 in 2,760 in the general population. In the same study, the frequency of DM2 was estimated to be 1 in 1,830. Suominen et al. (2011) stated that these estimates were significantly higher than previously reported estimates, which they cited as 1 in 8,000 for both DM1 and DM2.

Animal Model

Jansen et al. (1996) examined the effect of altered expression levels of DMPK by disrupting the endogenous Dmpk gene and overexpressing a normal human DMPK transgene in mice. They carried out an analysis of Dmpk gene expression by performing RNA in situ hybridization on whole-mount embryos and body sections of embryos to identify cell lineages that could potentially be affected by abnormal expression of DMPK. Jansen et al. (1996) reported that the results of nullizygous replacement mutations in Dmpk are extremely mild during all phases of mouse development and aging; the only change they noted was marginally altered muscle fiber size in muscles of the head and neck. The only histologic abnormality shown in the over-expressor model was transgene copy number-dependent cardiomyopathy. In these models other prominent features of myotonic dystrophy were lacking. They concluded that simple loss or gain of expression of DMPK was probably not the only crucial requirement for development of myotonic dystrophy.

Benders et al. (1997) studied the role of DMPK in myocyte ion homeostasis in wildtype and homozygous DMPK knockout mice generated by Jansen et al. (1996). Myotubes of knockout mice exhibited a higher resting intracellular calcium concentration than did myotubes of wildtype mice because of an altered open probability of voltage-dependent L-type calcium and sodium channels. Benders et al. (1997) observed smaller and slower calcium responses in myotubes of knockout, as compared to wildtype, mice after triggering with either acetylcholine or high external potassium.

Calcium flux was partially mediated by influx of extracellular calcium through the L-type calcium channel. Neither the content nor the activity of the sodium/potassium ATPase or the sarcoplasmic reticulum calcium ATPase were affected by the absence of DMPK. Benders et al. (1997) suggested that DMPK is involved in modulating the initial events in excitation-contraction coupling in skeletal muscle.

To ascertain if some or all of the symptoms of DM are the consequences of reduced levels of DMPK, Reddy et al. (1996) developed a strain of mice that carry a targeted disruption of the Dmpk gene. Analysis of skeletal muscle structure and function showed that Dmpk -/- mice develop a late-onset progressive skeletal myopathy characterized by decreased force generation, increased fiber degeneration and regeneration, and loss of sarcomeric organization. These changes occurred in mice between 3 and 7 months of age. Reddy et al. (1996) suggested that DMPK may be necessary for the maintenance of skeletal muscle structure and that a decrease in DMPK levels may contribute to DM pathology.

Gourdon et al. (1997) and Monckton et al. (1997) independently studied the behavior of the myotonic dystrophy CTG repeat in transgenic mice. Monckton et al. (1997) generated transgenic mouse lines that transmit a fragment of the human DM kinase gene, a 3-prime UTR-containing construct initially containing 162 CTG repeats. Gourdon et al. (1997) used a much larger genomic fragment (about 45 kb) as a transgene, originally derived from the DNA for a DM patient with 55 CTG repeats in the mutant allele. This cosmid clone not only housed the entire DM gene, but also contained sequences corresponding to the 2 genes immediately flanking the DM kinase gene. Both studies clearly documented intergenerational and somatic cell instability of the trinucleotide repeat in the transgenic mice.

Lia et al. (1998) studied somatic instability by measuring the CTG repeat length at several ages in various tissues of transgenic mice carrying a (CTG)55 expansion surrounded by 45 kb of the human DM region. These mice had been shown to reproduce the intergenerational and somatic instability of the 55 CTG repeat, suggesting that surrounding sequences and the chromatin environment are involved in instability mechanisms. As observed in some of the tissues of DM patients, there was a tendency for repeat length and somatic mosaicism to increase with the age of the mouse. Furthermore, Lia et al. (1998) observed no correlation between the somatic mutation rate and tissue proliferation capacity. Somatic mutation rates in different tissues were also not correlated to the relative intertissue differences in transcriptional levels of the 3 genes that surround the repeat: DMAHP (600963), DMPK, and 59. Similar studies by Seznec et al. (2000) with transgenic mice carrying greater than 300 CTG repeats demonstrated a strong bias towards expansions (vs contractions), similar sex- and size-dependent expansion characteristics as in humans, and a high level of instability (increasing with age) in tissues and in sperm.

Klesert et al. (2000) and Sarkar et al. (2000) independently developed mice with targeted disruption of the Six5 gene. Both animal models developed cataracts, leading Klesert et al. (2000) and Sarkar et al. (2000) to conclude that myotonic dystrophy represents a contiguous gene syndrome involving deficiency of both SIX5 and DMPK.

The CTG expansion causing DM results in transcriptional silencing of the flanking SIX5 allele. Sarkar et al. (2004) generated Six5 knockout and heterozygous mice by targeted disruption and demonstrated a strict requirement of Six5 for both spermatogenic cell survival and spermiogenesis. Leydig cell hyperproliferation and increased intratesticular testosterone levels were observed in the Six5 -/- mice. Although increased FSH (see 136530) levels were observed in the Six5 +/- and Six5 -/- mice, serum testosterone levels and intratesticular inhibin alpha (INHA; 147380) and inhibin beta-B (INHBB; 147390) levels were not altered in the Six5 mutant animals when compared with controls. Steady-state c-Kit (164920) levels were reduced in the Six5 -/- testis. The authors concluded that decreased c-Kit levels could contribute to the elevated spermatogenic cell apoptosis and Leydig cell hyperproliferation in the Six5 -/- mice. They hypothesized that the reduced SIX5 levels may contribute to the male reproductive defects in DM1.

Dmpk knockout mice show only mild muscle weakness and abnormal cardiac conduction; Six5 knockout mice develop cataracts only; neither mouse model develops myotonia. Mankodi et al. (2000) investigated the possibility that the pathogenic effect of the DM mutation is mediated by the mutant mRNA, i.e., that the nuclear accumulation of expanded CUG repeats is toxic to muscle fibers. They developed transgenic mice that express human skeletal actin (ACTA1; 102610) with either a nonexpanded (5-CTG) or an expanded (approximately 250-CTG) repeat in the final exon of the ACTA1 gene, midway between the termination codon and the polyadenylation site. Mice that expressed the expanded repeat developed myotonia and myopathy, whereas mice expressing the nonexpanded repeat did not. Thus, transcripts with expanded CUG repeats are sufficient to generate a DM phenotype. Mankodi et al. (2000) concluded that these results support a role for RNA gain of function in disease pathogenesis.

Mounsey et al. (2000) measured macroscopic and single channel sodium currents from cell-attached patches of skeletal myocytes from heterozygous (DMPK +/-) and homozygous (DMPK -/-) mice. In DMPK -/- myocytes, sodium current amplitude was reduced because of reduced channel number. Single channel recordings revealed sodium channel reopenings, similar to the gating abnormality of human myotonic muscular dystrophy, which resulted in a plateau of sodium current. The gating abnormality deteriorated with increasing age. In DMPK +/- muscle there was reduced sodium current amplitude and increased sodium channel reopenings identical to those in DMPK -/- muscle. The authors hypothesized that DMPK deficiency underlies the sodium channel abnormality in DM.

In tissues cultured from Dmt mice, Gomes-Pereira et al. (2001) noted the progressive accumulation of larger alleles as a result of repeat length changes in vitro, as confirmed by single cell cloning. The authors also observed the selection of cells carrying longer repeats during the first few passages of the cultures and frequent additional selective sweeps at later stages. The highest levels of instability were observed in cultured kidney cells, whereas the transgene remained relatively stable in eye cells and very stable in lung cells, paralleling the previous in vivo observations. No correlation between repeat instability and the cell proliferation rate was found, rejecting a simple association between length change mutations and cell division, and suggesting a role for additional cell-type specific factors.

Kanadia et al. (2003) found that mice with targeted deletion of exon 3 of the Mbnl1 gene (606516) developed overt myotonia with myotonic discharges on EMG at approximately 6 weeks of age. In addition to muscle abnormalities, the mice also developed ocular cataracts similar to DM1. These mice showed decreased expression and abnormal splicing of Clcn1, Tnnt2, and Tnnt3 (600692). Kanadia et al. (2003) concluded that Mbnl1 plays a direct role in splice site selection of different proteins and that manifestations of DM1 can result from sequestration of specific RNA-binding proteins.

In Mbnl1-deficient Drosophila embryos, Machuca-Tzili et al. (2006) found abnormal splicing of the Z-band associated proteins CG30084, which is the Drosophila homolog of ZASP/LDB3 (605906), and alpha-actinin. Studies of skeletal muscle tissue from 3 unrelated DM1 patients showed abnormal splicing of LDB3 but normal splicing of alpha-actinin-2 (ACTN2; 102573). The findings suggested that the molecular breakdown of Z-band structures in flies and DM1 patients may involve the MBNL1 gene.

Wang et al. (2007) generated an inducible and heart-specific mouse model of DM1 that expressed expanded human DMPK CUG-repeat RNA and recapitulated pathologic features of the human disorder, including dilated cardiomyopathy, arrhythmias, and systolic and diastolic dysfunction. The mice also showed misregulation of developmental alternative splicing transitions, including the Tnnt2 and Fxr1 (600819) genes. All died of heart failure within 2 weeks. Immunohistochemical studies showed increased CUGBP1 protein levels specifically in nuclei containing foci of DMPK CUG-repeat RNA. A time-course study showed that increased CUGBP1 cooccurred within hours of induced expression of the CUG repeat and coincided with reversion to embryonic splicing patterns. The results indicated that increased CUGBP1 is a specific and early event of DM1 pathogenesis and represents a primary response to expression of DMPK CUG-repeat mutant RNA.

Wheeler et al. (2007) reported that an antisense oligonucleotide targeting the 3-prime splice site of exon 7a of the Clc1 gene (CLCN1; 118425) reversed the defect of Clc1 alternative splicing in 2 mouse models of DM. By repressing the inclusion of this exon, the treatment restored the full-length reading frame of Clc1 mRNA, upregulated Clc1 expression, normalized Clc1 current density, and eliminated myotonic discharges. The findings supported the hypothesis that myotonia and chloride channelopathy observed in DM results from abnormal alternative splicing of CLC1.

Osborne et al. (2009) performed global mRNA profiling in transgenic mice that expressed CUG(exp) RNA, when compared with Mbnl1-knockout mice. The majority of changes induced by CUG(exp) RNA in skeletal muscle could be explained by reduced activity of Mbnl1, including many changes that are secondary to myotonia. The pathway most affected comprised genes involved in calcium signaling and homeostasis. Some effects of CUG(exp) RNA on gene expression were caused by abnormal alternative splicing or downregulation of Mbnl1-interacting mRNAs. However, several of the most highly dysregulated genes showed altered transcription, as indicated by parallel changes of the corresponding pre-mRNAs. Osborne et al. (2009) proposed that transdominant effects of CUG(exp) RNA on gene expression in this transgenic mouse model may occur at the level of transcription, RNA processing, and mRNA decay, and may be mediated mainly, but not entirely, through sequestration of Mbnl1.

Koshelev et al. (2010) expressed human CUGBP1 in adult mouse heart. Upregulation of CUGBP1 was sufficient to reproduce molecular, histopathologic, and functional changes observed in a DM1 mouse model that expressed expanded CUG RNA repeats (Wang et al., 2007) as well as in individuals with DM1. The authors concluded that CUGBP1 upregulation plays an important role in DM1 pathogenesis.

By inducing expression of human CUGBP1 in adult skeletal muscle of transgenic mice, Ward et al. (2010) showed that the pathogenic features of DM1 could be explained by upregulated CUGBP1 expression. Within weeks of induction of CUGBP1 expression, transgenic mice exhibited impaired movement, reduced muscle function, abnormal gait, and reduced total body weight compared with uninduced controls. Histologic analysis of transgenic muscle overexpressing CUGBP1 revealed centrally located nuclei, myofiber degeneration with inflammatory infiltrate, and pyknotic nuclear clumps. RT-PCR analysis revealed reversion to embryonic splicing patterns in several genes in transgenic muscle overexpressing CUGBP1. Ward et al. (2010) concluded that CUGBP1 has a major role in DM1 skeletal muscle pathogenesis.

Wheeler et al. (2012) showed that nuclear-retained transcripts containing expanded CUG repeats are unusually sensitive to antisense silencing. In a transgenic mouse model of DM1, systemic administration of antisense oligonucleotides caused a rapid knockdown of CUG expansion RNA in skeletal muscle, correcting the physiologic, histopathologic, and transcriptomic features of the disease. The effect was sustained for up to 1 year after treatment was discontinued. Systemically administered ASOs were also effective for muscle knockdown of Malat1 (607924), a long noncoding RNA that is retained in the nucleus. Wheeler et al. (2012) concluded that their results provided a general strategy to correct RNA gain-of-function effects and to modulate the expression of expanded repeats, long noncoding RNAs, and other transcripts with prolonged nuclear residence.

History

Anticipation--earlier onset and more severe manifestations in more recent generations--was described in myotonic dystrophy as a rather striking feature. Penrose (1948) concluded that it is probably an artifact of ascertainment. However, elucidation of the molecular defect (see above) indicates that the mutation can worsen progressively in successive generations. Julia Bell, in her extensive compilation of myotonic dystrophy families, noted the phenomenon, which she referred to as 'antedating.' The data of Bell (1947) were used by Penrose (1948) in his analysis. Both Bell and Penrose were aware of a low parent-child correlation. Penrose's conclusion was that anticipation was apparent rather than real and did not require a novel biologic explanation. He failed to consider the possibility that low parent-child correlation might itself be the result of anticipation.

In the days long before the gene was identified, it was feasible to perform amniocentesis in selected families to determine secretor status of the fetus and thereby predict inheritance of the allele for myotonic dystrophy based upon the DM-Se linkage. The affected spouse had to be heterozygous at the secretor locus and the linkage phase between DM and Se must be established; the unaffected spouse must not be homozygous secretor-positive. It is best if that spouse is secretor-negative, but useful information for counseling could be obtained if he is heterozygous for secretor. In some cases the secretor phenotype of the fetus could establish the genotype in the parents. Finally, recombination between DM and Se introduced a degree of uncertainty into the counseling (Schrott et al., 1973).

Caughey and Myrianthopoulos (1963) provided a monograph covering all aspects of myotonic dystrophy. Caughey and Myrianthopoulos (1991) privately published a second edition. The frontispiece is a Greek stamp commemorating Prince Ypsilante, a hero of Greek liberation who, along with his brother, was thought on good evidence to have had myotonic dystrophy.

Cattaino and Vicario (1999) suggested that Amenhotep IV, better known as Akhenaten, the heretical pharaoh, a king of the New Kingdom of Ancient Egypt, had myotonic dystrophy. Statues and reliefs of him showed abnormal features. He died at the age of about 36 years, without a male heir, although he had had 6 daughters by his principal wife. Perhaps because of religious reform, figurative art abandoned the classic style that had been almost immutable over the centuries and had imposed an idealized representation of the pharaoh, always vigorous and physically fit, with regular facial features showing an attitude of seraphic superiority. Surviving images from the time of Akhenaten are very different and have a realism never before seen in Ancient Egypt. Statues of Akhenaten show a long face, with thin and hollow cheeks, a half-open mouth, and lowered eyelids. Others had commented that the extremely long and thin neck reminded them of 'a swan's neck.' One statue shows gynecomastia and small genitals. Several reliefs demonstrate distal hypotrophy of the lower limbs with features of an upside-down bottle, or, as defined by Aldred (1988), of knickerbockers.

Tramonte and Burns (2005) reviewed early descriptions of myotonic dystrophy.