Our understanding of the genetics of myotonic dystrophy has been so entirely changed by the discovery that its mutational basis results from the expansion of an unstable trinucleotide repeat sequence, that it makes little sense to discuss any of the genetic aspects except in this context. However, the early studies, in particular the extensive family surveys of Fleischer,45 Bell,46 Thomasen,47 and Klein,48 were able to identify a series of unusual genetic features that demanded explanation, most of which have been resolved now that we know the mutational mechanism. These included whether autosomal dominant inheritance was an adequate explanation for the genetic basis; the extreme variability of the disorder within single families; parental origin effects in transmission of the disease, in particular the maternal transmission of congenital myotonic dystrophy; and, most notably, the phenomenon of anticipation, which required a period of 75 years from its first description in myotonic dystrophy to a satisfactory explanation for its occurrence. These aspects, summarized in Table 217-7, are considered in turn, taking the evidence from both classical formal genetic studies and from molecular analysis together. But first, the steps leading to the identification of the myotonic dystrophy gene and mutation are outlined.
Table 217-7: Inheritance of Myotonic Dystrophy-Summary of Unusual Features Requiring Explanation |Favorite Table|Download (.pdf) Table 217-7: Inheritance of Myotonic Dystrophy-Summary of Unusual Features Requiring Explanation
|Autosomal dominant inheritance-but extreme variation in expression despite lack of genetic heterogeneity. |
|Anticipation-progressively earlier onset and greater severity with successive generations. |
|Parent of origin effects-congenital form almost always maternally transmitted; first symptomatic generation usually of paternal origin. |
|Population variation-virtual absence in sub-Saharan Africa; isolates elsewhere with extremely high prevalence. |
Positional Cloning of the Myotonic Dystrophy Gene
The prolonged and eventually successful gene mapping and positional cloning studies of the myotonic dystrophy region of chromosome 19 were described in detail in the previous edition of this work49 and can now be considered as historical. Briefly, one can identify several phases of this work.
Early linkage studies with classical markers, allowing formation of a group of linked loci (myotonic dystrophy—Lutheran blood group—secretor locus).50–52
Assignment of the linkage group to chromosome 19.53
Detection of closely linked DNA markers54–56 and physical mapping57,58 of the relevant region of chromosome 19q.
Finding of markers showing linkage disequilibrium outside specific populations59 and allowing detailed molecular analysis of the critical region.60–62
Discovery of the unstable DNA sequence and its trinucleotide repeat nature as the mutational basis for myotonic dystrophy.
Isolation of the myotonic dystrophy protein kinase gene.
Identification of the overlapping myotonic dystrophy associated homeodomain protein [now known as SIX5 in the databases] (DMAHP) gene (see App. I, Chap. 1) and recognition that the trinucleotide repeat mutation affects its promoter region.
In addition to allowing the discovery of the myotonic dystrophy mutation and isolation of the gene, these positional cloning studies also established several important facts that are of general importance in understanding the genetic basis of the disorder. A significant heterogeneity in terms of multiple loci was excluded by the absence of families unlinked to the chromosome 19 locus, while the existence of a common haplotype of markers in most European, Israeli, and Japanese families59 first suggested that there might be very few original mutations. Also, the general studies of recombination along the chromosome showed that there was no extensive region of genetic instability in the neighborhood of the myotonic dystrophy gene.63
The disequilibrium data strongly suggested that the gene was located in a restricted region of around 200 kb.59 This region was also suggested by individual recombinants, and the combined evidence prompted a detailed molecular analysis of this entire length of DNA. By mid-1991 this region had essentially been cloned,61,62 and the problem was how to identify the myotonic dystrophy gene from the considerable number of genes (estimated at 10 to 20) likely to be present in this region.
Linkage disequilibrium played a further role in focusing attention on the correct region when a genomic sequence (D19S190; 59A) was identified64 that showed almost complete disequilibrium with myotonic dystrophy—74 of 75 unrelated affected patients had the same allele, in contrast to 140 of 232 normal individuals (the single apparent exception later proved to have PROMM). Because this genomic sequence was strongly conserved, it became an important candidate for the myotonic dystrophy gene.
At the end of 1991, the course of the search was radically altered by the finding that a specific abnormality was present in the DNA of individuals with myotonic dystrophy. The abnormality was seen both in a cDNA sequence65 and in the genomic sequence 59A described above.64 Exchange of materials by the two groups involved showed that both probes were detecting the same change, and that the abnormality, while specific to myotonic dystrophy, appeared to vary among patients, even within the same family. It was immediately recognized that this abnormality was likely to represent expansion of an unstable DNA sequence and that the sequence was probably within the myotonic dystrophy gene itself. The findings, together with those of a third group,66 were published in Nature in February 1992, thus representing the successful conclusion of almost a decade of gene mapping and positional cloning studies since the original localization of the gene to chromosome 19. Table 217-8 summarizes the main properties of the unstable trinucleotide repeat sequence, which is discussed in more detail below.
Table 217-8: The Myotonic Dystrophy Unstable Trinucleotide Repeat: Summary of Principal Features |Favorite Table|Download (.pdf) Table 217-8: The Myotonic Dystrophy Unstable Trinucleotide Repeat: Summary of Principal Features
|CTG repeat. |
|Location in 3′ untranslated region of myotonic dystrophy protein kinase gene. |
|Variable in normal population (less than 40 repeats). |
|Expansion in myotonic dystrophy extremely variable (50 to >5000 repeats). |
|Correlation of repeat number with severity of phenotype and early age at onset. |
|Intergenerational instability size and sex dependent, largely explaining observed anticipation and parent of origin effects. |
|Somatic instability, variable between tissues. |
An Expanded Trinucleotide Repeat Sequence in Myotonic Dystrophy
Six months prior to the recognition of the unstable sequence in myotonic dystrophy, the molecular basis of the disorder fragile X mental retardation had been found to be the presence of an expanded and unstable sequence in this gene on the X chromosome.67 Detailed analysis of the fragile X expansion showed this sequence to be a CGG repeat.68 Normal individuals have less than 50 copies of the repeat; clinically normal gene carriers have at least 50 copies; and clinically affected individuals have much larger expansions. The parallels with myotonic dystrophy had already been recognized at a clinical level and the identification of an unstable DNA sequence in the fragile X syndrome made it likely that a similar trinucleotide repeat sequence might be involved in myotonic dystrophy.
As indicated above, the unstable sequence was detected separately by a genomic and a cDNA clone, both of which proved to identify the same specific abnormality. Figure 217-4 illustrates the band pattern seen for the EcoRI polymorphism detected by these sequences in normal and affected individuals. The normal pattern is a two-allele insertion/deletion polymorphism with bands of 9 and 10 kb,69,70 individuals being either heterozygous (lane 5) or homozygous for one or the other band (lane 1). In myotonic dystrophy, one normal allele is seen, but the other (invariably a 10-kb allele) is increased in size to a variable degree.
The unstable DNA sequence in myotonic dystrophy. EcoRI-digested genomic DNA probed with pBB07. All individuals have a constant band of 15 kb (C). Normal individuals are either homozygous (lane 1) or heterozygous (lane 5) for bands of 9 and 10 kb. Affected individuals (lanes 2, 3, 4, 6) have one of these bands, but also show an additional larger band, whose size varies with the individual. (Adapted from Harley et al. 64)
The degree of increase seen in these initial studies ranged from an extra 150 bp to as much as 6 kb additional DNA, as illustrated in Fig. 217-4. Most important, this variation was seen within single kindreds, notably between generations, as shown below in Fig. 217-6. It can immediately be appreciated that the expansion of the sequence in successive generations agrees with the clinical observation of anticipation and with the genetic stability expected from an expanded trinucleotide repeat sequence.
Anticipation and myotonic dystrophy. Clinical features in a three-generation family, showing the increased severity in successive generations. A. Grandfather (age 57). Cataracts removed at age 48. Symptoms of myotonia since age 50, but no significant disability. Ptosis, facial weakness, and myotonia on examination. Myotonic dystrophy not diagnosed until birth of affected grandson. B. Mother (age 24). Myotonia since late teens; weakness of face and neck present on examination, together with marked myotonia. Only diagnosed after birth of affected son (right). No cataract. C. Son. Congenital myotonic dystrophy. Hydramnios during pregnancy; respiratory distress at birth necessitating ventilation. Subsequent improvement, but motor and mental development remains delayed. Marked facial and jaw weakness with hypotonia.
Anticipation and unstable DNA in myotonic dystrophy. DNA samples from the same three-generation family shown in Fig. 217-5 with myotonic dystrophy showing minimal DNA expansion in the mildly affected grandfather, lane 2 (CO.5 kb expansion), moderate expansion in the affected mother with adult onset, lane 3 (C2.5 kb), and large expansion in the congenitally affected son, lane 4 (C4 kb). Lanes 1 and 5 are from normal individuals. (From Harley et al. 64 Used by permission.)
The confirmation of the mutational defect in myotonic dystrophy as a trinucleotide repeat expansion had to await isolation of the gene,71–73 an event that occurred within a few months of the identification of the unstable sequence as a result of two different approaches, and the nature of which is described in more detail later. For the purposes of discussion of the mutation and its relationship to the disease, the relevant facts from these studies are that it is a CTG repeat and that it is located in the 3′ untranslated region of the gene that is now known as the myotonic dystrophy protein kinase gene (DMPK). More recently it was shown that the repeat also coincides with the regulatory sequences of a homeodomain transcription factor gene DMAHP .
Autosomal Dominant Inheritance and Myotonic Dystrophy
It was already noted that myotonic dystrophy was proposed to follow mendelian autosomal dominant inheritance by Fleischer as long ago as 1918,45 and that this was confirmed by the more systematic studies of Bell46 and Thomasen,47 both in 1948. However, the emphasis since then given to the puzzling and anomalous features of inheritance in the disorder makes it necessary to reexamine whether there really is a consistent autosomal dominant foundation.
First, other patterns of mendelian inheritance can be briefly dealt with and disposed of. X-linkage is clearly ruled out by male-to-male transmission (as is mitochondrial inheritance), while autosomal recessive inheritance is made improbable by the lack of consanguinity (although the level of this varies according to the particular population) and by the transmission through successive generations.
The various studies analyzing the segregation of the disorder in families were assessed by Harper5 who found that for both sibs and offspring of propositi, the proportion of affected offspring varied between 33.6 percent and 50.7 percent, suggesting that autosomal dominant inheritance was operating, but with a proportion of undetected gene carriers of up to 16 percent, this probably depending on the age range of the subjects and methods of the study. It should particularly be noted that no study showed a proportion of affected in significant excess of 50 percent; this was, however, strongly dependent on the propositi being removed to correct for ascertainment bias; the early study of Harper74 for example, showed 61.3 percent affected without removing propositi, but only 42.1 percent with it done.
Following the suggestion that there might be segregation distortion and possible meiotic drive for normal polymorphic alleles at the myotonic dystrophy locus (see below), two recent studies have claimed to show anomalous segregation for the disease allele also. Both Gennarelli et al.75 and Zatz et al.76 found a significant excess transmission of the abnormal allele to sons. However, in neither study were propositi distinguished, a factor of especial importance in view of the sex difference, because previous studies5 clearly showed a preferential ascertainment of affected males, again a finding that disappears when propositi are removed.
Since the identification of the CTG repeat at the myotonic dystrophy locus a series of studies has examined the segregation of normal alleles, largely as a means of explaining the generation of potentially unstable alleles to replace those lost as a result of myotonic dystrophy, a topic discussed later. Carey et al.77 found that there was preferential transmission of large repeats (19 and above) by males; however, reanalysis of these data by Hurst et al.78 and by Chakraborty et al.79 both showed that there was an excess of female transmission by heterozygotes of their larger repeat number allele.
Thus, the conclusions from segregation analysis overall are that, so far as myotonic dystrophy as a disease is concerned, the pattern fits that expected for autosomal dominant inheritance, with no evidence of anomaly. For normal variation at the myotonic dystrophy locus there may be preferential transmission of larger repeat size alleles, but further data are needed to be clear about this.
Anticipation has a special place in the genetics of myotonic dystrophy. Not only was it the first disorder for which it was demonstrated on the basis of valid clinical evidence, but it has always been the clearest example of the phenomenon. Full accounts of anticipation in myotonic dystrophy have been given elsewhere,80,81 as well as of its possible role in other disorders;82,83 anticipation is now recognized as a valuable indicator of a possible trinucleotide repeat expansion, but needs to be assessed critically.
Anticipation in myotonic dystrophy was first proposed by Fleischer in 1918,45 on the basis of thorough genealogic and clinical studies in which he not only was able to link families through apparently unaffected individuals, but also to show that there appeared to be progression of the disease in successive generations in terms of age at onset and severity. Although the validity of this anticipation was accepted over a considerable period, the need for a biologic mechanism was questioned by Penrose in 1948,84 who compared a series of different genetic disorders and showed varying degrees of anticipation in most, myotonic dystrophy being the most extreme example.
Penrose pointed out that a combination of ascertainment biases and variability could give the picture of apparent anticipation, and that most examples were merely the result of these biases (almost certainly a correct conclusion). For myotonic dystrophy, he suggested that an allelic modifying gene was an additional factor, together with its inherent variability. This led to the assumption that no particularly unusual biologic mechanism need be sought for anticipation in myotonic dystrophy and the subject was neglected for the next 40 years. By this time, it had become clear that the ascertainment biases were inadequate to explain the extreme anticipation seen in myotonic dystrophy;85,86 the severe congenital form of myotonic dystrophy had been recognized and shown to be maternally transmitted and parallels had been drawn with other disorders such as fragile X mental retardation. Thus the discovery of an unstable trinucleotide repeat expansion as the basis for both fragile X syndrome and myotonic dystrophy provided an immediate explanation for anticipation in these disorders.
The details of anticipation in relation to such factors as size of the expansion and sex of the transmitting parent are considered in the following sections, but the way in which anticipation can now be seen at the molecular as well as the clinical level is illustrated in Figs. 217-5 and 217-6, which both relate to the same family; Fig. 217-7 shows the process in a series of parent-child pairs. It can also be asked: how many generations are required for anticipation to proceed to genetic lethality in a kindred? De Die-Smulders et al.87 examined this in a large Dutch kindred in which myotonic dystrophy had originally been described 40 years previously, and found that anticipation had resulted in almost complete extinction of the gene within five generations, although it had proceeded at a different rate in different branches. Infertility of male patients and mental retardation associated with congenital myotonic dystrophy were the main reasons. Interestingly, and supporting other studies, no carriers of small repeat expansions were found in the fourth and fifth generations. While this is the usual pattern seen in families, it is not invariable, as evidenced by an American kindred88 in which relative stability was observed over three generations, with repeat numbers of 60 to 90 and no significant symptoms in most individuals.
Size of the CTG repeat in affected parent and offspring pairs, showing anticipation at the molecular level, with most extensions larger in the offspring than in the parent.
The initial studies identifying the unstable CTG repeat in myotonic dystrophy immediately identified its variable nature within families, the relationship to anticipation, and an approximate association between degree of severity and size of the expanded sequence. Since then numerous further studies have explored the extent to which genotype and phenotype are correlated, within terms of general severity of disease and age at onset, and also in relation to specific systems clinically involved.
The main clinical categories used have been congenital myotonic dystrophy; childhood onset disease (these two frequently combined); classical adult onset of neurologic disease; and minimal disease (often cataract only) in later life. Some studies have used severity of muscle disease only. Clearly, these groupings are strongly age related, and age at onset has itself been used as an indicator, which has the advantage of being quantifiable even if approximate.
The initial study of Harley et al.89 showed clear, though overlapping differences in expansion size for the different groups studied (Fig. 217-8). All congenital cases showed expansions of 2500 base pairs or greater, while in the minimally affected group all expansions were 250 base pairs or less, giving a complete separation of these two extreme groups. Cases with typical adult disease showed a wide variation of expansion size, overlapping with the other groups. Tsilfidis et al.90 also showed that congenital cases were associated with the largest expansions, as did Novelli et al.,91 Eguchi et al.92 (in a Japanese population), and Passos-Bueno et al.93 (in Brazilian families). For age at onset, Harley et al.94,95 found a strong inverse correlation with expansion size (Fig. 217-9), no sex difference being found.
Size of DNA expansion in myotonic dystrophy in relation to severity of disorder. Minimal: Neuromuscular abnormalities absent or insignificant; cataract is the principal feature. Classical adult: Progressive muscle wasting with myotonia; onset in adult life. Early childhood: Developmental delay and other serious childhood symptoms; onset not congenital. Congenital: Onset of myotonic dystrophy at or before birth. DNA expansion is clearly related to these categories of severity; while individual categories show overlap, there is a distinct separation between the minimal group and those with severe childhood disease. (From Harley et al. 89 Used by permission.)
The relationship of the age of onset of myotonic dystrophy to the size of the CTG repeat sequence. (From Harley et al. 89)
Gennarelli et al.96 have attempted to express these findings in a form that expresses a probability that a given number of CTG repeats will result in any of three categories of severity. While their categories are mainly determined by severity of muscle disease, they approximate to the three more widely used groups of minimal, classical onset neuromuscular, and severe childhood/congenital disease, and confirm the earlier conclusions that a CTG repeat of 100 or less will almost always be associated with minimal disease, while one of 1500 or more will have a very high chance of falling into the severe category. The final section of the chapter discusses those findings useful for genetic counseling.
In addition to these phenotype-genotype studies, a number of investigators have examined the relationship to specific areas of clinical abnormality in myotonic dystrophy, particularly the nonmuscular aspects. These include central nervous system involvement, especially IQ,97 cardiac defects,98 and a variety of endocrine99 and metabolic abnormalities.100–102 Almost all have found correlations comparable to the more general studies and the main conclusion to be reached is that no single system appears to be particularly closely related to repeat number, nor to be independent of it. This finding is of relevance in terms of the possible effects of other neighboring genes in the disease pathology (see below), because it suggests that no clinical feature can be separated out that might solely result from abnormality in another gene, as is often seen in microdeletion syndromes.
While all these studies agree in showing that the myotonic dystrophy phenotype is indeed strongly correlated with genotype as defined by repeat length, they equally show that other factors must also play an important part. The practical limitations in applying the statistical correlation to prognosis in individuals have already been noted and are discussed later in relation to genetic counseling. Twin data suggest that some of the factors are developmental, while Hamshere et al.103 suggested that the phenotype correlation with repeat number may not apply across all ranges and may relate to a threshold of repeat number influencing gene function. Clinically discordant sibs with the same repeat number have been reported,104 while Ashizawa et al.'s105 finding that anticipation might occur clinically even though the repeat number decreased, again indicates other important influences on phenotype. Sampling limitations and the use of different methods for determining repeat length are additional factors that need to be considered. In particular there remain doubts as to whether repeat length alone can explain the unusual clinical features and maternal transmission of congenital myotonic dystrophy.
So far these other possible factors have not been specifically identified, although an influence of the repeat number of the normal allele has been ruled out.106 Likely future candidates might be genes interacting with MDPK, the immediately adjacent gene loci, and other important genes involved in muscle and central nervous system development and repair.
Genetic Instability of the CTG Repeat: Intergenerational Aspects
The initial studies of the expanded CTG repeat in myotonic dystrophy all showed striking variation in repeat number between different individuals, even in the same family, raising immediate questions not only about the relationship with disease phenotype, but concerning the previously puzzling phenomenon of anticipation, already discussed, and the parental origin effects, especially in relation to the maternal transmission of congenital myotonic dystrophy. These aspects are examined in turn.
Parent of Origin Effects.
Prior to discovery of the mutation, parent of origin effects in the transmission of myotonic dystrophy were well established, as already noted. In particular, the congenital form of the disorder was known almost always to be transmitted maternally, while there was a suspicion of a predominant male excess of asymptomatic or minimally affected individuals in the earliest known generation of families. A series of studies has now examined these aspects in detail and the following conclusions seem clear.
Congenital myotonic dystrophy, associated with the largest expansions, is indeed transmitted maternally in the overwhelming majority of cases,23,24 but may occasionally be transmitted paternally. Ten such cases have been reported;107–110 the clinical features of the offspring are comparable to other congenital onset patients, but the affected father does not show a tendency to particularly severe disease or early onset (in contrast to the situation for juvenile Huntington disease).
The majority of parent-child transmissions where the repeat size decreases are male,105 as are the occasional instances where reversion of the expansion to the normal range has occurred.111,112 This suggests some mechanism that limits continued expansion of the repeat in male transmission of the repeat in male transmission that is not operating in female transmission. Because studies of sperm from male patients show a wide range of expansion sizes, there may be some selective disadvantage in fertilization by such sperm.
For smaller-sized repeats (less than 100 kb) expansion is greater for male transmissions than for female.113,114
Over the entire range of repeat lengths, there is no clear sex difference in expansion between the generations. Some studies have suggested a greater tendency to expansion in female transmission,115,116 while Brunner et al.113 found greater expansion in male transmissions overall. Harley et al.95 found no significant sex difference when the increase was expressed as a proportion of parental repeat size. It is likely that the variable results reflect interaction of the opposite parental effects involving small and large expansions.
The predominantly male origin of the disorder in the earlier generations of families has also been confirmed in a series of studies. Harley et al. showed a significant excess of male grandparents of congenital cases, as did Lopez de Munain et al.,117 while Brunner et al.113 found that genealogic links between families studied predominantly involved males. Their study also showed an excess of male transmission in the initial symptomatic generation. When combined with the results of previous studies, notably those from the Quebec kindred, this male excess was significant, whereas for previous asymptomatic transmissions there was no significant sex difference.
At first sight, the parent of origin effects may appear confusing and in contrast to what is seen in other trinucleotide disorders, in particular Huntington disease, where severe juvenile cases are mainly paternally transmitted.118 In fact, when the size-dependent nature of the parents affects is taken into consideration, any apparent conflict resolves,119 because the size range of repeats in severe cases of the polyglutamine (CAG) repeat disorders is comparable to those of minimally affected myotonic dystrophy patients. Thus, the pattern in this expansion range for all the disorders seems to be of greater instability in male transmissions, while the unique feature of myotonic dystrophy is the occurrence and viability of individuals with exceptionally large expansions, where the mainly female transmission appears to result from some selective disadvantage of sperm carrying large expansions, probably together with increasing male infertility in this range.
Expansion Size and Genetic Instability.
Along with the parental origin the main other factor identified as influencing instability in transmission between generations is the size of the expansion itself. A series of studies has shown that not only does the size of the expansion in the offspring correlate with that of the parent, but that it is related also to the increase between the generations, with an accelerating effect.95,113–115,120,121 Clearly, this is an important factor underlying the clinical observation of anticipation and it is clear that this trend is already established for the range of expansion size giving clinical manifestations. The variation in rate of progression in different kindreds or branches of a kindred has already been mentioned.110,111 What is less clear is the lower limit of significant instability—a point of practical importance in assessing risks for offspring when an individual with a borderline expansion is detected incidentally or as part of an extended family.
Genetic Instability: Developmental and Somatic Aspects
Studies on the intergenerational variation of the myotonic dystrophy gene mutation have principally been based on comparison of the CTG expansion in blood lymphocytes of affected adults. Such studies are considerably removed from the germ line cells that are the basis for any intergenerational effects, while they give no indication as to the developmental timing of the instability that is seen. In addition, lymphocytes are not necessarily representative of other body tissues, and are certainly not the cells most closely related to the clinical abnormalities of the disorder.
Fortunately, there is now evidence available on all these aspects of genetic instability that gives us a clearer picture of the dynamics of the processes involved.
While the ovum remains largely inaccessible to molecular studies, analysis of sperm allows direct analysis of the male germ line, although the azoospermia seen in many patients has been a limiting factor. Jansen et al.122 compared sperm and blood in 24 male myotonic dystrophy patients and found that in four of nine cases, the CTG expansion was less in sperm than in blood. They also found that when the repeat length in the blood of offspring was compared to that of paternal sperm, there was an increase in six of eight cases, suggesting a significant postmeiotic contribution to the instability in the offspring. Monckton et al.123 developed a technique for small-pool PCR analysis of sperm and other cells that enabled more detailed study and gave rather different results. In two of three males studied, there was a striking increase in the range of repeat sizes seen in sperm, including some extending down to the high normal repeat size range, even though the mean repeat size was greater in sperm than in blood. A third patient with a small expansion in blood (76 repeats) showed a greatly increased expansion in sperm (254 repeats) with none in the normal range.
While these data are still relatively few, they fit well with the clinical and overall genetic evidence on male intergenerational transmission, in particular with the greater tendency of the repeat to expand in transmission from mildly affected males with small repeats, and also with the observation that reduction in repeat size is a feature of male transmission. It is of interest whether data can be obtained from ova of affected females that will enable comparison.
Instability and Development.
Opportunities for studying instability of the myotonic dystrophy repeat during early embryonic development are limited, but the advent of preimplantation diagnosis has shown that expansions can be detected at the blastomere stage,124 as discussed later under “Molecular Diagnostics.” However, because abnormal embryos are not implanted during IVF, this does not provide information on how the expansion might alter subsequently.
Early prenatal diagnosis using chorion villus sampling at around 10-week gestation is now an established procedure and there are several reports on the identification of fetuses with expansion of the myotonic dystrophy repeat,125–127 including some with the very large repeats likely to have resulted in congenital myotonic dystrophy. One relevant observation on such samples is that they show little evidence of somatic variation, despite the very large number of repeats, Southern blotting showing a clear-cut band in contrast to the “smear” commonly seen in samples taken from such patients in later life.
A more detailed study of the repeat in early development was done by Worhle et al.,126 who analyzed a range of tissues from a 13- and a 16-week-old affected fetus, both with comparable size of expansions of 4 to 6 kb. The first showed no variation between different fetal tissues, whereas in the second, there was considerable variation of repeat size. Martorell et al.128 have taken this approach further, analyzing 7 cases of their own along with 10 previously reported cases at various stages of development from 10 weeks to the neonatal period. They confirmed that no tissue variability was seen before 13 weeks gestation, but that this was established by 16 weeks, showing no clear correlation with size of repeat.
An entirely different approach to the topic of developmental instability of the repeat sequence has come from the study of identical twins.129–131 Five such pairs were studied by molecular techniques; clinical features were extremely similar between the twins in all cases, supporting earlier reports from before isolation of the gene. In three pairs, the expansion of the repeat sequence also appeared to be identical, but in two pairs there were considerable differences in repeat size in blood, suggesting developmental instability.
It is relevant to compare the findings on developmental instability with the situation for fragile X mental retardation, where comparably large repeat expansions are also seen. Here, no tissue differences were found during fetal development132 and the repeat size in identical twins was the same.133 A further difference has been in finding that affected fragile X males with a full expansion show mainly premutation ranges of repeat size in gonadal tissue.134 This suggests that for fragile X, in contrast with myotonic dystrophy, instability is mainly prezygotic.
Somatic Instability in Later Life.
This has been extensively studied, the main approaches being the analysis of a wide range of tissues at autopsy, the detailed comparison of muscle with blood, and the dynamics of cultured cells.
The autopsy studies on adults122,136 have shown a wide variation in repeat numbers between tissues, this occurring to a greater degree than that already mentioned in relation to the fetal and neonatal cases. Most tissues show a greater degree of repeat expansion than seen in blood, but a consistent pattern has not emerged. One study137 reported a smaller expansion in cerebellum than in other parts of brain and other tissues, a finding that has also been noted for Huntington disease.138
The repeat expansion in muscle has received particular attention, partly because of availability of biopsy material, partly because of its relevance to the disease process. It might have been thought likely that the degree of expansion in muscle would show closer correlation with severity of muscle disease than does the expansion in lymphocytes, but this has not proved to be the case. A series of studies139–142 showed that muscle expansions are larger than in blood from the same patient, that there is considerable variation between sites, and that the correlation with severity is poorer than using blood.
The time course of genetic instability in later life was approached by studying samples (usually blood) taken at intervals. Martorell et al.143 showed that the repeat number increased over a 5-year interval for patients with a wide range of repeat numbers, although this increase was not seen in individuals who were asymptomatic when the first sample was taken.
It is now becoming possible to study the detailed cell dynamics of genetic instability, using small-pool and single-cell PCR techniques,123,144 and by studying cultured cells.126,144 Monckton et al.123 showed that a “smear” of repeat lengths seen in the blood of myotonic dystrophy patients could be resolved into a series of separate cell lines, each showing expansion from a specific starting point. There was no evidence of reduction in repeat length, agreeing with the earlier observation of a sharp lower size border of bands observed on Southern blotting and in contrast with the findings on sperm, studied by the same techniques and already mentioned. Worhle et al.126 analyzed cultured cells from fetal tissue and showed a progressive increase in repeat length with cell proliferation following a sigmoid time course. Cell culture studies are also proving of particular importance in elucidating the mechanisms of genetic instability.
While our understanding of the genetics of myotonic dystrophy has been radically changed by the recognition of genetic instability as the underlying basis for mutation, molecular studies have also shed light on a range of other aspects that need brief noting.
The recognition that homozygotes for Huntington disease exist and that they show no greater severity than those heterozygous for the mutation145 has given relevance to assessing the situation for comparable homozygotes in myotonic dystrophy. Surprisingly, the first two detected by molecular studies were asymptomatic, carrying two copies of small expansions.146 Six cases have now been reported,147–149 only one of which appears to be more severely affected than would be expected from expansion size.147 Because this case resulted from an incestuous mating, other loci may have been responsible for part of the phenotype and the overall conclusion so far is that, as for Huntington disease, homozygosity does not affect the clinical phenotype. This is extremely relevant to the mechanism of gene action and would suggest that simple haploinsufficiency is not an adequate explanation for causation of the disease.
The involvement of more than one locus was never likely except as an extremely rare event because linkage studies over several decades showed no evidence of families clearly unlinked to the locus established on chromosome 19. This was confirmed when the unstable repeat was identified and found to be responsible for the great majority of cases. However, when the very small number of individuals not showing such a repeat expansion was studied closely, and diagnostic and technical errors discounted, it became clear that a small group of patients showed a clinical picture that was distinguishable from, although similar to, myotonic dystrophy. This disorder, now known as proximal myotonic myopathy (PROMM), was discussed earlier and may prove of considerable importance in relation to myotonic dystrophy once we know more about its molecular basis.
After PROMM families have been removed, there still remains a handful of patients whose phenotype is indistinguishable from myotonic dystrophy but who show no repeat expansion. None of these patients has so far shown any other defect in the myotonic dystrophy protein kinase or neighboring genes, but one large family20 was recently shown to map to chromosome 3q, representing the first separate myotonic dystrophy locus. Careful clinical documentation150 shows the classical muscle and systemic features of myotonic dystrophy, but no clear evidence of anticipation; there is no evidence so far that a trinucleotide repeat expansion is involved. What the nature of the specific gene and protein is and the extent to which this locus is involved in PROMM families are of great interest. Rare families such as this could provide the key to the specific steps in pathogenesis of myotonic dystrophy.
The immense range of severity and age of onset in myotonic dystrophy makes this an extremely difficult area to study accurately. Reproduction of survivors with congenital onset is rare and many congenital cases are fatal in the neonatal period, making this form genetically lethal. At the other extreme, many patients have minimal disease with onset late in life after reproduction is complete, and such individuals often do not come to medical attention. Asymptomatic individuals now being detected who carry the mutation further complicate the situation. It can be imagined that results of studies may well depend on the different proportions of these groups that are represented in the series.
Most of the older studies, summarized by Harper,5 showed a decreased fertility in both sexes, with an overall relative fertility two-thirds of normal. Male fertility was reduced in most series more than female, but this was largely due to a high proportion of males failing to marry. The large Quebec isolate has been the subject of particular study and offers a unique opportunity of recording fertility of earlier generations because the genealogy is completely known for the preceding 10 generations.151 An initial study by Veillette et al.,152 based mainly on such marriages, showed reduced fertility comparable with previous series, but when older historical periods were analyzed, no difference in fertility was found from controls in each time period.153 It is likely that the majority of individuals included in this second study would have would have been symptomless carriers of the mutation, reinforcing the view that fertility is only likely to be impaired in those individuals with clinically significant disease.
A further factor relevant to fertility is the increased frequency of spontaneous early pregnancy loss in affected women,154,155 something distinct from perinatal deaths due to congenital myotonic dystrophy. However, such early losses are frequently replaced and there is no evidence that they preferentially involve pregnancies carrying the myotonic dystrophy mutation.
It can be seen that analysis of fertility in myotonic dystrophy is complex and difficult to interpret; it is not only of practical relevance, however, but of theoretical importance when considering how the mutation is maintained in the face of progressive and ultimately lethal instability, a topic that forms part of the population genetic studies on the disorder, which now require discussion.