Final Report Summary - DM AND CHROMATIN (Impact of the chromatin environment on the CTG repeat instability involved in myotonic dystrophy type 1)
In 1991, the first two unstable and expanded trinucleotide repeats (TNRs) were identified as the genetic cause of neurological diseases. These TNRs form an unusual group of mutations, because the repeat unit number can change and varies between individuals. The mutation rate is higher if the repeat is longer. If a TNR tract is longer, the risk of developing these disorders is higher. Furthermore, age of onset is often lower and the severity of the disorder tends to get worse with increasing repeat length. Instability towards longer repeats is seen both upon transmission to the next generation, and in body cells during life. Thus, successive generations are at higher risk of developing the disease.
Myotonic dystrophy type 1 (DM1) is a TNR disease, caused by an expanded CTG repeat located in the DMPK gene. DM1 is the most common form of adult muscular dystrophy. It is a truly multisystemic disorder, involving muscles deterioration, heart problems, neurological dysfunction, insulin resistance, cataracts and several other systems. The DM1-associated CTG repeat is specifically unstable, generally causing the clinical picture to become more severe in every generation.
To eventually be able to slow down progression of the disease, we strive to understand the mechanisms that cause repeat instability and what the effects of the CTG repeat are at a molecular level and how this leads to clinical symptoms. Although significant advances have been made over the past years, much remains.
This project aimed to advance our understanding of the effect of the unstable behaviour of the CTG repeat on the chromatin environment. Chromatin refers to the combination of specific proteins around which the deoxyribonucleic acid (DNA) is wrapped in the nucleus. Depending on the type of chemical group that can be attached to either these histone proteins or the DNA, the chromatin can adopt different conformations. Some conformations allow easy access to transcription factors: proteins that induce the transcription of a gene into ribonucleic acid (RNA) molecules, which can then be translated into a protein. Alternatively, genes that are not actively transcribed because their resulting proteins are not needed in the cell at that time, have a conformation that is more difficult to reach for transcription factors.
A mouse model for DM1 was previously developed in the host laboratory, which mimics both clinical and genetic aspects of DM1. For this project, this mouse model was used to gain more insight into chromatin dynamics in the presence of an expanded CTG repeat (500 or 1 400 CTGs), as compared to a normal, short (20 CTGs) repeat. Not only did I test the binding of specific chemical groups called histone modifications to the DNA close to the CTG repeat in adult hearts of these mice, I also studied binding of the transcription factor CTCF to two known binding sites located on each side of the CTG repeat. Furthermore, I looked into attachment of methyl groups to the 'C' nucleotides of the DNA around the CTG repeat. This is called DNA methylation and is known to generally suppress transcription activity. At the same time I measured the RNA levels of the genes located around the CTG repeat, to see if the transcriptional activity at this genetic locus matches the chromatin conformation.
I found that DNA methylation at the two CTCF binding sites (CTCFbs) is very low in mice with 20 CTGs, but quite prominent in mice with 500 CTGs and abundant in mice with 1 400 CTGs. I used a method with which the methylation pattern of DNA in individual cells can be analysed and I found that methylation was very different from one cell to another. This was a surprising finding, because colleagues hadn't looked in this detail, thus they tended to think in more general terms about absence or presence of CpG methylation.
It was known already that binding of the transcription factor CTCF may be inhibited by DNA methylation at its binding site, but nobody had thus far succeeded in analysing CTCF binding to the DM1 genetic locus in human or mouse tissue. I managed to do this and found that CTCF still binds at one of the CTCFbs, like in mice with 20 CTGs, despite the expanded CTG repeat and despite the methylation. Binding of CTCF around the CTG repeat may prevent regulatory DNA regions that lie in nearby genes to act on the DMPK gene. This could change DMPK transcription levels. But since I still saw CTCF binding in mice with expanded CTG repeats, it seemed like the effect of those regulatory regions on the DMPK gene had not changed. But, when we measured expression levels of the DM1 locus, we found that they were lower when the CTG repeat was expanded. So, we wondered, if not the binding of CTCF, what could cause the expression levels to change? Methylated C-nucleotides can attract chromatin-remodelling enzymes, which can then change the conformation of chromatin. I found that a specific histone modification that is known to be found in actively transcribed chromatin was less bound to DNA carrying expanded CTG repeats, as compared to the control tissue with 20 CTGs. Another histone modification known to be associated with silent chromatin was more often bound to the DNA with an expanded CTG repeat. Thus, overall we saw a less active chromatin conformation in hearts of mice with expanded CTG repeats, as compared to control-length CTG repeats, which was confirmed by lower DM1 expression levels. We generally observed more pronounced effects at CTG repeat lengths of about 1 400 CTGs than at 500 CTGs.
As a preliminary explanation as to why expanded CTG repeats induce chromatin remodelling and altered expression levels at the DM1 locus, we identified binding of PCNA very close to or at the CTG repeat, but not a bit further away. PCNA is a protein that is involved in many cellular functions. For our study, it is mostly relevant that it recruits many chromatin-remodelling enzymes. It has been shown that expanded CTG repeats, but not short control repeats, can serve as binding sites for PCNA. Although chromatin changes associated with expanded TNRs had been demonstrated before, to my knowledge this is the first study that indicates where the process might start.
Understanding the effects of CTG repeat expansion on cellular processes will provide new strategies for therapy. Current therapies are dissatisfactory. Comparing chromatin dynamics in tissues with different repeat lengths, and later in tissues exhibiting different instability rates, will allow to identify circumstances that favour instability or that are associated with stable repeat tracts. The ultimate goal is to halt disease progression by limiting repeat expansion. This will not only improve the quality of life of the patients, but also maintain their contribution to society and limit medical costs. Since the different TNR disorders share genetic properties, improvements in our understanding of any given disorder will most likely also have major implications for unravelling mechanisms underlying other TNR expansions. Worldwide, 1 in 8 000 people get diagnosed with DM1, amounting to over 60 000 people in the European Community. Taking together all TNR diseases, many more people will benefit from the scientific advances made in the field of TNR disorders, including this study.
Myotonic dystrophy type 1 (DM1) is a TNR disease, caused by an expanded CTG repeat located in the DMPK gene. DM1 is the most common form of adult muscular dystrophy. It is a truly multisystemic disorder, involving muscles deterioration, heart problems, neurological dysfunction, insulin resistance, cataracts and several other systems. The DM1-associated CTG repeat is specifically unstable, generally causing the clinical picture to become more severe in every generation.
To eventually be able to slow down progression of the disease, we strive to understand the mechanisms that cause repeat instability and what the effects of the CTG repeat are at a molecular level and how this leads to clinical symptoms. Although significant advances have been made over the past years, much remains.
This project aimed to advance our understanding of the effect of the unstable behaviour of the CTG repeat on the chromatin environment. Chromatin refers to the combination of specific proteins around which the deoxyribonucleic acid (DNA) is wrapped in the nucleus. Depending on the type of chemical group that can be attached to either these histone proteins or the DNA, the chromatin can adopt different conformations. Some conformations allow easy access to transcription factors: proteins that induce the transcription of a gene into ribonucleic acid (RNA) molecules, which can then be translated into a protein. Alternatively, genes that are not actively transcribed because their resulting proteins are not needed in the cell at that time, have a conformation that is more difficult to reach for transcription factors.
A mouse model for DM1 was previously developed in the host laboratory, which mimics both clinical and genetic aspects of DM1. For this project, this mouse model was used to gain more insight into chromatin dynamics in the presence of an expanded CTG repeat (500 or 1 400 CTGs), as compared to a normal, short (20 CTGs) repeat. Not only did I test the binding of specific chemical groups called histone modifications to the DNA close to the CTG repeat in adult hearts of these mice, I also studied binding of the transcription factor CTCF to two known binding sites located on each side of the CTG repeat. Furthermore, I looked into attachment of methyl groups to the 'C' nucleotides of the DNA around the CTG repeat. This is called DNA methylation and is known to generally suppress transcription activity. At the same time I measured the RNA levels of the genes located around the CTG repeat, to see if the transcriptional activity at this genetic locus matches the chromatin conformation.
I found that DNA methylation at the two CTCF binding sites (CTCFbs) is very low in mice with 20 CTGs, but quite prominent in mice with 500 CTGs and abundant in mice with 1 400 CTGs. I used a method with which the methylation pattern of DNA in individual cells can be analysed and I found that methylation was very different from one cell to another. This was a surprising finding, because colleagues hadn't looked in this detail, thus they tended to think in more general terms about absence or presence of CpG methylation.
It was known already that binding of the transcription factor CTCF may be inhibited by DNA methylation at its binding site, but nobody had thus far succeeded in analysing CTCF binding to the DM1 genetic locus in human or mouse tissue. I managed to do this and found that CTCF still binds at one of the CTCFbs, like in mice with 20 CTGs, despite the expanded CTG repeat and despite the methylation. Binding of CTCF around the CTG repeat may prevent regulatory DNA regions that lie in nearby genes to act on the DMPK gene. This could change DMPK transcription levels. But since I still saw CTCF binding in mice with expanded CTG repeats, it seemed like the effect of those regulatory regions on the DMPK gene had not changed. But, when we measured expression levels of the DM1 locus, we found that they were lower when the CTG repeat was expanded. So, we wondered, if not the binding of CTCF, what could cause the expression levels to change? Methylated C-nucleotides can attract chromatin-remodelling enzymes, which can then change the conformation of chromatin. I found that a specific histone modification that is known to be found in actively transcribed chromatin was less bound to DNA carrying expanded CTG repeats, as compared to the control tissue with 20 CTGs. Another histone modification known to be associated with silent chromatin was more often bound to the DNA with an expanded CTG repeat. Thus, overall we saw a less active chromatin conformation in hearts of mice with expanded CTG repeats, as compared to control-length CTG repeats, which was confirmed by lower DM1 expression levels. We generally observed more pronounced effects at CTG repeat lengths of about 1 400 CTGs than at 500 CTGs.
As a preliminary explanation as to why expanded CTG repeats induce chromatin remodelling and altered expression levels at the DM1 locus, we identified binding of PCNA very close to or at the CTG repeat, but not a bit further away. PCNA is a protein that is involved in many cellular functions. For our study, it is mostly relevant that it recruits many chromatin-remodelling enzymes. It has been shown that expanded CTG repeats, but not short control repeats, can serve as binding sites for PCNA. Although chromatin changes associated with expanded TNRs had been demonstrated before, to my knowledge this is the first study that indicates where the process might start.
Understanding the effects of CTG repeat expansion on cellular processes will provide new strategies for therapy. Current therapies are dissatisfactory. Comparing chromatin dynamics in tissues with different repeat lengths, and later in tissues exhibiting different instability rates, will allow to identify circumstances that favour instability or that are associated with stable repeat tracts. The ultimate goal is to halt disease progression by limiting repeat expansion. This will not only improve the quality of life of the patients, but also maintain their contribution to society and limit medical costs. Since the different TNR disorders share genetic properties, improvements in our understanding of any given disorder will most likely also have major implications for unravelling mechanisms underlying other TNR expansions. Worldwide, 1 in 8 000 people get diagnosed with DM1, amounting to over 60 000 people in the European Community. Taking together all TNR diseases, many more people will benefit from the scientific advances made in the field of TNR disorders, including this study.