Exploring the Muscleblind Sequestration Model for RNA-Mediated Diseases

Permanent Link: http://ufdc.ufl.edu/UFE0021695/00001

Material Information

Title: Exploring the Muscleblind Sequestration Model for RNA-Mediated Diseases
Physical Description: 1 online resource (113 p.)
Language: english
Creator: Yuan, Yuan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007


Subjects / Keywords: alternative, cug, developmental, dystrophy, interaction, mbnl, myotonic, protein, rna, splicing
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: 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.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yuan Yuan.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Swanson, Maurice S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021695:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021695/00001

Material Information

Title: Exploring the Muscleblind Sequestration Model for RNA-Mediated Diseases
Physical Description: 1 online resource (113 p.)
Language: english
Creator: Yuan, Yuan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007


Subjects / Keywords: alternative, cug, developmental, dystrophy, interaction, mbnl, myotonic, protein, rna, splicing
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: 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.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yuan Yuan.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Swanson, Maurice S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021695:00001

This item has the following downloads:

Full Text




2 2007 Yuan Yuan


3 To the loving and the beloved


4 ACKNOWLEDGMENTS I would like to thank m y mentor, Dr. Maurice Swanson, for giving me this exciting project to work on and superb intellectual support and guidance during the past six years. To think creatively and to attack scientific problems meticulously, are the two qualities he greatly encouraged and reinforced in me, which will bene fit me for the rest of my scientific career. I would also like to thank the past and present members of the Swanson lab, in particular Dr. Keith R. Nykamp, Dr. Patana Teng-Umnuay, Myrna Stenbe rg, Mike Poulos, Jason ORourke and Jihae Shin, for the fun intellectural discussions as well as their technical help. I would like to extend a special thanks to Dr. Jack Griffith and Dr. Sara h Compton for hosting me at University of North Carolina at Chapel Hill and teaching me electron mi croscopy. I also greatly appreciate the help of Dr. Krzysztof Sobczak for performing the RNA structural probi ng described in this dissertation. I would also like to thank my committee members, Dr. Alfred Lewin, Dr. John Aris, Dr. Richard Condit and Dr. Brian Burke, for providing me with valuable assistance during my graduate career. I am sincerely gr ateful to them. Last, but not least, I would lik e to thank my parents for their unconditional love and suppor t, for which I will be forever indebted.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES ................................................................................................................. ..........8 LIST OF FI GURES.........................................................................................................................9 LIST OF ABBRE VIATIONS........................................................................................................ 10 ABSTRACT ...................................................................................................................................13 CHAP TER 1 INTRODUCTION................................................................................................................. .15 Overview of Myotonic D ystrophy.......................................................................................... 15 Clinical Features of Myotonic Dystrophies .....................................................................15 Myotonic dystrophy type I (DM1) adult-onset ...................................................... 15 Myotonic dystrophy type I congenital form (CDM)............................................. 16 Myotonic dystrophy type II (DM2) .......................................................................... 16 Genetic Basis of Myotonic Dystrophies ..........................................................................17 Models for DM Pathogenesis .......................................................................................... 19 Haploinsufficiency m odels....................................................................................... 19 RNA dom inance models.......................................................................................... 20 Hum an MuscleblindLike Proteins.........................................................................................22 Structure and Phylogeny ..................................................................................................22 Expression Profile ........................................................................................................... 24 2 MATERIAL AND METHODS.............................................................................................. 27 Plasm ids and Constructs........................................................................................................ .27 Recom binant Protein Preparation........................................................................................... 28 Che mical and Enzymatic Analysis of RNA Structures.......................................................... 29 Isolation of RNA from mouse skeletal muscle and RT-PCR.................................................30 Cotransfection Assays ............................................................................................................31 Yeast Two-hybrid Analysis .................................................................................................... 32 Co-immunoprecipitation ......................................................................................................... 32 Electron Microscopy ...............................................................................................................33 Photocrosslinking ...................................................................................................................33 Filter-Bind ing Assays.......................................................................................................... ...35 Electrophoretic Mobility Shift Assays .................................................................................... 35


6 3 HUMAN MUSCLEBLIND-LIKE 1 PR OTEIN REGULATES DEVELOPMENTAL SPLICING SWITCHES.........................................................................................................37 Introduction: Misregulation of Alternativ e P re-mRNA Splicing in Myotonic Dystrophy and Other Human Diseases................................................................................................. 37 Overview of Alternative Splicing .................................................................................... 37 Alternative Splicing Regulation ...................................................................................... 37 Alternative Splicing and Hum an Diseases...................................................................... 41 Myotonic D ystrophy, Aberrant Splici ng and Muscleblind-like Proteins........................41 Results.....................................................................................................................................43 Skeletal Muscle Fast Troponin T (T nnt3) Undergoes a Postnatal Developmental Splicing Switch............................................................................................................ 43 MBNL1 Promotes Tnnt3 Fetal Exon Exclusion ............................................................. 43 MBNL1 Physically Interacts with T nnt3 Pre-mRNA..................................................... 44 MBNL1 Binding Sites Are Required for MBNL1-Mediated S plicing Regulation......... 45 Conclusions, Discussion and Future Work ............................................................................. 46 MBNL1 Protein Regulates Alternative Splicing .............................................................46 The Role of MBNL1 in DM Pathogenesis ......................................................................48 The Im portance of Developmental Splicing Regulation................................................. 48 Future W ork.....................................................................................................................49 4 POTENTIAL MECHANISM FOR FUNCTI ONAL SEQUESTRATION OF MBNL1 IN DM.....................................................................................................................................63 Introduction................................................................................................................... ..........63 Results.....................................................................................................................................64 MBNL1 Interacts with an Intronic Stem -Loop Structure Upstream of the Tnnt3 Fetal Exon.................................................................................................................... 64 MBNL1 Binds to the S tem Region of a Pathogenic dsRNA........................................... 66 Sim ilar Affinities of MBNL1 for Sp licing Precursor And Pathogenic RNAs................ 67 Visualization of MBNL1-CUGexp Complexes................................................................ 69 MBNL1 Self-interaction Mediated by the C-terminal Region ........................................69 MBNL1-RNA Com plexes Display Distinct Stabilities...................................................71 Conclusions, Discussion and Future Work ............................................................................. 72 MBNL1 Targets Sim ilar Binding Motifs in Splicing Precursor and Pathogenic RNAs........................................................................................................................... .72 MBNL1 Rings: Interactions with CUGexp RNA and Potential Mechanism for MBNL1 Functional Sequestration............................................................................... 74 Future W ork.....................................................................................................................77 5 DISCUSSION: INVOLVEMENT OF MUSC LEBLIND-LIKE PROTEINS IN RNAMEDIATED DISEASES AND POTENT IAL THERAPEUTIC APPROACHES................87 DM1 and DM2 ........................................................................................................................87 Congenital and Adult-Onset DM1 .......................................................................................... 88 Potential MBNL Involvem ent in Other RNA-Mediated Diseases......................................... 89 Spinocerebellar Ataxia T ype 8 (SCA8)...........................................................................90


7 Huntingtons Disease-like 2 (HDL2)..............................................................................91 Fragile X Trem or Ataxia Syndrome (FXTAS)............................................................... 91 Im plications for Potential DM Therapeutic Strategies........................................................... 92 APPENDIX: LIST OF PRIMERS .................................................................................................95 LIST OF REFERENCES ...............................................................................................................96 BIOGRAPHICAL SKETCH .......................................................................................................113


8 LIST OF TABLES Table page 3-1 Splicing factors and human diseases..................................................................................51


9 LIST OF FIGURES Figure page 1-1 Comparison of adult-onset DM1, co ngenital DM1 and DM2 symptoms..........................26 3-1 Types of alternative splicing events. ..................................................................................53 3-2 Splicing Mechanism ........................................................................................................... 55 3-3 Regulation of alternative splicing. ..................................................................................... 56 3-4 Illustration of aberrant sp licing patterns in DM. ................................................................57 3-5 Tnnt3 undergoes a postnatal developm ental splicing switch............................................. 58 3-6 MBNL1 promotes Tnnt3 fetal exon exclusion. .................................................................59 3-7 MBNL1 crosslinks to Tnnt3 pre-m RNA........................................................................... 60 3-8 MBNL1 binding sites are required for MB NL1mediated splicing regulation................. 61 3-9 DM pathogenesis m odel.................................................................................................... 62 4-1 MBNL1 recognizes a R NA hairpin upstream of the Tnnt3 fetal exon.............................. 79 4-2 MBNL1 binds throughout the dsCUG stem ......................................................................80 4-3 MBNL1 binds to pathogenic and splicing precursor RNAs with sim ilar affinities........... 81 4-4 Visualization of dsCUG a nd MBNL1-dsCUG complexes. ...............................................83 4-5 Self-association of MBNL1 proteins is m ediated by the C-terminal region..................... 84 4-6 MBNL1-RNA com plexes displa y distinct stabilities........................................................ 85 4-7 Model of complexes form ation between MBNL1 and splicing targets versus pathogenic RNAs............................................................................................................... 86


10 LIST OF ABBREVIATIONS CDM Congenital myotonic dystrophy CLCN1 Skeletal muscle chloride channel 1 cTNT/TNNT2 Cardiac troponin T CUGBP1 CUG repeats RNA binding protein 1 DM Myotonic dystrophy DM1 Myotonic dystrophy type I DM2 Myotonic dystrophy type II DMPK Dystrophia m yotonica protein kinase DMWD Dyotrophia myotonicacontaining WD repeat motif dpc Days post coitus EM Electron microscopy ESE Exonic splicing enhancer ESS Exonic splicing silencer F exon Fetal exon FTDP-17 Frontotemporal dementia with parkinsonism linked to chromosome 17 HD Huntington disease Hfq Host factor 1; E.coli homolog to the Sm protein hnRNP Heterogeneous ribonucleoproteins HSA Human skeletal muscle actin HSALR Transgenic mice carrying 250 CT G repeats in the 3-UTR of HSA transgene


11 HSASR Transgenic mice carry 5 CTG repeats in the 3-UTR of HSA transgene IR Insulin receptor ISE Intronic splicing enhancer ISS Intronic splicing silener MAPT Microtubule-a ssociated protein tau Mbl Muscleblind protein MBNL Muscleblind-like protein Nab2p Nuclear polyadenylated-RNA binding protein 2 Nova Neuro-oncological ventral antigen nPTB Neuronal polypyrimid ine tract binding protein NMD Nonsense-mediated decay PTB Polypyrimidine tract binding protein PTC Premature termination codon RAR Retinoic acid receptor gamma SCA Spinocerebellar ataxias Six5 Sine oculis-related homeobox 5 homolog snRNP Small nuclear ribonucleoprotein T5 The 500 nt region including Tnnt3 fetal exon and the upstream intronic sequence T5.1 The most 5 125 nt region of T5; it is en tirely intronic T5.45 The most 3 200 nt region of T5; it includes the en tire Tnnt3 fetal exon and 161 nt of the upstream intronic sequence


12 Tnnt3 Fast troponin T in skeletal muscle TTP Tristetraprolin U2AF U2 auxiliary factor UTR Untranslated region ZNF9 Zinc finger protein 9


13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXPLORING THE MUSCLEBLIND SEQUES TRATION MODEL FOR RNA-MEDIATED DISEASES By Yuan Yuan December 2007 Chair: Maurice S. Swanson Major: Medical Sciences--Genetics Nothing is static in living organisms. The life of an animal is spent adapting to everchanging external and internal e nvironments. Variati ons in the external environment can have profound effects on the behavior, fitness or ev en 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 a nd 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 post natally to transform neonates to adult animals, inducing both qualitative and qua ntitative changes in tissues. Ho wever, the regulation of these processes is less understood comp ared to those regulating prenatal development. Myotonic dystrophy (DM) provides an excellent tool to stu dy 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 embr yonic splicing patterns of certain pre-mRNAs persist in adult DM tissues, possi bly causing the clinical features of this disease. Thus, the


14 mechanistic basis of DM is dys regulation of postnatal alternat ive splicing, and elucidating DM pathogenesis could uncover the associated fact ors and pathways vital for postnatal splicing switches. When transcribed, mutant DMPK and ZNF9 microsatellites form stable RNA hairpins which sequester muscleblind-like (M BNL) proteins. Considerable data support the idea that DM is a MBNL loss-of-function disease. Howeve r, 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 deve lopmentally 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 hi gher affinity for pathogenic R NAs 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 term s of binding affinities and rec ognition preferences. We also employed electron microscopy to show that MBNL 1 forms oligomeric ring structures on CUG repeat RNA. While the ami no-terminal region of MBNL1 is essential for RNA binding, the carboxyl-terminus mediates self-i nteraction which we hypothesize st abilizes inter-ri ng stacking. While most protein-RNA interac tions are dynamic, we discovered that the intera ction between MBNL1 and pathogenic RNAs, but not a normal sp licing target, is unusua lly static. Based on this observation, we propose a model which explai ns how functional seques tration of MBNL is accomplished in DM.


15 CHAPTER 1 INTRODUCTION Overview of Myotonic Dystrophy Myotonic dystrophy (D M) is a multisystemic, autosomal dominant disease that affects 1 in 8,000 individuals worldwide (1). As perhaps the most variable of all human diseases, myotonic dystrophy can affect individuals of all ages ranging from infants to seniors. DM patients present an array of seemingly unrelated symptoms in a great variety of organs and systems including skeletal muscle, heart, brain, eye and the endocrine system ( 1). At least two types of myotonic dystrophies, myotonic dystrophy type I (DM1) and type II (DM2), have been characterized clinically and genetically in detail. Each of th ese DM types cause distinct yet similar clinical manifestations (2). The highly heterogeneous symptoms in DM1 and DM2 nonetheless reflect the unusual underlying genetic variations that have been pinpointed to two independent microsatellite expansions in the non-c oding regions of two unrelated genes ( 3-7 ). Although DM was one of the first genetic anticipation disord ers for which an unstable genetic element was identified, it was not until recently that the pathogenesis mechanism was unveiled ( 8). Clinical Features of Myotonic Dystrophies Myotonic dystrophy type I (DM1) adult-onset The hallm ark of the DM1 skeletal muscle pheno type is myotonia, which can be elicited in almost every symptomatic adult DM1 patient, an d also frequently found in presymptomatic individuals. Myotonia is defined as slow muscle relaxation after voluntary musc le contraction. In the clinic, it is readily tested in the hand muscles where, fo llowing a forceful contraction, there is a delayed ability to relax the grip ( 1). In DM1, myotonia is frequently accompanied by progressive muscle weakness and wa sting in the facial muscles as well as the distal muscles in


16 the limbs. Skeletal muscle sections reveal an increased number of centralized nuclei and split fibers in DM1 muscle, presumably indicati ng immaturity of muscle fiber development ( 1). Besides skeletal muscle involvement, DM1 patients present multisystemic manifestations including cardiac conduction defe cts, posterior iridescent ocular cataracts, irritable bowel syndrome, testicular atrophy, insulin resi stance, hypogammaglobulinemia, hypersomnia and personality disturbance. Cardiac conduction defects probably account fo r the increased sudden death rate observed in DM patients (1,2,9-11). Notably, genetic anticipa tion, characterized by an intergenerational increase in diseas e severity and earlier age of onset, is striking in DM1. Mildly affected women may give birth to severe ly and congenitally affected children. Myotonic dystrophy type I congenital form (CDM) Congenital DM1 presents a dis tinct clinical picture com p ared to adult-onset DM1. Affected infants are born with severe neurological, neuromuscular and musculoskeletal abnormalities including hypotonia (floppy baby), tali pes, craniofacial abnormalities, poor suckling and swallowing and often requi re respiratory support to survive ( 1 ). Unlike adult-onset DM1, mental retardation is a common accompanim ent and arguably the most important feature of CDM. Intriguingly, myotonia, the cardinal feature of adult-onset DM1, is never present in the first year of life in the CDM patients. Howeve r, CDM survivors in th eir adulthood eventually develop myotonia, among the full spectrum of adult-onset DM1 symptoms(1,9,10). Myotonic dystrophy type II (DM2) While DM1 can affect individuals of all ages, DM2 has a reported m edian age of onset at 48 years old( 12). Similar to DM1, myotonia, progressive muscle wasting and weakness are the key features of skeletal muscle involvement in DM2. But the pattern of muscle weakness in DM2 is significantly different from that in DM 1, in that DM2 patients frequently show proximal muscle weakness and wasting, such as in hip girdle muscles, as opposed to the distal muscle


17 weakness and wasting characterist ic of DM1 patients. In add ition, facial muscle weakness, which is prominent in DM1, is absent in DM2. Histologically, sections of DM2 muscles also reveal centralized nuclei as seen in DM1 muscles(1,9,10). DM1 and DM2 share multisystemic effects in the eye, heart and endocrine system. Yet one of the primary differences between DM1 and DM2 is the lack of a DM2 form that affects development of muscles and the central nervous system. To date, there has not been any definitive evidence indicating that mental retardation is associated with DM2. Figure 1-1 presents a brief comparison of the symptoms in DM1 and DM2. Genetic Basis of Myotonic Dystrophies Substantial progress in understanding the ge netic basis of DM1 was m ade in 1992, when six groups mapped and further confirmed that the DM1 mutation was an unstable (CTG)n microsatellite expansion in the 3-untransla ted region (3-UTR) of the dystrophia myotonica protein kinase ( DMPK ) gene (4-7,13,14). In addition, repe at size shows an inverse correlation with age of onset and a positive correlation with the severity of the disease, with 5-37 repeats in normal individuals, 50-1,000 in mild and classical adult-onset DM1 patients and >1,000 in the most severe cases, congenitally affected patients ( 15). Thus, the great genetic variability of the DM1 mutations is an important component of the extremely heterogeneous symptoms observed for DM1 patients. There is a marked tendency for expanded (CTG)n repeats to increase in size both in somatic tissues and in the germ line, in the latter case leading to more severe phenotypes in the offspring, thus accounting for the obser ved genetic anticipation in DM1 patients (12,1622). Expansion of CTG repeats can be dram atic in DM1, especially through maternal transmission; in some cases expansions of se veral thousand repeats can occur in a single generation (22,23). Studies in various systems and organisms including cell culture, bacteria, yeast and mice, have shown that the CTG repeat instability involves DNA replication-dependent


18 as well as -independent mechanisms ( 24-26). Several DNA mismatch repair proteins such as Msh2, Msh3 and Pms2 have been shown to play important roles in somatic and germline expansion of the repeats ( 27-32). Genetic anticipation has also been observed in a number of other diseases including Huntington disease (HD), spinocerebellar ataxias (SCA), schizophrenia and bipolar disorders. Interestingly, some, if not all, of these diseases are also caused by unstable repeat expansions ( 33-38). Similar to DM1, DM2 is caused by another unstable DNA element, a (CCTG)n tetranucleotide expansion located in the first intron of zinc finger protein 9 ( ZNF9 ) (3). ZNF9 encodes a nucleic acid binding protei n that is highly expressed in skeletal muscle and has been suggested to modulate th e activity of beta-myosin heavy ch ain and more recently, internal ribosomal entry site (IRES) mediated translatio n (39,40). Positioned at ~850 nt upstream of the 3 splice site of ZNF9 intron1, the repeat expansion has been shown not to affect ZNF9 mRNA processing or prot ein expression ( 41). While ZNF9 (CCTG)n repeats in normal individuals are short (5-26) and often interrupted by other sequences, the CCTG portion expands in affected individuals to ~75-11,000 repeats ( 42 ). Interestingly, although DM2 is generally a milder disease compared to DM1, the average size of the DM2 CCTG expansion is around 5,000, bigger than the longest documented DM1 CTG expansion (~4,000 repeats). Unlike DM1, there is a lack of positive correlation between the size of the repeats and the sever ity of the disease. Reports have shown that even people homozygous for large (CCTG)n expansions do not show more severe symptoms(12,43). In addition, th ere is no convincing evidence for genetic anticipation in DM2, although the CCTG tetranucleot ide expansions do display a high degree of instability in both somatic and germline cells ( 42 ).


19 Models for DM Pathogenesis For years after the d iscovery of the DM1 a nd DM2 mutations, one in the 3-UTR of the DMPK gene and the other in the first intron of ZNF9 a fundamental question about the pathogenic mechanism remained unanswered. Ho w do microsatellite expansions in these noncoding regions cause dominantly inherited diseases ? At least two types of models, which are not necessarily mutually exclusive, have been proposed for DM pathogenesis. Haploinsufficiency models Since the (CTG)n expansion resides in the DMPK gene, the DMPK haploinsufficiency model was initially proposed for DM1 pathogene sis. This model hypothesizes that a reduction of DMPK protein levels caused by the micros atellite expansion is responsible for the multisystemic clinical features of DM1. Ind eed there is evidence supporting reduced DMPK levels in adult DM1 patients ( 44-46 ). However DMPK levels in congenital DM1 patients appear to be unaltered compared to normal individuals ( 47). Furthermore, Dmpk heterozygous or homozygous knockout mice fail to re capitulate the cardinal phenotypes of human DM1 patients. Instead, these mice develop a cardiac conduction def ect and a mild, late-onset myopathy which is not characteristic of DM1 ( 48-50). Taken together, these resu lts suggest that although DMPK haploinsuffciency in adult DM1 patients may c ontribute to cardiac feat ures of DM1, loss-offunction of DMPK is not the sole requirement for the multisystemic progression of the disease. A second kind of haploinsufficien cy model proposes that (CTG)n repeats constitute a strong nucleosome positioning sequence, which could alter local chromatin structure and lead to reduced expression of neighboring genes, dyotro phia myotonica-containing WD repeat motif ( DMWD ) and sine oculis-related homeobox 5 homolog ( SIX5 ) (51,52). Interestingly, DMPK gene is located in a gene rich cluster where the DMWD DMPK and SIX5 genes occupy only a 20 kilobase (kb) region. The 3 terminus of the DMWD gene overlaps the DMPK promoter and the


20 CTG expansion overlaps the 5 promoter region of SIX5 Although experiments designed to test for DMWD and SIX5 expression in DM1 patients have so far yielded inconsistent results ( 53-56 ), the SIX5 haploinsufficiency model gain ed support from studies showing that Six5 heterozygous and homozygous knockout mice develop ocular catar acts (57,58). However, these cataracts are not the iridescent opacities with posterior location commonly seen in DM patients. Of course, the DMPK and SIX5 haploinsufficiency models only provide an explanation for DM1, and not DM2, pathogenesis. For DM2, neither ZNF9 gene expression nor the processing of ZNF9 mRNA is affected by the CCTG expansion (41). The high degree of clinical parallels between DM1 and DM2 suggest very similar underlying molecular events leading to these disorders. RNA dominance models In 1995, the Singer and Hous man labs discovere d that RNA transcripts containing either expanded CUG or CCUG repeats accumulate in nuc lear foci both in cultured cells and biopsied tissues ( 59). This was the first indication that mu tant RNAs are directly involved in DM pathogenesis. In 1996, Timchenko and colleague s proposed the RNA dominance model which hypothesized that mutant RNAs transcribed from the DM loci exert a toxic gain-of-function effect, possibly by disrupting other cellular proce sses such as transcription and RNA processing ( 60). The strongest support for the RNA domi nance model comes from a transgenic mouse model developed in 2000. Mankodi and colleagues generated transgenic mouse lines expressing either 5 or 250 CTG repeats inserted at the 3 -UTR of the human skeletal muscle actin gene which is not implicated in DM pathogenesis and expressed only in skeletal muscle. While the mice expressing (CTG)5 were normal, mice expressing the longer CTG repeats developed ribonuclear inclusions and disp layed myotonia and myopathic features characteristic of DM1 muscle ( 61). Furthermore, the toxic effect of the tran sgene is dependent on its expression level.


21 Mice with transgene insertions in heterochro matic regions appeared phenotypically normal, indicating a trans-dominant effect at th e RNA, instead of the DNA, level. Yet questions remain as to how the repeat-c ontaining RNAs elicit toxic effects. The RNA structure prediction program Mfold suggests that CUG and CCUG repeats form hairpin structures in vitro with G-C basepairs and mismatched pyrimidines ( 62). This prediction was later supported by crystallography and confirme d by RNA structural probing and electron microscopy ( 63-67). The length of these hairpins is proport ional to the number of repeats. Thus, cellular factors with high affinities for such hairpin structures could potentially be titrated by these abnormal RNAs away from their physiological targets. In an attempt to isolate such factors, Miller and colleagues identified human muscle blind-like (MBNL) proteins as strong candidates ( 68). MBNL proteins bind to CUG RNA ha irpins in a length-dependent manner. Moreover, MBNL proteins colocalize with CUG or CCUG containing RNAs in ribonuclear inclusions in DM myoblasts as well as ti ssues. These observations lead to the MBNL sequestration model, which proposes that DMPK and ZNF9 mutant RNAs disrupt normal MBNL function, and thus DM is a MBNL loss-of-function di sease. To test for this hypothesis, Kanadia and colleagues generated an Mbnl1 knockout mouse model which will be discussed further in the following chapter. Consistent with the MBNL sequest ration model, Mbnl1 loss-of-function in the mouse elicits key DM features, including myot onia, structural changes in muscle along with posterior dust-like cataracts ( 69 ). This result strongly argues th at titration of MBNL proteins underlies DM pathogenesis. Although the MBNL sequestration model for DM pathogenesis has gained considerable experimental support to date, a lternative hypotheses have been raised, including transcription factor sequestration proposed by Ebralidze and colleague ( 70 ). Upon expression of CTG repeat-


22 containing DMPK RNA in myocytes certain transcription factors, such as Sp1and retinoic acid receptor gamma (RAR ), relocalize from chromatin to the ribonuclear inclusions. Subsequent reduction in the mRNA levels was detected fo r skeletal muscle chloride channel 1 ( CLCN1 ), which harbors multiple Sp1 binding sites in its promoter region. Replenishment of Sp1 in mutant RNA expressing cells partially restored the CLCN1 mRNA level. Based on these data, Ebralidze and coworkers postulated that leaching of Sp1 by mutant RNAs c ontributes to reduced muscle chloride channel 1 levels, and resulted in myotonia in DM1 patients ( 70). In summary, our knowledge on DM pathogene sis has rapidly expanded over the past decade. The prevailing RNA dominan ce view is backed up by numerous in vitro and in vivo studies. Sequestration of human muscleblindlike proteins by expande d CTG or CCTG repeats may be the major mechanism underlying RNA toxicity. Human Muscleblind-Like Proteins Structure and Phylogeny Based on considerable sequence homology, hum an muscleblind-like proteins were named after their Drosophila homolog, muscleblind (mbl ), which, when mutated, results in abnormalities in the terminal differentiation of mu scle and photoreceptor cells (71,72). To date, muscleblind proteins have been found in a va riety of metazoan phyla, including nematoda, arthropoda, tunicata and verteb rata. While a single muscleblind gene was found for each invertebrate species, most vertebrates ha ve three, and in some case, such as Takifugu rubripes, up to five muscleblind homologs (73). Human MBNL family includes three members, MBNL1, 2 and 3, located on chromosomes 3, 13 and X chro mosome, respectively. Similar genomic organizations of the three human muscleblind-like genes suggest th at they originated from gene duplication events during evolution.


23 The most distinct structural feature of muscleblind proteins is the presence of tandem CCCH zinc finger motifs, composed of three cysteine and one histidine. Known for its ability to mediate nucleic acid-protein interactions, the CCCH motif is found in many RNA and DNAbinding proteins, including mouse tristetraprolin (TTP) and y east nuclear polyadenylated-RNA binding protein 2 (Nab2p) (74,75). The struct ure of an RNA-bound CCCH containing protein, zinc finger protein 36 C3H type-like 2 (ZNF36L2) has been solved using nuclear magnetic resonance (NMR). This structure indicates that electrostatic as well as hydrogen-bonding interactions between closely stack ed RNA bases and protein aromatic side chains play important roles in stabilizing the protein-RNA interaction ( 76 ). While all muscleblind proteins contain CCCH motifs, the number of the motifs in each pr otein varies from nematoda to vertebrata. Protostome species encode musclebind with onl y two CCCH motifs, whereas four such motifs are present in muscleblind proteins from Tunicat a and Vertebrata species, arranged as two pairs separated by a linker region of around 60 amino acid residues ( 73). Comparison of the individual CCCH motifs am ino acid compositions in Drosophila and human muscleblind proteins reveals that the first and third CCCH motif in human musc leblind proteins resemble the first CCCH motifs in Drosophila protein, while the second and fourth CCCH motifs in human resemble the second CCCH motif in Drosophila This suggests that the CCCH motifs duplicate, and function as pairs during evolution. Muscleblind proteins show strong func tional conservation among species. Although human and Drosophila muscleblind proteins display a distant 31% homology, human MBNL proteins can neve rtheless rescue Drosophila mbl mutants ( 77 ). Among vertebrate species, muscleblind proteins are also highly conserved in sequence. Mouse and human MBNL1 differ by only two amino acid residues. In terestingly, the high degree of conservation is found not only at


24 the protein level, but also at certain genomic regions. As shown by Bejerano and colleagues, MBNL1 and MBNL2 are both among a group of genes which have an uninterrupted >200 base pair (bp) genomic segments that are completely identical in human, mouse and rat. These so called ultraconserved regions in MBNL1 and MBNL2 overlap each other and are located around exon 7 of MBNL1 and the corresponding exon 6 of MBNL2 ( 78). Such strong conservation could be a result of potential mu tation cold-spots, alt hough the fact MBNL1 and MBNL2 are located in two distinct chromosoma l loci would argue agai nst this possibility. Alternatively, extreme negative selection, which us ually indicates functional importance, could be responsible for the ultraconservation. The me chanism for this absolu te sequence conversation in human, mouse and rat is not yet clear. Expression Profile Hum an muscleblind-like family members show di stinct expression patterns in adults. By Northern blot analysis, MBNL1 is abundantly expressed in h eart and skeletal muscle, while MBNL2 shows a more ubiquitous expression pattern w ith its mRNA detected in virtually every tissue examined. In contrast to MBNL2 a very specific pattern was seen in the case of MBNL3, which is expressed most strongly in placenta wh ile showing very weak signal in pancreas, kidney, liver and heart (68,79). The embryonic expression patterns of human MBNL genes, however, have to be extrapolated from the data obtained in mice ( 80). Using RNAs derived fr om embryos of different stages, RNA blot analysis revealed coor dinate temporal e xpression patterns of Mbnl with Mbnl1 peaking between 13.5-15.5 days post coitus (dpc), Mbnl2 between 17.5-18.5 dpc and Mbnl3 between 11.5-15.5 dpc. Furthermore, whole-moun t in situ hybridizat ion was performed on 9.5 dpc embryos and all three muscleblind proteins were highly expressed in the developing head region with prominent Mbnl2 and Mbnl3 expression in the mandibular and maxillary


25 components in the first branchial ar ch and neural tube. All three Mbnl genes were moderately expressed in forelimb buds. Interestingly, Mbnl2 also showed strong expression in the heart. When examined at a slightly later stage, Mbnl1 and Dmpk expression patterns overlap in the hindlimb and forelimb buds as well as the first and second branchial arch es. This observation supports the RNA dominance model, indicating that Dmpk mutant transcripts could titrate Mbnl1 during embryonic development and result in the developmental defects seen in CDM patients ( 80). Cumulatively, these results suggest that m yotonic dystrophy is caused by mutant RNAs which, in turn, lead to sequest ration of MBNL proteins. Ho wever, an important question remains. What is the missing link between MB NL loss-of-function and DM phenotypes? In other words, what is the physiologica l function(s) of MBNL proteins?


26 Figure 1-1 Comparison of adult-onset DM1, congenital DM1 and DM2 symptoms.


27 CHAPTER 2 MATERIAL AND METHODS Plasmids and Constructs To construct pGEX-6P-His, a XhoINot I fragm ent encoding the His6 tag from pGEXMBNL1 (NP_066368) was inserted into XhoI-Not I digested pGEX-6P-1 (GE Life Sciences). pGEX-6P-MBNL1-His was constructed by inserting a Bam HIXhoI fragment from pGEXMBNL1 into Bam HIXhoI digested pGEX-6P-His. The MB NL1 N-terminal region (residues 1253) was amplified using primers MSS2759 a nd MSS2760 (see Appendix B), digested with Bam HI and XhoI and inserted into pGEX-6P-His to cr eate pGEX-6P-MBNL1-N-His. For the yeast two hybrid system, either full length, amino te rminal (residues 1-264) or carboxyl terminal (residues 239-382) MBNL1 cDNAs were PCR am plified using the following primers and inserted into pGBKT7 or pGADT7 (Clontech, Mountain View, CA) at Sma I and Bam HI sites: 1) full length, primers MSS1163 (forward primer) a nd MSS1166 (reverse primer); 2) N-terminal region, MSS1164 (forward) and MSS1166 (reverse); 3) C-terminal region, primers MSS1163 (forward) and MSS1165 (reverse). To create pcDNA3-V5, primers MSS3045 and MSS3046 were subjected to a 10-cycle PCR reaction: 94oC 30 sec, 50oC 20 sec, 72oC 20 sec. The resulting DNA fragment was gel purified followed by digestion with Nhe I and Bam HI, and inserted into Nhe I-Bam HI digested pcDNA3.1(+) (Invi trogen, Carlsbad, CA). The Bam HIXhoI fragment from pGEX-6P-MBNL1-His was inserted into pcDNA3-V5 at Bam HI and XhoI sites to create pcDNA-V5-MBNL1. The Bam HIXhoI fragments from pGEX-6P-MBNL1-His and pGEX-6PMBNL1-N-His were inserted into pcDNA3.1(+) /myc-His A (Invitrogen) to create pcDNA3MBNL1-mycHis and pcDNA3-MBNL1-N-mycHis, respectively. The CUGBP1 (NP_006551) coding sequence was PCR amplified using primers MSS2699 and MSS2700 and inserted into pcDNA3.1(+)/myc-His A at Bam HI and XhoI sites. To construct pEGFP-C1-hnRNP A1


28 (NP_034577), the mouse hnRNP A1 coding region was amplified using primers MSS2002 and MSS2003. PCR product was digested with EcoRI and XhoI, and inserted into EcoRI-XhoI digested pEGFP-C1(Invitrogen). For pEGFP-C1MBNL1-N, the MBNL1 N-terminal coding region (residues 1-253) was amplified using primers MSS3074 and MSS3075, digested with XhoI and BamHI and inserted into pEGFP-C1. The Tnnt3 minigene was prepared by amplifying the mouse Tnnt3 genomic region between exons 8 and 9 using primers MSS1949 and MSS1950 and inserting the PCR product into pSG5 (Stratagene, La Jolla, CA) at the Eco RI site. The mutant Tnnt3 minigenes, pSG5-Tnnt3 10 and pSG-Tnnt3/gg, were generated by site-directed mutatgenesis using primers MSS2129/MSS 2130 and MSS3131/3132, respectively. Wild type pSG5-Tnnt3 (100 ng) (with 125 ng of each of the primers) was subjected to the following PCR reaction: 94oC 30 sec, 50oC 1 min, 72oC 8 min, 20 cycles using Pfu DNA polymerase (Stratagene). After Dpn I digestion, the PCR product was tran sformed into DH10B and mutants were identified by plasmid DNA sequencing. Using pSG5-Tnnt3 as a template, PCR fragments generated from primer pairs MSS1865/M SS1879 and MSS1884/MSS1866 were TOPO-cloned into pCR4-TOPO (Invitrogen) to make pTOPO-T5.1 and pTOPO-T5.45, respectively. Recombinant Protein Preparation For the prep aration of recombinant prot eins, BL21(DE3) RP containing pGEX-6P-1MBNL1 or pGEX-6P-1-MBNL1-N were grown to OD600=0.5 followed by induction with 1 mM IPTG for 2 h at 30oC. Cells were collected and resuspe nded in lysis buffer containing 25 mM Tris-Cl, pH 8.0, 0.5 M NaCl, 10 mM imidazole, 2 mM -mercaptoethanol, 2 mg/ml lysozyme, 10 g/ml DNase I, 5% glycerol, 0.1% Triton X-100 supplemented with protease inhibitors. The cell suspension was incubated on ice for 30 min with stirri ng prior to sonication and centrifugation at 12,000 X g. For protein purification, Ni-NTA-Sepharose (Amersham/GE


29 Healthcare, Piscataway, NJ) (12 ml) was inc ubated with the supernatant for 1 h at 4oC and washed three times with 40 ml of wash buffer containing 25 mM Tris-Cl, pH 8.0, 0.5 M NaCl, 20 mM imidazole, 0.1% Triton X-100, followed by th ree 10 ml elutions in 25 mM Tris-Cl, pH 8.0, 0.5 M NaCl, 250 mM imidazole, 0.1% Triton X-100. Subsequently, -mercaptoethanol was added (10 mM final concentration) to the elua te which was incubated with 2 ml glutathioneSepharose (Amersham) for 1 h at 4oC. After three washes (10 ml each) of buffer (WB) containing 25 mM Tris-Cl, pH 8.0, 300 mM NaCl, 5 mM -mercaptoethanol, 0.1% Triton X100, the glutathione-Sepharose beads were incubated with 4 ml WB containing 40 U of PreScission protease (Amersham) at 4oC overnight. The supernatant was collected following brief centrifugation and concentrated to 1-8 mg/ml). Chemical and Enzymatic Analysis of RNA Structures Transcriptio n reactions were carried out in a 50 l volume which contained 2 g of each DNA template, 1 mM NTPs, 3.3 mM guanosine, 60 units of ribonuclease inhibitor RNase Out (Invitrogen), 200 units of T7 RNA polymerase (Ambion, Austin, TX), 10 mM DTT, 40 mM Tris-HCl (pH 7.9), 6 mM MgCl2, 2 mM spermidine, 10 mM NaCl. The reaction was performed at 37oC for 2 h, the transcript was then purified on a denaturing 10% polyacrylamide gel and subsequently 5-end-labelled with T4 polynucleotide kinase and [32P]ATP (3000 Ci/mmol). The labeled RNA re-purified by electrophoresis on a denaturing 10% polyacrylamide gel. Prior to structure probing, the labeled RNA was subjected to a denatura tion/renaturation procedure in a reaction buffer containing 50 mM TrisHCl (pH 8.0), 60 mM KCl, 15 mM NaCl, 2 mM MgCl2 by heating the sample at 90oC for 1 min. and slowly cooling to 25oC. The RNA sample was then mixed with either a 25-fold molar excess of recombinant MBNL1 in 50 mM TrisHCl pH 8.0, 60 mM KCl, 15 mM NaCl, 2 mM MgCl2, 2% glycerol, 0,5 mM DTT, 50 g/mL BSA, or with


30 buffer only (control) and incubated 20 min at 25oC. The final concentration of (CUG)54 was 20 nM and MBNL1 was 500 nM. Under these reac tion conditions more >95% of RNA was bound with protein as revealed by filter binding assays Additional control samples were prepared by mixing RNA with MBNL1 protein previ ously denatured by heating at 75oC for 2 min. Limited RNA digestion was initiated by mixing 5 l of the RNA or RNA/protein sample (25,000 cpm) with 5 l of a probe solution containing either le ad ions or ribonuclease T1 in reaction buffer. The reactions were performed at 25oC for 20 min. and stopped by adding 20 volumes of 1XTE buffer followed by phenol/chloroform extraction. Precipitated RNAs were dissolved in a denaturation solution (7.5 M urea and 20 mM EDTA with dyes). To determine the cleavage sites, the products of RNA fr agmentation were separated on 10 % polyacrylamide gels containing 7.5 M urea, 90 mM Tris-borate buffer, and 2 mM EDTA, along with the products of alkaline hydrolysis and limited T1 nuclease digestion of th e same RNA. The alkaline hydrolysis ladder was generated by the incubation of the labe led RNA in formamide containing 0.5 mM MgCl2 at 100oC for 10 min. The partial T1 ribonuclease digestion of RNAs was performed under semidenaturing conditions (10 mM sodium citrate, pH 5.0; 3.5 M urea) with 0.2 unit/l of the enzyme during incubation at 55oC for 10 min. Electrophoresis was performed at 1800 V (gel dimensions, 30/50 cm). The products of the st ructure probing reactions were visualized by PhosphorImaging (Storm; Molecular Dynamics, Sunnyvale, CA) and analyzed by ImageQuant 5.2 (Molecular Dynamics). Isolation of RNA from mouse sk eletal muscle and RT-PCR Quadriceps isolated from 129 m ice at different post-natal ages were each homogenized in 0.5 ml of TRI reagent (Sigma, T9424) at 4oC. Total RNA was isolated according to manufacturer's instructions. The RNA was furt her treated with RQ1 DNase I followed by phenol extraction and ethanol precipitation. Revers e transcription was performed according to


31 manufacturer's instructions in a 20 l volume containing 1 X first strand buffer, 10 mM DTT, 5 g of RNA, 300 ng of random hexamers (Invitr ogen), 0.5 mM dNTPs and 200 U of Superscript II RNase Hreverse transcriptase (Invitrogen). PCR was performed in a 50 l reaction volume using 4 l of cDNA, 30 pmol of each primer (Primer1: TCTGACGAGGAAACTGAACAAG, Primer 2: TGTCAATGAGGGCTTGGAG), 10 Ci of -32P dCTP and 2.5 U of Taq (Invitrogen) for 27 cycles of 94oC for 30 sec, 50oC for 30 sec, 72oC for 30 sec, followed by one cycle at 72oC for 7 min. PCR samples were purified using a Qiag en PCR purification kit. Half of the eluate was digested with BsrBI and the digested and undigested PCR products were separated on a 8% polyacrylamide gel prior to autoradiography. Cotransfection Assays HEK293T or C2C12 cells were plated in a 6 we ll plate in antibiotic-free m edium to 30% confluency. About18-24 hr later, cells were transfected with 1 g of minigene and 2 g protein expression plasmid using Lipofectamine 2000 (Invitr ogen) following manufacturers protocol. Forty-eight hours post-transfection, cells were lysed in 0.2 ml of TRI reagent (Sigma), and total RNA was isolated following the manufacturers protocol. Revers e transcription was performed as described above except that oilgo(dT) 12-18 was used instead of random hexamers. The PCR reactions were performed using 4 l of cDNA and primers flanking pSG5 MCS (Primer1: AGAATTGTAATACGACTCACTATAGGGC, Primer 2: GCTGCAATAAACAAGTTCTGCTTT) for 27 cycles of 94oC for 30 sec, 55oC for 30 sec and 72oC for 20 sec. Fifteen microliters of PCR r eactions were separated on a 15% polyacrylamide gel at 75 volts for 16 hr followed by autoradiography.


32 Yeast Two-hybrid Analysis pGBKT7-M BNL1 together with one of the pGADT7 constructs (pGADT7-T Antigen, pGADT7-MBNL1, pGADT7-MBNL11-N, pGADT7-M BNL1-C) were transformed into yeast strain AH109. In addition, pGBKT7-p53 and pGADT7-T Antigen (Clontech) were cotransformed into AH109 as a negative control. Double transformants were selected on SD/-Trp/Leu plates. Expression of the myc-tagged GAL4 DNA binding domain (BD) and HA-tagged GAL4 activation domain (AD) fusion proteins we re confirmed by immunoblot analysis using mAb 9E10 and 16B12, respectively. To test for protein interactions, transf ormants were streaked onto SD/-Trp/-Leu/-His incubated at 30oC for 3-4 days and scored for growth. Co-immunoprecipitation To test for MBNL1-MBNL1 interactions in m ammalian cells, HEK293T cells were transfected with 5 g of pcDNA3-V5-MBNL1 alone as a control or 5 g of pcDNA3-V5MBNL1 together with either 5 g of pcDNA3-MBNL1mycHis or pcDNA3-MBNL1-N-mycHis. Cells were harvested 20-24 h post-transfection by trypsinization followed by neutralization in media contain 10% fetal bovine serum and two washes in 50 mM Tris-HCl, pH7.4, 150 mM NaCl. Cell pellets were resuspended in 50 mM Tris-HCl, pH7.4, 150 mM NaCl, 0.1% IGEPAL with protease inhibitors, and sonicated on i ce (3 X 5 sec). Cell debris was removed by centrifugation at 16,100 x g for 10 min at 4oC. Cleared lysates were treated with 200 g/mL RNase A for 20 min on ice (17) followed by another 10-minute centrifugati on. Cleared lysates were mixed with Dynabeads coupled to Protei n A (Invitrogen) precoated with rabbit anti-V5 polyclonal antibody (Novus, Little ton, CO) and incubated at 4oC for 2 hr. Dynabeads were washed three times with IPP150 buffer (50 mM Tris-HCl, pH7.4, 150 mM NaCl, 0.1% IGEPAL) and once with 50 mM Tris-HCl, pH7.4, 150 mM NaCl. Proteins were dissociated from the beads


33 by heating at 95oC for 2 min in 1X SDS-PAGE sample buffer. Proteins (50% of the immunoprecipitated proteins, 2.5% of the input) were separated on 12.5% SDS-PAGE gels. Immunoblotting was performed using mAb 9E 10 (1:1000) or mAb anti-V5 (1:1000, AbD Serotec). Electron Microscopy The CUG136 RNA, transcribed from pBC-CTG136 plasmid (R. Osborne, University of Rochester), was incubated with MBNL1 protein at molar ratio of 1:2.5 or 1:10 in a buffer containing 16 mM HEPES, 2 mM magnesium acetate, 0.16 mM EDTA 0.4 mM DTT, 1 mM ATP, 50 mM potassium acetate, a nd 16% glycerol for 30 min at 30 C. The resulting complexes were fixed in 0.6% glutaraldehyde for 5 min at ro om temperature and subsequently passed over a 2 ml column containing Bio-Gel A-5M (Bio-Rad, Hercules, CA) equilibrated with 0.01 M Tris (pH 7.6) and 0.1 mM EDTA to remove free prot ein and fixatives. Protein-RNA enriched fractions were incubated with 2.5 mM spermi dine and adsorbed on to glow charged carbon coated copper grids and dehydrat ed in a series of ethanol wa shes of 25, 50, 75, and 100% ethanol each for 5 min at room temperature. Samples were air dried prior to rotary shadow casting with tungsten. Protein-RNA complexes were visuali zed using a FEI Tecnai 12 electron microscope (FEI, Hillsboro, OR) at an accelerating voltage of 40 kV and images were captured using a 4k x 4K Gatan CCD camera using plate film or Ga tan digital image capturing software (Gatan, Pleasanton, CA). Plate film negatives were scanned using an Imacon scanner and supporting software (Imacon, Redmond, WA). Images we re photographed at a magnification of 52K. Photocrosslinking To prepare whole cell lysates for crosslinking, HEK293T c ells were grown in 10 cm plates and transfected with 10 g of pcDNA3-CUGBP1mycHis, pcDNA3-MBNL1mycHis or


34 pcDNA3-MBNL1-NmycHis. Cells were trypsinized and then neutralized in media containing 10% FBS followed by two additional PBS washes. Cells were then resuspended in 250 L of 20 mM HEPES-KOH (pH 8.0), 100 mM KCl, 0.1% IGEPAL and protease inhibitors, sonicated, and lysates centrifuged (13,200 rpm, 10 min, 4oC). Glycerol was added to the supernatants to a final concentration of 20%. RNAs for photocrosslinking were uniformly labeled with 40 Ci each of ( -32P)-GTP and ( -32P)-UTP (800 Ci/mMole) in the presence of 0.5 mM ATP and CTP, 0.02 mM GTP and UTP. Crosslinking was performed by incubating 0.1 pmol RNA with 15 L of HEK293T whole cell lysate in 25 l reactions containing 16 mM HEPES-KOH (pH 8.0), 65 mM potassium glutamate, 2 mM Mg(OAC)2, 0.4 mM DTT, 0.16 mM EDTA, 20 mM creatine phosphate, 2 mM ATP and 16% glycerol (final concentration). Reactions were incubated at 30oC for 15 min, transferred to pre-chilled PCR caps on ice and ph otocrosslinked in Stratalinker (Stratagene, La Jolla, CA) for 2.5 min (three times) with a 3 min interval between each irradiation. Samples were digested with 5 g of RNase A for 20 min at 37oC and immunopurified using the anti-myc monoclonal antibody 9E10 pre-coated protein A Sepharose (Amersham). Purified proteins were fractionated on 12.5% polyacrylamide gels containing SD S (SDS-PAGE) followed by autoradiography. A few modifications were made for the comm itment assay. Cold RNA competitors were synthesized using the Megashortscript T7 k it (Ambion), followed by purification using the Megaclear kit (Ambion) according to manufactur ers instructions. Unlabeled RNAs were resuspended to 20 M in final crosslinking buffer c ontaining 16 mM HEPES-KOH (pH 8.0), 60 mM KCl, 65 mM potassium glutamate, 2 mM magnesium acetate, 0.4 mM DTT, 0.16 mM EDTA, 20 mM creatine phosphate, 2 mM ATP and 16% glycerol. Ten l of unlabeled RNAs (200 pmol) were added to 15 l of HEK293T whole cell lysate e ither at the same time with 0.1


35 pmol radiolabeled RNAs and followed by 1 hr incubation at 30oC, or alternatively 15 min after the incubation of HEK293T whole cell lysate with radiolabeled RNAs, and followed by a 45-min incubation at 30oC. Reactions were photocrosslinked as described above and treated with 20 g of RNase A for 20 min at 37oC. The following immunoprecipitation and SDS-PAGE fractionation were performed as described above. Filter-Binding Assays Unifor mly labeled RNA was prepared as described previously (68). Calibration of the non-specific retention rate of th e nitrocellulose filter was pe rformed by incubating 0.01-0.1 nM RNA at 30C for 30 min in buffer containing 50 mM Tris-HCl (pH8.0), 40 mM KCl, 20 mM potassium glutamate, 15 mM NaCl, 0.5 mM DTT, 0.05 U/ L SUPER-asin (Ambion) followed by filtration through a Bio-Dot (B ioRad) apparatus containing a sandwich of nitrocellulose (BioRad) and Hybond-N plus (Amersham) membrane s followed by a single wash step with the same buffer. The membranes were UV-crosslinked, air dried and exposed to a phosphorimager screen. Non-specific retention on the nitro cellulose membrane was undetectable. Binding reactions were set up in the same buffer with 5 pM RNA and 3.13 X 10-12 M to 1.02 X 10-7 M of MBNL1/41-His, and incubated at 30C for 30 mi n. Each reaction was applied to the Bio-Dot apparatus followed by one wash with binding buffe r. Membranes were processed as described above and signals quantified using ImageQuant TL (Amersham). Standard deviations were calculated based on three independent experiment s and apparent dissociation constants were calculated using a one-site binding mode l and GraphPad Prism (v3.00) software. Electrophoretic Mobility Shift Assays RNA was uniform ly labeled with 40 Ci ( -32P)-GTP or UTP (800 Ci/mMole) in the presence of 0.5 mM ATP, 0.5 mM CTP, 0.02 mM GTP, 0.02 mM UTP and purified using a 5%


36 denaturing gel containing 8 M urea. Prior to use, purified RNA was heated at 65oC for 5 min in 50 mM Tris-HCl (pH8.0), 40 mM KCl, 20 mM potassium glutamate, 15 mM NaCl, 0.5 mM DTT, 0.5 U/ L SUPER-asin (Ambion) following by renatu ration at room temperature. Reactions (20 L) were assembled with 0.1 nM RNA and 0256 nM protein in 50 mM Tris-HCl, pH 8.0, 40 mM KCl, 20 mM KGlutamate, 15 mM NaCl, 15% glycerol, 0.5 mM DTT, 20 g/mL acetylated BSA. After incubation at 30oC for 30 min, reactions were immediately loaded onto a 4% polyacrylamide gel (80:1) containing 0.5 mM DTT and 5% glycerol wh ich had been pre-run at 150V for 1-2 h at 4oC. Gels were run in 0.5X TBE (pH 8.3) at 200V for 2 h, fixed and dried prior to autoradiography.


37 CHAPTER 3 HUMAN MUSCLEBLIND-LIKE 1 PROTEIN REGULAT ES DEVELOPMENTAL SPLICING SWITCHES Introduction: Misregulation of Alternative Pre-mRNA Splicing in Myotonic Dystrophy and Other Human Diseases Overview of Alternative Splicing Alterna tive splicing is a co-transcriptional pr ocess by which different forms of mRNAs are generated from the same gene. At least 70% of human genes undergo alternative splicing, which allows the Human Proteome Initiative (HPI) estimated 1 X 106 human proteins to be encoded by merely 2-2.5 X1 04 protein-encoding genes ( 81). There are five type s of alternative splicing (Figure 3-1), with cassette exon splicing being th e most common type (82) Alternative splicing provides an important mechanism for fine-tunin g protein functions. While protein isoforms usually share long stretches of sequence in co mmon, isoforms encoded by alternatively spliced mRNA species contain distinct sequences which poten tially enable them to carry out slightly, or even drastically, different tasks in the cell. In tricate regulation of alte rnative splicing can often be observed in a cell-type or developmental-st age specific manner. When this regulation goes astray, detrimental effects can occur a nd lead to various human diseases. Alternative Splicing Regulation The general m echanism of pr e-mRNA splicing must be introduced before discussing alternative splicing regulation. An intron of any pre-mRNA is defined by three major cisregulating elements, the 5 splice site, the 3 splice site and the branch point. All pre-mRNA splicing events follow an identical two-step proce ss (Figure 3-2). In the first step, the branch point adenosine carries out a nucleophilic at tack on the 5 splice site, forms a 5-2 phosphodiester bond with the nucleotide at the 5 end of the intron, which results in pre-mRNA


38 cleavage. In the second step, th e newly created 3-OH group of the 5 exon attacks the 3 splice site, which leads to ligation of the two exons and release of the intron lariat (83). Despite this chemical simplicity, these two tran sesterification reactions are carried out by a complicated and dynamic mega-dalton RNA-protei n complex called the spliceosome. The U2dependent spliceosome, also known as the majo r spliceosome, which carries out >99% of splicing in the cell, is asse mbled from U1, U2, U4/U5/U6 small nuclear ribonucleoproteins (snRNPs) and numerous auxiliar y protein factors which associ ate and dissociate with the spliceosome at different stages. During splicing, five spliceosome intermediates, E, A, B, B* and C complexes, have been observed (Figure 3-2). U1 snRNP is first recruited to the 5 splice site through base pairing with the 5 splice site, forming the spliceosomal E complex (Figure 32). Then with the help of the U2 auxiliary factor (U2AF), a heterodimer composed of 35 and 65 kDa subunits, U2 snRNP stably associates with the branch point sequence (BPS), thus forming the A complex. Subsequent recruitment of U4/U 5/U6 tri-snRNP generates the pre-catalytic B complex. Major rearrangements of RNA-RNA and RNA-protein interactions occur in the B complex, resulting in the destabliza tion of U1 and U4 snRNPs and the formation of the catalytic B* complex. The B* complex carries out the fi rst transesterification reaction before it is converted to the C complex, which catalyzes the second transesterfication step. Upon completion of the splicing reac tion, the spliceosome dissembles and the snRNPs are thought to be recycled into other spliceosomes (84). The three major sequences which contribute to defining an intron are usually short and degenerate in mammalian cells with the canon ical 5 and 3 splice site sequences being CAG/guragu (/ is the exon/intron boundary, R is A or G) and uuuyyyyunyag/G (y is C or U, N is any nucleotide), respectively (85). This very limited amount of sequence information poses a


39 great challenge for the spliceosome, which must recognize intron/exon boundaries with great accuracy and fidelity. This conundrum is solved by the integration of an extensive second layer of splicing regulatory elements, including intron ic splicing enhancers/silencers (ISE, ISS), exonic splicing enhancers/silencers (E SE, ESS) and a growing roster of trans -acting protein factors. Together, they modify the strength of individual splice sites, thus determining splice site selection and influencing both constitutive and alternative splice site usage (Figure 3-3). Although the line between constitutive and alterna tive splice sites is not necessarily clearcut, there are subtle qualitative differences between the two groups. In terms of the three primary sequence elements that define a splice s ite, the majority of alternative splice sites are suboptimal in that those sequences deviate more strongly from the consensus sequences compared to constitutive splice sites (82,86). Alte rnative 5 splice sites te nd to deviate from the consensus CAG/guragu at the +4 and +5 positio ns, which would most likely impair efficient interaction with U1 and U6 snRNPs, while altern ative 3 splice sites in general have more purine-rich polypyrimidine tracts which may reduce its affinity for U2AF65 (82,85-87). Interestingly, exons expressed exclusively in neur ons and muscle cells, as well as exons that are developmentally regulated, show even stronger de viation from the consensus compared to other classes of alternative exons (82,86). The suboptimal characteristics of alternat ive splice sites open a window for splicing regulation. Indeed, various intron ic and exonic elements act to promote or inhibit individual splice site utilization thr ough their interactions with protein fact ors. The best characterized ESEs are purine-rich sequences which facilitate spli ceosome assembly by binding to SR proteins, and the best studied ISSs and ESSs are the ones bound by heterogeneous ribonucleoproteins (hnRNPs) (88), such as hnRNP A1 and hnRNP I (also known as polypyrimidine tract binding


40 protein, PTB) (89-92). PTB can bi nd to pyrimidine rich sequences at the 3 splice sites of some alternatively spliced exons, including c-src, -actinin and human -tropomyosin pre-mRNAs, and repress splice site utilization by blocking U2AF recruitment (93). Yet, the simple view of defining SR proteins as splicing enhancers and hn RNPs as splicing silencers failed to explain the regulation of many alternative exons (94-97). Mo st importantly, the same splicing factor can exert opposite functions in the context of different pre-mRNAs, in many cases the outcome depends on the position of the factor binding sites relative to the regulated exons (98). A single exonic/intronic element together with its corresponding transacting factor(s) can regulate splicing autonomously. Ye t, it is not uncommon that an alternative exon is regulated by multiple, often antagonistic, elements (99). The relative concentrations of these transacting factors could dictate the final sp licing outcome. Therefore, ti ssue-specific or developmentalstage-specific splicing patterns mi ght arise, at least in part, as a result of the unique splicing factor expression profile in the cell. A good ex ample is the splicing re programming in neuronal differentiation induced by the PTB/neuronal PT B (nPTB) switch. PTB and nPTB are highly homologous proteins encoded by tw o paralogous genes. However, compared to PTB, nPTB is a significantly weaker splicing repressor for se veral neuron-specific exons (100-102). During neuronal differentiation, PTB prot ein levels decrease while nPTB increases, reprogramming splicing toward neuronal specific patterns (103,104). The alternative splice site choice can be infl uenced not only by the expression levels of certain splicing factors, but also by their posttranslational modifi cations (105), thus opening up a whole new venue for cell signaling in the regulati on of alternative splicin g (106). For example, M-phrase splicing repression during the cell cycle is partly regu lated by splicing factor SRp38. SRp38 is phosphorylated throughout the entire cell cycle except in the M-phase during which


41 SRp38 is dephosphorylated in its RS domain which allows tight binding to U1 70K protein. This interaction prevents U1 70K from binding to ot her SR proteins which are required to activate splicing (105). In summary, alternative splicing regulation involves a complex regulatory network, which requires positive and negative splicing regulators working in conjunction with relatively weak splice sites and flanking intronic and exonic regulatory sequences. Alternative Splicing and Human Diseases Dysregulate d alternative splicing is a majo r cause of human disease. Mutations in ciselements as well as trans -acting factors lead to diseaseassociated splicing patterns. Cis -acting mutations were originally underestimated to account for ~15% of human diseases. This estimate only reflects mutations that disrupt known spli ce sites (107). Recent re assessment of the figure revealed that >50% of disease-causing mutations result in aberrant sp licing (108). Based on a probabilistic model, it ha s been proposed that ~60% of human diseases could be caused by cis acting mutations that disrupt splicing codes (109). Although not as prevalent, mutations affecting trans-acting splicing f actors also affect the integrit y and dynamics of the splicing machinery and result in aberrant splicing. A number of examples are listed in Table 3-1. Myotonic Dystrophy, Aberrant Splicin g and Muscleblind-like P roteins Myotonic dystrophy (DM) provides an excellent example where mis-regulated alternative splicing causes inherited human di sease. At the molecular leve l, at least 21 pre-mRNAs have been shown to undergo aberrant splicing in DM tissues (110). Although the physiological readouts for the majority of the misregulated sp licing events remain to be elucidated, direct cause-effect connections have been drawn be tween several aberrant splicing events and characteristic clinical features of DM. Aberrant retention of exon 7a in CLCN1 mRNA introduces a pre-mature stop codon (PTC) whic h targets the mRNA to the nonsense-mediated


42 decay (NMD) pathway (111,112). Reduction in CLCN1 mRNA levels leads to loss of this major skeletal muscle chloride channel, accounting for the reduced Cl conductance and myotonia in DM patients. Another example of mis-splicing in DM is the insulin receptor (IR). Elevated IR exon 11 inclusion in DM skeletal muscle result s in predominant expression of an insulininsensitive isoform of IR which leads to the insulin insensitivity observed in DM patients (113,114). Although the splicing patterns of many pre-mRNAs are abnormal in DM, there is no general affect on splicing efficien cy or fidelity. In fact, DM se lectively affects a group of exons that undergo developmental splicing switches as illustrated in Figure 3-4. Interestingly the aberrant splicing pattern in DM invariably co rrelates with the corres ponding embryonic splicing pattern. Thus, the activ ity of a certain splicing factor which is responsible for maintaining normal adult splicing patterns may be missing in DM. Several studies have shown that CUG triplet repeat RNA binding protein 1 (CUGBP1) is able to regulate the alternative splicing of a number of misregulated exons in DM, includ ing cTNT exon 5, IR exon 11 and CLCN1 exon 7a (112,113,115). However, the embryonic or DM-lik e splicing patterns are seen upon CUGBP1 over-, instead of under-expressi on, suggesting that DM is not caused by CUGBP1 loss-offunction. In contrast, the MBNL1 sequestration m odel proposes that MBNL1 loss-of-function underlies DM pathogenesis, and is supported by both in vitro and in vivo evidence. Yet the normal function of MBNL1 was unknown when I bega n this project. As a working hypothesis, we proposed that MBNL1 functions to directly regulate developmental sp licing switches and loss of MBNL function leads to th e splicing defects in DM.


43 Results Skeletal Muscle Fast Tropon in T (Tnnt3) Undergoes a Po stnatal Developmental Splicing Switch Tnnt3 encodes a skeletal muscle specific tropon in T. Alternative splicing of fast troponin T (Tnnt3) is abnormally regulated in DM patients as well as in Mbnl1E3/E3 mice in that the fetal exon shows an elevated inclusion rate in DM patie nts (69). To test if alternative splicing of Tnnt3 is developmentally regulated, reverse tran scription-polymerase ch ain reaction (RT-PCR) analysis was performed using RNAs isolated from the quadriceps of wild type (WT) mice aged from embryonic day 14.5 (E14.5) to postnatal da y 28 (P28). PCR primers were positioned in exons 2 and 11. A number of exons represen ted as the black boxes in Figure 3-5 are developmentally regulated, and the alternative sp licing of these exons contributed to multiple PCR products (Figure 3-5, uncut). It is interesting to note that the ove rall splicing of Tnnt3 undergoes a developmental transition postnatally in that multiple splice isoforms gradually consolidate into a single major isoform that predominate in P28 mouse muscle. To assay specifically for the splicing transition of the T nnt3 fetal exon, PCR product s were digested with BsrBI, which had a unique cleavage site inside th e fetal exon. After Bs rBI digestion, isoforms including/excluding the fetal exon co uld be distinguished according to their sizes. In P2 mice, the fetal exon was included in > 90% of Tnnt 3 mRNAs whereas the fetal exon inclusion rate dropped nearly to zero in P28 mi ce, reflecting a narrow time window for this splicing transition. A high fetal exon inclusion rate in both neonates, as well as DM pati ents, is also consistent with the finding that DM patients adopt the fetal splicing pattern for a specif ic group of pre-mRNAs. MBNL1 Promotes Tnnt3 Fetal Exon Exclusion To test if MBNL1 can regulate the a lternative splicing of Tnnt3 fetal exon, a Tnnt3 minigene was constructed by inserting the genomic region between mouse Tnnt3 exon 8 and


44 exon 9 into the vector pSG5 (Stratagene). The mi nigene reporter was transfected into 293T cells together with plasmids expressing green fluorescent protein (GFP) or GFP-tagged RNA-binding proteins. One to two days post-transfection, R NAs were isolated from the cells and RT-PCR was performed using primers flanking Tnnt3 E8 and E9. In the presence of GFP alone, the minigene showed a default <5% fetal exon excl usion. Overexpression of either GFP-MBNL1 full length or N-terminus, which contains th e four CCCH motifs res ponsible for RNA binding (116), dramatically increased the fetal exon exclusion rate to >90% (Figure 3-6 A). This effect was specific to MBNL1, as equivalent amount s of GFP-CUG-BP1 or GFP-hnRNP A1 which often inhibits alternative exon inclusion(91,99,117) failed to significantly change the splicing pattern (Figure 3-6 A). Furthermore, the ex tent of fetal exon exclusion was MBNL1 dosage dependence. When increasing amounts of GFP-MB NL1 plasmids were cotransfected with the minigene, the percentage of fetal exon exclus ion positively correlated with the level of GFPMBNL1 proteins (Figure 3-6 B) Therefore, MBNL1 functionally promotes Tnnt3 fetal exon exlusion. MBNL1 Physically Interacts with Tnnt3 Pre-mRNA Although MBNL1 regulated Tnnt3 f etal exon sp licing, it remained unclear whether the effect was primary or secondary. Since RNAprotein interactions ar e a pre-requisite for alternative splicing regulation, potential MBNL1-Tnnt3 pre-mRNA interactions were examined using a UV crosslinking approach. HEK293T ce lls were transfected wi th protein expression plasmids encoding myc-tagged versions of CUGBP1, full-length MBNL1 (MBNL1 FL) or the MBNL1 N-terminal region (MBNL1 N). Followi ng transfection, cell lysa tes were incubated with radiolabelled Tnnt3 RNAs encompassing exons 7-9 or different 500-645 nucleotide (nt) subregions designated T1-6 (Figure 3-7 A), phot ocrosslinked with UV-light and digested with RNase A. Protein-RNA complexes were th en immunopurified with the anti-myc monoclonal


45 antibody (mAb) 9E10 and resolved by SDS-PAGE. Because these studies indicated that only Tnnt3 T5 RNA (500 nt) bound to both MBNL1 FL and MBNL1 N proteins, this region was further subdivided into T5.1-T5.45. Interest ingly, only T5.45 RNA (200 nt) crosslinked to CUGBP1 and MBNL1 proteins while T5.1 (125 nt), and the other subr egions (T5.2 and T5.3, data not shown), did not (Figure 3-7 B). In agreement with pr ior studies, CUGBP1 failed to crosslink to a CUGexp, (CUG)54, while both MBNL1 FL and MBNL1 N did (68). MBNL1 Binding Sites Are Required for MBNL1-Mediate d Splicing Regulation A recent mapping study identified several MBNL 1 binding sites containing a core element within human and chicken cardiac troponin T (cTNT/TNNT2) (118). Alignment of these sequences revealed a hexanucleotide consensu s motif (5-YGCUU/GY-3, Y-C orU). A highly similar element (TGCGCTT) was found in Tnnt3 pre-mRNA immediately upstream of the fetal exon 3 splice site. In order to test whether an MBNL1 bindi ng site was essential for MBNL1mediated splicing regulation, a 10 nucleotide region encompassing the TGCGCTT element was deleted in the Tnnt3 minigene (Figure 3-8 A). Crosslinking assays were then used to test whether wild type and mutant Tnnt3/98 RNA (a 98-nt subregion encompassing 83-nt of the 3 end of intron 8 and 15-nt of the F exon) was recognized by both CUGBP1 and MBNL1 (Figure 3-8 B, lower panel). Concurrently, wild type or mutant Tnnt3 minigenes were transfected into C2C12 cells to assay whether Tnnt3 splicing remained respon sive to MBNL1 overexpression (Figure 3-8 B, upper panel). These assays were performed using C2C12 myoblasts since they showed a higher default level of Tnnt3 F exon sk ipping compared to 293T cells. The resulting 10 mutant showed much reduced MBNL1 crossli nking compared to WT minigene (Figure 3-8 B, lower panel), suggesting at least one of the MBNL1 binding sites on Tnnt3 pre-mRNA was disrupted. When tested in the cotransfection assay, the 10 mutant showed a corresponding


46 impairment of F exon skipping promoted by MBNL1 overexpression (Figure 3-8 B, upper panel). Indeed, the 10 deletion eliminated F exon skipping in cells transfected with the Tnnt3 minigene alone or with CUGBP1-mycHis together with the minigene (Figure 3-8 B, RT-PCR) suggesting loss of MBNL1 binding and enhanced sp liceosome recruitment to the F exon region. Although MBNL1 binding was not completely elim inated by deleting the 10 nucleotides, the reduced fetal exon splicing response upon MBNL 1 overexpression suggests that the MBNL1 binding site was essential for MB NL1-mediated splicing regulation. Conclusions, Discussion and Future Work MBNL1 Protein Regulates Alternative Splicing MBNL1 bo und directly to an intronic sequen ce immediately upstream of the Tnnt3 fetal exon to promote fetal exon exclusion. Furthe rmore, an intact MBNL1 binding site was necessary for efficient enhancement of fetal exon exclusion promoted by MBNL1. Therefore, we concluded that MBNL1 functions as a potent alternative splicing factor that promotes the adult splicing patterns for specific target pre-mR NAs. This function of MBNL1 appears to be conserved from the Drosophia homolog, which regulates the splicing of -actinin (119). Although only one pre-mRNA, Tnnt3, was shown in this chapter to be regulated by MBNL1, a number of other transcripts including cTNT, IR and sarcop lasmic reticulum Ca2+-ATPase 1 (Serca1) have been shown to be direct MBNL1 targets (118,120). For al l of these pre-mRNAs, MBNL1 invariantly promotes the adult splicing pattern, regardless of whether splicing involves the inclusion of an adult-specific exon, as is th e case for IR and Serca1, or the exclusion of a fetal exon, as is the case for Tnnt3 and cTNT. Interestingly, the posit ions of MBNL1 binding sites relative to these regulate d exons correlate with the splicing patterns promoted by MBNL1. For example, in the case of Serca1, wher e MBNL1 promotes adult exon inclusion, MBNL1


47 binding sites have been mapped to intronic sequen ces downstream of the alternative exon (120). On the other hand, for cTNT and Tnnt3 where MBNL1 promotes exon exclusion, MBNL1 binds to intronic sequences immediately upstream of th e respective fetal exons. A parallel scenario was found in the case of another alternative splicing factor, neuro-oncological ventral antigen 1 (Nova1), where the splicing patterns promoted by Nova1 could be accurately predicted based on the positions of Nova binding sites (98,121). As with MBNL1, Nova1 usually promotes exon exclusion when binding immediately upstream of the regulated exon while it promotes exon inclusion when binding downstream of the alte rnative exon. Although th e mechanisms by which these two proteins regulate alte rnative splicing remain elusive, there could be significant similarities between them. For instance, both proteins may promote exon exclusion by binding upstream of the 3 splice site and blocking U2 AF recruitment and subsequent spliceosome assembly. Further studies are required to elucidate the mech anism by which MBNL1 promotes exon inclusion/exclusion. MBNL2 and MBNL3 have been shown to promote adult splicing patterns to the same extent as MBNL1 in the cases of cTNT, IR a nd Tnnt3 minigenes ((118) and data not shown). While MBNL3 expression is mainly confined to placenta among adult human tissues, MBNL2 expression, as assayed by Northe rn blotting, is higher than MBNL1 across a variety of tissues both in mouse and in human (79,80). Yet intriguingly, Mbnl2 knockout mice failed to recapitulate major DM symptoms (122). The reason for this failure is unknown. Due to the lack of a good MBNL2 antibody, the levels of MBNL2 prot ein in tissues remain to be assessed. One possibility is that MBNL2 translation is repr essed although it is being actively transcribed, so that the major MBNL protein across tissues is MBNL1. A similar situation has been described for PTB and nPTB, where both genes are actively transcribed in neurona l precursor cells, yet


48 nPTB is under translational repression by micr o RNAs (miRNAs), so that only PTB protein predominates in undifferentiated neuronal precurs ors and represses neuronal-specific splicing patterns (104,123). The Role of MBNL1 in DM Pathogenesis The MBNL sequestration m odel proposes that the expression of expanded CUG or CCUG repeats in DM are toxic because they titrate ce llular MBNL proteins, leading to their loss-offunction and subsequent DM phenotypes. Alt hough several lines of evidence support the hypothesis that pathogenic R NAs functionally sequester MB NL, there was a missing link between MBNL loss-of-function and DM phenotypes. In this study, I have demonstrated that MBNL1 acts as an alternative sp licing factor and promotes the adult splicing pattern of DMaffected transcripts. As i llustrated in Figure 3-9, pathogenic RNAs carrying expanded CUG or CCUG repeats titrate MBNL, which leads to reduced levels of available MBNL proteins in the nucleoplasm. Thus, antagonistic splicing factor s, such as CUGBP1, can more efficiently promote the fetal or DM-like splicing patterns of certain transcripts in adult tissues. The resulting ectopic expression of fetal isoforms in various adult tissues and organs eventually leads to the multisystemic symptoms observed for DM patients. The Importance of Developm enta l Splicing Regulation Although extensive work has been done on deve lopmentally regulated gene transcription, the developmental regulation of splicing is an area that has been overlooked in the past. More evidence has emerged during the last decade, which shows that a number of transcripts undergo splicing switches at a certain stage during development. Differ ences between fetal and adult splicing patterns, as in the case of CLCN1 have been shown to affect mRNA stability, leading to drastically different protein levels. Other mis-splicing events, as in the case of IR have been shown to generate functionally di stinct protein isoforms that ar e specifically compatible with


49 either fetal or adult tissue functions (111-113). Howe ver, the direct effects of the majority of the developmental regulated splicing events remains to be explored. Even for those splicing events with known functional consequen ces, little is know about their biological relevance. For example, why do neonatal muscles require fewer ch loride channels to function properly? Are there major differences in the molecular environmen t between fetal and adult tissues so that only fetal/adult isoforms would fit in the fetal/adult environment? How conserved are these events evolutionarily? At what time dur ing development do specific splic ing switches occur? Do all developmental splicing switches happen during th e same time window or are there multiple sets of such changes taking place during developmenta l progression? Are MBNL/CELF proteins the master regulators of all such sw itches? If not, what are the ot her pathways? Systematic and high-throughput approaches need to be utilized to identify de velopmentally regulated splicing events across tissues and organs in various model organisms. Functional grouping of these events could potentially shed light on their biological significance. If, hypothetically, the majority of these events involve proteins functioning in muscle contraction and neuronal signaling, the functional significance of these splicing switches might lie mainly in assisting animals to adapt to the distinct demands of fe tal versus adult lives associated with changing mobility and evolving tasks of learning and me mory. Combined with human genetic analysis, single nucleotide polymorphisms (SNPs) at these developmentally regulated splice sites could be assayed for their correlation with human diseases. Future Work Although it is now clear that MBNL1 functions as an alternative spli cing factor, the norm al splicing targets of MBNL1, as well as the mechanism by which it promotes exon inclusion/exclusion, rema ins unclear. Using an in vivo cross-linking and immunoprecipitation technique (124), a greater numbe r of MBNL1 targets, as well as MBNL1 binding sites on those


50 targets, can be identified in a more comprehens ive manner. It would be interesting to test whether those newly identified pre-mRNAs also undergo developmental splicing switches, and whether these switches are MBNL1-dependent. To address the mechanistic question, in vitro splicing assays of MBNL1 targets could be perfor med in the presence or absence of MBNL1 so that subsequent formation of spliceosomal inte rmediate complexes could be compared which may give clues to the mechanis m of MBNL1 mediated splicing re gulation. In the cases where MBNL1 promotes exon inclusion, additional factor s might be recruited by MBNL1 to enhance spliceosome assembly. Immunoprecipitation could be carried out to iden tify protein factors associated with MBNL1 in the cell which may mediate MBNL1-enhanced splicing events.


51 Table 3-1. Splicing fact ors and human diseases. Disease Affected splicing factor Comments Neuromuscular Diseases Myotonic dystrophy (DM) Muscleblind-like protein 1, 2 and 3 (MBNL1, MBNL2, MBNL3) Sequestration of human muscleblind proteins in DM foci disrupts MBNL mediated developmental stage-specific alterna tive splicing (68,69). Myotonic dystrophy (DM) CUG triplet repeat RNAbinding protein 1 and 2 (CUG-BP1, CUG-BP2) CUG-BP1 and CUG-BP2 re gulate alternative splicing in ways antagonistic to muscleblind proteins; elevated CUG-BP1 protein level has been reported in DM1 patients (112,115,125-127). Spinal muscular atrophy (SMA) Survival of motor neuron 1 and 2 (SMN1, SMN2) Deletion/mutations in SMN1 and SMN2 genes results in insufficient level of SMN, leading to abnormal snRNP biogenesis (128,129). Spinal cerebellum atrophy 2 (SCA2) Ataxin-2 Ataxin-2 is a RNA-binding protein that interacts with neuronal splicing factor A2BP1 (Fox-1) (130). Spinal cerebellum atrophy 8 (SCA8) MBNL? Drosophila muscleblind was identified as one of the poteintial modifier of a fly SCA8 model (131). Spinal cerebellum atrophy 10 (SCA10) Polypyrimidine tract binding protiein (PTB)? AUUCU repeat RNA may sequester PTB; potential splicing defects in patients have not been tested (132). Huntingtons diseaselike 2 (HDL2) MBNL MBNL1 shows decreased nucleoplasmic level while is enriched in CUG repeats containing nuclear foci (133).. Fragile X tremeor ataxia syndrome (FXTAS) Heterogeneous nuclear ribonucleoprotein A2 (hnRNP A2), MBNL1 hnRNP-A2 and MBNL1 are both enriched in the CGG repeat RNA containing nuclear inclusions (134). Fascioscapulohumoral dystrophy (FSHD) FRG1 Deletion of FRG1 transcriptional silencer leads to FSHD; mouse model overexpressing FRG1 phenocopy FSHD (135). Paraneoplastic opsoclonus-myoclonusataxia (POMA) Neurooncologic ventral antigen 1 and 2 (Nova-1 and Nova-2)) Nova is inactivated by autoantibodies; Nova knockout mice phenocopy the human POMA (136,137). Other Neurological Disorders Isolated case of mental retardation and epilepsy Ataxin-2 binding protein (A2BP1 (FOX-1)) A chromosome translocation disrupted A2BP1. Splicing patterns have not been tested in patients (138). Retinitis pigmentosa (RP) Precursor mRNAprocessing factor 3, 8, 31 (PRPF3, PRPF8, PRPF31) Loss of constitutive splicing factors PRPF3, PRPF8, PRPF31 leads to RP (139-141).. Cancer Leukemias and sarcomas Translocation liposarcoma (TLS) Fusion of TLS with ETS-related gene (ERG) or FUS eliminates its interaction with SR proteins, inhibiting SR protein mediated CD44 splicing (142,143). Azospermia RNA binding motif protein, Y-linked, family 1, member A1, RBMY interacts with Tra2 and regulates alternative splicing (144). Hepatocellular carcinoma HCC1 Autoantibodie s against HCC1 were detected in hepatocelluar carcinoma patients (145).


52 Table 3-1. Continued. Disease Affected splicing factor Comments Papilary renal cell carcinoma Splicing factor, prolineand glutaminerich (SFPQ) Fusion of SFPQ and p54nrb to TFE3 gene was identified in papillary renal cell carcinoma lines (146). Developmental Disorders Rett syndrome (RTT) methylation-dependent transcriptional repressor methyl-CpG binding protein 2 (MeCP2) MeCP2 regulates alternative splicing through interaction with RNA-binding protein YB-1 (147). Prader Willi Syndrome (PWS) HBII-52 Loss of HBII-52 correlates with aberrant serotonin receptor splicing in PWS patients (148). Hay-Wells syndrome p63 alpha p63 alpha mediates FGFR-2 alternative splicing through interaction with ABBP1 (149).


53 Figure 3-1 Types of alternative splicing events. Constitutive exons are shown in blue, alternative exons are shown in grey, purple, red, yellow, green and orange. Introns are shown as grey horizontal lines connecting exons. Th in grey lines indicate splicing events.




55 Figure 3-2 Splicing Mechanism. U1 snRNP is fi rst recruited to the 5 splice site through base pairing with the 5 splice site, forming the spliceosomal E complex. With the help of U2AF, U2 snRNP stably associates with the branch point sequence (BPS), which forms the A complex. Subsequent recruitment of U4/U5/U6 tri-snRNP generates the pre-catalytic B complex. Major rearra ngements of RNA-RN A and RNA-protein interactions occur in the B complex, resul ting in the destablization of U1 and U4 snRNPs and the formation of the catalytic B* complex. The B* complex carries out the first transesterification reaction before it is converted to C complex, which catalyzes the second transe sterfication step. Upon completion of the splicing reaction, the spliceosome dissembles and the sn RNPs are thought to be recycled into other spliceosomes.


56 Figure 3-3 Regulation of alternative splicing. Constitutive exons are shown as grey bars. The alternative exon is shown in blue. Introns are shown as black lines connecting the exons. Intronic/exonic splicing silencers (ISS, ESS) and corresponding transacting factors are shown in red. Intronic/ex onic splicing enhancers (ISE, ESE) and interacting factors are shown in green.


57 Figure 3-4 Illustration of aberrant splicing pa tterns in DM. Constitutive exons are shown in blue and alternative exons ar e shown in orange. Introns ar e shown as black horizontal lines connecting the exons. Black th in lines indicate splicing events.


58 Figure 3-5 Tnnt3 undergoes a postnatal developm ental splicing switch. In WT mice, RTPCR analysis of fast skeletal muscle tr oponin T (Tnnt3) mRNA showed alternative splicing of the fetal (F) exon as well as exons 4 through 8. Primers used for PCR are indicated as black arrows in Tnnt3 exon 2 and 11. Transition from inclusion to exclusion of the fetal exon occurred mainly between P2 and P16. The lower panel shows RTPCR products after digestion with BsrBI, which cleaves within the fetal exon. E14.5 is whole limb while P2-P28 are dissected quadriceps.


59 Figure 3-6 MBNL1 promotes Tnnt3 fetal exon exclusion. The Tnnt3 minigene was cotransfected into HEK293T cells with plasmids expressing GFP or GFP-tagged RNA-binding proteins. RT-PCR shows the ratios of Tnnt3 fetal exon inclusion/exclusion. Expression levels of GFP fusion proteins are shown by immunoblotting with anti-GFP and anti-MB NL1 antibodies. GAPDH is the protein loading control. A ) GFP-MBNL1 promotes Tnnt3 fetal exon exlusion. Overexpression of GFP-MBNL1 full length as well as the N-terminus increased fetal exon exclusion rate compared to overexpr ession of GFP alone. GFP-hnRNP A1 or GFP-CUGBP1 do not promote fetal exon excl usion while expressed at comparable levels with GFP-MBNL1. B ) Tnnt3 fetal exon exclusion is MBNL1 dosagedependent. Increasing amount s of GFP-MBNL1 were expressed in HEK 293T cells. Tnnt3 fetal exon exclusion rates display a corresponding increase with cellular MBNL1 levels. Overexpression of MBNL 1 does not decrease CUGBP1 levels.


60 Figure 3-7 MBNL1 crossli nks to Tnnt3 pre-mRNA. A ) Mapping of MBNL1 binding sites on Tnnt3. Illustration shows the Tnnt3 genomic region between exons 7 (E7) and 9 (E9) (thin lines, introns; black boxes, exons) and the subregions corresponding to the RNAs tested for MBNL1 crosslinking (thick lines, T1-6; thin lines, T5.1-T5.3; green line, T5.45). B ) Interaction sites between MBNL1 and CUGBP1 on Tnnt3 premRNA are positioned in a region which includes the F exon and upstream intron 8. HEK293T cells were transfected with myc-His tagged CUGBP1, MBNL1 FL or MBNL1 N followed by incubation of the corresponding cell lysates with 32P-labeled Tnnt3 T5.45, T5.1 or (CUG)54 RNAs and subsequent UV-light induced crosslinking and RNase A treatment. Labeled protei ns were detected by SDS-PAGE and autoradiography (upper panels, crosslink) wh ile protein levels in the lysates were detected by immunoblotting with the anti-myc mAb 9E10 (lower panel, protein). The primary structures of MBNL1 FL and MBNL1 N are illustrated below the immunoblot with the four CCCH motifs highlighted (shaded rectangles).


61 Figure 3-8 MBNL1 binding sites are required for MBNL1-mediated splicing regulation. A ) Schematic views of the 10 nt deletion. B ) The 10 mutant shows reduced MBNL1 crosslinking as well as MBNL1-mediated fetal exon exclusion. C2C12 myoblasts were used in the cotransfection assay since they show a higher default level of fetal exon exclusion compared to 293T cells.


62 Figure 3-9 DM pathogenesis model. Pathoge nic RNAs form hairpins and titrate MBNL proteins in DM cells. With reduced levels of available MBNL, antagonistic factors, such as CUGBP1, can more efficiently promote embryonic splicing patterns for certain pre-mRNAs in adult tissues, which leads to the DM phenotype.


63 CHAPTER 4 POTENTIAL MECHANISM FOR FUNCTIONAL SEQUE STRATION OF MBNL1 IN DM Introduction In the last chapter, evid ence was presented supporting the idea that MBNL1 functions as an alternative splicing factor that regulates developmental splicing switches. A model was suggested in Figure 3-8 in whic h pathogenic RNA repeats sequester MBNL1 protein, leading to its loss-of-function and the downstream splicing def ects. However, one question remains: what is the mechanism of MBNL1 functional sequestra tion in DM? In other words, what is the difference between the interactions between MBNL1 and pathogenic RNAs versus splicing targets? Several possibilities exist. Firs t, MBNL1 might be func tionally sequestered by CUG/CCUG repeats due to higher MBNL1 binding capacitie s on these pathogenic RNAs compared to those on MBNL1 normal splicing targ ets. However, this speculation is not compatible with the study by Ho et al where 960 CUG or CAG rep eats were transfected into COSM6 cells and assayed for their abilities to disrupt splicing {H o, 2005 #11}. While MBNL1 relocalizes to both CUG and CAG RNA foci, only (CUG)960 disrupted MBNL1-mediated splicing. The observation that C AG RNA repeats, which are of the same length and expression level as CUG repeats and thus might harbor co mparable numbers of MBNL1 binding sites, failed to functionally titrate MBNL1 pr otein suggests that the expanded length of pathogenic RNAs is not the key reason underlying MBNL 1 functional sequestration. Two hypotheses remain, which are both based on the potential qualita tive differences of MBNL1 interactions with pathogenic RNAs versus normal splicing targets. One proposes that MBNL1 is functionally sequestered due to its higher affinities for pathogenic RNA repeats compared to normal splicing targets. The other hypothesis proposes the di stinct stabilities of ribonucleoprotein (RNP) complexes formed be tween MBNL1 and pathogenic RNAs versus


64 normal splicing targets. However, little was kno wn about MBNL1-RNA interactions. Previous studies have suggested that the CCCH motifs (CX7CX4-6CX3H) in MBNL proteins are related to those found in tris-tetrapoplin (TTP)/TIS11, a factor that pr omotes the deadenylation and turnover of target mRNAs containing a 3 untrans lated region (3 UTR) class II AU-rich element (ARE) {Blackshear, 2002 #415}. Similar to TTP /TIS11d, the CCCH motifs in MBNL proteins are organized as tandem pairs separated by a 14-16-residue linker {M iller, 2000 #39; Nykamp, 2004 #420; Pascual, 2006 #225}. While TTP/TIS11d recognize single-stranded AU rich RNA sequences, MBNL1 has been shown to bind to double-stranded pathogenic RNAs as well as predicted single-stranded regions in cTNT pre-mRNA {Ho, 2004 #10}. However, the affinity of MBNL1 for different RNAs and the molecular details of MBNL1-RNA complexes remained to be characterized. In this chapter, a combinati on of chemical and biochemical methods were used to examine various aspects of MBNL1-RNA inte ractions. These resu lts support the proposed mechanism of MBNL1 func tional sequestration. Results MBNL1 Interacts with an Intronic Stem-Loop Structure U pstream of the Tnnt3 Fetal Exon As mentioned in the previous chapter, identification of MBNL1 binding sites on cTNT revealed a hexanucleotide consensus motif (5-YGCUU/GY-3), which was located in a predicted single-stranded (ss) RNA regions. To gether with prior observations that MBNL proteins are sequestered by CUGexp and CCUGexp hairpins in ribonuclear foci in DM cells {Mankodi, 2003 #131}{Mankodi, 2005 #124}{Mankodi, 2001 #38}{Miller, 2000 #39}, this finding suggests that the MBNL proteins bind to both ssRNA a nd dsRNA structural motifs. To determine if MBNL1 recognizes primarily ssRNA targets in other splicing precursors, the MBNL1 binding site on fast skeletal muscle troponin T (Tnnt3) RNA was mapped in greater detail. First, the structure of a 151-nt subr egion of T5.45 was chemically and enzymatically


65 probed. The Mfold {Zuker, 2003 #315} predicted stru cture of this RNA fragment is shown in Figure 4-1 A, right panel. A set of three wellcharacterized chemical and enzymatic probes was used for RNA structure probing. Lead ion (P b) induces RNA cleavage at ssRNA as well as relaxed dsRNA regions. Ribonuclease T1 cleaves G-nucleotides present in ssRNA regions while S1 nuclease also cleaves ssRNA regions but without base specificity. Two different concentrations of each probe were used in order to discriminate cleavages site with different susceptibilities. Based on the lead ion and nuc lease sensitivity pattern s, the Mfold predicted structure was confirmed. Interestingly, th e 10-nt region which was important for MBNL1mediated fetal exon exclusion as discussed in th e previous chapter, is located in an 18-nt stemloop region, as indicated by the strong S1 nu clease and RNase T1 cleavage signals at G94 and reduced nuclease sensitivities in the surround ing regions ranging from G82 to G107 (Figure 4-1 A). Deletion of the 10-nt region is predicted to eliminate the hairpin structure {Zuker, 2003 #315}. The crosslinking and splici ng results obtained with the 10 mutant indicated that the 18nt hairpin was a binding site for MBNL1 and th at MBNL1-hairpin interactions might promote fetal exon skipping. Because both sequence and stru ctural elements could contribute to efficient MBNL1 binding, a double C G and U G mutant was generated that eliminated the C-C and mismatch in the 18-nt hairpin, which also in creases the stability of this stem-loop, and furthermore substituted a G for a U in the loop. These mutations are not predicted to alter the overall folding pattern of Tnnt3/98. This double poi nt (gg) mutant showed considerably reduced MBNL1 crosslinking compared to wild type Tnnt 3 (Figure 4-1B) confirming that this region was an MBNL1 binding site. Interestingly, a similar decrease in MBNL1 crosslinking activity was also observed for the single C G stem substitution mutant (data not shown). While loss of MBNL1 binding should promote fetal exon splicing, inclusion activity was completely


66 eliminated (Figure 4-1C, right panel). Loss of fetal exon splicing activ ity could result from enhanced hairpin stability, or an increase in the purine content of the polypyrimidine (Py) region upstream of the fetal exon 3 splice site, and subsequent impairment of spliceosome asembly. Interestingly, another double mutant (G A and C U), which was predicted to preserved the 18-nt stem-loop structure while reducing the GC content in the stem, also showed reduced MBNL1 cross-linking compared to wild type (Figure 4-1B, au mutant) while MBNL1-induced fetal exon skipping activity wa s reduced similar to the 10 mutant (Figure 4-1C, au mutant). Overall, these studies suggest that MBNL1 pref ers to bind to a GC-rich stem-loop containing a pyrimidine mismatch(es) in a normal splicing target. MBNL1 Binds to the Stem Re gion of a Pathogenic dsRNA Muscleblind-like proteins were originally characterized as nuclear factors which are recruited by CUGexp RNAs {Miller, 2000 #39}. The predicte d double-stranded nature of these pathogenic RNAs has been validated by chemi cal and nuclease mapping, thermal denaturation and electron microscopy {Napierala, 1997 #479; Sobczak, 2003 #112; Michalowski, 1999 #111}. While the binding of MBNL proteins to CUGexp RNA is proportional to the predicted stem length, there is currently no direct experimental evidence th at MBNL binds directly to the CUGexp stem or that the RNA structure remains in the hairpin configuration following MBNL binding. Therefore, we initially used chemical and enzymatic structure probing of labeled RNAs to identify MBNL1 binding sites on (CUG)54 RNA. RNAs were 5 end-labeled, subjected to either lead ion (Pb)-induced hydrolysis or RNas e T1 digestion in the absence or presence of recombinant MBNL1, and the products were fractionated on denaturing polyacrylamide gels. As shown previously, lead ions cleave both ssRNA and relaxed dsRNA struct ures which yielded a uniform ladder that increased in intensity with increasing lead concentr ation (Figure 4-2, left


67 panel). As anticipated, addition of MBNL1 i nhibited strand cleavage in a concentrationdependent manner and densitometry analysis faile d to show significant re gional differences in the cleavage pattern by MBNL1 suggesting uniform binding of this protein throughout the stem region. In contrast to lead, RN ase T1 prefers to cleave after G nucleotides in single-stranded regions. Thus, incubation with RNase T1 resulted in strong cleavage at th e terminal loop (Figure 4-2, right panel, G26-G29). In terestingly, terminal loop clea vage was unaffected by MBNL1 addition while stem cleavage was uniformly inhi bited. Therefore, we concluded that MBNL1 interacts primarily with the stem region of CUGexp RNAs. Similar Affinities of MBNL1 for Sp licing Precursor And Pathogenic RNAs According to the RNA sequestration m odel, pathogenic RNAs out compete normal RNA binding targets for MBNL1 leading to loss of MB NL-mediated regulation of alternative splicing during postnatal development. However, it is not clear why MBNL1 accumulates on DM1 and DM2 expansion RNAs in ribonuclear foci. The mo st straightforward expl anation is that MBNL1 has a higher affinity for DM pathogenic RNAs compared to its physiological RNA splicing targets. Since the crosslinking of MBNL1 to the Tnnt3 fetal exon 3 splice site region and CUGexp RNAs appeared to be comparable (Figure 3-7B), we determined the relative affinities of MBNL1 for (CUG)54, (CAG)54 and Tnnt3/5.45 using recombinant MBNL1 protein in filter binding and gel shift assays. The MBNL1 proteins used for this study we re either MBNL1 FL or the C-terminal truncation mutant MBNL1 N. For the MBNL1 FL preparation, ~60% of the purified protein was full-length while the MBNL1 N protein prep aration was homogeneous as determined by Coomassie blue staining and immunoblot analysis (Figure 4-3A). Because MBNL1 shows a temperature-dependent binding profile in cell extr acts in the absence of ATP (data not shown), all recombinant protein binding studies were performed at 30C to maximize binding while


68 minimizing RNA degradation. As anticipated, MB NL1 FL showed a high affinity in the filter binding assay for (CUG)54 ( Kd=5.3 nM), but it also showed high affinities for the other RNAs examined including Tnnt3 5.45 (Tnnt5.45) ( Kd=6.6 nM) and (CAG)54 ( Kd=11.2 nM) (Figure 4-3B). While there is no statistically significant difference between the Kds of MBNL1 to (CUG)54 and T5.45 (p=0.119), the Kd values for MBNL1 binding to (CUG)54 and T5.45 are significantly different from the Kd of MBNL1 binding to (CAG)54 (p=0.028 and p=0.002, respectively). Given the ~two-fold differences in the values of these Kds, this result suggests that MBNL1 binds to (CAG)54 with a different, yet similar, affinity compared to (CUG)54 and T5.45. The similarities in the binding affinities of MBNL1 for (CUG)54 and (CAG)54 accounts for the prior observation that overexpre ssion of either of these repeat RNAs in COS-M6 cells results in the formation of nucl ear foci that colocaliz e with GFP-MBNL1 {Ho, 2005 #11}. In contrast, Tnnt3/T5.1 RNA, which did not crosslink to MBNL1 (Figure3-7B), bound poorly. Comparable affinities were obt ained when MBNL1-RNA complexes were analyzed by gel shift analysis (Figure 4-3C). Incubation of full-length MBNL1 (MBNL1 FL) with Tnnt3/T5.45 generated severa l protein-RNA complexes resolv ed by the polyacrylamide gel whereas significant binding to Tnnt3/T5.1 was only detectable at 256 nM and only gel excluded complexes were observed. Similar complexes were also formed with MBNL1 N (data not shown). Incubation of MBNL1 with (CUG)54 also resulted in the formation of several major complexes at protein concentr ations (4-16 nM) near the Kd determined by filter binding. At higher MBNL1 FL concentrations ( 64 nM) the majority of the resulting MBNL1-(CUG)54 complexes migrated at, or near, the top of the gel. The striking similarity in the binding affinities of MBNL1 for pathogenic and splicing precurs or RNAs prompted us to re-examine the interaction of this splicing factor with CUGexp RNA.


69 Visualization of MBNL1-CUGexp Complexes In a previous study, it was reported that th e MBNL splicing antagoni st, CUGBP1, binds to out-of-register ssCUG repeat s at the base of CUGexp RNA hairpins but not the A-form helical region while the dsRBD protein TRBP associat es with the stem region {Michalowski, 1999 #111}. To confirm that MBNL1 is a stem-bi nding protein, electron microscopy (EM) was performed. Purified (CUG)136 was examined following direct absorption to thin carbon foils, dehydration and tungsten shadowing. In contrast to ssRNA, (CUG)136 RNA formed rod-like segments as described previously for (CUG)130 (Figure 4-4A-C). To ex amine the structures of MBNL1-RNA complexes, MBNL1 was incubated with RNAs at two different RNA:protein molar ratios (1:2.5 or 1:10) and subsequently prep ared for EM. In the presence of RNA, purified MBNL1 formed a ring-shaped structure with a prominent central cavity and for (CUG)136MBNL1 complexes incubated at a ratio of 1:2.5, ~ 70% of the RNAs were bound by one of these MBNL1 rings (Figure 4-4 D-F). At hi gher protein levels (1:10), free (CUG)136 RNA was rarely detectable (6.4% of the RNAs in the field) wh ile >90% of the RNAs were bound by two or more MBNL1 rings (Figure 4-4 G-H). Also shown is a representative fiel d of dsCUG RNAs and MBNL1-(CUG)136 complexes (Figure 4-4I, RNA:protein is 1:2.5). Although the majority of MBNL1 rings were associated with RNA under our RNA assembly conditions, a few free rings were visualized in the background in the absence of associated (CUG)136 helices suggesting that MBNL1 may form a ring structure independent of RNA. We failed to visualize rings by negative staining, suggesting that this structure is disrupted by the ac idic conditions of the negative staining protocol. MBNL1 Self-interaction Mediated by the C-terminal Region A num ber of studies have demonstrated that MBNL1 accumulates in ribonuclear foci together with pathogenic RNA {Mankodi, 2005 #124}{Mankodi, 2001 #38}{Jiang, 2004


70 #32}{Mankodi, 2003 #131}. While additional pr oteins might also bind to CUGexp and CCUGexp RNAs, the number of ribonuclear foci in DM my oblasts declines significantly following loss of MBNL1, suggesting that this protei n is required for the formation and/or maintenance of these unusual nuclear structures {Dansithong, 2005 #18}. Since many RNA-binding proteins function as components of large multi-subunit complexes and some of these proteins self-interact via their auxiliary or non-RNA binding re gions{Moore, 2005 #564}{Cartegn i, 1996 #565}, we tested the possibility that MBNL1 proteins self-associate using the yeast two hybrid system. Although the amino terminal region of MBNL1 contains all fo ur CCCH motifs and is responsible for proteinRNA interactions {Kino, 2004 #21}, very little is known about the function of the C-terminal region. The MBNL1 FL protein sh owed strong homotypic interactio ns in this system (Figure 45A). Although MBNL FL failed to interact with the MBNL1 N-terminal region (1-264 amino acid residues), interactions between the fulllength protein and C-terminal region (239-382 amino acid residues) were readily detectable. This C-terminal region does not contain any known RNA binding, or other, structural motifs. To confirm that MBNL1 homotypic interactions occurred in a mammalian cell context, 293T cells were co-transfected with plasmids which expressed either V5-MBNL1 FL alone, V5MBNL1 FL and MBNL1 FL-myc, or V5-MBNL1 FL and MBNL1 N-myc. Twenty-four hours following transfection, V5-tagged MBNL1 was immunopurified from cell lysates using an antiV5 antibody and the precipitates were then immunoblotted using either anti-myc or anti-V5 antibodies. In agreement with the two hybrid analysis, the full-length V5 and myc-tagged proteins were associated in vivo while the N-terminal MBNL1 re gion (MBNL1 N-myc) failed to co-immunopurify with V5-MBNL1 FL (Figure 4-5B ). Interactions between V5-MBNL1 FL and MBNL1 N-myc were resistant to RNase treatment since digestion of the cell lysate with RNase


71 A did not affect the amount of MBNL1 N-my c in the V5-MBNL1 FL immunoprecipitate. We conclude that MBNL1-MBNL1 interactions occur in vivo and these inter actions are mediated by the C-terminal region. MBNL1-RNA Complexes Display Distinct Stabilities What leads to MBNL1 functional sequestrati on in DM if MBNL1 binds to pathogenic RNA and norm al splicing targets with similar affinities? One possibility, as mentioned in the Introduction, is that the complexes formed between MBNL1 and pathogenic RNAs are more stable compared to those formed between MBNL 1 and its normal RNA targets. A modified UV crosslinking assay was performed to test this hypothesis. A 2000 fold excess of non-labeled RNA was added to the nuclear extract either at the same time or 15 minutes following the addition of the uniformly radio-labeled RNAs to the reacions. RNA-pr otein mixtures were further incubated for 15 minutes, followed by UV irradiation and RNase A digestion. The percentage of protein remaining bound to the radiolabeled RNAs after the cold RNA challenge was calculated by comparing the amounts of cross linked protein in lanes concurrent vs. lanes after where there was no cold competitor. CUGBP1 crosslinks to T5.45 strongly in the absence of cold T5.45 competitor. However, in the presence of an excessive amount of nonlabeled T5.45 RNAs whether adde d concurrently or after the radiolabeled RNA, CUGBP1 crosslinking was considerably reduced to a si milar extent, indicating a dynamic interaction between CUGBP1 and T5.45 RNA. However, MBNL 1 displayed a much more static interaction pattern with all three RNAs tested, of which (CUG)54 repeats was able to retain 100% of the MBNL1 crosslinking signal even after the additi on of a 2000-fold excess of cold competitor, suggesting that the MBNL1-(CUG)54 complex is not prone to dissociation once formed. The (CAG)54 and T5.45 RNAs were less efficient in retaining MBNL1 compared to (CUG)54, since ~40% and ~60% MBNL1 dissociated from (CAG)54 and T5.45, respectively. In conclusion,


72 MBNL1 forms stable complexes with RNAs. However, the complexes formed between MBNL1 and CUG repeats display a considerably higher stability compared to those formed between MBNL1 and other RNAs. Conclusions, Discussion and Future Work MBNL1 Targets Similar Binding Motifs in Splicing Precursor and Pathogenic RNAs Although there is considerable evidence fo r the MBNL loss-of-function model for DM pathogenesis, the m olecular basis for MBNL1 sequestration by CUGexp and CCUGexp RNAs has not been elucidated. Since mutant RNA expans ions must compete with normal pre-mRNA, and possibly mRNA, binding sites for MBNL1 recruitm ent, effective MBNL1 sequestration might occur if the affinity of this protein for CUGexp and CCUGexp RNAs is greater than for its normal splicing targets. The binding analysis presente d here does not support this conjecture, since MBNL1 also possesses relatively high affinities for CAGexp and Tnnt3 precursor RNAs. Indeed, these binding studies provide an explanation fo r the formation of MBNL 1-containing ribonuclear foci in cells overexpressing CAGexp {Ho, 2005 #11}. Additionally, mapping of a binding site to a stem-loop structure in Tnnt3 in tron 8 just upstream of the fetal exon indicates that RNA recognition by MBNL proteins involves a comm on interaction mode fo r both pathogenic and normal pre-mRNAs: recognition of GC-rich hairpins containing pyrimidine mismatches. What is the physiological significance of the binding preference of MBNL1 for RNA stemloop structures? In this chapter, we provide evidence that MBNL1 is a splicing regulator which acts as an intronic splicing repressor by recogni zing a stem-loop near the Tnnt3 fetal exon 3 splice site, possibly resulting in th e stabilization of this secondar y structure and interference with U2AF recruitment. Based on this observation, we speculate that MB NL1 recognizes similar intronic stem-loop structures adjacent to the 5 splice sites of other MBNL1-regulated fetal exons


73 to inhibit U1 snRNP recruitmen t or MBNL1 may stabilize intera ctions between introns flanking alternative exons resulting in RNA looping and increased fetal exon skipping. The importance of RNA secondary structures in inherited disease and alternative splicing has been previously highlighted in frontotem poral dementia with parkinsonism linked to chromosome 17 (FTDP-17) which is caused by mutations in the MAPT gene encoding the microtubule-associated prot ein tau {Mankodi, 2003 #131}. Some FTDP-17 mutations destabilize a predicted stem-loop structure, which forms between the 3end of exon 10 and the 5 end of the downstream intron, resulting in an increase in U1 snRNP recruitment and E10 inclusion. Interestingly, MAPT exon 10 skipping increases in the DM brain suggesting that MBNL1 promotes exon 10 inclusion duri ng splicing {Kanadia, 2006 #157}. The MBNL1 binding preferences shown in this study suggest that this factor may also function as a splicing activator by recognizing RNA stem -loop structures in novel exoni c and/or intronic splicing enhancers or by blocking splici ng silencer elements by stabil izing overlapping RNA secondary structures. Another interesting question arose when MBNL1 binding site on Tnnt3 was mapped to an 18-nt stemloop structure. Previous study showed th at MBNL1 in HeLa nuclear extract crosslinks strongly to (CUG)74 and (CUG)97 but not to (CUG)<20 {Kino, 2004 #21}. Based on this observation, we proposed that below a certain length threshold (<20 repeats) the dsCUG helix was unstable in the cell extract and ssCUG was not a binding site for MBNL1. The results reported here support that proposal and demonstr ate that MBNL1 is primarily a dsRNA binding protein which recognizes relativ ely short GC-rich hairpins if the overall RNA secondary structure is stabilized by additi onal sequence interactions. This conclusion provides a plausible resolution for the apparently conflicting resu lts that overexpression of a GFP-DMPK 3-UTR


74 (CTG)5 transgene results in a DM phenotype {Mahadevan, 2006 #156} while mice expressing the HSASR (human skeletal -actin containing a 3-UTR with five CTG repeats) transgene are normal {Mankodi, 2000 #16}. For the DMPK 3-UTR, the (CTG)5 repeat is predicted to interact with an upstream region to form a GC-rich stem interrupted by several U-U and C-C mismatches while this repeat in the HSA 3-U TR is located in seque ntial unpaired loops. It is possible that this structural arrangement in the DMPK 3 -UTR promotes MBNL1 sequestration when GFPDMPK 3-UTR (CTG)5 RNA is overexpressed in transgenic mice. MBNL1 Rings: Interactions with CUGexp RNA and Potential Mechanism for MBNL1 Functional Sequestration Filter binding and gel shift assays indicated that the affinity of MBNL1 for Tnnt3/T5.45 is similar to those of (CUG)54 and (CAG)54 in cell-free binding reactions. While overexpression of either CUG or CAG repeats induces the formation of nuclear foci in cell culture, it is interesting to note that there was no detectable RNA foci formation following Tnnt3 minigene overexpression (data not shown). This obser vation argues that high affinity MBNL1-RNA interactions together with a bundant expression of MBNL1 target RNAs is not sufficient for ribonuclear foci formation. Of course, MBNL1 may be cleared from splicing target, but not pathogenic, RNAs during RNA pr ocessing and nuclear export or unusual interactions between MBNL proteins and CUGexp RNA might drive foci formation. In support of the latter possibility, we provide EM evidence that MBNL1 form s a tandem ring structure when bound to CUGexp RNA. The size of the rings is uniform with a di ameter of ~18 nm and since the MBNL1 isoform employed for these studies is 41 kDa, the ring st ructure must be an oligomeric complex. At a protein:RNA ratio of 2.5:1, most CUGexp helices were bound by a single ring while the majority of these hairpins were bound by at least two rings at a higher protei n:RNA ratio. It is not clear if the MBNL1 ring contains a hole but if a central cav ity exists it might allow threading of dsRNA.


75 When multiple rings were bound to a single RNA molecule they appeared to be tandemly stacked suggesting either a preferential ring-load ing site or potential ri ng-ring interactions. The latter possibility is s upported by the finding that MBNL1 self-i nteracts via its C-terminal region both in the yeast two hybrid system and in mammalian cells. In contrast, the MBNL1 N-terminal region encompassing the CCCH RN A binding motifs fails to inte ract with full length MBNL1 although RNA-binding activity is comparable to the full-length MBNL1 (data not shown). When MBNL1 N proteins were incubated with CUG136, similar ring structures were also detectable by EM although ring size was less uni form and there was considerably less stacking of MBNL1 N rings compared to MBNL1 FL (dat a not shown). A similar situation has been noted for the E.coli protein Hfq (Host factor 1) which is a single-strand RNA binding protein involved in the translational regula tion and stability of several RNAs (37). As visualized in the EM by negative staining, Hfq forms hexameri c rings. Similar to MBNL1, the RNA binding activity of Hfq resides in the N-terminal regi on and rings are still formed by a C-terminal truncated protein althou gh they are less stable. Does MBNL1 exist as a ring structure when bound to its normal RNA splicing targets? To address this question, EM analysis was performed on the MBNL1-Tnnt3/T5.45 complex. Although MBNL1 rings were observed, the result was inconclusive due to the difficulty in visualizing small and partially single-stra nded Tnnt3/T5.45 RNA by EM. However, MBNL1 forms large complexes with both CUG and Tnnt3/ T5.45 RNAs (Figure 4-4 C) so it is possible that MBNL1 forms a ring structure when bound to splicing regulatory sites. What leads to MBNL1 functional sequest ration on pathogenic RNAs? Based on the observations in this study, a model is proposed as illustrated in Figure 4-7. MBNL1 forms the ring structure around G-C rich stemloop RNA. The carboxyl terminus of MBNL1 is not


76 essential for ring formation, yet it is involved in stabilizing the stacking of the MBNL1 rings. When MBNL1 binds to expanded CUG or CCUG RNAs, multiple ring structures could be loaded onto the long RNA stemloop. With th e C-terminus stabilizing the MBNL1-MBNL1 interactions, the structure is thermodynamically st able and resistant to disruption. On the other hand, when MBNL1 binds to its splicing targets, only one ring structure could be accommodated due to the short length of the binding sites. In the absen ce of the reinforcing stacking interactions, the rings are prone to dissociate fr om the RNA. Thus, th e higher stabilities of MBNL1-pathogenic RNA complexes lead to the accumulation of MBNL1 on pathogenic RNA. Due to the extremely low off rate of MBNL1 molecules bound to pathogenic RNAs (Figure 46), the CUG or CCUG-bound MBNL1 molecules are in accessible to the norma l splicing targets. Interestingly, based on th e observation that (CUG)136 RNA could accommodate 3-4 MBNL1 rings, the footprint of ea ch MBNL1 ring is estimated to be 34 -45 CUG repeats, which is slightly lower than the DM1 pathogenic repeat range (>50), supporting the hypothesis that MBNL1 ring stacking is required for its functional sequestration. As demonstrated in Figure 4-6, 54 CAG re peats retain MBNL1 much less than CUG repeats, which could explain the re sults by Ho and colleagues that (CAG)960 failed to promote embryonic splicing patterns for cTNT and IR as (CUG)960 did {Ho, 2005 #11}. I think that structural differences between CAG and CUG re peats might underlie their varied ability to functionally titrate MBNL1. The U-U mismatch in CUG repeats do not to cause any distortion in the RNA sugar-phosphate backbone {Mooe rs, 2005 #227}. Although no structural study has been carried out for CAG repeat RNA, it is pos sible that the bulged purine mismatches in CAG lead to a different RNA helix stru cture and block MBNL1 ring stacking.


77 Future Work The hypothesis that th e extreme stability of the MBNL1-pathogenic RNA complex leads to MBNL1 functional sequestration requires further experimental support. Studies such as a plasmon resonance analysis need to be carried ou t to confirm the distinct stabilities of MBNL1RNA complexes seen in the competition assay. If indeed the difference in stabilities is confirmed, it would be interesting to ask what factors contribute to the enhanced stability of the MBNL1-pathogenic RNA complex. Is this stability mediated by MBNL1 ring stacking as illustrated in Figure 4-7? If so, once the length of the CUG repeats is reduced below a certain threshold where ring stacking is not allowed, would MBNL1-CUG complexes display a similar stability compared to the MBNL1-Tnnt3/T5.45 co mplex? Based on the considerably weaker stacking ability of MBNL1 N-terminual rings co mpared to MBNL1 full length observed in the EM, this model would predict th at the MBNL1 N-terminus is unable to form the ultra-stable complex with long CUG repeats as the full-length protein does. These proposed experiments are essential for validating our proposed hypothesis.




79 Figure 4-1 MBNL1 recognizes a RNA hair pin upstream of the Tnnt3 fetal exon. A ) Cleavage pattern (left) of the 5-end labeled Tnnt3 151-nt transc ript encompassing the fetal exon 3 splice site (110-nt of intron 8, 41-nt of fetal exon) obtained with use of three structure probes. Lanes are: Ci, incubation control or no probe added; Pb, lead ions (0.25, 0.5 mM); S1, S1 nuclease (1, 2 U/l and 1 mM ZnCl2 was present in each reaction); T1, RNase T1 (0.5, 1 U/l); F, fo rmamide (statistical ladder); T, guaninespecific ladder. The sequences forming the 18-nt stem-loop structure are also indicated. Also illustrated (right) is the proposed secondary structure model of the 151-nt transcript. The cleavage sites are indi cated for each probe used and the figure inset shows the probe designations and cleav age intensity classification. The fetal exon sequence is marked in upper case and intron 8 in lower case. The positions of two G substitutions in the 18-nt stem-loop are also indicated. B ) Photocrosslinking analysis showing reduced MBNL1, but not CUGBP1, binding to the Tnnt3 10 and gg mutants in contrast to wild type RNA. Photocrosslinking analysis was performed as described in Fig. 1B except only MB NL1 FL (MBNL1) protein was used. C ) Tnnt3 fetal exon skipping is impaired in the 10 Tnnt3 T5.45 mutant compared to wild type while fetal exon inclusion is el iminated in the gg double mutant. C2C12 cells were co-transfected with either a wild type, Tnnt3 10 or gg double point mutant splicing reporter plasmid and a protein expression pl asmid for either CUGBP1mycHis or MBNL1mycHis (full-length protein only). 32P-labeled splicing products, which included (+F) or excluded (F) the Tnnt3 fetal exon (black box), were detected by RT-PCR, using primers positio ned in Tnnt3 exons 8 and 9 (open boxes with arrows), followed by gel electrophoresis.


80 Figure 4-2 MBNL1 binds throughout the dsCUG st em. Structural analysis of 5-end labeled (CUG)54 transcript (20 nM) either in the pr esence (+) or abse nce (-) of 500 nM recombinant MBNL1 and either lead ion (0.25 or 0.5 mM, lanes Pb) or ribonuclease T1 (0.5 or 1 units/l, lanes T1); incuba tion control (no probe added, lane Ci); formamide (lane F); guanine-specific la dder obtained with RNase T1 compete digestion (lane T). The positions of select ed G residues are shown along the T1 ladder (G-residues of the corresponding CUG repeat ar e indicated). Below the gel panels is the color-coded densitometric analysis (ImageQuant) of cleavage patterns for the CUG stem obtained with lead ion (Pb) and T1 ribonuclease (T1) in the presence of MBNL1 FL (green), MBNL1 N (red) or in the absence of protein (blue).


81 Figure 4-3 MBNL1 binds to pathogenic and splicing precursor RNAs with similar affinities. A ) Purified recombinant MBNL1 proteins. Co omassie-stained gels (left panels) and immunoblots (right panels) of e ither full length or N-termin al proteins. Illustration shows the primary structures (line; C3H motifs are shown as shaded boxes) indicating the full-length (382 aa) and N-terminal ( 253 aa) MBNL1 proteins together with a Cterminal truncated protein (270 aa ) generated during expression in E. coli and an unknown 75 kDa protein (asterisk). B ) Nitrocellulose filter binding analysis of MBNL1 FL binding to (CUG)54 (red square), (CAG)54 (orange triangle), Tnnt3 5.1(T5.1, blue cross) and Tnnt3 5.45 (T5.45, gr een circle) RNAs (see Figure 3-7). C ) Electrophoretic mobility shift analysis of MBNL1 FL binding to Tnnt3 5.45, 5.1 or (CUG)54. The positions of the free RNA (bracket ), the gel origin (well) and MBNL1 FL concentrations (triangle, lanes are 0, 0.25, 1, 4, 16, 64, 256 nM) are indicated.




83 Figure 4-4 Visualization of dsCUG and MBNL1-dsCUG comple xes. Electron microscopy of either free (CUG)136 RNAs ( A-C ) or MBNL1-(CUG)136 complexes ( D-I ) at a protein:RNA ratio of either 2.5:1 ( D-F, I ) or 10:1 ( G-H ). As reported previously, purified dsCUG RNAs, which were directly adsorbed onto thin carbon foils followed by dehydration and rotary shadow casting with tungsten, are elongated rod-shaped structures (23). For analysis of complexes, MBNL1 protein and (CUG)136 RNA were incubated together at 30C, fixed with glutaraldehyde, passed over a gel filtration column and adsorbed on copper grids for ro tary shadowing. Under these conditions, MBNL1 has a distinct ring-shaped structure. Size bars (white line) are 40 nm ( H ) or 50 nM ( I ).


84 Figure 4-5 Self-association of MBNL1 protei ns is mediated by the C-terminal region. A ) Two hybrid analysis using yeast strains transformed with the following binding domain (BD) and activation domain (AD) plas mids: 1) GAL4 DNA binding domain (BD) plasmids with either p53 (activation c ontrol) or full-length MBNL1(MBNL1); 2) activation domain plasmids with either T-antigen (activat ion control), MBNL1 (residues 1-382), MBNL1 1-264 (N-terminal region) or MBNL 239-382 (C-terminal region). Functional interactions between th e proteins expressed from the BD and AD plasmids results in growth on the TrpLeuHisselection plate. B ) Coimmunopurfication of MBNL1 requires the Cterminal region. HEK293T cells were co-transfected for 24 h with plasmids expres sing tagged versions of either full-length or N-terminal MBNL1 (V5-MBNL1 FL alone, co-transfected V5-MBNL1 FL and MBNL1 FL-myc, co-transfected V5-MBNL1 FL and MBNL1 N-myc). Cell lysates were prepared and the V5-MBNL1 FL protein immunopurified ( V5) followed by SDS-PAGE (50% of IP sample) and immunodetection of MBNL1 FL-myc or MBNL1 N-myc using mAb 9E10 (top panel, input lanes represent 2.5% of total IP) or V5-MBNL1 FL (bottom panel, onl y the input lanes are shown).


85 Figure 4-6 MBNL1-RNA complexes display distinct stabilities. UV crosslinking assays were performed in the absence of competitors (non) or in the presence of a 2000-fold excess of non-labdeled RNAs. The cold RNAs were either added concurrently with the labeled RNAs (Concurrent) or added af ter the labeled RNAs has been incubated with the nuclear extract for 15 minutes (Aft er). Reactions were incubated for another 15 min followed by UV crosslinking and RNase A digestion.


86 Figure 4-7 Model of complexes formation be tween MBNL1 and splicing targets versus pathogenic RNAs. MBNL1 forms oligomeric ring structure upon binding to RNA. Normal splicing targets harbor short ha irpin structure which is long enough to accommodate only one MBNL1 ring. One th e other hand, pathogenic RNAs contain tandem MBNL1 binding sites. Stacking of MBNL1 rings mediated by its C-terminus stabilizes the complexes formed between MBNL1 and pathogenic RNAs.


87 CHAPTER 5 DISCUSSION: INVOLVEMENT OF MUSC L EBLIND-LIKE PROTEINS IN RNAMEDIATED DISEASES AND POTENTIAL THERAPEUTIC APPROACHES DM1 and DM2 By using CUG repeats as an exam ple for pa thogenic RNAs, my study mainly focuses on MBNL1 involvement in DM1 pathogenesis. Th e other type of myotonic dystrophy, DM2, is caused by a CCTG expansion in the first intron of ZNF9 Compared to DM1, DM2 is a milder disease in that the age of onset is generally much older, and the symp toms are less severe. However, the sizes of CCTG expansions in DM2 ar e usually much larger than those of the CTG expansions in DM1. In addition, the expression of ZNF9 is ubiquitous and about ten fold higher than that of DMPK (153). If MBNL1 sequestration by CUG or CCUG repeats is the major mechanism underlying both DM1 and DM2 pathoge nesis, what might account for the milder symptoms in DM2 given the larger and more abundant CCUG RNAs? Several possibilities exist. First, MBNL1 may have a higher affin ity for CUG repeats compared to CCUG repeats. Yet, to the contrary, MBNL1 displayed stronger interaction for CCUG repeats than CUG repeats in yeast three-hybrid system (116). This obs ervation was further supported by our preliminary filter binding result using an interrupted (CCUG)46 tract (data not shown). Surprisingly, the Kd of MBNL1 for this interrupted CCUG repeat was an order of magnitude lower than that of a CUG repeat of comparable length. Thus MBNL 1 appears to have a higher affinity for CCUG repeats. Another possibility is that CCUG RNA repeats possess a reduced ability to retain MBNL1 protein compared to CUG repeats. The stem loop predicted by Mfold which is formed by CCUG repeat RNA has two, instead of one, pyrimidin e mismatches between each G-C Watson-Crick basepair. This additional pyrimidine mismat ch could potentially affect the RNA sugarphosphate backbone conformation and/or surface elec trostatic potential, whic h could result in a


88 distinct MBNL1 binding mode ( 65). Our preliminary EM result suggests that MBNL1 rings formed on CCUG repeat are smaller in size comp ared to the ones formed on CUG repeats. Moreover, there was considerably less ring stac king on CCUG (data not shown). It would be interesting to test in the co mmitment assay whether expanded CCUG retains MBNL1 to a lesser extent than CUG repeats, and if CCUG display a reduced ability to affect MBNL1-mediated splicing switches compared to CUG repeats. If these results turn out as predicted, it could provide a mechanistic basis for the differen ce in disease severity between DM1 and DM2. Congenital and Adult-Onset DM1 Congenital DM patients present a very distinct array of sym ptoms compared to adult-onset DM1 patients, including hypotonia and severe mental retardation. Notably, myotonia, which is 100% penetrant among adult-onset DM patients, is absent in CDM patients. The fact that CDM patients recover some of the aforementioned symptoms during their childhood, and eventually develop adult-onset symptoms, suggests a differe nt pathogenesis mechanism for CDM. We have shown that five known MBNL1 targets all under go splicing switches during a similar postnatal time window in mice (122). If only those functions of MBNL1 involved in promoting adult splicing patterns were impaired in CDM patients, severe consequen ces should not be expected in newborns, since MBNL1 regulated splicing switches have not yet b een activated. Why, then, are congenital DM patients with larg e CTG expansions affected in infancy? A number of genes, including fibronectin, agrin, and alpha1(X1) collagen chain, have been shown to undergo prenatal splicing switches during embryogenesis (161-163). One possibility is that MBNL3 is required for certain prenatal splicing switches. Indeed, Mbnl3 expression levels peak in the early stages of mouse embryonic development followe d by a gradual decline du ring the later stages ( 80). Whole-mount in situ hybridization revealed high leve ls of Mbnl3 expression in the developing head region and limb buds in 9.5 dpc mouse embryos ( 80). If MBNL3 is important


89 for regulating prenatal al ternative splicing events, co-expressi on of mutant DMPK transcripts in these regions may result in MBNL3 titration, and subs equent failure to complete certain prenatal splicing switches. Such failures may lead to th e brain and muscle abnormalities observed for CDM patients. Functional sequestration of MBNL3 may require the accumulation of large amounts of mutant DMPK transcripts in the nucleus. Increased DMPK expression has been shown in CDM skeletal muscle, possibly th rough a CTCF-mediated chromatin insulator (164,165). Thus, compared to adultonset DM patients, higher leve ls of mutant RNAs combined with longer CTG tract in CDM can sequester more MBNL3 protein, potentially sufficient to cause MBNL3 loss-of-function. To test this hypothesis, Mbnl3 knockout mice are being generated currently. If mice l acking Mbnl3 recapitulate the CDM brain and muscle defects, it could provide a mechanistic explanation for th e CDM pathogenesis. Meanwhile, Mbnl3 binding targets can be identified using the CLIP met hod. Subsequent studies could be done to test whether these CLIP-tagged pre-mRNAs undergo prenatal splicing switches regulated by MBNL3. Potential MBNL Involvement in Other RNAMediated Diseases Although first identified in th e etiology of myot onic dystrophy, RNA-mediated disease pathogenesis is probably not unique to DM. In the past few years, researchers have identified a growing number of other dominantly inherited diseases possibly caused by toxic gain-offunction at the RNA level. Similar to myotoni c dystrophy, toxic RNAs in these diseases often result from transcription of microsatellite expa nsions in non-coding region s. Interestingly, many of these disease-causing microsatel lite expansions are also GC ri ch. Therefore, the transcribed RNAs in many cases are predicted to form GC -rich double-stranded stemloop structures with interspaced mismatches, reminiscent of the structures formed by (CUG)n and (CCUG)n RNAs. Based on our result that MBNL1 pr efers GC rich stemloops with pyrimidine mismatches, these


90 mutant RNAs could potentially sequ ester MBNL proteins in a manner similar to that described in DM. Spinocerebellar Ataxia Type 8 (SCA8) SCA8 is a slow progressive form of cereb ellar ataxia, characte rized by gait and limb ataxia, nystagmus and dysarthria (166,167). Th e SCA8 mutation was identified to be a CTG expansion residing in a noncoding gene on human chromosome 13q21 (166). It has been reported that in the general population, more than 99% of the alleles have 16-37 CTA/CTG combined repeats. Alleles with 107-127 CTG repeats are associated with SCA8 patients (166). However the inheritance pattern of SCA8 is complex displaying incomplete penetrance, which may due to the size of CTA tract pre ceding the CTG expansion or different SCA8 expression levels among patients (166-172). Although the mechanism of SCA8 pathogenesis remains an area of debate, one hypothesis suggests that the CUG containing R NA is toxic resulting in impaired cellular function. This idea has gained experimental support from transgen ic mouse and fly models. Transgenic mice expressing mutant SCA8 BAC developed a progressive neurological phenotype, which was not present in mice expressing the normal SCA8 transgene, suggesting th at the expanded SCA8 RNA is pathogenic (173). In a Drosophia model, retinal expression of both normal (9 CTG) and expanded (107 CTG) human SCA8 non-coding RNAs resulted in a rough-eye phenotype (131). Interestingly, a genetic modifier screen uncovered mutations in Drosophila muscleblind gene which enhanced the roug h-eye phenotype caused by SCA8 expression in a CTG length dependent manner suggesting possible interacti ons between muscleblin d proteins and SCA8 mutant transcripts (131). Further studies such as rescuing SCA8 transgenic mice by MBNL overexpression could be performed to determ ine whether MBNL1 is involved in SCA8 pathogenesis.


91 Huntingtons Disease-like 2 (HDL2) HDL2 is an autosom al dominant, progressive neurodegenerative disease characterized by chorea, dystonia, rigidity, bra dykinesia, psychiatric syndromes, dementia, and an inevitable decline to death (174-180). HDL2 is caused by a CTG expansion in the gene junctophilin-3 ( JPH3 ). Depending on the alternative splicing of the pre-mRNA, CUG repeats may encode polyleucine or polyalanine, or al ternatively reside in the 3-UTR (178). CTG repeat lengths in HDL2 display fairly tight normal versus pathog enic ranges with 6-28 triplets in normal individuals whereas 40-59 triple ts are found in HDL2 patients (179). RNA foci have been detected in HDL2 frontal cort ex using riboprobes specific for CUG repeats as well as junctophilin-3 mRNA (133). These RNA foci rese mble those found in DM in terms of size, nuclear location, and, importantly, MBNL1 colocalization (133). Assayed by immunofluorescence, nuclea r MBNL1 levels were reduced by an average of ~50% in foci containing cells, possibly leading to the aberrant splicing of MBNL1 targets in the affected brain region (133). These results suggest molecular parallels between DM and HDL2 pathogenesis. Fragile X Tremor Ataxia Syndrome (FXTAS) FXTAS is a late-onset neurodege ne rative disorder with core fe atures of action tremor and cerebellar gait ataxia (181). FXTAS is caused by a CGG expansion in the Fragile X Mental Retardation 1 (FMR1) gene (182). While full mutations of over 200 CGG repeats cause Fragile X Syndrome by silencing FMR1 transcription, CGG expansions in the permutation range of 55200 repeats lead to FXTAS, potentially thr ough a proposed RNA gain-of-function mechanism (182,183). Results from transgenic mouse and fly models expressing CGG repeats of premutation length lend support to the RNA-mediated pathogenesis model. Both mice and flies develop neurodegenerative phenotypes upon rCGG repeat expression (184,185).


92 Consistent with the RNA gain-of-function model, rCGG repeats were found in ubiquitinpositive intranuclear inclusions in FXTAS brains (183). These inclusions have been recently purified from post-mortem brain tissue of FXTA S patients by fluorescent-activated cell sorting (FACS) (134). Mass spectrometric analysis of the protein composition of these inclusions revealed more than twenty inclusion associated proteins, among which ar e structural proteins such as lamin A/C and neurofilament 3, stress -related proteins, and RNA-binding proteins including hnRNP A2/B1 and MB NL1. Immunofluorescent stai ning further confirmed the localization of MBNL1 to the ubiquitin positive nuclear incl usions (134). Although MBNL1 does not bind to rCGG repeats with high affin ity in filter binding assay (data not shown), considering the complexity of protein components in FXTAS inclusions, it remains possible that MBNL1 is indirectly recruited to the rCGG incl usions via other protein factors. Whether MBNL1 recruitment to rCGG foci plays a causative or secondary role in FXTAS pathogenesis requires careful examination using both genetic and biochemical approaches. Implications for Potential DM Therapeutic Strategies The ultim ate goal of studying DM pathogenesis is to not only understand what goes wrong in the disease, but also find ways to reverse disease progression. MBNL loss-of-function plays a causative role in DM pathogenesis. Thus, a st raightforward therapeutic strategy would be to supplement DM tissues with exogenous MBNL1 protei n to help maintain a dult splicing patterns. In fact, recombinant adeno-associated virus (rAAV) mediated rescuing experiment has been performed in the HSALR poly(CUG) mouse model for DM (159). Viral-mediated MBNL1 expression was induced in tibiali s anterior (TA) muscle by injecting rAAV. Myotonia in the injected TA muscle was reversed up to 43 weeks post-injection, and aberra nt splicing of several MBNL1 targets was successfully rescued. While effective for local applications, systemic rescue


93 of DM phenotypes using rAAV-medi ated gene therapy requires a large amount of viral vector, which may pose an obstacle for potential clin ical applications of this approach. Another therapeutic strategy would be to disrupt the interactio n between MBNL and pathogenic RNAs, thus enabling MBNL proteins to bind to th eir physiological targets and promote their adult splicing patterns. There are at least two ways to achieve this goal. The first strategy is based on the observation that MBNL1 has low affinities for perfectly base-paired dsRNAs (186). Thus, short stabilized (CAG)n or (CAGG)n could be introduced into DM cells and transform the CUG or CCUG hair pin into perfectly paired CAG CUG or CCUG CAGG which would no longer be able to titrate MBNL. Several comme rcially available nucleic acid modifications, such as the locked-nucleic acid (LNA), could be used to enhance the thermostability of the CAG CUG or CCUG CAGG hybrid. The toxicity of these modified nucleic acids requires careful assessment and their delivery methods await further improvements. With the rapid development of small molecule drug screens, thousands of chemicals could be screened for their abilities to inhibit MBNL binding to CUG or CCUG repeats, which could provide another promising avenue to achieving a molecular therapy for DM. However, a key to the successful clinical applications of these dr ugs will rely on their abilities to preserve the interactions between MBNL1 and normal RNA targets while preferentially disrupting the interactions between MBNL1 and pathogenic RNAs. Results from this study suggest that MBNL1 interacts with pathogenic and normal RNA ta rgets in a very similar manner, in terms of recognition properties and binding affi nities. Thus, it is highly likely that small molecules which disrupt MBNL1-CUG or CCUG bind ing would also inhibit MBNL1 binding to its physiological targets. A different screening strategy is propos ed here based on the model illustrated in Figure 4-7. If MBNL1 C-terminus-mediated ring st acking is essential for MBNL1 functional


94 sequestration, chemicals could be screened for th eir abilities to disrupt MBNL1 C-terminal selfinteractions. Since the MBNL1 N-terminus alone promotes adult splicing patterns to the same extent as full-length MBNL1 (Figure 3-6), such chemical agen ts would probably not affect MBNL1 regulated splicing events. However, ba sed on the model in Figu re 4-7, loss of MBNL1 C-terminal interactions would decrease ring stacking, which would lead to reduced stability of MBNL1-pathogenic RNA complexes, and s ubsequently alleviat e MBNL1 functional sequestration. One important cons ideration of this st rategy is the toxicity of naked CUG and CCUG repeats. Once stripped of MBNL proteins CUG and CCUG repeat R NAs may be able to interfere with other cellular pr ocesses through interactions with protein or RNA factors with relatively high affinities. This possibility can be tested by crossing poly(CUG) or poly(CCUG) transgenics with Mbnl1-/Mbnl2-/double knockouts and assaying for additional phenotypes compared to the transgenic mice.




96 LIST OF REFERENCES 1. Harper, P.S. (2001) Myotonic Dystrophy. 2. Ranum, L.P. and Day, J.W. (2004) Myotonic dystrophy: RNA pathogenesis comes into focus. Am. J. Hum. Genet. 74, 793-804. 3. Liquori, C.L., Ricker, K., Moseley, M.L., Jacobsen, J.F., Kress, W., Naylor, S.L., Day, J.W. and Ranum, L.P. (2001) Myotonic dystrophy type 2 caused by a CCTG expansion in intron 1 of ZNF9. Science, 293, 864-867. 4. Harley, H.G., Brook, J.D., Rundle, S.A., Cr ow, S., Reardon, W., Buckler, A.J., Harper, P.S., Housman, D.E. and Shaw, D.J. (1992) Expansion of an unstable DNA region and phenotypic variation in myotonic dystrophy. Nature, 355, 545-546. 5. Buxton, J., Shelbourne, P., Davies, J., Jones, C., Van Tongeren, T., Aslanidis, C., de Jong, P., Jansen, G., Anvret, M., Riley, B. et al. (1992) Detection of an unstable fragment of DNA specific to individua ls with myotonic dystrophy. Nature, 355, 547-548. 6. Brook, J.D., McCurrach, M.E., Harley, H.G., Bu ckler, A.J., Church, D., Aburatani, H., Hunter, K., Stanton, V.P., Thirion, J.P., Hudson, T. et al. (1992) Molecular basis of myotonic dystrophy: expansion of a trinucleotide (CTG) re peat at the 3' end of a transcript encoding a protei n kinase family member. Cell 68, 799-808. 7. Aslanidis, C., Jansen, G., Amemiya, C., Shu tler, G., Mahadevan, M., Tsilfidis, C., Chen, C., Alleman, J., Wormskamp, N.G., Vooijs, M. et al. (1992) Cloning of the essential myotonic dystrophy region and mapping of the putative defect. Nature, 355 548-551. 8. Larkin, K. and Fardaei, M. (2001) Myotonic dystrophy--a multigene disorder. Brain Res. Bull. 56, 389-395. 9. Machuca-Tzili, L., Brook, D. and Hilton-Jones, D. (2005) Clinical and molecular aspects of the myotonic dystrophies: a review. Muscle Nerve 32, 1-18. 10. Day, J.W. and Ranum, L.P. (2005) RNA pa thogenesis of the myotonic dystrophies. Neuromuscul. Disord., 15 5-16. 11. Cho, D.H. and Tapscott, S.J. (2007) Myot oinc dystrophy: Emerging mechanisms for DM1 and DM2. Biochimica et Biophysica Acta 195-204. 12. Day, J.W., Ricker, K., Jacobsen, J.F., Rasmussen, L.J., Dick, K.A., Kress, W., Schneider, C., Koch, M.C., Beilman, G.J., Harrison, A.R. et al. (2003) Myotonic dystrophy type 2: molecular, diagnostic an d clinical spectrum. Neurology 60, 657-664. 13. Fu, Y.H., Pizzuti, A., Fenwick, R.G., Jr., King, J., Rajnarayan, S., Dunne, P.W., Dubel, J., Nasser, G.A., Ashizawa, T., de Jong, P. et al. (1992) An unstable triplet repeat in a gene related to myotonic muscular dystrophy. Science 255, 1256-1258.


97 14. Mahadevan, M., Tsilfidis, C., Sabourin, L., Shutler, G., Amemiya, C., Jansen, G., Neville, C., Narang, M., Barcelo, J., O'Hoy, K. et al. (1992) Myotonic dystrophy mutation: an unstable CTG repeat in th e 3' untranslated region of the gene. Science 255 1253-1255. 15. Harley, H.G., Rundle, S.A., Reardon, W., My ring, J., Crow, S., Brook, J.D., Harper, P.S. and Shaw, D.J. (1992) Unstable DNA sequence in myotonic dystrophy. Lancet 339 1125-1128. 16. Ashizawa, T., Dubel, J.R. and Harati, Y. (1993) Somatic instability of CTG repeat in myotonic dystrophy. Neurology 43, 2674-2678. 17. Martorell, L., Johnson, K., Boucher, C.A. a nd Baiget, M. (1997) So matic instability of the myotonic dystrophy (CTG)n repeat during human fetal development. Hum. Mol. Genet. 6, 877-880. 18. Wong, L.J. and Ashizawa, T. (1997) Instabi lity of the (CTG)n repeat in congenital myotonic dystrophy. Am. J. Hum. Genet. 61, 1445-1448. 19. Wong, L.J., Ashizawa, T., Monckton, D.G., Caskey, C.T. and Richards, C.S. (1995) Somatic heterogeneity of the CTG repeat in myotonic dystrophy is age and size dependent. Am. J. Hum. Genet. 56, 114-122. 20. Martorell, L., Monckton, D.G., Gamez, J., Johnson, K.J., Gich, I., Lopez de Munain, A. and Baiget, M. (1998) Progression of somatic CTG repeat length heterogeneity in the blood cells of myotonic dystrophy patients. Hum. Mol. Genet. 7, 307-312. 21. Abbruzzese, C., Costanzi Porrini, S., Mariani, B., Gould, F.K., McAbney, J.P., Monckton, D.G., Ashizawa, T. and Giacanelli, M. (2002) Instability of a premutation allele in homozygous patients with myotonic dystrophy type 1. Annals of Neurology, 52, 435-441. 22. Duthel, S., Bost, M., Ollagnon, E., Vial, C., Petiot, P., Chazot, G. and Vandenberghe, A. (1999) CTG instability in myotonic dystrophy: molecular genetic analysis of families from south-eastern France with characteristic s of intergenerational variation in CGT repeat numbers. Annales de Genetique 42, 151-159. 23. Brunner, H.G., Bruggenwirth, H.T., Nillesen, W., Jansen, G., Hamel, B.C., Hoppe, R.L., de Die, C.E., Howeler, C.J., van Oost, B.A., Wieringa, B. et al. (1993) Influence of sex of the transmitting parent as well as of pare ntal allele size on the CTG expansion in myotonic dystrophy (DM). Am. J. Hum. Genet. 53, 1016-1023. 24. Lahue, R.S. and Slater, D.L. (2003) DNA repair and trinuc leotide repeat instability. Front. Biosci. 8, s653-665.


98 25. Parniewski, P. and Staczek, P. (2002) Mo lecular mechanisms of TRS instability. Advances in Experiment al Medicine and Biology, 516, 1-25. 26. Cleary, J.D. and Pearson, C.E. (2003) The contribution of cis-el ements to diseaseassociated repeat instability: clin ical and experimental evidence. Cytogenet. Genome Res. 100, 25-55. 27. van den Broek, W.J., Nelen, M.R., Wansink, D.G., Coerwinkel, M.M., te Riele, H., Groenen, P.J. and Wieringa, B. (2002) Somatic expansion behaviour of the (CTG)n repeat in myotonic dystrophy knock-in mice is differentially affected by Msh3 and Msh6 mismatch-repair proteins. Hum. Mol. Genet. 11, 191-198. 28. Foiry, L., Dong, L., Savouret, C., Hubert, L., te Riele, H., Junie n, C. and Gourdon, G. (2006) Msh3 is a limiting factor in the forma tion of intergenerational CTG expansions in DM1 transgenic mice. Human Genetics 119, 520-526. 29. Savouret, C., Garcia-Cordier, C., Megret, J., te Riele, H ., Junien, C. and Gourdon, G. (2004) MSH2-dependent germinal CTG repeat expansions are produced continuously in spermatogonia from DM1 transgenic mice. Mol. Cell. Biol. 24, 629-637. 30. Manley, K., Shirley, T.L., Flaherty, L. and Messer, A. (1999) Msh2 deficiency prevents in vivo somatic instability of the CAG rep eat in Huntington disease transgenic mice. Nat. Genet. 23, 471-473. 31. Kovtun, I.V. and McMurray, C.T. (2001) Trinucleotide expansion in haploid germ cells by gap repair. Nat. Genet. 27, 407-411. 32. Gomes-Pereira, M., Fortune, M.T., Ingr am, L., McAbney, J.P. and Monckton, D.G. (2004) Pms2 is a genetic enhancer of trinuc leotide CAG.CTG repeat somatic mosaicism: implications for the mechanism of triplet repeat expansion. Hum. Mol. Genet. 13, 18151825. 33. Stober, G., Franzek, E., Lesch, K.P. and Beckmann, H. (1995) Periodic catatonia: a schizophrenic subtype with major gene effect and anticipation. European Archives of Psychiatry and Clinical Neuroscience, 245, 135-141. 34. Bassett, A.S. and Husted, J. (1997) Anticipat ion or ascertainment bias in schizophrenia? Penrose's familial mental illness sample. Am. J. Hum. Genet. 60, 630-637. 35. McInnis, M.G., McMahon, F.J., Chase, G.A ., Simpson, S.G., Ross, C.A. and DePaulo, J.R., Jr. (1993) Anticipation in bipolar affective disorder. Am. J. Hum. Genet. 53, 385390. 36. McInnis, M.G. (1996) Anticipati on: an old idea in new genes. Am. J. Hum. Genet. 59, 973-979.


99 37. Ridley, R.M., Frith, C.D., Crow, T.J. a nd Conneally, P.M. (1988) Anticipation in Huntington's disease is inheri ted through the male line but may originate in the female. Journal of Medical Genetics 25, 589-595. 38. Manto, M.U. (2005) The wide spectrum of spinocerebellar ataxias (SCAs). Cerebellum (London, England) 4, 2-6. 39. Flink, I.L. and Morkin, E. (1995) Alternativ ely processed isoforms of cellular nucleic acid-binding protein interact with a suppressor region of the human beta-myosin heavy chain gene. J. Biol. Chem. 270, 6959-6965. 40. Gerbasi, V.R. and Link, A.J. (2007) The myot onic dystrophy type 2 pr otein ZNF9 is part of an ITAF complex that prom otes cap-independent translation. Mol. Cell Proteomics 6, 1049-1058. 41. Margolis, J.M., Schoser, B.G., Moseley, M.L., Day, J.W. and Ranum, L.P. (2006) DM2 intronic expansions: evidence for CCUG accumulation without flanking sequence or effects on ZNF9 mRNA proces sing or protein expression. Hum. Mol. Genet. 15, 18081815. 42. Liquori, C.L., Ikeda, Y., Weatherspoon, M., Ricker, K., Schoser, B.G., Dalton, J.C., Day, J.W. and Ranum, L.P. (2003) Myotonic dys trophy type 2: human founder haplotype and evolutionary conservation of the repeat tract. Am. J. Hum. Genet. 73, 849-862. 43. Schoser, B.G., Kress, W., Walter, M.C., Hallig er-Keller, B., Lochmuller, H. and Ricker, K. (2004) Homozygosity for CCTG mutation in myotonic dystrophy type 2. Brain 127, 1868-1877. 44. Fu, Y.H., Friedman, D.L., Richards, S., Pearlman, J.A., Gibbs, R.A., Pizzuti, A., Ashizawa, T., Perryman, M.B., Scarlato, G., Fenwick, R.G., Jr. et al. (1993) Decreased expression of myotonin-protein kinase messenger RNA and protein in adult form of myotonic dystrophy. Science 260, 235-238. 45. Novelli, G., Gennarelli, M., Zelano, G., Pi zzuti, A., Fattorini, C., Caskey, C.T. and Dallapiccola, B. (1993) Failure in detecting mR NA transcripts from the mutated allele in myotonic dystrophy muscle. Biochemistry and Molecula r Biology International 29, 291297. 46. Hofmann-Radvanyi, H., Lavedan, C., Rabes, J.P., Savoy, D., Duros, C., Johnson, K. and Junien, C. (1993) Myotonic dystrophy: absence of CTG enlarged transcript in congenital forms, and low expression of the normal allele. Hum. Mol. Genet. 2, 1263-1266. 47. Narang, M.A., Waring, J.D., Sabourin, L. A. and Korneluk, R.G. (2000) Myotonic dystrophy (DM) protein kinase levels in congenital and adult DM patients. Eur. J. Hum. Genet. 8, 507-512.

PAGE 100

100 48. Jansen, G., Groenen, P.J., Bachner, D., Ja p, P.H., Coerwinkel, M., Oerlemans, F., van den Broek, W., Gohlsch, B., Pette, D., Plomp, J.J. et al. (1996) Abnormal myotonic dystrophy protein kinase levels pr oduce only mild myopathy in mice. Nat. Genet. 13, 316-324. 49. Reddy, S., Smith, D.B., Rich, M.M., Leferovich J.M., Reilly, P., Davis, B.M., Tran, K., Rayburn, H., Bronson, R., Cros, D. et al. (1996) Mice lacking the myotonic dystrophy protein kinase develop a late onset progressive myopathy. Nat. Genet. 13, 325-335. 50. Berul, C.I., Maguire, C.T., Gehrma nn, J. and Reddy, S. (2000) Progressive atrioventricular conduction block in a mouse myotonic dystrophy model. J. Interv. Card. Electrophysiol. 4, 351-358. 51. Wang, Y.H., Amirhaeri, S., Kang, S., Wells, R. D. and Griffith, J.D. (1994) Preferential nucleosome assembly at DNA triplet repeats from the myotonic dystrophy gene. Science 265, 669-671. 52. Otten, A.D. and Tapscott, S.J. (1995) Tr iplet repeat expansio n in myotonic dystrophy alters the adjacent chromatin structure. Proc. Natl. Acad. Sci., U S A, 92, 5465-5469. 53. Klesert, T.R., Otten, A.D., Bird, T.D. and Tapscott, S.J. (1997) Trinucleotide repeat expansion at the myotonic dystrophy locu s reduces expression of DMAHP. Nat. Genet. 16, 402-406. 54. Hamshere, M.G., Newman, E.E., Alwazzan, M., Athwal, B.S. and Brook, J.D. (1997) Transcriptional abnormality in myotonic dys trophy affects DMPK but not neighboring genes. Proc. Natl. Acad. Sci., U S A 94, 7394-7399. 55. Frisch, R., Singleton, K.R., Moses, P.A., Gonzalez, I.L., Carango, P., Marks, H.G. and Funanage, V.L. (2001) Effect of triplet re peat expansion on chro matin structure and expression of DMPK and neighboring genes, SIX5 and DMWD, in myotonic dystrophy. Molecular Genetics and Metabolism 74, 281-291. 56. Thornton, C.A., Wymer, J.P., Simmons, Z., McClain, C. and Moxley, R.T., 3rd. (1997) Expansion of the myotonic dystrophy CTG re peat reduces expression of the flanking DMAHP gene. Nat. Genet. 16, 407-409. 57. Klesert, T.R., Cho, D.H., Clark, J.I., Mayl ie, J., Adelman, J., Snider, L., Yuen, E.C., Soriano, P. and Tapscott, S.J. (2000) Mi ce deficient in Six5 develop cataracts: implications for myotonic dystrophy. Nat. Genet. 25, 105-109. 58. Sarkar, P.S., Appukuttan, B., Han, J., Ito, Y., Ai, C., Tsai, W., Chai, Y., Stout, J.T. and Reddy, S. (2000) Heterozygous loss of Six5 in mice is sufficient to cause ocular cataracts. Nat. Genet. 25 110-114.

PAGE 101

101 59. Taneja, K.L., McCurrach, M., Schalling, M., Housman, D. and Singer, R.H. (1995) Foci of trinucleotide repeat transcripts in nuclei of myotonic dystrophy cells and tissues. J. Cell Biol. 128 995-1002. 60. Timchenko, L.T., Miller, J.W., Timchenko, N. A., DeVore, D.R., Datar, K.V., Lin, L., Roberts, R., Caskey, C.T. and Swanson, M.S. (1996) Identification of a (CUG)n triplet repeat RNA-binding protein and its expression in myotonic dystrophy. Nucleic Acids Res. 24, 4407-4414. 61. Mankodi, A., Logigian, E., Callahan, L., McClain, C., White, R., Henderson, D., Krym, M. and Thornton, C.A. (2000) Myotonic dystr ophy in transgenic mice expressing an expanded CUG repeat. Science 289 1769-1773. 62. Zuker, M. (2003) Mfold web server for nucle ic acid folding and hybridization prediction. Nucleic Acids Res. 31, 3406-3415. 63. Michalowski, S., Miller, J.W ., Urbinati, C.R., Paliouras, M., Swanson, M.S. and Griffith, J. (1999) Visualization of double-stranded RNAs from the myotonic dystrophy protein kinase gene and interacti ons with CUG-binding protein. Nucleic Acids Res. 27, 35343542. 64. Sobczak, K., de Mezer, M., Michlewski, G ., Krol, J. and Krzyzosiak, W.J. (2003) RNA structure of trinucleotide repeats associated with hu man neurological diseases. Nucleic Acids Res. 31, 5469-5482. 65. Mooers, B.H., Logue, J.S. and Berglund, J.A. (2005) The structural basis of myotonic dystrophy from the crystal stru cture of CUG repeats. Proc. Natl. Acad. Sci., U S A 102, 16626-16631. 66. Tian, B., White, R.J., Xia, T., Welle, S., Turner, D.H., Mathews, M.B. and Thornton, C.A. (2000) Expanded CUG repeat RNAs form hairpins that activate the double-stranded RNA-dependent prot ein kinase PKR. RNA (New York, N.Y) 6, 79-87. 67. Napierala, M. and Krzyzosi ak, W.J. (1997) CUG repeats present in myotonin kinase RNA form metastable "slippery" hairpins. J. Biol. Chem. 272 31079-31085. 68. Miller, J.W., Urbinati, C.R., Teng-Umnuay, P ., Stenberg, M.G., Byrne, B.J., Thornton, C.A. and Swanson, M.S. (2000) Recruitment of human muscleblind proteins to (CUG)(n) expansions associated with myotonic dystrophy. The EMBO Journal, 19, 4439-4448. 69. Kanadia, R.N., Johnstone, K.A., Mankodi A., Lungu, C., Thornton, C.A., Esson, D., Timmers, A.M., Hauswirth, W.W. and Swan son, M.S. (2003) A muscleblind knockout model for myotonic dystrophy. Science, 302, 1978-1980.

PAGE 102

102 70. Ebralidze, A., Wang, Y., Petkova, V., Ebra lidse, K. and Junghans, R.P. (2004) RNA leaching of transcription factors disrup ts transcription in myotonic dystrophy. Science 303, 383-387. 71. Artero, R., Prokop, A., Paricio, N., Begemann, G., Pueyo, I., Mlodzik, M., Perez-Alonso, M. and Baylies, M.K. (1998) The muscleblind gene participates in the organization of Zbands and epidermal attachments of Dros ophila muscles and is regulated by Dmef2. Dev. Biol. 195, 131-143. 72. Begemann, G., Paricio, N., Artero, R., Kiss, I., Perez-Alons o, M. and Mlodzik, M. (1997) muscleblind, a gene required for photoreceptor differentiation in Drosophila, encodes novel nuclear Cys3His-type zinc -finger-containing proteins. Development, 124, 43214331. 73. Pascual, M., Vicente, M., Monferrer, L. and Artero, R. (2006) The Muscleblind family of proteins: an emerging class of regulators of developmentally pr ogrammed alternative splicing. Differentiation 74, 65-80. 74. Gomperts, M., Pascall, J.C. and Brown, K.D. (1990) The nucleotide sequence of a cDNA encoding an EGF-inducible gene indicates the existence of a new family of mitogeninduced genes. Oncogene 5, 1081-1083. 75. Anderson, J.T., Wilson, S.M., Datar, K.V. and Swanson, M.S. (1993) NAB2: a yeast nuclear polyadenylated R NA-binding protein essential for cell viability. Mol. Cell. Biol. 13, 2730-2741. 76. Hudson, B.P., Martinez-Yamout, M.A., Dyson, H.J. and Wright, P.E. (2004) Recognition of the mRNA AU-rich element by the zinc finger domain of TIS11d. Nat. Struct. Mol. Biol. 11, 257-264. 77. Monferrer, L. and Artero, R. (2006) An inte rspecific functional complementation test in Drosophila for introductory genetics laboratory courses. The Journal of Heredity 97, 6773. 78. Bejerano, G., Pheasant, M., Makunin, I., St ephen, S., Kent, W.J., Mattick, J.S. and Haussler, D. (2004) Ultraconserved elements in the human genome. Science, 304, 13211325. 79. Fardaei, M., Rogers, M.T., Thorpe, H.M., La rkin, K., Hamshere, M.G., Harper, P.S. and Brook, J.D. (2002) Three proteins, MBNL, MBLL and MBXL, co-localize in vivo with nuclear foci of expanded-repeat tr anscripts in DM1 and DM2 cells. Hum. Mol. Genet. 11, 805-814. 80. Kanadia, R.N., Urbinati, C.R., Crusselle, V.J ., Luo, D., Lee, Y.J., Harrison, J.K., Oh, S.P. and Swanson, M.S. (2003) Developmental expression of mouse muscleblind genes Mbnl1, Mbnl2 and Mbnl3. Gene Expr. Patterns, 3, 459-462.

PAGE 103

103 81. Wang, G.S. and Cooper, T.A. (2007) Splicing in disease: disruption of the splicing code and the decoding machinery. Nat. Rev. Genet ., 8, 749-761. 82. Stamm, S., Zhu, J., Nakai, K., Stoilov, P., Stoss, O. and Zhang, M.Q. (2000) An alternative-exon database a nd its statistical analysis. DNA and Cell Biology 19, 739-756. 83. Padgett, R.A., Grabowski, P.J., Konarska M.M., Seiler, S. and Sharp, P.A. (1986) Splicing of messenger RNA precursors. Annual Review of Biochemistry 55, 1119-1150. 84. Staley, J.P. and Guthrie, C. (1998) Mech anical devices of th e spliceosome: motors, clocks, springs, and things. Cell 92, 315-326. 85. Wu, S., Romfo, C.M., Nilsen, T.W. and Green M.R. (1999) Functional recognition of the 3' splice site AG by the splicing factor U2AF35. Nature, 402, 832-835. 86. Stamm, S., Zhang, M.Q., Marr, T.G. and Helfman, D.M. (1994) A sequence compilation and comparison of exons that are alternatively spliced in neurons. Nucleic Acids Res. 22, 1515-1526. 87. Clark, F. and Thanaraj, T.A. (2002) Categor ization and characterization of transcriptconfirmed constitutively and alternatively spliced introns and exons from human. Hum. Mol. Genet. 11, 451-464. 88. Kanopka, A., Muhlemann, O. and Akusjarvi, G. (1996) Inhibition by SR proteins of splicing of a regulated adenovirus pre-mRNA. Nature, 381, 535-538. 89. Cooper, T.A. and Mattox, W. (1997) The regulat ion of splice-site sele ction, and its role in human disease. Am. J. Hum. Genet. 61, 259-266. 90. Damgaard, C.K., Tange, T.O. and Kjems, J. (2002) hnRNP A1 controls HIV-1 mRNA splicing through cooperative bindi ng to intron and exon splicing silencers in the context of a conserved secondary structure. RNA (New York, N.Y) 8, 1401-1415. 91. Tange, T.O., Damgaard, C.K., Guth, S., Valcarcel, J. and Kjems, J. (2001) The hnRNP A1 protein regulates HIV-1 tat splicing via a novel intron silencer element. The EMBO Journal 20, 5748-5758. 92. Carstens, R.P., Wagner, E.J. and Garcia-Blanco, M.A. (2000) An intronic splicing silencer causes skipping of the IIIb exon of fibroblast growth factor receptor 2 through involvement of polypyrimidine tract binding protein. Mol. Cell. Biol. 20, 7388-7400. 93. Wagner, E.J. and Garcia-Blanco, M.A. ( 2001) Polypyrimidine tract binding protein antagonizes exon definition. Mol. Cell. Biol. 21, 3281-3288.

PAGE 104

104 94. Lou, H., Helfman, D.M., Gagel, R.F. and Berget, S.M. (1999) Polypyrimidine tractbinding protein positively regul ates inclusion of an al ternative 3'-terminal exon. Mol. Cell. Biol. 19 78-85. 95. Hofmann, Y. and Wirth, B. (2002) hnRNP-G promotes exon 7 inclusion of survival motor neuron (SMN) via direct interaction with Htra2-beta1. Hum. Mol. Genet. 11, 2037-2049. 96. Chou, M.Y., Rooke, N., Turck, C.W. and Blac k, D.L. (1999) hnRNP H is a component of a splicing enhancer complex that activates a c-src alternative exon in neuronal cells. Mol. Cell. Biol. 19 69-77. 97. Min, H., Chan, R.C. and Black, D.L. ( 1995) The generally expressed hnRNP F is involved in a neural-specifi c pre-mRNA splicing event. Genes & Development 9, 26592671. 98. Ule, J., Stefani, G., Mele, A., Ruggiu, M., Wang, X., Taneri, B., Gaasterland, T., Blencowe, B.J. and Darnell, R.B. (2006) An RNA map predicting Nova-dependent splicing regulation. Nature, 444, 580-586. 99. Zhu, J., Mayeda, A. and Krainer, A.R. (2001) Exon identity established through differential antagonism between exonic spli cing silencer-bound hnRNP A1 and enhancerbound SR proteins. Mol. Cell 8, 1351-1361. 100. Ashiya, M. and Grabowski, P.J. (1997) A neuron-specific splicing switch mediated by an array of pre-mRNA repressor sites: evidence of a regulatory role for the polypyrimidine tract binding protein and a brain-specific PTB counterpart. RNA (New York, N.Y) 3, 9961015. 101. Irwin, N., Baekelandt, V., Goritchenko, L. and Benowitz, L.I. (1997) Identification of two proteins that bind to a pyrimidine-rich sequence in the 3'untranslated region of GAP-43 mRNA. Nucleic Acids Res. 25, 1281-1288. 102. Markovtsov, V., Nikolic, J.M., Goldman, J. A., Turck, C.W., Chou, M.Y. and Black, D.L. (2000) Cooperative assembly of an hnR NP complex induced by a tissue-specific homolog of polypyrimidine tract binding protein. Mol. Cell. Biol. 20, 7463-7479. 103. Coutinho-Mansfield, G.C., Xue, Y., Zhang, Y. and Fu, X.D. (2007) PTB/nPTB switch: a post-transcriptional mechanism for programming neuronal differentiation. Genes & Development, 21, 1573-1577. 104. Boutz, P.L., Stoilov, P., Li, Q., Lin, C.H., Ch awl a, G., Ostrow, K., Shiue, L., Ares, M., Jr. and Black, D.L. (2007) A post-transcriptional regulatory switch in polypyrimidine tractbinding proteins reprograms alternativ e splicing in developing neurons. Genes & Development, 21, 1636-1652.

PAGE 105

105 105. Shin, C., Feng, Y. and Manley, J.L. (2004) Dephosphorylated SRp38 acts as a splicing repressor in response to heat shock. Nature, 427, 553-558. 106. Shin, C. and Manley, J.L. (2004) Cell signa lling and the control of pre-mRNA splicing. Nature Reviews, 5, 727-738. 107. Krawczak, M., Reiss, J. and Cooper, D.N. (1992) The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Human Genetics 90, 41-54. 108. Pagani, F. and Baralle, F.E. (2004) Genomic variants in exons and introns: identifying the splicing spoilers. Nat. Rev. Genet. 5, 389-396. 109. Lopez-Bigas, N., Audit, B., Ouzounis, C ., Parra, G. and Guigo, R. (2005) Are splicing mutations the most frequent cau se of hereditary disease? FEBS Letters 579 1900-1903. 110. Osborne, R.J. and Thornton, C.A. (2006) RNA-dominant diseases. Hum. Mol. Genet. 15 Spec No 2 R162-169. 111. Mankodi, A., Takahashi, M.P., Jiang, H., Beck, C.L., Bowers, W.J., Moxley, R.T., Cannon, S.C. and Thornton, C.A. (2002) Expanded CUG repeats trigger aberrant splicing of ClC-1 chloride channel pre-mRNA and hyperexcitability of skeletal muscle in myotonic dystrophy. Mol. Cell 10, 35-44. 112. Charlet, B.N., Savkur, R.S., Singh, G., Philips, A.V., Grice, E.A. and Cooper, T.A. (2002) Loss of the muscle-specific chloride channel in type 1 m yotonic dystrophy due to misregulated alternative splicing. Mol. Cell, 10, 45-53. 113. Savkur, R.S., Philips, A.V. and Cooper, T. A. (2001) Aberrant regulation of insulin receptor alternative splicing is associated with insulin resistance in myotonic dystrophy. Nat. Genet. 29, 40-47. 114. Moxley, R.T., Griggs, R.C. and Goldblatt, D. (1980) Muscle insulin resistance in myotonic dystrophy: effect of supraphysiologic insulinization. Neurology 30, 1077-1083. 115. Philips, A.V., Timchenko, L.T. and Cooper, T.A. (1998) Disruption of splicing regulated by a CUG-binding protein in myotonic dystrophy. Science 280 737-741. 116. Kino, Y., Mori, D., Oma, Y., Takeshita, Y., Sasagawa, N. and Ishiura, S. (2004) Muscleblind protein, MBNL1/EXP, binds specifically to CHHG repeats. Hum. Mol. Genet. 13, 495-507. 117. Rooke, N., Markovtsov, V., Cagavi, E. and Bl ack, D.L. (2003) Roles for SR proteins and hnRNP A1 in the regulation of c-src exon N1. Mol. Cell. Biol. 23, 1874-1884.

PAGE 106

106 118. Ho, T.H., Charlet, B.N., Poulos, M.G., Singh, G., Swanson, M.S. and Cooper, T.A. (2004) Muscleblind proteins regulate altern ative splicing. The EMBO Journal, 23, 31033112. 119. Vicente, M., Monferrer, L., Poulos, M.G., Houseley, J., Monckton, D.G., O'Dell K, M., Swanson, M.S. and Artero, R.D. (2007) Muscle blind isoforms are functionally distinct and regulate alpha-actinin splicing. Differentiation 75, 427-440. 120. Hino, S.I., Kondo, S., Sekiya, H., Saito, A., Kanemoto, S., Murakami, T., Chihara, K., Aoki, Y., Nakamori, M., Takahashi, M.P. et al. (2007) Molecular mechanisms responsible for aberrant splicing of SERCA1 in myotonic dystrophy type 1. Hum. Mol. Genet. 16, 2834-2843. 121. Sharma, S. and Black, D.L. (2006) Maps, c odes, and sequence elements: can we predict the protein output from an alternatively spliced locus? Neuron 52, 574-576. 122. Lin, X., Miller, J.W., Mankodi, A., Kanadia, R.N., Yuan, Y., Moxley, R.T., Swanson, M.S. and Thornton, C.A. (2006) Failure of MBNL1-dependent post-natal splicing transitions in myotonic dystrophy. Hum. Mol. Genet. 15, 2087-2097. 123. Makeyev, E.V., Zhang, J., Carrasco, M.A. and Maniatis, T. (2007) The MicroRNA miR124 promotes neuronal differentiation by trigge ring brain-specific alternative pre-mRNA splicing. Mol. Cell 27, 435-448. 124. Ule, J., Jensen, K., Mele, A. and Darnell, R.B. (2005) CLIP: a method for identifying protein-RNA interaction sites in living cells. Methods (San Diego, Calif.) 37, 376-386. 125. Timchenko, N.A., Cai, Z.J., Welm, A.L., Reddy, S., Ashizawa, T. and Timchenko, L.T. (2001) RNA CUG repeats sequester CUGBP1 an d alter protein leve ls and activity of CUGBP1. J. Biol. Chem., 276, 7820-7826. 126. Savkur, R.S., Philips, A.V., Cooper, T.A., Da lton, J.C., Moseley, M.L., Ranum, L.P. and Day, J.W. (2004) Insulin receptor splicing alteration in myotonic dystrophy type 2. Am. J. Hum. Genet. 74, 1309-1313. 127. Dansithong, W., Paul, S., Comai, L. a nd Reddy, S. (2005) MBNL1 is the primary determinant of focus formation and aber rant insulin receptor splicing in DM1. J. Biol. Chem. 280, 5773-5780. 128. Lefebvre, S., Burlet, P., Viollet, L., Bertra ndy, S., Huber, C., Belser, C. and Munnich, A. (2002) A novel association of the SMN protei n with two major nonribosomal nucleolar proteins and its implication in spinal muscular atrophy. Hum. Mol. Genet. 11, 10171027.

PAGE 107

107 129. Srivastava, S., Mukherjee, M., Panigrahi, I ., Shanker Pandey, G., Pradhan, S. and Mittal, B. (2001) SMN2-deletion in childh ood-onset spinal muscular atrophy. American Journal of Medical Genetics 101 198-202. 130. Underwood, J.G., Boutz, P.L., Dougherty, J.D., Stoilov, P. and Black, D.L. (2005) Homologues of the Caenorhabditis elegans F ox-1 protein are neurona l splicing regulators in mammals. Mol. Cell. Biol. 25, 10005-10016. 131. Mutsuddi, M., Marshall, C.M., Benzow, K. A., Koob, M.D. and Rebay, I. (2004) The spinocerebellar ataxia 8 noncoding RNA causes neurodegeneration and associates with staufen in Drosophila. Curr. Biol., 14 302-308. 132. Wakamiya, M., Matsuura, T., Liu, Y., Schuste r, G.C., Gao, R., Xu, W., Sarkar, P.S., Lin, X. and Ashizawa, T. (2006) The role of ataxin 10 in the pathogenesis of spinocerebellar ataxia type 10. Neurology 67, 607-613. 133. Rudnicki, D.D., Holmes, S.E., Lin, M.W., Thornton, C.A., Ross, C.A. and Margolis, R.L. (2007) Huntington's disease--lik e 2 is associated with C UG repeat-containing RNA foci. Annals of Neurology, 61 272-282. 134. Iwahashi, C.K., Yasui, D.H., An, H.J., Gr eco, C.M., Tassone, F., Nannen, K., Babineau, B., Lebrilla, C.B., Hagerman, R.J. and Hage rman, P.J. (2006) Prot ein composition of the intranuclear inclus ions of FXTAS. Brain 129, 256-271. 135. Gabellini, D., D'Antona, G., Moggio, M., Prelle, A., Zecca, C., Adami, R., Angeletti, B., Ciscato, P., Pellegrino, M.A., Bottinelli, R. et al. (2006) Facioscapulohumeral muscular dystrophy in mice overexpressing FRG1. Nature 439, 973-977. 136. Ule, J., Ule, A., Spencer, J., Williams, A ., Hu, J.S., Cline, M., Wang, H., Clark, T., Fraser, C., Ruggiu, M. et al. (2005) Nova regulates brain-sp ecific splicing to shape the synapse. Nat. Genet. 37, 844-852. 137. Jensen, K.B., Dredge, B.K., Stefani, G ., Zhong, R., Buckanovich, R.J., Okano, H.J., Yang, Y.Y. and Darnell, R.B. (2000) Nova-1 regulates neuron-specific alternative splicing and is essentia l for neuronal viability. Neuron, 25, 359-371. 138. Bhalla, K., Phillips, H.A., Crawford, J., McKenzie, O.L., Mulley, J.C., Eyre, H., Gardner, A.E., Kremmidiotis, G. and Callen, D. F. (2004) The de novo chromosome 16 translocations of two patients with a bnormal phenotypes (mental retardation and epilepsy) disrupt the A2BP1 gene. Journal of Human Genetics, 49, 308-311. 139. McKie, A.B., McHale, J.C., Keen, T.J., Ta rttelin, E.E., Goliath, R., van Lith-Verhoeven, J.J., Greenberg, J., Ramesar, R.S., Hoyng, C.B., Cremers, F.P. et al. (2001) Mutations in the pre-mRNA splicing factor gene PRPC8 in autosomal dominant retinitis pigmentosa (RP13). Hum. Mol. Genet. 10, 1555-1562.

PAGE 108

108 140. Vithana, E.N., Abu-Safieh, L., Allen, M. J., Carey, A., Papaioannou, M., Chakarova, C., Al-Maghtheh, M., Ebenezer, N.D., Willis, C., Moore, A.T. et al. (2001) A human homolog of yeast pre-mRNA splicing gene PRP31, underlies autosomal dominant retinitis pigmentosa on chromosome 19q13.4 (RP11). Mol. Cell 8, 375-381. 141. Chakarova, C.F., Hims, M.M., Bolz, H., AbuSafieh, L., Patel, R.J., Papaioannou, M.G., Inglehearn, C.F., Keen, T.J., Willis, C., Moore, A.T. et al. (2002) Mutations in HPRP3, a third member of pre-mRNA splicing factor ge nes, implicated in autosomal dominant retinitis pigmentosa. Hum. Mol. Genet. 11, 87-92. 142. Yang, L., Embree, L.J. and Hickstein, D.D. (2000) TLS-ERG leukemia fusion protein inhibits RNA splicing mediated by serine-arginine proteins. Mol. Cell. Biol. 20, 33453354. 143. Ichikawa, H., Shimizu, K., Hayashi, Y. and Ohki, M. (1994) An RNA-binding protein gene, TLS/FUS, is fused to ERG in human my eloid leukemia with t(16;21) chromosomal translocation. Cancer Research 54, 2865-2868. 144. Venables, J.P., Elliott, D.J., Makarova, O.V., Makarov, E.M., Cooke, H.J. and Eperon, I.C. (2000) RBMY, a probable human sper matogenesis factor, and other hnRNP G proteins interact with Tr a2beta and affect splicing. Hum. Mol. Genet. 9, 685-694. 145. Imai, H., Chan, E.K., Kiyosawa, K., Fu, X.D. and Tan, E.M. (1993) Novel nuclear autoantigen with splicing factor motifs id entified with antibody from hepatocellular carcinoma. J. Clin. Invest. 92, 2419-2426. 146. Clark, J., Lu, Y.J., Sidhar, S.K., Parker, C., Gill, S., Smedley, D., Hamoudi, R., Linehan, W.M., Shipley, J. and Cooper, C.S. (1997) Fusi on of splicing factor genes PSF and NonO (p54nrb) to the TFE3 gene in papillary renal cell carcinoma. Oncogene 15, 2233-2239. 147. Young, J.I., Hong, E.P., Castle, J.C., Cres po-Barreto, J., Bowman, A.B., Rose, M.F., Kang, D., Richman, R., Johnson, J.M., Berget, S. et al. (2005) Regulation of RNA splicing by the methylation-dependent transcriptional repressor methyl-CpG binding protein 2. Proc. Natl. Acad. Sci., U S A, 102, 17551-17558. 148. Kishore, S. and Stamm, S. (2006) The snoR NA HBII-52 regulates alternative splicing of the serotonin receptor 2C. Science 311, 230-232. 149. Fomenkov, A., Huang, Y.P., Topaloglu, O., Brechman, A., Osada, M., Fomenkova, T., Yuriditsky, E., Trink, B., Sidransky, D. and Ratovitski, E. (2003) P63 alpha mutations lead to aberrant splicing of keratinocyte growth factor receptor in the Hay-Wells syndrome. J. Biol. Chem. 278, 23906-23914. 150. Ho, T.H., Savkur, R.S., Poulos, M.G., Mancini, M.A., Swanson, M.S. and Cooper, T.A. (2005) Colocalization of muscle blind with RNA foci is sepa rable from mis-regulation of alternative splicing in myotonic dystrophy. Journal of Cell Science, 118, 2923-2933.

PAGE 109

109 151. Blackshear, P.J. (2002) Tristetraprolin a nd other CCCH tandem zinc -finger proteins in the regulation of mRNA turnover. Biochemical Society Transactions 30, 945-952. 152. Nykamp, K.R. and Swanson, M.S. (2004) Toxic RNA in the nucleus: unstable microsatellite expression in neuromuscular disease. Progress in Molecular and Subcellular Biology 35, 57-77. 153. Mankodi, A., Teng-Umnuay, P., Krym, M., He nderson, D., Swanson, M. and Thornton, C.A. (2003) Ribonuclear inclusions in skelet al muscle in myotonic dystrophy types 1 and 2. Annals of Neurology 54, 760-768. 154. Mankodi, A., Lin, X., Blaxall, B.C., Swans on, M.S. and Thornton, C.A. (2005) Nuclear RNA foci in the heart in myotonic dystrophy. Circulation Research 97, 1152-1155. 155. Mankodi, A., Urbinati, C.R., Yuan, Q.P ., Moxley, R.T., Sansone, V., Krym, M., Henderson, D., Schalling, M., Swanson, M.S. and Thornton, C.A. (2001) Muscleblind localizes to nuclear foci of aberrant RNA in myotonic dystrophy types 1 and 2. Hum. Mol. Genet. 10, 2165-2170. 156. Jiang, H., Mankodi, A., Swanson, M.S., M oxley, R.T. and Thornton, C.A. (2004) Myotonic dystrophy type 1 is associated with nuclear foci of mutant RNA, sequestration of muscleblind proteins and deregulated alternat ive splicing in neurons. Hum. Mol. Genet. 13, 3079-3088. 157. Moore, M.J. (2005) From birth to death: the complex lives of eukaryotic mRNAs. Science 309 1514-1518. 158. Cartegni, L., Maconi, M., Morandi, E., Cobian chi, F., Riva, S. and Biamonti, G. (1996) hnRNP A1 selectively interacts through its Gly-rich domain with different RNA-binding proteins. J. Mol. Biol. 259 337-348. 159. Kanadia, R.N., Shin, J., Yuan, Y., Beatti e, S.G., Wheeler, T.M., Thornton, C.A. and Swanson, M.S. (2006) Reversal of RNA mi ssplicing and myotonia after muscleblind overexpression in a mouse poly(CUG) model for myotonic dystrophy. Proc. Natl. Acad. Sci., U S A 103, 11748-11753. 160. Mahadevan, M.S., Yadava, R.S., Yu, Q., Balijepalli, S., Frenzel-McCardell, C.D., Bourne, T.D. and Phillips, L.H. (2006) Reve rsible model of RNA toxicity and cardiac conduction defects in myotonic dystrophy. Nat. Genet. 38, 1066-1070. 161. Morris, N.P., Oxford, J.T., Davies, G.B., Smoody, B.F. and Keene, D.R. (2000) Developmentally regulated alternative splicing of the alpha1(XI) collagen chain: spatial and temporal segregation of isoforms in the cartilage of fetal rat long bones. J. Histochem. Cytochem. 48, 725-741.

PAGE 110

110 162. Thomas, W.S., O'Dowd, D.K. and Smith, M.A. (1993) Developmental expression and alternative splicing of chick agrin RNA. Dev. Biol., 158, 523-535. 163. Muro, A.F., Chauhan, A.K., Gajovic, S., Iac oncig, A., Porro, F., Stanta, G. and Baralle, F.E. (2003) Regulated splicing of the fibron ectin EDA exon is esse ntial for proper skin wound healing and normal lifespan. J. Cell Biol. 162, 149-160. 164. Filippova, G.N., Thienes, C.P., Penn, B.H., Cho, D.H., Hu, Y.J., Moore, J.M., Klesert, T.R., Lobanenkov, V.V. and Tapscott, S.J. (2001) CTCF-binding sites flank CTG/CAG repeats and form a methylation-sensitive insulator at the DM1 locus. Nat. Genet. 28, 335-343. 165. Sabouri, L.A., Mahadevan, M.S., Narang, M., Lee, D.S., Surh, L.C. and Korneluk, R.G. (1993) Effect of the myotonic dystrophy (D M) mutation on mRNA levels of the DM gene. Nat. Genet. 4, 233-238. 166. Koob, M.D., Moseley, M.L., Schut, L.J., Benzow, K.A., Bird, T.D., Day, J.W. and Ranum, L.P. (1999) An unt ranslated CTG expansion causes a novel form of spinocerebellar ataxia (SCA8). Nat. Genet. 21, 379-384. 167. Day, J.W., Schut, L.J., Moseley, M.L., Durand, A.C. and Ranum, L.P. (2000) Spinocerebellar ataxia type 8: clin ical features in a large family. Neurology 55, 649-657. 168. Topisirovic, I., Dragasevic, N., Savic, D ., Ristic, A., Keckarevic, M., Keckarevic, D., Culjkovic, B., Petrovic, I., Romac, S. and Kostic, V.S. (2002) Genetic and clinical analysis of spinocerebellar ataxia ty pe 8 repeat expansion in Yugoslavia. Clinical Genetics, 62 321-324. 169. Ikeda, Y., Shizuka, M., Watanabe, M., Okamoto, K. and Shoji, M. (2000) Molecular and clinical analyses of spinocereb ellar ataxia type 8 in Japan. Neurology 54 950-955. 170. Cellini, E., Nacmias, B., Forleo, P., Piacentini, S., Guarnieri, B.M., Serio, A., Calabro, A., Renzi, D. and Sorbi, S. (2001) Genetic and clinical analysis of spinocerebellar ataxia type 8 repeat expansion in Italy. Archives of Neurology, 58, 1856-1859. 171. Stevanin, G., Herman, A., Durr, A., Jodice, C., Frontali, M., Agid, Y. and Brice, A. (2000) Are (CTG)n expansions at th e SCA8 locus rare polymorphisms? Nat. Genet. 24 213; author reply 215. 172. Moseley, M.L., Schut, L.J., Bird, T.D., Koob, M.D., Day, J.W. and Ranum, L.P. (2000) SCA8 CTG repeat: en masse contractions in sperm and intergenerational sequence changes may play a role in reduced penetrance. Hum. Mol. Genet. 9, 2125-2130. 173. Moseley, M.L., Zu, T., Ikeda, Y., Gao, W., Mosemiller, A.K., Daughters, R.S., Chen, G., Weatherspoon, M.R., Clark, H.B., Ebner, T.J. et al. (2006) B idirectional expression of

PAGE 111

111 CUG and CAG expansion transcripts and intr anuclear polyglutamine inclusions in spinocerebellar ataxia type 8. Nat. Genet. 38, 758-769. 174. Walker, R.H., Rasmussen, A., Rudnicki, D., Holmes, S.E., Alonso, E., Matsuura, T., Ashizawa, T., Davidoff-Feldman, B. and Mar golis, R.L. (2003) Huntington's disease-like 2 can present as chorea-acanthocytosis. Neurology 61, 1002-1004. 175. Walker, R.H., Morgello, S., Davidoff-F eldman, B., Melnick, A., Walsh, M.J., Shashidharan, P. and Brin, M.F. (2002) Auto somal dominant chorea-acanthocytosis with polyglutamine-containing neuronal inclusions. Neurology 58 1031-1037. 176. Walker, R.H., Jankovic, J., O'Hearn, E. and Ma rgolis, R.L. (2003) Phenotypic features of Huntington's disease-like 2. Mov. Disord., 18, 1527-1530. 177. Stevanin, G., Camuzat, A., Holmes, S.E., Julien, C., Sahloul, R., D ode, C., Hahn-Barma, V., Ross, C.A., Margolis, R.L., Durr, A. et al. (2002) CAG/CTG repeat expansions at the Huntington's disease-like 2 locus are ra re in Huntington's disease patients. Neurology 58, 965-967. 178. Holmes, S.E., O'Hearn, E., Rosenblatt, A., Callahan, C., Hwang, H.S., IngersollAshworth, R.G., Fleisher, A., Stevanin, G., Brice, A., Potter, N.T. et al. (2001) A repeat expansion in the gene encoding junctophilin-3 is associated with Huntington disease-like 2. Nat. Genet. 29, 377-378. 179. Margolis, R.L., Holmes, S.E., Rosenblatt, A., Gourley, L., O'Hearn, E., Ross, C.A., Seltzer, W.K., Walker, R.H., Ashizawa, T., Rasmussen, A. et al. (2004) Huntington's Disease-like 2 (HDL2) in North America and Japan. Annals of Neurology, 56, 670-674. 180. Margolis, R.L., O'Hearn, E., Rosenblatt, A., Willour, V., Holmes, S.E., Franz, M.L., Callahan, C., Hwang, H.S., Troncoso, J.C. and Ro ss, C.A. (2001) A disorder similar to Huntington's disease is associated with a novel CAG repeat expansion. Annals of Neurology 50, 373-380. 181. Berry-Kravis, E., Abrams, L., Coffey, S.M ., Hall, D.A., Greco, C., Gane, L.W., Grigsby, J., Bourgeois, J.A., Finucane, B., Jacquemont, S. et al. (2007) Fragile X-associated tremor/ataxia syndrome: Clinical featur es, genetics, and testing guidelines. Mov. Disord., 22, 2018-2030. 182. Oostra, B.A. and Willemsen, R. (2003) A fragile balance: FMR1 expression levels. Hum. Mol. Genet. 12 Spec No 2 R249-257. 183. Hagerman, P.J. and Hagerman, R.J. (2004) The fragile-X premutation: a maturing perspective. Am. J. Hum. Genet. 74, 805-816. 184. Willemsen, R., Hoogeveen-Westerveld, M., Reis, S., Holstege, J., Severijnen, L.A., Nieuwenhuizen, I.M., Schrier, M., van Unen, L., Tassone, F., Hoogeveen, A.T. et al.

PAGE 112

112 (2003) The FMR1 CGG repeat mouse displays ubiquitin-positive in tranuclear neuronal inclusions; implications for the cer ebellar tremor/ataxia syndrome. Hum. Mol. Genet. 12, 949-959. 185. Jin, P., Zarnescu, D.C., Zhang, F., Pearson, C.E., Lucchesi, J.C., Moses, K. and Warren, S.T. (2003) RNA-mediated neurodegeneration caused by the fragile X premutation rCGG repeats in Drosophila. Neuron, 39, 739-747. 186. Warf, M.B. and Berglund, J.A. (2007) MBNL binds similar RNA structures in the CUG repeats of myotonic dystrophy and its premRNA substrate cardiac troponin T. RNA (New York, N.Y), 13, 2238-2251.

PAGE 113

113 BIOGRAPHICAL SKETCH Yuan Yuan was born in Wuhan, China on August 25th. She is the only child of Lizhi Yuan and Xinyun Zhu. Yuan attended Wuhan Univer sity from 1997 to 2001, during which time she pursued a Bachelor of Science degree in biologi cal sciences. After graduation, Yuan came to University of Florida and joined the Interdiscip linary Program in Biomedical Sciences (IDP) in the College of Medicine. In May of 2002, Yuan chose to perform her doctorate research under the guidance of Dr. Maurice Swanson. She completed her graduate work and doctoral dissertation in December of 2007, and will move to Rockefeller University to begin her postdoctoral studies with Dr. Robert Darnell.