| Nothing is static in living organisms. The life of an animal is spent adapting to ever-changing external and internal environments. Variations in the external environment can have profound effects on the behavior, fitness or even survival of an animal. Programmed development has been optimized during the course of evolution to increase animal survival rate under these variable external conditions. In mammals, metamorphosis happens in embryogenesis during which genes are switched on and off to initiate and regulate the formation of various organs and tissues. The end result of this regulation is a fully developed fetus prepared for the transition from the maternal to the outside environment. Programmed developmental events continue to take place postnatally to transform neonates to adult animals, inducing both qualitative and quantitative changes in tissues. However, the regulation of these processes is less understood compared to those regulating prenatal development. Myotonic dystrophy (DM) provides an excellent tool to study the fetal to adult transition. DM is an autosomal dominant neuromuscular disease caused by abnormal (CTG)n or (CCTG)n microsatellite expansions in the non-coding regions of the DMPK or ZNF9 genes, respectively. A unique molecular feature of DM is that embryonic splicing patterns of certain pre-mRNAs persist in adult DM tissues, possibly causing the clinical features of this disease. Thus, the mechanistic basis of DM is dysregulation of postnatal alternative splicing, and elucidating DM pathogenesis could uncover the associated factors and pathways vital for postnatal splicing switches. When transcribed, mutant DMPK and ZNF9 microsatellites form stable RNA hairpins which sequester muscleblind-like (MBNL) proteins. Considerable data support the idea that DM is a MBNL loss-of-function disease. However, how MBNL sequestration caused DM was unclear. In this study, we characterized MBNL1 as one of the first alternative splicing factors which specifically function in developmentally regulated splicing.;The mechanistic basis for MBNL functional sequestration in DM was another unsolved puzzle in DM pathogenesis. The abilities of (CUG)n and (CCUG)n RNAs to titrate MBNL suggested that MBNL has a higher affinity for pathogenic RNAs versus its normal splicing targets. However, through a combination of RNA structural probing, filter binding and gel mobility shift assays, we demonstrated similar MBNL1 binding properties to pathogenic and splicing target RNAs, both in terms of binding affinities and recognition preferences. We also employed electron microscopy to show that MBNL1 forms oligomeric ring structures on CUG repeat RNA. While the amino-terminal region of MBNL1 is essential for RNA binding, the carboxyl-terminus mediates self-interaction which we hypothesize stabilizes inter-ring stacking. While most protein-RNA interactions are dynamic, we discovered that the interaction between MBNL1 and pathogenic RNAs, but not a normal splicing target, is unusually static. Based on this observation, we propose a model which explains how functional sequestration of MBNL is accomplished in DM. |
