PDF hosted at the Radboud Repository of the Radboud University Nijmegen

The following full text is a publisher's version.

For additional information about this publication click this link. http://hdl.handle.net/2066/85871

Please be advised that this information was generated on 2021-09-29 and may be subject to change. SFRS N UCE N BAN CELLS BRAIN AND MUSCLE IN ISOFORMS o a ia in n fnto • ralph j.a. oude ophuislocalization • and function

ISBN 978-90-8891234,-4 2011 9 789088 912344 > DMPK ISOFORMS IN MUSCLE AND BRAIN CELLS LOCALIZATION AND FUNCTION

Voor het bijwonen van de openbare verdediging van het proefschrift van

RALPH J.A. OUDE OPHUIS

DMPK ISOFORMS IN MUSCLE AND BRAIN CELLS LOCALIZATION AND FUNCTION

op

vrijdag 1 april 2011 om 13:00u precies in de Aula van de Radboud Universiteit Nijmegen aan de Comeniuslaan 2 te Nijmegen

Na afloop van de verdediging is er een receptie ter plaatse

PARANIMFEN

Susan Mulders [email protected]

Rinske van de Vorstenbosch r.vandevorstenbosch(§) ncmls.ru.nl DMPK ISOFORMS IN MUSCLE AND BRAIN CELLS LOCALIZATION AND FUNCTION ISBN-13 978-90-8891234-4

ISBN-10 90-8891-234-3

Printed by Proefsohriftmaken.nl || Printyourthesis.com

Published by Uitgeverij BOXPress, Oisterwijk DMPK ISOFORMS IN MUSCLE AND BRAIN CELLS LOCALIZATION AND FUNCTION

Een wetenschappelijke proeve op het gebied van de Medische Wetenschappen

Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. mr. S.C.J.J. Kortmann, volgens besluit van het college van decanen in het openbaar te verdedigen op vrijdag 1 april 2011 om 13:00 uur precies

door Raphaël Johannes Antonius Oude Ophuis geboren op 24 oktober 1978 te Sint-Oedenrode Promotor Prof. dr. B. Wieringa

Copromotores Dr. D.G. Wansink Dr. J.A.M. Fransen

Manuscriptcommissie Prof. dr. H.G. Brunner Prof. dr. J.A.M. Smeitink Dr. F.J.M. van Kuppeveld

The studies presented in this thesis were performed at the Department of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, The Netherlands. Financial support was obtained from the Prinses Beatrix Fonds, the Stichting Spieren voor Spieren, the Muscular Dystrophy Association (MDA) and the Association Française contre les Myopathies (AFM). TABLE OF CONTENTS

Chapter 1 9 General introduction • 1 Myotonic Dystophy type 1 11 • 2 DM1 molecular pathogenesis 12 • 3 DMPK protein 16 • 4 DMPK splice isoforms 18 • 5 Tail-anchored proteins 22 • 6 DMPK substrates and function 25 • 7 Aim and outline of this thesis 28

Chapter 2 31 DMPK protein isoforms are differentially expressed in myogenic and neu­ ral cell lineages

Chapter 3 49 Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

Chapter 4 77 A tail-anchored myotonic dystrophy protein kinase isoform induces peri­ nuclear clustering of mitochondria, autophagy and apoptosis

Chapter 5 97 Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes

Chapter 6 121 Summarizing Discussion

References 132

Nederlandse samenvatting 146

Abbreviations 150

Dankwoord 152

Curriculum vitae 156

Publications 157 i CHAPTER 1

GENERAL INTRODUCTION

GENERAL INTRODUCTION Í

• 1. Myotonic Dystophy type 1 (DM1)

DM is the most common genetic determined muscular dystrophy in adults at a prevalence of 1:8000 (101) and was first described by Steinert (266), and Batten and Gibb (10) about one century ago. We now know that this multisystemic neuromuscular disorder can be divided into two subtypes with different prevalence: DM type 1 (DM1, also known as Steinert dis­ ease; MIM 160900) and DM type 2 (DM2, also known as proxim al myo­ tonic myopathy, MIM 602668). DM symptoms vary in severity and mostly manifest themselves as muscle related. Prominent symptoms include myotonia, distal and proxim al muscle weakness and atrophy, gastroin­ testinal transit problems, and cardiac aberrations like conduction de­ fects and arrhythmias. Cardiac problems are generally considered the leading cause of death in DM. Furthermore, the central nervous system can be affected, and cataract and endocrine problems can be involved, illus trating the multisystemic character of the disease (summarized in Fig. 1A). This plethora of symptoms together with long asymptomatic

Figure 1 • DM1 molecular A Muscular pathogenesis: symptoms and myotonia +++ locus. distal muscle weakness +++ (A) Characteristic symptoms proximal muscle weakness ++ of DM1. The frequency of muscle atrophy ++ facial muscles ++ symptoms is indicated by a hip girdle muscles + scale from + to +++. Typical finger flexors +++ skeletal muscle features are neck muscles +++ illu s tra te d in the draw ing foot dorsiflexors ++ quadriceps +++ darkest shading indicates cardiac arrythmia ++ most severely affected skel­ gastrointestinal problems + etal m uscle areas. (B) DM1 uterus + locus organization. The (CTG)n repeat tra ct is Non-muscular cataract ++ located in a gene-dense re ­ hypogonadism ++ gion on human chromosome insulin insensitivity + 19q13 in exon 15 of the hypersomnia ++ DMPK gene, which overlaps mental retardation + with the promoter region of congenital DM + the SIX5 gene. Rectangles indicate exons (grey, DMWD; + present; ++ frequent; +++ usual black, DMPK; white, SIX5). The straight line represents B locus (19q13) N orm al 5 < (CTG)n < 35 introns and intergenic se­ DM1 quences. Arrows denote start P re m u ta tio n 35 < (CTG)n < 49 l L (CTG)n ^ and direction of transcription. M ild 50 < (CTG)n < 150 Disease classification anc C la ssic 100 < (CTG)n < 1000 corresponding number of -faotifr] i in //—tf-E—i-o o (CTG)n repeats is indicated DMWD DMPK SIX5 C o n g e n ita l (CTG)n > 1000

to the right. 1 • General Introduction 12 1 • General Introduction h ms peaet n bs spotd xlnto fr h molecular m the for explanation supported best and prevalent most The 1, 2, 8) Te n rsl i goa mirglto o gn expres­ gene of isregulation m global is result end The 181). 129, 111, CT) mtto i ito 1 f h zn fne poen (N9 mapping (ZNF9) 9 a protein later, finger years zinc the of Several 1 175). 86, intron (26, in 1B) (Fig. mutation (CCTG)n gene DMPK the of 3'-UTR ht h mtto laig o M i lclzd n h 3-T o te DMPK the of 3'-UTR discovery the in Upon belong. localized is DM1 to disorders leading mutation neurodegenerative the that other several and patterns. expression and functions protein and pie atr o te uceln fmiy MN 1 2 n 3 wihn the ithin w 3) and 2 1, of (MBNL ily presence fam revealed muscleblind patients DM1 the reten­ of from ribonuclear nuclear factors of tissue brain proteins, ofsplice Analysis 58). and (54, Length nuclear muscle (132). of aggregates in binding foci nuclear of ith misregulation w formation gene and to tion correlated is contributing repeat transcripts way the this soluble in that factors, speculated was it lately, Just 225). 65, (43, sion pathology. DM1 of nature ic tisystem successive in and (187) un­ tissues becomes it Once somatic 116). in (101, triplets, CTG CTGs expansion 35 in of 1000 over of resulting threshold a stable, repeat crosses a repeat (CTG)n carry the individuals affected r­ ta re mental hypotonia, ortality, m neonatal high by characterized form h rpa ae orltd Fg 1) A uafce pro cris short a carries person unaffected 1B). An (Fig. correlated 225). (167, are DM2 of repeat cause the the as identified was 3q21 chromosome to otiig og U-eas r as al t bn tasrpin n splice and transcription bind to able also are CUG-repats long containing be cannot structure be must protein abnormal pathogenesis of of formation mechanisms because special that operational realized was it gene, also which to disorders of family heterogeneous a disorder, expansion pathogenesis molecular 2. DM1 • mRNA DMPK about knowledge extensive gain to importance paramount of 21 RA pathogenesis 2.1. RNA • earlier appear symptoms and severity generations that between meaning increases anticipation, causes disease of expansion Repeat generations. classical the ith w Patients allele. affected their in CTGunits 35-150 carry (101). distress congenital severe respiratory a as and occur dation also can DM1 DM2, to Contrary (94). order utntns ies, RX, aiu tps fsioeeelraai (94) DNA ataxia or spinocerebellar of icrosatellite m types or various mini- FRAXA, disease, unstable Huntington's an as classified now is DM1 rtis o om ioula poen opee o anra ntr (75, nature nuclear abnormal of ith w (CUG)n complexes aggregate large protein they where containing ribonuclear nucleus, form products mechanism, to the in proteins trapped gain-of-function transcription are toxic formed repeats RNA-based newly an which is in DM1 of pathology mul- the explain to postulated were hypotheses several Hence, involved. 187). 102, (31, life in severely most the and CTGs 100-1000 of repeats have phenotype DM1 dis­ diagnose to difficult a DM make onset of age variable a and periods eet opsd f -5 T tilt, hl midy fetd patients affected ildly m while triplets, CTG 5-35 of composed repeat the in 19q13 chromosome on groups several by discovered was DM1 ing n re o etr nesad h mutsse c hrce o D1 t is it DM1 of character ic ultisystem m the understand better to order In n h ery ieis f h teteh etr te uain underly­ mutation the century twentieth the of nineties early the In o D1 i hs en on ta svrt o smtm ad egh of length and symptoms of severity that found been has it DM1, For

ribonucleoprotein (RNP) aggregates. It is therefore commonly believed that the amount of MBNL available for proper regulation of splicing is reduced. Although not found in nuclear foci but believed to bind soluble, expanded transcripts, other RNA processing factors able to bind (CUG) n repeats are upregulated or are behaving abnormally, like CUG-binding protein 1 (CUG-BP1) a member of the CUG-BP and ETR3-like (CELF) fam ilies of factors (132, 278) or HnRNP-H (139). Furthermore, several other proteins not involved in splicing such as protein kinase R (276), transcription factor SP1 (specific protein 1), signal transducers and ac­ tivators of transcription (STAT1 and 3), the retinoic acid receptor gamma subunit (58, 65) and Nkx2.5 (306) may also be involved in DM1 via an RNA-mediated mechanism. In DM1 tissues, multiple splice abnormalities have been described (58), including in muscle: splicing of the mRNAs for chloride channel 1 (ClC-1 ) (40, 180), sarcoplasmic/endoplasmic reticulum Ca2+ ATPase 2 (143), cardiac troponin T (110, 219), insulin receptor (IR) (248), and the ryanodine receptor (143). In brain, processing of the mRNAs for Tau (164, 253), and amyloid precursor protein (129) is misregulated. Two of these aberrant processing events have been associated with typical muscle symptoms: myotonia and insulin resistance are associated with mis- splicing of the ClC-1 and the IR respectively (58). Alterations in splicing mostly involve a shift in expression from an adult to an embryonic mode of splicing, with production of isoform products typical for the immature state. Both reduction of available MBNL and increase in CUG-BP1 expres­ sion support this shift in splice mode, since MBNL induces adult-type splicing whereas CUG-BP1 is known to promote embryonic splice modes (58).

• 2.2. Small noncoding RNAs In recent years it has become clear that the non-coding DNA sequences in the genome are not just noncoding or "stuffer” DNA, but that they contain essential sequences encoding e.g. microRNAs (309), which have been found to be important in the regulation of embryogenesis (308), protection against cell stress and pathogenic insult and in gene regula­ tion in general (309). In a paper by Malinina it was hypothesized that the hairpin structure of the CTG-repeat in the mutant DMPK gene transcript, when processed by the RNAi machinery, could give rise to CUG containing small noncod­ ing RNAs (178). In turn, these sm all noncoding RNAs could serve in breakdown or blocking of translation of CUG- or CAG-containing mRNAs. Although the CUG containing mRNAs are not fully complementary to the CUG containing sm all noncoding RNAs, each U-U mismatch is flanked at each side by two C-G pairs providing sufficient stabilization for the formation of near complementary double stranded RNA molecules. Pre­ dicted candidates that could be affected are MBNL1 (CUG-repeat) or the myosin phosphatase-Rho interacting protein (CAG-repeat), where the silencing effect of the CUG containing small noncoding RNAs would be most effective against the CAG sequence (178). Concrete evidence to support that this hypothetical mechanism is in fact active in DM1 was provided by Krol et al. (153) who showed that transcripts containing long CUG repeats, like the DMPK transcript, can be a target for Dicer. The 1 1 • General Introduction

13 14 1 • General Introduction h (T) rpa i te MK ee s oae wti a p iln, which island, CpG a within located is gene DMPK the in repeat (CTG)n The ut ontem f MK I fc, h (T) rpa oelp wih the ith w overlaps repeat (CTG)n the fact, In DMPK. of downstream just 17 25.T suy h rl o SX i D1 ahlg i mr dti Six5- detail more in finding pathology DM1 the in by SIX5 of role the strengthened study To 275). (147, further been has idea This 245). 213, (107, n seea msl ad ets al tsus fetd n M pathology, DM1 in affected heart tissues eye, ll in a testis, expression and Its 147). muscle (103, region skeletal and promotor/enhancer SIX5 ex­ in unaf­ reduction and consistent a patients DM1 reveal in not did DMWD of however, profiling individuals, fected expression Comparative DMPK of 1B). (Fig. 247) 153, 147, 125, transcription (77, SIX5 and altering in DMWD hypothesis genes the to involved led be neighboring may repeat and the of expansion downstream repeat This just 152).that 147, 83, located (22, site sensitive structure chromatin condensed more a in sulting emnl idct ta a erae n I5 xrsin a b involved be may expression SIX5 in decrease a that indicate seemingly 50% about by reduced is patients DM1 in SIX5 of expression mRNA that repeat downregulate to pathway interference RNA the in used are which il n h seuaie hs, oee, n te re oe f neighboring of role true the and however, phase, which 245), (144, speculative the in spermatogenesis in till s role a plays SIX5 that suggesting with patients also that fact the Moreover, DM1. in pathology ocular the r ee 9 n ua) lctd ut ptem f MK hv be found been have DMPK, of upstream just located human), in 59 gene or hyper­ DNase a of disruption a of finding ­ re the an area, ith w by the together across interrupted binding is observation, nucleosome island CpG increases the this When (22). repeat kb 3.5expanded over to extends effects gene 2.3. Neighboring • pd aaat t n al ae 16 25 23. oehr tee findings these Together, 243). 245, (146, age knockout early an Homozygous at 243). (146, cataract oped generated been have mice deficient noncoding all sm that evidence provide results These mRNA. containing otoes o ti tpc ean. hr sm suis eot increased report now, studies Until some tissues. Where patient remains. in topic levels this on expression DMPK controversy identify to dertaken debate. protein under DMPK of till s expression 2.4. Aberrant • therefore is is work pathology this of disease DM1 Much in DMPK of patients. DM1genes in hypogonadism to linked be could (224), patients DM1 in found as type same the of not were mice deficient rdcs eeae b Dcr r sot U rpas ht c a siRNAs as act that repeats CUG short are Dicer by generated products mlcts I5 s n motn cniae ee o ivleet n DM1 in involvement for gene candidate important an as SIX5 implicates 83). 71, (4, dismissing etiology found, disease been DM1 has in between levels involvement for correlation no expression candidacy DMWD DMWD's Furthermore, and group. length patient repeat DM1 the in pression pathology. DM1 in implicated tissues (127), testis and 299) (127, brain in pathogenesis. DM1 of development the in role a play could RNAs MK xrsin ohr so a erae 3, 3 8, 7, 0, 298). 206, 176, 85, 73, (35, decrease a show others expression, DMPK Six5- in observed cataracts the Since DM1. in formation devel­ allele, cataract Six5 in one only lack that mice also interestingly, and, mice vr ic te icvr o te MK ee uain suis ee un­ were studies mutation, gene DMPK the of discovery the since Ever fertility, male ex­ impaired SIX5 also genome abnormal of showed mice KO unrelated involvement Six5 Finally, causal completely (224). a simple of pression against basis to the argues on mutation SIX5-deficiency of cataracts develop contribution DM2 the about doubt till s is there however, h gn ecdn hmooan rncito fco SX i located is SIX5 factor transcription homeodomain encoding gene The rdcs f h DW gn (omel kon s M-9 n mouse in DMR-N9 as known erly (form gene DMWD the of Products

Reduced expression is consistent with a haploinsufficiency model, where mRNA derived from the DM1 allele is retained in the nucleus in nuclear foci and unavailable for translation into DMPK protein products (54, 182, 272). To investigate the effect of haploinsufficiency and the function of DMPK, DMPK KO mouse lines were generated, independently by two labs, including ours (126, 227). Replacement of the first seven exons of the DMPK gene by a neomycin resistance cassette resulted in complete ab­ sence of DMPK protein (298). KO mice were viable and only showed mild myopathy in head and neck muscles at an advanced age (126). Skeletal muscle cells derived from these mice demonstrated impaired calcium homeostasis (13). The DMPK KO mouse generated by Reddy et al. via a similar strategy displayed a late onset progressive skeletal myopathy, characterized by a decrease in force generation and increased fiber degeneration (227). Upon closer inspection, these mice were shown to display cardiac conduction defects including first, second and third de­ gree atrioventricular block. The development of third degree heart block closely resembles the condition found in DM1 patients, suggesting a potential role for DMPK haploinsufficiency in DM1 etiology (16). Further­ more, DMPK KO mice showed increased insulin resistance (170) resulting in a significantly higher body weight when they were fed a high-fat diet (169). Loss of DMPK expression might therefore be involved in the devel­ opment of type II diabetes in DM1 patients (170). To examine cell biological and physiological consequences of DMPK overexpression, a transgenic overexpressor mouse model (Tg26) has been generated. Approxim ately 25 copies of the entire human DMPK gene including regulatory flanking segments were tandemly integrated into the genome of this model. This resulted in a continuous surplus of DMPK transcripts, resulting in 5-10 fold excess of human DMPK pro­ tein over endogenous mouse DMPK protein levels (126). Tg26 transgenic mice demonstrated hypertrophic cardiomyopathy, reminiscent of human cardiac hypertrophy, and neonatal mortality indicating a detrimental ef­ fect of DMPK overexpression during pregnancy (126). Additional studies revealed cardiac dysrhythmia, myopathy in skeletal muscle and hypoten­ sion of smooth muscle suggesting a role for DMPK in a dose- or activity- dependent manner safeguarding the physiological functioning of skeletal muscle (211). Since DMPK was found to be involved in myogenesis, hav­ Figure 2 • DMPK and NDR families of serine/threonine ing either too much or too little of this protein could induce aberrant protein kinases. Phylogenetic tree of the DMPK (DMPK, MRCKa, -ß and % -y, ROCK-I and II, and CRIK) -----DMPK...... (100) and NDR families (NDR1 MRCKa...... (71) and -2 and L a ts l and -2) of — MRCKß...... (69) protein kinases. The struc­ — MRCKy...... (70) tu re of the tree is basec on homology comparison ---ROCK I...... (52) across the serine/threonine ------ROCK II...... (51) protein kinase domain, using ------CRIK..... (47) ClustalW (1 84, 274). Mouse ------NDR1...... (46) sequences were used and percentage sequence identity ------NDR2...... (45) relative to the DMPK kinase ------LATS1....(42) domain is given. The tree ------LATS2...... (43) is drawn to scale. Adapted

from W a nsink et al. (297). 1 • General Introduction

15 16 1 • General Introduction G (APdpnet rti kns/rti kns Gpoen iae C) kinase G/protein kinase kinase/protein protein (cAMP-dependent AGC (297). organization structural comparable a share members family DMPK (184). o h srn/hrnn kns dmi va ln lne (3 183). (33, linker long a via domain kinase serine/theronine the to both DMPK resemble 2 and NDR1 183). (33, 2) (Fig. 2 and 1 suppressor) remains It model. mechanistic one byjust explained easily not is thology only. level protein features the multisystemic the of complexity full the hand, other the Ontein. iia i dmi ognzto, MK ooos ifr osdrby in considerably differ homologs Although DMPK 313). domain (232, organization, length kinase domain variable in of serine/threonine region similar a coiled-coil and a by N-terminus followed leucine-rich a with of group the to belonging kinase protein serine/threonine a as sified characterized by an N-terminal ubiquitin-associated domain, connected connected domain, ubiquitin-associated N-terminal an by tumor (large characterized LATS and 2 and 1 Dbf2-related) (nuclear NDR domain kinases of homology citron and domain cysteine-rich domain, homology ki­ Cdc42-binding kinase-related dystrophy myotonic homologs, closest members family • 3.1. DMPK pa­ DM1 of nature multisystemic the that clear become has it time Over protein • 3. DMPK at effects by explained be not certainly can patients DM1 in observed in domain structure and length. LATS1 and 2, on the other hand, are are hand, other the on 2, and LATS1 length. and structure domain in the of products RNA and side. by protein of side gene properties DMPK investigate to important pro­ DMPK of levels aberrant to DM1 correlated in be found symptoms somehow could several together, patients Taken (36). maturation muscle presence of additional domains like a GTPase-binding domain, plextrin plextrin and domain, domain GTPase-binding a coiled-coil like variable domains the additional are of presence differences main The length. 2) (Fig. member the up archetype the make is (173) itself (CRIK) DMPK kinase which for citron and subfamily, containing (232) DMPK II coiled-coil and I (ROCK) rho-associated 208), (166,kinase y its and ß with a, together (MRCK) nase DMPK kinases, AGC of group the s­ Within cla (183). is kinases mammals, in found only protein young evolutionary an DMPK, More distant relatives of DMPK are members of the NDR subfamily subfamily NDR the of members are DMPK of relatives distant More UCE NON-MUSCULAR MUSCLE oge brain stomach testis tongue muscle skeletal esophagus iprg eyes diaphragm bladder intestine heart Table 1 • DMPK tissue tissue DMPK • 1 Table ad a p te d fro m W a n s in k et et k in ). s 7 9 n a (2 W al. m fro d te p a ad of n ssio re p ex w o h s at th distribution. ies in m ic e and hum ans, ans, hum and ­ d e tu ic s m m in fro ies tained b O s PK. an DM rg o or s e u s tis of Listing Figure 3 • The DMPK gene A ATG (CTG)n gives rise to different, i> I alternatively spliced, serine/ threonine protein kinases. (A) The DMPK gene consists 2 34 5 6 7 8 9 10 11121314 15 < of 15 exons and is strongly conserved between man and mouse (human DMPK gene, drawn to scale). Rectangles with VSGGG motif tail 1 tail 3 indicate exons; the straight line represents introns. Exonic parts that are subject KÎHÎÏ > to a lte rn a tiv e s p lic in g -i.e., 12 V 1213 14 15 a lte rn a tiv e u se of 5' and 3' 13 14 s p lic e s ite s- in ex o ns 8 and w/o VSGGG motif tail 2 14 are in d icate d in b lack. (B) Detailed illustration of alternative splice modes ir the DMPK gene. Inclusion VSGGG of 15 n u c le o tid e s in exon 8 C motif r e s u lts in p re se n c e of the extension Leu-rich to kinase coiled C-terminal VSGGG motif (left panel). domain Ser/Thr kinase domain domain coil tail Alternative use of four n u c le o tid e s in exon 14 is DMPK A responsible for two different tail 1 open reading frames (ORF), DMPK B defining C-terminal tail 1 and tail 2 (middle panel). Asterisks indicate stop DMPK C c o d o n s in the O RFs. Sk ip pin g tail 2 of exons 13 and 14 is a DMPK D smooth-muscle specific event, w h ic h re s u lts in a stop codon at the beginning DMPK E of exon 15 and expression of a short two-amino-acid DMPK F C-terminus (tail 3) (right panel). (C) Domain organiza­ tion of the six major DMPK • 3.2. Cell type and tiscue distribution of DMPK expression isoforms A through F (drawn DMPK prote in is r xpres sed ir a wid e range of tiss ues, with hi gd est to scale). All isoforms have a leucine-rich N-terminus express ¡0 n in smo o th mus o le, Ske stomach and hlad de r (Fig. 3) (95, followed by a serine/threo­ 2 11, 114 4). In other muscle type s like hea rt, skeletal muscl e, tongue and nine protein kinase domain, diaphragm, DMPK is also found (Table 1) (95, 126, 174, 244). Further including an extension to the kinase domain, an specification ot the exptession pnttert in sl

17 18 1 1 • General Introduction Transcripts from the DMPK gene are subject to extensive alternative alternative extensive to subject are gene DMPK the from Transcripts ATPase and in the T-tubules of type I skeletal muscle fibres (283). In car car - In (283). fibres muscle skeletal I type of T-tubules the in and ATPase eos -) a ahlcl oldci dmi (xn 1-2 ad sev­ and 10-12) (exons domain coiled-coil a-helical an 2-8), (exons was found that exon 8 contained an alternative 5' splice site, resulting resulting site, splice 5' it alternative study an this In contained 8 products. exon gene hDMPK that DMPK found mouse transgenic was a endogenous in and investigated transgenic was transcripts DMPK of splicing 296). (286, below further outlined as residence DMPK for was DMPK (218). al. et Pham by found was junction neuromuscular the find­ immune-histochemical interpreting in taken be should care and of development the with improved been has situation this recently, until DMPK the of Products findings. these Cross-reaction explained 285). proteins 283, 218, (159, kDa DMPK-related 84to to 42 from ranging weights, pie rdcs 9) Te 5 xn o te MK ee noe protein domain a kinase protein encode gene and serine/threonine a DMPK mouse the N-terminus, of exons leucine-rich between 15 a with The (95). conserved products largely splice is that process a in splicing, fluorescent with studies from came DMPK of localization subcellular 284) 218, (68, cytosolic or 284) as 256, 240, DMPK 205, (64, classified bound prediction membrane hydrophobicity a because also results, these spinal adult like types cell other In 283). (205, junctions gap to sociated overexpressor mouse model, allowing direct comparison of human DMPK DMPK human of comparison direct allowing model, mouse alternative study, overexpressor one In 13-15). (exons 3) (Fig. C-termini different eral isoforms DMPK of splicing • 4.1. Alternative isoforms splice • DMPK 4. locations mitochondrial main ER, the as identified studies were these cytosol In and lines. (MOM) cell membrane outer transfected permanently or ER the at (68), cytosol the in localized DMPK brain, in neurons motor 284), (205, reticulum ic sarcoplasm the in found SerCa the was near DMPK 256), 240, muscle, (64, diac reticulum ic at sarcoplasm protein the the at of detected expression no panel a DMPK using against However, antibodies transduction. signal monoclonal inof role a implicating ceptor Not job. difficult and tedious a DMPK against antibodies high-affinity 42-55 the so range, size kDa 65-80 the in fall to predicted were gene n rdcin f MK RA ih ioom A C n E o wtot (iso­ without or E) and C A, (isoforms with mRNA DMPK of production in re­ acetylcholine the near the in muscle, DMPK found skeletal 301) of 174, (11, junction groups DMPK, of Several neuromuscular reports. localization early in ings subcellular the about remains confusion 218).bodies, (159, specificity DMPK true with antibodies monoclonal man. Alternative splicing results in six major isoforms and several minor minor several and isoforms major six in results splicing Alternative man. or transiently in vectors cDNA from variable expressed for isoforms, evidence will tagged as protein Independent isoforms 296). 286, splice (95, multiple below of can DMPK existence discussed for the be by described explained localization in best be variation the Nowadays, protein. interpret to difficult been has it Overall (8). microtubules dendritic near s­ a was DMPK where 301) 240, 218, (174, discs intercalated the to next specific of production the made (159) ß and DMPK MRCKa towards like Reactivity no 218). 159, (95, were homologues soon products true antibodies considered first-generation longer the by recognized proteins kDa eas mn ery tde md ue f h frtgnrto anti­ first-generation the of use made studies early many Because forms B, D and F) a 15-nucleotide stretch encoding a VSGGG amino acid motif (Fig. 3B and C). A second alternatively spliced region encompasses exons 12-15, resulting in the use of different ORFs for the production of DMPK proteins with different C-termini. Inclusion of exons 12-15 results in DMPK isoforms A and B containing a hydrophobic C-teminus (tail 1), with a molecular size of ~70 kDa (Fig. 3B and C). When an alternative splice acceptor site in exon 14 is used instead of the normal 3' splice site of intron 13, the first four nucleotides of exon 14 are excluded from the mRNA. This use results in a frame shift in the ORF, which translates into a less hydrophobic tail (tail 2) of nearly the same size (Fig. 3B and C). Fusion of exons 12 and 15, i.e., with skipping of exons 13-14, leads to a frame shift in the ORF with a stop codon appearing at the beginning of exon 15. At the protein product level this gives rise to a truncated C-ter- minus composed of only two amino acids (tail 3), with a molecular size of approximately 60 kDa (for both DMPK E and F, see Fig. 3B and C) (95). Alternative splicing is used differentially in different tissues. Long DMPK isoforms A-D are predominantly expressed in skeletal muscle, heart and brain, whereas short isoforms E-F are mainly found in smooth muscle (95, 211). All the above mentioned mature transcripts contain the (CUG)n repeat in the 3'-UTR, either of normal length when transcribed from the healthy allele or of increased length when originating from the mutated allele.

• 4.2. Protein domains

• 4.2.1. Leucine-rich N-terminal domain The typical modular structure of DMPK and the different sequence motifs in the various modules is well preserved between members of the AGC subfamily of kinases. A leucine-rich N-terminal region of about 70 amino acids is commonly found in MRCKs, ROCKs, CRIK and DMPK (Fig. 3C). This domain may serve in controlling the activity of the adjacent serine/ threonine kinase domain in these proteins, either by facilitating linking to other signaling proteins, by being a scaffold for interacting proteins, or by inducing oligomerization of the protein. Directly, or indirectly, this domain may also be involved in directing the subcellular localization of the kinase (184). Specific evidence for a role of the leucine-rich domain in modulating DMPK activity has been provided by Wansink et al. (297). The elucidation of the crystal structure of the DMPK kinase domain (in­ cluding the N-terminus) revealed that the leucine-rich N-terminal end could also be involved in DMPK dimerization (67, 89).

• 4.2.2. Serine/threonine protein kinase domain The kinase domain in DMPK (Fig. 3C) has all the characteristic features of catalytic domains in other kinases, especially the highly abundant subclass of serine-threonine specific kinases. Typically, a kinase domain consists of 11 major conserved subdomains separated by insertions and deletions (100, 297). In an active conformation ("on” state), the kinase domain is folded into two lobes forming the active cleft. A glycine-rich sequence (GXGX0G, 0 is usually phenylalanine or tyrosine) in the cleft is a component of the highly conserved phosphate-binding loop, called 1 1 • General Introduction

19 20 1 1 • General Introduction h srn/henn poen iae oan f MK s olwd y an by followed is DMPK of domain kinase protein serine/threonine The T (117). ATP loop. P the of nitrogens chain main the by controlled is catalysis for ATP a-helical coiled-coil region (Fig. 3C) (26, 95). The coiled-coil structure structure coiled-coil The 95). (26, 3C) (Fig. region coiled-coil a-helical hydrophobic the when that shown been has it kinases some For 124, 296). kinases AGC other 40 approximately in found is arrangement similar A (49, 50), and was later found in other proteins like tropomyosin, myosin myosin tropomyosin, like proteins other in found later was and 50), (49, iei cmlxs 27. utmrzto o DP ifune substrate influences DMPK of Multimerization (287). complexes timeric long contain family DMPK the of members Most 297). 173, (166, family show­ sections), later (see found been have substrates itself DMPK protein the natural in few site no observed, Although was (296). site DMPK of among lysines phosphoacceptor or the to autophosphorylation arginines N-terminal three least residues at of five the presence prefers and serine of the dimerization Finally, in 296). involved 270, is (165, domain a kinase MRCK the of and II extension ROCK C-terminal like members family C-terminal the in located motif VSGGG 117, The 210). (17, loop activation has domain additional this of role the kinases these of several in and of blockage in or residues 273). 210, (117, site site binding active of substrate ATP the displacement in results which dis­ frequently most The structures. various adopt to able therefore and t oiin 0, hc mks otc wt te ad popae of phosphates ß and a the with contact makes which 100, position at n fbioe (8 9, 8) Cie-ol oan ae fe peet in present often are domains Coiled-coil 188). 96, a-keratin (28, in fibrinogen element and structural main the as 1953 in described first was -op cpbe f idn AP 6, 0, 1, 7) Poe ainet of alignment Proper 117, 273). 100, (67, ATP binding of capable P-loop, ein f MK tef s uh malr n ecmass ny 6 resi­ ~65 only coiled-coil encompasses The and (297). aller sm acids (95). much is amino dues itself 600 DMPK least of at region of DMPK the regions of members coiled-coil all in importantly, more and proteins, eukaryotic • region 4.2.3. Coiled-coil (67). region N-terminal leucine-rich the to binding by DMPK DMPK other by used also is which mechanism a isoforms, the of DMPK activation different Activation in (296). involved be autophosphorylation to and found trans- both was in domain DMPK of kinase the of extension activation. kinase in involved be to presumed is which domain, ditional lobe, N-terminal the in helix aC or loop rearrangements activation major the either of involve disorder inactivation or of mechanisms cussed n lmtd rdcie oe o te osnu sequence. consensus the of a power only predictive substrates limited ing putative all Besides sequence. consensus the matches the of end C-terminal the of hindrance steric by inhibited be can kinase the phosphorylating by kinase the phophoinositide-de- activate recruits can it which 84, kinase-1, (17, phosphorylated pendent is phosphorylation motif by motif, phosphorylation activated be phosphorylation can hydrophobic which ki­ conserved the of a extension FXX[F/Y][S/T][Y/F], contains C-terminal domain so-called nase This extensively. studied been state constraint chemically a in not are state "off” the in kinases protein For DMPK this function is served by the highly conserved lysine residue residue lysine conserved highly the by served is function this DMPK For The coiled coil region of DMPK is responsible for the formation of mul­ of formation the for responsible is DMPK of region coil coiled The meitl Ctria o te iae oan DP cnan a ad­ an contains DMPK domain, kinase the of C-terminal Immediately DMPK is a serine/theonine protein kinase that favors threonine over over threonine favors that kinase protein serine/theonine a is DMPK structure, defined a adopt kinases which in state "on” the in Unlike binding and increases protein kinase activity (199, 287, 289). Likewise, the a coiled-coil regions in ROCK II and MRCK influence kinase activity, although in these two family members activity is inhibited by coiled-coil interactions (5, 270). Additionally, coiled-coil mediated multimerizing is involved in anchoring of isoform DMPK C to its specific localization at the mitochondrial outer membrane (286, 287).

• 4.2.4. C-terminal anchor The six major DMPK isoforms found in mouse and humans carry three

Figure 4 • Subcellular localization of DMPK splice isoforms. N-terminal YFP-DMPK fusion proteins were expressed in D M PK KO m y o b la s ts and visualized by confocal laser scanning microscopy. ER was counter-stained using a calreticulin antibody; mi­ tochondria were visualizec using MitotrackerRed. (A, A') YFP-mDMPK A is associated to the endoplasmic reticu­ lum. (B, B') YFP-mDMPK C is located at mitochon­ dria. (C) YFP-mDMPK E is lo c a te d in the cyto so l. (D, D') YFP-hDMPK A and (E) distinct C-terminal tails (Fig. 3C) (95, 296). Tails 1 (DMPK A and B) and YFP-hDMPK C are both asso­ ciated to mitochondria. Only 2 (DMPK C and n) are comprised of 96 and £3 7 amino acids roscectively, YFP-hDMPK A induces strong and are considerably longer than tail 3 (DM PP E and F), which consist of mitochondrial aggregation onty twa amino acid rdaidues. Ac mentioned earlier, sevaoal studies have around the nucleus. Bar = 10 pm. shown ttiat tail domains of the differe nt DMPK ^oforms influence acti v- ity of thn photein Dinase domrin, or its sdbstratd epecificity (1S9, °96). Furthermore, C-terminal tails are involved targeting of DMPK isoforms to a specific subcellular localization. In mouse, DMPK tail 1 induces ER localization, presence of tail 2 results in targeting to the MOM and a cytosolic localization is observed for tail 3 (Fig. 4A-C) (286, 296). It has been found that all information necessary for targeting to the ER is present in the tail 1, whereas for tail 2, additional presence of the coiled-coil region is a requisite for proper MOM targeting (286). Unlike tail 3, which does not contain targeting in­ formation (Fig. 4C), tail 1 and 2 anchor firmly at their target membranes. Although there is strong homology between mouse and human DMPK genes, a species difference was observed for isoforms A and B: mouse DMPK A and B target to the ER, the human orthologues are localized at the MOM (Fig. 4D and E). Furthermore, transient expression of hDMPK A induces mitochondrial clustering around the nucleus, ultimately leading to cell death. Expression of mouse and human DMPK C and D also results in localization to the MOM, but gives not rise to perinuclear mitochondrial clustering (286). Computer analysis revealed hydrophobic regions in tail 1 and 2, classifying them as membrane anchors. This specifies long iso­ forms A-D as tail-anchored (TA) proteins. It is still uncertain in what way DMPK isoforms are integrated in the lipid bilayer (this thesis) and if they require additional factors to facilitate membrane targeting. 1 1 • General Introduction

21 22 1 • General Introduction TA proteins end up in many cellular membranes they initially only target target only initially they membranes cellular many in up end proteins TA stretch a and domain N-terminal oytosolio a by defined are proteins TA 6, 5) n aotss (137). apoptosis and 254) (69, in eurd o tasoain f A rti truh hi tre me ­ em m target their through protein TA of translocation for required tion hampgring without anchor membrane the to appended lie can that acids their to closer N- anchor their membrane expose a also contain but which proteins, cytosol the to transmembrane terminus II type from them the from emerges region contain hydrophobic not the do since proteins TA and N-terminus. sequence, the signal from a acids amino 30 within cellular numerous in roles key play and cell the of membranes all at agt o R r O. o ec dfeet oprmns f h secretory the of compartments different reach To MOM. or ER anchor to to able target are proteins TA few A membranes. of number limited a to co-translational translocation (18). Some of these features distinguish distinguish features these of Some (18). translocation co-translational anchoring allows which region, C-terminal the in residues hydrophobic of proteins Tail-anchored 5.• brane is contained within the C-terminal region (24, 207, 209). Although Although 209). 207, (24, region C-terminal the within contained is brane 47). eventually 24, (18, will domain translocation II polar must type one in C-terminal although result the extending residues, 85 to that up mind in reach can bear translocation sequence. msmbraee signal cleavable cotrans- a are containing proteins pi without Recent si h pers transmembrane a translocated a ngI II thag Shu sho e axim wn lationally hy type mbem of er d amino rophilie Furthermore, N-terminus. to subject not are proteins TA translation, of termination upon ribosome located usually is that anchor an by membrane the to bound are proteins translocation protein 122), (119, transport vesicular including processes occur proteins TA 19, 222). (18, membranes of bilayer phospholipid the in n utpe ebae; ioui 2 5) n Bl (4) fr xml, can example, for (141), Bcl2 and (55) 2 mitofusin membranes; multiple in ih h Ntria bl o te rti lctd n h ctsl TA cytosol, the in located protein the of bulk N-terminal the With sn tucto aayi i hs en hw ta treig informa­ targeting that shown been has it analysis truncation Using T arg etin g to the MOM (2 ) or or ) (2 MOM the to g etin arg T pt r gese et e s e rg o B m fro d te ap d A e m iso x pero d e p lo ve e d lly fu e ran b m e m l a m o is x ro e p e th is y a w th a p of cytic o x e ts - o n d e n e rtm a e p th m o c r e th o to the m fro d e s a le re n e h W ( . ~ ). 8 (1 l a In ). w arro rey g d tte o a (d into in a n m o d ratio b u s atu m is by th in of a d m e o w d b llo su a fo ), into ER (3 n d e liz sertio in ia c e p s via is e m TA o s for y a w h ug o e tiv alth a rn ct, lte a dire an o ls a is ) (4 ­ s n tra r la u ic s e v by d e v ie h c a sert in n ca ) d N-end -en C k c la rey b g ith and w re tu c tru (s re c tly targ e t to c h lo ro p la s ts . . ts s la p ro lo h c i­ to d t e o ls targ a s tly in c te re ro p TA ts, plan i­ x ero p the ch a re to s in te ro p ). w rro a k c la b d tte o (d port n atio c slo ran T ). (1 ER the into protein TA a e m o s o rib cells. al anim in proteins TA of Routing • 5 Figure pathway, TA proteins first insert into the ER and via vesicular transport translocate to their final destination (Fig. 5). How this journey is effectu­ ated is determined by the lipid composition of the different membranes in de secretory pathway i.e., by the cholesterol concentration (18, 24) < Figure 6 • TA features ER determine targeting of TA- proteins to the ER, MOM or peroxisomes. Organelle-specific targeting of TA-proteins, which de­ pends mainly on the length of the hydrophobic stretch and presence of flanking c h a rg e d a m in o a cid s. TA proteins are depicted with a folded cytosolic domair followed by a variably- sized hydrophobic domain (rounded grey rectangle). Charged residues flanking the hydrophobic domain are indicated as + and -. (A) A short hydrophobic domain flanked by positively ch a rg e d re s id u e s d riv e s TA proteins to the MOM. (B/C) Loss of either of these fea­ tures -a short hydrophobic together with the: length and hydrophobicity of the membrane anchor domain or flanking chargec (20, 27). Overall in animal cells, TA proteins either insert into the ER re s id u e s - re s u lts in ta rg e t­ ing to the ER. (D) C-termini membrane, MOM, and in soma cases directly into Ihe peroxisomal mem­ with intermediate fea­ brane (Fig.5) (18, 20). In tine following section possible mechanisms and tures -slightly lengthened anchoring requirements will be further discussed. hydrophobic domain and/ or reduced positive charge- induce targeting to either • 5.1. Targeting to the ER membrane the MOM or ER. (E) For Although ER targeting requirements are variable, targeting to the ER is translocation to peroxisomes, a short hydrophobic domain controlled by a relatively long (>20 amino acids) stretch of hydrophobic is flanked by a moderate residues with high hydrophobic strength, flanked by a variable amount positive charge. Sequence- of charged residues (Fig. 6) (20). Insertion of TA proteins into the ER specific features like a membrane is direct and an ATP-dependent process, although the amount recognition sequence (black oval) facilitate direct inser­ of ATP required for membrane insertion differs between different TA pro­ tion into the peroxisome teins. Cytochrome b5 is able to insert into protease-treated microsomes (E). Adapted from Borgese in the near absence of ATP, while addition of ATP does not increase inte­ et al (20). gration into the ER membrane significantly (24, 305). It is hypothesized that this ability to independently insert into the ER membrane requires an ER membrane anchor with a more moderate hydrophobicity. Such an arrangement will give the protein higher solubility in the cytosol (18). TA proteins with a highly hydrophobic tail, like synaptobrevin 2 (Syb2), are more prone to form aggregates in the cytosol and therefore need ATP consuming chaperones to remain soluble and to insert properly into the membrane (24, 155). Interestingly, the ER membrane has been shown to be the most permissive membrane for accepting TA proteins. TA proteins, in vivo targeted to the MOM, are able to insert into microsomes in vi­ tro, demonstrating that ER membrane is capable of taking up many TA proteins when it does not need to compete with other membranes (18). TA proteins that target to the peroxisome can also travel via the ER, in the so-called pER mediated route (Fig. 5), as has been shown 1 1 • General Introduction

23 24 1 1 • General Introduction The general consensus for MOM targeting is dependent on a membrane membrane a on dependent is targeting MOM for consensus general The lhuh xetos xs () a Bx0-9Bx0-2-TMD-x0-1Bx0-6B a (7), exist exceptions stands Although B (where and characteristics similar showed which identified was Get3 Asna-1, post- interact can Asna-1 activity. ATPase with protein a Asna-1/TRC40, 265). Another interaction that facilitates ATP-dependent integration of TA of integration ATP-dependent facilitates that interaction other by Another 265). supported not and however, controversial, are results These (1). Sec61ß, and in this way was involved in insertion into the ER membrane membrane ER the into insertion in involved been was way have this in and insertion Sec61ß, of ways Several involved. are chaperones which a contains and hydrophobicity mild only has proteins TA some-targeted direct Alternatively, 203). 142, (66, peroxidase ascorbate and Pex15p for tde i mmain n yat el, hc rvae ta treig of targeting that revealed which cells, yeast and mammalian in studies tions solely based on hydrophobicity and charge of relevant sequences sequences relevant of charge and hydrophobicity on decreased to based leads solely - tions length TMD increasing or charges has positive of sequence tion consensus acid) amino any for x and residues 6). basic (Fig. for acids amino basic by flanked hydrophobicity, moderate of anchor receptor ER the to binding via insertion protein TA facilitate to found ­ was em m the of insertion and substrate TA the of release allowing hy­ activity, ATP of involvement that demonstrates This pathways. ­ em m translocation ER the at receptor and proteinaceous proteins TA a of to region protein the deliver protein membrane-anchoring thereby the cytosolic with kDa 40 the translationally is factor targeting well-defined most the f h poen cnrbt t treig 3 18 257). 118, (3, targeting to contribute protein, the of predic­ Still, 119). 113, 61, (18, ER the to rerouting even or binding MOM MOM • the 5.2. to Targeting known was much Not isoforms. DMPK of characteristics and distribution 268). 251, 189, (114, Get1/Get2 complex for homologue yeast a Recently, (265). membrane the into anchor was brane model mechanistic a findings these on Based essential. is resulted drolysis Asna of mutants ATPase-defective of Expression (265). brane 264, (155, SRP the or GTP not but ATP on dependence and showed who Syb2 groups with interact could (SRP) particle recognition signal the GTP netgtd n i ws hw b Hg' gop ta tgte wt AP or ATP with together that group, High's by shown was it and known not is itinvestigated cases of majority the for although chaperones, molecular (18). 6) (Fig. C-terminus peroxi­ in charged the domain positively to tail binding the via Surprisingly, hFis 108). (99, and 5) Pex26 (Fig. for receptor found Pex19 was targeting peroxisomal in inserting TA protein into the MOM. The answer came from two recent recent two involved from classical is came the answer complex) The whether MOM. TOM (i.e., the into raised protein TA been machinery inserting in has translocation question the mitochondrial 5), (Fig. (18) ER may not be accurate and additional signals, probably in the N-terminus N-terminus the in probably signals, additional and accurate be not may dele­ f.e. by - residue one either altering that found was sequence it this in as residue essential each ofis positioning Correct (137). proposed been period. thesis this of start the the at anchoring understanding in membrane DMPK value regarding great of is proteins TA of mechanisms ing Asna-1 stimulates membrane ER the on receptor a whereby proposed, other influence not did but insertion, protein TA on effect inhibitory an in Finally, (2). Hsc70 and Hsp40 chaperones molecular involves proteins h dpnec o AP o E treig sal sget a ed for need a suggests usually targeting ER for ATP on dependence The ic tasoain o h MM s iet n de nt cu va the via occur not does and direct is MOM the to translocation Since nraig nweg o mmrn acoig n udryn target­ underlying and anchoring membrane of knowledge Increasing several TA proteins occurs independent of the TOM translocation ma­ chinery. Absence of multiple components of the TOM complex in either system did not interfere with MOM targeting, suggesting the existence of a fully independent machinery for MOM translocation (138, 254). In the mammalian cell system MOM targeting was independent of ATP and cytosolic molecular chaperones, even though in some situations chaper­ ones could facilitate integration. Presumably, this was caused by folding requirements of the N-terminal domain of the TA proteins (254). In yeast membrane composition influenced TA protein integration, since elevating ergosterol levels inhibited membrane insertion (138). In summary, MOM targeting is determined by the composition of the TA protein's C-terminus and by membrane lipid composition. Molecular chaperones may facili­ tate the process, but it does not involve the classical TOM translocation machinery.

Figure 7 • DMPK, activators and substrates. Scheme of known activators and substrates of DMPK. Substrates are function­ ally categorized into three groups of clearly distinct biological significance: i or homeostasis (light grey), actin cytoskeleton (medium grey), and splicing (dark grey). Also the pathogenic phenotypes that may be associated with DMPK substrate dysregulation are depicted. Abbreviations: PLM = phospholemman, PLN = phospholamban, DHPR = di­ DM1 myotonia cardiac myotonia motility conduction hydropyridine receptor, SkM1 sym ptom s contractility disorders, defects, = skeletal muscle voltage- cataract, myotonia, insulin metabolic gated sodium channel 1, resistance alterations SRF = serum response factor, MYPT = myosin phos­ phatase targeting subunit, CUG-BP = CUG binding • 6. DMPK substeates and function protein, MKBP = DMPK binding protein, HSP25/B1 = heat shock protein 25/B1, DMPK research over the oears has provided a limited number of clues aB-Cryst. = aB-Crystallin. for the protein's biological significance. Direct and indirect evidence is available for a role in actin cytoskeleton dynamics and ion homeostasis, as will be described below.

• 6.1. DMPK and actin-myosin dynamics DMPK family members MRCKs, ROCKs and CRIK play a role in actomyosin cytoskeleton dynamics, in processes like cell motility, cytokinesis and muscle contraction. In these activities these enzymes are under the control of members of the family of sm all GTPases, Rho, Cdc42 and Rac (74, 92, 173, 232). Based on these findings, an analogous situation was predicted for DMPK, but based on sequence homology, DMPK does not contain a typical GTPase binding domain. Nevertheless, Rac involvement in phosphorylation and activation of DMPK was reported in one study (Fig. 7) (255). When DMPK was isolated from cells exposed to the G- 1 1 • General Introduction

25 26 1 • General Introductor lee mdlto o seea msl vlaegtd a canl by channels Na* voltage-gated muscle skeletal of modulation Altered abnormalities found in DM1 (79, 195, 200, 226). Since complete heart heart complete Since 226). 200, 195, (79, DM1 in found abnormalities voltage-gated cardiac the and DMPK of coexpression Surprisingly, 201). those seen in DMPK-deficient skeletal muscle (162). These data provide provide data These (162). to muscle similar skeletal channels Na* DMPK-deficient of in seen reopenings those late multiple showing thoroughly, and amplitude current Na* reduced mice KO DMPK heterozygous and (37, manner dependent phosphorylation a in decay, current accelerated based acto-myosin in role DMPK's establish studies to follow-up but necessary role predicted, A thus 282). is clearly 232, (74, are processes apoptosis these all and in DMPK for cytokinesis formation, fiber stress enhanc­ thereby complex, phosphatase myosin the of deactivation in phosphorylated sult only DMPK but (94), DMPK bind to shown were MRLC light and regulatory phospha­ myosin the of myosin of status subunit a is phosphorylation the in MYPT 7). involved (Fig. tase (MYPT) subunit targeting ay o lrf ti point. this clarify to sary trafficking improper (130), blebbing lens apoptotic in DMPK for are role a question this on answers functions The biological activity? the DMPK are by then promoted What in are activity. that signaling DMPK of G-protein of Thus, regulation (255). involvement 7) to the point (Fig. findings pathway kinase, these kinase Raf by MAP Fur­ collectively, the in activated (29). enzyme and cell Ras-activated phosphorylated non-treated be the from also can isolated DMPK DMPK ofthermore, that to compared two ada anraiis bt ehnsi dtis n hs poess re­ processes these in and details myotonia like mechanistic but symptoms in DMPK abnormalities, of cardiac involvement the for evidence homozygous of myocytes Na* skeletal alter and not did muscle site, skeletal In phosphorylation (37). identical currents an containing channel, and Na* amplitude current Na* voltage-gated peak lower skeletal significantly with in a together resulted DMPK channels human of coexpression homeostasis ion and• 6.2. DMPK definitely. more dynamics actomyosin enhances and bly assem actin-filament promotes MRLC of MYPT both screen yeast-two-hybrid a In 296). 287, (204, (MRLC) phosphatase chain myosin the via exerted be may cytoskeleton actin the on yokltn y MK drcl o idrcl, u frhr ok s neces­ is work further but indirectly, or term directly long DMPK, by synaptic cytoskeleton decremental and 170) (169, receptor insulin the of an unsolved. main in currents Na* DM1, in investigated occur were mice KO also DMPK heterozygous death and homozygous of cardiac hearts channel sudden Na* and reproducing block found, were reopening channel Na* increased DM1 underlying that mechanism shown have possible oocytes a Xenopus in as Experiments 201). (37, proposed pathogenesis been has DMPK size, and shape cell contraction, muscle influencing thus interactions, phosphorylation increased turn, In MRLC. of re­ state would phosphorylated activity the ing DMPK manner, this In 296). 204, (94, MRLC not MYPT, oetain 22. hs efcs ol al e ikd o euain f the of regulation to linked be all could effects These (252). potentiation members. family kinase other the for than clear less activator protein nte DP tre, hc i a addt fr en ivle i myo- in involved being for candidate a is which target, DMPK Another Studies using DMPK overexpression- and knockout mice revealed revealed mice knockout and overexpression- DMPK using Studies Independently, there is also evidence to suggest that effects of DMPK DMPK of effects that suggest to evidence also is there Independently, GTP- y -S, its activity was increased by about a factor factor a about by increased was activity its tonia is phospholemman (PLM ) (Fig. 7), a protein mainly expressed in heart, skeletal muscle and liver. PLM functions as a regulator of the Na*/ Ca2* exchanger and Na*/K* ATPase (90). Upon expression in Xenopus oo­ cytes it induces Cl- currents, which are lowered upon coexpression with DMPK. This happens in a phosphorylation-dependent manner and is the result of a reduced amount of PLM at the oocyte membrane (199). In DM1 patients, the PLM-DMPK interaction can be an alternative explanation for reduced Cl- conductance and occurrence of myotonia. Furthermore, PLM is associated with heart failure, suggesting a possible involvement of DMPK in DM1 related atrioventricular block (15). It has furthermore been proposed that DMPK is involved in volt­ age-dependent Ca2* release in muscle. In DM1 patient muscle cells an increased cytoplasmic calcium concentration ([Ca2*]c) was observed, caused by an increased influx of Ca2* through the voltage-operated nifedipine-sensitive Ca2*, i.e., the dihydropyridine receptor (DHPR) (Fig. 7) (121). In myotubes derived from DMPK KO mice also an elevation in [Ca2*]i was observed, likely due to an altered "open” probability of the DHPR, reminiscent of the symptoms found in DM1 patients (13). Interest­ ingly, DMPK KO myotubes were found to have a reduced response upon triggering by acetylcholine compared to their wild type counterparts (13). This could be a direct effect of DMPK on the DHPR since the ß-subunit of the DHPR might be a phosphorylation substrate (277). Earlier studies in cardiac muscle from DM1 patients and DMPK KO mice suggested that DMPK is critical in the modulation of cardiac con­ tractility and maintenance of proper cardiac conduction (16, 101). Sup­ port for this idea came when DMPK was found highly enriched at the sarcoplasmic reticulum (SR) where it colocalizes with the ryanodine receptor and phospholamban (PLN) (Fig. 7). PLN, the muscle-specific SR Ca2* ATPase (SERCA2a) inhibitor, was capable of physically interacting with DMPK (134). In vitro experiments have shown that DMPK can use PLN as a substrate for phosphorylation, a finding which is consistent with the prediction of a phosphoacceptor site in PLN by Wansink et al (134, 296). When hypophosphorylated, PLN binds and inhibits SERCA2a, thereby impairing Ca2* sequestration into the SR. Exactly this was found in ventricular homogenates of DMPK KO mice (134). Taken together, these findings suggest that DMPK may be involved in the regulation of initial events that govern excitation-contraction coupling in skeletal or cardiac m uscles (13, 134).

• 6.3. Alternative DMPK effectors An alternative activator of DMPK may be DMPK binding protein (MKBP), which is localized in the cytosol and also found enriched in the neu­ romuscular junctions of skeletal muscle (Fig. 7) (218, 269). MKBP is thought to be involved in stress response in muscle. MKBP binding to DMPK and its selective upregulation in DM1 patients could thus be in­ volved in DM1 pathogenesis (269, 297). In a study using m ass spectrometry, DMPK interacting proteins were found which comprise parts of the contractile and myofibrillar m achin­ ery, or are either mitochondrial proteins, heat shock proteins or other proteins (78). Like MKBP, proteins involved in stress response have been identified like aB-crystallin and heat shock protein 25 and B1 (HSP25/ 1 1 • General Introduction

27 28 1 1 • General Introduction h dsoey ht CGn eet xaso i te MK ee s h root the is gene DMPK the in expansion repeat (CTG)n that discovery The Abnormal CUG-BP1 regulation is thus becoming a prominent candidate candidate prominent a becoming thus is regulation CUG-BP1 Abnormal files and distribution of long and short DMPK isoforms in various muscle muscle various in isoforms DMPK short and long of distribution and files tissue-type distribution of different DMPK gene products (i.e., mRNA and and mRNA (i.e., products gene DMPK different of distribution tissue-type biology. DMPK chapters of different the in aspects studies different on the isoform focus protein speculation, mere than abnormal ground with solid association an be also could there that also may transcripts, DMPK of repeat (CUG)n expanded the to binds that vivo, in and cell vitro in non-myogenic and SRF myoblasts C2C12 in phosphorylate to activity its able is increasing thereby DMPK that know we 7). (Fig. MKBP like fectors ye ad ri rgos te ise ta ae ot fetd n M. Spe- DM1. in affected most are that tissues the regions, brain and types patho­ DM1 to contributes mRNA DMPK expanded possibility toxic the on only concentrate to not is thesis that PhD this in described study the additive an have could CUG-BP1 of or toxicity, phosphorylation RNA-based repeat by abnormal caused whether uniquely is regulation abnormal One pathogenesis. olecular m DM the to and biology DMPK to complexity production from the mutant and normal DMPK genes. To put this on more more on this To put genes. DMPK but normal chapter), and this mutant 2, the from (Section production above described echanisms m via genesis underly­ for search intense an for start the was pathology DM1 of cause thesis this of outline and • 7. Aim DM1 in levels protein of CUG-BP1 inhibition to Since 280). leading 248, 233, ultimately (157, mRNAs, myogenesis multiple of processing proper direct a as CUG-BP1 of Identification (279). DMPK of control the under be Although protein. interacting DMPK as identified was ef­ are transcription they gene that speculated be can it DMPK, of activators or effectors rti) I Catr , e ecie o w aaye te xrsin pro­ expression the analyzed we how describe we 2, Chapter In protein). been has expression (120). gene 7) (Fig. a-actin role proposed A skeletal and interaction. this of cardiac of significance upregulation in the about known is much not lines, are proteins these if determined be to remains it Although (78). HSPB1) n dsae ehnss n ay aoaois n h wrd Te i of aim The world. the in laboratories many in mechanisms disease ing studies. future for subject is this role, whether question The pathobiology. DM explaining 248). (110, for DM1 of typical mechanism features 1, heart channel and chloride muscle T, skeletal to muscle-specific troponin the have leading and cardiac gene 1 model receptor, half­ insulin mouse the protein of myotubularin-related overexpressing missplicing CUGBP1 is CUG-BP1 demonstrated a increasing now CUGBP1 with PKC, that Studies by found (156). altered been DM1 of lives in has effect it an be hand hyperphosphorylated also other the could On this increased, phosphorylation. globally are affecting patients thereby nucleus, the may in this turn, CUG-BP1 In of CUG-BP1. decreased, of accumulation are state itself an to DMPK lead of hypophosphorylated levels a in the which resulting in proposed, of layer was another model adds (233), 7) (Fig. DMPK for substrate phosphorylation Splice and processing factor CUG-BP1, discussed above as the factor factor the as above discussed CUG-BP1, factor processing and Splice a-actin regulate to known protein a (SRF), factor response Serum is, e edd etr nih i te nrclua ad el and cell- and intracellular the in insight better needed we First, cifically, we report on detailed studies on this topic in muscle and brain- derived cell culture systems. DMPK is considered the first tail-anchored protein kinase. Alternative splicing of precursor DMPK mRNA gives rise to multiple isoforms, encoding kinases with different C-terminal membrane anchors. Chapter 3 describes how structural requirements for membrane targeting of these C tails and the membrane anchoring topology and dynamics of the entire protein were studied. In Chapter 4, we report on the finding that overexpression of the hDMPK A isoform results in mitochondrial clustering. The region responsible for this clustering was identified and the physiological consequences associated with this mito­ chondrial clustering were analyzed. In Chapter 5, we provide evidence for a link between the activity of individual DMPK isoforms and cytosolic and mitochondrial Ca2* handling in myotubes. In the same chapter, we also report on possible influences that DMPK-based changes in Ca2* handling have on the cellular energy state. Finally, Chapter 6 summarizes our find­ ings, explaining how results from our studies provide more insight in the potential link between DMPK biology and DM1 pathology. 1 1 • General Introduction

29 r 'i CHAPTER 2 L . J

DMPK PROTEIN ISOFORMS ARE DIFFERENTIALLY EXPRESSED IN MYOGENIC AND NEURAL CELL LINEAGES

Ralph J.A. Oude Ophuis, Susan A.M. Mulders, René E.M.A. van Herpen, Rinske van de Vorstenbosch, Bé Wieringa, and Derick G. Wansink

Department of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Muscle and Nerve 40(4):545-55, 2009 ► 32 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages To appreciate its normal and possible pathobiological role, we analyzed analyzed we role, pathobiological possible and normal its appreciate To • ABSTRACT ht r afce i DM1. a in as affected DMPK are identifies that muscle work or This neural of neurons. lines myoblasts, cell hippocampal in related cultured not in and but origin, detectable astrocytes expressed also cortical highly was myotubes, protein skel­ were DMPK and isoforms regions. brain diaphragm cytosolic heart, short in Long whereas tissues. dominated muscle, mouse in DMPK etal expression isoform splice membrane-anchored DMPK of manifestations, patterns DM1 the to contributor main the is toxicity mRNA-based DM protein kinase (DMPK) gene. It is commonly accepted that DMPK DMPK that accepted commonly is It gene. the of caused region (DMPK) disorder untranslated kinase 3' the in protein neuromuscular DM a segment is (CTG^CAG)n (DM1) unstable 1 an by type dystrophy Myotonic kinase with pronounced expression in diverse muscle and neural tissues tissues neural and muscle diverse in expression pronounced with kinase diverse in present were types isoform Both stomach. and bladder in protein. DMPK the of significance the about known is much not however • INTRODUCTION

Myotonic dystrophy type 1 (DM1) is a complex, multisystemic disorder which affects many tissues including skeletal muscle, heart, gastro­ intestinal tract and brain (101). The disease is caused by cell type- specific expression of an abnormal (CTG^CAG)n segment located in the 3' untranslated region of the DM protein kinase (DMPK) gene (94). The (CTG^CAG)n segment expands over generations in DM1 families and in somatic tissues during aging. In tissues like muscle and brain, (CUG)n- expanded transcripts become trapped in cell nuclei, forming abnormal RNA protein aggregates (300). The aggregation process -or early steps therein- is considered to be the main contributor to disease problems, because it compromises transcription or processing of specific mRNAs or exerts global cell stress (65, 300). Nevertheless, it remains to be seen whether all disease aspects, including the selective muscle wasting and loss of endocrine function, can be explained by the RNA toxicity model. There may also be a causative role for an imbalance in DMPK protein level or isoform distribution as well. There is considerable controversy about the fate and role of DMPK protein in tissues of patients with DM1 (summarized by Wansink et al. (297, 298)). Moreover, circumstantial evidence from animal and cell model studies connects DMPK to problems similar to those in DM1 mani­ festation. Lack of DMPK in knockout (KO) mouse models affects heart physiology (16, 126), viability and calcium regulation in skeletal muscle cells (13, 200) and intermediate memory potentiation in brain (252). In addition, our group demonstrated that overexpression of hDMPK gene products results in hypertrophic cardiomyopathy, cardiac dysrhythmia, myotonia and hypotensic stress (126, 211). DMPK protein occurs as six major splice isoforms, named A-F, in man and mouse (95). Mouse DMPK A and C, the prototype long C-tailed isoforms, localize to the endoplasmic reticulum (ER) or mitochondrial outer membrane (MOM), respectively (296). Strikingly, the human DMPK orthologues of these isoforms both localize to the MOM (286). Forced expression of the human DMPK A isoform in different animal and human cell types alters mitochondrial morphology, leading to perinuclear aggre­ gation of mitochondria with possible consequences for cell dynamics or viability regulation (Oude Ophuis et al., manuscript in preparation) (286). DMPK has been classified as a member of a subfamily of the large group of AGC protein kinases, to which also ROCK-I/-II, MRCKa/ß/Y and citron kinase belong (135, 232, 297). These kinases have been implicated in smooth muscle contraction, stress fiber formation, neurite retraction and cytokinesis (232, 260). DMPK has a presumed role in modulating actomyosin dynamics through phosphorylation of the regulatory subunit 2 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

33 34 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages a complementary approach at the protein level, we report on expression expression on report we level, protein the at approach segments, (CUG)n complementary expanded a of dose the and effects pathogenic specific study hybridization situ in an In products. gene DMPK all of distribution profiling of DMPK isoforms in tissues known to be prominently involved involved prominently be to known tissues in using Here, isoforms allele. DMPK of mutant the profiling from variants splice tissue- DMPK becomes between all in DM1 present where correlations DMPK finding on tissues that mainly all shown not was but was focus it many, The in manifest. co-workers, expressed and is Reddy by mRNA tissues mouse on in dihydropy­ Ca2* of transients the of regulation the in ß-subunit involved is with the of DMPK receptor, interactions rimidine phosphorylation direct and Through homeostasis, ion in currents. Cl-phospholamban engaged and is Na* Ca2*, DMPK to patients addition, regulating In DM1 of (170). susceptibility diabetes II increased type indicating membrane, cell the cytoskeleton the to Alterations 296). (204, (MYPT) phosphatase myosin of n ies pathology. disease in cell-type and tissue of picture detailed a have to important is it level 277). 135, (134, muscle skeletal and heart to receptor insulin the of trafficking in implicated are DMPK by induced o ul udrtn D1 ies eilg a bt te N ad protein and RNA the both at etiology disease DM1 understand To fully • MATERIALS AND METHODS

• Mice and tissues Tissues were isolated from wild type (WT; 129Ola/C57BL6 background), DMPK knockout (KO; 129Ola/C57BL6 background) and Tg26-hDMPK (TG; Fvb background) mice generated in our laboratory (126). Directly after killing the animals by cervical dislocation, the following m uscle-cell containing organ and tissue samples were individually excised and snap frozen in liquid nitrogen: ventricle and atrium from heart, bladder, stom­ ach, diaphragm and tongue, and soleus, gastrocnemius, tibialis anterior from the hind limb. Likewise, from the brains of each of three mice, the cortex, brain stem, cerebellum, hippocampus, striatum, hypothalamus and olfactory bulb regions were collected, snap-frozen and pooled. Also the brain-connected pituitary gland was isolated. All procedures involv­ ing animals were approved by the Animal Care Committee of the Radboud University Nijmegen Medical Centre and conformed to the Dutch Council for Animal Care.

• Primary myoblasts and myotubes Conditionally immortalized myogenic cell lines were derived from the calf muscle complex from WT and KO mice harboring the H-2Kb-tsA58 allele as described (197). WT and KO myoblasts were ring cloned, and selected for myotube formation ability. Myoblasts were propagated at 33°C on gelatin-coated dishes in DMEM (GibcoBRL, Gaithersburg, MD) supplemented with 20% (v/v) FCS, 50 pg/ml gentamycin and 20 units IFN-Y/ml. Myotube formation was induced by placing a confluent myo­ blast culture grown on Matrigel (BD Biosciences, the Netherlands) at 37°C in DMEM supplemented with 5% horse serum (HS) and 50 pg/ml gentamycin and maintaining these conditions for up to nine days.

• Primary hippocampal neurons and cortical astrocytes Primary cultures of cortical astrocytes and hippocampal neurons were established essentially as described (57, 128), Both cell populations (>95% pure) displayed expected characteristic morphologies. Astrocyte identity was verified by GFAP immunostaining (data not shown). Primary neurons were cultured without an astrocyte feeder layer and were main­ tained for up to ten days in culture.

• Established cell lines Mouse neuro-2A (N2A) neuroblastoma cells, human SK-N-SH neuro­ blastoma cells and mouse C2C12 myoblasts were grown as described elsewhere (286, 295). Rat PC12 pheochromocytoma cells were grown on collagen-coated dishes in DMEM supplemented with 10% FCS and 5% HS 2 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

35 36 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages (3 % (w/v)) and lysate of DMPK KO myoblasts were added to the blocking blocking the to added were myoblasts KO DMPK of lysate and (w/v)) % (3 glioblastoma U373 CO2. Human % 5 and 37°C at Netherlands) the (Sigma, (52). To decrease the background on blots using the B79 antibody, BSA BSA antibody, B79 the using blots on background the decrease To (52). antibody ß-tubulin 218), (159, MANDM1 antibody DMPK monoclonal (95), 217 (46, ). acid retinoic pM 20 and FCS 2% with supplemented MEM in hours 48 fe adto o rcmiat iu, rs nw eim ih composition with medium new fresh virus, recombinant of addition after pro­ to coupled (95), B79 antibody DMPK using immunoprecipitated was NP40 in ice on homogenized were mice TG and KO WT, from brains Whole Immunoprecipitation • or myotube and Myoblast buffer. sample SDS-PAGE with mixed were homog­ were 1% tissues Brain including homogenizer. EDTA) dounce mM 1 glass-glass Germany), a using (Roche, SDS cocktail inhibitor protease for differentiated were CO2. cells CCL-131 5% and 37°C at FCS neuroblasto­ CCL-131 10% Mouse with confluence. ~75% at harvested and 37°C at • Transduction blotting. western and SDS-PAGE by analyzed and ECL, by generated were signals and used, was UK) Laboratories, search blotting western by distribution DMPK of • Analysis tion procedure have been given in van Herpen et al. (286) Three hours hours Three (286) al. transduc­ et the Herpen of van Details in given cultures. been cell have myo­ other for procedure 37°C astrocytes, tion at primary and neurons, myoblasts primary of transfer gene mediated lysis NP40 in washed were Immunoprecipitates beads. A-Sepharose tein AR). X-OMAT (Kodak film to exposure by followed antibody GFP USA), Gilbertsville, (Rockland, antibody kinase pyruvate B79 antibody DMPK-specific used: were antibodies following the tection buf­ and sample above. cleared SDS described in were as buffer Lysates lysed sample were respectively. buffer, with mixed culture lysis in NP40 or cells fractions directly fer supernatant astrocyte and resulting neuron the and 4°C, at 14,000g centrifugation by at cleared min 10 were 1xfor Lysates buffer. lysis PMSF, NP40 in ice mM on enized 1 microcystin, pM 2 vanadate, mM 0.1 pyrophosphate, confluence near until days 2 for and non­ 37°C cultured and at beating HL-1 were FCS Mouse 10% confluence. cardiomyocytes with beating ~75% at supplemented harvested DMEM CO2 and in 5% cultured were cells supplemented MEM in subconfluent until days 2 for grown were cells ma brain of postnatal day 0 (~1.5 mg protein, determined as described (23)) (23)) one of described as homogenate determined protein, cleared mg (~1.5 the 0 half day to postnatal equal of 14,000g.brain at sample min 10 a for in protein centrifugation by DMPK cleared were Lysates buffer. lysis immunode­ For USA). (Millipore, membrane PVDF to blotting western by lss n moue ws efre i DE fr hus t 3C for 33°C at hours 3 for DMEM in performed was adenoviral-vector myotubes E, and and C A,blasts YFP-mDMPK of expression transient For buffer sample SDS-PAGE in solubilized were proteins bound and buffer, ImmunoRe- I g (Jackson G HRP-conjugated antibody, secondary As buffer. USA), Iowa, of University Bank, Hybridoma Studies (Developmental E7 sodium mM 1 NaF, mM 25 (1% NaCl, buffer mM lysis 150 7.5, pH NP40 in ice Tris-HCl, on mM 50 homogenized NP40, were tissues muscle Frozen Protein lysates were separated on 8% SDS-PAGE gels and transferred transferred and gels SDS-PAGE 8% on separated were lysates Protein as indicated was added, and cultures were maintained for 20-48 hours.

• Immunofluorescence microscopy Cells were grown on glass coverslips, coated with gelatin (myoblasts), Matrigel (myotubes) or poly-L-lysine (Sigma; neurons and astrocytes). Myoblasts, neurons and astrocytes were washed once with phosphate- buffered saline (P B S ) and fixed in PBS containing 2% (w/v) formaldehyde 20-48 hours after transduction and permeabilized in PBS containing 0.5% (w/v) NP40. Sam ples were processed for immunofluorescence micros­ copy using standard procedures. Mitochondria were visualized either by an anti-cytochrome c oxidase antibody (262) or Mitotracker Red, CMXROS (Invitrogen, the Netherlands). The ER was stained by an anti-calreticulin antibody (Sigma, the Netherlands). Images were obtained with a Bio-Rad MRC1024 confocal laser-scanning microscope equipped with an argon/ krypton laser, using a 60X 1.4 NA oil objective and LaserSharp2000 a c­ quisition software. Myotubes were not fixed but stained with Mitotracker Red, CMXH2ROS (Invitrogen, the Netherlands), and while they were placed at 37°C on a temperature-controlled stage of a Zeiss LSM510-Meta confocal micro­ scope (running software release 3.2), images were acquired using the appropriate argon laser lines and a 63X 1.4 NA oil objective.

• RNA isolation and RT-PCR RNA w as isolated using the Aurum Total RNA Mini Kit (Bio-Rad, Hercules, CA, USA), according to the manufacturer's protocol. Semi-quantitative RT-PCR analysis was performed. Approximately 0.5 pg RNA served as template for cDNA synthesis with random hexamers using the SuperScript first-strand synthesis system (Invitrogen) in a total volume of 20 pl. One pl of the cDNA reaction mixture was used in PCR analysis according to standard procedures. In RT- control experiments, reverse transcriptase was omitted. Cycle number was 24 for ß-actin and 26 for DMPK. The following primer pairs were used: human/mouse/rat beta-actin forward primer 5'-GCTAYGAGCTGCCTGACGG-3' and reverse primer 5'-GAGGCCAG- GATRGAGCC-3'; human DMPK exon 7 forward primer 5'-ACGGCGGAGA- CCTATGGCAA-3' and exon 9 reverse primer 5'-TCCCGAATGTCCGACAGTGT-3'; Mouse/rat DMPK exon 7 forward primer 5'-AAGAATTCGCCGAGACATATGC- CAAGATT-3', exon 9 reverse primer 5'-AATCTCGAGTCCTGCATGTCTGACAGC- GT-3', exon 12 forward primer 5'-AAAGAATTCGAGACCTGGAGGCGCATGT-3' and exon 15 reverse primer 5'-AAACTCGAGCGAACAGGAGCAGGCAAC-3'. PCR products were analyzed on 1.5-2.5% agarose gels and stained by ethidium bromide. Signals were imaged on the Epi Chemi II Darkroom (UVP BioImaging systems, Cambridge, United Kingdom), and images were collected using the Labworks 4.0 software (UVP BioImaging systems, Cambridge, United Kingdom). 2 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

37 r 3 8 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages • RESULTS sue samples from the DMPK knockout (KO) mouse served as negative negative as served mouse (KO) Tis­ products. knockout both of DMPK the composed from are samples samples TG sue in obtained signals so 211). (126, type tissue the on depending mDMPK, smooth in predominantly occur F) and (E isoforms short encoding scripts tissues muscle in protein • DMPK r srnt i hat daham n tnu, n - nniiae - rela­ - unanticipated - and tongue, and diaphragm heart, in strength ary iey ek n kltl uce Fg 1). (Fig. muscle skeletal in weak tively the loaded, protein total of basis the on compared When (126). controls separated, transgene be not DMPK could protein human DMPK The the mouse on weight. endogenous based the and molecular product immunoblot their in on isoforms difference DMPK clear short and long endogenous todistinguish compared higher multiple eight-fold from to promoter two is hDMPK hDMPK of own and its Production by copies mice. driven gene is (WT) mice type TG in wild in the of protein that to gene hD- next the lineage of mDMPK mouse expression (TG) endogenous Tg26-hDMPK examined our in we protein, transgene signals MPK reliable low-abundance obtain to a and DMPK, presence purposes for for types comparative muscle For distributed protein. mouse are DMPK of different products analyzed protein we the correspondingly, whether verify To 211). tran­ (95, and muscle, muscle skeletal and heart in expressed mainly are A-D) forms ale wr etbihd ht RA ecdn ln DP ioom (iso­ isoforms DMPK long encoding mRNAs that established work Earlier MK inl a cery togs i badr n soah o intermedi­ of stomach, and bladder in strongest clearly was signal DMPK sn or w DP-pcfc nioy 7 (5, e ee be to able were we (95), B79 antibody DMPK-specific own our Using A 7 k-S psSort tMPK~7S kD-rS used to monitor proteir controls. GPS = Gastroc­were included as negative Samples from a KO mouse and analyzed on western were resolved by SDS-PAGE of WT (A) or TG mice typ (B)es. muscle smooth and loading. --6f7 = -ubulin staining ikoformsD-F (~6 7 (DkO (95, parentmole cp(a e w r DMPKightst isoformsA- muscleD (ap- complex; L =nemius long Plantaris Soleus blot using DMPK a antibody. Proteins in tissue cardiac lysates skeletal, in ex­ pression protein DMPK • 1 Figure In WT and TG ventricle and atrial segments of heart, long and short DMPK isoforms were expressed in almost equal amounts (Fig. 1A and B). Total DMPK protein expression, i.e. transgenic and endogenous level combined, in TG heart was considerably higher than in WT heart. In skeletal muscles, in gastrocnemius, fast twitch tibialis anterior, and also in slow twitch soleus muscle of both WT and TG mice, long DMPK isoforms predominated. Weak expression of short DMPK isoforms of about 10-40% of total signal intensity was observed as well, at vari­ able long:short signal ratios for different muscle types in WT and TG (compare e.g., tibialis and soleus). When compared to the tubulin control also total DMPK signal levels varied, with the highest expression in WT found in soleus and in tibialis anterior in TG. In diaphragm of WT mice, a mixed fiber type muscle, expression of long DMPK isoforms predominated. In diaphragm of TG mice, absolute DMPK expression was considerably increased, and the signal level of long and short isoforms appeared about equal (Fig. 1A and B). In tongue, also a mixed fiber muscle, short and long isoforms were present at equal levels in both TG and WT mice (Fig. 1A and B). In tongue of WT mice, also a faster migrating protein immunoreactive to our anti-DMPK B79 antibody was noted (Fig. 1A). By far the highest expression of DMPK, almost exclusively of the short isof or m ,w as found in bladder and stomach of WT and TG mice, both tissues with a high content of smooth muscle cells (Fig. 1A and B).

Figure 2 • DMPK protein A expression in brain and individual brain regions. Lysates of brain regions anc the pituitary gLanC from one- cP C? 'Ç'’ Ô' year-old TG mice (A) were L analyzed by western blotting *S using a DMPK antibody (B79).Protern Loading was verified by pyruvate kinase PK (PK) immunoreactivity. Whole brain extracts from one-year-old KO mice (BKO) and TG mice (BTG) served B as controls. (B) Immunopre- cipitated DMPK from whole BKO BWT BTG brains of KO, WT and TG 0 21 42 0 21 42 21 42 mice isolated at postnatal —• days 0, 21 and 42 detected short exp) by the MANDM1 DMPK an­ tibody. To better appreciate long exp) signal intensities, a long and a short exposure are shown. L = long DMPK isoforms and S = short DMPK isoforms (95, 296).

• DM PK protein in brain anO eistinct tihain regions Northernblot, RT-i CRand ISH ptrBios eemnnstrateb tOmt RNA products fnoro tie it SPKgnneare ex nnessed at )ow lenelin th e central nervous byntnm.(95, 2n4) S inga dhn htdnc|sooun D M P ° groteio (evel in total WT brain lysate was also very low and difficult to detect on western blot (95, 211), and data not shown we first concentrated on the analysis of 2 • DMPK protein isoforms are differentially exx44sed in mysgenic and neural cell lineages cell neural and mysgenic in exx44sed differentially are isoforms protein DMPK • 2

39 40 2 • DMPK protein isoforms am differentially expressed in myogenic and neural cell lineages nioy MNM (5, 1) I areet ih rvos idns ex- findings, previous with agreement In 218). (159, MANDM1 antibody, mice TG and WT KO, of brains whole from originating protein of amounts contained gland shown). not pituitary (data the isoforms that DMPK long confirmed only mice almost WT of Analysis 2A). mice. overexpressor in TG samples dn mc(2°e br g h Hr - An lll (9) n alwd to allowed and (197) c ol -t lele H-ry sA5n rbori the ng mica(12u°les n0 od a myotu bes ant meoblayts primory protei nincultured •DMPK were isoforms DMPK Long brain. TG for obtained were signals strong pressio pressio antibody DMPK by immunoprecipitation to 42 and 21 0, days postnatal of (Fig. observed was isoforms long of expression predominant a gland- M C a yopony¡n s f oro uM ioom appeared isoforms uMM) however, o observe, r id lo d r o y .W b ^ e of lb us u t r a m d Strik­ ase expressed. yaosphornlyr¡cn uMoC r4yfubtn uac were ym tostm isotoums snM differentiated ctrng DMoU cr as ell lorg w s a e ire u yt m a W ) ^ SA m g (Fie. m ptoUfeuati mhotubuo rltrs.ln in e eliHetrctiat during strength signal 2B). in (Fig. increase development of clear weeks a six first the was there and predominant, B79, followed by detection on western blot with a second DMPK-specific DMPK-specific second a with blot western on detection by followed equal B79, subjecting by profiles DMPK followed we development, brain ing brain regional from lysates total examining by distribution protein DMPK tem , n o a p ta re e C Tiflerenru ¡4 exerrn smn Imed o u s observed between between observed s u o Imed smn exerrn ¡4 e C Tiflerenru re ta p a o n , tem WT of complexes muscle ly a c e h t eom f d e isolat e er w te s la b o m Ptimery o noe mr sniie sa fr oioig sfr sit dur­ shifts isoform monitoring for assay sensitive more a invoke To In all areas examined -most prominent in olfactory bulb and pituitary pituitary and bulb olfactory in prominent -most examined areas all In nofDMPKw A s e b u t o y m s t s n l b o y m B O K T W O K T W O K T W O K T W O K T W O K T W O K T W W-#W—* M — w —

s e b u t o y m s t s a l b o y m 3 5 7 9 7 5 3 1 C P

b RT- T R b 1 C 11 rather but brain, WT in detectable barely still as 4I 2 I151 12 ■4SI 1 15 112 I14Í+/-4 nt)l 13 1 4 115 114 13intron 1+ 113 12 * 4 S 4 L 4 T The asterisks indicate rare a The main alternative splice formeh DMRaAeegae0from weight of 63 kDa. with predicted a molecular encoding protein a isoform resulta of incorporation of are indicated on theor right. short (S) DMPK isoforms and myotubes after 1confluent and myoblast culture differentiation medium. L = (see K0 samplerO tPefrster oith cross-reactingproteins expression (T). Note that was assessed by westerr and 9 days of differentiation expression in proliferating allowed to differentiate into cells. part of intron 13, potentially premature stop codonDMPK isoform as containing a products encoding long (L) 7 days of differentiation. emlifereting eoSa^te,) aT-PSSaKalyais par- a wa 2mSa^)aprn i-puptfitahee soforms^e i = ^ ahort M [ lane [JMI^issfonms S aae ic proliferpfihn, maS ^i^tomf o t ^ ^ a D u migretmg e c J e r ^ 3nep S tDhtlo ppr7 e s o a t bed DMPK isoforms comigrated ing was related to tubulinblotting. Protein lacda myotubes after 1a C,myoblast 1, 7 culture (C)myoblasts a.3d (P), a confluent myotubes. (A) DMPK protein mice were cultured ancmuobla sts fromConditionally WT and KO immortalizec myogenic primary in ex­ pression protein DMPK • 3 Figure slightly increased in myotubes at later time points of differentiation (7-9 days), as illustrated by the more intense upper band of the doublet signal (296). Expression of short DMPK isoforms was low in differentiating myo- tubes and became almost undetectable after nine days of differentiation (Fig. 3A). Presence of short isoforms in myoblasts could not be deter­ mined due to a strong cross-reacting signal also observed in KO cells (probably bovine serum albumin since myotubes cultured in the presence of horse serum did not show a signal; Fig. 3A). To shed more light on the repertoire of DMPK isoforms in myoblasts, we also performed an RT-PCR analysis of the alternatively spliced exon 12-15 region (Fig. 3B). This analysis confirmed that transcripts encoding the short isoforms were present in myoblasts and that expression was only just detectable in myotubes (Fig. 3B). Extrapolating this to the protein level, our data indi­ cate that short isoforms are present at low level in myoblasts, but they disappear during terminal differentiation of myofibers, leaving expression of only long DMPK isoforms.

• DMPK protein in cultured primary neurons and astrocytes DMPK expression was also analyzed in primary astrocytes from brain cortex and neurons from the hippocampus isolated from E16.5 WT and KO mice and cultured in vitro. DMPK protein was clearly detectable in cortical astrocytes but not in hippocampal neurons (Fig. 4A). Long iso­ forms were predominant, but short isoforms also occurred at low signal strength. We wondered whether the absence of a DMPK signal in neurons was due to the limited sensitivity of our western blotting procedure or to complete absence of transcription of the DMPK gene in this cell type. We therefore determined mRNA levels in the primary neurons and as­ trocytes using RT-PCR. This revealed DMPK mRNA at low signal intensity

Figure 4 • DMPK expres­ sion in primary neurons and AB astrocytes. Neurons Astrocytes Neurons Astrocytes (A) Protein lysates of primary neurons and WT KO WT KO RT- KO WT KO WT a s tro c y te s w e re a n alyz e d on DMPK DMPK western blot using antibody B79 against DMPK(95) and PK an anti-pyruvate kinase Actin antibody (PK) as a loading

control. DMPK protein c o u I c be d e te cte d in W T a stro cyte s, but not neurons. L = long in neurons and much higher intensity in primary astrocytes (Fig. 4B), DMPK isoforms (~74 kDa) corroborntmg the w estera b lot findinas.T ak en combined, these res ults and S = short DMPK iso­ suggesttnat DMPK mRSA onb proteinisgresentin primoryseuoons dut forms (~67 kDa) (95, 296). at lowlovats compared to ustrarrtns. (B) RT-PCR using RNA from primary neurons and astro­ cytes demonstrated presence • DMPK preae iain estaniiodeDeelll inesof myogeric e r nnupaI origin of D M PK tra n s c rip ts in WT To confirm Uerdngsand orev¡de(:urtOer dookgronna for bM kKoteUies in astrocytes and also neurons. The doublet signal repre­ a set hiologicaleeUir^wealse sna^zaS DMbK gratrm expression in a sents alternative splicing numoec of entabUoheh melt lin esof moogenic nr neprslor¡o¡n.l-ILm cor­ of 15 nucleotides encoding di omoocetes; bolO leating up) pon-naot¡ogea^ants(OU; d17f anCCAC12 the VSGGG sequence (95). Beta-actin served as control myoblasts expressed long as well as short DMPK isoforms (Fig. 5A). In

for RNA input. 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

41 42 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages lhuh e sue ht h srn sga o wsen lt bsd on based blot, western on signal strong the that assume we Although here expression,althoughinterpretation DMPK r a le c d e w o h )s A 2 ,N 1 3 1 gis sm ctslc n ncer akrud tiig o YFP-mDMPK for staining background nuclear and cytosolic some DMPK against astrocytes, primary In 7A). (Fig. distribution by cytosolic a caused again adapted probably staining, nuclear and cytosolic Some 7A). (Fig. this attribute We seen. also was staining instead; nuclear used and was cytosolic counterstain some mitochondrial a myotubes, in staining 26 296). (286, cells primary in isoforms DMPK of localization • Subcellular e in u n e tg n e s re p tre o n id d ls a n ig ts lo b rn te s e w K P M D e tth a th g stin e g g u s lpig f F-DP C ws lo en ee Fg 7) YPmMK E YFP-mDMPK 7A). (Fig. here seen also was C, YFP-mDMPK of clipping relatively the of release in results This protein. fusion the of clipping to the from different clearly pattern, ER distribu­ A reticular mDMPK typical myotubes, In a (286). showed 6A) tion (Fig. locations predicted the os noig F-agd esos f DP A kon o nhr n the in anchor to (known A mDMPK of versions and YFP-tagged cells encoding tors muscle skeletal primary end, this To (286). locations expected slightly migrated that signal weak very second, a revealed examination RT-PCR. this y b d e s lve s e so o ss a ).T A 5 also again signal was strong expression A N R additional cross-reactingprotein(Fig. m signal PK the M an w lo cells e confusion,D tb parebackground s PC12 ju (com d In n u signal fo s a myotubes). w KO in across-reacting observed peredby ham CCL- U373, was SK-N-SH, (PC12, tested lines cell neural the of none contrast, eiui ad F-DP C ih iohnra mre Mttakr Red Mitotracker marker mitochondrial with C and YFP-mDMPK and cytosolic reticulin free-floating, a as 6C). remains (Fig. protein which nuclear moiety, YFP resistant mDMPK of distribution cytosolic and C DMPK of distribution mitochondrial erl el fo DP K mc wr tasue wt aeoia vec­ adenoviral with transduced were mice KO DMPK their to from targeted cells were neural identified now cells host different the in pressed 5A). (Fig. DMPK represent may which slowly, more line. cell this in transcribed is gene DMPK the that indicating RT-PCR, A DMPK A, C and E localized to the ER, MOM and cytosol, respectively, respectively, cytosol, and MOM ER, the to localized with E neurons, in and seen C as A, distribution DMPK similar largely a exhibited isoforms but mitochondria, with overlapped ER localization C well-defined a in YFP-mDMPK resulted 6B). tested Fig markers ER the of none (since E ER), mDMPK C (known to anchor in the MOM) or mDMPK E (cytosolic) (cytosolic) E mDMPK or MOM) the in anchor to (known C mDMPK ER), ex­ transiently isoforms DMPK individual if investigated we Finally, closer protein, cross-reacting a to in corresponds behavior, signal strong migration its a showed cells U373 strongly surprise, B), .5 our (Fig cells protein.To N2A and DMPK PC12 for found were cts u d ro p R C P o N In primary neurons, YFP-mDMPK A colocalized with ER marker cal- cal- marker ER with colocalized A YFP-mDMPK neurons, primary In xrsin n ybat soe lclzto o DP A C n E at E and C A, DMPK of localization showed myoblasts in Expression * * * * * ■ AVofrV/> PK M D Actin tid e s en co d in g th e VSGGG VSGGG e th g in d co en s e tid cted ete d re e PK w M D ts . rip lls c e s c n tra l) tro n o c e tiv and m o u se (C 2 C 1 2 ) D M PK PK M D . ) s 2 n 1 o C lic 2 ) p 3 (C m 7 a 3 (U se u o m an m u ­ h and rim p e th n e for e tw e rs b e reflect s e s c le n p re m a iffe s d U373 12 n C2C e e tw size e b and ct s u e d c ro n P re ). iffe 5 d (9 e c n e u q e ­ s o le c u n e 15 tiv a rn of lte a g in ts n lic p se s re p re l a n ig s 12, PC m fro ec d rm rive erfo e p d A s a RN w g in on 9 n an to sp 7 i­ m RT-PCR e exon s e A tiv ) tita B n ( a u N2A q P). R C and , w 373 rro U (a lly ia c e p s e ­ s s o r c a at th by Note . d s k rke a ris m te s PK a M re D a tic ls n a e n th ig u s s A rm fo iso 296). PK , 5 M (9 D short = S rn ste e w on d e N2A) alyz n a ours), h re e w 48 for d ate 373, U , ir -SH rig -N o SK , 2 brain 1 C (P r o ) 2 1 C 2 (C (HL-1 c ia d r a c of s te a s y L ) (A or cardiac muscle, skeletal n salse cl lns of lines cell established in expression DMPK • 5 Figure n e u ro b la s to m a s . The d o u ble t t ble u o d The . s a m to N2A s la b ro and u but e n 2 1 a PC m to s in la b not lio g 3 7 3 U in i­ s o (p 12 C2C and 2A N , 3 7 3 U lines, ll e c e present th of lso a y n a is m ^ in s e d b cte tu te o e y d m O K protein g ctin a re and s rm fo o is PK M D . ression long exp = L PK M D for blot ti­ n re iffe (d h 8 4 ), g Dif tin ra life ro (p L131 CC Pro L131 CC le c s u m l ­ n ta o le (n e k s NB )), g HL-1 atin e b ), g tin a e (b B origin. brain C (Fig. 7B). Also here, some clipping of the YFP part has occurred (Fig. 7C). Overall, these trensient expression studies demonstrate authentic localization of mDMPK isoforms in all distinct cell types tested.

Figure 6 • Subcellular local­ A ization of individual DMPK ------m yoblasts------splice isoforms in myoblasts and myotubes. Splice isoforms mDMPK A, C or E fused to YFP were e x p re sse d in KO m y o b la s ts (A) and myotubes (B) using adenoviral expression vec­ tors. Confocal images were taken 20 hours (myoblasts) or 48 hours (myotubes) after transduction. To identify subcellular structures, cells were counterstained for ER marker calreticulin or mitochondrial marker Mito- Tracker Red. In myoblasts, Y F P - m D M P K A, C and E were found at the ER, mito­ ch o n d ria and in the c yto so l respectively (A). In myotubes YFP-mDMPK C, but not YFP- m D M P K A and E, lo ca liz e d at mitochondria (B). Bars ^ B A: 10 pm. B a rs in B: 25 pm. ------myotubes (C) Western blotting was used to ve rify in te g rity of expressed fusion proteins. Recombinant proteins were visualized using a DMPK and a YFP antibody. Molecular weights of marker proteins are indicated. Asterisks indi­ cate C-terminally truncated mDMPK products as a result of non-enzymatic breakdown (2 8 6 ).

D M P K Y F P 2 • DMPK protein iso forms are ifferentially d ex pressed in myogenic and neural cell lineages

43 44 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages A ------neurons to th e ER, m ito c h o n d ria and and ria d n o h c ito m ER, e th to gnal t 04 kDa. P F 30-40 Y by at d ls te a n g tra ig s pin s illu clip , d rre protein ccu o e m so C and head w rro a An . teins d ro p icate d r in e rk a r m re a la u c of le o M ts and h ig e w PK M D antibody. a P F Y ere w sing a u s te y d c e liz tro a s a isu v in d sse re p x e teins. pro n sio rity fu teg in d sse re rify p e v ex to of used s a w Bars: . g in and stain r a le c lic u o n s to y c d e w o h s o ls a - FP Y xidase. o c e arker m m ro h c l to ria y c d n o h c ito m or for d e in lls e ta c rs te n u s, re co ctu re stru e w r identify la llu To e c . n b u tio s c u d s n tra r cal fte a nfo o C rs. using cto ve ) B ( l ira s v te o y n c e d tro a s a eurons n and KO ) (A in d sse re A, p x e K P M D m s rm fo o is e lic types. p S cell primary brain- derived in isoforms splice n k-teaed asr yes. cyte stro a also d ate tre - ck antibody, o m P F in Y the and ith w PK M d g D cte tin te c e a d re - s s tein cro pro a s te a ic d in teins ro p t n a in b m o c e R s te y c tro s a y an and m C, s n ro . K eu P ly n e M tiv D c m e - p s P F re Y l, For so cyto d e e th liz ca in lo E and C A, K P M D m lin u tic lre a c r e rk a m R E hours 20 n e tak re e w s e g a im re e w P F Y to sed fu E or C DMPK individual of local­ ization Subcellular • 7 Figure In th e c a s e of Y F P - m D M P K A A K P M D m - P F Y of e s a c e th In 10 pm. (C ) W e s te rn b lotting lotting b rn te s e W ) (C pm. 10 • DISCUSSION

Muscle and brain are prominently implicated in DM1. FISH analyses have demonstrated presence of DMPK transcripts in these and other tissues, however the occurrence of splice products was not addressed (126, 244). Here, we analyzed the distribution of long and short DMPK protein iso­ forms (95, 296) in muscles of different fiber type and in discrete regions and cell types of the brains of mice. Since isoform-specific antibodies are not available for DMPK, we have used classification of signals in dif­ ferent molecular weight categories to profile expression. Human DMPK- overexpressing transgenic (TG) mice aided in the detection of DMPK in tissues with low expression, nonetheless we realize that cautious interpretation is needed. Effects of transgene integration site, species differences in promoter specificity and post-translational control may af­ fect expression level of different human and mouse DMPK gene products.

• DMPK protein in muscle tissues and cultured muscle cells All muscle types investigated expressed long and short DMPK isoforms, albeit at different levels for different muscle types from WT and TG origin. Cardiac involvement in DM1 concerns conduction defects leading to arrhythmia and heart block and has been correlated with (CTG^CAG) n expansion or DMPK gene product dosage (16, 101, 211). Studies in transgenic mice demonstrated that DMPK is involved in ion homeostasis in heart, influencing Ca2+, Na+ and Cl- currents (16, 134, 211). Combined with our current findings that long and short DMPK isoforms are highly expressed in both ventricle and atrium, this leaves open the possibility that both toxicity of (CUG)n-expanded mRNAs and mutational effects on DMPK protein level or distribution contribute to DM pathology. As we find long and short DMPK faithfully expressed in HL-1 cardiomyocytes, we consider these cells a reliable experimental model for further studies into the pathophysiological significance of DMPK mRNA and protein. Smooth muscle involvement in DM1 is typically noted by malfunc­ tioning of the gastrointestinal tract and vacuolization in the bladder of patients (101). Smooth muscle abnormalities have not been reported in KO mice, but overexpression of hDMPK affects arterial smooth muscle tone (211). As we report high expression of virtually only short DMPK iso­ forms in both stomach and bladder here, we propose that further studies should now be focused primarily on how these particular isoforms con­ trol cell integrity globally throughout all smooth muscle cell-containing tissues in DM1 patients and normal controls. Weakness and wasting of distal skeletal muscles and myotonia are hallm arks of DM1 (101), with type 1 (slow twitch) fibers being most strongly affected (9, 289). We show here that long and short DMPK iso- 2 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

45 46 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages There is presently no other work to which our results can be compared, compared, be can results our which to work other no presently is There (101, 193). Our results from TG brain suggest that both long and short short and long both that suggest brain TG from results Our 193). (101, mice, long DMPK isoforms dominated. A possible explanation for this fact fact this for explanation possible A dominated. isoforms DMPK long mice, cells brain-derived and brain in expression • DMPK study to myotubes KO in 296) (286, A-D isoforms DMPK individual with long that think now we (126), somites of region myotome the in already Myoblasts vitro. in differentiated myotubes in abundance low very at n mr eaoae tde ae hrfr necessary. therefore are studies elaborate more and DMPK neurons. adult in expression DMPK of presence or absence about un­ observation This age. of weeks 3 were animals until brain whole diaphragm and tongue. These tissues are of mixed fiber-type and are are and in fiber-type mixed expressed of are mix­ strongly A also tissues 304). was These (229, tongue. isoforms and respectively DMPK diaphragm muscle, short muscles, and long mixed gastrocnemius of muscles and and ture fast tibialis, skeletal in slow, soleus, mostly typical including dominate types, fiber isoforms all of long but coexist, forms eie fo E65 mro, n/r h dfeet ernl el ie ana­ lines neurons cell neuronal hippocampal different the primary and/or our in embryos, E16.5 from expressed derived yet not be either could place takes mainly that brain process a develop­ terminal the astrocytes, that during speculate to DMPK-expressing expression of tempting is itment DMPK and in relevant, is increase development an that at derscores DMPK expressed lineage, progenitor cells tumor astrocytic cells, the glioblastoma from U373 originating Also cortex. primary the from (145). cytes tube neural the from originate not does TG it and that in WT regions, both of gland pituitary the In striatum. and bulb olfactory hypersomnia and disturbances emotional and cognitive retardation, tal isolation. in role experiments their transfection complementation of using are we significance Currently, biological ment. the of study for models our that attractive and demonstrate faithful thus findings Our aging. and development muscle exist to before reported were iso­ products DMPK gene long of DMPK As expression of significantly. affected forms process myotubes the to nor fate fast- athis maturation and adopt neither but fusion and (158), imprint phenotype commitment precursor their twitch lose usually vitro in cultured s ht iutr tsu (.. te neir oe dfes rm te brain other from differs lobe) anterior the (e.g., tissue pituitary that is men - includes CNS the of involvement DM1, adult-onset and congenital In of trajectory entire the in stages all at expressed remain isoforms DMPK occurred isoforms DMPK short vivo, in muscles mature in isoforms DMPK da ht eetidcd islcn o cran rncit i nuos is neurons in transcripts certain of missplicing repeat-induced that idea mlctd n eprtr dsrs o sec ad wloig rbes in problems swallowing and (101). speech or patients DM1 distress respiratory in implicated lyzed may all have reverted to an non-expressing embryonic phenotype. phenotype. embryonic non-expressing an to reverted have all may lyzed statements firm make tissue- however, overall cannot, the we to point mostly this At contributes increase. 1-6, level weeks postnatal between in protein DMPK observe not did we that here note of is It level. low stem, brain including regions, all across expressed are isoforms DMPK develop­ muscle skeletal during isoforms, long the represent line particular in cell DMPK, C2C12 the or myoblasts-myotubes WT immortalized n motn fnig s h ietfcto o DP poen n astro­ in protein DMPK of identification the is finding important An urnl, ot ok n ri ivleet n M i bsd n the on based is DM1 in involvement brain on work most Currently, In contrast to the persistence of a significant percentage of short short of percentage significant a of persistence the to contrast In the dominant event and is the basis of the neuropsychiatric problems in DM1 (129, 253). DMPK transcripts aggregated in RNP foci were revealed in neurons in post-mortem DM1 tissue (129), but cell-type differences in accumulation and aggregation behavior of RNAs may contribute to dif­ ferential visibility in neurons and astrocytes. Given our results, we sug­ gest that study of the fate and effects of normal and mutant DMPK gene products in astrocytes now also merits serious attention. As a prelude to such studies, we compared the distribution fate of mDMPK isoforms in transiently transduced primary brain cells to that in muscle cells and to already known behavior in other cell types (286). In all cell types analyzed, cytosolic or membrane localization was exactly as anticipated for DMPK E or DMPK A/C isoforms, respectively (286). Hence, our observations indicate that location of DMPK isoforms does not rely on cell-type dependent rearrangements in membrane composition (25) or cytoarchitecture. Based on our findings regarding the inter- and intracellular distribution patterning of DMPK isoforms, more detailed studies into the normal and pathobiological role of DMPK should now become feasible. 2 2 • DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages

47 r 'i CHAPTER 3 L . J

TOPOLOGY AND SPECIFICITY OF MEM BRANE INSERTION OF TAIL- ANCHORED DMPK ISOFORMS

Ralph J. A. Oude Ophuis, Lisette Hetterschijt, Daan de Gouw, Mietske Wijers, Jack A. M. Fransen, Bé Wieringa, and Derick G. Wansink

Department of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Submitted for publication 50 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms alacoe (A poen ae ebae rtis hrceie b a by characterized proteins membrane are proteins (TA) Tail-anchored • ABSTRACT a distinctly different C-terminus and require presence of a coiled-coil coiled-coil a of presence require and C-terminus different distinctly a its whereas (ER), A reticulum DMPK Mouse endoplasmic the splicing. with tail its alternative by via formed We associates genes understood. DMPK poorly human is and domain tail the of composition and sequence and HA double-tagged fusion proteins revealed that DMPK A isoforms isoforms A DMPK that revealed YFP proteins with fusion studies topology double-tagged Finally, HA protein. and translated newly of pool solic topol­ and specificity How C-terminus. their in region hydrophobic single hs eemn te oain oinain ad rsmby oe o dis­ of role- interface. ER-MOM presumably the at -and (Nout-Cout isoforms orientation DMPK cytosol location, tail-anchors tinct the the different of face determine C compositions thus DMPK sequence of unique ends The both orientation). whereas orientation), ae hi Ctriu oine twrs h ognle ue ( -C (N lumen organelle the towards oriented C-terminus their have isoforms, DMPK the between differs the that diffusion indicated membrane carry analysis lateral of isoforms FRAP MOM.extent C the in DMPK insertion proper for mouse domain and influ Human - A. moderately DMPK only tail. of region entire organ­ the targeting hydrophobic of that the ence and hydrophobicity flanking and binding residues length to Charged membrane related determines is in tail the specificity in difference elle species-specific (hArg600-mAla602) this position (MOM). non-conserved mitochondria one of membrane outer the to binds orthologue human mouse the of products protein-isoform of behaviour acid anchoring on amino by report determined are proteins TA of insertion membrane of ogy u ta nn o te nhrd MK qiky xhne wt te cyto­ the with exchanges quickly DMPKs anchored the of none that but at acid amino the of nature the that demonstrated analysis Mutational ■3 u ir out ' • INTRODUCTION

Tail-anchored (TA) proteins are a structurally and functionally diverse class of proteins which associate with the phospholipid bilayer of organ­ elles like endoplasmic reticulum (ER), mitochondria, peroxisomes and chloroplasts. For this membrane association, TA proteins use a stretch of hydrophobic amino acids situated close to their C-terminus (19), while their N-terminal part faces the cytosol. TA proteins are involved in a wide range of cellular processes, including apoptosis (228), protein transloca­ tion (18), vesicle fusion (155) and mitochondrial morphology regulation (234, 31 1). Membrane targeting of TA proteins occurs posttranslationally, after their hydrophobic membrane region emerges from the ribosome upon termination of translation (19). When newly synthesized TA proteins use this mechanism to insert into the ER, they may become redistributed via vesicular transport to other destinations within the secretory path­ way (18, 66, 203). Lipid composition of the membrane of residence is then an important parameter for retention in specific cellular organelles (25, 138). Targeting of TA proteins to the mitochondrial outer membrane (MOM) or membranes of chloroplasts and peroxisomes occurs directly from the cytosol (18). Generally, membrane anchoring depends on ATP, and requires chaperone involvement. Chaperones that serve TA proteins as client-proteins are Asna-1 (265) and Hcs70-Hsp40 (2). Also the sig­ nal recognition particle may be involved in tail anchoring, although its precise role is not clear (1). The tail region of TA proteins is thought to contain all necessary in­ formation for membrane specificity (18). MOM-targeted TA proteins usu­ ally contain a membrane anchor of at most 20 hydrophobic residues flanked by basic -i.e. positively charged- amino acids. With relatively few exceptions, ER TA proteins have a longer hydrophobic region and do not require specific basic amino acids. Precisely how these physical and chemical characteristics determine the selectivity of membrane associa­ tion remains unclear The mode of association has only been studied for a few TA proteins, mainly with focus on proteins that localize to the ER (18). Cytochrome b5, for example, contains an anchor that spans the membrane, such that its C-terminus resides in the ER lumen (25, 154). Even less is known about the transmembrane topology of MOM-targeted TA proteins (3, 138, 292). Two recently identified TA proteins are isoform products of the myo­ tonic dystrophy protein kinase (DMPK) gene (286), which causes myoton­ ic dystrophy type 1 (DM1) when the (CTG)n repeat in its 3' untranslated region expands beyond a certain threshold (26, 176). As the gene is only expressed in mammals, DMPK isoforms can be considered evolutionary 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

51 52 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms A n C eit ht ar dfeet ye o etne C emn, both termini, C extended of types different carry that exist C) and (A serine-threonine kinases, including myotonic dystrophy kinase-related kinase-related dystrophy myotonic including kinases, serine-threonine determine differential modes of membrane anchoring of DMPK. Using Using DMPK. of anchoring membrane of modes differential determine evn a mmrn bnig oan. neetnl, rael specific­ organelle Interestingly, domains. binding membrane as serving fet n ebae behaviour. membrane determine on composition, regions effect tail acid across amino distribution how report hydrophobicity and we length tail analyses, truncation-mutation f raie sot E (SR srcue o prncer lseig of clustering perinuclear or formation cause structures 286). can (216, cells, (OSER) mitochondria ER transfected smooth in whereas vitro, organized ER, inof the A with DMPK of to target associates pression both A) C DMPK (mDMPK A human DMPK and mouse mouse MOM, the While species-specific. even isoforms DMPK long two splicing, alternative to and Due ß, (33). a, of kinase citron (MRCKs) subfamily a of kinases members Cdc42-binding archetype the are They (286). proteins new MKs fiiy o mmrn tp, t isrin oooy a wl a its as well as topology, insertion its type, membrane for affinity DMPK's t otoou i hmn (DP A lclzs o h MM Etpc ex­ Ectopic MOM. the to localizes A) (hDMPK is humans in and orthologue types its DMPK these between differs anchoring membrane of ity Here, we study the structural requirements of the tail-domains that that tail-domains the of requirements structural the study we Here, y , h kns a n ß and ß, and a kinase Rho • MATERIALS AND METHODS

• Cell culture and DNA transfection C2C12 myoblasts and N2A cells were grown to subconfluency in Dulbec- co's minimal essential medium supplemented with 10% (v/v) foetal calf serum and maintained at 37oC under a 5% CO2 atmosphere. Cells were transiently transfected with various expression plasmids in presence of Lipofectamine (protocol as specified by the manufacturer, Invitrogen, the Netherlands and maintained in culture for an additional 20 hours prior to analysis.

• DNA vectors EYFP-DMPK expression vectors and mutation and deletion constructs were obtained by cloning appropriate PCR fragments into p EYFP-C1 (Clontech; see Supplementary Experimental for details). Next to the ORFs for different mouse and human DMPK A or C isoforms, expression plas­ mids also contained the corresponding 3'-UTR regions of mouse or hu­ man DMPK mRNAs. Plasmid inserts were verified by DNA sequencing. For visualization of the ER, we used vector pDsRed2-ER, encoding Discosoma sp. red fluorescent protein with a calreticulin-derived ER targeting sig­ nal at its N-terminus and a KDEL ER retention signal at its C-terminus (Clontech) (191).

• Immunofluorescence analysis of intracellular protein distribution C2C12 myoblasts and N2A cells were grown and transfected on glass cover slips, fixed in PBS containing 2% (w/v) formaldehyde and per- meabilized in PBS containing 0.5% (w/v) Nonidet P-40. Sam ples were processed for immunofluorescence microscopy using standard proce­ dures. A rabbit anti-cytochrome c oxidase antibody was used to visualize mitochondria (262). Subcellular localization of YFP-DMPK fusion proteins was scored independently by two examiners blinded to sample identity, whereby location was categorized as either (i) cytosolic, (ii) ER (with or without OSER structures), (iii) mitochondrial (with or without perinuclear clustering of mitochondria).

• Immunofluorescence analysis of membrane topology by selective mem­ brane permeabilization Insertion topology was investigated by selective permeabilization of the plasma membrane which was done as follows: C2C12 myoblasts and N2A cells grown on cover slips were incubated with 3 pg/ml (C2C12) or 6 pg/ml (N2A) digitonin (Sigma) in cytomix buffer (120 mM KCl, 0.15 mM CaCl2, 25 mM Hepes/KOH pH 7.6, 2 mM EGTA and 5 mM MgCl2) containing 0.3 M sucrose (Merck) at 4oC for 15 minutes (179). To permeabilize all 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

53 54 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms rnfce CC2 n NA el wr lsd n P4 lssufr (50 lysisbuffer NP-40 in lysed were cells N2A and C2C12 Transfected Rce ad m P F fr mnts t o. rd lsts ee cen­ were lysates 4oC. Crude at minutes 5 for SF) PM mM 1 and (Roche) mn ais DP poen eune fo vros pce wr col­ were species various from sequences protein DMPK acids. amino • Bioinformatics AR). X-OMAT an­ secondary peroxidase-conjugated transferred and horseradish by followed electrophoresis antibody, gel SDS-polyacrylamide by separated SSplarlmd gl lcrpoei ad etr blotting western and electrophoresis gel • SDS-polyacrylamide microscopy Immunoelectron • mobile and values T1/2 and 4 Prism GraphPad using fitted were Curves analysis photobleaching after recovery • Fluorescence Laser- and a objective software. oil NA 1.4 with acquisition 60x a obtained using Sharp2000 laser, were argon/krypton Images an with antibody. secondary fluorescent a with 3.5 pm region was photobleached at full laser power and fluorescence fluorescence and x power pm 3.5 laser A full objective. at oil NA 1.4 63x photobleached a was and region lines pm 3.5 laser argon nm 488 a (Stressgen) isomerase disulfide anti-protein oxidase, 12CA5), c (anti-HA; calf temperature room foetal at anti-cytochrome anti-hemagglutinin hour 1 (v/v) antibodies: for 10% following incubated the of in one were with they temperature Then room PBS. in at hour serum 1 for incubated and bicity using the GES hydrophobicity scale (70) and a window of seven seven of window a and (70) scale hydrophobicity GES the using bicity (Kodak film to exposure was and Immunodetection chemiluminescence Laboratories). enhanced by performed ImmunoResearch anti-GFP an using (Jackson tibodies immunodetection for (Millipore) collected, membranes were PVDF to 4oC. Supernatants at so­ 14.000g at mM cocktail 1 minutes inhibitor 10 NaF, for protease mM trifuged x 25 1 NP-40, vanadate, (w/v) mM 0.1 1% NaCl, pyrophosphate, dium mM 150 Tris-HCl, mM aadhd ad rcse fr muood eeto ad lcrn mi­ electron and detection (286). immunogold described for as croscopy processed and taraldehyde frame). per s (1 series time determined. a fractions as recorded (1%). power was laser low at recovery cells whole Fluorescence scanning by monitored was recovery Zeiss a of ­ se stage calf foetal (v/v) temperature-controlled a Dulbecco's 10% on 37oC at with red-free placed phenol in supplemented while rum imaged medium were essential Cells minimal wells). (Wilco dishes hour 1 for incubated were Sys­ cells (R&D PBS with (HtrA2/Omi) wash protein-2 minutes 30 a After requirement tems). temperature anti-high or (Serva) X-100 Triton (w/v) 0.3% with incubated were cells membranes, etd s ecie i Splmnay Experimental. Supplementary in described as lected hydropho- average an as residue each for calculated was Hydrophobicity gl, H .) otiig % wv fradhd ad .1 (/) glu- (w/v) mM 2 0.01% and and EGTA, mM 10 formaldehyde (w/v) HEPES, hour 1 mM 1% for 25 fixed containing were PIPES, 6.9) pH mM MgCl2, (60 buffer pEYFP-hDMPK(R600A) PHEM with in transfected cells N2A glass-bottom on grown were myoblasts C2C12 transfected DMPK-YFP equipped microscope laser-scanning confocal MRC1024 (Biorad) BioRad S50Mt cnoa mcocp (unn sfwr rlae .) using 3.2) release software (running microscope confocal LSM510-Meta PBS with washed were cells 4oC. Subsequently, at minutes 25 for PBS in • Statistics All values are presented as mean ± SEM. Statistical significance of T1/2 and mobile fractions in FRAP analysis was assessed by a student's t- test, taking YFP-mDMPK E as a reference. A Chi-square test was used to compare localization of point mutants examined in Fig. 3. 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

55 ► 56 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms Ctria. yrfIr^s 's¡n H)f r^r^iont r r (i^5)) x ir ra .r^^^rr^titoun^t; )if (H ¡on i'vs eyeraf^Irrt^is C-terminal. A are domain coiled-coil the and anchor tail its in region hydrophobic A • RESULTS atnal se f ot l ci )^5 ( g 1) Fl^ä^t YFS^-^li fc^i‘ Fclb^fäi^^tS S)Z^r5W S; ll 1A). e c ig. ^ lDM (F dost P hf tsne l a n t na cdara mro nleHeF i xes. c asn ewe ttie gelwsen nanson ecm axiease. si nyloeaHoesF marnot doadríal oc F m eurd o mtcodil agtn o hmn MK C. DMPK human of targeting mitochondrial for required necss seay fuo thin anshorm g.Fisse g.Fisse anshorm thin fuo seay necss myoblasts, ^2Cie ( in ísed m exv p^re^^ins was fusion P F Y s a ed express oazd l)h M^ a c id s ohc ffS ID1. - B I . ^ 1E-G), F ( dit]. ( f2f|S s lonaUeation e ^ ^ e c othec csr tiFd n e ck h .an ^M M hs ) ul lonaUzsd Memtirane a nchoring Di' hDMPK isoforms andtruncated derivatives derivatives andtruncated isoforms hDMPK Di' nchoring a Memtirane 7 53 R 630 HR 583 577 YFP A YFP P YFP- hD M PK C C PK M hD YFP- B « r r y a l g l s g ■h'. R tpanspqsgaaqeppalpe p l a p p e q a a g s q p s n a p pt w w g l a a w p p p v l c f l p s c s c p f ___ h FU > ■ Ó. ■ ‘>C on^o ls o elslztd wt fCe witd eolusslizbtidn o) somnd^tox lesr t merge mto ' Y FP-W W W K]=^ YFP-hDMPK YFP-hDMPK K]=^ W W FP-W Y % YFP ^ F p ------heM^aC r e s 8 lt e d ic a cyrohelic cyrohelic a ic d e lt 8 s e r heM^aC F-DP -3-3) C ) C-534-630) YFP-hDMPK F-DP O'53 C CO1'YFP-hDMPK 583) F-DP M CYFP-hDMPK 1 D | p 1 / m Á

ci421' 630) "y P M w as as u

h sbeLLr LocaLiza­ subceLLuLar The th e C -term inaL taiL resuLted resuLted taiL inaL -term C e th the of IncLusion ) M - (K tosoL. using d e in rm te e d s a w tion l­ ca o L and subceLLuLar ins te ir ro e p th n sio fu P F Y targeting. coiLed coiL d o m ain b e sid e s s e sid e b ain m o d coiL ­ y c coiLed the in ­ n ed o h c resid ito m but to t dria, e rg C(5 ta 34-630)) K not P M D h did - P F Y t n ta u (m K P M D h - P F Y utant M ) G K - P M (E D h - P F Y ) D e - m (B ro ch yto ase. c xid o r ­ e rk c ito a m m for riaL d n d o e ch in sta - o c re e ins w te pro n sio fu of after n y p ssio o re c p s x e ro ic m caL fo n co C-reLated of K P M n D h tio ta n t n se re re p iffe re d tic a m e h c of s e c n e u q se ry a rim P ) (A mitochondrial for domain coil region hydrophobic the quires as 1 pm. 10 LocaLization. Bars, riaL d n o ch ito m in aLone 2 taiL C J) - H PK ( M D cytosoL. an m u H the in HR, Located the Lacks C(1 h ic ria. -583), d h n w o ch ito m to LocaLized C CeLLs yobLasts. m 12 C2C in C=cytosoL. ), ce n e ria, u d q n o se ch M = ito m (boxed ic b region o h p ro = HR yd h ization. and s u in rm te - C C's K P M D h coiled plus C-terminus the in re­ C DMPK Human • 1 Figure localization of YFP fusion proteins with only the HR sequence in the tail segment (mutant YFP-hDMPK C(534-630), Fig. 1H-J) or a tail segment plus coiled coil domain immediately upstream (mutant YFP-hDMPK C(421-630), Fig 1K-M) revealed that both elements are necessary for proper mito­ chondrial targeting. Of note, we had difficulties in defining the precise location of the YFP-hDMPK C(534-630) mutant because in cells transfected with the vector for this mutant also some free YFP tag was produced (Supplementary Fig. S1). Probably, the YFP-hDMPK C(534-630) product is more susceptible to proteolytic clipping, explaining also the nuclear staining in some cells.

Figure 2 • A 12-amino- 577 587 HR1 600 HR2 619 629 acid hydrophobic region in parvprpglsEalslllfavvlsTrAaalgciglvahagoltavw Rrpgaarap its C-terminal anchor is 1------1 I------1 responsible for association of hDMPK A with mitochondria. Y F P ^ ^ » YFP-hDMPK A(534-629) M/CM (A) Primary sequence of

hDMPK A's C-terminus and Y F P * ^ ------WMAMKW YFP-hDMPK AC1'600) M/CM

schematic representation of Y F P * YFP-hDMPK A(600-629) C/N different hDMPK A-reLated Y F P « YFP-hDMPK A(587-629) C/PA YFP fusion proteins anc their subceLLuLar LocaL­ ization. HR = hydrophobic region (boxed se­ quences), M = mitochondria, CM = cLustered mitochon­ dria, C=cytosoL, N = nucLeus, PA=protein aggregates. The subceLLuLar LocaLization was determined using confocal microscopy after expression of fu sio n p ro te in s in C2C12 m yobLasts. CeLLs w ere co-stained for mitochon­ driaL marker cytochrome c oxidase. (B-D) YFP-hDMPK A LocaLized to mitochondria and caused cLustering of mitochondria around the nucLeus. (E - G ) TaiL 1 aLone

(mutant YFP-hDMPK a (534-629)) was sufficient for targeting to mitochondria. (H-J) DeLetion of the ten uLtimate C-terminaL amino acids, incLuding three arginines, did not affect mitochondriaL LocaLization. (K-M) DeLetion of HR2 pLus additionaL uLtimate C-terminaL residues stiLL resuLted in a m ito ­ chondriaL LocaLization. (N-P) Fused to YFP, the uLtimate 30 amino acids, i ncLuding HR2, LocaLized to the cytosoL and nucLeus. (Q-S) IncLusion of aLso HR1 did not restore mitochondriaL LocaLiza­ tion, but induced cytosoLic protein aggregates (arrows). Bars, 10 pm. 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

57 58 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms h osrain ht pcfc nhrn o hMK as hs morpho­ a has also A hDMPK of anchoring specific that observation The 1 segments across multiple species indicated that cow, pig and sheep sheep and pig cow, that indicated species multiple across segments 1 (18). Yet, when we removed this stretch of amino acids from the sequence sequence the from acids amino of stretch this in removed rich we when Yet,(18). particularly is segment decapeptide This S2). Fig. (Supplementary is information targeting A(534-629) MOM all that YFP-hDMPK demonstrating paet idctn ta tee eius o o cnrbt treig in­ targeting contribute not do residues these that indicating apparent, segment tail-only the for observed was behaviour distribution Similar the of % 0 5 around in MOM the to targets A hDMPK Full-length geting. targeting mitochondrial for 2 HR not but HR1 requires tail A's DMPK • Human ih nrae hdohbct (i. AB. ned tsig f hs muta- HR this of long one testing to Indeed, HR2 3A-B). and (Fig. HR1 fuse hydrophobicity the fact in of increased would centre with A the in hDMPK in position mutation equivalent the at (A602) residue alanine 3C-D Fig. in S3. shown Fig. are mutations point GES Supplementary the distinct to of and effects (according and 3A-B Fig. (70)) in given scale is orthologues both of effects segments TA between between comparison a coupling in known the detail analyzed more in arrangements, we function Here and membrane (286). structure anomalous structures induces OSER that and as work ER earlier the from know with we fact, In function. and structure anchor characteristics anchor tail determine HR the of a hydrophobicity and as • Length formed aggregates in protein mainly large occurred probably it There 2Q-S). complexes, (Fig. amorphous localization but resides cytosolic thus a mitochondria, 1 A(587-629)) adopted with PK tail associate of (YFP-hDM not HR2 did and information Targeting 2N-P). (Fig. nucleus and anchoring tail in important be to known residues prolines, and arginines long HR in mDMPK A (Fig. 3B). We predicted that introduction of a R600A R600A a an of introduction replacing that and predicted We A 3B). (Fig. hDMPK A in mDMPK in HR HR2 long and HR1 separating R600 of ence A DMPK in anchors membrane of properties hydrophobic and lengths of specifically associates A DMPK tail of between counterpart murine relationship the of a be expression must there that suggests (216), region HR1. of hydrophobicity strong the of result HR1 only carrying protein fusion YFP a contrast, In HR1. in cytosol the predominantly throughout distributed and while 2K-M), associate (Fig. longer no did mitochondria to latter anchor the to A(600-629)) able first PK still the was that protein (YFP-hDM product only fusion showed fusion A(1-600)) motif PK YFP HR2 a the or (YFP-hDM HR2 carrying lacking mutant cation 2H-J). (Fig. formation became A(1-619)), localization in change (YFP-hDMPK no A YFP-hDMPK of 2A-D). (Fig. to 286) (216, leads area anchoring perinuclear the in mitochondrial mitochondria of cases, of cytosolic clustering a majority shows the population In tar­ cell the ofdistribution. half subcellular other for the while responsible cells, C2C12 was ref) for (95) (see 1 tail C-terminal n smlr a, e xmnd hc pr o hMK , h ioom with isoform the A, hDMPK of part which examined we way, similar a In isoforms of human and mouse. A hydrophobicity analysis of the relevant relevant the of analysis hydrophobicity A mouse. and human of isoforms perinuclear the in mitochondria of clustering to i.e., leads effect, logical trun­ a using analysis Differential 2A). (Fig. residue arginine single a by species other in found acids amino C-terminal ultimate tail of ten the comparison lack Sequence 2E-G). (Fig. segment 1 tail the in included Tail 1 contains two regions with high hydrophobicity (HRs), separated separated (HRs), hydrophobicity high with regions two contains 1 Tail srkn dfeec bten os ad ua DP A s h pres­ the is A DMPK human and mouse between difference striking A Figure 3 • Point mutations A 591 595 602 in the C-termini of DMPK A mDMPK A(R591L) RIPRPGLSEA l C l L l FAAA l AA a ATLGCTGLVAYTGGLTPVWCFPGATFAP 2.03(24) isoforms modulate organelle mDMPK A(L595R) RIPRPGLSEARC l Lr If AAA l Aa AATLGCTGLVAYTGGLTPVWCFPGATFAP 1.88(18) specificity and membrane mDMPK AaA602R) RIPRPGLSE a RC l L l Fa AALARAATLGCTGLVAYTGGLTPVwCFPGATFAP 2.33ai0)/1.5ia20) clustering effects.

YFP-DMPK A fusion proteins mDMPK A RIPRPGLSEARCLL l FAAALAAAATLGCTGLVAYTGGLTPVWCFPGATFAP 2.01a22) mutated in their C-termini hDMPKA RVPRPGLSEA l S l L l FAVVLS rAAALGCIGLVAHAGQLTAVwIRRPGAARAP 2.28ai2)/1.37a20) w e re ex p re sse d in C2C12 579 i : i j i j 629 myobLasts and anaLyzed for hDMPK AaL589R) RVPRPGLSEARSLL l FAVVLSRAAALGCIGLVAHAGQLTAVwRRPGAARAP 2.29(1^/1.37(20) their targeting characteris­ hDMPKAaL593R) RVPRPGLSEALSLL FAVVLSRAAALGCIGLVAHAGQLTAVwIRRPGAARAP 2.12a5)/2.32a6)/1.37a20) tics by confocaL microscopy. hDMPKAaR600A) 2.11(24) (A) ALignment of amino acid 589 593 600 sequences of DMPK A C- B terminaL mutants examined. — mA - hA — mAR591L 3 Residue numbering is indi­ “ hA 3 ■■ hAR600A IV v -- mAL595R 2 ■” mAA602R 3 ■ hAL593R cated separateLy for mouse (top) and human (bottom) jëcjl 600 610 620 ° 59q eoö ^ ö io 620 0 |59p 60u 610 020 ° forms. Hydrophobic regions i (HRs) are boxed, positiveLy U\j' -2 f -2 W -2 ch a rg e d re s id u e s a re boLd, -4' -4 point m u tatio n s are in red. -5- -5 -5 Average hydrophobicity C s c o re s (G E S scaLe (7 0 )) of the HRs are indicated at the right, incLuding the size of the various hydrophobic re g io n s in itaLic, b e tw e e r n / w brackets. (B) Hydrophobic­ ity pLots of the C-termini O SER netsosl of the proteins Listed m (A ) i LLustrate e ffe c ts of point m u tatio n s on LocaL hydrophobicity. (C) TypicaL exampLes of five categories of subceLLuLar LocaLization which were scored for wt and mutant proteins Listed in (A): MOM (with or without cLustered mitochondria), ER (with our without OSERs) or cytosoL. Fifty C2C12 ceLLs were anaLyzed for each protein (n = 2) (D). tio n d em onstrated a s Cift inlc^oaPon fuom tn^ MOM to eden ER anti nausee

O SE1? formotioo in aboet 5% of cel°s (F ¡o. 3D and Supplem antaoy Fic|.

o :mfd-To. h OMoOutrng tUes rK-ansM mutaitiuio is m D M PK R n by vetnitO th e

.MOit MR w vs sptid ¡m two om M lleiregioogi resotted ¡E on woa°aito oMi(t

nddut ER to M e M in 50% rf 6hw p ills and a K .ttroU c [esu6¡uait;¡on iw thiw

otdes flantm t . Thts ointoita w ap hicihlf s.m il^ar do thai f-w rD M d- A de¡f.

oU^ M) out rem arkt bty, th em MMPK 5\(M°rs m dOanf i adonrO no m itoehr -. drialclosterrng (^ g .SO 2n M D ^ wfopf l e m e n t a r y F ^ . 3 3 ^ To address influence of HR structure in a different way, we cre­ ated a mDMPK mutant that carries a shorter HR with lower average hydrophobicity and an extra positive charge in the N-terminal flanking sequence (YFP-mDMPK A(L595R)). The majority of cells expressing this mu­ tant showed a staining of the MOM, again with hardly any mitochondrial clustering (Fig. 3D, Supplementary Fig. S3). In an attempt to elongate the HR, we replaced R591 in mDMPK for a leucine (the corresponding resi­ due at the equivalent position in hDMPK). Presence of this elongated HR ■ now with a negative charge (E589) flanking the N-terminal side, like in the hDMPK sequence - did, however, not change the association with the

ER or the frequency of OSER formation (Fig. 3D, Supplementary Fig. S3). 3 • Topology and specificity of membrane insertion of tail-onchorod DMPP isoforms

59 ► 60 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms 12 amino acids, which specified MOM association and mito-clustering mito-clustering and association MOM specified which acids, amino 12 soito characteristics. association that imply findings our p<0.001). Collectively, S3; Fig. Supplementary and splits which L593R, mutation of introduction Remarkably, p<0.001). S3; N- the of exchange and region hydrophobic the of Shortening ability. changes in amino acid composition can strongly influence membrane membrane influence strongly can composition acid subtle that amino show data in hy- the high the changes Furthermore, on depends segment. HR1 indeed the of clustering drophobicity mitochondrial hDMPK-induced and association Fig. introduction mitochondrial Supplementary 3D, (Fig. via one localization reduced cytosolic positive a into strongly protein a the drove mutation for L589R charge a of negative flanking terminal nrae n O ascain wt rdcd lseig blt (i. 3D (Fig. ability clustering reduced with significant a in association, resulted MOM in stretches, hydrophobic increase aller sm even two in HR1 iia aaye wr promd o hMK ' H1 emn of segment HR1 A's hDMPK for performed were analyses Similar B A C A YFP-mDMPK F-DP 69± . .7±0.06 ± 0.37 0.02 ± 0.17 0.09 ± 0.97 0.01± 0.32 4.3 16.9 ± 0.10 ± 0.79 2.4 10.8 ± 1.1 2.4 ± C YFP-hDMPK 3.6 15.2 ± A YFP-hDMPK 16.02.3 + E YFP-mDMPK C YFP-mDMPK A YFP-mDMPK rbec e e 115sec sec 4 sec 0 prebleach ..... * ■ □I - . T1/2 (s) ± sem (s) ± T1/2 / i ,/ # # ■ • ie (sec.) time □ oiefato sem ± fraction Mobile • □ — . ' ' r i f A K P M D m - P F Y of n tio c FP- Y fra r fo tly ed n a tain ific ob n ig s at th r iffe m d fro s e lu a v ll A five of s e s ly a n a P A R F m fro ­ ex re e w s in te ro p K P M D - P F Y (n = 3; tw o or th re e c e lls per per t). lls en e c erim p x e e re th or o tw =(n 3; - FP Y d e liz a c lo for lly tia n n re ctio (T1/2) fra iffe d e ile tim b o m lf a h s. and rm ry e v isofo o c e R K P M ) D (C - P F Y t n re e tim iffe d t s in a g a plotted , s e lu a v sities, ten in t n e pm. c s re 3.5 o lu x F ) pm B ( 3.5 is a re a a g in w o h s s e e g tiv a ta n im se re of p re s A rie e s ) P (A A FR . is s to ly a n a d cte je b u s and pool. cytosolic a with ex­ change quick no but diffusion DMPK isoforms show lateral lateral show isoforms DMPK • Tail-anchored 4 Figure m D M P K E, ex ce pt th e m o b ile ile b o m e th pt ce ex E, K P M D m B. in n w o sh s rm fo o is PK M D h c a le b - re p to d lize a rm o n d e h c a le B . is s ly a n during a A P A K R ­ P F s M s D re p m - x P e F Y t s ing la b o y m 12 C2C ts s la b o y m 12 C2C in d sse re p • Membrane-anchored DMPK isoforms show lateral diffusion, but no rapid exchange with a cytosolic pool Biochemical extraction experiments demonstrated that mDMPK mem­ brane anchoring is robust and can only be disrupted by treatment with detergents (286). To examine membrane association dynamics in another manner, we subjected YFP-DMPK fusion proteins to FRAP analysis (Fig. 4A). YFP-mDMPK E, a cytosolic splice isoform (296), was used as refer­ ence and showed a rapid recovery with a T1/2 of 2.4 s, without any ap­ preciable immobile fraction (Fig. 4B-C). ER-resident YFP-mDMPK A dis­ played a much slower recovery (Fig. 4A-C) and a mobile fraction of 0.79, implicating hardly any exchange with a cytosolic pool of newly synthe­ sized protein but rather recovery via lateral diffusion in the membrane (239). Recovery of MOM isoforms YFP-hDMPK A and C and YFP-mDMPK C was even lower. Both YFP-DMPK C isoforms showed similar mobile frac­ tions of 0.32 and 0.37 and T1/2 of 15.2 and 16.9 seconds, respectively (Fig. 4B and C). YFP-hDMPK A’s mobile fraction was only 0.17, implicating a very static nature of the protein, when present on clustered mitochon­ dria. These results indicate that TA DMPK isoforms are firmly attached to the membrane and show only migration through lateral diffusion.

• Membrane topology of DMPK tail anchors Finally, we performed a topology study on all tail-anchored mouse and human DMPK isoforms. To discriminate between situations in which the C-terminus traverses the organellar membrane (Nout-C ) or is associ­ ated with the outer leaflet of the lipid bilayer (Nout-Cout), we generated YFP-DMPK-HA fusion constructs, with DMPK flanked by an N-terminal YFP-moiety and a C-terminal HA-tag. By western blotting analysis of cell fractions and immunofluorescence microscopy, we first demonstrated that tagging with the HA moiety did not alter DMPK's membrane anchor­ ing ability (Supplementary Fig. S1 and S4). Topology of DMPK proteins was then analyzed using digitonin-induced selective permeabilization of the plasma membrane in combination with complete permeabilization by Triton X-100 (see Supplementary Fig. S5). Conspicuously, anti-HA immu­ nofluorescence revealed that the HA-tag at the C-terminus of YFP-mD- MPK A-HA did not produce any specific signal in digitonin-permeabilized cells, indicating that the C-end was localized at the luminal side of the ER in C2C12 cells (Fig. 5A). In contrast, under the condition of selective permeabilization with digitonin (Fig. 5B and C), the HA tag at the C-termini of mouse and hu­ man YFP-DMPK C yielded clear signals and therefore must be exposed (in)to the cytosol. In the majority of cells expressing YFP-hDMPK A-HA, no HA-tag signal was detectable after selective permeabilization of the cell membrane, suggestive of an N -C topology (Fig. 5D). ' cytosol intermembrane space “ o j ^ o ' We know that in a minority of cells the MOM had become leaky due to the initiation of apoptosis, explaining the presence of anti-HA signal in some digitonin-treated cells (216). Global membrane permeabilization with Triton X-100 resulted in colocalization of YFP fluorescence and HA-immunostaining in all cells. Essential similar topology findings were obtained for transfected N2A cells, a neuroblastoma cell line, thus con­ firming our conclusions (Supplementary Fig. S6). 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

61 62 3 • Tope ra y and specificity™ membrane insertion of t^ B anchored DMPK orms A B D YFP-mDMPK A-HA YFP-mDMPK YFP-hDM PK A-HA PK YFP-hDM YFP-hDMPK C-HA YFP-hDMPK F 1.H xmerge x HA . 1 YFP ’-'- O A merge HA - A } }A * • j* ( ' \ L > . 4 tve cels di ayed d e y la p is d lls e c e itiv s o p - P F Y tag HA e th lls e c d te c fe s n tra s a w s u in rm te - C the at th ­ s e c c a s a w HA-tag e th s a w y d o tib n a A H ti- an ized n An recog re e w lls e c d cte fe i i C=cyt l, o s to y e c = h t .C y h d t o i tib w n l a A a H n ti- g n i a s r a e l all c a n tio a iliz b a Triton e m r fter e A p 0 . 0 1 le - ib X s s e c c a s a w the In ) (D MOM. e th of e id s ^ n ito ig d ild m r fte a le ib s e e (s e n ra b m e m ER the of t­ a ic d in le ib s s e c c a not s a w fusion of s u in HA-tag -term C e e th th ct at te e d to sed u fusion A H - K P M D - P F Y of r tio a iliz b a e rm e p e tiv c le e s DMPK membrane-anchored =lme. r, 0 pm. 10 ars, B en. lum L= of • Topology 5 Figure ort o YFP-hDMPK AHA A A-H K P M D h - P F Y of rity jo a m lic o s to y c e th at d te a c lo g icatin d in , n atio iliz ab e rm e p teins ro p n sio fu For C C) PK M D and (B both ). n tio side tra s l a illu in m lu e s a th w at s u in d rm e te liz - a C c lo e th A at th A-H ing K P M D m - P F Y the f o g , a t lls e - c A H d e in- n iliz ito b a dig e rm In e p ) (A teins. pro . e c n e c s re o flu P F Y ir e th by ­ s n ra T . ts s la b d o y cte m sfe n tra 12 C2C d tly te n a ie s stig n ve in tra in s a w s in te y ro g p lo o p to n sertio in S5), e n re u ra b Fig m e m e e (s l co to ro p , ased b in- n ito ig d a g sin U isoforms. I = in te r m e m b r a n e s p a c e , , e c a p s e n a r b m e m r I = te in • DISCUSSION

The molecular principles that govern membrane insertion and organelle selectivity of C-termini of tail-anchored proteins are still rather poorly understood. Here, we used long DMPK isoforms, which show both splice- form and species-dependent variation in intracellular location, as mod­ els to learn more about sequence characteristics that control anchoring of proteins in the ER or the MOM. Earlier studies of our group had already classified DMPK as a member of the functionally highly variable group of TA proteins and identified DMPKs of different species as proteins with a particular variable anchoring behaviour. Interspecies difference in protein targeting as noted for DMPK A has been described for at least one other protein, i.e., the equilibrative nucleoside transporter 1 (161). Examples of differential location of splice variants of TA proteins are also known, e.g. the VAMP-1A protein targets to the endomembrane sys­ tem, whereas the VAMP-1B protein, a product of the same VAMP-1 gene, locates at the MOM (119). Also SLMAP-TA1 locates to the ER, whereas its splice-counterpart SLAMP-TA2 locates to the ER and MOM (32). Our study of human DMPK C anchoring presented here, together with study of mouse DMPK C anchoring earlier (286), now definitively es­ tablishes that presence of a coiled coil domain next to the tail domain is required for proper insertion. This suggests that, in addition to the moderately hydrophobic tail anchor (average score of 1.54, GES scale), some form of cooperative binding in multimeric complexes, association with other membrane proteins, or direct binding of the coiled coil domain to the membrane is crucial (221). Alternatively, presence of the coiled coil domain may help to drive DMPK C's tail in a membrane-receptive conformation, even though protein structure software predicts that the tail region has characteristics of an unstructured random coil. A con­ sensus for MOM-anchored proteins shows that their TAs are of moderate hydrophobicity, flanked by basic residues (18). Human DMPK C contains basic amino acids only at the N-terminal border of its HR. The corre­ sponding residues have been shown to be important in anchoring of the mouse orthologue (286). A prominent and conspicuous characteristic of DMPK C from both species is the high (25-30%) proline content in and C-terminal of the HR. Proline residues in HRs have been implicated in sorting of inner membrane proteins in mitochondria (192). In addition, proline residues have a strong impact on the formation of a-helices and the efficiency of membrane insertion (109). Helix-breaking proline resi­ dues in DMPK C's HR may thus prevent formation of long, transmembrane a-helices and be responsible for the unique Nout-Cout topology in the MOM. To our best knowledge, DMPK C isoforms are the first TA proteins described to display an Nout-Cout orientation. This topology implies that 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

63 64 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms A and C do not fully obey currently postulated general consensus rules rules consensus isoforms general DMPK postulated human currently that obey fully show not do C results and A these together, Taken (20). A 22, u eprmna eiec fr hs da s tl needed. still is idea this for evidence experimental but (292), excluded. be thus can 138) (25, suggest that HR1 contributes by becoming embedded in the outer leaflet leaflet outer the in embedded becoming by contributes HR1 that suggest impairs than rather A(L593R)) hDMPK enhances mutant in (like segments lipid the spans that A hDMPK of HR2 the is it that assume we intermem­ sequences, the in C-terminus their of exposure showed all determined inves­ be to remains has proteins ER lumen ER soluble the in with exposed interactions region functional C-terminal any all sm the Whether 305). and murine all in mutant each for identical properties was intrinsic behaviour by partitioning determined entirely is behaviour anchoring cific consensus the from predictions with accordance in not and Surprisingly, point HR some the for although A, flanking DMPK of residues case the in Basic important (109). less residues appeared hydrophobic sufficient of segment TA the in residue (A602R/R600A) charged single one of ping translated newly of pool cytosolic a with exchange rapid that suggests f h MM mc lk i pooe fr yrpoi dmi 1 n GDAP1 in 1 domain hydrophobic for proposed is like much A(1-600))? MOM, hDMPK the of re­ We be mutant sequence (compare HR1 targeting MOM complete the direct efficient by could for not If then quired how conclusion. spanning, this membrane corroborates localization, mitochondrial hydrophobic two in HR1 of separation that finding The 302). (118, domain ER regular other of that to (239). indicated proteins comparable is mobility A membrane mDMPK membrane of for mobility data the that published to Comparison tigated. 154, 230, (18, A mDMPK Nout-Cin for an orientation found we published, subcellular that observation the on based spe­ this domain that C-terminal however, DMPK's of propose, do We proteins. TA MOM-targeted for of C-terminal residues charged positively proteins, TA region. the gain MOM-targeted of for or loss hydrophobicity of total and effects length in between alterations or discriminate to charge, of difficult was it bymutants flanked when TA the within tolerated shown be have may Others residue location. charged membrane in a il­ shift that a nicely induced mutants was these tendency This protein 137). a (18, A localize to DMPK prone mouse more like are ER, isoform, the this to in present as gions al­ which conformation a adopts or (286) proteins inserted once MOM-localized tail, other DMPK's that propose We occur. not will (220) protein cross­ than rather membrane, the in embedded is anchor membrane the bilayer, since the HR1 is probably too short to act as a transmembrane transmembrane a as act to short too probably is acid HR1 for amino the here linear of since bilayer, length observed we segment of what to estimations on similar Based 292), A. 141, 138,hDMPK (3, space brane types cell in different in differences membranes like target of factors, extrinsic composition of protein or Effects lipid tested. types cell human hDMPK of targeting MOM proper for important be to not appeared HR2 A(R600A). A(A602R) ­ hDMPK ap mDMPK Sw and of behaviour the by lustrated (109). bilayer lipid the in anchoring with strong very a interactions lows multimerization, in engaged be may membrane, the in analysis FRAP during observed fraction immobile large very The it. ing The few MOM-resident TA proteins for which membrane topology was was topology membrane which for proteins TA MOM-resident few The As for all ER-resident TA proteins for which membrane topology was was topology membrane which for proteins TA ER-resident all for As u aayi o DP A ofre ta ln, ihy yrpoi, re­ hydrophobic, highly long, that confirmed A DMPK of analysis Our Apparently, stasis of DMPK A production and membrane anchoring is essential. We observed here and earlier that mitochondrial cluster­ ing (and to a lesser extent OSER formation) can already be induced at moderate levels of ectopic DMPK A expression (see also (216)). Taken together with findings reported here, we thus assume that induction of organelle clustering and protein aggregation is an intrinsic capacity of the hDMPK A tail itself, and not a simple effect of expression levels that go far beyond the physiological threshold. Our mutational analysis dem­ onstrated that only the HR1 in hDMPK A's tail anchor region was already sufficient for induction of aggregates. Interruption of HR1 in full length hDMPK A by mutating a leucine to an arginine residue (mutant hDMPK A(L593R)) strongly reduced its mitochondrial clustering ability. HR1's high hydrophobicity may induce formation of aggregates when proper target­ ing and timing of anchoring fails. Evidence has been provided that an­ choring of highly hydrophobic TA proteins should be swift (spontaneous insertion (138)) or guided by chaperones in order to prevent aggregation in the cytosol (18, 221). Possibly, presence of an adjacent coiled coil domain affects the efficiency or order of steps that help the protein domain towards correct structural assembly, with or in organelle mem­ branes. For mitofusin 2 (234) it has been shown that presence of a coiled coil domain affects tethering behaviour of opposing membranes. Induced dense packing of mitochondrial membrane structures (216) may further potentiate abnormal interaction-assembly-anchoring of hDMPK A, thus explaining the strongly impaired FRA P mobility of hDMPK A. In summary, our data support that length and hydrophobicity of the membrane anchor is the dominant factor determining organelle specific­ ity of DMPK splice isoforms. Evolutionary adaptation of the membrane anchor leading to altered membrane location between species may im­ plicate different functions for mouse and human DMPK A isoforms. Alter­ natively, differential membrane localization may be of minor importance if DMPK functions at ER-MOM contact sites (91). Future study is needed to reveal details about the relation between intracellular membrane partitioning behaviour and biological function for DMPK. 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

65 66 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms ACAAC3 lctd n h mlil coig ie f h pEF- 1 p EYFP-C the of site cloning multiple the in located TAACAACAAC-3' XhoI with digested was obtained product PCR The mutation). Tyr584Stop F-D K (-8) A C famn ws mlfe b st-ietd mu­ site-directed by amplified was C(1-583). fragment PK PCR YFP-hDM A (R) f ua ti 1 nldn te 'UR s C-mlfe from PCR-amplified as w 3'-UTR the including 1 tail human of (HR1) 1 verse primer 5'-GGCGGTACCTAAGCGTAATCTGGAACATCGTATGGGTAGGGGGC- 5'-GGCGGTACCTAAGCGTAATCTGGAACATCGTATGGGTAGGGGGC- primer verse the into subcloned and purified gel BglII, with trimmed was DNA sulting pEYFP-C1. of site BglII the into subcloned of site BglII the into subcloned and purified gel pEYFP-C1. BglII, with in trimmed located was 5'-TGCAATAAACAAGTTAACAACAAC-3' BglII primer a reverse adding and site 5'-GTCAGATCTATGGAAGTGGAGGCCGAGCAG-3' primer ward of site BglII the into subcloned and purified gel pEYFP-C1. BglII, with in trimmed located was 5'-TGCAATAAACAAGTTAACAACAAC-3' primer reverse and site ligated was mutation the containing fragment kb 1.7 resulting a the and (generates re­ 5'-GGAAAGCGCCTCCGCTAGGCCAGGCCTAGG-3' and primer verse 5'-CCTAGGCCTGGCCTAGCGGAGGCGCTTTCC-3' primer forward with • DMPK cDNA cloning and site-directed mutagenesis mutagenesis site-directed and cloning cDNA DMPK • AGTGT-' dig n Atg n Kn st. hs rget was fragment This site. KpnI and re­ HA-tag and an adding 5'-GCCGCTGGCAGACACAGTT-3' GAAGGTGGCTC-3' primer pEYFP- forward template from with A PCR-amplified was mDMPK A mDMPK of region coding the re­ The backbone. pEYFP-C1 the of site cloning multiple the in located p EYFP-hD- template from PCR-amplified was 3'-UTR the including 1 tail 5'-TGCAATAAACAAGT- primer 5'-GTCAGATCTCGTGCCGC- reverse and primer site BglII forward a with adding A CGCCCTG-3' pEYFP-hDMPK template DNA resulting The backbone. pEYFP-C1 the of site cloning multiple for­ the with C pEYFP-hDMPK template from PCR-amplified was 3'-UTR the DNA resulting The backbone. pEYFP-C1 the of site BglII cloning a multiple the adding 5'-GGAAGATCTGCTGTCACGGGGGTCCC-3' primer forward (286) C PK pEYFP-hDM template from polymerase Pfu using tagenesis gI st o pEYFP-C1. of site BglII 5'-TGCAATAAACAAGTTAACAACAAC-3' primer reverse add­ and site 5'-GTGAGATCTGAGGCGCTTTCCCTGCTC-3' BglII a primer ing forward with A MPK and purified gel BglII, with trimmed was DNA resulting The backbone. with C PK pEYFP-hDM template from PCR-amplified was 3'-UTR the ing C. pEYFP-hDMPK of site XhoI the into • SUPPLEMENTARY DATA F- MP AH. famn ecmasn te ia 3 90 p of bp 900 3' final the encompassing fragment A A-HA. PK DM YFP-m YFP-hDMPK A(587-629). A fragment encompassing HR1 and 2 of human human of 2 and HR1 A(587-629). encompassing YFP-hDMPK fragment A region hydrophobic A(600-629). encompassing fragment YFP-hDMPK A including 2 C(421-630). tail human YFP-hDMPK encompassing fragment A includ­ 2 tail human C(534-630). encompassing YFP-hDMPK fragment A trimmed with XbaI and KpnI resulting in a 250 bp fragment. Another fragment of about 1.1 kb encompassing the 3'-UTR of mDMPK A was PCR-amplified with forward primer 5'-GCCGGTACCTGAACCCTAAGACTC- CAAGC-3’ adding a KpnI site and reverse primer 5'-TGCAATAAACAAGT- TAACAACAAC-3' located in the multiple cloning site of pEYFP-C1. The resulting DNA was trimmed with KpnI and SalI resulting in a 800 bp fragment. The fragments were subcloned into pEYFP-mDMPK A digested with XbaI and SalI. YFP-m DM PK C-HA. A fragment encompassing the final 3’ 900 bp of the coding region of mDMPK C was PCR-amplified from template pEYFP-mDMPK C with forward primer 5'-GCCGCTGGCAGACACAGTT-3' and reverse primer 5'-GGCGGTACCTAAGCGTAATCTGGAACATCGTATGGG- TAGGGTTCAGGGGGCGAAGG-3' adding an HA-tag and KpnI site. This frag­ ment was trimmed with XmnI and KpnI resulting in a 450 bp fragment. Another fragment of about 1.1 kb encompassing the 3’-UTR of mDMPK C was PCR-amplified with forward primer 5’-GCCGGTACCTGAACCCTA- AGACTCCAAGC-3' adding a KpnI site and reverse primer 5'-GCCGG- TACCTAAGACTCCAAGCCATCTTTC-3’ located in the multiple cloning site of pEYFP-C1. The resulting DNA was trimmed with KpnI and SalI resulting in a 800 bp fragment. The fragments were subcloned into pEYFP-m DM PK C digested with XmnI and SalI. YFP-hDM PK A-HA. A fragment encompassing the final 3’ 900 bp of the coding region of hDMPK A was PCR-amplified from template pEYFP- hDMPK A with forward primer 5’-GCCGCTGGTGGACGAAGGG-3’ and re­ verse primer 5'-GGCGGTACCTAAGCGTAATCTGGAACATCGTATGGGTAGGGAGC- GCGGGCGGCTC-3' adding an HA-tag and KpnI site. This fragment was trimmed with SstI and KpnI resulting in a 500 bp fragment. Another fragment of about 1.1 kb encompassing the 3’-UTR of hDMPK A was PCR-amplified with forward primer 5’-GCCGGTACCTGAACCCTAGAACT- GTCTTCG-3’ adding a KpnI site and reverse primer 5-TGCAATAAACAAGT- TAACAACAAC-3' located in the multiple cloning site of pEYFP-C1. The resulting DNA was trimmed with KpnI and SalI resulting in a 800 bp fragment. The fragments were subcloned into pEYFP-hDMPK A digested with SstI and SalI. YFP-hDMPK A(1-619). A PCR fragment was amplified by site-directed mutagenesis from template pEYFP-hDMPK A with forward primer 5’-CT- CACCGCAGTCTGACGCCGCCCAGGAG-3’ and reverse primer 5’-CTCCTGGGC- GGCGTCAGACTGCGGTGAG-3' (generates a Trp620Stop mutation). The re­ sulting product was digested with XbaI and EcoRI and the 1 kb fragment containing the mutation was ligated into pEYFP-hDMPK A digested with XbaI and EcoRI. YFP-hDMPK A(1-600). A PCR fragment was amplified by site-directed mutagenesis from template pEYFP-hDMPK A with forward primer 5’-GTT- GTTCTGTCTCGTTGAGCCGCCCTGGG-3' and reverse primer 5'-CCCAGGGCG- GCTCAACGAGACAGAACAAC-3' (generates an Ala601 Stop mutation). The resulting product was digested with XbaI and EcoRI and the 1 kb frag­ ment containing the mutation was ligated into pEYFP-hDMPK A digested with XbaI and EcoRI. YFP-hDMPK A(R600A). A PCR fragment was amplified by site-directed mutagenesis from template pEYFP-hDMPK A with forward primer 5’-CC- GTTGTTCTGTCTGCTGCCGCCGCCCTGG-3’ and reverse primer 5’-CCAGGGCG- 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

67 68 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms E1217 Sempiu tieelnau, AQ1118 Rattus AAQQ01017148; tridecemlineatus, Spermophilus PE01124197; A A 1 kb fragment containing the mutation was ligated into pEYFP-m DM PK A PK DM pEYFP-m into ligated was mutation the containing fragment kb 1 each species (95): Ochotona princeps, AAYZ01229401; Myotis lucifugus, lucifugus, Myotis in used AAYZ01229401; was 14 princeps, exon of Ochotona border (95): the at deduced species site each were acceptor splice 3' sequences protein alternative DMPK databases. genome and script • Bioinformatics DP A ietd ih bI n HindIII. and XbaI with digested A mDMPK Leu595Arg an (generates primer reverse and 5’-CAGAGCAGCGGCGAACCGGAGCAGGCAACG-3’ 5’-CGTTGCCTGCTCCGGTTCGCCGCTGCTCTG-3’ primer HindIII. and XbaI with digested A mDMPK primer reverse and 5’-CGCTGCTCTGGCTCGGGCCGCCACACTGG-3’ primer muta­ Leu589Arg a (generates CAGGAGCAGGGAACGCGCCTCCGATAG-3’ EcoRI. and XbaI with rm h floig uloie aaae cesos asmn ta the that assuming accessions, database nucleotide following the from query as used were these) from derived sequences protein (and quences (216). earlier the 1 kb fragment containing the mutation was ligated into pEYFP- pEYFP- into ligated was mutation the and containing HindIII and fragment XbaI kb with 1 digested was the product resulting forward The mutation). with plasmid A PK DM pEYFP-m pEYFP- template into from ligated mutagenesis ed was mutation the and containing HindIII and fragment XbaI kb with Ala602Arg 1 digested was an the product resulting forward The (generates mutation). with plasmid A PK 5'-CCAGTGTGGCGGCCCGAGCCAGAGCAGCG-3' DM pEYFP-m template from mutagenesis ed HindIII. the and and XbaI HindIII with and digested XbaI with muta­ digested Leu Arg591 was primer an product resulting forward The (generates tion). 5'-GAA- with A primer PK DM CAGGAGCAGGCAAAGCGCCTCGGATAG-3' reverse pEYFP-m and template 5'-CTATCCGAGGCGCTTTGCCTGCTCCTGTTC-3' from mutagenesis ed EcoRI. and a XbaI with digested A (generates pEYFP-hDMPK prim­ prim­ forward with reverse A 5’-CAGAACAACGGCGAATCGGAGCAGGGAAAGCG-3’ PK and pEYFP-hDM er 5’-CGCTTTCCCTGCTCCGATTCGCCGTTGTTCTG-3’ template from er mutagenesis ed EcoRI. 1 and the and XbaI with EcoRI digested and XbaI with digested primer was product forward resulting with The tion). 5'-GAA- A primer PK reverse pEYFP-hDM and template 5'-CTATCGGAGGCGCGTTCCCTGCTCCTGTTC-3' from mutagenesis ed digested A frag­ kb PK 1 the pEYFP-hDM into and ligated EcoRI was and mutation XbaI the with containing digested ment The mutation). was product Arg600Ala an resulting (generates GCGGCAGCAGACAGAACAACGG-3’ os (M021) n hmn N_049 DP ncetd s ­ se tran­ protein, EST, in nucleotide DMPK (blast.ncbi.nlm.nih.gov/blast.cgi) searches (NM_004409) BLAST in human and (NM_032418) Mouse cR ad h 1 b rget otiig h mtto ws iae into ligated was mutation the containing and XbaI fragment with kb 1 digested the was and product EcoRI resulting The mutation). Leu593Arg A pEYFP-hDMPK into ligated was mutation the containing fragment kb F- MP AL9R. PR rget a apiid y site-direct­ by amplified was fragment A(L595R). PCR PK DM A YFP-m site-direct­ by amplified was fragment A(L593R). PCR PK A YFP-hDM F-DP AA0R. PR rget a apiid y site-direct­ by amplified was fragment A(A602R). PCR A YFP-mDMPK site-direct­ by amplified was fragment A(R591L). PCR A YFP-mDMPK site-direct­ by amplified was fragment A(L589R). PCR A YFP-hDMPK Cloning of YFP-hDMPK A(534-629) and YFP-hDMPK C-HA was described described was A(534-629) C-HA YFP-hDMPK of YFP-hDMPK Cloning and norvegicus, XM_218411; Pan troglodytes, XM_512759; Macaca mulatta, XM_001111220; Microcebus murinus, ABDC01131798; Otolemur garnettii, AAQR01015434; Canis familiaris, XM_541551; Felis catus, AANG01678636; Bos taurus, XR_028836; Sus scrofa, CF791315, DT333544, and BE233962, Ovis aries, EE772416. No true DMPK orthologues could be found in Gallus gallus (chick) or Danio rerio (zebrafish). 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

69 70 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms A B • SUPPLEMENTARY FIGURES 0 D > kDa 100 2 D > kDa 32 C2C12 2 D > kDa 32 7 D > kDa 47 r j / / / / // / / / / / / / / j < « A

2 D > kDa 32 7ka > kDa 47 > kDa 65 J? ^ <& 4¡r 4¡r ^ s lNi Ns> N2A seset f rpr expres­ proper of Assessment to p o lo g y in C 2C12 (le ft) anc anc ft) (le 2C12 C in y g lo o p to teins ro p r e rk a m of ts h ig e w blotting. western proteins via fusion DMPK of sion • 1 Figure Supplementary s ig n a ls re p re se n tin g YFP- YFP- g tin n se re p re ls a n ig s nt- onal ti­ n a l a n lo c o n o m A ti-H an e ran b m e m y stud to w used rro a ith w d icate d in r la are u c le o M y). d o tib an l a n lo c l­ a c lo in used s in te ro P ) (A r a (Da). (kD s ead h w rro a p o ly c lo n a l a n tib o d y and and y d o tib n a l a n lo c ly o p ly­ o p P F G ti- n (a ts s la b o y m p ro te ins are in d icated w ith ith w icated d in are ins te ro p ization s tu d ie s in C2C12 C2C12 in s ie d tu s ization 2 (i ) l ant-GFP P F G ti- n (a lls e c t) h (rig N2A M o le c u la r w e ig h ts of m a rk e r r e rk a m of ts h ig e w r la u c proteins. le o M n sio fu A H PK- icate M d D in s w rro a k c la B body). s tein ro P ) (B a). D (k s ead h Supplementary Figure 2 • DMPKA Sequence alignment of C- termini of DMPK isoforms A 600 and C from different species. Ochotona princeps DMPK protein sequences Myotis lucifugus Spermophilus tridecemlineatus were deduced from the fol­ Rattus norvegicus lowing nucleotide database Mus musculus accessions, assuming that Homo sapiens the alternative 3' splice ac­ Pan troglodytes ce p to r site at the b o rd e r of Macaca mulatta Microcebus murinus exon 14 was used in each Otolemur garnettii species [24] (Ochotona prin­ Canis familiaris ceps, AAYZ0 1 22940 1; Myotis Felis catus lucifugus, AAPE0 1 1 24 1 97; Bos taurus Spermophilus tridecem- Sus scrofa Ovis aries lineatus, AAQQ01017148; Rattus norvegicus, XM_218411; Mus musculus, N M _0 324 1 8; H om o sapien s, DMPK C N M _0 0440 9; Pan tro g lo d ytes, 600 XM_5 1 2759; Macaca Ochotona princeps mulatta, XM _00 1 1 1 1 220; Myotis lucifugus Microcebus murinus, Spermophilus tridecemlineatus ABDC0 1 1 3 1 798; Otolemur Rattus norvegicus garnettii, AAQR0 1 0 1 5434; Mus musculus Homo sapiens Canis familiaris, Pan troglodytes XM _54 1 55 1; Felis catus, Macaca mulatta AANG0 1 678636; Bos taurus, Microcebus murinus XR_028836; Sus scrofa, Otolemur garnettii CF791315, DT333544, Canis familiaris Felis catus and BE233962; Ovis ar­ Bos taurus ies, EE7724 1 6). No DMPK Sus scrofa orthologs could be found Ovis aries in G a llu s g a llu s (c h ic k ) or Danio rerio (zebraf- ish). Numbering refers tc am in o acid s e q u e n c e in m an. Identical amino acids are sh o w n in w h ite on a b lack background. X = sequence unknown; - = alignment gap. 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

71 72 3 ‘ Topolo gy and specificity of membrane insertion of tail-anchooed DMPK isoforms s (R 600 A) KA PK M D m A K P M D h A (R600a> w a s co n firm e d by by d e firm n co (R600a>A s a w The ER. e th to d te e targ l­ a (L593R) c A lo K P M D h - P F Y ­ ith w ria d n o h c ito m (L595R) s at A a w K P M found D m - P F Y SER O tein d uce d in and ER e th to s u in -term C ir e th in s s n in tio te ro p n sio fu A K P M D - P F Y point A DMPK of Targeting w o rk. B ars, 10 pm. (S- T ) ER ER ) T (S- pm. t­ 10 e n ER ars, B l ra rk. e o h w rip e p the of ­ in ta s l ria d n o h c ito m a lic o lls s e to c y c a in ly in a m R d 9 8 lte 5 u L s A ) K J- ( l. so cyto and in n tatio u m R 2 . 0 g 6 in A stain An lic I) - o s (G to y c any out ro­ p Fusion ) F - (D s. re ctu tru s - FP Y ) -C (A ER. the e liz a u is R v E - d e sR D a Co­ of n ria. d n ssio o re ch p x ito e m e liz a u is v ese th of n tio a s e liz g a a c im lo e the tiv ta of n se re p re re a ressed p ex tly n ie s n tra ­ re ta e u w m point le g in s g in rry a c mutants. • 3 Figure Supplementary as 20 nm. y. p 250 sco icro m n Bars, tro c le e a n u m im K P M D h - P F Y of n tio a liz a c lo t n e m e rg la n e an s w o h s rt se in (R600A) A K P M D ) h - R P - F (P Y t s a w utan uced. M red g rin ly te d s e lu rk c a m l and ria d n o ria h d c n o ito h c m ito m to ) -O (M ized . d rve se b o s a w of ing rity o in m a In . n atio liz a c lo re­ A PK M D h in n tatio u MOM m the to shift a n tio in a liz a c d lo lte su in re A K P M D m d e liz a c lo s (R591L) a A w to K P M D d m e s u s a w protein r e rk a m to sed u s a w ti­ y n d a o tib n An a COX teins. pro t tan u m layed isp D . lls e c 12 C2C in Supplementary Figure 4 • C2C12 N2A Proper subcellular targeting of YFP-DMPK-HA fusion proteins. C2C12 (A-H) and N2A (I-P) cells were transiently transfected with constructs encoding YFP-DMPK-HA fusion proteins. YFP-DMPK constructs without an HA- tag were used as control. Subcellular localization of DMPK fusion proteins was assessed via YFP flu o re s c e n c e . (A, B, I and J) YFP-mDMPK A-HA and YFP-mDMPK A localized at Y F P- m D M P K ^ th e ER. (C, D, G, H, K, L, C E and P) YFP-DMPK C-HA, like YFP-DMPK C, localized at m ito c h o n d ria and sh o w e d no impaired anchoring capacity in e ith e r c e ll type. (E, F, M and N) YFP-hDMPK A-HA

located at mitochondria and YFP-hDMPK A , YFP-hD M PK A-HA induced the same clustering G % H phenotype as YFP-hDMPK A. Bars, 10 pm.

YFP-hDMPK C — YFP-hDMPK C-HA — 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

73 74 3 • Topology and specificity of me mbrane insertion of tail-anchored DMPK isoforms A 3 and 6 ^g/ml digitonin ^g/ml 6 and 3 B C . \ . ' X O C _ X O C . ” M • « . "¿'à* 4 3 < * t t - Y r - ;V - V ’ '.j:;-- ■

•

neurobl a m o t s la b o r u e n A 2 N •• •• ■ DI - I PD DI ii- I PD C 2 C w 12 A . « ... -A ast s la b o y m J V Â IP 0.3% TritonX-1000.3% tu bu lin . The ER lu m e n w a s s a w n e m lu ER The . lin bu tu a m s la p the e iliz b a e rm e p to anti-C O X or an ti-H trA 2/O m i i m 2/O trA ti-H an ith w or d e X O stain anti-C s a w e c a p s The I. PD ti- n a ith w d e st in a stain g a d by cte ire d d te y tra d s o n o tib n m a e d an s a w l o s to y c (3 lls e c C2C12 to d Thus lie p p a s. e ran b m e m e r iliz la ab e llu e rm c e p ll to a used s a w used re e w s n tio tra n e c n o c n atio iliz ab e rm e the p g e tin ic tiv p c e d le e s e m e h c S ) (A treatment. by digitonin membrane cell the of permeabilization • 5 Selective Figure ental Supplem epectv y. r, 0 pm. 10 ars, B . ly tive c e p re e n ra b m e rm te ­ in m e m and er n e in n ra b l ria d n o h c ito m the of ility ­ ib s ito s ig e d c c A l /m g (C). m ) in (6 n anc ) lls B e ( c ) in N2A n ito ig d l /m g lly m fu s s e c c u s s a w l co to ro p 0 0 X-1 Triton only. e n ra b m e m igitonin d w Lo . le cip rin p Supplementary Figure 6 • To­ A pology of tail-anchored DMPK Y FP- m D M PK A-HA isoforms in N2A cells. Using a selective, digitonin- based permeabilizatior protocol, membrane topology of YFP-DMPK-HA fusion proteins was investigated m transiently transfected N2A neuroblastoma cells. Trans­ fected cells were recognized by their YFP fluorescence. An anti-HA antibody was used to detect the C-termi­ nus of fusion proteins. (A) In B digitonin-permeabilized cells YFP-mDMPK C-HA the HA-tag of YFP-mDMPK A-HA was not acces­ sible indicating that the C-terminus was localized at the lumenal side of the ER membrane (see schematic representation of the protein in th e m e m b ran e ). (B and C) For both DMPK C fusion proteins the HA-tag was ac­ cessible after mild digiton^ permeabilization, indicating that the C-terminus was C located at the cytosolic side YFP-hDMPK C-HA of the MOM. (D) A mixed population was observed after YFP-hDMPK A-HA tra n s fe c tio n : in on e-th ird of digitonin-treated cells the HA tag was accessible tc the HA-antibody, whereas in the other two-third no sig­ nal was detected. C=cytosol, I = innermembrane space, L= lum en. B ars, 10 pm..

YFP-hDMPK A-HA

i

YFP ___ i < merge K,„J w „ . YFP __ HA merge jr a & Hi HA merge 3 3 • Topology and specificity of membrane insertion of tail-anchored DMPK isoforms

75 r 'i CHAPTER 4 L . J

A TAIL-ANCHORED MYOTONIC DYSTROPHY PROTEIN KINASE ISOFORM INDUCES PERINUCLEAR CLUSTERING OF MITOCHONDRIA, AUTOPHAGY AND APOPTOSIS

Ralph J. A. Oude Ophuis1, Mietske Wijers1, Miranda B. Bennink2, Fons A. J. van de Loo2, Jack A. M. Fransen1, Bé Wieringa1, Derick G. Wansink1

departm ent of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands 2Rheumatology Research and Advanced Therapeutics, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

PLoS ONE. 25;4(11):e8024, 2009 78 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis Background: Studies on the myotonic dystrophy protein kinase (DMPK) (DMPK) kinase protein dystrophy myotonic the on Studies Background: • ABSTRACT nlssuig F-DP uin osrcs eeld htte protein's the that revealed constructs fusion A YFP-hDMPK using analysis induces expression A hDMPK changes, morphomechanical to addition variety a in occurred Clustering region. perinuclear the in clustering and A hDMPK the gene, human the of product protein major one of pression products. protein DMPK of function the about known is little Surprisingly might be a codetermining factor in manifestation of specific DM1 features features DM1 of specific manifestation in expression factor isoform codetermining bea splice might aberrant that speculate We expressed. is gene cluster­ mitochondrial evoke to sufficient and necessary was domain tail Truncation apoptosis. by accompanied space potential, intermembrane membrane tochondrial mitochondrial of loss In like cytoskeleton. changes tubulin intact physiological an by enhanced was and types cell of models. disease and patients DM1 in level RNA the at expansion ofrepeat of fate on the concentrated mainly far thus have products gene and gene isoform with a long tail anchor, results in mitochondrial fragmentation fragmentation mitochondrial in results anchor, tail long a with ex­ isoform transient that here demonstrate We Findings: Methodology/Principal n patients. in hDMPK the where cells in controlled tightly be to needs isoform A DMPK ofthe level expression the that suggest data Our Conclusion/Significance: behavior.ing mi­ the from c cytochrome of leakage and activity autophagy increased consequences and level DNA the at (CTG)n repeat the in mutation length • INTRODUCTION

The myotonic dystrophy protein kinase (DMPK) gene is involved in myo­ tonic dystrophy type I (DM1) when it is mutant and contains an unstable (CTG)n segment in its 3' terminal exon (176). DMPK encodes several ser­ ine/threonine protein kinases, believed to be involved in ion homeostasis and remodeling of the actin cytoskeleton (13, 95, 199, 296). Up till now, emphasis in most DM1 studies was on the pathobiological significance of toxic RNA products from the mutant DMPK gene. Only relatively few studies have addressed individual protein products from the DMPK gene, including their normal structure function relationship (135, 297). Constitutive and regulated modes of alternative splicing exist for DMPK pre-mRNA and result in the expression of six major DMPK splice isoforms, conserved between mouse and man. Individual isoforms are characterized by presence of either one of two types of long C-termini (tail versions 1 or 2; DMPK isoforms A to D) or a rather short C-terminus (tail 3; isoforms E and F), combined with absence or presence of an internal VSGGG-motif (A vs B, C vs D, E vs F) (95). DMPK isoforms A-D are typical tail-anchored proteins with a membrane segment in their C- terminus. These isoforms are mainly expressed in heart, skeletal muscle and brain. Isoforms E and F are cytosolic proteins, predominantly found in smooth muscle cells (95, 215, 296). Previously, we demonstrated that tail anchors in DMPK A/B and DMPK C/D drive binding to specific organellar membranes (259). In mouse, this results in binding of mDMPK A and B to the endoplasmic reticulum (ER) and in binding of mDMPK C (and D) to the mitochondrial outer membrane (MOM). In humans, hDMPK A/B and C/D have also dis­ tinct tails, but these isoforms all anchor to the MOM. Isoform hDMPK A is unique in that its transient expression causes mitochondrial morphol­ ogy to become abnormal, eventually leading to cell death via an as yet unidentified mechanism (286). Mitochondria form an elaborate network with variable morphology and spatial distribution, tightly controlled by the physiological state of the cell and dependent on cell type and metabolic needs (12). Organellar form and function in this network are regulated by fission and fusion with important bearing on the internal distribution of energy metabolites, the mode of sequestration of intracellular Ca2+ ions (261) and perhaps even apoptosis signaling (312). MOM-associated proteins like mitofusins 1 and 2 (Mfn1 and 2) and OPA1 or hFis control mitochondrial fragmenta­ tion or perinuclear localization (115, 234, 246, 311). Various diseases, either coupled to acquired or inherited defects in bioenergetic circuits or to abnormalities in the fission-fusion machinery have been associated with abnormal mitophysiology (148). Also in DM1 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

79 80 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis 15. ee e eot n n apc, osqecs f xrsin of expression of consequences aspect, one on report we Here (135). ciiy Utmtl, eoain f iohnra ih DP A eut in results A hDMPK with mitochondria of decoration Ultimately, activity. mi­ determining in role A's hDMPK analyze we deficiency, DMPK without model mouse DM1 ina (CTG)11 transgene hDMPK a from products protein os f elfntoa itgiy n i te ntain f apoptosis. of initiation the in and integrity cell-functional of loss autophagic increased with associated was revealed clustering analysis mitochondrial biochemical that induction and for Microscopy sufficient is clustering. tail perinuclear of C-terminal We protein's cell. the of the that integrity demonstrate functional the use and function By and or fate MOM. with tochondrial the cells, to binding cultured in its experiments particular in transfection-complementation of isoform, A organs other hDMPK and the brain muscle, in viability cell of loss insulin and Ca2+ homeostasis, DM1resistance ion of defective including features typical to manifestation, disease contribute thus could involvement- chondrial tolerance workload reduced a caused and cristae aberrant of formation dysfunction mitochondrial and form mitochondrial abnormal patients, n ie (211). mice in and space subsarcolemmal the in mitochondria of accumulation induced and RNA of overexpression Furthermore, 284). (258, described been have uniaie n qaiaie set o DP booy va mito­ -via biology DMPK of aspects qualitative and Quantitative • MATERIALS AND METHODS

• Antibodies and chemicals Staining with anti-cytochrome c oxidase antibody (262) was used to visu­ alize mitochondria and monitor apoptosis. Monoclonal ß-tubulin antibody E7 (Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, Iowa) was used to stain the tubulin cytoskeleton. TexasRed-conjugated- phalloidin staining (Molecular Probes, Breda, The Netherlands) was used to visualize F-actin. Manipulation of the cytoskeleton was performed by incubating cells for 18 hours in cell culture medium supplemented with 3 |j M nocodazole (Sigma, Zwijndrecht, The Netherlands) or 0.6 pM cytochalasin D (Sigma). Mouse monoclonal antibody 12CA5 was used to detect the HA epitope tag. For western blotting DMPK-specific antibody B79 (95), ß-tubulin antibody E7 and LC3b polyclonal antibody (Cell Sig­ nalling, Beverly, Massachusetts) were used. Apoptosis was inhibited by addition of 100 j M of the pan-caspase inhibitor z-vad-fmk (R&D systems, Abingdon, UK) to the culture medium for 16-72 hours.

• Cell culture and DNA transfection C2C12 myoblasts (ATCC #CRL-1772), N2A neuroblastoma cells (ATCC #CCL-131 ) and HeLa cells (ATCC #CCL-2) were grown subconfluent in DMEM supplemented with 10% FCS and maintained at 37oC under a 5% CO2 atmosphere. C2C12 and N2A cells were transiently transfected with expression plasmids (specified below) using Lipofectamine (Invitrogen, Breda, the Netherlands) as specified by the manufacturer. HeLa cells were transiently transfected using polyethyleneimine. After transfection cells were maintained in culture for an additional 8-72 h prior to analy­ sis. Alternatively, we used adenoviral vector-based DNA transduction for expression of single hDMPK isoforms in DMPK knockout (KO) myoblasts (286), since transfection with lipofectamine or polyethyleneimine does not achieve sufficient efficiency. To force C2C12 myoblasts to use mitochondrial oxidative phosphory­ lation for the production of ATP, cells were cultured in DMEM without glucose, supplemented with 10 mM galactose, 1 mM sodium pyruvate, 2 mM glutamine and 10% dialyzed FCS. Cells were maintained at 37°C un­ der a 5% CO2 atmosphere. To maximally stimulate glycolytic metabolism, C2C12 myoblasts were cultured in DMEM, 10% FCS with 10 mM glucose and 1 pM of the mitochondrial uncoupler FCCP (Sigma, Zwijndrecht, The Netherlands) and maintained at 37°C under a 2% O2 atmosphere. Cells were allowed to adapt to these culture conditions for 5 days, transfected with equal amounts of plasmid DNA encoding YFP-hDMPK A or C and then maintained in culture for an additional 20 hours before the number of YFP-positive cells was analyzed. 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

81 Autophagy was induced by placing cells in nutrient-starvation me­ dium Earle's Balanced Salt Solution (EBSS) for two hours.

• Plasmids and adenoviral vectors for DMPK expression EYFP-DMPK expression vectors were obtained by cloning appropriate DMPK cDNA segments into pEYFP-C1 (Clontech, Saint-Germain-en-Laye France) and pSG8AEco vectors (286, 296). Inserts of all expression plasmids were composed of relevant domains from the open reading frames plus the adjacent 3'-UTR of DMPK cDNAs. All segments obtained via PCR were verified by DNA sequencing. MonoYFP-hDMPK A. To introduce a L221K mutation into YFP, pEYFP- C1 vector was PCR-amplified with forward primer 5'-GATCACATGGTCCT- TAAGGAGTTCGTGACC-3' and reverse primer 5'- GGTCACGAACTCCTIA- AGGACCATGTGATC-3' (mutation underlined). The resulting fragment was digested with BsrGI and NheI, gel purified and subcloned into pEYFP-C1. A BglII excised hDMPK A fragment was introduced into the BglII site. YFP-hDMPK A(K100A). A PCR fragment was amplified by site-directed mutagenesis from a template pEYFP-hDMPK A plasmid with forward primer 5'-GGCCAGGTGTATGCCATGGCAATCATGAACAAGTGGGAC-3' and re­ verse primer 5'-GTCCCACTTGTTCATGATTGCCATGGCATACACCTGGCC-3' (mu­ tation underlined). The resulting PCR product was digested with BspEI and the fragment containing the mutation was subcloned between two BspEI sites of pEYFP-hDMPK A. YFP-hDMPK A(534-6l9). A fragment encompassing human tail 1 includ­ ing the 3'-UTR was PCR-amplified from pEYFP-hDMPK A with forward primer 5'-GGAAGATCTGCTGTCACGGGGGTCCC-3' adding a BglII site (under­ lined) and reverse primer 5'-TGCAATAAACAAGTTAACAACAAC-3' located in the multiple cloning site of the pEYFP-C1 backbone. The resulting DNA was trimmed with BglII, gel purified and subcloned into the polylinker of vector pEYFP-C1.

YFP-hDMPK a ®34-62®*3'-™. A fragment encompassing human tail 1 excluding the 3'-UTR was PCR-amplified from pEYFP-hDMPK A with for­ ward primer 5'-GGAAGATCTGCTGTCACGGGGGTCCC-3' adding a BglII site (underlined) and reverse primer 5'-CGAATTCTCAGGGAGCGCGGGCGG-3' located at the stop codon containing an EcoRI site (underlined). The resulting DNA was trimmed with BglII and EcoRI, gel purified and sub­ cloned between the BglII and EcoRI restriction sites in the polylinker of pEYFP-C1. YFP-3'-UTR. A fragment encompassing the hDMPK A 3'-UTR was PCR- amplified from pEYFP-hDMPK A, with forward primer 5'-GGAAGATCTT- GAACCCTAGAACTGTCTTC-3’ adding a BglII site (underlined) and reverse primer 5'-TGCAATAAACAAGTTAACAACAAC-3' located in the multiple clon­ ing site of the p EYFP-C1 backbone. The resulting DNA was trimmed with BglII, gel purified and subcloned into the polylinker in vector pEYFP-C1. hDMPK C-HA. A fragment encompassing the final 900 bp of the cod­ ing region of hDMPK C was amplified by PCR from pEYFP-hDMPK C with forward primer 5'-GCCGCTGGTGGACGAAGGG-3' and reverse primer 5'-GC- CGGTACCTAAGCGTAATCTGGAACATCGTATGGGTAGGGTTCAGGGAGCGCGG-3' adding an HA tag (italics) and KpnI site (underlined). The PCR fragment was trimmed with SstI and KpnI resulting in a 500 bp fragment. A frag­ ment of about 1.1 kb encompassing the 3'-UTR of hDMPK C was PCR am ­

82 plified with forward primer 5-GCCGGTACCTGAACCCTAGAACTGTCTTCG-3' adding a KpnI site (underlined) and reverse primer 5'-TGCAATAAACAAGT- TAACAACAAC-3' located in the multiple cloning site of pEYFP-C1. The PCR product was trimmed with KpnI and SalI resulting in a 800 bp frag­ ment. The two fragments were subcloned into pEYFP-hDMPK C digested with SstI and SalI. The pEYFP-hDMPK C-HA construct was digested with BglII and ligated into the BglII sites of pSG8AEco resulting in pSG8AEco- hDMPK C-HA. E1/E3-deleted serotype 5 adenoviral vectors encoding YFP-hDMPK isoforms A and C under the control of a CMV IE-promoter were generated using the AdEasy Vector System (106), as described (41). In brief, cD- NAs encoding YFP-hDMPK isoforms including their 3'-UTR, were obtained from pEYFP-hDMPK A or C and cloned into transfer vector pShuttle-CMV. In a second step, pShuttle-YFP-hDMPK plasmid was recombined with the viral DNA plasmid pAdEasy-1 in E. Coli strain BJ5183. Next, viral particles were generated in N52.E6 (249) viral packaging cells (93). Ad­ enoviral vectors were purified using CsCl gradient purification and stored at -80oC. The viral titer and plaque forming units were determined (41).

• Immunofluorescence microscopy Cells were grown on glass coverslips, fixed in PBS containing 2% (w/v) formaldehyde ~20 h after transfection and permeabilized in PBS contain­ ing 0.5% (w/v) Nonidet P-40 substitute. Samples were processed for im­ munofluorescence microscopy using standard procedures. Images were obtained with a Bio-Rad MRC1024 confocal laser scanning microscope (Biorad, Hercules, California) equipped with an argon/krypton laser, us­ ing a 60x 1.4 NA oil objective and LaserSharp2000 acquisition software.

• Time lapse imaging Transfected KO myoblasts or N2A cells were stained for 30 minutes using 1 pM MitoTracker Red (CM-H2XROS; Invitrogen). After washing with acqui­ sition medium (DMEM without phenol red supplemented with 10% (N2A) or 20% FCS (KO myoblasts)), dishes were mounted into a temperature- controlled incubation chamber on the stage of an inverted microscope (Axiovert 200 M; Carl Zeiss, Jena, Germany) equipped with a 63x, 1.25 NA Plan NeoFluar oil-immersion objective. Images were acquired every three minutes in three Z-directions at 33°C for myoblasts and 37°C for N2A cells during which cells were maintained in standard culture medium without phenol red. YFP was excited at 510 nm with an acquisition/il­ lumination time of 200 ms and fluorescence light was directed through a 545AF35 emission filter (Omega Optical, Brattleboro, VT). MitoTrackerRed was excited at 568 nm (Polychrome IV) with an acquisition/illumina­ tion time of 200 ms and fluorescence light was directed by a 525DRLP dichroic mirror (Omega) through a 564AF65 emission filter (Omega). A CoolSNAP HQ monochrome CCD-camera (Roper Scientific Photometrics, Vianen, The Netherlands) was used and no bleed through was detected. Hardware was controlled by and images were analyzed using Metamorph 6.2 software for mean fluorescence intensity of the mitochondrial area.

• Western blotting Cells were lysed on ice in NP40 lysis buffer (1% NP40, 50 mM Tris-HCl, 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

83 84 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis h Ntelns ad rnfce wt pYPhMK ad pEYFP-hD- and A pEYFP-hDMPK with transfected and Netherlands) The (Omega) and detected as above. YFP was excited at 470 nm (Polychrome (Polychrome nm 470 at excited was YFP above. as detected and (Omega) X-OMATAR). (Kodak film etos y lcrn irsoy a dn i a iia manner. similar a in done was microscopy electron by (Leica, sections software spher­ more or acquisition one containing Manager cells of Image percentage The Germany). IM500 Solms, Leica ana­ and and (Sigma-Aldrich) system Eukitt Sections using blue. glass toluidine cover with a on stained mounted and cut were 1% were analysis. pm 0.5 EM of for Sections described as embedded and fixed were Samples EM and LM by quantification Autophagosome • microscopy Electron • (149). area mitochondrial ofthe intensity fluorescence mean Corpo­ for Devices software (Molecular software 6.0 Metafluor using controlled was Mannheim, (Roche, cocktail inhibitor protease 1x PMSF, mM 1 vanadate, and examined in a JEOL JEM1010 electron microscope (JEOL, Welwyn Welwyn (JEOL, microscope electron JEM1010 JEOL a in examined and medium. embedding as 812 Epon pure to Epon and ethanol of mixture a filter emission 565ALP a through (Omega) mirror dichroic 560DRLP a supple­ DMEM in nM 100 at (Invitrogen) TMRM ester methyl ylrhodamine (MMP) potential membrane Mitochondrial • to exposure by followed ECL, by generated were signals and used, was oid autophagosome structures was quantified. Quantification on ultrathin ultrathin on Quantification quantified. was structures autophagosome oid (310). membranes of visibility increase to included was tetroxide osmium kV. 80 at operating UK) City, Garden de­ were cells buffer, in ferrocya- washing potassium After buffer. and 1% buffer and cacodylate M 0.1 tetroxide cacodylate in M nide osmium 0.1 1% in in h 1 glutaraldehyde for post-fixed 2% in fixed were Cells 6.2 Metamorph using analyzed were Images PA). hardware Downingtown, All ration, detected. was a through bleed-through (Omega) No filter. mirror emission dichroic 505DRLPXR 565ALP a by directed was light supple­ DMEM acquisition/ an red-free with nm 555 phenol at excited in was FCS. 10%TMRM maintained with mented were cells the which of stage during the on mounted were dishes FCS. Wilco 10% with mented anti­ secondary As membrane. PVDF to blotting western and by SDS- gels SDS-PAGE 8% with on transferred mixed separated were were 10 Lysates buffer. fractions for sample PAGE supernatant centrifugation and by 4°C at 14,000g cleared at were min Lysates EDTA). mM 1 Germany), 0.1 mM pyrophosphate, sodium mM 1 NaF, mM 25 NaCl, mM 7.5, 150pH ye uig ih mcocp wt a ec D60BCR00 imaging DM6000B/CTR6000 Leica a with microscopy light using lyzed rnl ctt, isd ad onesand ih ed irt, i dried, air citrate, lead 3% aqueous with with counterstained contrasted and cut, rinsed, acetate, were nm) uranyl (60-80 sections Ultrathin via transferred and ethanol aqueous of series ascending an in hydrated fluorescence and ms 100 of time acquisition/illumination an with IV) by directed was light fluorescence IV, TILL and (Polychrome Germany) Gräfelfing, monochromator Photonics, a using ms 100 of time illumination temperature room at performed were Experiments microscope. inverted tetrameth­ using minutes 30 for stained were cells h, ~20 C. After MPK B.V., Amsterdam, wells (Willco dishes Willco in cultured were cells HeLa UK) Laboratories, ImmunoResearch (Jackson IgG HRP-conjugated body, • Statistics Data are expressed as mean ± s.e.m. Between group comparison was performed by two-tailed unpaired Student's t-test. Differences between groups were considered significant when P<0.05. * P<0.05, ** P<0.01, *** P<0.001. Statistical analyses were performed with GraphPad Prism 4 software. The number of replicates and the number of cells counted per experiment are described in the figure legends. 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

85 86 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis A®34-62® confirmed that clustering capacity was entirely contained within within contained entirely was A®34-62® capacity clustering that confirmed Since required. not is activity enzymatic that idea the with 1D), consistent mitochondria. to associated are proteins A Fg 1, oet ae) smlr o el wt MMbud F-DP C YFP-hDMPK MOM-bound with cells to similar panel), lowest 1A, (Fig. transfection including conditions, experimental with somewhat varied YFP-hDMPK expressed that cells all virtually in found was fluorescence • RESULTS with cytosolic YFP-hDMPK A displayed a normal elongated appearance appearance elongated normal a displayed A YFP-hDMPK cytosolic with tl osre we te iaeda K0A uat a epesd (Fig. expressed was mutant K100A kinase-dead the when observed still involve­ exclude to incorporated was variant YFP L221K monomeric with was constructs YFP-fusion of series a phenomenon, this in involved are mitochondrial perinuclear for responsible is A hDMPK region of The tail • c dsrbto ad ucin n el (3. h 3-T o hMK mRNA, hDMPK of 3'-UTR The (53). cells in function and distribution uct 1D). (Fig. A’s C-terminus YFP-hDMPK hDMPK mutant of expression Indeed, responsible be clustering. must domain tail mitochondrial the for C ter­ that the in hypothesized only A we hDMPK (286), from minus differs it though even morphology, drial clustering mitochondrial this of A's Use hDMPK (259). affect not portion did YFP tag the of YFP capacity monomeric dimerizing the of ment 1D). (Fig. Tagging myoblasts C2C12 in expressed 1C) and (Fig. designed hDMPK where cells in induced only suggest are changes observations these mitomorphological together, that Taken cells. mock-transfected or cells in mitochondria all Virtually shown). not (data duration and dose categories different the of in Frequencies network a mitochondria. showed few elongated only contained andnormal, 25% cells mitochondria of 1B). Approximately (Fig. fragmented majority appearance possessed for 1A the clustered Fig. a that with (see A revealed mitochondria hDMPK classes) MOM-bound with YFP- morphological cells of mitochondrial Classification contrast, C. without In fluorescence observed. cytosolic was a binding -40% remaining mitochondrial the for while YFP- of hD­ cells, Transfection (215). proteins DMPK for line cell a host myoblasts, this C2C12 of of of natural aspects transfection used range we qualitative detail, broad and more a in in phenomenon quantitative analyze distribution To (286). lines mitochondrial cell affect and MOM the to Earlier, we had observed that hDMPK A has the capacity to localize localize to capacity the has A hDMPK that observed had we Earlier, DP C lo nhr a mtcodi bt os o cag mitochon­ change not does but mitochondria at anchors also C hDMPK was clustering the of mitochondrial property Importantly, 1D). (Fig. intrinsic itself an indeed moiety is DMPK this that demonstrating behavior, YFP-positive all of -60% in decoration mitochondrial in resulted A MPK clustering To investigate whether the entire protein or only domains of hDMPK A hDMPK of domains only or protein entire the whether investigate To t s el nw ta h 3-T o mN cn eemn gn prod gene - determine can mRNA of 3'-UTRa the that known well is It when overexpressed, has been shown to be detrimental to cardiomyo­ cytes and myofibers (177). We therefore verified whether mitochondrial behavior differed between situations where hDMPK A protein was ex­ pressed from constructs with or without its normal 3'-UTR (Fig. 1C and E), but found no differences. Also, a control vector containing a YFP ORF followed by the DMPK 3'-UTR gave the anticipated cytosolic and nuclear YFP distribution. Thus, our data indicate that the 3'-UTR is not involved in hDMPK A-induced mitochondrial clustering. In most cell types where the DMPK gene is naturally expressed, isoforms A and C are present in approximately equal amounts (95, 215, Figure 1 • Perinuclear Mulders and Wansink, data not shown). To test whether presence of clustering of mitochondria is induced by the C-terminal hDMPK C could prevent hDMPK A from inducing mitochondrial clustering, domain of hDMPK A. we used N2A cells, well known for their high transfection efficiency, to (A,B) C2C12 myoblasts generate sufficient numbers of cells that coexpressed hDMPK A and C tra n sie n tly expressing YFP- hDMPK A or C fusior in a double transfection. Isoform hDMPK C failed to modulate formation proteins or mock-transfected of mitochondrial clusters in all doubly transfected cells, indicating that cells were stained with hDMPK A effects are dominant (Fig. 1F). a cytochrome c oxidase antibody to visualize mito­ ABYFP 100 I fragmented mito chondria. Typical examples of classes of m itochondrial 80 I I clustered mito morphology are shown m elongated mito 60 (A). M itochond rial d is trib u ­ tion w as classified as 40 fragmented, clustered or elongated. Frequencies of appearance are listed as percentage of the to tal YFP- YFP- hDMPKA hDMPKC num ber of c e lls expressing MOM-associated hDMPK A C YFP kinase coil tail 3-UTR (n = 3, ~50 cells analyzed YFP-hDM PKA i i i . x n = i per experiment). Arounc monoYFP-hDMPKA 40% of YFP-hDMPK A- K100A expressing cells showed a YFP-hDMPKA(K100A) I I IJ cytosolic expression. These

cells contained mito­ YFP-hDMPKA(534-629) chondria with a typically elongated shape, but were YFP-hDMPKA(534-629)A3'-UTR I I I disregarded in the analysis. YFP-3'-UTR (C) Schem atic representation of YFP constructs used for D E transfection expression with protein domains, mutations v > and 3'-UTR indicated. (D) ■ V C2C12 myoblasts expressing YFP fusion proteins were t y " stained with a cytochrome c hD M PKA _ mito merge oxidase antibody to visual­ T\ ize their ability to induce clustering of m itochondria. (E) Expression of constructs v N i. ■ •• s

with altered 3'-UTRs dem­ monoYFP _ mito merge onstrated that the hDMPK A

3'-UTR w as not involved in f r 'C m itoch on d rial clustering. (F) t Co-expression of YFP-hDMPK f c V A and hDMPK C-HA in N2A V (K100A) — merge F c e lls resulted in m itochon­ drial clustering, whereas änT & cells only expressing hDMPK C-HA exhibited normal, < < elongated mitochondria. A & ^ A

Bars, 10 pm. (534-629) _ mito merge YFP-hDMPKA hDMPK C-HA merge 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

87 88 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis The cytoskeleton controls interactions in the mitochondrial network, network, mitochondrial the in interactions controls cytoskeleton The B A F-DP A r epeso dd o rva ay vr efc o struc­ on effect overt any reveal not did expression C or A YFP-hDMPK displaying perinuclear accumulation of mitochondria was significantly significantly was mitochondria of accumulation perinuclear displaying clustered displayed that cells transfected of (mea­ percentage treatment D the as cytochalasin by sured altered not was clustering of quency mitochondrial clustered alter not did D cytochalasin by disorganization immediately and transduction adenoviral by isoforms C or A hDMPK with cytoskeletal-mitochondrial the at be could role its 297), (135, that behavior reasoned actomyosin affect towe believed also is and and dynamics andmitochondrial toaffect microtubules appears motor A hDMPK along (39).Since proteins, microfilaments transported actin fission/fusion are Mitochondria membrane proteins. of adaptor host a by facilitated clustering mitochondrial reduces depolymerization Microtubule • eue, vn huh h famne apaac o te mitochondrial the of appearance fragmented the though even reduced, cells A-expressing hDMPK of fraction nocodazole-induced the a disorganization, with cells microtubular in contrast, In 2C). Fig. mitochondria; fre­ the Also 2A). (Fig. cells isoform-transfected C and decoration A YFP-hDMPK in between patterns comparison by depolymerization- revealed as F-actin morphology nocodazole. or D cytochalasin with treated reconstituted were (286) myoblasts KO mouse DMPK (6, 307). tochondria oftu­ absence the what examined also B). We and 2A (Fig. integrity tural ui o atn iaet wud o o h prncer cuuain f mi­ of accumulation perinuclear the to do would filaments actin or bulin interface. Comparison of the actin and microtubule cytoskeletons in cells with with cells in cytoskeletons microtubule and actin the of Comparison The num ber of transduced transduced of ber num The ~30 c e lls per experim ent, ent, experim per lls e c ~30 percent­ the of tification nocodazole or D chalasin to A-expressing YFP-hDMPK YFP- ith w transduced r t te ment f cy­ of t n e tm trea t u o ith w or expressing lls e c of tal to ount the am of percentage as clustered contain t tha lls e c clustering. rial d on itoch m of by unaffected as w lls e c of erization Depolym dria. affect not did cytoskeleton ti­ n a anti-tubulin an stained ith w as w cytoskeleton fluorescent by d alize visu or (A) D presence cytochalasin the of in adenoviruses clustering. mitochondrial w ith nocodazole, followed followed ent nocodazole, treatm ith w hours after 12-16 a itochondria m ith w lls e c clustered transduced age B; and A in shown ages (im itochondria m of n utio The istrib d ented. till s fragm appeared itochondria m but cells, P<0.05). (n = 3, wash-out hours 8 a by Quan­ wash-out. of Effect nocodazole (D) P=0.01). ment, ri­ e xp e per n cells ~= 3, 10 0 ith w MOM, the at A hDMPK expressed are itochondria m Bars, ent. treatm nocodazole C-transduced YFP-hDMPK in m clustering rial d n o ch ito m decreased icrotubules m itochon­ m of n calizatio lo actm the of Disruption body. r la bu icrotu m The phalloidin. was F-actin (B). nocodazole C-expressing or A hDMPK were yoblasts m KO DMPK perinuclear A-induced hDMPK enhances cytoskeleton bular microtu- intact •. An 2 Figure 10 pm. (C) Quantification Quantification (C) pm. 10 network remained (Fig. 2B and C). When nocodazole was washed out at 12-16 hours after transduction and expression was continued for an additional 8 hours, the percentage of cells with mito-clusters was sig­ nificantly higher than among cells that were continuously treated with nocodazole (Fig. 2D). Nocodazole had no overt effect on mitochondrial location or morphology in hDMPK C-expressing cells. To us, these obser­ vations suggest that hDMPK A-induced fragmentation and clustering are at least partly uncoupled effects and that microtubular infrastructure only affects the extent of clustering.

• Mitochondrial morphology changes soon after hDMPK A translocation To obtain insight in spatio-temporal aspects of mitochondrial morphology alterations, KO myoblasts were transduced with YFP-hDMPK isoforms an d s ubjected to live imaging. Mitochondria were visualized via Mito- Tracker Red staining and images were taken every three minutes. YFP- hDMPK A located in the cytosol at first appearance, but translocated

Figure 3 • Mitochondrial and concentrated at the MOM A OEPcODMPK A aeoUleoC clustering is a rapid process within four hours, after which and occurs already at low fragmentation and clustering hDMPK A expression. • ^ 1 r * KO myoblasts were trans­ of mitochondria occurred (Fig. _ o n 4:36 0 6:00 n 7:42 n duced with YFP-hDMPK 3A). YFP-hDMPK C appeared di­ fusion proteins (top panels), ; rectly on mitochondria without while mitochondria stained .' i inducing mitochondrial clus­ w ith M itoTracker Red r 1 aiCo y . UK (bottom panels). Images ters (Fig. 3B). In transfected were collected every three B OEP-hDMPK C aeoUleoC N2A cells, initial YFP-hDMPK A minutes. The tim e of first staining in the cytosol was less appearance of YFP signal was set at t = 0; other time evident and hDMPK A appeared points are indicated in the at the MOM without intermedi­ top panels. (A) YFP-hDMPK ate accumulation, directly after A w as first detected in the __ o n 2:09 n 7:06 n 10:42 n cytoplasm. Soon YFP- expression onset. Size changes hDMPK-decorated mitochon­ and clustering of mitochondria dria appeared which ther could be detected as early as started to cluster, eventu­ a lly resulting in severely six hours after transfection in aggregated mitochondria these cells (Fig. 3C). Combining surrounding the nucleus. C OEP-lDMPK A N2A these data with findings in other (B) YFP-hDMPK C directly cell lines (286) demonstrates appeared on m itochondria which maintained their that timing of expression and elongated, reticular structure. sorting of DMPK isoforms may (C) In N2A cells, YFP-hDMPK ■ ^^|6:06h differ between cell types, but A expression emerged m a similar fashion as m the sequence of events that is KO myoblasts, except that induced once DMPK A is as­ mitochondrial clustering IB sociated with mitochondria is occurred almost immediately and cytosolic staining was qualitatively similar. less pronounced. •Abnormal mitochondrial ultrastructuredevelopsijp on hDMPK A expression W e eerfnrm edelrctron m ictoacopn (EM) on N2A ce lls expressing hDMPK A to mnestlciate effects or mitocleonelna l ultraotructure and cell archi- tnnpnre. YFP-hDMPK A-trensfected oells svoso easy recognizable by the concentrctioN of m ltoneopMaa in thejoxtanunlere region (Fig. 4A)., while other puets cl fne neU socio left nom°l.efdl.h voir lof mitochondria (Fig. 4B). Clustered mitochondria showed a rounded morphology, often with disor- 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

89 90 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis YFP-hDMPK C mitochondria contained a normal, intact cristae structure, structure, cristae intact normal, a contained mitochondria C YFP-hDMPK gis tis ak on, noig ht tcodil nect an integcity M fenc- itochondtial thatm kn thi backg owiug stound, Against YFP-hDMPK C-transfected cultures. In control cells or cells expressing expressing cells or cells control In cultures. C-transfected YFP-hDMPK Cutrn o iohnra eut i los;:; in eesults mitochondria of function of Clustering • in cells of majority the in loss viability gradual a to point picture our collectively A influenced DMPK have highest may with this o cells ki offn Although au below). loss etic of We toph (see s expression. ag in of selective yexpression ad period o d to h u ur cells, findingata p en 16-24 ofctio i the this ool the tire nn within attribute level A hDMPK reduced observed a repeatedly we blot, On shown). not ex­ (data ofcells in N2A hours 16-24confirmed after observed was form LC3b-II the to conversion could contain­ cells of remnants fraction The mitophagy. of mitochondrial indicative cases, recognized, be some still In mitochondria. au­ clustered These 4A). (Fig. expression A hDMPK after structures autophagic of (Fig. 4C). localization disperse a had and elongated more were protein using western blotting (136). In transduced myoblasts significant significant myoblasts transduced In (136). blotting western using protein YFP- with cells in 5-6% and expression A YFP-hDMPK after 12-14% to in r dnmicl tnu ad ilU cue t t tr phkaiol.ogical lt'r k c e th to cou^ed tiklsUy and tinkud iucfly dynam are tion ns pool. a fected r t the were observations These 4D). (Fig. C not but A, of YFP-hDMPK pression of areas within and near found were and size in varied tophagosomes density electron of decrease a and structure cristae absent or ganized ing autophagic structures was assessed by LM and EM and amounted amounted and C. hDMPK EM and LM by assessed was structures autophagic ing in seen never were appearance this with Cells 4A). (Fig. matrix the in s scn idctr f uohg, e esrd ovrin ofLC3b conversion measured we autophagy, of indicator second a As Besides abnormal mitochondria, we identified an increased number number increased an identified we mitochondria, abnormal Besides tive and positive controls, controls, positive and tive cultured Cells expression. A a ith w blotting estern w for 8, for C or A expressing YFP-hDMPK yoblasts m m fro ­ ito m clustered contained ented, lls fragm e c A-expressing s odn control. used as loading w as staining antibody a using verified as w sion YFP-hDMPK ing w follo conver­ sion LC3b increased an . reticulum ic endoplasm itochondria m and of loose n utio and istrib d structure proper the cristae Note orientation. the of ent fragm A cells. rial itochond M (C) (B). ria ­ d com ere w ll e c the of other areas contrast, In (arrow). the of density electron electron by analyzed as w C or A YFP-hDMPK ith w morphology. accompanied is clustering at ouin ES) o 2 for (EBSS) Balanced Solution Salt Earle's in cells, cultured nutrient-starved conditions and al rm o n under nucleus. the near chondria MK nioy Tbln (T) Tubulin antibody. DMPK expres­ DMPK respectively. nega­ as used ere w hours, showed and antibody LC3b Lysates (D) pm. 1 Bars, for here included is nucleus C-expressing al hDMPK norm in as w orphology m itochon­ m of devoid letely p observed as w and itophagy m head) (arrow atrix m ith w together often lost as w artly p structure Cristae YFP-hDMPK (A) microscopy. re tu c tru ltras u and s isoform transfectec ere w cells N2A ultrastructural aberrant by Mitochondrial • 4 Figure 16 or 24 hours w ere used used ere w hours 24 or 16 state (12, 38), we investigated hDMPK A effects on cell viability further. To force utilization of relevant metabolic energy pathways, C2C12 myo­ blasts were kept in media of different composition, including high or low oxygen supply, and monitored after transfection with YFP-hDMPK A or C. Cell survival was similar for YFP-hDMPK A-expressing cells growing under normal conditions or high glucose conditions with low oxygen plus use of mitochondrial uncoupler FCCP (to maximally prevent mitochon­ drial OXPHOS activity; Fig. 5A). This outcome can be explained because normal growth of cells in culture already strongly relies on glycolytic metabolism (238) and therefore cannot be further promoted. In glucose- deprived medium supplemented with galactose, forcing cells to rely on mitochondrial oxidative phosphorylation (186), hardly any YFP-hDMPK A- expressing cell survived (Fig. 5A). This suggests that clustering of mito­ chondria is detrimental for cell viability, especially in aerobically-poised situations when cells are forced to use mitochondrial TCA/OXPHOS activ­ ity. As a control, YFP-hDMPK C-expressing cells showed no differences in cell survival under these culture conditions (Fig. 5A).

Figure 5 • Human DMPK A expression affects mito­ chondrial function and cell ■ - ' | | | viability. (A) C2C12 myoblasts were grown under different culture conditions and DMEM + + + + + + transfected with YFP-hDMPK glucose + - + + A or C. The fraction of FCCP - - + - 2% O2 - - + viable YFP positive cells galactose - + - + was determined after 2C pyruvate + + + + hours. When supplied with YFP-hDMPK A YFP-hDMPK C galactose and pyruvate, YFP- hDMPK A-expressing cells showed a significantly lower » viability than YFP-hDMPK YFP-hDMPK A C-expressing c e lls (P<0.01, n = 3, >100 cells analyzed per experiment). (B) The MMP was determined ir HeLa c e lls expressing YFP- hDMPK A or C. YFP-hDMPK A-expressing cells without clustered mitochondria showed a clear MMP signal (upper panels). No TMRM s ign al w as found in YFP- Formatiow of an electrochemicaL proton gradient across the inner mi- hDMPK A-expressing cells with clustered mitochondria tochendrial membrane drives ATP production bn OXPHOS (242). Tn com­ (middle panels, asterisks). pare pos sitile effe cts of D Is! P K isoforms en this c oupling , w e used stain- YFP-hDMPK C-expressing inw of YFP-hDMPK A and C-expressinn HeLa ceUs with the fluorescent cells demonstrated a clear dne TMRM, which provides readoet for tlte mitochondrial membraee mitochondrial signal (lower a panels). Bars, 10 pm. (C) potential UMMP). HeLa cells were chosen because thnse have deen esed The MMP in YFP-hDMPK before for MMP determination and gave sufficient TMRM signal strength A-expressing cells was for quantitative analysis (59). Before measurement, depolarization with almost completely abolishec and significantly lower FCCP as described by Distelmaier et al. (60) was used to control that than in YFP-hDMPK C and TMRM accumulation did not involve quenching, becoming a confounding non-transfected (NT) cells factor in MMP measurement. For many other cell lines the 100 nM TMRM (P<0.001, n = 3, >35 c e lls per experiment). concentration appeared toxic or TMRM did not accumulate well enough in mitochondria to generate adequate signals (data not shown). YFP-hDMPK 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

91 92 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis -oiie el wt n o ol ml mtcodil lseig showed clustering mitochondrial mild only or no with cells A-positive A-expressing HeLa cells showed a significantly lower MMP than cells cells than MMP lower significantly a showed cells HeLa A-expressing f el epesn hMK (i. A. ute spot a otie by obtained was support Further 6A). (Fig. C hDMPK expressing cells of of periods prolonged after rapidly died cells- expressing C YFP-hDMPK xrsig el (i. B lower 6B, (Fig. cells C- YFP-hDMPK expressing for observed was after occurs probably (168), event apoptosis-triggering known well a sug­ this To us panels). middle (Fig.6B, c cytochrome contain did ed fragment­ only were that chondria surviving of number the increased imme­ significantly z-vad-fmk inhibitor transfection apoptosis after with diately treatment that was this conjecture supporting One finding mechanism. apoptosis-based not an by explained -but myoblasts KO A-expressing YFP-hDMPK that observed We apoptosis undergo mitochondria clustered with Cells • 5C). (Fig. cells untransfected in that from significantly not different was not cells (data Red C-expressing YFP-hDMPK in MitoTracker MMP compound The shown). related the clustered internalize to in able were signal of lack The panels). upper 5B, (Fig. MMP normal a the observation that little or no staining of cytochrome c, a mitochondrial c, mitochondrial a cytochrome of staining noor little that observation the be could loss this whether wondered and transfection-complementation aes. F-DP YFP-hDMPK YFP-hDMPK C A panels). release clus­ c cyt o No chrome tering. mitochondrial of onset the c, ofcytochrome release that gests they since up TMRM, take to inability anto due be not could mitochondria YFP-hDMPK C). Interestingly, and 5B (Fig. C YFP-hDMPK expressed that nemmrn poen wa peet n lsee mtcodi i YFP- in mitochondria myoblasts KO clustered in A-expressing hDMPK present w as protein, intermembrane viability the on seen was effect no whereas cells, A-expressing hDMPK B YFP-hD M PK C PK M YFP-hD A PK M YFP-hD A PK M YFP-hD w a y u w F Fg 6, pe pnl) wie mito while panels), upper 6B,(Fig. 6 4 8 6 4 8 6 4 8 6 4 48 24 16 48 24 16 48 24 16 48 24 16 + z-vad-fmk + z-vad-fmk + z-vad-fmk + A discrete, m itochond rial rial itochond m panels). discrete, le A idd (m ented fragm a displayed A YFP-hDMPK cells C-expressing and A per counted cells >90 were hours 16 at Values aintained m and transduction of l a iv rv u s ll ce on tested also observed in YFP-hDMPK YFP-hDMPK in was observed also staining c e cytochrom appeared present itochondria m as w tohsn staining l ria d (upper clustered were itochon­ m ria d hen w staining cytochrome c cytosolic diffuse, expressing e Cells c). cytochrom (cyt c for stained ere w YFP-hDMPK (B) ent). experim n = 3, (P<0.05, 100% at set after ly iate d e m im was applied Z-vad-fmk was Cexpressed. YFP-hDMPK effect no hen w had but cells, A-expressing death ll YFP-hDMPK ce of reduced atly re Z-vad-fmk g 24 hours. 16, 48 and after counted ere w ovi­ en ad A C-expressing or YFP-hDMPK ith w duced apoptosis- of effect The (A) epesn cls Bars, cells. expressing C itochon­ m ar cle A panels). ent. experim the of ­ e ain m tim re ing the for present cells YFP-positive ruses. ­ s n tra yoblasts m KO DMPK was z-vad-fmk r ito hib in apoptosis. induces A hDMPK of Expression • 6 Figure 0 pm. 1 0 • DISCUSSION

We demonstrate here that expression of MOM-anchored hDMPK A results in fragmentation of the mitochondrial network and perinuclear clustering of morphologically-altered mitochondria. In addition to these morphome- chanical changes, physiological changes like loss of MMP, increased autophagosomal activity and leakage of cytochrome c accompanied by apoptosis were induced. It is difficult to bring hierarchical ordering in these events, but we propose that the fate of mitochondria in cells with hDMPK A is specified as follows: based on our observations with live-cell imaging and the finding that microtubule disruption attenuated clustering, we suppose that induction of morphological change is the ini­ tiating step that marks mitochondria for subsequent anomalous transport and clustering. Upregulation of autophagy may be an intermediate phase to rescue defective energy and clear mitochondrial waste (140). The fact that the same sequence of events in mitochondrial fragmen­ tation and perinuclear clustering with globally similar final outcome oc­ curred in all mouse and human cell types examined is important. Slight variation in the speed and magnitude of events was observed between cell types, but this can most easily be explained by differences in cytoar- chitectural arrangement, pathophysiological thresholds or stress-coping ability. Toxicity of hDMPK A was also observed in KO myoblasts in which DMPK was reconstituted to apparently normal expression levels by com­ plementation. We observed mitochondrial clustering already after several hours of hDMPK A expression when ectopic protein levels were still relatively moderate. Therefore, we postulate that induction of abnormal mitochondrial behavior is an intrinsic capacity of hDMPK A and cannot be simply explained as a result of overexpression. Mutational analysis demonstrated that hDMPK A's tail region was sufficient to induce mito­ chondrial fragmentation and clustering. The conclusion that this is the only segment with detrimental effects was corroborated by the observa­ tion that MOM-binding of hDMPK C, which differs from hDMPK A only in its C-terminus, had no effect on mitochondrial physiology and trafficking. Interesting parallels can be drawn to the morphophysiological ef­ fects induced by ectopic or overexpression of other MOM proteins, like TA proteins h Fi s (123, 311), Miro GTPases (80, 82) and Mfn 1 and 2 proteins (115, 234, 241). All these enzymes are involved in mitochondrial dynamics. Although similar in nature, changes evoked by these MOM proteins cannot be simply lumped together, since it is unclear how mi­ tochondrial clustering occurs, and clear mechanistic differences can be noted. For example, enzymatic activity appears irrelevant for Mfn 1 and 2 (1 15, 234, 241) and hDMPK A (this paper), but is necessary for hFis (123) and Miro 1 and 2 (80). We demonstrate that hDMPK A effects do 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

93 94 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis -i.e., not one large continuum- suggesting that lipid-protein or lipid-lipid lipid-lipid or lipid-protein that suggesting continuum- large -i.e., one not Loss of cytochrome c and disappearance of MMP was observed when when observed was MMP of disappearance and c cytochrome of Loss (34, 80, 82, 271). Disbalance in any of these systems may create overtly overtly create may systems these of any in Disbalance 271). 82, 80, (34, phy (38, 148). Also in DM1 patient muscle, mitochondrial irregularities irregularities mitochondrial muscle, patient DM1 in Also 148). (38, phy only can we moment this At studies. interaction biophysical and analyses hDMPK not but A, hDMPK of region tail the in abundant highly are acids transmembrane in peptides in pairs Pro/Gly or Gly/Pro and residues Val have fragmentation or formation vesicle curvature, membrane abnormal data EM Our remodeling. and curvature membrane induce to able are clustering A hDMPK yet (234), effects 2 Mfn determine not does skeleton apoptosis is still unclear (123, 163). We observed that hDMPK A-induced hDMPK that observed 163). We (123, unclear still is apoptosis as mitochondria of clustering and fragmentation with effects, similar en dniid 28 24. tde o T2-DP mc, overexpressing mice, Tg26-hDMPK on Studies 284). (258, identified been atro­ optic and dominant fission and in 2A involved subtype genes in Charcot-Marie-Tooth like mutations fusion, by caused manifestations clustering mitochondrial before apoptosis. of just onset or the to during contribute must occur indicate to that observations events these take that We clustered. became mitochondria in intact largely still was MMP and occur not evidence did provide leak c and death cytochrome cell tothat lead ultimately effects morphological con­ cycle has cell fragmentation and mitochondrial form Although (196). reported mitochondrial been has between trol relationship a even composition acid anchor. amino the of membrane theof importance the about permutation by speculative remain obtained be can tail A's hDMPK in sequences specific with peptide interactions lipid of involvement about knowledge C. amino Further these Exactly 214). (112, index fusogenic high a to add domains of formation by followed clustering and interaction distortion in protein-protein result of that anchors membrane of effects Direct A. hDMPK of round and small are mitochondria A-decorated hDMPK that demonstrate proteins membrane in regions hydrophobic evi­ mounting certain is that There itself. indicating anchor dence membrane A hDMPK the of property outcome. main involved differentially be may proteins machinery motor to diverse and mitochondria complex links the also that involve and MOM may the proteins in these by areas distinct mediated Reactions structures. cell other creat­ in role own its has speculate protein to MOM tempting TA-containing is distinct It every microtubules. that of integrity on depends cyto­ partly the of status 2 (151). The and 1 associations Mfn these on but rely proteins, clearly between effects interactions coiled-coil require not ae en on, lhuh h cue f hs mlomtos a nt yet not has malformations these of cause the although found, been have 172). Recently, (115, appearance morphological to coupled reciprocally is Leu/ that shown been has It involved. be of also could properties fusogenic anchor A's hDMPK abnormal to Coupling 231). (45, described been anchor tail the of presence in affected abnormally become interactions DP -eoae mtcodi ta wr sil n h famne state. fragmented the in still were that mitochondria A-decorated hDMPK in role active its apoptosis, undergoing cells in frequently observed been with interactions mitochondrial for microenvironment appropriate an ing Aberrant mitochondrial morphology is linked to multiple disease disease multiple to linked is morphology mitochondrial Aberrant Many studies have shown that the physiological state of mitochondria ofmitochondria state physiological the that shown have studies Many intrinsic an probably is clustering and fragmentation Mitochondrial tail-anchored hDMPK isoforms especially in heart and skeletal muscle, showed disorganized cristae structure and symptoms like reduced work­ load tolerance, atrophy, hypertrophic cardiomyopathy, myotonic myopathy and hypotension, reminiscent of DM1 symptoms (211). We do not expect that the normal physiological function of hDMPK is directly related to regulation of mitochondrial clustering. Like hDMPK C, hDMPK A is predominantly expressed in skeletal muscle myofibers (95, 215), a cell type with a rigidly compartmentalized cytoarchitectural infrastructure and well-defined mitochondrial distribution across intra- myofibrillar and subsarcolemmal areas. This typical robust cytoarchitec- tural organization may render muscle cells more resistant to fluctuations in hDMPK A expression, but based on our findings we speculate that also for myotubes it is of paramount importance that hDMPK A protein levels are tightly regulated and strictly kept within the normal physiological range. Expression levels of completely other proteins, for example VEGF, also require tight regulation as both reduced and increased expression can result in pathogenesis, manifesting itself in embryonic lethality or cancer, respectively (76). Changes in DMPK mRNA synthesis or pre-mRNA splicing induced by the presence of expanded (CUG)n transcripts, when resulting in only a slight disbalance in hDMPK A expression level, could thus be a contributing factor in the pathophysiology of DM and explain mitochondrial abnormalities. In more extreme situations like congeni­ tal DM with very large (CTG)n expansions, where abnormal chromatin modification and defective bidirectional transcriptional control across the DM locus may directly couple to unbalanced DMPK production, the toxic effects of DMPK A may even directly contribute to special disease features (44, 77). Further research into the role of hDMPK A in health and disease is thus warranted and also needed to validate this model and its predictions. A more precise study of the balance in DMPK splice isoforms in different categories of DM1 patients is a necessary first step.

• Acknowledgments We thank Werner Koopman (Dept. of Biochemistry, NCMLS, Radboud Uni­ versity Nijmegen Medical Centre) for discussions and helpful advice. 4 4 • A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy and apoptosis

95 r ' i CHAPTER 5 L . J

MYOTONIC DYSTROPHY PROTEIN KINASE ISOFORMS A AND C EXERT DIFFERENTIAL CONTROL OVER CA2+ MOBILIZATION IN MYOTUBES

Ralph J.A. Oude Ophuis1, Derick G. Wansink1, Frank Oerlemans1, Mietske Wijers1, Jack A.M. Fransen1, Miranda B. Bennink4, Fons A.J. van de Loo4, Peter H.G.M. W illem s3, Werner J.H. Koopman2,3 and Bé Wieringa1

departm ent of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands 2Microscopical Imaging Centre, Department of Cell Biology, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, The Netherlands 3Department of Biochemistry, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, The Netherlands 4Rheumatology Research and Advanced Therapeutics, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands 98 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes study system. We found the mitochondrial-outer-membrane-associated mitochondrial-outer-membrane-associated the found We system. study cell different in expressed as isoforms, DMPK extended C-terminally and complemented and mock-complemented cells behaved similarly during during similarly behaved DMPK- cells stimulation. mock-complemented and [Ca2+]m acetylcholine upon causing complemented Ca2+ maximal uptake, in mitochondrial decrease a affected isoform, DMPK SR-bound ad­ now to up studies Most control. viability cell and dynamics skeletal eei ae warranted. are genesis state. cell on effects via in C exerted and not A and isoforms direct, DMPK of rather role is Ca2+ the handling that conclude We cytoarchitec- or make-up. balance tural energy in differences overt without myogenesis, C, mDMPK active enzymatically of presence or absence [Ca2+]c [Ca2+]c acetylcholine with peak rate and either decay correlated to maximal in myotubes the Variation of ionomycin. or exposure after stores calcium intracellular Ca2* of the from mobilization the affect to C isoform mDMPK our as variants, DMPK single with myoblasts-myotubes transduction-complementation DMPK-deficient of adenovirus-mediated using cells, in Ca2* inmuscle DMPKs handling individual of role on the report Here, we types. of truncated mixtures natural of bystudy significance biological dressed events, ofcellular variety in a implicated been have gene bythis encoded • ABSTRACT ute suis no psil rl o DP dsaac i D1 patho­ DM1 in disbalance DMPK of role possible a into studies Further an isoforms. variant, DMPK A mDMPK other the of the of expression any of transduction-mediated presence Likewise, or absence with not but cyto- actin-based homeostasis, ion cellular of regulation the including products gene. Different (DMPK) kinase protein dystrophy myotonic expansion thein (CTG)n by a repeat caused is (DM1) 1 type Dystrophy Myotonic • INTRODUCTION

Myotonic dystrophy protein kinase (DMPK) is the protein product of the gene involved in myotonic dystrophy type 1 (DM1), a frequently occur­ ring autosomally-inherited, multisystemic disorder which affects skeletal muscle, heart, brain, eyes and the gastrointestinal tract (101). Disease manifestation in DM1 is caused by (CTG)n repeat expansion in exon 15 of the 3'-UTR of the DMPK gene (26, 86, 94), the length of which is related to disease severity. Evidence has been provided that expression of DMPK transcripts with long expanded (CUG)n repeats plays an important role in DM1 disease induction. Presence of DMPK transcripts from the mutant allele thereby causes abnormal intranuclear titration or mislocation of splice and transcription factors, with dominant trans-effects on the ex­ pression of products of many other genes (300). Up to now, the direct involvement of abnormal DMPK protein expres­ sion, which may explain DM1 features like muscle wasting, neurological problems or loss of endocrine function, has not received a lot of atten­ tion. As a result of alternative splicing, DMPK is expressed as a small set of protein isoforms. These isoforms all have serine/threonine protein kinase activity, but they differ in the absence/presence of an internal VSGGG-motif and in the length and sequence content of the C-terminus. In human and mouse, the DMPK C-terminal tail exists as three variants, the longest of which can serve as tail anchor for binding DMPK to either the mitochondrial outer membrane (MOM) or the ER membrane (95, 286). Moreover, all DMPK isoforms contain a coiled-coil segment involved in dimerization and association to other cellular structures (286, 287). Based on the linear sequence characteristics of their kinase domain, DMPKs are classified as members of the AGC (cAMP-dependent, cGMP- dependent and protein kinase C) subfamily of protein kinases. Structure- function homology analysis revealed that DMPK displays highest homol­ ogy with myotonic dystrophy kinase-related Cdc42-binding kinases a, ß, and y, Rho kinase a and ß, and citron kinase (297). Since these kinases all are involved in remodeling of the actin-myosin cytoskeleton (130, 170, 255), this suggests that also DMPK could play a role in cell dynamics and cell shape plasticity. Interestingly, DMPK appears to play an important role in cellular ion homeostasis as well, because two small transmembrane proteins that interact with ion-pumping P-type ATPases, phospholamban (PLN; (134)) and phospholemman (PLM or FXYD1; (200)), have been identi­ fied as targets. Moreover, DMPK may also phosphorylate the ß-subunit of the dihydropyridine receptor (DHPR) Ca2+ channel, which modulates Ca2+-release from the sarcoplasmic reticulum (SR) upon skeletal muscle depolarization (277). In keeping with these findings, overexpression of 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

99 100 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes T pouto i rqie fr hpn srolsi C2 taset and Ca2+ transients sarcoplasmic shaping for required is production ATP 1, 2, 9) aan onig o h sm direction. same the to pointing again 298), 226,(13, abnormal Ca2+ currents and Na+ channel openings in myotube cultures cultures myotube in openings Na+ channel and Ca2+ currents abnormal ciiy n meit vcnt o te els ot motn clim stores. calcium important most cell's the of vicinity immediate in activity n h mgiue f a+ ees ad ieis f eutk i intracel­ in re/uptake of kinetics C DMPK and of Ca2+ of influence release magnitude regulatory the Ca2+on differential a dis­ intracellular of demonstrating by aspects and tribution isoforms DMPK active and catalytically excitability muscle over control of 150). (56, level relaxation another forms therewith to mitochondrial dimension Ca2+-stimulated another ER-mi- Conversely, [Ca2+]c adds (81). it functional because control regional and important is structural coupling this tochondrial muscle, In (98). between Ca2+ efficient exchange organelles jux­ allowing these closely be tethers, can structural by mitochondria and taposed SR that evidence provided other of groups studies Elaborate (216). signaling [Ca2+]c and cium mobilization anchor­ mDMPK's because triggered was myo­ question this in mature in Ca2+ Interest affect handling tubes. differentially C) and A (DMPK forms revealed 298), (227, abnormalities conduction cardiac and myopathy etal (PM) membrane Na+ plasma Ca2+ 121). (13,and concentration depolarized sarcoplasmic a elevated and exhibit muscle potential patients skeletal DM1 cultured from and fibers cells muscle resting that observation the od edn t clua hprxiaiiy 21. rl fr MK n ion in DMPK with for compatible role Ca2+ also is A (211). homeostasis, particular in and hyperexcitability homeostasis, Ca2+ over­ cellular sarcoplasmic to triggers leading muscle load skeletal mouse in DMPK human ua soe ad f MK o te Ca2+ mitochondria. of the on loading A DMPK of and stores lular cal­ cytosolic to coupled to affects coupled appear is A that and hDMPK processes viability of autophagy, of cell and initiation fate overexpression distribution ectopic that mitochondrial both kinase reported its we places (286) Earlier membranes mitochondrial or SR either to ing skel­ mild display which mice, (KO) out knock DMPK of analysis Finally, ee w cnim h sgetd eainhp ewe peec of presence between relationship suggested the confirm we Here, iso­ DMPK long individual whether question the on focus we Here, • MATERIALS AND METHODS

• Cell culture Normal and DMPK-deficient immortalized myogenic cell lines were de­ rived from the calf muscle complex of mice (postnatal day 10) that harbored one H-2Kb-tsA58 allele and carried either two DMPK wild type (WT) or two DMPK KO alleles, following procedures as described (56, 197). Myoblasts were ring cloned and selected for myotube formation ability. Myoblasts were propagated at 33°C in gelatin-coated 6-wells plates in DMEM (GibcoBRL, Gaithersburg, MD) supplemented with 20% (v/v) FCS, 50 pg/ml gentamycin and 20 units IFN-y/ml. Terminal dif­ ferentiation with myotube formation was induced by placing a confluent myoblast culture grown on Matrigel (BD Biosciences, the Netherlands) at 37°C in DMEM supplemented with 5% horse serum (HS) and 50 pg/ ml gentamycin and by maintaining these conditions for up to five days.

• Plasmids and vectors E1/E3-deleted serotype 5 adenoviral vectors encoding YFP-mDMPK A, C, E and kinase dead mutants (K100A mutation (298)) under the control of a CMV IE-promoter were generated using the AdEasy Vector System (106), as described (41). In short, full-length cDNAs (including ORF and 3'-UTR) encoding EYFP-mDMPK A, C, E or kinase-dead mutant proteins were cloned as restriction fragments into the pShuttle-CMV transfer vector. In a second step, pShuttle-YFP-mDMPK plasmids were recombined with the viral DNA plasmid pAdEasy-1 in E. coli strain BJ5183. Next, viral par­ ticles were generated in N52.E6 (249) viral packaging cells (93). Finally, adenoviral vectors were purified using CsCl gradient purification and stored at -80 °C. The viral titer and plaque forming units were determined (41) prior to use in transduction experiments.

• Adenoviral transduction of immortalized myotubes For transient expression of YFP-mDMPK A, C and E, and kinase-dead mu­ tants, myotubes differentiated for 3 days, were transduced with adenovi- ral-vectors in DMEM for 3 hours at 37 °C as described previously (286). Three hours after addition of recombinant virus, medium was replaced with fresh DMEM supplemented with 2% HS and 50 pg/ml gentamycin and cultures were maintained for an additional 48 hours (total of five days of differentiation).

• Cytosolic and mitochondrial calcium measurements Myoblasts were seeded on Matrigel-coated glass coverslips. Terminal differentiation into myotubes was induced by placing them at 37°C on DMEM supplemented with 2% HS and 50 pg/ml gentamycin for a total 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

101 102 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes YFP was excited using a 488 nm laser, T1/T2A dichroic and 522DF35 522DF35 and dichroic T1/T2A laser, nm 488 a using excited was YFP KCl, NaH2PO4, mM 5 mM NaHCO3, 1 mM 10 NaCl, mM (125 solution (HT) ennal Te ehrad) qipd ih n ro/rpo lsr us­ laser, argon/krypton an with equipped Netherlands) The Veenendaal, pat­ banding sarcomeric visualize to used was (88) per- antibody were anti-titin Cells transduction. after hours 48 formaldehyde (w/v) 2% (AC h; acetylcholine pM 20 containing solution HT with superfused was 1.3 40x, a with equipped Germany) Jena, Zeiss, (Carl microscope verted slips coated with Matrigel (BD Biosciences, the Netherlands), washed washed Netherlands), the Biosciences, (BD Matrigel with coated slips microscopy Immunofluorescence • ir­ m dichroic 560DRLP a excitation, nm 540 using similarly visualized Optical, (Omega mirror dichroic 430DCLP a via directed was emission and Jena, Zeiss, (Carl System Cultivation POC-R Cell closed 7.4) a in pH glucose, placed mM and 10 and HEPES, mM CaCl2, 10 MgSO4, 1.8 mM mM 2 hoc n 55P msin itr Frrd a ectd sn a 4 nm 647 a using excited was di­ Far-red T1/T2A filter. laser, emission nm 568 585LP a and using chroic excited was TexasRed filter. emission Laboratories, (Biorad microscope laser-scanning confocal MRC1024 rad visualize to used was (Invitrogen) TexasRed Phalloidin-conjugated tern. rabbit A processed procedures. and standard P-40 using Nonidet microscopy (w/v) 0.5% immunofluorescence for containing PBS in meabilized containing PBS in fixed and (PBS) saline phosphate-buffered with once (Mo­ software 6.0 Metafluor with controlled was hardware microscopy Gräfelf­ Photonics, was IV; TILL YFP cells. (Polychrome without monochromator coverslip a using the onexcited and position a Fura-2 at Both fluorescence (Omega). the filter emission 565ALP a and (Omega) ror exposure/ an using Netherlands) The Vianen, Scientific, (Roper camera mM 0.5 absence added the and CaCl2 in without Ca2+ (Invitrogen) medium (i.e., HT ionomycin Ca2+ extracellular ionophore of the of pM 1 of in­ M 200 Axiovert an of stage the on placed was latter The Germany). (w/v) 0.025% of presence the in respectively, Invitrogen) ester; methyl ester; acetoxymethyl (Fura-2 AM Fura-2 pM 3 to exposed were ([Ca2+]m),myotubes concentration sarcoplasmic in calcium [Ca2+] changes free ([Ca2+]c) mitochondrial monitor To and transduction. including days five of ing a 60x 1.4 NA oil objective and LaserSharp2000 acquisition software. software. acquisition LaserSharp2000 and objective oil 1.4 NA 60x a ing application by investigated was Ca2+ content Sarcoplasmic Invitrogen). culture the Ca2+ release, sarcoplasmic trigger To objective. Fluar F NA sd o dtcin f irtbl. mgs ee band ih Bio­ a with obtained were was Images Iowa) of microtubuli. University of Bank, detection antibody for Hybridoma ß-tubulin used Studies monoclonal mouse (Developmental the E7 and microfilaments F-actin cover­ glass millimeter 24 on transfected and grown were Myotubes USA). Sunnyvale, Devices, lecular All filter. emission 535AF26 a and dichroic 505DRLPXR a Germany), ing, of subtraction by corrected background were signals emission Rhod-2 was Rhod-2 s. 1 of interval sampling a and ms 200 CCD of a time onto integration (Omega) filter emission 510WB40 a and USA) Brattleboro, light (Ca2+-unbound form) nm 380 excited and (Ca2+-bound nm form) alternately 340 using was (Polychrome Fura-2 Germany). monochromator Gräfelfing, a Photonics, by IV; TILL provided was light Excitation EGTA). HEPES-Tris colorless a with After washed medium. were culture in myotubes 37°C at adherent min loading, 30 for acetoxy­ (Invitrogen) (Rhodamine-2 Pluronic-F127 AM Rhod-2 pM 5 or USA) Eugene, Invitrogen, lasers T1/T2A dichroic and 680DF32 emission filter. To determine mito­ chondrial colocalization, myotubes were stained with 1 pM Mitotracker Red CM-XH2ROS (Invitrogen) for 30 min at 37°C and imaged at 37°C using the temperature-controlled stage of a Zeiss LSM510-Meta confo- cal microscope (Carl Zeiss) using the appropriate argon laser lines and a Plan Neofluor 40X/1.3 oil DIC objective. YFP was excited using a 488 nm laser, 545NFT dichroic and a BP500-530 emission filter. Mitotracker Red was excited using a 568 nm laser, 545NFT dichroic and LP560 nm emission filter

• Cell lysates, SDS-PAGE and Western blotting Myotubes were lysed on ice in NP40 lysis buffer containing 1% NP40, 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 25 mM NaF, 1 mM sodium pyro­ phosphate, 0.1 mM vanadate, 1 mM PMSF, 1x protease inhibitor cocktail (Roche, Mannheim, Germany) and 1 mM EDTA. Lysates were cleared by centrifugation for 10 min (14,000g) at 4°C. The obtained supernatant fractions were mixed with SDS sample buffer, separated on 8% SDS- PAGE gels and transferred to PVDF membrane (Millipore, USA). For im­ munodetection, DMPK-specific polyvalent antiserum B79 (95) was used as a primary antibody. HRP-conjugated goat anti-rabbit I g G was used as a secondary antibody (Jackson ImmunoResearch Laboratories, UK). For visualization of ECL signals were detected using a Kodak X-OMAT AR film.

• Electron microscopy For EM analysis, cells were fixed in 2% (w/v) glutaraldehyde in 0.1 M cacodylate buffer and postfixed for 1 h in 1% osmium tetroxide and 1% potassium ferrocyanide in 0.1 M cacodylate buffer. After being washed in buffer, cells were dehydrated in an ascending series of aqueous ethanol and subsequently transferred via a mixture of ethanol and Epon to pure Epon 812 embedding medium. Ultrathin gray sections (60-80 nm) were cut, contrasted with aqueous 3% uranyl acetate, rinsed, and counter­ stained with lead citrate, air dried and examined in a JEOL JEM1010 electron microscope (JEOL, Welwyn Garden City, UK) operating at 80 kV.

• HPLC analysis Samples were prepared using a perchloric acid extraction method de­ scribed previously (14). Briefly, myotubes and myoblasts (in triplicate) were washed three times in PBS and lysed in 1.4 ml of perchloric acid (Fluka, Germany). Suspensions were homogenized on ice, centrifuged at 1,200g at 4°C and supernatants were transferred to new Eppendorf tubes. Pellets were used for protein analysis with the Mico BCATM Protein Assay Kit (Pierce, USA) to correlate data to protein input. Supernatants were neutralized using KOH to reduce metabolite oxidation and remove ClO4- ions from the solution. After spinning down the potassium perchlorate crystals, supernatants were freeze-dried in a Speed Vac Concentrator (Savant, USA). Pellets were dissolved in milli-Q, filtered over a Millex-HV 0.45 pm filter and 50 pl of each sample was used for the HPLC analy­ sis as described previously (267) using a Shimadzu LC20 HPLC system (Shimadzu, Japan) with a Supelcosil LC-18-T, 15 cm x 4.6 mm (3 pm particles) column (Supelco, USA). 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

103 104 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes / (9) Uls sae ohrie dt fo mlil eprmns are experiments multiple from data otherwise, stated Unless (290). 1/A [Ca2+] ([Ca2+]c), sarcoplasmic rest­ free the resting the of measure a As te ea tm cntn () a cluae uig h euto: = K equation: the using calculated was (K) constant time decay the A a mono-exponential function (56, 150), R(t) = R(P)+R(t=0)e(-A/t), where where R(P)+R(t=0)e(-A/t), = R(t) 150), (56, function mono-exponential a in­ of rate the of measure a as used and cell each for calculated was MA). Northampton, 6.1 (Microcal, OriginPro and USA) CA, Jolla, La ware, statistics and analysis Data • a cluae fr ah el n ue a a esr o mxml sarco­ maximal of measure signal a as ratio used and fura-2 cell each ACh-induced for calculated maximal was The points. data stimulatory means of an unpaired two-tailed Student's T-test using a significance significance a using T-test by Student's performed two-tailed was unpaired analysis an of means Statistical s.e.m. ± averages as expressed From s). (in constant time the is A and s) (in time the is t signal, fura-2 by described adequately is sar­ in [Ca2+]c. decrease of decrease rate this of coplasmic the myotubes In measure a as used was signal ratio fura-2 ACh-induced [Ca2+]c. the of decrease of rate sarcoplasmic in The crease signal ratio fura-2 in increase [Ca2+]c. ACh-induced of plasmic rate The level to which R declines, R(t=0) is the maximal ACh/ionomycin-induced ACh/ionomycin-induced maximal the is R(t=0) declines, R post-stimulatory which the to is R(P) t, level timepoint at signal ratio fura-2 the is R(t) Soft­ 5.1 (GraphPad Prism GraphPad using analyzed was data Numerical ee o P<0.05. of P<0.01. P<0.05; * level ** pre­ 10 from cell each for calculated was signal ratio fura-2 average ing • RESULTS

• DMPK isoforms localize to different cellular compartments in myotubes To study how distinct long DMPK isoforms (A-D) affect sarcoplasmic Ca2+ handling we applied adenoviral transduction (i.e., complementation) of DMPK-knockout (KO) myoblasts/myotubes with individual YFP-tagged mouse DMPKs (mDMPKs). We chose DMPK KO cells because these pro­ vide a natural cellular context and lack the complex mixed background of endogenous DMPK isoforms (215). To allow proper interpretation of complementation experiments, mDMPKs were introduced into cells either as kinase active (normal) or inactive (Kd, kinase dead) variants. Both mDMPK A and A Kd displayed a reticular localization that did not overlap with mitochondria (Fig. 1A; top panels). In contrast, mDMPK C and C Kd showed a mitochondrial pattern in addition to cytosolic staining (Fig. 1A; middle panels). DMPK variants without the long C- terminal tail (mDMPK E and E Kd) displayed a cytosolic localization (Fig. 1 A; bottom panels). Thus, in complemented mature DMPK KO myotubes, mDMPKs with different C-termini display a different subcellular localiza­ tion, confirming findings in other cell types with a less highly organized infrastructure (215, 286). Of note, the mDMPK localization pattern did not depend on whether the protein was enzymatically active or inactive. SDS- PAGE analysis of complemented cells revealed that all mDMPK variants were expressed to similar levels and migrated at their expected size (~ 100 kDa; Fig. 1B and data not shown). Production of fusion proteins with an aberrant structure due to proteolytic processing was therefore not a concern in our experiments.

• Effects of mDMPK expression on cytosolic calcium handling and store filling In an earlier study, we analyzed myotubes that were formed by in vitro differentiation of primary myoblasts from WT or DMPK KO mice (13) and found that kinetics of ACh-induced sarcoplasmic Ca2+ ([Ca2+]c) transients differed between WT and KO myotubes. Here, with myotubes generated from immortalized myoblasts derived from WT/SV40Tagts or DMPK KO/ SV40Tagts doubly-transgenic mice, we corroborate these findings. DMPK absence or presence affected kinetics of ACh- or ionomycin (Iono)-in- duced Ca2+ transients differentially (Fig. 2). ACh was used to induce calcium release from the intracellular SR Ca2+ stores via the activation of voltage-gated Ca2+ channels, imitating muscle stimulation (in vitro). Use of Iono in Ca2+ free medium with EGTA enabled us to monitor effects of DMPK isoforms on the filling state of intracellular SR Ca2+ stores. Basal [Ca2+]c in KO myotubes was consistently, but not significantly, higher than basal [Ca2+]c in WT myotubes, both in medium with and with- 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

105 A Figure 1 • DMPK isoform expression and localization in myotubes. Confocal images were ac­ quired 48 hrs after adenovi­ ra l transduction of DMPK KC myotubes (at differentiation day three) with YFP-tagged mDMPK A/C/E isoforms. Cells were co-stained for mitochondrial marker Mito­ tra c k e r Red. (A) YFP-mDMPK A did not colocalize with Mitotracker Red and showec a reticular pattern, similar to YFP-mDMPK A Kd.Both YFP-mDMPK C and its Kd mutant localized at mito­ chondria, but showed also some cytosolic and nuclear staining. Arrows indicate clear colocalization betweer YFP-mDMPK C/C Kd and M ito tracke r Red. A cytosolic localization was adopted by YFP-mDMPK E and Kd fusion proteins. Bar, 25 pm. * , V (B) Western blot analysis of : Vei.' YFP-mDMPK fusion proteins expressed in DMPK KC myo­

m e rg e tubes using an anti-DMPK antibody. Fusion proteins run at the appropriate molecular , i t w eight of about 100 kDa. Note that YFP-mDMPK Kd isoform s do not show the "í» x ». . S': extra band associated with YFpsmDMPK C KO mita» m e rg e autophosphorylation (indi­ cated by asterisks) (296). Æ 4 t ? Æ Ê ê a f 1

'P-'1 ’r ^»mDMPK E Kd áte'' . . merge

aw ? a ,0? di' ha op sp Ly on ed ithi .S'0 O S °0r vt0 < # < # o'# o'# o ^ K,é J r <$r < f <$P <$P J r / ^ ^ / / / /

100 kOe *

out oaloium before stimulation (Fig. 2A, B, G and H). Furthermore, a clear correlation between basal [Ca2*]c and residu al YO [Ca2*]c I evels af t^r stimulcition sA/as nbseeved (Fig. 3A eind S). Tgls cttrrelaition ailss held for KO myoKuben comptemented w ith in dividual D Man isoforms (Fig. FA and Be and for mynt:udes w irhi [Sd nasonts, w hicA dsei[^in u lirdvrl ib bli/b ACh sti mul^gictn ex pen mest s osty (Fig. 3A). p0 mF0 Fat ivo comptementnüon w ith acSive snS Kd vatianys tifsa^ f)e rno( ttieO Po Oiseyimieete Oelwuen ttfs- siSSe efl /unhn oC D(eICi< o/ry tho caloiom-hon3üng machinery, which could

106 Figure 2 -Effects of DMPK expression on cytosolic A B 0.60- lono calcium handling and store • WT filling in myotubes. o KO Fura-2 AM-loaded WT and DMPK-KC myotubes differentiated for five days were stimulated with 20 pM acetylcholine (ACh) (A) or 1 pM ionom ycin (lono) (B) for about 60 s (b lac k bar). [Ca2+]c w as monitored by digital imaging microscopy 0 50100 150 200 under constant superfuso Time (s) D using Ca2+-containing (A) C 1.4­ or Ca2+-free m edium w ith 1.2 0.5 mM EGTA (B). Typical : 1.0 tran sien ts are shown. WT i 0.8 and KC myotubes showed an £ 0.4- f 0.6 altered [Ca2+]c handling w ith s lig h tly elevated basal [Ca2+] : 0.4

c le ve ls for KC m yotubes (A), 0.2 which was more pronounced 32 88 35 19 33 16 20 12 40 30 25 22 0.0 17_ in Ca2+-free m edium (B). WT KO mA mA mC mC mE mE mA/C WTmC mE mA/C DMPK-KC myotubes were Kd Kd Kd transduced with adenoviral f 160- vectors encoding active _ 140- or inactive YFP-mDMPK w - isoforms and then subjected Ñ 120­ to ACh (C, E, G) or lone E 100- (D, F, H) stim ulatio n. The "O§ 80- obtained curves were ana­ o 60 - lyzed and effects induced £ 40- by the expression of DMPK § 20 - isoforms were quantified. UL 0 The maximal peak amplitude WT KO mA mC mE mA/C (C, D), Ca2+ decay tim e Kd Kd Kd constant (E, F) and basal GH ■ Pre-Iono [Ca2+]c before stim ulatio n □ Y0 together w ith residual [Ca2+] c Y0 (G, H) were determined for ACh-stimulated (C, E, G) or lono-stimulated (D, F, H) myotubes. The num ber of ce lls used for all parameters is shown at the bottom of each bar in C and D. mA/C denotes ce lls expressing both YFP- WT KO mA mC mE mA/C Kd Kd Kd mDMPK A and C. WT indicate significance w ith respect to WT myotubes. be caused by either phosphorylation modification or by a mere structural association to protein targets. From these observations, we can draw one tentative conclusion: although expression of certain isoforms mildly affects the setting of balance between basal Ca2+ influx and extrusion, none of the DMPKs permanently compromises the ability of myotubes to restore this balance to their own resting-state value, after one round of stimulation. The maximal Ca2+ amplitude upon treatment with either ACh or lono did not markedly differ between cohorts of WT and KO myotubes and KO myotubes complemented with different isoforms. Differences in compari­ son to WT tubes appeared only marginally significant in two cases, for mDMPK A Kd-expressing myotubes with ACh stimulation and mDMPK A- expressing myotubes with lono stimulation (Fig. 2C and D). 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

107 108 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes C B A fe Ah ope o eaiey lw ea rts fe ln tetet in treatment lono after rates decay slow relatively to couple ACh after also were myotubes WT and C, Fig. A/C and D mDMPK for and 2C (Fig. results stimulation. lono Clustering 3C). upon cells heights the peak were lowest these and that in similarity clustered and showed doubly-complemented myotubes A/C WT mDMPK complemented, singly C mDMPK ment were plotted (Fig. 3D), suggesting that relatively fast decay rates rates decay fast relatively that suggesting treat­ 3D), (Fig. lono and plotted ACh were after ment Ca2* transients the of rates decay when found stimulation ACh upon heights peak ratio Fura-2 highest the showed that F E osiuul, orlto aayi rvae ta C2 rsoss in Ca2* responses that revealed analysis correlation Conspicuously, 0.80- 0.85­ 0.90­ 0.95­ 1.00­ 1.05­ 1.10­ 1.15­ 1.20-i 0.46- 0.48­ 0.50­ 0.52­ 0.54­ 0.56­ 0.58­ 0.60­ 0.62-, 0 4 8 2 6 0 34 30 26 22 18 14 10 ACh-induced fura-2 ratio decay (s) decay ratio fura-2 ACh-induced 0.95 1.00 1.05 1.10 1.15 1.20 1.20 1.15 1.10 1.05 1.00 0.95 ACh-induced peak fura-2 ratio fura-2 peak ACh-induced A m i-i i-i mA/C Basal fura-2 ratio fura-2 Basal mXljmCKd C m j l X m Í m 7 T O K D 0.40- 0.42­ 0.44­ 0.46­ 0.48­ 0.50­ 0.52­ 0.54­ 0.56­ 0.58­ 0.60­ 0.62-, 0.25 0.30 0.35 0.40 0.45 0.45 0.40 0.35 0.30 0.25 0 0 0 10 4 160 140 120 100 80 60 ooidcdfr- ai ea (s) decay ratio fura-2 Iono-induced ACh-induced fura-2 ratio decay (s) decay ratio fura-2 ACh-induced Iono O "?"/1 / " ""E? Basal fura-2 ratio fura-2 Basal Í-" ACh-induced m a xim a l peak peak l a xim a m ACh-induced lono- and ACh-induced lono-in- and ACh-induced and ACh- for [Ca2+]c Y0 l a fin ine determ to analysis tion paet sotd circles). (spotted apparent D­ m C, mDMPK of g clusterin constant: decay and e valu constant: decay and e valu values: peak l a xim a m duced ­ la rre co a in used ere w myotubes. stimulated Iono- and ACh- of analysis P AC n W vaues is s e alu v WT and A/C MPK clear A R2 P=0.06. = 0.79, between relationship r a e lin The (F) P=0.0008. R2 = 0.91, between relationship r a e lin The (E) R2 P=0.015. constants: = 0.90, decay induced between relationship r a e lin The (D) P=0.011. R2 = 0.91, between relationship r a e lin The (C) R2 P=0.017. = 0.89, lono: R2 P=0.015; = 0.77, ewe bsl [Ca2+]c and basal ined determ between as w linear The relationship B) (A, between eters. s param ship relation r a e lin lono or ACh r ithe e ­ u ith tim w s lls lated ce the loaded of Fura-2 data tal en xperim E Correlation • 3 Figure lono-induced m a xim a l peak peak l a xim a m lono-induced ACh: myotubes: Iono-treated cells that express mDMPK C (Fig. 2E and F). Finally, analysis of peak values versus decay rates of the Fura-2 signal ratios revealed that Ca2* release and Ca2* sequestration were correlated and occurred to highest values and fastest rates in the group of ACh-stimulated mDMPK C, A/C and WT myotubes (Fig. 3E). Analysis of lono stimulation also placed mDMPK C, A/C and WT myotubes into one cluster. However, here the rela­ tionship was reversed, as lowest maximal peak values appear correlated with slowest decay for this treatment (Fig. 3F).

• DMPK A expression alters mitochondrial calcium uptake To study whether the effects described above indeed could in part be explained by influence of single DMPK isoforms on mitochondrial Ca2* handling, [Ca2+]m was monitored using rhodamine-2 AM (Rhod-2) stain­ ing in WT, KO and complemented myotubes. After ACh stimulation, the [Ca2+]m increased at a similar rate for KO and WT myotubes, reaching a similar fluorescent signal Aincrease (Fig. 4A). In KO myotubes with m- DMPK A - but not A Kd or any of the other mDMPK isoforms - the maximal peak height in the Rhod-2 signal tracing was significantly lower than for all other myotubes types analyzed (Fig. 4B). The shape of the [Ca2+]m transient in cells that co-expressed mDMPK A and C had a similar ap­ pearance, again with significantly reduced peak height in comparison to WT myotubes (Fig. 4A and B). For clarity reasons, the spreading in traces for isoforms mDMPK C, C Kd, E and E Kd were not included in Fig. 4A

Figure 4 • DMPK isoform ex­ A pression alters mitochondrial Ca2+ transients. (A) Rhod-2-loaded WT and KO myotubes, differentiated for five days, were stimu­ lated w ith 20 pM ACh for 60 s. [Ca2+]m w as momtorec by digital imaging mi­ croscopy under constant superfusion. [Ca2+]m was calculated as the signal Aincrease compared tc basal level. DMPK KO myotubes transduced with adenoviral vectors encoding different active and inactive YFP-mDMPK isoforms were also used. Traces show average [Ca2+]m after super­ B imposing all appropriate ex­ periments. (B) The Aincrease maximal peak is displayed showing [Ca2+]m uptake. For myotubes transduced with DMPK A vector or for myotubes transduced with a m ixture of DMPK A and C WT* vectors, the maximal peak of [Ca2+]m is significantly reduced (com parison to WT). The number of cells tested is shown at the bottom of each bar. 19 8 15 10 16 WT KO mA mA mC mC mE mE mA/C

Kd Kd Kd 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

109 110 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes dt nt hw) Cmaio o te ae f C2] dcy i as not [Ca2+]m also of did decay rate the of myotubes Comparison different shown). not the (data between differences significant no revealed (81), olwn Ah tmlto (upeetr Fig. 2). (Supplementary directly myotubes, of stimulation ACh movements or­ offollowing contractile shifting by to due caused decay difficult, position the for ganelle notoriously tracings appeared linear of transients of rendering phase the be as should unreliable recordings the considered here However, differences. conspicuous reveal uptake calcium [Ca2*]c-induced mitochondrial high of period brief the kinet­ of Comparison isoforms. these for found was height peak maximal u ae hw i Splmnay i. , ic n sgiiat aito in variation significant no 1, since Fig. Supplementary in shown are but c o [a+m cuuain ie, ae f hd2 inl nrae during increase signal Rhod-2 of (i.e., rate [Ca2+]m of ics accumulation mDMPKA/ ci ii actin/titin E K P M titin D -m P F Y actin /C A K P M D -m P F Y mDMPKAttnactin/titin titin C K P M D -m P F Y A K P M D -m P F Y _ T W KO A % *s¿-, w W ■ actin ci -ttnactin/titin titin - actin ci ii actin/titin titin actin actin f ÆOk if w I I V V I ■' ii actin/titin titin ii actin/titin titin ' ' U ? « IF 9' ' '

c tubes were found. Z-bands Z-bands found. were tubes for co-stained were tubes using orphology m in tions were differentiation day of at s three isoform YFP-mDMPK Z-bands indicated by arrows. arrows. by myotubes. indicated ll a Z-bands in are found ere structures w eric sarcom of differences icros­ m obvious No copy. electron by lysis a an Bar, structures. eric sarcom Myo­ microscopy. confocal ­ ra lte a for differentiation of ( expression. DMPK by affected not is ogy as 1 pm. 1 Bars, yo­ m C-transduced PK YFP- mDM and KO WT, between visualize to titin and F-actin days five after investigated with transduced yotubes m morphol­ Myotube • 5 Figure 10 pm. (B-D) U ltra s tru c tu ra l l ra tu c tru s ltra U (B-D) pm. 10 a ) T oue ad KO and yotubes m WT 5 * Myotonie dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes 112 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes TEM confirmed formation of sarcomeric structures at regular appearance, appearance, regular at structures ofsarcomeric formation confirmed TEM A (Supplementary Fig. 3). Further analyses at the ultrastructural level with with level ultrastructural the at analyses Fig. 3). Further (Supplementary cytoarchitectural normal a revealed also staining Phalloidin 5A). (Fig. r lkl nt u t atrtos n yoacietr o myotubes. of cyto-architecture in alterations to due not likely are organelles visible clearly to next Z-lines, dense electron fibers with 4). Fig. Actin-myosin appeared and Supplementary and 5B-D transduced, KO, (Fig. DMPK for myotubes WT similarly malformations, apparent no with cultures between differences obvious no pres­ with of myotubes the independent of staining organization, sarcomeric distinct analyzed with myotube-types the structure of any of cross­ structures the in in sarcomeric abnormalities any differentiation striated of observe 4-7 not did day we 250), sarcomeric between 160, of (48, occurs vitro maturation usually of which indicator as structures, pattern banding F-actin and a plays DMPK that proposed been has it Ca2* of Since shaping transients. ofexpression or absence by the influenced not is development Myotube • D,AP n APcnetw on n sgiiat ifrne i metabo­ in differences the on significant no found effects GDP,we content ATP, ADP and differential had isoforms these whether investigated we state metabolic on expression DMPK Influence of • lbl nrei sae f ytbs Uig PC o te nlss f GTP, of analysis the for HPLC Using myotubes. of state energetic global respectively, A, and C DMPK active of presence with varied transients m [Ca2+]c over control dynam­ tight for Ca2+-ATPases of essential ATP is proper fueling (131). Conversely, of dehydrogenases TCA stimulation through machinery. excitation-contraction intact an indicating treatment, ACh to overall an gave staining Anti-tubulin isoforms. DMPK of absence or ence and infrastruc­ correlated be Ca2+ myotube in could changes the handling of characteristic and maturation ture and the organization of whether the investigate to out development in set we 87), (30, maturation parameter muscle in role important an reciprocally, also the but, of is fate, development system) the in muscle (e.g., as of development excitation-contractile regulator infrastructural important muscle an is Calcium Hence, this suggests that any differences in calcium handling observed observed handling calcium in differences any that respond to suggests this Hence, capable were organization. measured myotubes all ultrastructural virtually similar a Furthermore, show C all and A, mDMPK myotubes KO and WT, E-expressing that support data These mitochondria. like titin typical the Using myotubes. maturing in regulation coordinate under ie otn bten T K moue ad ytbs rndcd ih the with transduced myotubes and WT, KO myotubes between content lite [Ca2+]c both [Ca2+] of and profiles the Since 263). (133, myotubes in ics m¡t In 500 ° C 1500 = 1000 2000 niiul MK isoforms DMPK individual 0 2 < WT hnra [a*m ndrcl cnrl AP eeain y OXPHOS by generation ATP [Ca2*]m controls inchondr¡a, directly " < kg < mA < mC < mE A CW K m m m mA/C mE mC mA KO WT mA/C A B 600 400 200 0 YFP-mDMPK isoform s were were s isoform YFP-mDMPK a ls e eprment). experim per ples sam days. five for myotubes differentiated in concentrations eter­ d to used as w lysis a an myotubes. in concentrations nucleotide influence not does en ± e n 2 3-4 2, = (n sem ± Means GDP and GTP ADP, ATP, ine m HPLC B) (A, investigated. with transduced KC and yotubes m WT of lites etabo M expression DMPK • 6 Figure different DMPK isoforms (Fig. 6). Undifferentiated WT and KO myoblasts also did not differ in energetic state (Supplementary Fig. 5). 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

13 114 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes (EC-coupling) system (171) -at least to the maximal state of maturity maturity of state maximal the to least -at (171) system (EC-coupling) analysis of time the around levels expression protein keeping hours, 48 on anticipated as was myotubes in E) (mDMPK cytosol and C) (mDMPK paramount of is design cytoarchitectural its Furthermore 244). (215, hw) Tu, e xet o vi oestrto a sts t hc DMPK which at sites at oversaturation avoid to expect we Thus, shown). of days overexpres­ three local undergo of myotubes effects where protocol a circumvent adapted to (211). We sion important it considered also cells are Myotubes Ca2+ myotubes. in namely role, handling of this aspect n K-opeetd ytbs W as ntd iiaiy ewe the between similarity noted also We myotubes. KO-complemented and u tasue moue. ale, lo o paet auain differ­ in maturation apparent normal no near also is Earlier, vitro- in myotubes. differentiation transduced upon our achieved be can that acquisi­ findings in these Ca2+; us To shown). elicited not without data or contractions (with medium spontaneous tion of frequency and number similarly, proceed to appeared F-actin and titin striated of formation the abil­ in handling implicated calcium been turn, in and, has 87) (30, protein myogenesis of the because programming important DMPK of levels particularly near-normal consider We reside. normally would proteins not (data protein endogenous of that of offour-fold maximum a within to the of preference we location, natural ectopic of about concerns Besides locations isoforms. DMPK the major three reflect patterns assume these therefore We 286).that (215, types cell other in of findings basis the partitioning the that found we isoforms, DMPK of expression transient system. model our as myotubes, into ofdifferentiating KO capable DMPK myoblasts, immortalized ofSV40TAgts use adopted (62).We apparatus tion one on focused We isoforms. DMPK physi­ individual of the investigated significance we (286),ological cytosol- of and MOM -ER, control the and compartments architecture membrane organellar and cellular etal, • DISCUSSION ne ti rgm moue epes h YPmMK sfrs o only for isoforms YFP-mDMPK the express myotubes out. regime carried is this Under transduction adenoviral before differentiation, undisturbed Using now. until available are myotubes in distribution isoform DMPK expressed prominently most are gene DMPK the of products which in cytoskel- the of arrangement the in role putative a have proteins DMPK niae ht eeomn o fntoa ectto-otato coupling excitation-contraction functional a of development that indicate KO WT, immortalized for maturation of degree comparable a to leading with costamerogenesis, and Myofibrillogenesis maturation. of degree lar state. differentiation to coupled is myotubes in ity mitochondria A), (mDMPK ER over variants DMPK single of behavior contrac­ Ca2+-driven excitation the of functioning proper for importance different to target C-termini different with DMPKs As distribution. ion u srcua aaye cnimd ht ytbs ece a simi­ a reached myotubes that confirmed analyses structural Our u t lc o ioomseii DP atbde, o od aa on data good no antibodies, DMPK isoform-specific of lack to Due ences were observed between primary WT and KO myotubes (13) and an absence of gross morphological or structural alterations was reported for muscle of KO mice (227). Taken combined, we consider this strong evidence against a major role of DMPK isoforms in sarcomere assembly, muscle development and differentiation. Interference by DMPK isoforms with cytoarchitectural arrangement in myotubes is therefore only a minor concern for this study. Our observations with regard to increased basal [Ca2+]c and altered maximal [Ca2+]c peaks and decay constant of calcium transients in KO myotubes compared to WT control cells are largely reminiscent of find­ ings with myotubes derived from primary WT and KO myoblasts (13), albeit that differences upon stimulation with ACh appeared somewhat less pronounced. The quantitative, but not qualitative nature of these dif­ ferences can be most easily explained by cell-lineage and mouse-strain background differences. Comparison cannot be drawn to other findings for singly expressed DMPKs, because the experiments and methods used for correlation analysis presented here are unique to this study. In general, three types of phenotypic effects on calcium mobilization -i.e., selective effects on basal and induced cytosolic concentration, ER store filling, and intra-mitochondrial concentration- should be distin­ guished for individual DMPK isoforms. Presence of the catalytically active DMPK C isoform, which is a common feature of mDMPK C, mDMPK A/C and WT myotubes, rendered the release of Ca2+ from, and sequestering into, ACh-responsive stores more efficient. For Iono-released Ca 2+, active DMPK C seemed to have opposite effects, coupled to lower peak values at cytosolic entry and slower reuptake. Presence of DMPK A, as in DMPK A or DMPK A/C complemented cells, selectively affected peak uptake of Ca2+ in mitochondria after ACh stimulation. What could explain these findings and the differential effects of ACh and lono thereon? Various studies have indicated that the movement of Ca2+ among different com­ partments -cytosol, ER, mitochondria and extracellular space- depends on the nature of the agent used to deplete the stores, and on the cell type and manner in which stores are structurally arranged. ACh stimulation of myotubes with a fully-developed excitation-contraction coupling occurs via the natural way of triggering of voltage-gated Ca2+ channels, with subsequent calcium release at special calcium release sites in which SR, T-tubules and mitochondria are in close apposition (237). The process of lono-induced release of calcium in myotubes is much less well under­ stood, but the ability of this ionophore to induce general membrane per­ foration and stimulate the hydrolysis of phophoinositides evidently plays a role. For ER stores, even specialized subcompartments with distinctly regulated depletion and entry of calcium have been proposed (291). Our findings are thus best explained if we assume that presence of DMPK C -or DMPK A- has a selective modulating effect on the spatial ar­ rangement of these ER subcompartments, on the mitochondrial orienta­ tion with respect to the landscape of calcium release and uptake events, and/or the efficiency or positioning of release channels and Ca2+-ATPases in ER and mitochondrial membranes. What remains puzzling, however, is that DMPK C, which is anchored to the mitochondrial outer membrane (MOM), seems to exert its main effects on the kinetics of exit and (re)en- try of calcium in intracellular stores that are not mitochondria. Likewise, 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

115 116 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes ad -n te ptoeprl set teef ae edd I any In needed. are thereof- aspects spatiotemporal the -and C and A prnr f EC) r L (ate o N+ hne) poen i te cal­ the in proteins Na+ of channel), voltage- (partner PLN (ß-subunit), or SERCA) DHPR of the (partner pump, SERCA the like channels, and vial approaches. available hte te iuto i mc rpeet a sd-tp n vlto” and evolution” in "side-step a represents mice in situation the whether up is taken it before mitochondria, and SR between contact of close sites single complementation with DMPK A, or double complementation with with complementation double why or A, explain DMPK easily with cannot we complementation Furthermore, single mice. in important of only are localization mitochondrial the that assume to have therefore may we myotubes, remains a complicated technical challenge with the currently currently the with challenge technical complicated a remains myotubes, for targets potential as identified been have that network cium-handling PLM like proteins partner associating pumps Ca2+ or Na+gated channels, resident and/or selective of plasma-membrane or ER- possibility the of the on activity be of should modulation attention special study future yooi clim handling. calcium cytosolic ef­ clarify to done dramatic be to have needs may work More isoform behavior. A DMPK mitochondrial on human fects the the of beyond shown have expression [Ca2+]m. we on slightly that effects Earlier only dominate may although range, A, natural DMPK of overexpression ectopic A DMPK human because isoform, A DMPK mouse of the property peculiar in role at a cytosol play the into that events released on effects Ca2+ normally is mitochondria. selective in Ca2+ main entry its has ERmembrane the in anchored is which A, mDMPK [Ca2+]m that our indicate findings ihy yai cls ih dne n hgl odrd nrsrcue like infrastructure ordered highly and dense a in with cells certainly vivo, in dynamic highly regu­ of phosphorylation by analysis channels and Proper pumps ion of (297). lation vitro in phosphorylation DMPK-mediated o DP A ooous n te seis r lctd n ivle in involved and located are species other in homologues A DMPK how a be may ER-location that before reported We (98). mitochondria the by in KO myotubes, but results in distinct behavior. It is still possible that that possible still is It behavior. or distinct WT in in seen results as but [Ca2+]m myotubes, KOin mobilization mimic not does C, and A DMPK SR the via effects A DMPK and Ca2+ (286) in handling dominant is DMPK view, of point evolutionary an C. From DMPK like MOM, just the to binds ute suis n h aprn tasognla efcs f DMPKs of effects trans-organellar apparent the on studies Further • SUPPLEMENTARY FIGURES

Supplementary Figure 1 • A DMPK expression alters mito­ chondrial Ca2+ transients. (A,B) Rhod-2-loaded WT and KO myotubes, differenti­ ated for 5 days, were stimu­ lated w ith 20 pM ACh for 60 s. [Ca2+]m w as m onitorec by digital imaging mi­ croscopy under constant superfusion. [Ca2+]m was calculated as the signal Dincrease compared tc basal level. DMPK KO myotubes transduced with adenoviral vectors encoding different active and inactive YFP-mDMPK were also examined. Traces show the average [Ca2+]m ± sem after superimposing all appropri­ ate experiments. Time (s) B

160 WT KO 140 mC I T 120 mC Kd mE ^ 100 mE Kd 80

■S i 60 CM 40 a: 20

0 0 20 40 60 80 100 Time (s) 5 5 •M y o to nystrophy ic d protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

117 118 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over Ca2+ mobilization in myotubes Y F P -m D M P K E K P M D -m P F Y T W KO tions in m orphology using using orphology m in tions were s isoform YFP-mDMPK KO and yotubes m WT DMPK upon myotubes in tion localiza­ tubulin and Actin different m yotube cultures. cultures. yotube for m found different were s n lteratio a structure. eric sarcom and organization skeletal visualize yto c to lin u b F-actin tu for and co-stained Cells ere w microscopy. confocal • 3 Figure Supplementary thereby ACh, ith w d te la u tim s intensity fluorescence ere w e tim different five at ages stimulation. acetylcholine • 2 Figure Supplementary a, 0 pm. 10 Bar, orphological m apparent No ­ ra lte a for investigated with transduced yotubes m expression. isoform pm. 10 Bar, ination. determ intensity fluorescence pering ham when move m yotubes M indicated red. is measured is regior The shown. are points ove­ m affecting ment myotube of ple Exam upon movement Myotube C2] mesrmet. m­ Im ents. easurem m [Ca2+]m Supplementary Figure 4 • Myotube morphology is not affected by DMPK expression. (A-C) Ultrastructual analysis of myotubes using electron microscopy. No obvious differences between YFP- mDMPK A, C and A/C-trans­ duced myotubes were found. Z-bands of sarcom eric stru ctu re s w ere found in all myotubes. Z-bands indicated by arrows. Bars, 1 pm.

Supplementary Figure 5 • Nucleotide levels in WT and DMPK KO myoblasts. The energy-metabolite content of WT and KO myoblasts was determined.

(a , B) HPLC an alysis was used to determ ine ATP, ADP, GTP and GDP concentrations in m yoblasts. Means ± sem

(n = 2, 3-4 sam ples per WT KO WT KO experiment). 5 5 • Myotonic dystrophy protein kinase isoforms A and C exert differential control over mobilization Ca2+ in myotubes

119 i CHAPTER 6

SUMMARIZING DISCUSSION

• SUMMARIZING DISCUSSION

Myotonic dystrophy type 1 (DM1) is caused by the expansion of a (CTG) n triplet repeat in exon 15 of the DMPK gene (175). Since the repeat sequence is not situated in the protein-coding region of the gene, the length increase of the transcript does not result in production of a mu­ tant protein. Rather unconventional mechanisms must thus be involved in disease manifestation of DM1. Nowadays it has been commonly accepted that a dominant RNA pathogenesis mechanism -a process by which ab­ normal transcripts induce pathology- may be the main cause of the ma­ jority of symptoms (129, 300). The RNA toxicity mechanism has received increased attention since the discovery that a second form of myotonic dystrophy, DM type 2 (DM2), is caused by a (CCTG)n repeat expansion in intron 1 of the ZNF9 gene (225). Transcripts bearing an expanded (C/CTG) n repeat capture proteins involved in splicing, transcription or mRNA ex­ port and thereby induce misregulation of several genes in trans (65, 181). So is there no need to invoke DMPK protein products in our theo­ ry about the complex clinical manifestation of DM1? Based on current knowledge, we feel this possibility can still not be excluded. For example, expansion of the (CTG)n repeat may affect the rate of DMPK transcription or the mode and activity of alternative splicing or transport of DMPK transcripts in cis and thus directly influence the level of expression or the composition of the DMPK isoform repertoire. Any aberrancy in the expression of DMPK protein isoforms could have an additive role in one or more of the symptoms in the multisystemic pattern of DM1 pathol­ ogy. Only very few well-controlled in-depth studies have actually dealt with this matter (135, 297). Also still relatively little is known about the expression and biological function of alternative protein variants of the DMPK gene. This PhD study aimed to improve our level of knowledge on DMPK protein structure and function by investigating the localization and functional properties of different DMPK splice isoforms in more detail.

• DMPK splicing and isoform expression Primary transcripts of the DMPK gene in both man and mouse are subject to alternative splicing. Most DMPK splice modes are evolutionarily con­ served and result in the expression of multiple isoforms, six of which are considered major isoforms (95). One of the differences between isoforms is the presence or absence of a five-amino-acid VSGGG motif C-terminal of the kinase domain. The relative expression of VSGGG-encoding tran­ scripts slightly increases during myogenic differentiation (S. Mulders, PhD Thesis, Radboud University Nijmegen). For our studies, we decided to concentrate on the protein products encoded by these most abundant mRNAs (DMPK A, C and E), which are the isoforms with highest kinase 6 6 • Summarizing Discussior

123 124 6 • Summarizing Discussion (long) and cytosolic (short) isoforms. We found that the ratio between between ratio the that found We isoforms. (short) cytosolic and (long) has tissues patient DM1 and normal in expression DMPK about versy do­ protein N-terminal their expose characteristically proteins TA(18). el (296). cell protein recognized often studies early in used samples tissue Antibodies lacking. well-matched were and tools appropriate because partly ago, profile expression DMPK • TA only and first the is DMPK 223). (21, proteins TA for targets potential proteins C DMPK (286). MOM the at localizes human from A DMPK and (296). activity tay tt cnetain o ln ad hr DP ioom vre be­ varies isoforms DMPK short and long of concentrations state steady and intratissue specificity, tis­ cell-type and on expression information of Although technique isoforms. DMPK ratios this (E-F) thewith short of and (A-D) long picture of distribution reliable sue a obtain to core ipa dfeeta sbtae pcfct a te ifrn lctos n the in locations different the at specificity substrate differential development display during protein DMPK for requirements functional 2). suggests This different (Chapter myogenesis during changes and tissues tween membrane-associated between discrimination easy enables and reliable approach blotting Western a chose we Here, studied. well not were els exist. not still therefore and results) difficult are unpublished al. et isoforms DMPK (Wansink make some to in against present antibodies selectively Unfortunately, domains (297). protein kDa 68-72 of weight predicted a has DMPK molecular whereas kDa, 42-54 from size in ranging products decades two about described first was gene DMPK the since ever existed the on knowledge detailed obtain to important is it pathology, complex for specificity and below). 3; see (Chapter im­ requirements binding domain targeting membrane of C-terminal their in understanding our variation proved sequence artificial and natural are membrane plasma and membrane nuclear the including membranes acids amino hydrophobic of segment a contain and cytosol the into main proteins (TA) tail-anchored of group the to belong A-D isoforms DMPK MOM. the to bind both mouse and human from membrane ER the at localizes mouse from A DMPK isoform. A DMPK DMPK the and membranes in E, intracellular to and C bind A, DMPK C to typically and A DMPK properties that association membrane different fers u ptenn i te MK pie sfr poie o ata poen lev­ protein actual or profiles isoform (244), DM1 splice in DMPK the pathology in show patterning that but tissues of range a in expression RNA DMl's to expression DMPK altered of contribution a identify to order In date. to literature in reported kinase in different tissues, because the long and short DMPK isoforms probably probably isoforms DMPK short and long the because tissues, different in more quantitatively is blotting Western lost, is patterning intracellular protein DMPK the for specific antiserum polyclonal high-affinity a using contro­ A tissues. and types cell different in pattern expression DMPK intracellular Various membrane. a in inserts which terminus, C the in for occurs targeting membrane Species-specific cytosol. in the E remains lentv slcn, y hc dfeet -emn ae eeae, con­ generated, are C-termini different which by splicing, Alternative Study of the organelle-selective membrane behavior of DMPKs with with DMPKs of behavior membrane organelle-selective the of Study ada ivleet n M hs en orltd ih oh (CTG) both with correlated been has DM1 in involvement Cardiac n iu yrdzto aayi o mue ise dmntae DMPK demonstrated tissues mouse on analysis hybridization situ In long membranes, to association C-terminal typical the of Because n repeat expansion and DMPK gene product dosage (16, 101, 211). We found that long and short isoforms exist in about equal concentrations in cardiac tissue and HL-1 heart cells in culture. If this suggests that tight regulation has cardio-physiological relevance, aberrant expression levels of splice isoforms as may occur in DM1, could possibly contribute to cardio pathology. Weakness and wasting of distal skeletal muscles and myotonia are other hallmarks of DM1 (101), where mostly type I (slow twitch) fibers are affected (289). Several of these disease features are also found in DMPK KO mice and Tg overexpressor mice, again suggesting that tight control of DMPK protein expression may be necessary to prevent disease development (126, 211). We found that in skeletal muscle tissue long DMPK isoforms predominate, being expressed about three-fold more than short isoforms. Strikingly, DMPK protein was expressed at comparable levels in both type I and II muscles: in soleus muscle (predominantly type I) only slightly higher amounts of DMPK were found than in tibialis anterior (mainly type II) and gastrocnemius (mixed type I and II muscle). These observations corroborate findings reported in an ISH study (244). Comparison of expression levels of DMPK protein or RNA between muscle types, therefore, gives not a simple explanation for the predisposition to type I muscle pathology in DM1. Gastrointestinal disturbances might be related to DMPK expression levels in smooth muscle tissue, although no direct evidence for DMPK protein involvement exists up to now, as no apparent smooth muscle abnormalities have been reported in KO mice (298). Smooth muscle involvement in DM1 is mainly characterized by malfunctioning of the gastrointestinal tract, inducing symptoms like abdominal pain, vomiting, nausea and diarrhea (101, 198). DMPK is also expressed in layers of the ileum (244), in stomach and bladder (215). Other than vacuolization in bladder, no obvious cellular pathology has been found in smooth muscle tissue of DM1 patients (101). DMPK protein in stomach and bladder was predominantly made up of short DMPK isoforms, pointing to a distinct smooth muscle cell type-specific role of these cytosolic isoforms. Short DMPK isoforms E and F can be regarded as truncated relatives of MRCK and ROCK (297), proteins involved in cell motility. Therefore, se­ lective knockdown or overexpression of DMPK E-F in smooth muscle cell lines in cell migration- or cell spreading assays could help increasing our understanding of their normal role in smooth muscle biology. Pathology of the central nervous system in DM1 includes symp­ toms like mental retardation, character changes, apathy, cognitive and emotional disturbances and hypersomnia (101, 193, 235). Brain-related symptoms are most severe in patients with the congenital form of DM1 and can lay a great burden on family and carers. At a structural level, brain abnormalities like progressive white matter loss, cortical brain atrophy and ventricular dilation have been described in DM1 patients (193). In one study using DMPK KO mice, absence of DMPK was linked to neuronal plasticity defects (252). In a second, however, no cognitive and behavioral abnormalities could be detected in DMPK-deficient mice (190). Using transgenic mice expressing multiple copies of the human DMPK gene (211), we were able to demonstrate expression of both long and short DMPK isoforms in many brain regions including brain stem, ol- 6 6 • Summarizing Discussion

125 126 6 • Summarizing Discussion A rtis te yrpoi sqec i te -emns f og DMPK long of C-terminus the in sequence hydrophobic the proteins, TA organelle and insertion membrane govern that principles molecular The srcts hn n ern, t es i sm-ue el ouain kept populations cell semi-pure in least at neurons, in than astrocytes tece ad erae fiiny f ebae neto (0) These (109). insertion membrane of efficiency decrease and stretches 286). We used interspecies differences and isoform variation of the tail tail the of variation isoform and differences interspecies used We 286). other in Like understood. poorly rather still are proteins ofTA selectivity anchoring membrane and localization DMPKsubcellular • longer no can DM1 in symptoms brain-related the in ofastrocytes role a types major all in transcripts DMPK mutant and normal of fate of study aggregates RNP abnormal of presence show that appeared have studies whether question (72). The tissue neonatal to compared cerebellum and ne f cie-ol oan 26.Ti rgo my id iety o the to directly bind may region This (296). domain coiled-coil a of ence oan n MK t ivsiae ebae eetvt ad oe f mem­ of mode and selectivity membrane investigate to DMPKs in domain are astrocytes of populations certain As patients. DM1 of cells brain of (CUG)n-repeat containing foci, scaffolding ribonuclear of neuroanatomical presence now, supportive Until a anymore role. not just and with functioning, cells brain as of considered complexity enormous the in origin role neural of types cell all in protein- and RNA DMPK of -both role ofthe products gene investigation detailed pri­ more ona that findings indicate our RT-PCR. by cells Still, mary transcripts DMPK of levels low tect brain of stages later at study. DMPK for further needs role now therefore important development more a implies this RT- confirming development, during WT in increased isoforms DMPK expression study to DMPK used bemice. to had methods sensitive immunoprecipitation expression, low like the to Due striatum. and bulb factory eius a atrte retto o mmrn aco (7 28 result­ (97, 288) anchor membrane ofa orientation the alter can residues ofa-helical structure regular the in breaks cause can anchor membrane an play can region hydrophobic the flanking residues basic domain, coil orientation. this display to described protein TA MOM-localized is first C the DMPK Nout-Cout knowledge, a best DMPK For observed our composition. To topology. we C, lipid cytoarchi- in membrane differences differences anticipated or dependent tecture cell-type by affected or determined n utr. oee, s a aray la fo peiu suis (129), de­ could we studies and neurons in previous silenced from completely clear not is expression already DMPK was as However, culture. in rn acoig n oe eal Catr 3). (Chapter detail more in anchoring brane (18, signal targeting and anchor membrane a as both serves isoforms excluded. be (194) disease Alzheimer's and Parkinson's like diseases brain in involved comparative recommend the highly on we However, here patients. of reported tissue findings our brain of basis postmortem in astrocytes in No neurons. for reported been only has MBNL1, like factors splice bound distinct a with cells as seen increasingly are Astrocytes justified. fully is cortex brain mouse adult in expression higher showed that results PCR ing in embedding of the membrane anchor in the lipid bilayer rather than rather bilayer lipid the in anchor membrane of the embedding ining the in residues proline Furthermore, anchoring. tail in role the of important contribution the Besides proteins. or membrane (298) other homo-oligomers with of interact formation pres­ the on induce (236), dependent is membrane Clipid DMPK of anchoring that reported we Earlier, hgl iprat idn i ta DP poen s oe bnat in abundant more is protein DMPK that is finding important highly A e on ta te ebae oain f A MK sfrs s not is isoforms DMPK TA of location membrane the that found We spanning it. Currently, we think that it is the combination of the above sequence features that determines the topology of DMPK C binding. To know all details, more biophysical work on this topic is necessary. We investigated specific requirements for ER and MOM anchoring by analyzing effects of sequence differences between the mouse and hu­ man form of DMPK A. We found that the length of the hydrophobic region was more important than the presence of positively charged residues C-terminal of the hydrophobic region. Indeed, it has been reported that membrane width can influence localization of TA proteins in the secre­ tory pathway, suggesting that the length of the hydrophobic region may become important in selecting the preferred membrane site for anchor­ ing (236). Based on this idea, we were somewhat surprised to find that extending the hydrophobic region of mDMPK A did not result in altered organelle localization. Although both the mouse and human DMPK A isoforms display an Nout-Cin topology, we can only speculate about the exact membrane as­ sociation of the tail anchor of the human DMPK A isoform. Based on the length of the hydrophobic elements in the C terminus, hydrophobic region 2 (HR2) would be able to span the MOM lipid bilayer. As HR1 is probably too short to act as a transmembrane domain (302), it may be embedded in the outer leaflet of the MOM (292). Increased MOM as­ sociation of mutant hDMPK A(L593R) suggests that HR1 is able to bind to negatively charged lipids in the MOM. Introduction of a basic residue in the membrane anchor is probably tolerated because there are sufficient hydrophobic residues surrounding it (109). The curvature of membranes can recruit amphipathic helices present in the tail anchor of TA proteins (104). It has been postulated that protein complexes at the MOM can induce membrane curvature by the transfer of specific lipids (e.g., phosphatidylserine, phosphotidylethanolamine). In turn, these events can contribute to the recruitment of TA proteins, like h Fis, for initiation of autophagosome formation (97) or possibly pro­ motion of mitochondrial fission or fusion. Curved membranes can thus operate as scaffolds for the recruitment of specific proteins. Whether membrane curvature is important in DMPK membrane anchoring remains to be determined. Translocation of TA proteins to their target membrane can occur un­ assisted or assisted and is mediated by specialized protein complexes, being either ATP dependent or independent. Proteins assisting in MOM targeting have not yet been described, however it is known that an­ choring of several MOM-targeted TA proteins is independent of the TOM machinery (18). Insertion of TA proteins into the ER membrane has been studied more elaborately and resulted in the discovery of several fac­ tors that may be involved, like the signal recognition particle (SRP), the Hsp40-Hsc70 complex, the Asna-1/TRC40 polypeptide (223) and the Bat3 complex (185). It has been postulated that TA proteins with a highly hydrophobic membrane anchor are more prone to aggregate and thus require fast chaperone-regulated translocation. For example, increasing the hydrophobicity of the cytochrome b5 membrane anchor rendered the protein unable to independently insert into its target membrane (18). Since tail 1 of DMPK A is very hydrophobic, we speculate that it indeed requires chaperones for proper targeting and membrane insertion. Fur- 6 6 • Summarizing Discussion

127 128 6 • Summarizing Discussion 3, 4) Te lncl ypos f M rsml D1 ypos and symptoms DM1 resemble CMT of symptoms clinical The 148). (38, in resulting effects, in similar result may coupling this in disbalance Any several only after even phenomenon, this induce to sufficient is alone 1 as r irevrnet fr iohnra itrcin wt ohr cel­ other with interactions mitochondrial for microenvironments or ways after autophagy of level the in increase an find we of As (97). sites MOM, autophagy the At interactions. lipid-lipid or lipid-protein in rations mut o DP, n mgt pclt ta -s cue fatrd DMPK ofaltered cause a -as that speculate might one DMPK, of amounts determining are enzymatic cytoskeleton the of domain, status the coiled-coil nor a protein of of the presence activity the neither as clustering, induce to able are proteins membrane in regions hydrophobic certain expression DMPKA by behavior clustering Membrane • pathology in Alzheimer's and because these cells express substantial substantial express cells these because and brain to linked Alzheimer's in are astrocytes pathology Since (294). morphology mitochondrial inducedof Alzheimer's malformations Also (101). DM1 mitochondrial with with associated been misdiagnosis has of disease history a has even diseases (CMT) 2A human subtype to Charcot-Marie-Tooth and linked is Atrophy Optic Dominant mitochondria, like of mor­ clustering mitochondrial like Abnormal (196).phology, cycle be cell can of appearance stages change and different mitochondria during heterogeneous cell, single functionally and a within morphologically Even (172). appearance cal 3). (Chapter clustering A(L593R) hDMPK reduced to ofmutant expression as rise gave clustering mitochondrial the A, hDMPK of case the In clustering. and fragmentation (80).mitochondrial proteins motor possibly machineries, or structures cellular distinct are proteins these that speculated be could it Therefore structures. path­ lular separate in involved are proteins MOM all if as appears It factors. (234). 2 and 1 Mfn and (80), GTPases like Miro (123), proteins hFis MOM other ef­ of proteins TA the clustering overexpression for mitochondrial reported been 1. have Similar fects tail A hDMPK of expression upon (42). membrane mitophagy of mitochondrial initiator of loss known in a potential, that results fact also the by A hDMPK of supported is expression idea MOM. This the at autophagosomes of like processes specific of sites initiation be can curvature aber­ membrane indicating mitochondria, round and small decorates A demonstrate hDMPK data EM 231). (45,that Our remodeling and curvature membrane oftail expression as itself, domain tail the in localized is capacity tering therefore would chaperones discovered be iso­ to DMPK newly or other known and this with of forms interaction possible the of investigation ther hDMPK A might induce membrane curvature and subsequent formation formation of subsequent and expression ectopic curvature that membrane induce speculate to might A like hDMPK we expression, A hDMPK by amyloid precursor protein via modification of Drp1, a known mediator mediator known Drp1, a of modification via protein precursor amyloid by of determinant main the is anchor membrane the of hydrophobicity high and mitochondria the of areas distinct between links forging in involved mitochondrial induce to able are proteins TA these how clear not is It that indicating evidence mounting is There expression. transient of hours clus­ membrane of (216). The expression mitochondria of the that clustering to leads demonstrated A have hDMPK we 4) (Chapter thesis this In interest. great ofbe h pyilgcl tt o mtcodi i culd o morphologi­ to coupled is mitochondria of state physiological The Mitochondria started to round up and cluster around the nucleus nucleus the around cluster and up round to started Mitochondria expression- astrocyte mitochondrial morphology or function could be­ come aberrant in DM1, possibly leading to astrocyte cell death and white matter abnormalities in patients. Perhaps similar arguments can be used for muscle pathology, as long DMPKs like the A and C isoforms are mainly expressed in this tissue. Muscle cells have a highly organized cytoarchitecture, which may render them vulnerable to structural-organellar disorganization. Thus, we an­ ticipate that also in muscle the concentration and the ratio of expression of long hDMPK A and C isoforms must be maintained within a strict range. Speculative arguments in favor of a role for mitochondrial abnormalities in connection to DMPK misexpression are even further strengthened by findings in the TG mouse that carries multiple copies of the hDMPK gene. Tissue analysis of this model showed an aberrant mitochondrial mor­ phology with disorganized cristae structure and symptoms like reduced workload tolerance, atrophy and cardiomyopathy (211). Whether the role of DMPK misexpression, in particular hDMPK A overexpression, is indeed special for DM is a topic for further study in this mouse model. Ultimately, we w ill also need studies of DMPK isoform levels and ratios in DM pa­ tients with different repeat lengths and comparison to healthy controls to shed more light on the contribution of the DMPK mitochondrial axis to DM1 pathogenesis.

• DMPK function Still not much is known about the physiological role of DMPK. Especially the role of the separate isoforms remains to be investigated. Several studies have shown that DMPK is localized at the neuromuscular junction or the intercalated disk and gap junctions (135). Furthermore, altered Ca2* handling was found in mytubes derived from DMPK KO mice, compared to WT myotubes, resembling Ca2* handling in myotubes derived from DM1 patients (13), indicating a role in ion homeostasis in muscle. We have in­ vestigated the contribution of individual DMPK isoforms on Ca2* handling in myotubes (Chapter 5). Similar results were found for myotubes with (the MOM-localized) enzymatically active DMPK C isoform (WT, mDMPK C, coexpression of mDMPK A and C). Upon acetylcholine stimulation these myotubes were able to release Ca2* from - and sequester Ca2* into - ACh- responsive stores most efficiently. The opposite effect was found for han­ dling of Ca2* released by ionomycin stimulation. Myotubes expressing the mDMPK A isoform (mDMPK A, coexpression of mDMPK A and C) showed a decreased ability to take up Ca2* into mitochondria. Remarkably, WT myotubes and DMPK KO myotubes, showed similar behavior in mitochon­ drial Ca2* handling, so mDMPK A expression must be apparently care­ fully dosed to avoid dominant effects with reduced mitochondrial uptake. Another unexplained and somewhat surprising finding was that mDMPK C (located at the MOM) appears most important in the regulation of Ca2* release and uptake from the sarcoplasmic reticulum (SR; a special type of ER in muscle cells), whereas mDMPK A (located at the SR) has its most dominant effects on mitochondrial Ca2* uptake. The cytoarchitectural arrangement of the excitation-contraction ap­ paratus and the positioning of organelles in myotubes is of great im­ portance regarding myotube functioning. As Ca2* is normally released at areas of close contact between SR and mitochondria, the membrane 6 6 • Summarizing Discussion

129 130 6 • Summarizing Discussion 22 22 23 sol b dvlpd ih ae n nt otiue o a to contribute not and care with developed be should 293) 212, (202, decreases than rather increases knowledge new This control. viability contact SR-MOM initiate to proposed been has also VDAC Since (297). ufcec o DP ioom ai aernis cnrbt t D1 pathol­ DM1 to contribute aberrancies, ratio isoform DMPK or sufficiency gene DMPK the possibly including genes, several of expression gene of Concluding remarks • DMPK of role indirect or direct possible a include also thus should work targets. DMPK potential phosphorylation the these of addressing state required are experiments additional cells phospholem- of phosphorylation DMPK Finally, Ca2* in handling. variation and DHPR the of b3-subunit the VDAC-1, in identified phosphoacceptor were consensus DMPK sequences lab our in performed study previous a and effective treatment w ill become available and a patient's experience experience improve. patient's a may and DM1 with available safe become then, ill Only w products. treatment protein effective DMPK and of expression the in disbalance peutic strategy for DM1 based on breakdown of expanded DMPK mRNA mRNA DMPK expanded of breakdown on based DM1 for activity strategy protein peutic DMPK in disbalance a of involvement of probability the long individual of role the that evidence provide I study, PhD this Inogy? Future (281). patients DM1 in reported been study, have PhD this levels in ROS investigated elevated not was Ca2* of me­ controls transients control (105) oxidant-stress ofdiated STIM1 involvement protein Ca2* Although handling. oxidant-stress ER-resident mitochondrial the Interestingly, of (51). been ROS have regulation to dependent receptor susceptible be ryanodine to the and described SERCA of Both available with residues. activity reacting the by cysteine Ca2* decrease in homeostasis or involved increase proteins to certain Reactive able cell. the of are (ROS) state redox species the oxygen via affected be also can transients muscle in DMPK of role the understand fully to order In (297). interac­ Ca2* of DHPR through influx the ions via the myotubes increasing of state thereby Na* channels, with depolarization tion the alter could (PLN) man to contribute and MOM and SR between con­ sites could C of contact mDMPK tightness bythe VDAC of trol phosphorylation altered (51), formation Ca2* in homeostasis involved proteins all receptor, ryanodine cardiac the in Furthermore, (DHPR). receptor dihydropyridine of the ß-subunit the or at proteins on activity its exert to it enable would C mDMPK of location itself. Could this cis-acting mechanism, potentially resulting in haploin- haploin- in resulting potentially mechanism, cis-acting this Could itself. misregulation to leading (CTG)n repeat a of expansion by caused is DM1 control. stress oxidant mitochondrial-mediated on isoforms n set o D1 ahlg. s cneune ay oeua thera­ molecular any consequence, a As pathology. DM1 of aspects in cell to and Ca2* handling to biology, mitochondrial and may ER behavior- to linked anchoring be membrane their on -depending isoforms DMPK (SERCA) Ca2* SR ATPase the of inhibitor an phospholamban for like proposed been haveDMPK, substrates potential SR. or Several MOM the both et o hshrlto, euain f a* adig bhvo o Ca2* of behavior Ca2* of handling, regulation phosphorylation, to Next 6 * Summarizing Discussion 132 6 • References 1. Abell, B. M., M. R. Pool, O. Schlenker, I. Sinning, and S. High. 2004. Signal recognition recognition Signal 2004. High. S. and Sinning, I. Schlenker, O. Pool, R. M. M., B. Abell, 1. • REFERENCES 9 Bree N, . a ilsa P Sfinii M Ybl ad . aao. 03 Boeei of Biogenesis 2003. Makarow. M. pro­ and Yabal, M. tail-anchored guide Soffientini, P. ils ta How billasca, kinases ram B 2007. S. N., protein Colombo. Borgese, S. Ser/Thr of and 19. specificity billasca, ram B S. N., nalling Sig Borgese, 2003. al. et 18. Nebreda. R. A. Gehrmann, and M., J. R. iller, M Biondi, C. and al rm no 17. Greenwood, J. in Aronovitz, regulation J. M. Na+ Ca2+ and Maguire, C. T. I., C. myocytes Berul, Cardiac xtrac­ 16. 2006. E 1 Despa. 994. S. and Ferrari. M., R. D. and Bers, Curello, S. 15. Pedersini, P. 1 Cargnoni, 997. A. Ceconi, C. ieringa. P., W B. bearing and Bernocchi, its and Veerkamp, H. 14. J. itochondrion m Oerlemans, T. F. the of Groenen, re P. J. tu A., c A. tru ltras U Benders, 2008. 13. Rossignol. R. Ex­ and G., 2005. al. Benard, et Benzi, L. 12. Benedetti, A. P.87-205. 32:1 Simi, S. Brain Carratore, atrophica. Del R. yotonica M 1 Barsanti, 909. C. P., Gibb. P. Beffy, H. and E., F. 11. Batten, 10. is dystrophy Myotonic 1999. Brook. D. J. and Hamshere, G. M. Newman, E. M., azzan, lw A 4. . bl, . . C Rb, . enci J C Yug ad . ih 20. s-tasainl nte­ in translational ost- P 2007. High. S. and Young, C. J. Leznicki, P. Rabu, C. M., B. Abell, 2. 1 Bree N, n E Fsn. 00 Treig ah y o Cti-nhrd rtis Bio­ proteins. C-tail-anchored ­ of com ays pathw proteins: Targeting tail-anchored 2010. of le Fasana. ta E. The and N., 2003. Borgese, Pedrazzini. E. 21. and Colombo, S. N., Borgese, 20. al. et Leber, B. Penn, Z. L. Cruz-Aguado, A. J. Dlugosz, P. J. Soucie, L. G., E. M. Annis, 7. resi­ proline conserved A 2002. Lithgow. T. and Beilharz, T. Gabriel, K. Egan, B. R., Allen, 3. 9. Bassez, G., E. Chapoy, S. Bastuji-Garin, H. Radvanyi-Hoffman, F. J. Authier, J. F. Pellissier, Pellissier, F. J. Authier, J. F. al. Radvanyi-Hoffman, H. et Neumeyer, Bastuji-Garin, M. S. A. Chapoy, E. Potvin, G., A. Bassez, Beaulieu, C. 9. Stringer, L. J. Iyer, D. A., , anyam and Balasubram shape rial 8. itochond m between relationship The The 2006. 1 999. Scorrano. L. Kaibuchi. K. and V., and Anesti, Matsuura, Y. Kaneko, 6. T. ura, Nakam N. Chihara, K. M., Amano, 5. hm Bohs Acta. Biophys chim 9. MK oae trtons r ul i ati rcul cnuto a maii n a in s alitie rm o n ab conduction r la u tric n e v trio a in lt su re s n lteratio a dosage DMPK 1999. 10:1313-42. Signal Redox cytoskeletor in Antioxid protein the bioenergetics. of and role function on possible antibody: novel a ith w ined determ lls ce 2 type of bination com the by predicted be can the in dystrophy kinase yotonic m protein 2 Type dystrophy 2008. al. myotonic et of expression in changes ental m evelop D 1998. expandec the to adjacent le lle a DMWD the from RNA of l e v le reduced a ith w associated gration of tail-anchored proteins is facilitated by defined m o le c u lar chaperones. J Cell Cell J chaperones. lar u c le o m defined by facilitated is proteins tail-anchored of gration alacoe poen. ice Sc rn 31:1238-42. Trans Soc Biochem 1 9:368-75. Biol proteins. Cell Opin tail-anchored Curr -13. 372:1 J destinations. ir Biochem e th to interactions. teins ediated docking-site-m through tissue 1 15-22. 00:3 cardiac Sci in acol harm P nucleotides J hearts. pyridine g and failin purine, phosphate, creatine of assay and tion 25. branes em m eabilize erm p to erize m oligo 03. that ers 24:2096-1 onom J m EMBO anning 1757:692-9. apoptosis. ultisp m Acta during s rm fo Biophys Bax Biochim 2005. cytoskeleton. the and function 18-32424. 274:324 tail-anchored other and Tom7 of ain dom spanning brane- em transm the in present is due atce dae ps-rnltoa treig i ekroe. MO 23:2755-64. J EMBO eukaryotes. in targeting post-translational ediates m particle Sci 1 20:1 743-5 1. 1743-5 20:1 Sci os mooi dsrpy oe. Ci Ivs 103:R1-7. Invest Clin J model. dystrophy myotonic mouse muscle l 29:742-53. ta le Int e k s Biol an hum Cell in kinase differentiation. protein during ents dystrophy em ­ 9 yotonic 1 m rearrang :3 7 of 6 n eurol N calizatio Exp lo and europathol N pression J atrophy. scattered and nucleation tral n e c er fib 394:309-25. uscle m eurol N Comp J system. nervous central rat 8:1491-7. Genet Mol Hum 51 4:347-50. Lett repeat. FEBS translocase. TOM the of subunits protein n fo h ctsl n loig o a e rn. Cl Bo 161:1013-9. Biol Cell J brane. mem a for looking and cytosol the from ing 222:374-9. Biochem nal A atography. chrom liquid ance eostasis Ca2+ high-perform hom the of reversed-phase 10 by 0-7.0:144 odulation m Invest the in Clin J cells. involved is uscle m kinase l ta le e k protein s in dystrophy Myotonic Chem iol B J activity. Rho-kinase regulates egatively n Rho-kinase of s u in rm te COOH 22. Boucher, C. A., S. K. King, N. Carey, R. Krahe, C. L. Winchester, S. Rahman, et al. 1 995. A novel homeodomain-encoding gene is associated with a large CpG island interrupted by the myotonic dystrophy unstable (CTG)n repeat. Hum Mol Genet 4:1919-25. 23. Bradford, M. M. 1 976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248­ 54. 24. Brambillasca, S., M. Yabal, M. Makarow, and N. Borgese. 2006. Unassisted translocate of large polypeptide domains across phospholipid bilayers. J Cell Biol 1 75:767-77. 25. Brambillasca, S., M. Yabal, P. Soffientini, S. Stefanovic, M. Makarow, R. S. Hegde, et al. 2005. Transmembrane topogenesis of a tail-anchored protein is modulated by membrane lipid composition. EMBO J 24:2533-42. 26. Brook, J. D., M. E. McCurrach, H. G. Harley, A. J. Buckler, D. Church, H. Aburatani, et al. 1 992. M o le cu lar basis of m yotonic dystrophy: expansion of a trin u cle o tid e (CTG) repeat at the 3' end of a transcript encoding a protein kinase family member. Cell 69:385. 27. Bulbarelli, A., T. Sprocati, M. Barberi, E. Pedrazzini, and N. Borgese. 2002. Trafficking of tail-anchored proteins: transport from the endoplasmic reticulum to the plasma membrane and sorting between surface domains in polarised epithelial cells. J Cell Sci 1 1 5:1 689-702. 28. Burkhard, P., J. Stetefeld, and S. V. Strelkov. 200 1. Coiled coils: a highly versatile protein folding motif. Trends Cell Biol 11:82-8. 29. Bush, E. W., S. M. Helmke, R. A. Birnbaum, and M. B. Perryman. 2000. Myotonic dystrophy protein kinase domains mediate localization, oligomerization, novel catalytic activity, and autoinhibition. Biochemistry 39:8480-8490. 30. Bush, E. W., C. S. Taft, G. E. Meixell, and M. B. Perryman. 1 996. Overexpression of myo­ tonic dystrophy kinase in BC3H1 cells induces the skeletal muscle phenotype. J Biol Chem 27 1:548-52. 31. Buxton, J., P. Shelbourne, J. Davies, C. Jones, T. Van Tongeren, C. Aslanidis, et al. 1 992. Detection of an unstable fragment of DNA specific to individuals with myotonic dystro­ phy. Nature 355:547-8. 32. Byers, J. T., R. M. Guzzo, M. Salih, and B. S. Tuana. 2009. Hydrophobic profiles of the tail anchors in SLMAP dictate subcellular targeting. BMC Cell Biol 10:48. 33. Caenepeel, S., G. Charydczak, S. Sudarsanam, T. Hunter, and G. Manning. 2004. The mouse kinome: discovery and comparative genomics of all mouse protein kinases. PNAS 1 0 1:1 1 707-1 1 7 1 2. 34. Cai, Q., and Z. H. Sheng. 2009. Moving or stopping mitochondria: Miro as a traffic cop by sensing calcium. Neuron 6 1:493-6. 35. Carango, P., J. E. Noble, H. G. Marks, and V. L. Funanage. 1 993. Absence of myotonic dys­ trophy protein kinase (DMPK) mRNA as a result of a triplet repeat expansion in myotonic dystrophy. Genomics 1 8:340-8. 36. Carrasco, M., J. Canicio, M. Palacin, A. Zorzano, and P. Kaliman. 2002. Identification of in­ tracellular signaling pathways that induce myotonic dystrophy protein kinase expression during myogenesis. Endocrinology 1 43:30 1 7-25. 37. Chahine, M., and A. L. George, Jr. 1 997. Myotonic dystrophy kinase modulates skeletal muscle but not cardiac voltage-gated sodium channels. FEBS Lett 4 1 2:62 1 -624. 38. Chan, D. C. 2006. Mitochondria: dynamic organelles in disease, aging, and development. Cell 1 25:1 24 1 -52. 39. Chang, D. T., and I. J. Reynolds. 2006. Mitochondrial trafficking and morphology m healthy and injured neurons. Prog Neurobiol 80:24 1 -68. 40. Charlet, B. N., R. S. Savkur, G. Singh, A. V. Philips, E. A. Grice, and T. A. Cooper. 2002. Loss of the muscle-specific chloride channel in type 1 myotonic dystrophy due to misregu- lated alternative splicing. Mol Cell 1 0:45-53. 41. Chartier, C., E. Degryse, M. Gantzer, A. Dieterle, A. Pavirani, and M. Mehtali. 1 996. Efficient generation of recombinant adenovirus vectors by homologous recombination in Esch­ erichia coli. J Virol 70:4805-1 0. 42. Chen, H., and D. C. Chan. 2009. Mitochondrial dynamics--fusion, fission, movement, and mitophagy—in neurodegenerative diseases. Hum Mol Genet 18:R1 69-76. 43. Cho, D. H., and S. J. Tapscott. 2007. Myotonic dystrophy: emerging mechanisms for DM1 and DM2. Biochim Biophys Acta 1 772:1 95-204. 44. Cho, D. H., C. P. Thienes, S. E. Mahoney, E. Analau, G. N. Filippova, and S. J. Tapscott. 2005. Antisense transcription and heterochromatin at the DM1 CTG repeats are constrained by CTCF. Mol Cell 20:483-9. 45. Chou, T., K. S. Kim, and G. Oster. 200 1. Statistical thermodynamics of membrane bending- mediated protein-protein attractions. Biophys J 80:1 075-87. 46. Claycomb, W. C., N. A. Lanson, Jr., B. S. Stallworth, D. B. Egeland, J. B. Delcarpio, A. Bahinski, et al. 1 998. HL-1 cells: a cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc Natl Acad Sci U S A 95:2979-84. 6 6 • References

133 134 6 • References 9 Cik F H 1 5. s lh-krtn cie ci? aue 170:882-3. Nature coil? coiled a keratin alpha- Is 1 952. H. et F. Crick, lexander, A E. I. 49. the Hardeman, C. in E. Ghoddusi, proteins M. Kizana, cytosolic E. of ell, role axw M The L. A. T., S. 2009. Cooper, Borgese. N. 48. and Longhi, R. F., S. Colombo, 47. 2 Eiso, . T Ase, . dto, . . ls D J Wt, . nrt e a. 00 Inde­ 2000. al. et Anvret, M. Watt, J. D. ells, W J. D. Edstrom, L. Ansved, T. M., Eriksson, 72. he­ r e y ila sb n tra nonpolar Identifying 1986. Goldman. A. and Steitz, A. T. M., D. an, Engelm 70. 71. Eriksson, M., T. Ansved, L. Edstrom, M. Anvret, and N. Carey. 1 999. S im ultan eo us analysis analysis us eo ultan im S 1999. Carey. N. and Anvret, M. Edstrom, L. Ansved, T. M., Eriksson, 71. 9 Ed, . H Ymmt, n M Eai 20. ucinl oprto ad eaain of separation and cooperation Functional 2003. l­ ve e Esaki. D M. and 2000. a. Yamamoto, H. T., Takashim S. Endo, and 69. Yamada, T. Structure Harada, 2009. K. Knapp. S. Arahata, K. and Fedorov, Motonaga, O. K. A., Pike, C. Endo, A. Niesen, H.68. F. 1 997. Amos, al. A. M., et J. ani, Elkins, Subram S. 67. Distel, B. Snyder, B. leaching W. RNA Berg, 2004. den van M. P. Junghans. R. Kwast, L. Y., and a, Elgersm Ebralidse, K. yo­ m 66. Petkova, of V. Wang, Y. A., Localization 1996. Ebralidze, Epstein. F. expres­ H. 65. and kinase il­ Harati, m Y. protein Casey, new L. D. Myotonic the Ma, L. 1 into 996. W., P. 0s 1 Epstein. 95 Dunne, F. H. the and 64. from Casey, L. D. Ma, coupling L. lar W., llu P. e c b su Dunne, the of 63. Excitation-contraction ts an in 2006. F. A. eterm d lar cu Dulhunty, le o M 62. 2007. Kumar. S. et and Mayatepek, Dorstyn, E. L. J., Swarts, G. of H. Doumanis, n Jong, de 61. S. calizatio lo A. Testa, R. rial E. itochond M from Koopman, J. F., W. 2003. system: aier, Falchi. nervous M. A. istelm D the and in 1 60. Diana, A. Isola, R. dystrophy astrocytes Liu, and G., S. Myotonic Diaz, neurons 2008. pal 59. Cisneros. B. Hippocam and B., M. ulturing Leon, C 1 de 998. Koop­ Dotti. J. C. W.58. and and Liane, M. ieringa, M., W B. tc Hoop, m s, de lu u illem tic W re H. P. 57. ic Errington, endoplasm J. R. ers th te Fransen, 2 A. J. J., A. Mitofusin Groof, de 2008. 56. Scorrano. L. and M., O. Brito, Expan­ de 1997. 55. Housman. E. D. cytoplasm. and the in Singer, H. n R. calizatio Taneja, L. lo K. mRNA McCurrach, for E. M. M., B. Pathways Davis, 2006. A 54. 1 Singer. 999. H. R. and K., Hendriks. W. and zaplinski, C calcium ieringa, W B. 53. unication: Fransen, m J. com local Schepens, J. ijers, W Crystallo- M. E., itochondrial -m cta A /ER Cuppen, SR 2009. 52. coiled-coils. ple Hajnoczky. G. sim and G., Csordas, alpha-helices: of 51. packing The 1953. C. H. F. Crick, 50. rns ice Si 3 1:687-93. Sci Biochem Trends metl xrsin f ytnc ytoh poen iae n ri ad t rlvne to relevance 1 16:3259-67. its Sci and Cell J brain 1 in 1elles. 3-20. 00:5 kinase (Berl) protein europathol N dystrophy Acta 1. phenotype. myotonic 1 l of 8:782-9 a ic Sci lin c expression ic Protein ental kinase. endoplasm pm o the of protein yotonica m proliferation causes dystrophia of S.cerevisiae, in 303:383-7. ly b Science assem e dystrophy. peroxisom myotonic for in quired n tio transcrip isrupts d factors n tio transcrip of single the at potential 5. 1 brane 4:907-1 em m rial 29-38. 73:1 A itochond m of Cytometry level. quantification ll e c organelle Life 2008. al. 352-62. 1787:1 Acta Biophys Biochim ROS. and myosin fast lt u ad of and l a iv rv u nuclei s 1. sustained eripheral p 58:200-1 ith w enables m Cytoskeleton tu yotubes m otil bstra M su st Cell la b ro differentiated fib a expression. hly on hig co-culture contractile, C2C12 2004. al. of expression of the three m yotonic dystrophy locus genes in ad u lt s k e le ta l m uscle uscle m l ta le e k s lt u ad in genes locus dystrophy yotonic m three the of expression of 115:32 -53. rnsoc or i poen mpr it miohnra te obeme rn budd organ­ bounded brane em double-m the itochondria, m into port im protein in rs to ca slo tran Cell disease. ith w n alteratio its and uscle m l 1 ta -8. le e k 225:28 s in Commun kinase Res protein Biophys Biochem dystrophy lenses. tonic bovine and human in sion 8-26. 86:1 Res Neurosci J 1. s. vol. echanism m handbook, lar u c ry le o m orato lab to A ic clin Biology: the Cell 63, 1 54-1 p. brain, l ta fe from dystrophy yotonic m of region untranslated 3' the in repeat e tid o le c u trin CUG a of sion 6:689-697. graphica sam ples: the CTG expansion correlates in ve rsely w ith DMPK and 59 expression levels, levels, expression 59 and DMPK ith w rsely ve in correlates expansion CTG the ples: sam edn rglto o te ooi dsrpy lcs ee psntly n drn adult during and postnatally genes locus 1 dystrophy yotonic m the of regulation pendent 1. 16:7326-4 J EMBO brane. mem m lu u tic re 1 1 9-25. 20:3 Biol Cell istochem H probes. dihydrofluorescein by species oxygen reactive ic sarcoplasm the of fill 277:5275-84. re Chem al tim p Biol o J for muscle. ssential l e ta is le e k s in system Ca2+ store kinase m lu u Sci tic re creatine Acad The Natl 2002. Proc man. 0. 456:605-1 transcripts. of Nature retention itochondria. m nuclear in lts su re ts rip sc tran kinase protein eepeso o Px5, popoyae prxsma i e a me rn poen e­ re protein brane em m ral teg in al peroxisom phosphorylated a Pex15p, of verexpression O 33:52-63. ytoskeleton C otil M 33:763-72. Differ Physiol Death acol Cell Pharm BUFFY. Exp and Clin DEBCL . lennium ologues hom Bcl-2 Drosophila the of n calizatio lo 94:7388-7393. A S U ( 112 Sci Cell J cells. l lia e ith p e in PTP-BL of ent confinem apical 19):3299-308. Pt governs ain dom FERM 122:2383-92. Sci Cell J bilayers. phospholipid into proteins tail-anchored of insertion but not DMAHP levels. Hum Mol Genet 8:1 053-1 060. 053-1 8:1 Genet Mol Hum levels. DMAHP Chem not but Biophys Biophys Rev Annu proteins. brane em m of sequences acid ino am in lices skeletal muscle regeneration. J Biol Chem 275:1 9964-1 9969. 73. Eriksson, M., B. Hedberg, N. Carey, and T. Ansved. 200 1. Decreased DMPK transcript levels in myotonic dystrophy 1 type IIA muscle fibers. Biochem Biophys Res Commun 286:1 1 77-1 1 82. 74. Etienne-Manneville, S., and A. Hall. 2002. Rho GTPases in cell biology. Nature 420:629­ 35. 75. Fardaei, M., M. T. Rogers, H. M. Thorpe, K. Larkin, M. G. Hamshere, P. S. Harper, et al. 2002. Three proteins, MBNL, MBLL and MBXL, co-localize in vivo with nuclear foci of expanded-repeat transcripts in DM1 and DM2 cells. Hum Mol Genet 1 1:805-8 1 4. 76. Ferrara, N. 2004. Vascular endothelial growth factor as a target for anticancer therapy. Oncologist 9 Suppl 1:2-10. 77. Filippova, G. N., C. P. Thienes, B. H. Penn, D. H. Cho, Y. J. Hu, J. M. Moore, et al. 200 1. CTCF-binding sites fla n k CTG/CAG repeats and fo rm a m ethylation-sensitive in s u lato r at the DM1 locus. Nat Genet 28:335-43. 78. Forner, F., S. Furlan, and S. Salvatori. 2010. Mass spectrometry analysis of complexes formed by myotonic dystrophy protein kinase (DMPK). Biochim Biophys Acta 1 804:1 334­ 41. " 79. Franke, C., H. Hatt, P. A. Iaizzo, and F. Lehmann-Horn. 1 990. Characteristics of Na+ channels and Cl- conductance in resealed muscle fibre segments from patients with myotonic dystrophy. J Physiol 425:39 1 -405. 80. Fransson, S., A. Ruusala, and P. Aspenstrom. 2006. The atypical Rho GTPases Miro-1 and Miro-2 have essential roles in mitochondrial trafficking. Biochem Biophys Res Commun 344:500-1 0. 81. Franzini-Armstrong, C. 2007. ER-mitochondria communication. How privileged? Physiology (Bethesda) 22:26 1 -8. 82. Frederick, R. L., J. M. McCaffery, K. W. Cunningham, K. Okamoto, and J. M. Shaw. 2004.

Yeast Miro GTPase, Gem1p, regulates mitochondrial morphology via a novel pathway. j Cell Biol 1 67:87-98. 83. Frisch, R., K. R. Singleton, P. A. Moses, I. L. Gonzalez, P. Carango, H. G. Marks, et al. 200 1. Effect of triplet repeat expansion on chromatin structure and expression of DMPK and neighboring genes, SIX5 and DMWD, in myotonic dystrophy. Mol Genet Metab 74:28 1 -9 1. 84. Frodin, M., T. L. Antal, B. A. Dummler, C. J. Jensen, M. Deak, S. Gammeltoft, et al. 2002. A phosphoserine/threonine-binding pocket in AGC kinases and PDK1 mediates activation by hydrophobic motif phosphorylation. EMBO J 2 1:5396-407. 85. Fu, Y. H., D. L. Friedman, S. Richards, J. A. Pearlman, R. A. Gibbs, A. Pizzuti, et al. 1993 Decreased expression of myotonin-protein kinase messenger RNA and protein in adult form of myotonic dystrophy. Science 260:235-8. 86. Fu, Y. H., A. Pizzuti, R. G. Fenwick, Jr., J. King, S. Rajnarayan, P. W. Dunne, et al. 1 992. An unstable triplet repeat in a gene related to myotonic muscular dystrophy. Science 255:1 256-8. 87. Furling, D., D. Lemieux, K. Taneja, and J. Puymirat. 200 1. Decreased levels of myotonic dystrophy protein kinase (DMPK) and delayed differentiation in human myotonic dystro­ phy myoblasts. Neuromuscul Disord 1 1:728-35. 88. Furst, D. O., M. Osborn, R. Nave, and K. Weber. 1 988. The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy:

a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. j Ce ll Biol 1 06:1 563-72. 89. Garcia, P., Z. Ucurum, R. Bucher, D. I. Svergun, T. Huber, A. Lustig, et al. 2006. Molecular insights into the self-assembly mechanism of dystrophia myotonica kinase. FASEB J 20:1 1 42-5 1. 90. Geering, K. 2006. FXYD proteins: new re g u lato rs of Na-K-ATPase. Am J Physiol Renal Physiol 290:F24 1 -50. 91. Giorgi, C., D. De Stefani, A. Bononi, R. Rizzuto, and P. Pinton. 2009. S tru ctu ral and

functional link between the mitochondrial network and the endoplasmic reticulum. Int j Biochem Cell Biol 41:1817-27. 92. Gomes, E. R., S. Jani, and G. G. Gundersen. 2005. Nuclear movement regulated by Cdc42, MRCK, myosin, and actin flow e stablishes MTOC p olarization in m igrating cells. Cell 1 2 1:45 1 -63. 93. Graham, F. L., J. Smiley, W. C. Russell, and R. Nairn. 1 977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol 36:59-74. 94. Groenen, P., and B. Wieringa. 1 998. Expanding complexity in myotonic dystrophy. BioEs­ says 20:90 1 -9 1 2. 95. Groenen, P. J., D. G. Wansink, M. Coerwinkel, W. van den Broek, G. Jansen, and B. Wier­ inga. 2000. Constitutive and regulated modes of splicing produce six major myotonic dystrophy protein kinase (DMPK) isoforms with distinct properties. Hum Mol Genet 9:605-1 6. 96. Gruber, M., and A. N. Lupas. 2003. Historical review: another 50th anniversary--new 6 6 • References

135 136 6 • References 2.Jcb, . . A A Bnes A Osehf J H Vekm, . a Me, . . ees e al. et Wevers, A. R. Mier, van P. Veerkamp, H. J. Oosterhof, Novel A. 2003. Benders, A. al. A. E., et A. eigel, W L.Jacobs, N. 121. Wei, L. Rowan, G. B. A Fluck, M. 1 998. elaguli, B attenberg. N. D., W W. B. Iyer, and 120. Vadas, M. Gamble, J. Khew-Goodall, Y. S., l­ u M T. R. Isenmann, and 119. Dyer, M. Cell J. kinases. Andrews, W. D. protein of Henderson, P. M. lasticity p Pelitire, M. ational S. T., Y. rm confo Hwang, The cor­ 118. 2002. The 1 992. al. Kuriyan. et J. and Surh, M., L. Huse, 117. Mahadevan, M. of Jacob, P. ettler, M overexpression G. by C. Tsilfidis, induced A., Hunter, 116. clustering graphic rial crystallo itochond X-ray M ary in 2007. lim Yoon. re Y. P and Yu, P., T. 2009. Sha. B. that Huang, 115. and Fu, signal Z. of Jin, Z. Qian, X. Li, Characterization J. J., Hu, 2002. 114. Mihara. K. novo De and 2004. al. Sakaguchi, M. et Suzuki, Stelzer, H. C., W. Horie, Stalz, H.113. al, graw A P. Ollesch, J. 2005. Weise, K. Cooper. W., A. M. T. and Hofmann, Swanson, 112. S. M. Mancini, A. M. express­ Poulos, G. M. mice Savkur, S. R. H., Transgenic T. Ho, 2005. 111. Cooper. A. T. and Rec­ strong, rm A 2005. L. D. al. et Bundman, D. H., T. Andersson, peroxi­ Ho, H. 110. the of Boekel, J. Lundin, Topogenesis C. aier, peroxisomes. ihlm B of K. Kim, H. T., Biogenesis Hessa, 2005. 109. of Erdmann. R. haracterisation and I., C 1 997. Heiland, ells. W 108. J. D. and plified sim A Johnson, 1 J. K.998. Hoyle, C. Vogelstein. B. Carne, S. K., and S. et Heath, Kinzler, 107. W. K. athy, Yu, Bhanum J. D. C. Costa, da Wang, T. L. Y. Lien, Zhou, C. C., S. Y. T. He, , an 106. ram ka llilan a M K. al. et Irrinki, M. K. Kunding, H. B. J., A. home- Bolinger, Hawkins, Y. P. 105. the of Madsen, L. K. analysis Larsen, J. Functional Bhatia, K. V. S., 2000. N. myotonic in Johnson. Hatzakis, J. K. 104. UK. and nticipation A London, 1 992. inchester, W L. C. E., Saunders, Shaw. S. J. D. WB 37. Harris, and vol. 103. kinase ed, Reardon, ily: W. 3rd Harley, superfam G. H. S., dystrophy, P. kinase Harper, Myotonic 1. protein 102. 200 S. P. Harper, eukaryotic 101. The 1995. Hunter. T. and K., S. Hanks, 100. 9 Hlah A, . adrf S Lrne, . oekaz R Vlme-net R Edan e al. et Erdmann, R. er-Engert, Volkm R. Rosenkranz, K. Ca2+ Lorenzen, local S. of Landgraf, C. A., achinery m The Halbach, 2000. 99. Pacher. P. and al. et Madesh, M. Kim, K. P. Csordas, G. G., Sougrat, R. Hajnoczky, Mitra, K. 98. Satpute-Krishnan, P. Rambold, S. A. W., D. Hailey, 97. Md ee 29:774-9. Genet Med J 9. h clim hmesai ad h me rn ptnil f utrd sl cells uscle m cultured of potential brane em m the and eostasis hom calcium The 1990. 649-60. 9:1 Cell Biol Mol signal. targeting drial n cho 109:275-82. Nature translocon. m lu ticu re ic endoplasm the by helices e n 1. ra b em sm 433:377-8 tran of Hum ognition locus. DM urine m the at gene ain-encoding eodom hom a mDMAHP, J of homeostasis. expression rial d on itoch m rs alte and STIM1 activates tathionylation S-glu 2010. al. 187 -8. 28:1 Res Acids Nucleic SIX5. protein ain odom 7. 1 19:2508-1 Sci Cell J sites. PEX19-binding inal -term C through occurs 1:69-81. rncito. ohmity 42:7477-86. istry ­ ito m iochem B a contains transcription. (VAMP-1) protein-1 or brane m em lu m ticu re ic vesicle-associated endoplasm of r ithe e to b5 splice-isoform e cytochrom protein brane em m tail-anchored death. ll e c and dysfunction rial d on itoch 1. m 65:489-9 causes Mfn2 Commun Cryst protein Biol fusion Struct rial F d Biol on Sect Mol itoch m the brane. rystallogr C mem Acta outer Get3. rial d yeast on of itoch m studies alian m fusion. am m brane to em m proteins driving peptides C-tail-anchored e n ra b directs em sm tran le ib x fle ally 1 18:2923-2933. ation Sci rm fo n Cell co of J design dystrophy. myotonic in splicing tive 272:2362-72. J FEBS proteins. atrix m and brane 4. em m 95:2509-1 al A som S U Sci Acad atl N Proc adenoviruses. binant recom generating for system motifs. anchoring protein and helices phipathic am recruit branes em m curved How 51:10-6. 2009. Genet Hum J Am problem. old an on ht lig new dystrophy: PEX15 and PEX26 proteins brane em m al Pt peroxisom 529 Physiol J tail-anchored the of mitochondria. and Targeting m 2006. lu u tic re ic starvation. during sarco-endoplasm biogenesis between e g allin ign autophagosom s for branes em m supply itochondria M 2010. ctltc dman rcue n casfcto. AE J 9:576-596. J FASEB classification. and tructure s ain dom (catalytic) phosphorylation target in the serum response factor MADS box reg ulates alpha-actin alpha-actin ulates reg box MADS factor response serum the in target phosphorylation 9. 16:3002-1 Cell Plant itochondria. m dystrophy. myotonic in n atio lific p am repeat e tid o cle u trin CTG ith w onset of age of relation 28:679-85. Sci Biochem Trends coils. coiled in icities period len. 2004. Novel targeting sign als m ediate the sorting of different isoform s of the the of s isoform different of sorting the ediate m als sign targeting Novel 2004. len. 86:289-302. Biol Cell J Eur 13:1615-25. Cell ­ 1. a rn 10 lte 1:1 4776-8 a A of S U Sci is-regulation Acad m atl N from Proc separable is foci RNA Hum ith w d dystrophy. uscleblin m yotonic m of in olocalization observed C is-regulation m 547. 1539-1 4:1 splicing Genet Mol reproduce CUG-BP1 ing 6:6 51-7. Genet Mol 1 1-405. 90:39 Biol Cell 1. 5:835-4 Biol Chem Nat 14 1:656-67. Cell from patients with myotonic dystrophy. Biochim Biophys Acta 1 096:1 4-9. 122. Jahn, R., and R. H. Scheller. 2006. SNAREs--engines for membrane fusion. Nat Rev Mol Cell Biol 7:63 1 -43. 123. James, D. I., P. A. Parone, Y. Mattenberger, and J. C. Martinou. 2003. hFis1, a novel com­ ponent of the mammalian mitochondrial fission machinery. J Biol Chem 278:36373-9. 124. Jang, Y. J., S. Ma, Y. Terada, and R. L. Erikson. 2002. Phosphorylation of threonine 210 and the role of serine 137 in the regulation of mammalian polo-like kinase. J Biol Chem 277:44 1 1 5-20. 125. Jansen, G., D. Bachner, M. Coerwinkel, N. Wormskamp, H. Hameister, and B. Wieringa. 1 995. Structural organization and developmental expression pattern of the mouse WD- repeat gene DMR-N9 immediately upstream of the myotonic dystrophy locus. Hum Mol Genet 4:843-852. 126. Jansen, G., P. J. Groenen, D. Bachner, P. H. Jap, M. Coerwinkel, F. Oerlemans, et al. 1 996. Abnormal myotonic dystrophy protein kinase levels produce only mild myopathy in mice. Nat Genet 13:316-24. 127. Jansen, G., M. Mahadevan, C. Amemiya, N. Wormskamp, B. Segers, W. Hendriks, et al. 1 992. Characterization of the myotonic dystrophy region predicts multiple protein isoform-encoding mRNAs. Nat Genet 1:261-6. 128. Janssen, E., J. Kuiper, D. Hodgson, L. V. Zingman, A. E. Alekseev, A. Terzic, et al. 2004. Two structurally distinct and spatially compartmentalized adenylate kinases are ex­ pressed from the AK1 gene in mouse brain. Mol Cell Biochem 256-257:59-72. 129. Jiang, H., A. Mankodi, M. S. Swanson, R. T. Moxley, and C. A. Thornton. 2004. Myotonic dystrophy type 1 is associated with nuclear foci of mutant RNA, sequestration of muscleblind proteins and deregulated alternative splicing in neurons. Hum Mol Genet 1 3:3079-88. 130. Jin, S., M. Shimizu, A. Balasubramanyam, and H. F. Epstein. 2000. Myotonic dystrophy protein kinase (DMPK) induces actin cytoskeletal reorganization and apoptotic-like bleb- bing in lens cells. Cell Motil Cytoskeleton 45:1 33-48. 131. Jouaville, L. S., P. Pinton, C. Bastianutto, G. A. Rutter, and R. Rizzuto. 1 999. Regulation of mitochondrial ATP synthesis by calcium: evidence for a long-term metabolic priming. Proc N atl Acad Sci U S A 96:1 3807-1 2. 132. Junghans, R. P. 2009. Dystrophia myotonia: why focus on foci? Eur J Hum Genet 1 7:543­ 53. 133. Kahlert, S., and G. Reiser. 2000. Requirement of glycolytic and mitochondrial energy sup­ ply for loading of Ca(2+) stores and InsP(3)-mediated Ca(2+) signaling in rat hippocampus astrocytes. J Neurosci Res 6 1:409-20. 134. Kaliman, P., D. Catalucci, J. T. Lam, R. Kondo, J. C. Gutierrez, S. Reddy, et al. 2005. Myotonic dystrophy protein kinase phosphorylates phospholamban and regulates calcium uptake in cardiomyocyte sarcoplasmic reticulum. J Biol Chem 280:80 1 6-2 1. 135. Kaliman, P., and E. Llagostera. 2008. Myotonic dystrophy protein kinase (DMPK) and its role in the pathogenesis of myotonic dystrophy 1. Cell Signal. 136. Karim, M. R., T. Kanazawa, Y. Daigaku, S. Fujimura, G. Miotto, and M. Kadowaki. 2007. Cytosolic LC3 ratio as a sensitive index of macroautophagy in isolated rat hepatocytes and H4-II-E cells. Autophagy 3:553-60. 137. Kaufmann, T., S. Schlipf, J. Sanz, K. Neubert, R. Stein, and C. Borner. 2003. Character­ ization of the signal that directs Bcl-x(L), but not Bcl-2, to the mitochondrial outer membrane. J Cell Biol 1 60:53-64. 138. Kemper, C., S. J. Habib, G. Engl, P. Heckmeyer, K. S. Dimmer, and D. Rapaport. 2008. Integration of tail-anchored proteins into the mitochondrial outer membrane does not require any known import components. J Cell Sci 1 2 1:1 990-8. 139. Kim, D. H., M. A. Langlois, K. B. Lee, A. D. Riggs, J. Puymirat, and J. J. Rossi. 2005. HnRNP H inhibits nuclear export of mRNA containing expanded CUG repeats and a distal branch point sequence. Nucleic Acids Res 33:3866-74. 140. Kim, I., S. Rodriguez-Enriquez, and J. J. Lemasters. 2007. Selective degradation of mito­ chondria by mitophagy. Arch Biochem Biophys 462:245-53. 141. Kim, P. K., M. G. Annis, P. J. Dlugosz, B. Leber, and D. W. Andrews. 2004. During apoptosis bcl-2 changes membrane topology at both the endoplasmic reticulum and mitochondria. Mol Cell 1 4:523-9. 142. Kim, P. K., R. T. Mullen, U. Schumann, and J. Lippincott-Schwartz. 2006. The origin and maintenance of mammalian peroxisomes involves a de novo PEX16-dependent pathway from the ER. J Cell Biol 1 73:52 1 -32. 143. Kimura, T., M. Nakamori, J. D. Lueck, P. Pouliquin, F. Aoike, H. Fujimura, et al. 2005. Altered mRNA splicing of the skeletal muscle ryanodine receptor and sarcoplasmic/en­ dop lasm ic re tic u lu m Ca2+-ATPase in myotonic dystrophy type 1. Hum Mol Genet 1 4 :2 1 8 9 ­ 200. 144. Kirby, R. J., G. M. Hamilton, D. J. Finnegan, K. J. Johnson, and A. P. Jarman. 200 1. Dro­ sophila homolog of the myotonic dystrophy-associated gene, SIX5, is required for muscle 6 6 • References

137 138 6 • References 168. Liu, X., C. N. Kim, J. Yang, R. Jem m erson, and X. Wang. 1 996. Induction of apoptotic apoptotic of Induction 1996. Wang. X. and 1. 200 erson, m al. Jem R. et Yang, J. Naylor, L. Kim, S. N. C. X., Kress, Liu, W. 168. F. Jacobsen, J. t­ la re Moseley, - L. ase M. kin Ricker, K. L., C. dystrophy Liquori, Myotonic 167. 1998. Lim. L. and ROKa Manser, E. kinase Tan, I. Chen, Q. X. T., RhoA-binding Leung, 166. p160 The 1996. Lim. L. and Andreadis, Manser, A. E. Chen, X.-Q. T., Delacourte, Leung, A. 165. Belarbi, K. aschke, the of Schraen-M S. Roles Dhaenens, 2004. M. C. O., Youle. Leroy, J. R. 164. and Smith, L. C. ­ b ski, A Karbow 2003. M. al. et Jeong, Y. S. J., Y. Lee, Moorman, R. 163. J. Reddy, S. Q.Wang, Mistry, J. rial D. Patel, K. itochond m M. C., H. the Lee, of 162. media: Identification -free 2006. serum in Unadkat. D. yoblasts J. m and of Zhang, H. Lai, Y. ifferentiation D W., E. Lee, 2000. 161. a Purslow. P. of P. and A., M. Characterization Lawson, 160. 2000. Morris. E. G. and Nguyen, anious, M. Arm T. S. C. Pham, C. Y. T., L. Bombach, L. K. Lam, 159. Elborne, V. Scalise, D. proteins Guthrie, D. R. A., binding W. RNA of boise, ily LaFram m fa 158. CELF The 1. 200 Cooper. A. T. and steady-state Charlet, N. N., A. Increased Ladd, 157. 2007. Cooper. A. T. and Wang, S. G. M., 1 995. N. Rapoport. artinez, A. T. cu-M and cyto­ Kuyum of ann, 156. iedenm W B. topology situ ann, In Hartm E. 1 996. Ito. ilger, A. and hnert-H A G. U., Ri- Mitoma, Kutay, J. 155. 2007. Honsho, M. Krzyzosiak. J. Kinoshita, W. R., J. and Kuroda, Mezer, 154. de M. Sobczak, K. 1 995. ykowska, al. M A. et Fiszer, A. J., Giacanelli, M. Krol, 153. Carango, P. Roeder, E. 2004. Chan. C. D. Abbruzzese, C. and a, McCaffery, Ashizaw M. R., T. J. Chen, Krahe, H. 152. Kaiser, T. J. H. Detmer, P. A. S. T., and Koshiba, Engelen, 151. van G. B. Kuppevelt, van H. T. 2008. Oosterhof, s. A. illem W H. P. Renders, M. and J., W. eitink, Koopman, Sm A. 150. J. Esseling, itochondrial M J. J. 2008. aier, istelm D F. Bossy-Wetzel. E. J., W. and expan­ Koopman, repeat 149. Schwarzenbacher, R. Trinucleotide 1 Perkins, 997. G. B., A. Tapscott. J. S. Knott, Mice 148. and Bird, 2000. D. T. al. Otten, et D. A. R., T. Snider, L. Klesert, an, 147. Adelm 9:224-44. J. Med Maylie, Biol J. Clark, Oral I. Rev J. Cho, H. Crit D. R., T. Klesert, Neuro-osteology. 146. 1 998. I. Kjaer, 145. Transport route for synaptobrevin via a novel pathw ay of insertion into the endoplasm ic ic endoplasm the into insertion of ay pathw novel a via synaptobrevin for route Transport d d4-idn kns at a a d4 efco i po tn ctseea rorganiza­ reo cytoskeletal oting prom in effector Cdc42 a as acts kinase Cdc42-binding ed ally rm o n ab 84:852-9. Res event Neurosci splicing a J I. type inclusion, 2/3 dystrophy exons yotonic m Tau in represses enhanced ETR-3 2006. al. et 30-7. 1 67:1 Organs Tissues Cells 9:2167-73. line-specific. Genet ll e Mol c are Hum edia heart. m and odified m uscle of m in effects gene ly sive clu x e uscle-specific st m o ll alm e c llite expressed te a s on 8. 307-1 35:1 phenotype Cardiol and Cell origin Mol uscle J m of expression. Effect 2003. al. et 120:828-33. Biochem J brane. mem m lu ticu re ic endoplasm the in b5 e chrom 1. 46:304-1 Methods handling. calcium and 1 1:1044-9. Biol Curr ent. developm gonad and 293:864-7. 1 8:130-40. Biol Cell Mol tion. 1 1. 1-1 5:500 Cell Biol Mol sis. Chem Biol ­ a J lic p e. in im fialurid (hENT1): of 1 r xicity to rte o sp rial d tran on itoch m nucleoside in rative ilib differences u q e 6700-6. 1:1 28 human interspecies for the of tions signal targeting 28:68-78. Cell Mol tion. 25:575-86. Cell Mol targets. Physiol role Cell possible Physiol a J muscle: Am l ta le e synthesis. k s ATP an rial d hum on in itoch m val o of m Ca2+ re 263-9. of ing 285:C1 prim Ca2+-dependent Upregulation for 2003. s. m ille W 8. 9:505-1 16:402-6. Neurosci Genet Rev Nat Nat DMAHP. of neurodegeneration. expression in reduces entation Genet locus fragm Nat dystrophy dystrophy. myotonic myotonic the for at s n sion tio lica p im cataracts: develop 09. 05-1 25:1 Six5 in deficient program in cell-free extracts: requirem ent for dATP and cytochrom e c. Cell 86:1 47-57. 86:1 Cell c. e cytochrom and dATP for ent requirem extracts: cell-free in program apopto­ in Opa1 and Drp1, Fis1, ediators m fusion and dystrophy fission myotonic rial in itochond m deficient alian myocytes m 1 2:1 am 47-57. m cardiac Genomics urine m Physiol in kinase. gating protein channel Na al rm no is DMPK, kinase, protein dystrophy yotonic m the that s show panel antibody onoclonal m 14:217-23. J EMBO brane. mem m lu u tic re 305:858-62. Science 28:1-14. complexes. Genomics itofusin m by processing. tethering rial d on itoch m of basis ral ctu tru S Myotonic dystrophy type 2 caused by a CCTG expansion in intron 1 of ZNF9. Science Science ZNF9. of 1 intron in expansion CCTG a by caused 2 type dystrophy Myotonic cytoskeleton. the of reorganization the in involved is and ily 1 1 m 3-5327. 6:53 fa Biol kinase Cell a of Mol ber em m a is Cell Mol splicing. ative ltern a regulated lly ta n e m p lo ve e ryla­ d ho sp and erpho hyp 2285-96. 1:1 cell-specific in Biol PKC-mediated to licated p due im are is 1 dystrophy yotonic m in CUGBP1 of ls ve le specific silence t tha repeats r shorte into hairpins and repeat n t tio le trip transcrip DMPK cleaves on dicer expansion bonuclease repeat e tid o cle u trin dystrophy myotonic of Effect orphology m potential, brane em m rial d on itoch m of analysis ll ce live Computer-assisted 169. Llagostera, E., M. C. Carmona, M. Vicente, R. M. Escorihuela, and P. Kaliman. 2009. High- fat diet induced adiposity and insulin resistance in mice lacking the myotonic dystrophy protein kinase. FEBS Lett 583:2 1 2 1 -5. 170. Llagostera, E., D. Catalucci, L. Marti, M. Liesa, M. Camps, T. P. Ciaraldi, et al. 2007. Role of myotonic dystrophy protein kinase (DMPK) in glucose homeostasis and muscle insulin action. PLoS ONE 2:e1134. 171. Lorenzon, P., A. Giovannelli, D. Ragozzino, F. Eusebi, and F. Ruzzier. 1 997. Spontaneous and repetitive calcium transients in C2C12 mouse myotubes during in vitro myogenesis. Eur J Neurosci 9:800-8. 172. Lyamzaev, K. G., D. S. Izyumov, A. V. Avetisyan, F. Yang, O. Y. Pletjushkina, and B. V. Chernyak. 2004. Inhibition of mitochondrial bioenergetics: the effects on structure of mitochondria in the cell and on apoptosis. Acta Biochim Pol 5 1:553-62. 173. Madaule, P., M. Eda, N. Watanabe, K. Fujisawa, T. Matsuoka, H. Bito, et al. 1 998. Role of citron kinase as a target of the small GTPase Rho in cytokinesis. Nature 394:49 1 -494. 174. Maeda, M., C. S. Taft, E. W. Bush, E. Holder, W. M. Bailey, H. Neville, et al. 1 995. Identi­ fication, tissue-specific expression, and subcellular localization of the 80- and 71-kDa forms of myotonic dystrophy kinase protein. J Biol Chem 270:20246-9. 175. Mahadevan, M., C. Tsilfidis, L. Sabourin, G. Shutler, C. Amemiya, G. Jansen, et al. 1 992. Myotonic dystrophy mutation: an unstable CTG repeat in the 3' untranslated region of the gene. Science 255:1 253-5. 176. Mahadevan, M. S., C. Amemiya, G. Jansen, L. Sabourin, S. Baird, C. E. Neville, et al. 1 993. Structure and genomic sequence of the myotonic dystrophy (DM kinase) gene. Hum Mol Genet 2:299-304. 177. Mahadevan, M. S., R. S. Yadava, Q. Yu, S. Balijepalli, C. D. Frenzel-McCardell, T. D. Bourne, et al. 2006. Reversible model of RNA toxicity and cardiac conduction defects in myotonic dystrophy. Nat Genet 38:1 066-70. 178. Malinina, L. 2005. Possible involvement of the RNAi pathway in trinucleotide repeat expansion diseases. J Biomol Struct Dyn 23:233-5. 179. Man, W. C., M. Miyazaki, K. Chu, and J. M. Ntambi. 2006. M em brane topology of mouse stearoyl-CoA desaturase 1. J Biol Chem 28 1:1 25 1 -60. 180. Mankodi, A., M. P. Takahashi, H. Jiang, C. L. Beck, W. J. Bowers, R. T. Moxley, et al. 2002. Expanded CUG repeats trigger aberrant splicing of ClC-1 chloride channel pre-mRNA and hyperexcitability of skeletal muscle in myotonic dystrophy. Mol Cell 1 0:35-44. 181. Mankodi, A., P. Teng-Umnuay, M. Krym, D. Henderson, M. Swanson, and C. A. Thornton. 2003. Ribonuclear inclusions in skeletal muscle in myotonic dystrophy types 1 and 2. Ann Neurol 54:760-768. 182. Mankodi, A., C. R. Urbinati, Q. P. Yuan, R. T. Moxley, V. Sansone, M. Krym, et al. 200 1. Muscleblind localizes to nuclear foci of aberrant RNA in myotonic dystrophy types 1 and 2. Hum Mol Genet 1 0:2 1 65-2 1 70. 183. Manning, G., G. D. Plowman, T. Hunter, and S. Sudarsanam. 2002. Evolution of protein kinase signaling from yeast to man. Trends Biochem Sci 27:5 1 4-20. 184. Manning, G., D. B. Whyte, R. Martinez, T. Hunter, and S. Sudarsanam. 2002. The protein kinase complement of the human genome. Science 298:1 9 1 2-34. 185. Mariappan, M., X. Li, S. Stefanovic, A. Sharma, A. Mateja, R. J. Keenan, et al. 2010. A ribosome-associating factor chaperones tail-anchored membrane proteins. Nature 466:1 1 20-4. 186. Marroquin, L. D., J. Hynes, J. A. Dykens, J. D. Jamieson, and Y. Will. 2007. Circumventing the Crabtree effect: replacing media glucose with galactose increases susceptibility of HepG2 cells to mitochondrial toxicants. Toxicol Sci 97:539-47. 187. M artorell, L., D. G. Monckton, J. Gamez, K. J. Johnson, I. Gich, A. Lopez de Munain, et al. 1 998. Progression of somatic CTG repeat length heterogeneity in the blood cells of myotonic dystrophy patients. Hum Mol Genet 7:307-3 1 2. 188. Mason, J. M., and K. M. Arndt. 2004. Coiled coil domains: stability, specificity, and bio­ logical implications. Chembiochem 5:170-6. 189. Mateja, A., A. Szlachcic, M. E. Downing, M. Dobosz, M. Mariappan, R. S. Hegde, et al. 2009. The structural basis of tail-anchored membrane protein recognition by Get3. Nature 46 1:36 1 -6. 190. Matynia, A., C. H. Ng, W. Dansithong, A. Chiang, A. J. Silva, and S. Reddy. 2010. Muscle- blind, but not Dmpk or Six5, contributes to a complex phenotype of muscular and motivational deficits in mouse models of myotonic dystrophy. PLoS ONE 5:e9857. 191. Matz, M. V., A. F. Fradkov, Y. A. Labas, A. P. Savitsky, A. G. Zaraisky, M. L. Markelov, et al. 1 999. Fluorescent proteins from nonbioluminescent Anthozoa species. Nat Biotechnol 1 7:969-73. 192. Meier, S., W. Neupert, and J. M. Herrmann. 2005. Proline residues of transmembrane domains determine the sorting of inner membrane proteins in mitochondria. J Cell Biol 1 70:88 1 -8. 193. Meola, G., and V. Sansone. 2007. Cerebral involvement in myotonic dystrophies. Muscle 6 6 • References

139 140 6 • References 196. Mitra, K., C. Wunder, B. Roysam, G. Lin, and J. Lippincott-Schwartz. 2009. A hyperfused hyperfused A 2009. r­ u c Na Lippincott-Schwartz. J. uscle m and l ta Lin, G. le ke S 1. Roysam, 200 B. Mounsey. C. P. Wunder, K., J. and Mitra, 196. Reddy, Specific S. 2009. Moorman, Hol. R. M. E. J. J., D. and Mistry, Speijer, D.195. Aronica, E. Berge, den van A. S. J., Middeldorp, 194. 199. Mounsey, J. P., J. E. John, 3rd, S. M. H elm ke, E. W. Bush, J. Gilbert, A. D. Roses, et al. al. et Roses, D. A. Gilbert, J. in Bush, W. E. inadequacy ke, elm H al n M. S. tritio u N 3rd, John, 2005. E. P., J. J. Tarnopolsky. A. Mounsey, M. 199. and MacDonald, R. 1 994. J. B., al. et Motlagh, Jat, S. P. 198. tis, talio A P. Peckham, M. Pagel, N. C. Beauchamp, R. J. E., J. Morgan, 197. 216. Oude Ophuis, R. J., M. W ijers, M. B. Bennink, F. A. van de Loo, J. A. Fransen, B. W ieringa, ieringa, W B. Fransen, A. J. Loo, de van A. F. Bennink, B. M. ieringa, W B. ijers, W M. J., R. Vorstenbosch, Ophuis, de Oude van R. 216. Herpen, van E. R. al. et Mulders, A. Gerwert, S. K. J., R. ann, Ophuis, errm H A. Oude 215. and Hofmann, W. M. Tissue 1 998. i. Nikolaus, akam J. Kaw K. Poschner, C. and B. J., Ikeda, K. Ollesch, 214. Kobayashi, HTS- M. Saito, 2010. T. al. et Takizawa, H., T. Chow, K. Ohto, S. 213. Sigal, J. Y. Garcia, E. M. Sharif, O. Vargas, L. A., D. O'Leary, al. 212. et Hodgson, M. D. actin-related of Behfar, A. Kane, C. G. structures Reyes, S. Crystal Perez-Terzic, C. F., 2004. D. Pollard. D.O'Cochlain, T. 211. and of Littlefield, Targeting S. R. 1 993. J., B. Shore. C. G. Nolen, 210. and Korsmeyer, J. S. Dystrophy Yong, W. V. Myotonic illar, M G. D. Human M., the Nguyen, of 209. syn- of Expression rm fo 2004. spliced Leung. T. ly e and tiv a Lim, rn L. lte a Tan, I. Y., an Ng, of 208. ent ecruitm R 1999. dystrophy illi. Cam De Myotonic P. and 2000. Y., Nemoto, Korneluk. G.207. R. and protein Sabourin, A. L. dystrophy Waring, D. Myotonic J. A., M. 1 999. Narang, Salvatori. 206. S. and Furlan, S. dystro­ Marin, O. Myotonic 1. Biral, 200 D. I., al. et Mussini, membrane- Epstein, 205. F. H. al Ito, M. peroxisom Hirano, K. for Liu, F. signals Zhang, R. A., sorting The Muranyi, J. S. 2000. 204. Trelease. N. R. Kuik-Romeijn, and van T., R. P. Croes, J. H. Mullen, 203. heeler, W M. T. Broek, den 1995. van al. J. et W. A., S. Roses, D. A. Mulders, 202. Gilbert, J. Horne, T. L. muscle 3rd, l ta le John, E. ke S J. Xu, P. P., 2000. J. Mounsey, Moorman. R. 201. J. and Reddy, S. Ai, W. C. Mistry, J. D. P., J. Mounsey, 200. Bo Ce 268:25265-8. Chem Biol J rs-ecs t nuoia n- i Azeme. LS N 4:e7663. ONE PLoS er. Alzheim in ent-L neurofilam ith w cross-reacts t l 20. ti-nhrd ytnc ytoh poen iae sfr nue peri­ induces isoform kinase protein dystrophy myogenic in myotonic expressed lly tail-anchored tia 40:545-55. A n re iffe 2009. Nerve d al. are et Muscle s isoform lineages. ll e protein c DMPK eural n 2009. and ansink. W G. D. and 4:9-18. Genomics Chem Curr 1. Type Dystrophy Myotonic in Splicing t berran A itochondrial m a of ain dom PDZ the ith 1 w 1-3006. 8:299 J EMBO interaction protein. the by ria brane em m itochond m outer to 2 aptojanin 31:713-8. Nerve Muscle dystrophy. lar scu u m ith w lts u ad 2008. Secondary s tructure and d istrib u tio n of fusogenic LV-peptides in lipid membranes. membranes. lipid in LV-peptides fusogenic 42:141-8. of iol B n tio Dev u J istrib d Int and products. tructure s gene ily m Secondary fa Six 2008. of n utio istrib d ental pm evelo d cardio­ hypertrophic into ulates accum DMPK an hum of overexpression Transgenic 2004. 1 64. 1 56-34 279:34 Chem Biol J Binding. Sp1 and tion 47:383-92. Cytochem istochem H J junctions. gap in xicity to RNA of rsal ve re ediated oligonucleotide-m Triplet-repeat 2009. al. et Kimpe, de Mol Hum kinase. protein dystrophy myotonic in Chem Biol deficient J mice kinase. in protein gating dystrophy channel myotonic sodium for substrate a is an m 275:23362-7. hospholem P 2000. cer lseig o miohnra atpay ad ppoi. LS N 4:e8024. ONE PLoS apoptosis. and autophagy, itochondria, m of clustering uclear n 105627-32. 1:1 A S U Sci Acad Natl Proc ADP. or ATP bound ith w plex com 2/3 protein Dev lines. ll e c utation-specific m and tissue-specific of phase. S erivation d into the for entry and system odel buildup m E cyclin ulates reg 6:153-8. G1-S at Genomics achieved hysiol P state rial deficiency. d on Six5 itoch m for heterozygous mice in rents myopathy, m yotonic m yopathy and hypotension tra its of m yotonic dystrophy. Hum Mol Mol Hum dystrophy. yotonic m of its tra hypotension and yopathy m yotonic m myopathy, 7­ 8:50 Genet Hum J Eur patients. DM lt u ad and congenital in ls ve le 512. kinase protein (DM) inhibits and subunit targeting 493:80-84. phosphatase Lett FEBS yosin m the activity. phosphatase phosphorylates myosin kinase protein phy 101 39 6:15-20. A S U Sci Acad atl N Proc dystrophy. yotonic m u n srct sbye eeld y fiiy uiid FP nioy uprfe srm serum unpurified antibody; GFAP purified affinity by revealed subtype astrocyte an hum 36:294-306. Nerve u Bohs 37:435-45. J Biophys Eur of odulators M olecule M ll a m S Identify to Assay Cell-Based Patient-Derived patible Com 8. 13:2505-1 Genet sequence. anchor al n sig COOH-terminal a by brane em m Methyla- outer DNA rial d on oter rom P itoch m by the to Regulated Bcl-2 Is {gamma} Kinase Cdc42-binding Kinase-related and m lu ticu re ic sarcoplasm ith w associated is uscle m cardiac rat in expressed 6337-44. 275:1 kinase Chem Biol J tail. inal C-term its in ith w are peroxidase ascorbate bound kinase. protein myotonin an hum by channels sodium uscle m l ta le e k C s of odulation M 1 3-20. 9:23 Genet a antigen: T ile b la o rm e th a carrying ice m transgenic 162:486-498. m fro Biol derived lines ll e c 1Myogenic 1 960-5. 06:1 A S U Sci Acad atl N Proc i Ivs 95:2379-84. Invest lin j

217. Pelloux, S., J. Robillard, R. Ferrera, A. Bilbaut, C. Ojeda, V. Saks, et al. 2006. Non-beating HL-1 cells for confocal microscopy: application to mitochondrial functions during car­ diac preconditioning. Prog Biophys Mol Biol 90:270-98. 218. Pham, Y. C., N. Man, L. T. Lam, and G. E. Morris. 1 998. Localization of myotonic dystrophy protein kinase in human and rabbit tissues using a new panel of monoclonal antibodies. Hum Mol Genet 7:1 957-65. 219. Philips, A. V., L. T. Timchenko, and T. A. Cooper. 1 998. Disruption of splicing regulated by a CUG-binding protein in myotonic dystrophy. Science 280:737-4 1. 220. Picard, D., E. Suslova, and P. A. Briand. 2006. 2-color photobleaching experiments reveal distinct intracellular dynamics of two components of the Hsp90 complex. Exp Cell Res 3 1 2:3949-58. 221. Pitts, K. R., M. A. McNiven, and Y. Yoon. 20 04. Mitochondria-specific function of the dynamin family protein DLP1 is mediated by its C-terminal domains. J Biol Chem 279:50286-94. 222. Rabu, C., and S. High. 2007. Membrane protein chaperones: a new twist in the tail? Curr Biol 17:R472-4. 223. Rabu, C., V. Schmid, B. Schwappach, and S. High. 2009. Biogenesis of tail-anchored proteins: the beginning for the end? J Cell Sci 1 22:36 0 5-1 2. 224. Ranum, L. P., and J. W. Day. 2002. Myotonic dystrophy: clinical and molecular parallels between myotonic dystrophy type 1 and type 2. Curr Neurol Neurosci Rep 2:465-70. 225. Ranum, L. P., and J. W. Day. 2004. Pathogenic RNA repeats: an expanding role in genetic disease. Trends Genet 20:506-1 2. 226. Reddy, S., D. J. Mistry, Q. C. Wang, L. M. Geddis, H. C. Kutchai, J. R. Moorman, et al. 2002. Effects of age and gene dose on skeletal muscle sodium channel gating in mice defi­ cient in myotonic dystrophy protein kinase. Muscle Nerve 25:850-7. 227. Reddy, S., D. B. Smith, M. M. Rich, J. M. Leferovich, P. Reilly, B. M. Davis, et al. 1 996. Mice lacking the myotonic dystrophy protein kinase develop a late onset progressive myopa­ thy. Nat Genet 1 3:325-35. 228. Reed, J. C., H. Zha, C. Aime-Sempe, S. Takayama, and H. G. Wang. 1 996. Structure-function analysis of Bcl-2 family proteins. Regulators of programmed cell death. Adv Exp Med Biol 406:99-1 1 2. 229. Reggiani, C., L. Brocks, P. Wirtz, H. Loermans, and G. te Kronnie. 1 992. Myosin isoforms m hindlimb muscles of normal and dystrophic (ReJ129 dy/dy) mice. Muscle Nerve 15:199­ 208. 230. Renthal, R. 2009. Helix insertion into bilayers and the evolution of membrane proteins. Cell Mol Life Sci 67:1 077-88. 231. Reynwar, B. J., G. Illya, V. A. Harmandaris, M. M. Muller, K. Kremer, and M. Deserno. 2007 Aggregation and vesiculation of membrane proteins by curvature-mediated interactions. Nature 447:46 1 -4. 232. Riento, K., and A. J. Ridley. 2003. Rocks: multifunctional kinases in cell behaviour. Nat Rev Mol Cell Biol 4:446-456. 233. Roberts, R., N. A. Timchenko, J. W. Miller, S. Reddy, C. T. Caskey, M. S. Swanson, et al. 1 997. Altered phosphorylation and intracellular distribution of a (CUG)n triplet repeat RNA-binding protein in patients with myotonic dystrophy and in myotonin protein kinase knockout mice. Proc Natl Acad Sci U S A 94:13221-13226. 234. Rojo, M., F. Legros, D. Chateau, and A. Lombes. 2002. Membrane topology and mitochon­ drial targeting of mitofusins, ubiquitous mammalian homologs of the transmembrane GTPase Fzo. J Cell Sci 1 1 5:1 663-74. 235. Romeo, V., E. Pegoraro, C. Ferrati, F. Squarzanti, G. Soraru, A. Palmieri, et al. 2010. Bram involvement in myotonic dystrophies: neuroimaging and neuropsychological comparative study in DM1 and DM2. J Neurol. 236. Ronchi, P., S. Colombo, M. Francolini, and N. Borgese. 2008. Transmembrane domain­ dependent partitioning of membrane proteins within the endoplasmic reticulum. J Cell Biol 181:105-18. 237. Rossi, A. E., S. Boncompagni, and R. T. Dirksen. 2009. Sarcoplasmic reticulum-mitochon­ drial symbiosis: bidirectional signaling in skeletal muscle. Exerc Sport Sci Rev 37:29-35. 238. Rossignol, R., R. Gilkerson, R. Aggeler, K. Yamagata, S. J. Remington, and R. A. Capaldi. 2004. Energy substrate modulates mitochondrial structure and oxidative capacity m cancer cells. Cancer Res 64:985-93. 239. Rotman-Pikielny, P., K. Hirschberg, P. Maruvada, K. Suzuki, I. E. Royaux, E. D. Green, et al. 2002. Retention of pendrin in the endoplasmic reticulum is a major mechanism for Pendred syndrome. Hum Mol Genet 1 1:2625-33. 240. Salvatori, S., D. Biral, S. Furlan, and O. Marin. 1 997. Evidence for localization of the myo­

tonic dystrophy protein kinase to the terminal cisternae of the sarcoplasmic reticulum. j Muscle Res Cell Motil 1 8:429-40. 241. Santel, A., S. Frank, B. Gaume, M. Herrler, R. J. Youle, and M. T. Fuller. 2003. Mitofusin-1

protein is a generally expressed mediator of mitochondrial fusion in mammalian cells. j 6 6 • References

141 142 6 • References 267. Stocchi, V., L. Cucchiarini, F. Canestrari, M. P. Piacentini, and G. Fornaini. 1 987. A very very A 1 987. Fornaini. G. and Piacentini, P. M. Canestrari, F. Cucchiarini, L. V., Stocchi, 267. 6. tdodr, . . P H Jp H P Wike, . . pebre, n T Walma . 1 994. n. an allim W T. and Eppenberger, M. H. -depen­ inkler, -calcium W P. low H. and Jap, H. P. M., High- A. 2008. Stadhouders, uscle Hajnoczky. G.262. nonm and and Csordas, G. uscle m Szanda, G. A., ooth sm Spat, of 261. sitivity Ca2+ sen 2003. al. Somlyo. et V. A. and P., Pedrazzini, A. E. Somlyo, Colombo, S.260. Lombardo, F. 1. 200 Francolini, al. M. et Hegde, Bcl- S. R. L., Lombardi, E. V. targets Snapp, Renna, R.259. M. FKBP38 r Manca, ito L. hib M. in Tedeschi, D. calcineurin Mancuso, M. G., Inherent Siciliano, 2003. 1 997. 258. al. et Nakayama. I. K. Saitoh, N. and M., Sasagawa, N. Shirane, 257. Yamamoto, M. Kameda, Raf-1 N. and Ishiura, Rac-1 S. M., 2000. a, okaw Epstein. Shim F. H. 256. and Dunne, W. P. TOM-independent and Walch, T. E. factor- Wang, W. M., Cytosolic izu, 2006. Shim 255. et Mihara. K. Wattez, and A. Otera, H. K., Maurage, A. C. for Setoguchi, Ghestem, 254. role A. A aschke, 2003. Schraen-M S. Epstein. F. H. and Sablonniere, B. N., ieringa, W B. The Sergeant, Kasten, 253. 2008. R. M. al. et McIntosh, D. Schmitt, A. E., D. P. H. Schulz, alska, 252. akw R M. Denic, V. 2000. Ven. Schmid, V. der Metz, van J. M., F. P. and Schuldiner, ann, 251. errm H H. Reimann, ary J. prim of Klasen, C. n Furst, atio O. D. R., transform Schroder, 250. Efficient 2000. receptor Kochanek. S. and insulin of Hertel, G., S. regulation Schiedner, t 249. berran A 1. 200 Cooper. A. T. t le and trip DNA Philips, A. S., R. 2003. Savkur, Festenstein. 248. R. e t a and v a c Webster, Z. lo b u Sharpe, T. s tial Everett, C. A., ifferen D cell Saveliev, 2003. 247. atogenic Endo. H. sperm and for in Kagawa, Y. required Seo, mRNA N. is pk Dm Six5 of Hamamoto, T. M., 2004. lysis a Reddy. Satoh, an S. 246. n and izatio Han, rid of J. hyb Paul, loss S. S., situ P. In Sarkar, Heterozygous 2004. 245. 488-93. Reddy. 283:1 2000. S. al. and Science et Han, Tsai, S., J. siecle. W. P. de C. Ai, Ito, fin Sarkar, Y. 244. the Han, at J. Appukuttan, B. S., P. phosphorylation Sarkar, xidative O 243. 1999. M. Saraste, 2.24 266. Steinert, H. 1 909. Myopathologische Beitrage 1. Uber das klin ische und anatom ische ische anatom und ische klin das sla­ sttran Uber o 1. p for Beitrage factor targeting a Myopathologische of 1 909. H. into Identification Steinert, 266. 2007. insertion Hegde. S. R. protein and S., Tail-anchored Stefanovic, 265. 2002. Stirling. C. J. al. and et Brownsword, Ruitenbeek, W. J. G. J., Steel, Heerschap, 264. A. Haan, de A. Oerlemans, F. Benders, A. K., Steeghs, 263. es E rqie a oe posta ltona mehns ih s needn o te SEC the of independent is hich w echanism m al n slatio sttran o p cytosolic and novel a rial requires itochond 03. m ER 89:93-1 yeast Cell combined ith w deficiencies. uscles m kinase in creatine Ca2+ responses ltered A 1997. 163:257-69. in aturation m mRNA tau icrotubule-associated m brain an hum of ysregulation D 1. 200 al. intermyofi- the of ation form 1 the 34:634-45. for prerequisite a is Z-discs ith w plectin of ssociation A Nat dystrophy. myotonic in resistance insulin ith w associated is splicing tive a rn alte Res Biophys 3. Biochem 422:909-1 itochondria. m in s isoform OPA1 protein in-related dynam the of 14:497-506. Disord uscul eurom N tissues. mouse lt u ad at o-ar eesdpae PC ehd o te eaain f h ms significant most the of separation the for 04. method 37:58-1 HPLC ervenheilkd N Z reversed-phase Dtsch ion-pair fast Myotoniker. der unds skelschw u M des i me rn poen neto it te R Cl 12: 1 147-59. 28:1 Cell ER. the into insertion protein brane em m l a n tio 1 -63. 44:5 alcium C Cell signalling. calcium rial d on itoch m of s echanism m dent Biol Cell J interactions. protein affinity low by cisternae ER stacked of ation Form 2003. 5:28-37. Biol Cell Nat apoptosis. its hib in and ria itochond m to 2 11:2105-16. Ther Gene Hum production. vector 13:1421-31. Genet Mol Hum iogenesis. sperm and l a iv rv u s tc ndra moahe. NS 9 1:5089-93. PNAS myopathies. rial d on itoch m Rev Physiol phosphatase. myosin and kinases, proteins, G by 325-58. 83:1 odulated m II: yosin m 110:2 43-55. Genet Mol 1 77-86. 89:1 Hum 1. type europhysiol N J dystrophy plasticity. yotonic m synaptic in kinase protein dystrophy yotonic m ciey Bice sr 41: 91-2 . 20 14- 19 :1 1 4 istry iochem B achinery. m repeats m ediate heterochrom atin-protein-1-sensitive variegated gene silencing. Nature Nature silencing. gene variegated atin-protein-1-sensitive heterochrom ediate m repeats 25:110-114. Genet Nat cataracts. lar cu o cause to sufficient is mice in Six5 tcodra cetn kns: mao cnttet f tooia icui s en m seen ns inclusio athological p of constituent ajor m a kinase: creatine rial itochond M dystrophy. yotonic m in expansion e tid o cle u 0. trin CTG 56:405-1 and ull B lactate Res 285-95. 1 50:1 Brain exercise l­ a Pathol c Q10, lo J is Am Coenzyme muscle. dystrophy l ta le yotonic e m k s of of m product lu ticu gene re ic kinase: protein plasm sarco protein dystrophy the in yotonin m myotonic of ized activate isoform 25:5635-47. Novel pathways, J EMBO 475:273-7. signaling proteins. Lett distinct FEBS brane of em m kinase. ponents outer com rial d on kinases, itoch m C-tail-anchored of port im Cell brane. em m ER the into proteins tail-anchored of insertion 80:455-64. ediates m Invest plex com Lab GET adenoviral cytoskeleton. for in lines desm ll r ce rilla b new of generation Ad5: of functions E1 by niocytes am an hum 29:40-47. Genet 300:482-93. Commun 1 1 6:2763-74. Sci Cell B i I c

nucleotides and their degradation products in human red blood cells. Anal Biochem 1 67:1 8 1 -90. 268. Suloway, C. J., J. W. Chartron, M. Zaslaver, and W. M. Clemons, Jr. 2009. Model for eu­ karyotic tail-anchored protein binding based on the s tructure of Get3. Proc N atl Acad Sci U S A 1 06:1 4849-54. 269. Suzuki, A., Y. Sugiyama, Y. Hayashi, N. Nyu-i, M. Yoshida, I. Nonaka, et al. 1 998. MKBP, a novel member of the small heat shock protein family, binds and activates the myotonic dystrophy protein kinase. J Cell B iol 1 40:1 1 1 3-1 1 24. 270. Tan, I., K. T. Seow, L. Lim, and T. Leung. 200 1. Intermolecular and intramolecular interac­ tions regulate catalytic activity of myotonic dystrophy kinase-related Cdc42-binding kinase a. Mol Cell Biol 2 1:2767-2778. 271. Tanaka, Y., Y. Kanai, Y. Okada, S. Nonaka, S. Takeda, A. Harada, et al. 1 998. Targeted disruption of mouse conventional kinesin heavy chain, kif5B, results in abnormal peri­ nuclear clustering of mitochondria. Cell 93:1 1 47-58. 272. Taneja, K. L., M. McCurrach, M. Schalling, D. Housman, and R. H. Singer. 1 995. Foci of trinucleotide repeat transcripts in nuclei of myotonic dystrophy cells and tissues. J Cell Biol 1 28:995-1 002. 273. Taylor, S. S., D. R. Knighton, J. Zheng, L. F. Ten Eyck, and J. M. Sowadski. 1 992. Structural framework for the protein kinase family. Annu Rev Cell Biol 8:429-62. 274. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1 994. CLUSTAL W: improving the sensi­ tivity of progressive multiple sequence alignment through sequence weighting, position- specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673-80. 275. Thornton, C. A., J. P. Wymer, Z. Simmons, C. McClain, and R. T. Moxley, 3rd. 1 997. Expan­ sion of the myotonic dystrophy CTG repeat reduces expression of the flanking DMAHP gene. Nat Genet 1 6:407-9. 276. Tian, B., R. J. White, T. Xia, S. Welle, D. H. Turner, M. B. Mathews, et al. 2000. Expanded CUG repeat RNAs form hairpins that activate the double-stranded RNA-dependent protein kinase PKR. RNA 6:79-87. 277. Timchenko, L., W. Nastainczyk, T. Schneider, B. Patel, F. Hofmann, and C. T. Caskey. 1 995. Full-length myotonin protein kinase (72 kDa) displays serine kinase activity. Proc Natl Acad Sci U S A 92:5366-70. 278. Timchenko, L. T., J. W. Miller, N. A. Timchenko, D. R. DeVore, K. V. Datar, L. Lin, et al. 1 996. Identification of a (CUG)n triplet repeat RNA-binding protein and its expression in myo­ tonic dystrophy. Nucleic Acids Res 24:4407-1 4. 279. Timchenko, L. T., N. A. Timchenko, C. T. Caskey, and R. Roberts. 1 996. Novel proteins with binding specificity for DNA CTG repeats and RNA CUG repeats: implications for myotonic dystrophy. Hum Mol Genet 5:115-21. 280. Timchenko, N. A., R. Patel, P. Iakova, Z. J. Cai, L. Quan, and L. T. Timchenko. 2004. Over­ expression of CUG triplet repeat-binding protein, CUGBP1, in mice inhibits myogenesis. J Biol Chem 279:1 3 1 29-39. 281. Toscano, A., S. Messina, G. M. Campo, R. Di Leo, O. Musumeci, C. Rodolico, et al. 2005. Oxidative stress in myotonic dystrophy type 1. Free Radic Res 39:77 1 -6. 282. Totsukawa, G., Y. Wu, Y. Sasaki, D. J. Hartshorne, Y. Yamakita, S. Yamashiro, et al. 2004. D istinct roles of MLCK and ROCK in the regulation of m em brane p rotrusions and focal adhesion dynamics during cell migration of fibroblasts. J Cell Biol 1 64:427-439. 283. Ueda, H., S. Ohno, and T. Kobayashi. 2000. Myotonic dystrophy and myotonic dystrophy protein kinase. Prog Histochem Cytochem 35:1 87-25 1. 284. Ueda, H., M. Shimokawa, M. Yamamoto, N. Kameda, H. Mizusawa, T. Baba, et al. 1 999. Decreased expression of myotonic dystrophy protein kinase and disorganization of sarco­ plasmic reticulum in skeletal muscle of myotonic dystrophy. J Neurol Sci 1 62:38-50. 285. van der Ven, P. F., G. Jansen, T. H. van Kuppevelt, M. B. Perryman, M. Lupa, P. W. Dunne, et al. 1 993. Myotonic dystrophy kinase is a component of neuromuscular junctions. Hum Mol Genet 2:1 889-94. 286. van Herpen, R. E., R. J. Oude Ophuis, M. Wijers, M. B. Bennink, F. A. van de Loo, J. Fransen, et al. 2005. Divergent mitochondrial and endoplasmic reticulum association of DMPK splice isoforms depends on unique sequence arrangements in tail anchors. Mol Cell Biol 25:1 402-1 4. 287. van Herpen, R. E., J. V. Tjeertes, S. A. Mulders, R. J. Oude Ophuis, B. Wieringa, and D. G. Wansink. 2006. Coiled-coil interactions modulate multimerization, mitochondrial bind­

ing and kinase activity of myotonic dystrophy protein kinase splice isoforms. FEBS j 273:1 1 24-36. 288. Vergeres, G., and L. Waskell. 1 995. Cytochrome b5, its functions, structure and membrane topology. Biochimie 77:604-20. 289. Vihola, A., G. Bassez, G. Meola, S. Zhang, H. Haapasalo, A. Paetau, et al. 2003. Histopatho- logical differences of myotonic dystrophy type 1 (DM1) and PROMM/DM2. Neurology 60:1 854-7. 290. Visch, H. J., G. A. Rutter, W. J. Koopman, J. B. Koenderink, S. Verkaart, T. de Groot, et 6 6 • References

143 144 6 • References 296. W ansink, D. G., R. E. van Herpen, M. M. Coerw inkel-Driessen, P. J. Groenen, B. A. Hem- Hem- A. B. Groenen, J. P. inkel-Driessen, Coerw M. C. M. Herpen, anova, Akhm van A. E. G., R. D. Splinter, D. ansink, W of 296. Schlager, A. M. balance paired Im Vorstenbosch, de 2009. van al. R. F., B. et Smith, A. Wanschers, M. 295. Perry, G. Cooper. Li, A. X. T. Lee, and G. H. Su, B. X., Wehrens, H. X. Wang, the 294. of Mathur, N. function Sarma, S. and artinez, Targeting cu-M 2009. Kuyum N. M. Suter. S., G. U. and Wang, 293. Niemann, A. Ruegg, ic M. M., K. endoplasm The Wagner, 1292. 992. al. et Panzeri, C. M. Nori, A. Martini, A. Podini, P. Villa, A. P., Volpe, 291. 299. W esterlaken, J. H., C. E. Van der Zee, W. Peters, and B. W ieringa. 2003. The DMWD protein protein DMWD The 2003. ieringa. W B. dystrophy and myotonic Peters, W. for Zee, der odels m Van E. C. H., mouse J. Transgenic esterlaken, W 2003. 299. ieringa. W B. G., and D. pathophysi­ and ansink, W al 298. orm N 2006. ieringa. W B. and Herpen, van A. M. E. G., R. D. ansink, W 297. 313. Zhao, Z. S., and E. Manser. 2005. PAK and other Rho-associated kinases--effectors w ith ith w kinases--effectors Cell Mol Rev Rho-associated Nat other and apoptosis. PAK in 2005. fission rial Manser. d E. on and itoch S., Z. M Zhao, 2005. 313. protein ski. rial d on Karbow M. itoch m and by J., The R. 2003. Youle, 312. autophagy McNiven. A. M. onitoring M and 2009. Oswald, J. B. biogen­ Eskelinen. L. A E. Krueger, icroRN W. E. and m Y., , Yoon, and 311. kitalo Jo E. interference Vihinen, H. P., (miRNAs). RNA RNAs ediated la-Anttila, Y micro 310. Intron-m intronic 2009. in Lin. L. S. and Y., S. perspectives Ying, 309. Current 2006. Lin. L. S. icrotubules M and Y., 1 S. 996. Ying, Nurse. 308. P. and Toda, T. Eddison, M. Verde, F. Harata, D. P., M. Yaffe, 307. 12-residue- A 1996. Heijne. von G. and Rapoport, al. A. T. et Kutay, U. Staines, A. Grahn, W. E. P., Hincke, M. hitley, uscu­ W 302. eurom erville, n Som the J. M. at foci K. Tamai, Waring, D. ibonuclear R J. J., E. 2007. hiting, W Thornton. 301. A. C. and Krym, C. M. M., T. heeler, W 300. 0.Ydv, . . C D FezlMcadl, . u V Siiaa, . . ukr J Pyia, et Puymirat, J. Tucker, L. A. Srinivasan, V. Yu, Q. cCardell, Frenzel-M D. C. S., R. Yadava, 306. Charac­ 1999. Johnson. J. K. and Clark, J. B. Sermoni, A. Ferrier, K. R. L., C. inchester, W 303. 305. Yabal, M., S. B ram billasca, P. Soffientini, E. Pedrazzini, N. Borgese, and M. M akarow. 2003. 2003. akarow. M M. and Borgese, N. Pedrazzini, E. Soffientini, P. and th grow billasca, ram B S. M., Postnatal 1Yabal, 983. 305. Reintjes. G. A. and Peer, G. P. Loermans, M. H. P., irtz, W 304. T. A shiza w a (ed.), Genetic In s ta b ilitie s and N eurological Diseases, 2nd ed. Elsevier- Elsevier- ed. 2nd Diseases, eurological N and s ilitie b ta s In Genetic (ed.), a w shiza A T. Translocation of the C te rm in u s of a tail-anchored protein across the endoplasm ic ic endoplasm the across protein tail-anchored a of s u in rm te C the of Translocation olog ical significance of myotonic dystrophy protein kinase, p. 79-97. In R. D. W e lls and and lls e W D. R. In 79-97. p. CA. Diego, kinase, San Inc., protein Press, ic dystrophy Academ myotonic of significance ical olog electron m icroscopy in M a m m alia n cells. Methods Enzym ol 452:1 43-64. 452:1 ol Enzym Methods cells. n alia m m a M in icroscopy m 13. electron 487:387-4 Biol Mol Methods esis. rodent and an hum in protein (DMK) kinase dystrophy myotonic of Characterization 1995. 142-6. 89:6 A ATP agonist-induced restores Na+-Ca2+ exchange rial itochond m of Inhibition 2004. al. al. 2008. RNA to xicity in myotonic m u scu lar dystrophy induces NKX2-5 expression. Nat Nat expression. NKX2-5 induces dystrophy lar scu u m myotonic in xicity to RNA 2008. al. rm te ooi dsrpy DM1 gn rgo i deel ment l rgltd n is and regulated lly ta n e pm lo eve d is region gene 1) M (D dystrophy yotonic m 1 the 00:230-42. Res from Genome Cytogenet (DM1). 1 type 31 3:3408-20. Res Cell Exp cells. neuronal in rt o transp myotonic of odel m mouse a in 1 1 phenotype 9:3797-806. Invest cardiac Clin the J s 1. rate lio type e am dystrophy inhibition PKC 2009. su rp risin g ly diverse m echanism s of regulation. Biochem J 386:20 1 -1 4. 386:20 J Biochem regulation. of s echanism m diverse ly g risin rp su 23:5409-20. Biol Cell Mol DLP1. protein like in- ynam d the terizatio n of the expression of DMPK and SIX5 in the hum an eye and im p licatio n s for for s n licatio p im and eye an hum the in SIX5 and DMPK of expression the of n terizatio differentiation of m uscle fib res in the mouse. I. A histo chem ical and m o rp hom etrical etrical hom rp o m and ical chem histo A I. mouse. the in res fib uscle m of differentiation 501. 4:e5160. ONE PLoS tail-anchor. its S on U Sci dependent Acad are Natl GDAP1 Proc fibers. factor uscle reticulum fission m l ic rial ta d le on e k s endoplasm itoch of m of m n lu utio ticu re istrib d 279:40328-36. ic plasm Chem connection: sarco Biol m lu u the J tic in re ic deficiency. I arkers lasm m plex sarcop - com an ticulum re hum in Ca2+ handling and production dae tcodra diti ton n iso yat Po Nt Aa Si S 9: 1664­ 4 1 6 6 93:1 A S U Sci Acad Natl Proc yeast. fission in n utio istrib d rial 8. itochond m ediate m 072. 063-1 4:1 Genet Mol Hum system. nervous tral n e c and uscle m 1 1:16-27. 97 Res Brain areas. brain synapse-dense in inently rom p most present protein dystrophy yotonic m controls splicing tive a rn lte A 2003. ieringa. W B. and 03. mings, 29:9090-1 Neurosci J disease. r's e im lzhe A in fusion and fission rial d on itoch m mem brane. J Biol Chem 27 1:7583-6. 27 Chem Biol J brane. mem re tic u lu m m em brane in yeast m utants defective in sig n al peptide-driven translocation. J J translocation. peptide-driven al n sig in defective utants m yeast in brane em m m lu u tic re 1 8:48 -492. Genet Mol Hum dystrophy. yotonic m in pathogenesis iae tutr, ny tc ciiy ad l ar oaiain Ml el il 54 9­ 48 :5 3 2 Biol Cell Mol localization. r la llu e c b u s and activity, atic retrograde enzym in structure, interaction kinase Bicaudal-D1 Rab6B the for role A 2007. al. et Hoogenraad, C. hFis1 reg ulates m itoch on d rial fission in m am m alian cells through an interaction with with interaction an through cells alian 6:657-63. m am m Biol in fission rial d on itoch m ulates reg hFis1 B 40:61-8. Genet reticulum ic endoplasm the to synaptobrevin anchor to sufficient is il ta polyleucine long 1 2-7. 7:24 Disord uscul eurom N 1. type dystrophy myotonic in junction r la il hm 278:3489-96. Chem Biol netgto o nor l uce J nt 3 (t 109-26. ):1 (Pt 137 Anat J muscle. al rm o n of investigation oe Si 13:5-15. Sci iomed j

146 6 • Nederlandse samenvatting (NB: bij de mens aan mitochondriën). Een minder hydrofobe C-terminus C-terminus hydrofobe minder Een mitochondriën). aan mens de bij (NB: ankert membraan ER het in die C-terminus hydrofobe een B bevatten enA NEDERLANDSE• SAMENVATTING a d ioomn eid zc en SG-euni an e ene van einde het aan VSGGG-sequentie een zich bevindt isovormen de helft de In van coiled-coil-domein. een en kinasedomein serine/threonine een bekend. weinig ei- maar nog individuele tevens is van DMPK-gen het functie van en expressie witproducten de Over mogelijkheid. deze aan indirect- of -direct namelijk kan DMPK-gen het in (CTG)n-repeat de van een met DMPK-gen mutant een van afkomstig Transcripten DM1. van resulteert en DMPK-gen het van deel eiwitcoderende het buiten situeerd die gevonden repeatexpansies personen aangedane bij (CTG)n-tripletten 35 worden tot 5 voorkomt, van repeatlengte een mensen gezonde in Waar eshled slc-svre vn e DP-ii i me dti bestu­ detail meer in DMPK-eiwit het van splice-isovormen verschillende eiwitten. verankerde C-terminaal zogenaamd van aminozuren twee slechts van C-terminus F, een en die E isovormen voor mitochondriële het aan bindt die D, en C isovormen bij gevonden wordt er. m e eae o ht MKewt oeik e rl a see bij spelen kan rol een mogelijk DMPK-eiwit het of bepalen te Omdeerd. N-terminus, hoofdeiwi- leucine-rijke een zestal een allen van bevatten expressie isovormen de Deze in tisovormen. resulteert wat splicing, tieve besteed eiwit- DMPK aandacht hebben van studies expressie weinig de ook Slechts beïnvloeden. uiteindelijk producten daarmee en transcripten Expansie rol. pathobiologische een abnormale ook zelf heeft DMPK-gen het Mogelijk van genen. expressie andere transcriptie- aan en van verscheidenheid splice- expressie de grote van een bepaalde misregulatie tot leidt staat hetgeen vangen te in af factoren zijn repeat geëxpandeerde de symptomen van abnor­ merendeel het een aan door ligt grondslag geïnduceerd ten al­ istranscript, wordt maal ziektebeeld Inmiddels een eiwit. waarbij proces mutant een dominant-RNA-pathogenese-mechanisme, een een dat van geaccepteerd expressie gemeen de in niet daardoor DMPK-gen. het van exon Laatste CTG- het in vele uit tandem- in bestaande nucleotide-tripletten sequentie DNA -een repeat (CTG)n-tripLet een lend voor lokalisatie en functie en komt voor in drie varianten. Isovormen Isovormen varianten. drie in voor bepa­ komt en meest en functie het is lokalisatie DMPK voor lend van C-terminus uiterste De domein. kinase het mRNA DMPK van oftransport splicing, transcriptie, veranderde tot leiden ge­ is repeatsequentie De repeats. duizenden enkele tot oplopen kunnen n i pofcrf zj d lklste n ucinl egncapn van eigenschappen functionele en lokalisatie de zijn proefschrift dit In groep de in ingedeeld A-D DMPK's de worden membranen aan A-D isovormen van intracellulaire binding van gevonden wijze typische de wordt Vanwege bezitten. lokalisatie cytosolische Een (MOM). buitenmembraan van expansie de door veroorzaakt wordt (DM1) 1 type dystrofie Myotone rmie MK rncitn in nehvg a uterie alterna­ uitgebreide aan onderhevig zijn transcripten DMPK Primaire de DM1-pathogenese is het belangrijk te weten in welke weefsels van het lichaam en in welke relatieve hoeveelheden het eiwit voorkomt. In Hoofd - stuk 2 is de expressie van het DMPK-eiwit onderzocht en in het bijzonder aandacht besteed aan de verhouding van de lange (A-D) en korte (E, F) isovormen. Het onderzoek heeft aangetoond dat de ratio van de DMPK- isvormen in verschillende spierweefsels variabel is en dat de verhouding gevonden in skeletspieren overeenkomt met die in primaire spiercellijnen. Daarnaast is gekeken naar de DMPK-expressie in verscheidene brein- regio's. Lange DMPK-isovormen komen in brein het meest tot expressie en nemen in hoeveelheid toe tijdens breinontwikkeling. Expressiestudies in primaire neurale cellijnen lieten zien dat DMPK voornamelijk voor­ komt in corticale astrocyten. Nogal verrassend was de observatie dat primaire hippocampale neuronen slechts een zeer lage DMPK-expressie lieten zien. Deze vinding impliceert dat de breinsymptomen gevonden in DM1 gerelateerd zouden kunnen zijn aan veranderde DMPK-expressie of processing in tenminste beide celtypes of zelfs in astrocyten alleen en niet in neuronen. In een studie van YFP-DMPK fusie-eiwitten in primaire spier- en breincellen werden geen verschillen in subcellulaire lokalisatie gevon­ den tussen DMPK isovormen. Dit suggereert dat de bij DMPK-binding betrokken membraaneigenschappen en targeting machinerie -en daar­ mee de subcellulaire lokalisatie van DMPK's- niet celtype of celsoort afhankelijk zijn. De subcellulaire targeting van DMPK is verder onderzocht in Hoofd­ stuk 3 aan de hand van transductie-expressiestudies met muize- en humane DMPK-isovormen. Voor de humane DMPK C isovorm werd lokali­ satie op het MOM gevonden. Binding bleek bepaald te worden door zowel het coiled-coil domein als ook door de C-terminus. Eenzelfde lokalisatie was ook gevonden voor muis DMPK C en is dus niet soortafhankelijk. Voor lokalisatie van de humane DMPK A isovorm op het MOM is slechts de C-terminus vereist. Deze eigenschap wijkt af van die van de ortholoog van DMPK A van de muis, welke lokaliseert op het ER. Analyse wijst uit dat op aminozuurpositie 600 van de humane DMPK A-sequentie een arginine (R) aanwezig is in plaats van een alanine (A) residu in DMPK A van de muis. Dit zorgt ervoor dat het hydrofobe domein in de C-terminus van hDMPK A onderbroken wordt, waardoor twee gesplitste hydrofobe regios aanwezig zijn. Kennelijk zorgt dit voor een MOM-lokalisatie. Na­ dere analyse van de C-termini van de muize- en humane DMPK A liet zien dat een lange hydrofobe regio leidt tot ER-lokalisatie. Het opbreken of verkorten van de hydrofobe regio resulteert in een lokalisatie op het MOM of in het cytosol. Het grote belang van individuele aminozuurposi- ties in C-ankers is hiermee nogmaals aangetoond. In een topologiestudie waarbij DMPK gemerkt werd met fluorescente antilichamen na selectieve permeabilisatie van het celmembraan hebben we aangetoond dat zowel bij de muize- als de humane DMPK C isovorm het uiterste deel van het C-terminale uiteinde zich in het cytosol bevindt. Voor de muize- en de humane DMPK A-isovorm is het uiteinde van de C-terminus in het ER- lumen (muis) of de mitochondriële intermembrane space (IMS; mens) gelokaliseerd. De expressie van de humane DMPK A isovorm op het MOM kan daarbij specifiek aggregatie van mitochondriën induceren. In cellen met zulke geaggregeerde mitochondriën leek de C-terminus van hDMPK 6 6 • Nederlandse samenvatting

147 148 6 • Nederlandse samenvatting zc i ht yoo t bvne. arcinik ef dt e ae met maken te dit heeft Waarschijnlijk bevinden. te cytosol het in zich A (mDMPK C, mDMPK A en C dubbel transfectie, en WT) werden opvallend opvallend werden WT) en transfectie, dubbel C en A C, mDMPK (mDMPK do­ coiled-coil het Ook speelt. rol geen hierbij daarin, (CUG)n-repeat voor primaire myotubes. Voor alle mDMPK C-expresserende myotubes myotubes C-expresserende mDMPK alle Voor geobserveerd myotubes. vroeger als primaire gevonden WT voor en KO DMPK tussen verschillen mi- en cytosolaire de op isovormen E en C afwezigheid of A, aan mDMPK van afzonderlijke van langetermijneffecten om mogelijk eerst het voor SV40TAg temperatuur-sensitief een voor codeert welke gen een met zen dit op wij gaan 5 Hoofdstuk In DMPK. zonder spiercellen in veranderd erg isovormen DMPK verschillende de van ratio de ook en sieniveau's in­ een door gefaciliteerd wordt en na op uren expressie enkele A na hDMPK al van treedt aanvang Aggregatie aggregatie. mitochondriële voor voor is verantwoordelijk zelf A we hDMPK konden van C-terminus de dat M.b.v. truncatie-analyse celdood.aantonen uiteindelijk en aggregatie aal (MMP). Opvallend was het feit dat wanneer de mitochondriën slechts slechts mitochondriën functie de wanneer dat membraanpotenti- feit mitochondriële het was mitochondriële van de Opvallend (MMP). van aal verlies verlies een het liet andere onder zien, werk Dit bestudeerd. aggregatie zijn mitochondriële van gevolgen fysiologische de Ook voorkomen. vretle lseig a efce gvne, e vrage maximale verlaagde tegen­ met een gevonden, effecten werd van van ionomycine clustering met verwijdering overgestelde stimulatie versnelde Bij een Ca2* optrad. en Ca2*cytosolisch piek maximale verhoogde een kwalitatieve dezelfde werden myotubes door de In bestuderen. geïmmortaliseerde te gebruikte ons gedetailleerd calciumhuishouding tochondriële het maakt celsysteem dit van Gebruik terug. differentiatie vermogen hun terminale cellen detot krijgen en tempera­ hoge Bij geinactiveerd eiwit het TAg ongelimiteerd. wordt tuur groeien myoblasten afgeleide temperatuur lage muizen van relatief bijdeze eiwit dit van aanwezigheid In eiwit. mui­ transgene van afgeleid daarvoor werden myotubes WT en myotubes van betrokkenheid de naar beschreven studie een wordt en verder thema expres- de van regulatie dat aan geven resultaten Deze mitochondriën. Ook energieproductie. aan verlies een voor waarbij compenseren reactie te een hier tracht het cel de betreft Waarschijnlijk aggregatie. chondriële apoptose. voor is marker een hetgeen IMS, de uit vrijko­ c het voor cytochroom werd van gevonden men deze ook en wat bezitten iets MMP aggregatie, na normale verliezen een pas wel nog deze waren gefragmenteerd microtubu­ dit van structuur de van Verstoring netwerk. microtubuli tact belang van niet een bleek bezit, niet dan eigenschappen al dimeriserende met welke wordt, mein, vertaald de eiwitdomein van het structuur waarvan afwijkende een mRNA dat en aggregatie mitochondriële deze mitochondriële door gevolgd wordt proces dit en mitochondriën van tatie eagik in or e otma fntoee vn e cel. de van functioneren optimaal het voor zijn belangrijk fragmen­ tot leiden te eerst blijkt A hDMPK van Expressie 4. Hoofdstuk aggregatie. na integriteit mitochondriële van verlies het identieke responsen waargenomen, waarbij na acetylcholine-stimulatie acetylcholine-stimulatie na waarbij waargenomen, responsen identieke KO DMPK Primaire calciumresponse. de bij isovormen DMPK individuele functionele langer niet van opruimen het bij zijn betrokken autofagie kan mito- tijdens toe autofagie nam heeft Verder en gevolg. apoptose tot tot dus celdood leidt A uiteindelijk hDMPK van expressie verhoogde Een niet deze kan maar aggregatie mitochondriële de vertraagt netwerk laire i suis i ht eldn s eed a d climeuai is calciumregulatie de dat bekend is verleden het uit studies Uit in onderzocht nader is aggregatie mitochondriële van fenomeen Het cytosolische calcium en vertraagde verwijdering, weer uniek voor cellen met mDMPK C. Verder werd voor de mDMPK A isovorm een verandering geconstateerd wat betreft mitochondriële Ca2*-regulatie in de vorm van een lagere maximale piek na acetylcholine-stimulatie. Wij concluderen dat de gevonden effecten te maken hebben met de lokaal membraan-ge- bonden enzymatische activiteit van DMPK, en dat deze niet het indirecte gevolg zijn van een door DMPK-isovormen geïnduceerd verschil in dif- ferentiatiegraad of globale energiestatus. Onze resultaten zijn verrassend omdat (i) de MOM-gelokaliseerde DMPK C isovorm zijn effect sorteert op de snelheid van vrijkomen en opname terug in de SR Ca2* stores van Ca2* en (ii) de SR-gelokaliseerde DMPK A isovorm juist een effect heeft op de maximale mitochondriële Ca2* opname. Het vrijkomen van calcium in het cytosol en opname van calcium in het ER vindt voornamelijk plaats in gebieden waar ER en mitochondriën zich dicht bij elkaar bevinden. We moeten dus aannemen dat de effecten van mDMPK C activiteit tussen deze organellen kan worden uitgewisseld. Ook moet in onze verklaring rekening gehouden worden met de mogelijkheid dat lichte overexpressie van de DMPK isovormen -wat onvermijdbaar is in onze experimentele opzet- de vindingen heeft beïnvloed. Verder kunnen er diersoortverschil- len zijn in de rol van DMPK, waarbij de ER-lokalisatie van de DMPK A van muis kan duiden op een specifieke rol voor deze isovorm alleen in dit zoogdier.

Samenvattend; in dit proefschrift tonen we aan dat DMPK isovormen specifiek tot expressie komen in een groot aantal spierweefsels en brein- regio's. Deze isovormen bevinden zich op verschillende locaties in de cel, gekoppeld via gedefinieerde membraanankers. Dit maakt DMPK tot een aantrekkelijk modelsysteem voor het bestuderen van de (patho)biolo- gische rol en wijze van membraanbinding van C-terminaal verankerde eiwitten. De verschillen in lokalisatie van de zeer homologe muis- en humaan DMPK A-isovormen maken dit nog extra interessant. De cel- en weefsel-afhankelijke expressie van isovormen impliceert een alterna­ tieve functie voor DMPK op verschillende locaties. Mogelijk modificeert DMPK op de diverse locaties alternatieve substraten en is een juiste verdeling over intracellulaire membranen belangrijk. Mogelijk heeft een disbalans in de expressie van DMPK isovormen daardoor aanzienlijke consequenties voor de fysiologische integriteit van bepaalde cellen. Voortgezet onderzoek naar DMPK-expressieniveaus en spliceproducten in DM1 -patiënten zal voor de toekomst waardevolle informatie kunnen op­ leveren betreffende de mogelijke betrokkenheid van een DMPK-disbalans in de manifestatie van DM1. 6 6 • Nederlandse samenvatting

149 150 6 • Abbreviations AGC cAMP-dependent protein kinase/protein kinase kinase kinase/protein protein cAMP-dependent Acetylcholine AGC ACh RCC2 rlaeatvtd Ca2* influx Ca2* release-activated 1 Channel Chloride CRAC ClC-1 ABBREVIATIONS• DP os Mooi Dsrpy rti Kinase Protein Dystrophy Myotonic mouse mDMPK Kinase 1 Citron protein CUG-binding 2A subtype Charcot-Marie-Tooth CUGBP1 ETR3-like CUG-BP and CRIK CMT CELF1 YTMoi Popaae agtn Subunit Targeting Phosphatase Myosin kinase Chain Light Dbf2-Related Regulatory Nuclear Myosin Cdc42-Binding Kinase-Related Dystrophy Myotonic Membrane Outer kinase NDR Mitochondrial Potential Membrane Mitochondrial MYPT Protein MRLC Binding DMPK Mitofusin MRCK MOM Muscleblind Suppressor Tumor MMP Large Knockout MKBP dead Kinase Receptor Mfn Insulin lonomycin MBNL LATS Tris Hepes protein-2 KO requirement Protein Shock temperature High Heat Kd Serum Horse IR Region Hydrophobic lono Kinase Reticulum Protein HtrA2/Omi Dystrophy Endoplasmic Myotonic human HT Microscopy Electron HSP HS Solution Salt Balanced HR Earle's Kinase Protein hDMPK coupling Dystrophy Myotonic 2 type Excitation-Contraction Dystrophy Myotonic ER 1 type Dystrophy Myotonic EM Receptor EC-coupling Dihydropiridine EBSS DMPK DM2 DM1 DHPR C2] iohnra [Ca2*] Mitochondrial [Ca2*] free Sarcoplasmic [Ca2*]m [Ca2*]c /rti kns C kinase G/protein Kinase ORF Open Reading Frame OSER organized smooth ER PBS Phosphate-Buffered Saline PLM Phospholemman PLN Phospholamban PM Plasma membrane PMCA Plasma membrane Ca2*-ATPase Rhod-2 AM Rhodamine-2 acetoxymethyl ester RNP Ribonucleoprotein ROCK Rho-associated Coiled-coil Containing Kinase ROS Reactive Oxygen Species SERCA Sarco/endoplasmic reticulum Ca2*-ATPase SP1 Specific Protein 1 SR Sarcoplasmic Reticulum SRF Serum Response Factor SRP Signal Recognition Particle STAT Signal Transducers and Activators of Transcription Syb2 Synaptobrevin 2 TA Tail-Anchored WT Wild type ZNF9 Zinc Finger Protein 9 < 6 6 • Abbreviations

151 152 6 • Dankwoorc j' enwod a wre. e i en ek, neesne n f n toe en af en interessante leuke, een is Het worden. kan beantwoord 'ja' e eeedn e enthousiasme. en begeleiding je een boekje dit dat gezorgd ervoor hebben inzicht en raad Je geduld. je moe­ heb spiegel de in maal menig ik waarin tijd Een geweest. tijd zware een met die vraag een eindelijk Nu gespookt. hoofd mijn door ook zeker ee en tk idr xeietn ean ane e it n e week­ het in niet je wanneer gedaan experimenten minder stuk een zeker was het altijd erg gezellig in het u-tje, menig onderwerp is de revue revue de is onderwerp menig u-tje, het in gezellig erg altijd het was erg heel natuurlijk Rick ik wil Verder geworden. is geheel en samenhangend promotieonderzoek mijn tijdens gekregen heb Bedankt ik beginnen. die te vrijheid de promotieproject voor mijn Celbiologie bij om gelijkheid afkomen?” ooit dan het "Gaat vraag: de heeft vragen wetenschappelijke alle praktische hulp bij het M&N verhaal, voor mij echt een milestone milestone een echt mij voor verhaal, M&N het bij hulp praktische alle rsoi, e voled de it er elrn a ga. eak voor Bedankt gaan. zal mi­ van verloren geworden meer fan niet ik die ben jou voorliefde Door een bedanken.croscopie, graag ik daar. en wil hier Jack Ook experimentje een voor tijd zelfs tussentijds met gepasseerd, Daarnaast geweest. leerzaam erg is heb, voorgehouden dieje malen enkele spiegel de Ook me geleerd. veel ik heb doen, experimenten van manier mo­ de voor bedanken Bé graag ik wil Allereerst gebeurd. hulp nodige terug op plezier met en geleerd heb van veel ik waar maar kijken, ten vele Naast definitief. ineens en zolang gekomen waar plots moment is Een is, afgekomen. gehikt toch tegenaan dan is 'boekje' het plaats, • DANKWOORD up n eelged en b-j o zj tj i lid od oka regent al goed, ook altijd is tijd op zijn bbq-tje (een gezelligheid en geclusterde de voor hulp bedanken groep met DM de van EM een de rest cellen graag ik wel of Ook bevat. hoofdstuk wil ieder plaatje bijna myotubes dat wel is wat resultaat eens Het mitochondriën. maar nog met dragen zijn. wil paranimf je dat menig Leuk zuiden. voor het ook richting Bedankt is. treinreis hoofdstuk gezellige gepubliceerde eerste mijn omdat samenwerking de voor bedankt Ook hebt. gedaan mij voor 'effe' wat end drijvende de van een bent je Susan, toren! de van verdieping gezelligste a e oiejrn ad ekn s a eneik e hleuate p zijn op halleluja-tje een eindelijk dan is werken hard jaren nodige deNa ute E e atfgsmn eln He wt ee ka i we aan­ weer ik kwam keren wat Heel tellen. autofagosomen en EM uurtjes bedanken. Rick: van jouw begeleiding, gedrevenheid en je pragmatische pragmatische je en gedrevenheid begeleiding, jouw van Rick: bedanken. de zonder helemaal niet natuurlijk is werkje dit van komen stand tot Het van stukje gelezen) dankwoord. meest het schrijven: blijkbaar te (en boekje het laatste het om dus tijd Nu kijk. rsna n h cr fr M! edrwl k ra Rnk bdne voor bedanken Rinske graag ik bij succes wil Veel zijn. Verder DM1! for wilt cure the en paranimf je dat Prosensa blij ben Ik verhaal. M&N het bij ik had Verder initiatief. je voor bedankt sfeer, goede deze achter krachten Daarnaast wil ik ook graag Mietske heel erg bedanken voor de vele vele de voor bedanken erg heel Mietske graag ook ik wil Daarnaast de verdieping, 6e de op gehad zin mijn naar erg altijd het heb Ik het). Walther, ik wacht op een volgende productie met jou in de hoofdrol en Lieke, veel succes met de afronding van jouw boekje. René, ik heb het onderzoek van je overgenomen. Bedankt voor die basis, het wegwijs maken op het lab en je samenwerking, oa. bij het M&N verhaal. De sfeer op het lab is altijd erg goed geweest, daar hebben de nodige labdagen, paaslunches, kerstdiners en filmavonden zeker aan bijgedra­ gen. Ook tijdens de lunch en vrijdagmiddagborrel heb ik veel gelachen. Dit heeft er in ieder geval voor gezorgd dat ik me meteen heb thuis ge­ voeld op de 6e. Ik wil iedereen van Celbiologie, nog present of inmiddels verkast, bedanken voor een goede tijd. Marieke, Mirthe, Magda, Irene, Ad, Edwin, Mariska, Yvet, Marloes, Lieke vdB, Femke, Helma, Sharita, Gönöl, Gerda, Jan S. en Remco bedankt! Wilma, ons helaas veel te vroeg ontval­ len. Je gezelligheid zal ik nooit vergeten. Huipie, bedankt voor de hulp tijdens het microscopen, wanneer er weer eens een schuifje of zo ver­ keerd stond en er niks te zien was. Op naar de zeven min ;-). Daarnaast, Frank bedankt voor de HPLC metingen op de myoblasten en myotubes, en de hulp bij het bijbehorende bepalen van celvolumes met jazeker, een schuifmaat. Basic celbiologie dacht ik zo! Wiljan, je commentaren tijdens werkbesprekingen hebben zeker geholpen en mocht je nog een artikeltje nodig hebben, ik hoor het wel :-). Jan K. en Bas, erg fijn dat ik bij een van jullie kon pitten na weer een kerstdiner of labstapavond wanneer ik niet helemaal meer capabel was om nog naar Wageningen of later Den Bosch terug te keren. Mijn tijd op de flex-plek tijdens het schrijven van dit proefschrift is ook erg gezellig geweest. Gert-Jan, Linda, Monique, Ineke en Michiel bedankt voor de goede sfeer. Ook de 'de Boertjes' Alwin, Godfried, Marieke en Peter bedankt. De Moldieren wil ik natuurlijk ook niet vergeten. Jullie hebben er allemaal aan bijgedragen dat de 6e ver­ dieping de beste verdieping van de toren is. Jessica nog bedankt voor het carpoolen vanuit Wageningen, het reizen was erg gezellig zelfs met de onvermijdelijke files voor de Waalbrug. Ook mijn studenten Daan en Lisette wil ik bedanken voor hun inspan­ ningen zodat hoofdstuk 3 is geworden tot wat het nu is. Ik vond het erg gezellig om met jullie samen te werken en jullie te begeleiden, hier heb ik een hoop van geleerd. Daan, veel succes nu met jouw promotie. Lisette, ik heb weinig studenten rond zien lopen met meer motivatie en arbeidsethos dan jij. Houd ook tijd over voor leuke dingen :-). Verder wil ik nog Fons en Miranda bedanken voor hun samenwerking met het ontwikkelen voor de DMPK adenovirussen. Deze zijn zeer veel gebruikt voor dit werkje. Daarnaast ook Werner heel erg bedankt voor de input bij hoofdstuk 5. Je hebt ervoor gezorgd dat het maximale uit de data gehaald kon worden.

Ook vrienden en familie hebben natuurlijk een belangrijke rol gespeeld. Jullie hebben ervoor gezorgd dat er naast de altijd aanwezige druk van het proefschrift ook de mogelijkheid is geweest om er even niet aan te denken en gewoon lol te hebben. Ook al waren de ontmoetingen niet altijd met even grote regelmaat is het toch altijd erg gezellig en wil ik graag iedereen bedanken. De club uit Rooi wil ik bedanken voor alle jaren vriendschap, lol en gezelligheid. Joost en Maureen, Bart en Birgit, Ruud en Aukje, Roel en Marlou, Danny en Kim en Mathieu en Margot, na de hele rits bruiloften en geboortes nu eens tijd voor wat anders, een promotie- 6 6 • Dankwoord

153 154 6 • Dankwoord jaardagen, kroeg en vrijgezellenfeestjes. Maurice en Annemieke: wanneer wanneer en Annemieke: Maurice vrijgezellenfeestjes. en kroeg jaardagen, omgn l ebn orean o e ot rmvrn N me i het ik moet Nu promoveren. moet hoeje voorgedaan hebben al sommigen f Eneik s r er id m ek dne t de. k ik i na de naar uit kijk Ik doen. te dingen leuke om tijd weer er is Eindelijk af. onmis­ was boekje het van layouten het met hulp Ookje jaren. afgelopen ligt en volgehouden ik heb :-). hierdoor geloven in Mede keer een ook zelf allemaal toch ik waar leggen te uit goed om lastig toch het vragen. was oprechte en Soms interesse steun, is. Voorjullie gezellig altijd het waar eens weer ik en Hester dat keren de tijdens Ook steun. en zelligheid uln e e ogne uiar aod houden? avond culinaire volgende de we zullen n neik e Sin n ij bdnt or le eelged idn ver­ tijdens gezelligheid alle voor bedankt Jinji en Stijn Inge, Robert en en Frank Annelieke en Marieke, en Hans waarmaken. gaan eens maar dan waarvan bedanken, ik wil Wageningen uit studievrienden de Ook feestje. Cheers, Kus... leven. ons in fase nieuwe genen? toch of opvoeding nou het Was boekje. een toch er er je ga dan kunt het je dat hoort keer duizend je Als mij. in geloof ende niet nu hoeft dat Maar duurde. zolang het klaar! is 't meer, waarom en was bezig mee ar J gdl i ari o d pof etl ma dn s e n toch nu het is dan maar gesteld proef de op aardig is geduld Je baar. houden. te in en beetje een worden proberen ons moest zullen We gelegd. geschilderd worden er enlaminaat verhuizen gaan te hadden besloten ar u s e dn oh ct id m f e lie e ht ls e heffen. te glas het en sluiten te af om tijd echt toch dan het is nu Maar etr leed dn j vo a j led, tu e etosam de enthousiasme en steun liefde, je al voor je dank lieverd, Hester, lief­ steun, onvoorwaardelijke je basis, solide een voor je dank Mam, thuis, tweede mijn voor bedankt en Ama, Eefje, Joep Henriette, Pappa, ge­ belangstelling, jullie voor bedankt Alwin, en Imke Froukje, Hans, Dankwoord 156 6 • Curriculum vitae studenten Geneeskunde/Biomedische Wetenschappen. Geneeskunde/Biomedische studenten afgerond. 2004 in Bioprocestechnologie studie de werd crf. eued dz proe eeede ap te HLO-studenten twee Ralph begeleidde periode deze Gedurende schrift. oeeshpe e Fraooi an e Rdl Mgu Isiut te Instituut Magnus Rudolf het aan Farmacologie en rowetenschappen voor onderwijs praktisch bij hij assisteerde en hoofdvakstage hun met Dr. en Jack Wansink Dr. Rick Wieringa, Bé dr. Prof. van Nijmegen. in supervisie Onder Radboud St. UMC het van Celbiologie afdeling de op derzoek in Boyle Dr. Laboratory David van groep Health Diagnostic Molecular de bij Animal (Australië) Geelong Australian het aan stage buitenlandse een Dr. en Schaap Dr. Peter van begeleiding aan onder Universiteit Micro-organismen Wageningen de Industriële van Genetica Moleculaire devakgroep Wageningen de aan Bioprocestechnologie studie de met Sint-Oedenrode. begonnen te hij is 1978 oktober 24 op geboren is Ophuis Oude Ralph CU• R R I C U L U MV I T A E Utrecht in de groep van Prof. dr. Roger Adan. Prof. Roger dr. van groep de in Utrecht proef­ dit in staat beschreven dat uit onderzoek het hij voerde Fransen Na Amsterdam. in afde­ de bij Instituut Kanker maanden vijf Nederlands en het aan afde­ Maastricht Immunologie de UMC ling bij het van maanden vier Immunologie gedurende ling onder Jacobs Heinz Dr. afstudeervak van tweede eenbegeleiding volgde Hierna Wijk-Basten. van Danielle bij afstudeervak een hij verrichtte doctoraalfase de Tijdens Universiteit. Veghel in College Zwijsen het aan diploma VWO zijn van behalen het Na id mat 09 s i wrza as oto o d adln Neu­ afdeling de op Postdoc als werkzaam hij is 2009 maart Sinds n ar 20 sate ap as uir nezee en promotieon­ een onderzoeker junior als Ralph startte 2004 maart In • PUBLICATIONS

OUDE OPHUIS, R. J., M. Wijers, M. B. Bennink, F. A. van de Loo, J. A. Fransen, B. Wieringa, and D. G. Wansink. 2009. A tail-anchored myotonic dystrophy protein kinase isoform induces perinuclear clustering of mitochondria, autophagy, and apoptosis. PLoS ONE 4:e8024.

OUDE OPHUIS, R. J., S. A. Mulders, R. E. van Herpen, R. van de Vorsten bosch, B. Wieringa, and D. G. Wansink. 2009. DMPK protein isoforms are differentially expressed in myogenic and neural cell lineages. Muscle Nerve 40:545-55.

van Herpen, R. E., J. V. Tjeertes, S. A. Mulders, R. J. OUDE OPHUIS, B. Wieringa, and D. G. Wansink. 2006. Coiled-coil interactions modulate multimerization, mitochondrial binding and kinase activity of myotonic dystrophy protein kinase splice isoforms. FEBS J 273:1124-36.

OUDE OPHUIS, R. J., C. J. Morrissy, and D. B. Boyle. 2006. Detection and quantitative pathogenesis study of classical swine fever virus using a real time RT-PCR assay. J Virol Methods 131:78-85.

van Herpen, R. E., R. J. OUDE OPHUIS, M. Wijers, M. B. Bennink, F. A. van de Loo, J. Fransen, B. Wieringa, and D. G. Wansink. 2005. Divergent mitochondrial and endoplasmic reticulum association of DMPK splice isoforms depends on unique sequence arrangements in tail anchors. Mol Cell Biol 25:1402-14.

OUDE OPHUIS, R. J., L. Hetterschijt, D. de Gouw, M. Wijers, J. Fransen, B. Wieringa, and D. G. Wansink. Topology and specificity of membrane insertion of tail-anchored DMPK isoforms. Submitted 6 6 • Publications

157