The Repeat Expansion Diseases are a group of genetic disorders that result from an increase in the number of repeats in a specific tandem repeat array. These disorders include Fragile X syndrome (FXS) and Friedreich's ataxia (FRDA). FXS is the most common heritable cause of mental retardation. It is caused by expansion of a CGG/CCG-repeat in the 5' UTR of the FMR1 gene. This expansion leads to a decrease in the translatability of the FMR1 mRNA and hypermethylation of the promoter. Carriers of FXS """"""""premutation alleles"""""""" also show a much higher incidence of ovarian and cerebellar dysfunction than individuals with the """"""""full mutation"""""""". FRDA is a degenerative disease associated with cerebellar dysfunction, cardiomyopathy, and diabetes. It is caused by GAA/TTC-repeat expansion in the first intron of the frataxin gene. We are interested in both the mechanism of expansion and the consequences of expansion in these disorders. We have previously shown that the DNA repeats responsible for FXS and FRDA form secondary structures that may contribute not only to the instability of these sequences in the human genome, but also to some of the consequences of expansion in these disorders. We have now shown that the repeats rsponsible for yet another Repeat Expansion Disease, Progressive myoclonus epilepsy Type I also form secondary structures, in particular i-motif structures and tetraplexes. The fact that the ability to form secondary structures is a common property of unstable sequences supports a role for such structures in disease etiology. We have previously shown that the FXS repeats are not intrinsically prone to expansion in transgenic mice. This suggests that sequence context or differences in trans-acting factors might play a role in expansion in humans. Since FMR1 is an X-linked gene, and expansion in FXS is only seen on maternal transmission, it had been suggested that X-inactivation and/or maternal imprinting might play a role. We have now shown that the failure to detect expansion in our CGG/CCG-containing mice is not due to the absence of maternal imprinting, since CGG/CCG-repeats embedded in a transgene than confers a cis-acting imprinting signal also did not expand. We also shown that failure to see expansion in mice is not due to hypervigilant p53 DNA damage surveillance or repair since mice nullizygous for p53 do not show expansions either. Other trans-acting factors currently being tested include the Werner's and Bloom's syndrome helicases, 2 enzymes thought to act to unwind secondary structures during replication or repair. We have extended our previous work on the regulation of the FMR1 gene to show that the FMR1 promoter is intrinsically bent. This bending is exacerbated by extrinsic bending caused by USF1/USF2 heterodimers and NRF-1, the transcription factors that we have shown to be the most important for maximal FMR1 gene transcription. USF1/USF2 and NRF-1 each have a single consensus binding site in the FMR1 promoter. These binding sites are located on opposite ends of the promoter. In addition to these consensus binding sites, both sets of factors also bind to non-consensus sites at the other end of the promoter. Binding to the non-consensus site requires binding to the consensus site. We propose a model in which both intrinsic and extrinsic DNA bending facilitates this simultaneous binding, leading to DNA looping and the juxtaposition of all of the important transcription factors with the basal transcription machinery at the transcription start site. Such a promoter architecture may be particularly important for the regulation of genes like FMR1 which lack both a TATA-sequence and an Initiator element. We have previously proposed a model for the transcription defect in FRDA in which the RNA polymerase is trapped by a purine:purine:pyrimidine triplex that forms in the FRDA-repeat during transcription. This model allows us to make 2 predictions, one is that an RNA:DNA hybrid will form between the template DNA strand and the nascent RNA. We have shown that such a hybrid forms both in vitro and in vivo. Since such a hybrid is sensitive to RNase digestion, it may contribute to the transcription deficit in FRDA by leading to increased transcript degradation. The second prediction we can make based on our Triplex model is that agents that block triplex formation should alleviate the transcription deficit. We have shown that oligonucleotides complementary to the non-template strand that prevent triplex formation lead to an ~8-fold increase in the transcription of synthetic templates containing 88 FRDA repeats. In contrast oligonucleotides complementary to the template strand or those that are not homologous to the repeat have no effect. This suggests possible novel therapeutic approaches to alleviating the symptoms of FRDA.
Showing the most recent 10 out of 18 publications
