MDF is pleased to welcome Dr. Laura Ranum, Ph.D., to its Scientific Advisory Committee (SAC). Dr. Ranum, who joined the SAC in summer 2015, is an internationally known investigator of disorders that result from repeat expansion mutations, such as those that cause type 1 and type 2 myotonic dystrophy (DM1 and DM2). More recently, she discovered that repeat expansion mutations can produce unexpected RNAs and proteins. These discoveries were big surprises and have changed the understanding of how expansion mutations work.
Dr. Ranum received her Ph.D. in cell biology from the University of Minnesota in 1989 and carried out postdoctoral studies with Dr. Harry Orr in the Department of Laboratory Medicine and Pathology at that institution. In 1994, she joined the faculty of the University of Minnesota Department of Neurology as an assistant professor.
From 2003 to 2010 Dr. Ranum was a professor in the Department of Genetics, Cell Biology and Development at the University of Minnesota and the research director of the Paul and Sheila Wellstone Muscular Dystrophy Center.
In 2010 She moved to the University of Florida in Gainesville, to become the founding director of the Center for NeuroGenetics and a professor of Molecular Genetics and Microbiology and Neurology in the university’s College of Medicine. We spoke with Dr. Ranum in October 2015:
MDF: When did you first become interested in nucleotide repeat expansion disorders?
LR: I was working on spinocerebellar ataxia type 1 – SCA1 – as a postdoc in Harry Orr’s lab at the University of Minnesota. In 1993, we identified a CAG repeat expansion mutation as the cause of SCA1. That came on the heels of the 1992 identification of the CTG repeat expansion [on chromosome 19] as the cause of type 1 myotonic dystrophy -- DM1. DM1 was the third repeat expansion disorder to be identified. SCA1 was the fifth, after the triplet repeat identification in Huntington’s disease [HD], which was also in 1993.
It was all very exciting. We had suspected that SCA1 could be an expansion mutation, and it was. It was a very electric time.
The first discussions of spinal-bulbar muscular atrophy [SBMA] and fragile X syndrome [two other triplet repeat expansion disorders] started to open up the possibility that variability in neurological diseases such as DM1 and SCA1 could be explained by the length of the repeat expansion.
Curiously, the triplet repeat expansions for HD and SCA1, which are both dominant diseases, were in protein coding regions. But the expansion in DM1, which is also a dominant disease, was in an untranslated region of the gene, so it didn’t fit the same pattern.
MDF: When did you first become involved with the search for a second DM locus?
LR: That was probably in 1995 or so, shortly after I started as a faculty member in the Department of Neurology at the University of Minnesota. Dr. John Day, a neurologist there, was seeing a family in the clinic with DM but without the expected CTG expansion. John knew I had worked on SCA1, and we decided to work together [to identify the cause of this family’s disorder].
We began by collecting blood samples from a large family in Minnesota and mapped that gene to chromosome 3 in 1998. At a meeting in Naarden in the Netherlands, we began a collaboration with Dr. Kenneth Ricker, who was one of the first investigators to describe families from Germany who had DM but without the expected CTG expansion. I was interested in studying DNA samples from the additional families Dr. Ricker had collected because I thought that this would help us pinpoint and find the DM2 gene. So we expanded the project, with Dr. Ricker, to Germany.
In 2001, we published the identification of the gene defect in type 2 myotonic dystrophy – DM2. [The paper, Liquori et al., Science, Aug. 3, 2001, described the identification of the DM2 gene as a CCTG repeat in an intron of the ZNF9 gene on chromosome 3.] We were very excited, and the scientific community as a whole was excited, that a CCTG tetranucleotide repeat located in an intron had been found to underlie DM2. The interpretation was that, because the DM1 and DM2 diseases were so similar and both were expansions in noncoding regions, these were both RNA gain-of-function disorders.
That was the interpretation at the time. We now have a collaborative Program Project grant [from the National Institutes of Health] to investigate RNA mechanisms and also to test if other mechanisms might be involved. Investigators on this multi-investigator grant include me, Dr. Maurice Swanson at the University of Florida, Dr. John Day at Stanford University, and Drs.Timothy Ebner and Brent Clark at the University of Minnesota.
MDF: So that 2001 paper gave a lot of support to the RNA toxic gain-of-function hypothesis for both forms of DM?
LR: Yes, especially after the findings that the CUGBP1 protein and the MBNL protein were affected by the DM1 and DM2 RNA expansions and that they in turn affected splicing of various genes.
MDF: When did you first begin to think about bidirectional transcription of expanded repeats?
LR: I had continued to work on SCA genes, and in 1999, we published that an untranslated CTG expansion underlies spinocerebellar ataxia type 8 -- SCA8. [See Koob et al., Nature Genetics, April 1999.] It was a controversial discovery, because the expansion mutation did not cause disease in all people who carried it.
Eventually, we developed a mouse model of SCA8 with the untranslated CTG expansion. The mice developed SCA8 disease features, but when we investigated the details, we found an unexpected result: In addition to the CUG expansion, there was a CAG expansion RNA made in the opposite direction, and it was translated into a polyglutamine expansion protein that accumulated in both our mouse model and human autopsy brains. We published that in 2006. [See Moseley et al., Nature Genetics, July 2006.]
MDF: When did you first start to think about translation without the usual start codon?
LR: This discovery also came from our SCA8 work and was completely unexpected. To understand the contribution of the polyglutamine protein versus the CUG RNA in SCA8, we removed the ATG initiation codon. Surprisingly, we found that mutating the only ATG start codon did not, as had been expected, prevent polyglutamine from being made. At the time, the ATG initiation codon was thought to be essential to start translation and also that this start signal would ensure only one protein would be made.
To make a long story short, we discovered in studies of cells that CUG and CAG expansion RNAs can make three different repeat expansion proteins each – meaning that, if a disease-causing expansion mutation expresses RNAs in both directions, up to six additional, unexpected proteins could be made. We named this novel type of protein production repeat-associated non-ATG translation, or RAN translation.
So, in SCA8, first we found a CUG expansion RNA. Then we found the gene mutation ran in both directions, producing CUG and CAG RNAs in both directions; and later, that each RNA can make three mutant proteins. It got really exciting, as we showed that these unexpected RAN proteins are made in mice and also in patients.
Initially we reported the discovery of RAN translation in SCA8 and in DM1. [See Zu et al., Proceedings of the National Academy of Sciences USA, Jan. 4, 2011.] We and others have now found RAN proteins in a growing number of additional expansion diseases including Huntington’s Disease [HD], C9ORF72 amyotrophic lateral sclerosis [ALS] and frontotemporal dementia [FTD], and in fragile X tremor ataxia syndrome [FXTAS].
Now we are investigating the impact of RAN translation in DM1 and DM2. We know DM1 can express a polyglutamine protein, but a lot of questions remain. How frequently is the polyglutamine protein found in patients? Are other RAN proteins also found in DM1? Does RAN translation also occur in DM2? Do RAN proteins substantially contribute to DM1 and DM2? We don’t yet know, but we’re working hard on these questions.
MDF: Most therapy development in DM1 centers on antisense oligonucleotides to block or destroy the expanded CUG repeats in the DMPK RNA. If that were accomplished, could bidirectional transcription or RAN translation still have an effect on the disease course? Is getting rid of the CUG RNA expansion enough to treat the disease?
LR: An antisense oligo, such as Ionis Pharmaceutical's DMPKRx, is a great idea because it is designed to destroy the CUG RNA, and this would also get rid of any possible CUG RAN proteins. I am hopeful that this will be enough to make a big impact on the disease. But there could still be additional problems caused by remaining RNA and proteins not removed by the drug, such as CAG RNA [resulting from bidirectional transcription of the CTG repeat] and RAN proteins. We don’t yet know the impact that the CAG RNAs or RAN proteins have on DM1 at this point, but we’re working hard on this problem and will need to keep this in mind.
MDF: Is there a mouse model for DM2?
LR: Yes. We’ve developed a mouse model of DM2 and presented it at IDMC 10 [the 10th International Myotonic Dystrophy Consortium, held in Paris in June 2015]. The model overexpresses CCUG RNA conditionally. We’re working on publishing this DM2 mouse model.
MDF: What special contributions do you expect to make to the MDF Scientific Advisory Committee?
LR: I have a lot of research experience in DM1 and DM2 and a broad perspective from working on these and other expansion diseases. There’s a lot you can learn from looking across diseases, and I can bring this perspective to the table.