Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
TREATMENT OF MYOTONIC DYSTROPHY
Document Type and Number:
WIPO Patent Application WO/2015/158740
Kind Code:
A1
Abstract:
The present invention relates to compositions and methods for treating myotonic dystrophy.

Inventors:
FURLING, Denis (8 bis, rue Raspail, Champigny Sur Marne, F-94500, FR)
SERGEANT, Nicolas (9 rue Jean Jaurès, Ronchin, Ronchin, F-59790, FR)
CAILLET-BOUDIN, Marie-Laure (31 rue de la Paix, Tournai, Tournai, B-7500, BE)
ARANDEL, Ludovic (20 avenue Miss Cavell, Saint Maur Des Fosses, Saint Maur Des Fosses, F-94100, FR)
Application Number:
EP2015/058111
Publication Date:
October 22, 2015
Filing Date:
April 14, 2015
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ASSOCIATION INSTITUT DE MYOLOGIE (Bâtiment Babinski, 47 boulevard de l'Hôpital, Paris, 75013, FR)
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE (3 rue Michel Ange, Paris, 75016, FR)
INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALE (101 rue de Tolbiac, Paris, 75013, FR)
UNIVERSITE PIERRE ET MARIE CURIE (PARIS 6) (4 Place Jussieu, Paris, Paris, 75005, FR)
International Classes:
A61K38/17; A61P21/00
Domestic Patent References:
WO2010044894A12010-04-22
Other References:
C. M. CHAMBERLAIN ET AL: "Mouse model of muscleblind-like 1 overexpression: skeletal muscle effects and therapeutic promise", HUMAN MOLECULAR GENETICS, vol. 21, no. 21, 1 November 2012 (2012-11-01), pages 4645 - 4654, XP055152726, ISSN: 0964-6906, DOI: 10.1093/hmg/dds306
H. TRAN ET AL: "Analysis of Exonic Regions Involved in Nuclear Localization, Splicing Activity, and Dimerization of Muscleblind-like-1 Isoforms", JOURNAL OF BIOLOGICAL CHEMISTRY, vol. 286, no. 18, 6 May 2011 (2011-05-06), pages 16435 - 16446, XP055152599, ISSN: 0021-9258, DOI: 10.1074/jbc.M110.194928
ERIN PENNOCK FOFF ET AL: "Therapeutics development in myotonic dystrophy type 1", MUSCLE & NERVE, vol. 44, no. 2, 23 August 2011 (2011-08-23), pages 160 - 169, XP055152590, ISSN: 0148-639X, DOI: 10.1002/mus.22090
Attorney, Agent or Firm:
SEKHRI, Redha et al. (Becker & Associes, 25 rue Louis le Grand, Paris, 75002, FR)
Download PDF:
Claims:
CLAIMS

1. A modified MBNL polypeptide having an YGCY binding property, and having a reduced splicing activity as compared to wild-type MBNL protein, for use as a medicament.

2. The modified MBNL polypeptide for use according to claim 1, wherein said polypeptide binds CUG repeats.

3. The modified MBNL polypeptide for use according to claim 1 or 2, which is derived from MBNL 1 , MBNL2 or MBNL3.

4. The modified MBNL polypeptide for use according to any one of claims 1 to 3, which is derived from MBNL1. 5. The modified MBNL polypeptide according to any one of claims 1 to 4, lacking the C- terminal domain of the wild-type MBNL protein.

6. The modified MBNL polypeptide for use according to any one of claims 1 to 5, derived from the MBNL1 protein and lacking the amino acids corresponding to the encoding exons 5 to 10 of the MBNL1 mRNA, the modified MBNL polypeptide having in particular the referenced sequence shown in SEQ ID NO: 2 and 3, or being a non- functional YGCY-binding variant thereof.

7. The modified MBNL polypeptide according to any one of claims 1 to 6, having a splicing activity reduced by at least 75 % as compared to the wild-type MBNL protein.

8. A nucleic acid molecule encoding the modified MBNL polypeptide as defined in any one of claims 1 to 7. 9. A genetic construct, in particular a viral vector genome, comprising the nucleic acid molecule according to claim 8 operably linked to control sequences.

10. The genetic construct according to claim 9, which is a lentivirus- or AAV-derived genome.

11. A viral vector comprising the genetic construct according to claim 9 or 10.

12. The viral vector according to claim 11, which is an AAV vector having a serotype 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or 11 AAV capsid.

13. The viral vector according to claim 12, which is an AAV vector having a serotype 9 AAV capsid. 14. A modified MBNL polypeptide as defined in any one of claims 1 to 7, the nucleic acid molecule according to claim 8, the genetic construct according to any one of claims 9 to 10, or the viral vector according to claim 11 or 12, for use in the treatment of myotonic dystrophy disease, in particular for the treatment of DM1 or DM2, or disorder causes by the abnormal sequestration of MBNL.

15. The viral vector for use according to claim 14, which is an AAV vector, in particular an AAV9 vector, which is intended to be administered intramuscularly or directly in the CNS or by any conventional route. 16. The viral vector for use according to claim 15, wherein the AAV vector is intended to be administered as a single injection.

Description:
TREATMENT OF MYOTONIC DYSTROPHY

FIELD OF THE INVENTION

The present invention relates to compositions and methods for treating myotonic dystrophy.

BACKGROUND OF THE FNVENTION Myotonic dystrophy type 1 (DM1), one of the most common neuromuscular disorders in adult, is an inherited autosomal dominant disease caused by an unstable CTG expansion located in the 3' untranslated region (UTR) of the dystrophia myotonica protein kinase (DMPK) gene (Brook et al. 1992). The number of CTG varies from fifty to more than several thousand of repeats in affected patients whereas unaffected individuals have less than 38 repeats, and globally there is a correlation between the size of the CTG repeats, the severity of the disease and inversely with the age of onset (Hunter et al. 1992; Tsilfidis et al. 1992). Clinical features of DM1 are variable but commonly include myotonia, progressive muscle weakness and atrophy as well as cardiac conduction defects but also extra-muscular symptoms such as cognitive dysfunctions, cataract, hypogonadism and endocrine deficiencies (Harper 2001).

The pathogenic CTG tract is transcribed and gives rise to RNAs containing expanded CUG repeats (CUGexp-RNAs) located in the 3' UTR of the DMPK transcripts, which are responsible for a toxic RNA gain-of- function mechanism in DM1 pathogenesis (Klein et al. 2011). CUGexp-RNAs that are retained in nuclei as discrete aggregates or foci alter the function of RNA splicing factors members of the MBNL and CELF families resulting in alternative splicing misregulation of a specific group of transcripts in affected DM1 tissues (Taneja et al. 1995; Ranum and Cooper 2006). Abnormal regulation of splicing events leads mainly to the re-expression of a fetal splicing pattern in DM1 adult tissues, and missplicing events affecting the CLC-1, INSR and BIN1 pre-mRNAs have been associated respectively with myotonia, insulin resistance and muscle weakness (Savkur et al. 2001; Charlet et al. 2002; Mankodi et al. 2002; Fugier et al. 2011). A recent study performed on a cohort of fifty DM1 patients confirmed forty-two splicing defects in affected skeletal muscles, and showed that these splicing changes were specific to DM1 when compared to other muscle disorders, and mainly attributable to MBNLl loss-of- function (Nakamori et al. 2013).

MBNLl is a member of the muscleblind-like RNA-binding protein family including MBNLl, -2 and -3 (Pascual et al. 2006), and is the major MBNL protein expressed in adult skeletal muscle (Kanadia et al. 2003; Holt et al. 2009). MBNLl like the other MBNL protein paralogues binds to expanded CUG repeats with high affinity, and colocalizes with nuclear foci of CUGexp-RNA in DM1 muscle cells (Miller et al. 2000; Fardaei et al. 2001). Sequestration of MBNLl in these ribonucleoprotein complexes due to the large number of CUG repeats in mutant RNAs leads to its loss-of- function, and consequently to alternative splicing misregulation of several target pre-mRNAs, including MBNLl itself. Consistent with this hypothesis, Mbnll knockout mice reproduces most of the deregulated splicing events observed in muscle samples of DM1 patients or DM1 mouse model expressing CUGexp- RNAs (Mankodi et al. 2000; Kanadia et al. 2003; Lin et al. 2006; Du et al. 2010). Moreover overexpression of functional and full length MBNLl (isoform 40 or 41) in the skeletal muscles of DM1 mice is sufficient to correct splicing defects and abolish myotonia, hallmarks of DM1 disease (Kanadia et al. 2006; Chamberlain and Ranum 2012). Furthermore, in WO2010/044894 it is proposed to administer a MBNL protein or a functional variant thereof, i.e. a variant that retains de biological activity of a MBNL protein, in the form of a chimeric polypeptide conjugated with a targeting moiety. This document does not disclose the use of a non- functional MBNL polypeptide for the treatment of DM1. In addition, disruption of MBNL2, which is prominently expressed in the brain, deregulates specific splicing events in mice that are similarly misregulated in human DM1 brains supporting a prominent role of MBNL2 loss-of-function in the pathological changes in the human disease (Charizanis et al. 2012). Taken together, these results support MBNL loss-of- function as a key mechanism involved in RNA toxicity induced by expanded CUG repeats in DM1.

Modified oligonucleotide antisens approaches that interfere with CUGexp-RNAs to release MBNLl from the foci have already been proposed for reversing splicing misregulations and myotonia in a DM1 mouse model. However, alternate and efficient means for reversing splicing misregulations and counteracting clinical symptoms such as myotonia in myotonic dystrophy are still needed.

SUMMARY OF THE INVENTION The purpose of the present invention is to provide new tools and methods for myotonic dystrophy treatment. The present invention is based on the evidence provided herein that a modified MBNL polypeptide ectopically expressed, in particular through the use of viral vectors, is effective to counteract CUGexp-RNA toxicity both in vitro and in vivo.

An aspect of this invention relates to a modified MBNL polypeptide. As provided below, the modified MBNL polypeptide of the invention has a reduced splicing activity, in particular a splicing activity reduced by at least 50%, in particular by at least 60%, 70%>, 75%, 80%, 85% or by at least 90% or even by at least 95%, when compared to splicing activity of full-length MBNL protein but maintains its YGCY binding property. In particular, the modified MBNL polypeptide of the invention is able to bind pathological CUG repeat. Moreover, the modified MBNL polypeptide used herein can counteract the CUGexp-RNA toxicity by releasing sequestered endogenous MBNL such as MBNLl and MBNL2 from the CUGexp-RNA aggregates in order to restore the function of these endogenous MBNL proteins.

The modified MBNL polypeptide of the present invention is used in the treatment of myotonic dystrophies. Another aspect relates to a method for treating myotonic dystrophies, comprising administering to a subject in need thereof an effective amount of a modified MBNL polypeptide according to the invention. A further aspect of the invention is the use of a modified MBNL polypeptide as described herein, for the manufacture of a medicament for use in the treatment of myotonic dystrophies.

Another aspect disclosed herein is the use of the modified MBNL polypeptide according to the invention for displacing an endogenous MBNL protein, such as endogenous MBNLl or MBNL2, from CUG repeats in a cell or organism in need thereof, and thereby reversing deregulated splicing events induced by the CUGexp-RNA expression. In a particular embodiment, the modified MBNL polypeptide is provided to the cell or organism using a viral vector. LEGEND OF THE FIGURES

Fig 1: Genomic DNA organization of human MBNLl gene. Order and names of exons are shown as well as the length in nucleotides of each exon. Clear gray boxes represent UTR. Color boxes represent cassette exons. Empty boxes represent constitutive exons. Alternative splicing of MBNL1 alternative cassettes leads to more than ten iso forms including MBNL143 or MBNL140. The length in amino acids (aa) is indicated and the Dark gray boxes represent C 3 H 1 zinc finger motifs. Two are located in MBNL1 exon 2 and the two others are located in MBNL1 exon 4. The modified MBNL1 polypeptide (herein referred as ACT3) is the truncated MBNL1 construct lacking the C-terminal domain following the fourth C 3 H 1 zinc finger motif.

Fig 2: ACT3 colocalizes with nuclear CUGexp-RNA aggregates in vitro. GFP, GFP-ACT3 or GFP-MBNL140 constructs were co-transfected with 960CTG repeats in Hela cells. CUGexp- RNA foci were visualized by FISH using a Cy3-CAG7 probe.

Fig 3: Expression of various MBNL1 iso forms as well as ACT3 construct restores DM1 deregulated splicing of Tau exon 2/3 minigene. The Tau exon 2/3 minigene comprises the two alternate cassettes 2 and 3 insert in the psvIRB splicing reporter minigene (Tran et al. 2011). MBNL1 35 or 38 or 41 or 43 iso forms (panel A) or GFP-ACT3 panel B) were co-expressed with Tau exon 2/3 minigene and a plasmid containing 960 interrupted CTG repeats in Hela cells, as previously described (Tran et al. 2011). The percentage of inclusion of Tau E2 was calculated and established following RT-PCR using primers surrounding Tau exon 2 and exon 3. Fig 4: Nuclear localization of ACT3 is required to modulate splicing events. A) GFP-ACT3 constructs containing or not a nuclear export signal (NES) were co-expressed or not with 960 CTG repeats in Hela cells. B) Inclusion of cTNT exon 5 or IR exon 11 was assessed by RT- PCR after co-tranfection of MBNL or GFP constructs and cTNT exon 5 or IR exon 11 minigenes in Hela cells. C). Inclusion of Tau exon 2 was analyzed in Hela cells co-transfected with Tau exon 2/3 minigene, 960 CTG repeats and MBNL or GFP constructs.

Fig 5: ACT3 is able to displace MBNL1 from CUG repeats in vitro. Recombinant MBNL1 40 (or ACT3) protein was cross-linked to in vitro transcribed 32 P RNA containing 95 CUG repeats in the absence or presence of incremental concentrations of recombinant ACT3 (MBNL 140) protein.

Fig 6: ACT3 colocalizes with nuclear CUGexp-RNA in human DM1 muscle cells. Primary DM1 muscle cells were transduced with lentiviral vectors containing the cDNA encoding the GFP-ACT3. CUGexp-RNA foci were visualized by FISH using a Cy3-CAG 7 probe. Fig 7 A: ACT3 normalizes missplicing events in differentiated human DM1 muscle cells. Primary human DM1 and non-DMl muscle cells were transduced or not with lentiviral vectors expressing GFP-ACT3 or GFP alone. Splicing profile of BIN1 exon 11, LDB3 exon 7 and DMD exon 78 transcripts were analyzed by RT-PCR.

Fig. 7B: Differentiated muscle cells (Control, DM1 or DM1 expressing C) were transfected or not with MBNLl siRNA directed again the C-terminus exon 9 (present in MBNLl but not in ACT3 sequences). Splicing profile of DMD exon 78 transcripts was analyzed by RT-PCR.

Fig 8: Intramuscular injection of adeno-associated virus of serotype 9 (AAV9) containing the GFP-ACT3 encoding cDNA normalizes splicing misregulations in DM1 mice. Gastrocnemius muscles of HSA-LR mice were injected with AAV9 GFP-ACT3 (1.10 11 vg; n= 6) and analyzed after 6 weeks. Contralateral muscles were injected with saline. Splicing profile of Sercal exon 22, Mbnll exon 7 and Clcnl exon 7a were analyzed by RT-PCR.

Fig 9: GFP-ACT3 colocalizes with nuclear CUGexp-RNA foci in vivo. FISH-IF were performed to detect CUGexp-RNA foci and GFP-ACT3 on muscle sections of HSA-LR mice injected with AAV9 GFP-ACT3.

Fig 10 : GFP-ACT3 displaces Mbnll from nuclear CUGexp-RNA foci in vivo. FISH-IF were performed to detect CUGexp-RNA foci, endogenous Mbnll and GFP-ACT3 on muscle sections of HSA-LR mice injected with AAV9 GFP-ACT3 or saline. The peak of intensity for each component was measured along an arbitrary lane crossing foci observed within the nucleus.

Fig 11 : Intramuscular injection of AAV9 GFP-ACT3 abolishes myotonia DM1 mice. Force relaxation of HSA-LR Gastrocnemius muscles injected with AAV9 GFP-ACT3 (1.10 11 vg; n= 6) or saline (contralateral muscles) was measured 6 weeks post-injection. Force relaxation was also determined in Gastrocnemius muscles of FVB wt mice.

Fig 12: No signs of muscle degeneration in FVB wt mice expressing AAV9 GFP-ACT3. Tibialis anterior muscles of FVB wt mice were injected with AAV9 GFP-ACT3 (1.10 11 vg; n=6) and analyzed by IF after 3, 4 or 6 weeks. Contralateral muscles were injected with empty AAV9 MCS. Embryonic MyHC and laminin antibodies were used to detect regenerating fibers and muscle fibers respectively. Nuclei were stained with Dapi. Fig 13: Expression of AAV9 GFP-ACT3 alone did not deregulate alternative splicing in wt mice. Tibialis anterior muscles of FVB wt mice were injected with AAV9 GFP-ACT3 (1.10 11 vg, n=6) and splicing profiles of Clcnl exon 7a or Sercal exon 11 were analyzed after 3, 4 or 6 weeks post transduction. Contralateral muscles were injected with empty AAV9 MCS. Fig 14A: Intramuscular injection of AAV9 GFP-ACT3 normalises splicing misregulations in DM1 mice. Tibialis anterior muscles of HSA-LR mice were injected with AAV9 GFP-ACT3 (1.10 11 vg; n= 6) and analyzed after 6 weeks. Contralateral muscles were injected with AAV9 MCS. Splicing profiles of Clcnl exon 7a, Sercal exon 11 and LDB3 exon 11 were determined by RT-PCR.

Fig. 14B: Intramuscular injection of AAV9-V5-ACT constructs. NLS-V5-ACT3 and V5- ACT3 constructs normalize splicing misregulations in DM1 mice in contrast to V5-ACT(-3) lacking exon 3. Tibialis anterior muscles of HSA-LR mice were injected with AAV9-V5- ACT constructs (5.10 10 vg; n= 3) and analyzed after 6 weeks. Contralateral muscles were injected with PBS. Splicing profiles of Clcnl exon 7a and Sercal exon 11 were determined by RT-PCR.

Fig 15: ACT3 restores MBNL2-splicing dependent events. MBNL constructs (MBNL1, panel A ; MBNL2, panel B ; ACT3, panel C) were co-expressed with hTau exon2 minigene and 960 CTG repeats in T98G cells, as described in (Carpentier et al, 2014). Inclusion of Tau E2 was analyzed by RT-PCR. The graph indicates the percentage of Tau exon 2 exclusion (averaged ± S.E.M. for at least three independent experiments). Significant differences are indicated by asterisks: *, p < 0.05; **, p < 0.01, p < 0.001. 18S transcripts were used as internal controls to verify the amounts of RNA. The efficiency of DT960 transfection was verified by RT-PCR of the 3'UTR of the human DMPK gene. Panel D shows the mutant MBNL1 sites that have been mutated (bold grey) in the Mut MBNL construct. DETAILED DESCRIPTION OF THE INVENTION

The modified MBNL polypeptide of the invention is able to bind the MBNL YGCY RNA- motif, with "Y" representing a pyrimidine (uridine or cytosine). In particular, the modified MBNL polypeptide of the invention is able to bind UGCU-motif, which is the building block of the pathological DM1 expanded CUG repeats. In a particular embodiment, the modified MBNL polypeptide of the invention includes or not the amino acids corresponding to exon 3 of the MBNLl mRNA (accession number NM 021038). In a further embodiment, the modified MBNL polypeptide of the invention lacks the amino acids of SEQ ID NO: l (SEQ ID NO:l : TQSAVKSLKRPLEATFDLGIPQAVLPPLPKRPALEKTNGATAVFNTGIFQYQQALAN MQLQQHTAFLPPGSILCMTPATSVVPMVHGATPATVSAATTSATSVPFAATTANQIPII S AEHLT SHKY VTQM) corresponding to exons 5 to 10 of the MBNLl mRNA. As used herein, the term "MBNL" denotes all paralogue members of the muscleblind-like RNA-binding protein family and includes in particular MBNLl, -2 and -3. In a particular embodiment, the modified MBNL polypeptide according to the invention is derived from the human MBNLl protein sequence. In an embodiment, the modified MBNL polypeptide is a MBNLl protein having the exon 3 encoded sequence but lacking the amino acid sequences encoded by exons 5 to 10 of the MBNLl gene. In a specific embodiment, the modified MBNLl-derived polypeptide is referred to as ACT3 having the following amino acid sequence:

MAVSVTPIRDTKWLTLEVCREFQRGTCSRPDTECKFAHPSKSCQVENGRVIACFDSL KGRCSRENCKYLHPPPHLKTQLEINGRNNLIQQKNMAMLAQQMQLANAMMPGAPL QPVPMFSVAPSLATNASAAAFNPYLGPVSPSLVPAEILPTAPMLVTGNPGVPVPAAAA AAAQKLMRTDRLEVCREYQRGNCNRGENDCRFAHPADSTMIDTNDNTVTVCMDYI KGRCSREKCKYFHPPAHLQAKIKAAQYQVNQAAAAQAAATAAAM (SEQ ID NO: 2).

In an embodiment, the modified MBNL polypeptide is a MBNLl protein lacking both the exon 3 encoded sequence and the sequences encoded by exons 5 to 10 of the MBNLl gene. In a specific embodiment, the non- functional MBNLl-derived polypeptide is referred to as ACT having the following amino acid sequence:

MAVSVTPIRDTKWLTLEVCREFQRGTCSRPDTECKFAHPSKSCQVENGRVIACFDSL KGRCSRENCKYLHPPPHLKTQLEINGRNNLIQQKNMAMLAQQMQLANAMMPGAPL QPVVCREYQRGNCNRGENDCRFAHPADSTMIDTNDNTVTVCMDYIKGRCSREKCKY FHPPAHLQAKIKAAQYQVNQAAAAQAAATAAAM (SEQ ID NO: 3).

In another embodiment, the modified MBNL polypeptide is a MBNL2 protein encoded by the amino acid sequences of exons 2 to 5 of the MBNL2 protein. In a specific embodiment, the modified MBNL2-derived polypeptide is referred to as MBNL2-ACT3 having the following amino acid sequence:

MALNVAPVRDTKWLTLEVCRQFQRGTCSRSDEECKFAHPPKSCQVENGRVIACFDSL KGRCSRENCKYLHPPTHLKTQLEINGRNNLIQQKTAAAMLAQQMQFMFPGTPLHPV PTFPVGPAIGTNTAISFAPYLAPVTPGVGLVPTEILPTTPVIVPGSPPVTVPGSTATQKL L RTDKLEVCREFQRGNCARGETDCRFAHPADSTMIDTSDNTVTVCMDYIKGRCMREK CKYFHPPAHLQAKIKAAQHQANQAAVAAQAAAAAATVM (SEQ ID NO:4).

As used herein a "variant" of the modified MBNL polypeptide of the invention is a protein having the same or similar binding properties to the YGCY motif, in particular to CUG repeats, as the wild-type MBNL protein it is derived from (in particular MBNL1, 2 or 3) or as the modified MBNL protein of SEQ ID NO: 2, SEQ ID NO:3 or SEQ ID NO:4 as shown above, and wherein said variant has a reduced splicing activity as compared to the wild-type MBNL protein. In other terms, the modified MBNL polypeptide used in the present invention is a non- functional MBNL polypeptide, that has low or even no splicing activity as compared to the wild-type parent MBNL protein. The use of such a non- functional MBNL polypeptide with regard to its splicing activity has never been reported in the prior art for the treatment of a myotonic dystrophy, since all previous therapeutic attempts were carried out using a functional MBNL protein, i.e. a protein that has all the features of the wild-type protein including its ability to bind to the YGCY motif and its splicing activity. Indeed, previous attempts were aimed at providing to the treated cell overexpression of a MBNL protein that will compensate for the loss of free and functionally available endogenous MBNL proteins which has been sequestered onto pathological repeats. The strategy applied in the present disclosure is unrelated. Instead of providing overexpression of a functional MBNL protein, the inventors propose to introduce into cells in need thereof a modified MBNL protein, i.e. a non- functional variant MBNL protein having reduced or even no splicing activity, to compete, displace, and replace thereon endogenous MBNL protein(s) to avoid the negative consequences of their sequestration. As provided in the experimental part of this disclosure, the results obtained with the present strategy have been extremely satisfying. In a particular embodiment, the variant according to the invention may have a sequence at least 50 %, in particular at least 60%, 70 %, 80 %, 90 % and more particularly at least 95 % or even at least 99 % identical to the amino acid sequence corresponding to exons 1 to 4 of the wild-type MBNL protein (e.g. of MBNL1, 2 or 3) or to the amino acid sequence shown in SEQ ID NO: 2, 3 or 4.

In a particular embodiment of the invention, the modified MBNL polypeptide of the invention is not a chimeric peptide consisting of a MBNL polypeptide and a targeting moiety.

The present invention implements and modified MBNL polypeptide which has almost no splicing activity, or otherwise said as a reduced activity, as compared to wild-type MBNL protein. By "almost no activity" or "reduced activity", it is herein intended to describe a splicing activity, which is reduced by at least 50%, in particular by at least 60%>, 70%>, 75%>, 80%), 85%o, 90%) or even at least 95%> as compared to the splicing activity of wild-type MBNL protein. Such activity may be determined according to methods well known by those skilled in the art such as the use of minigenes to analyze the alternative splicing of cTNT exon 5, IR exon 11 and Tau exon 2 (Tran et al., 2011). In a particular embodiment, the modified MBNL protein of the invention may comprise a localization sequence such as a nuclear localization sequence (NLS) or a nuclear export signal (NES). A representative NLS has the sequence represented in SEQ ID NO: 5: PK KR V. A representative NES has the sequence represented in SEQ ID NO: 6: LPPLERLTLD. The present disclosure includes any modified MBNL polypeptide as described above, combined with such a localization sequence, in particular with a NLS or NES such as those specifically mentioned above.

The invention further relates to a pharmaceutical composition comprising the modified MBNL polypeptide of the invention or a variant thereof.

Another aspect of the invention is a nucleic acid sequence comprising or consisting of a nucleotide sequence encoding a modified MBNL polypeptide according to the invention. The invention further relates to a genetic construct consisting of or comprising a nucleotide sequence as defined herein, and regulatory sequences (such as a suitable promoter(s), enhancer(s), terminator(s), etc ..) allowing the expression (e.g. transcription and translation) of a modified MBNL polypeptide according to the invention in a host cell. The genetic constructs of the invention may be DNA or RNA, and are preferably double-stranded DNA. The genetic constructs of the invention may also be in a form suitable for transformation of the intended host cell or host organism, in a form suitable for integration into the genomic DNA of the intended host cell or in a form suitable for independent replication, maintenance and/or inheritance in the intended host organism. For instance, the genetic constructs of the invention may be in the form of a vector, such as for example a plasmid, cosmid, YAC, a viral encoding vector or transposon. In particular, the vector may be an expression vector, i.e. a vector that can provide for expression in vitro and/or in vivo (e.g. in a suitable host cell, host organism and/or expression system). In a preferred but non-limiting aspect, a genetic construct of the invention comprises i) at least one nucleic acid sequence of the invention; operably linked to ii) one or more regulatory elements, such as a promoter and optionally a suitable terminator; and optionally also iii) one or more further elements of genetic constructs such as 3'- or 5'-UTR sequences, leader sequences, selection markers, expression markers/reporter genes, and/or elements that may facilitate or increase (the efficiency of) transformation or integration or subcellular localization or expression of the modified MBNL polypeptide, such as nuclear localization signal (NLS) or nuclear export signal (NES). In a particular embodiment, the genetic construct corresponds to the genome of a recombinant viral vector. Suitable viral vectors used in practicing the present invention include retroviruses, lentiviruses, adenoviruses and adeno-associated viruses. In particular, the invention relates to a lentivirus comprising a nucleic acid sequence encoding a modified MBNL polypeptide according to the invention. In another particular embodiment, the invention relates to an AAV vector, in particular an AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10 or AAV11 vector, in particular an AAV9 vector, comprising a nucleic acid sequence encoding a modified MBNL protein according to the invention. The AW vector may be a pseudotyped vector, i.e. its genome and its capsid may be derived from different AAV serotypes. For example, the genome may be derived from an, AAV2 genome and its capsid proteins may be of the AAV1, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10 or AAV11 serotype.

Another aspect relates to a modified MBNL polypeptide according to the invention, for use as a medicament. The modified MBNL polypeptide of the invention is an useful therapeutic agent, in particular in the treatment of a disease or disorder linked to a sequestration of a MBNL protein, or to a deregulated function of a MBNL member such as MBNL1, or other paralogue members (including MBNL2 and MBNL3). In a preferred embodiment, the modified polypeptide of the invention is used for the treatment of a myotonic dystrophy such as DM1 and DM2, or any disease where a loss of MBNL function (e.g. sequestration, aggregation, mutations...) may be rescued by ectopic delivery of the modified MBNL polypeptide of the invention. In a further aspect, the invention relates to a modified MBNL polypeptide as described above, for use in a method for the treatment of a myotonic dystrophy.

As used herein, the term "treatment" or "therapy" includes curative and/or preventive treatment. More particularly, curative treatment refers to any of the alleviation, amelioration and/or elimination, reduction and/or stabilization (e.g., failure to progress to more advanced stages) of a symptom, as well as delay in progression of a symptom of a particular disorder. Preventive treatment refers to any of: halting the onset, delaying the onset, reducing the development, reducing the risk of development, reducing the incidence, reducing the severity, as well as increasing the time to onset of symptoms and survival for a given disorder.

It is thus described a method for treating myotonic dystrophies in a subject in need thereof, which method comprises administering said patient with a modified MBNL polypeptide according to the invention, or with a nucleic acid sequence encoding said modified MBNL polypeptide.

Within the context of the invention, "subject" or "patient" means a mammal, particularly a human, whatever its age or sex, suffering of a myotonic dystrophy. The term specifically includes domestic and common laboratory mammals, such as non-human primates, felines, canines, equines, porcines, bovines, goats, sheep, rabbits, rats and mice. Preferably the patient to treat is a human being, including a child or an adolescent.

For the uses and methods according to the invention, the modified MBNL polypeptide, the nucleic acid, the genetic construct, or the viral vector (such as a lentiviral or AAV vector) may be formulated by methods known in the art. In addition, any route of administration may be envisioned. For example, the modified MBNL polypeptide, the nucleic acid, the genetic construct and the viral vector (such as a lentiviral or AAV vector) may be administered by any conventional route of administration including, but not limited to oral, pulmonary, intraperitoneal (ip), intravenous (iv), intramuscular (im), subcutaneous (sc), transdermal, buccal, nasal, sublingual, ocular, rectal and vaginal. In addition, administration directly to the nervous system may include, and are not limited to, intracerebral, intraventricular, intracerebroventricular, intrathecal, intracistemal, intraspinal or peri-spinal routes of administration by delivery via intracranial or intravertebral needles or catheters with or without pump devices. It will be readily apparent to those skilled in the art that any dose or frequency of administration that provides the therapeutic effect described herein is suitable for use in the present invention. In a particular embodiment, the subject is administered a viral vector encoding a modified MBNL polypeptide according to the invention by the intramuscular route. In a specific variant of this embodiment, the vector is an AAV vector as defined above, in particular an AAV9 vector. In a further specific aspect, the subject receives a single injection of the vector.

Additionally, standard pharmaceutical methods can be employed to control the duration of action. These are well known in the art and include control release preparations and can include appropriate macromolecules, for example polymers, polyesters, polyamino acids, polyvinyl, pyrolidone, ethylenevinylacetate, methyl cellulose, carboxymethyl cellulose or protamine sulfate.

In addition, the pharmaceutical composition may comprise nanoparticles that contain the modified MBNL polypeptide of the present invention.

The below examples illustrate the invention without limiting its scope. EXAMPLES MATERIAL AND METHOD

Plasmids and viral production

The plasmid containing the 3'UTR of DMPK with 960 interrupted CTGs was a minigene vector also under the control of the CMV promoter (kindly gift from T. Cooper, Baylor College of Medecine, Houston TX, USA). The sequence of MBNL1 full-length variant constructs and truncated ACT3 used in this study were previously described (Tran et al. 2011). NES was derived from the REV protein of HIV (Fischer et al. 1995) and fused to the ACT3 construct. MBNL1, ACT3 and ACT3-NES splicing activity was assessed using three minigenes previously described: the RTB300 minigene containing exon 5 of human cTNT (hcTNT); the INSR minigene containing exon 11 of human insulin receptor and the pSVIRB/Tau minigene containing exons 2 and 3. All plasmids DNA were double-strand sequenced at GATC Biotech (France) and purified using the Nucleobond® AX endotoxin free kit (Macherey Nagel, Germany). The cDNA coding for GFP-ACT3 or for the GFP protein containing both a Kozak consensus sequence were cloned in the SIN-cPPT-PGK-WHV or pSMD2 transfer vectors. Lentiviral and AAV9 vectors were obtained as previously described (Caillierez et al. 2013; Francois et al. ; Fugier et al. 2011) and stored frozen at -80°C. Recombinant GST-MBNL1 and ACT3 proteins were produced and UV-cross-linking experiments performed as describe before (Laurent et al. 2012, Tran et al. 2011). The hTau minigene and Mut MBNL construct are described in (Carpentier et al., 2014). Briefly, the hTau E2 minigene consists of exon 1, exon 2 and exon 4 sequences of the human MAPT gene inserted into the pEGFPNl eucaryote expression vector (Clontech). The exon 2 is preceded and followed by 878 and 2100 nucleotides of the intronic sequences 2 and 3 of the human MAPT gene, respectively (detailed in Carpentier et al, 2014). The Seq250 E2 250 MBNL1 mutated sites in figure 15D represents the 250 nucleotides of the intronic sequences surrounding exon 2 for which the MBNL binding sites are mutated (sequences in bold grey). This mutant minigene is no more responsive to MBNL splicing regulatory activity. These minigenes were generated by GeneArt® (Gene Synthesis company) and the sequence of plasmid preparation was verified by double strand sequence by GATC (Biotech, Constance, Germany).

Cell culture, transfection and infection

HeLa cells were grown in monolayer cultures in 6 well plates in Dulbecco's Modified Essential Medium (DMEM) (Invitrogen) supplemented with 10% fetal calf serum (FBS) at 37°C in a humidified C0 2 (5%) incubator. Cells grown to ~ 70% confluence were transiently co-transfected with ^g of minigene plasmid DNA, ^g of CUG repeats and 3 μg of MBNL plasmid DNA in triplicate, using FuGENE HD transfection reagent (Roche Diagnostics) according to the manufacturer's instructions. Human muscle cells were isolated from skeletal muscle biopsies as described (Furling et al, 2001), in accordance to French legislation on ethical rules. Wild-type (WT) and DM1 myoblasts were grown in HAM's F10 medium supplemented with 20 % FBS and 5 μg/mL gentamycin (Invitrogen), at 5% C02 and 37°C. lOOng Ρ24/μ1 were used to transduce 2xl0 5 human muscle cells. Vector transduction was performed overnight in the presence of 4 μg/ml of polybrene (Sigma). To trigger differentiation, growth medium was removed from subconfluent cultures and replaced by DMEM medium supplemented with 10 μg/mL insulin (Sigma).

In vivo gene transfer and experiments

HSA-LR mice were obtained from C. Thornton and control FVB mice from Janvier. All mouse procedures were done according to experimental protocols approved by the Ethic Committee on Animal Resources at the Centre d'Exploration Fonctionnelle of Pitie- Salpetriere animal facility and under appropriate biological containment. The gastronemius or tibialis anterior muscles of adult mice were injected respectively with 30 to 100 μΐ of physiological solution containing or not AAV9 vectors. For each mouse, one muscle was injected with AAV GFP-ACT3 and the contralateral muscle was injected with control AAV containing any transgene (MCS) or GFP or vehicle alone (PBS). Six weeks after injections, the isometric contractile properties of the muscles were measured as previously described (Mouisel et al. 2006). Then, the mice were killed, muscles were collected and snap-frozen in liquid nitrogen-cooled isopentane and stored at -80 °C.

Fluorescent in situ hybridization (FISH) and immunofluorescence

Fluorescent in situ hybridization (FISH) was done as described using a Cy3-labeled 2-0- methyl RNA (CAG)7 probe. Combined FISH-immuno fluorescence (IF) experiment was done as described (Francois et al.) using polyclonal MBNL1 (Everest Biotech.) or GFP (Invitrogen) antibodies followed respectively by secondary Cy5- or Alexa 488-conjugated antibodies. Pictures were captured using a Leica confocal microscope and software (Leica microsytems), and processed with ImageJ software. Immunofluorescence on muscle section was done as described using embryonic MyHC (Novocastra) and Laminin (Novocastra) antibodies.

Protein extraction and western blot analysis

Western blotting was performed with standard methods as described previously (Tran et al. 2011) using anti-GFP (Santa Cruz) or anti- GAPDH (Tebu-Bio) antibodies. RNA extraction and semi-quantitative analysis

Total RNA was isolated using a total RNA extraction kit (Nucleospin® RNA II kit, Macherey Nagel) or TRIzol reagent (Invitrogen) according to the manufacturer's protocol. RNA concentration was determined by measuring the absorbance at 260 nm by using the Nanodrop (Labtech). RT-PCR was performed using 1 μg of total RNA using random hexamers and the M-MLV reverse transcriptase (Invitrogen) according to standard protocols. No DNA amplification was observed in the RT controls. PCR was carried out as previously described. The reaction products were resolved by electrophoresis using a 5% or 8% polyacrylamide gel, and bands were stained with SYBR Gold (Invitrogen). The intensity of SYBR Gold luminescence was measured using a Fluorolmager scanner (Claravision). PCR experiments were repeated at least three times.

Statistical analyses

Statistical analyses were performed using unpaired t-test with two tails P values, with the help of Prism 6 Software (GraphPad Software Inc.).

RESULTS

In a previous study focused on binding-affinity and splicing activity of the different MBNLl iso forms, we showed that truncated MBNLl construct lacking the C-terminal domain (ACT3, Fig. 1) keeps YGCY binding property with a slightly lower affinity when compared to full length MBNLl isoforms but has a dramatic reduction of its splicing activity due to the absence of the sequence encoded by exons 5 to 10 (Tran et al. 2011). To evaluate if the nonfunctional polypeptide ACT3 polypeptide is still able to bind to pathogenic CUG repeats, Hela cells were co-transfected with expanded CUG repeats and MBNLl- or GFP -tagged ACT3 constructs. As observed in figure 2, GFP-ACT3 colocalizes with the nuclear foci of CUGexp-RNA as observed for the full length MBNLl . We next examined its effect on DM1 deregulated splicing events by co-expressing GFP-ACT3 or MBNLl constructs with expanded CUG repeats and analyze the Tau exons 2/3 splicing using a splicing reporter minigene in Hela cells (Fig 3). In the presence of CUG repeats, inclusion of Tau exon 2 is significantly reduced as observed in DM1 patients, and over-expression of various MBNLl isoforms that have similar splicing activities (Tran et al. 2011) restores partially Tau exon 2 inclusion (Fig 3A). However, over-expression of the GFP-ACT3 construct that has more than 80% reduction of its splicing activity compared to MBNLl also corrects splicing changes of Tau exons 2/3 minigene to similar extend than MBNLl (Fig 3B). This result suggests that the GFP-ACT3 may interact with CUG repeats to release functional MBNLl involved in the regulation Tau exon 2 splicing since it is unlike that the residual splicing activity of GFP- ACT3 is sufficient to restore a normal splicing of Tau in the presence of CUG repeats. To confirm this hypothesis, we generated a GFP-ACT3 construct fused with a strong nuclear export signal (NES) derived from the REV protein of HIV (Fischer et al. 1995). As expected, the GFP-ACT3-NES has a complete cytoplasmic localization when compared to GFP-ACT3, which displayed a nucleo-cytoplasmic localization (Fig 4A). Due to efficient nuclear export and exclusive cytoplasmic localization, GFP-ACT3-NES has no more residual splicing activity as shown using hcTNT exon 5 and IR exon 11 minigenes (Fig 4B). By contrast, co- expression of GFP-ACT3-NES with CUG repeats and Tau exons 2/3 splicing reporter minigenes is still able to restore a normal splicing of Tau as observed previously with the full length MBNLl and GFP-ACT3 (Figure 4C). As shown in figure 4A, GFP-ACT3-NES colocalizes with the CUGexp-RNA foci but the free remaining unbound GFP-ACT3-NES is efficiently exported out of the nucleus, and therefore is no more available for alternative splicing regulation. All together, our results indicate that ACT3 reverses the deregulated splicing events in the presence of pathogenic CUG repeats by saturating the CUG binding sites and releasing sufficient quantity of functional MBNLl from the CUGexp-RNA foci. To confirm that ACT3 can bind to CUG repeats and compete with MBNLl binding, recombinant MBNLl protein was cross-linked to in vitro transcribed 32 P RNA containing 95 CUG repeats in the absence or presence of growing concentrations of recombinant ACT3 protein, or vise-et-versa (Fig. 5). In both conditions, incremental concentrations of recombinant ACT3 or MBNLl were able to reduced respectively, the amount of recombinant MBNLl or ACT3 indicating that ACT3 is able to compete and displace efficiently MBNLl from CUG repeats.

To assess if ACT3 can interact with CUGexp-RNA in DM1 cells and modulate DM1 molecular features as alternative splicing mis regulation, human primary muscle cell cultures of DM1 and non-DMl patients were transduced or not with lentiviral vectors expressing either GFP-ACT3 or GFP. As shown in figure 6, GFP-ACT3 colocalizes with the nuclear CUGexp-RNA foci in DM1 muscle cells. We next investigated the effects on splicing mis regulation of the DMD, BIN1 and LDB3 transcripts, which are abnormally spliced in differentiated DM1 muscle cells (Fig. 7A) (Francois et al. 2011). Expression of GFP-ACT3 significantly normalized the splicing profiles of these transcripts in DM1 cells, whereas it did not affect their splicing in control non-DMl cells. These results confirm that expression of GFP-ACT3 is able to reverse molecular changes induced by toxic CUGexp-R As in DM1 muscle cells. Moreover, it also shows that modified GFP-ACT3 alone does not modify the splicing of endogenous targets. To further decipher the mechanism of action of the modified GFP-ACT3, we silenced MBNL1 using siRNA in GFP-ACT3 treated muscle cells. As shown in figure 7B, GFP-ACT3 required MBNL1 activity to fully restore DMD splicing profile in DM1 muscle cells. It is of note that loss of MBNL1 in control muscle cells alters DMD splicing profile. This result indicates that in DM1 cells, GFP-ACT3 has not a direct activity on splicing. Thus, ACT3 requires the release of functional MBNL1 from the expanded CUG- RNAs to correct splicing changes in DM1 cells.

The capacity of ACT3 to neutralize in vivo the RNA toxicity induced by the expanded CUG repeats was next tested in the DM1 mouse model (HSA-LR) expressing 220 CTG in the 3'UTR of the human skeletal actin gene (Mankodi et al. 2000). These mice accumulate CUGexp-RNA in the nuclei of their skeletal muscle fibers and display missplicing events as well as myotonia. Gastronemius (GAS) muscle of HSA-LR mice was injected intramuscularly with AAV9-GFP-ACT3 vectors whereas contralateral GAS was injected with saline solution. After 6 weeks, the contraction properties of these muscles were measured in situ, mice were thereafter sacrificed and the muscles were taken for histological and biochemical analysis. Among the splicing changes in the HSA-LR mice that are similar to those observed in DM1 patients, we examined the splicing mis regulation of Sercal, Mbnll and Clc-1. As showed in figure 8, injection of AAV9-GFP-ACT3 corrected the splicing pattern of these transcripts when compared to HSA-LR contralateral muscles and restores an almost complete normal splicing profile when compared to FVB wt mice. Noticeably, AAV9- GFP-ACT3 had no impact on the endogenous splicing of these transcripts in FVB wt mice thus confirming that the ACT3 construct had almost no splicing regulatory activity in vivo and (Fig. 13) or in vitro concentration equivalent to the WT MBNL1 protein (Tran et al, 2011). Together, our results suggest that ACT3 competes with endogenous MBNL for abnormal binding to expanded CUG repeats leading to the release of sequestered and functional MBNL from the CUGexp-RNA aggregates. To support our data indicating that ACT3 releases enough functional Mbnll from CUGexp-RNA foci to restore normal alternative splicing profiles in HSA-LR mice, we monitored their nuclear localization on muscle sections (Fig 9). As expected, ACT3 colocalizes with the CUGexp-RNA foci in the myonuclei of AAV9-GFP- ACT3 injected HSA-LR mice. In contrast, Mbnll :CUGexp-R A foci co localization (as indicated by the peak of intensity) that overlaps in control HSA-LR mice was largely reduced in AAV9 GFP-ACT3 injected mice (Fig 10), confirming that ACT3 replaces MBNL1 into the foci and displaces enough endogenous MBNL1 to restore functional MBNL-dependant splicing activity in DM1 mice.

At the physiological level, it has been established that myotonia observed in this DM1 mouse model results from abnormal splicing of muscle-specific chloride channel Clc-1 exon 7a (Wheeler et al. 2007). Myotonia that is characterized by muscle hyperexcitability that leads to persistent electrical discharges and delayed force relaxation. Since Clc-1 exon 7a missplicing was almost completely normalized by ACT3 expression, its effect on muscle force relaxation was determined after induced-contraction (Fig 11). Significant increased of force relaxation was measured in HSA-LR muscles when compared to wt FVB mice confirming the myotonia previously established by electromyography in these DM1 mice. Myotonia reveals by abnormal force relaxation was abolished in the GAS muscle of HSA-LR mice injected with AAV9-GFP-ACT3 when compared to contralateral muscles whereas no significant changes in muscle strength and muscle histology was detected (data not shown). In addition, Tibialis Anterior (TA) muscles of wt mice were also injected with AAV9-GFP-ACT3 or an empty AAV9-MCS and sacrificed after 3, 4 and 6 weeks (Fig 12). These muscles showed no signs of toxicity or muscle regeneration/degeneration as indicated by the almost absence (less than 1%) of central nuclei, embryonic myosin heavy chain re-expression as well as abnormal size of the muscle fibers. In addition, the splicing profile of genes that are abnormally spliced in DM1 is not perturbed in wt mice by either ACT3 expression or AAV9 transduction (Fig 13). Finally, injection of AAV9-GFP-ACT3 in TA muscles of HSA-LR mice also corrects the splicing misregulation of several DM1 genes when compared to contralateral muscles injected with empty AAV9-MCS (Fig. 14A). Moreover, addition of a nuclear localization signal (NLS) to the ACT3 construct does not modify ACT3 efficacy in HSA-LR mice (Fig. 14B). In contrast, the removal of exon 3 from the ACT3 construct prevents its splicing correction activity.

Since MBNL2 is also able to bind to expanded CUG repeats leading to its sequestration, we examined whether splicing changes related to MBNL2 deficiency can be corrected by ACT3. As showed in figure 15A by co-transfection of hTau E2 minigenes with expanded CUG repeats and MBNL- or GFP -tagged ACT3 constructs in T98G cells, overexpression of MBNLl is not enable to correct the defective splicing of Tau exon 2. In contrast, overexpression of MBNL2 reverse this deregulated splicing event induced by the presence of expanded CUG repeats (Fig 15B) indicating that the missplicing of hTau E2 minigene induced by the presence of expanded CUG repeats is due to MBNL2 rather than MBNLl deficiency. A CT3 is also able to rescue the defective splicing of Tau exon 2 (Fig 15C). The rescue effect observed with either MBNL2 or ACT3 overexpression was abrogated while using a MBNL mutated minigene (Fig 15B and C) with mutated MBNL sites surrounding Tau exon 2 (Fig 15D) demonstrating that the rescue is not independent of MBNL or not due to an indirect effect. Therefore, A CT3 can rescue both MBNLl- and MBNL2 -deregulated splicing events by releasing several MBNL paralogues from expanded CUG repeats.

DISCUSSION

In this study we provided evidences that non- functional MBNL (ACT3), which is almost devoid of splicing activity is effective to counteract CUGexp-RNA toxicity both in vitro and in vivo. Thus, intramuscular administration of AAV vectors expressing ACT3 proteins corrects both alternative splicing misregulation and myotonia in DM1 mice. ACT3 expressing only the RNA-binding domain of MBNLl interacts with the pathogenic CUG repeats and releases sequestered MBNLl from the nuclear CUGexp-RNA foci. This mechanism restores endogenous functional MBNLl in DM1 muscle cells and corrects DM1 -associated phenotypes in vivo. This finding supports the development of a modified, non- functional MBNLA gene therapy approach as an alternate or complementary therapeutic approach for DM1.

Based on the ability of MBNL to bind to expanded CUG repeats with high affinity, we propose to use MBNLl RNA-binding domain as a bait to block deleterious interaction of poly-CUG binding proteins to pathogenic repeats. To test this hypothesis, we generated a modified, non- functional MBNL (ACT3) that contains only the MBNLl RNA-binding domain and lacks the C-terminal domains encoded by exons 5 to 10 that are responsible for MBNLl splicing regulatory activity, MBNL nucleocytoplasmic shuttling and most possibly MBNL oligomerisation (Tran et al. 2011). Our results confirm that ACT3 maintains its ability to bind to CUG repeats and co localizes with CUGexp-RNA in muscle cells, both in vitro and in vivo. As shown by in vitro crosslink assay, ACT3 displaces MBNLl from expanded CUG tracts suggesting that in vivo the binding of ACT3 to pathogenic DM1 repeats is able either to block deleterious interaction of MBNLl as well as other unidentified poly-CUG binding proteins or displace sequestered MBNLl from the nuclear CUGexp-RNA foci. As a consequence, release of functional MBNLl will reverse DMl-misregulated events.

Normalization of alternative splicing misregulation by ACT3 either in DM1 muscle cells or in skeletal muscle of DM1 mice supports the ability of ACT3 to target pathogenic CUG repeats and block access to endogenous MBNLl . However, since in vitro assays have showed that YGCY binding property of ACT3 are similar or slightly lower than MBNLl, we wondered whether ACT3 is able to directly modulate the MBNLl -regulated events. This seems unlike because splicing activity of ACT3 due to the lack of MBNLl exon 5 to 10 is dramatically reduced when compared to MBNLl using an in vitro minigene assay, and similar results were obtained with a ACT3-NES construct that has no splicing activity due to its strong nuclear export signal. But above all, no splicing changes were detected in wt mice or control human cells expressing ACT3. Rather, our results argue in favor of a release of endogenous MBNLl from the nuclear CUG-exp-RNA foci in muscle cells expressing ACT3 that restore functional MBNLl endogenous activity. While MBNLl is sequestered and colocalized with CUGexp- RNA foci in control HSA-LR mice, its localization is less associated with the nuclear foci than ACT3 in HSA-LR injected mice. Sequestration of ACT3 by CUGexp-RNA displaces endogenous MBNLl from these abnormal structures resulting in the correction of MBNLl - misregulated events in the DM1 mice.

Our AAV-ACT3 strategy is the first gene therapy approach that target CUGexp-RNA to inhibit deleterious of poly-CUG binding proteins and correct their toxic effects in vivo. To date MBNL proteins are almost the only proteins in DM1 human tissue samples that were found sequestered in nuclear foci, and recently, splicing abnormalities present in affected muscles of DM1 patients were mainly associated to functional loss of MBNLl (Nakamori et al. 2013). Depletion of functional MBNL splicing factors due to their abnormal binding and sequestration by CUGexp-RNA leads to alternative splicing misregulation of specific subset of transcripts and ultimately to pathological changes in DM1 tissues. Thus MBNLl -regulated events were associated to DM1 skeletal muscle defects whereas MBNL2 -regulated events were misregulated in DM1 brain. In addition, overexpression of functional MBNLl (isoforms-40 and -41) using AAV vectors is sufficient to reverse missplicing and myotonia in DM1 mice as confirmed by doubly transgenic HSA-LR:MBNLl-OE mice (Kanadia et al. 2006; Chamberlain and Ranum 2012). This strategy of functional MBNLl overexpression compensates for the loss of MBNLl in DM1 mice cells by increasing artificially the level of functional MBNLl . Thus MBNLl isoforms-40 and -41 were successfully overexpressed in muscles however up to 10 different MBNLl iso forms with various expression profiles and tissue-specific patterns were described. The function of different iso forms is not completely established yet as showed by the recent report indicating that MBNLl isoform-43 can interact with Src family kinase (Wang et al. 2012; Botta et al. 2013). In addition, MBNLl that regulates alternative splicing events is also involved in other RNA processes like mRNA decay and miRNA biogenesis (Rau et al. 2011; Masuda et al. 2012). ACT3 that targets CUGexp-RNA will circumvent the question of which iso form of MBNLl should be overexpressed since sequestered endogenous MBNLl proteins are released in a tissue-specific manner. Moreover, missplicing of MBNLl it-self that changes the MBNLl iso forms ratio in DM1 tissues is corrected in the muscle tissue of DM1 mice expressing ACT3. Besides it is not known whether MBNLl overexpression can restore or compensate for the loss of other MBNL paralogues such as MBNL2. We can presume that ACT3 will release other MBNL proteins from CUGexp-RNA and correct MBNL-misspliced events in other tissues than skeletal muscle. Our in vitro results indicate that ACT3 is most probably able to compensate for the loss of MBNL2 in a DM1 context (Fig. 15). In fact defective splicing of hTauE2 minigene in the presence of CUGexp-RNA can be restore by either MBNL2 or ACT3 but not by MBNLl overexpression suggesting that ACT3 is also able to counteract MBNL2- misregulated events in DM1. Therefore, together our results show that ACT3 can counterbalance the effect of CUGexp-RNA deregulated targets which or either regulated by MBNLl, MBNL2 or both. In contrast to a functional MBNLl overexpression strategy that will compensate for the loss of free and functionally available endogenous MBNLl due to their sequestration in CUGexp-RNA aggregates, the non- functional ACT3 will bind to expanded CUG repeats to release endogenous MBNLl from CUGexp-RNA aggregates and restore their cellular localization and function.

Among the therapeutic approaches currently under development for DM1, various modified oligonucleotides or small compounds targeting the mutant CUGexp-RNAs have shown promising beneficial effects in vivo (Mulders et al. 2009; Warf et al. 2009; Wheeler et al. 2009; Garcia-Lopez et al. 2011; Sobczak et al. 2012; Wheeler et al. 2012; Leger et al. 2013). Most of these strategies that reverse the muscle phenotype of DM1 mice share a common feature: release of sequestered MBNL paralogues from the CUGexp-RNA foci that leads to its cellular redistribution/relocalization and restores functional MBNL paralogues, resulting ultimately to correction of DM1 -associated phenotypes. This mechanism was described for strategies that cause either degradation of the CUGexp-RNAs or steric block of the expanded CUG repeats. Here we propose a novel AAV-ACT3 gene therapy for DM1. A single injection of AAV-ACT3 was efficient to neutralize RNA toxicity in DM1 mice. In contrast to synthetic oligonucleotides or small compounds that require repeated treatments, AAV vectors have been shown to persist several years in muscles (Rivera et al. 2005) allowing permanent expression of non- functional ACT3 that can counteract the continuous expression of toxic CUGexp RNA and trigger a long-lasting effect. Thus, we propose this new gene therapy approach as a valuable alternate or complementary therapeutic approach for DM1.

References Botta, A., A. Malena, et al. (2013). "MBNL 142 and MBNL 143 gene isoforms, overexpressed in DM1 -patient muscle, encode for nuclear proteins interacting with Src family kinases." Cell Death Pis 4: e770.

Brook, J. D., M. E. McCurrach, et al. (1992). "Molecular basis of myotonic dystrophy: expansion of a trinucleotide (CTG) repeat at the 3' end of a transcript encoding a protein kinase family member." Cell 69(2): 385.

Chamberlain, C. M. and L. P. Ranum (2012). "Mouse model of muscleblind-like 1 overexpression: skeletal muscle effects and therapeutic promise." Hum Mol Genet 21(21): 4645-54.

Charizanis, K., K. Y. Lee, et al. "Muscleblind-like 2-mediated alternative splicing in the developing brain and dysregulation in myotonic dystrophy." Neuron 75(3): 437-50.

Charlet, B. N., R. S. Savkur, et al. (2002). "Loss of the muscle-specific chloride channel in type 1 myotonic dystrophy due to misregulated alternative splicing." Mol Cell 10(1):

45-53.

Caillierez R, Begard S, et al. (2013). "Lentiviral delivery of the human wild-type tau protein mediates a slow and progressive neurodegenerative tau pathology in the rat brain".

Mol Ther. (7): 1358-68.

Carpentier C, Ghanem D, et al. (2014). "Tau exon 2 responsive elements deregulated in myotonic dystrophy type I are proximal to exon 2 and synergistically regulated by MBNL1 and MBNL2". Biochim Biophys Acta. 1842(4):654-664. Du, H., M. S. Cline, et al. (2010). "Aberrant alternative splicing and extracellular matrix gene expression in mouse models of myotonic dystrophy." Nat Struct Mol Biol 17(2): 187- 93.

Fardaei, M., K. Larkin, et al. (2001). "In vivo co-localisation of MBNL protein with DMPK expanded-repeat transcripts." Nucleic Acids Res 29(13): 2766-71.

Fischer, U., J. Huber, et al. (1995). "The HIV-1 Rev activation domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs." Cell 82(3):

475-83.

Francois, V., A. F. Klein, et al. (2011). "Selective silencing of mutated mRNAs in DM1 by using modified hU7-snRNAs." Nat Struct Mol Biol 18(1): 85-7.

Fugier, C, A. F. Klein, et al. (2011). "Misregulated alternative splicing of BIN1 is associated with T tubule alterations and muscle weakness in myotonic dystrophy." Nat Med 17(6): 720-5.

Furling. D, L. Coiffier, et al (2001). "Defective satellite cells in congenital myotonic dystrophy" Human Molecular Genetics 10 (19): 2079-87

Garcia-Lopez, A., B. Llamusi, et al. (2011). "In vivo discovery of a peptide that prevents

CUG-RNA hairpin formation and reverses RNA toxicity in myotonic dystrophy models." Proc Natl Acad Sci U S A 108(29): 11866-71.

Harper, P. S. (2001). Myotonic dystrophy Third Edn. , W.B. Saunder, London.

Holt, I., V. Jacquemin, et al. (2009). "Muscleblind-like proteins: similarities and differences in normal and myotonic dystrophy muscle." Am J Pathol 174(1): 216-27.

Hunter, A., C. Tsilfidis, et al. (1992). "The correlation of age of onset with CTG trinucleotide repeat amplification in myotonic dystrophy." J Med Genet 29(11): 774-9.

Kanadia, R. N., K. A. Johnstone, et al. (2003). "A muscleblind knockout model for myotonic dystrophy." Science 302(5652): 1978-80.

Kanadia, R. N., J. Shin, et al. (2006). "Reversal of RNA missplicing and myotonia after muscleblind overexpression in a mouse poly(CUG) model for myotonic dystrophy."

Proc Natl Acad Sci U S A 103(31): 11748-53.

Klein, A. F., E. Gasnier, et al. (201 1). "Gain of RNA function in pathological cases: Focus on myotonic dystrophy." Biochimie 93(11): 2006-12.

Laurent FX, Sureau A, et al. (2012). "New function for the RNA helicase p68/DDX5 as a modifier of MBNL1 activity on expanded CUG repeats". Nucleic Acids Res.

40(7):3159-71. Leger, A. J., L. M. Mosquea, et al. (2013). "Systemic delivery of a Peptide-linked morpholino oligonucleotide neutralizes mutant RNA toxicity in a mouse model of myotonic dystrophy." Nucleic Acid Ther 23(2): 109-17.

Lin, X., J. W. Miller, et al. (2006). "Failure of MBNL1 -dependent post-natal splicing transitions in myotonic dystrophy." Hum Mol Genet 15(13): 2087-97.

Mankodi, A., E. Logigian, et al. (2000). "Myotonic dystrophy in transgenic mice expressing an expanded CUG repeat." Science 289(5485): 1769-73.

Mankodi, A., M. P. Takahashi, 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 10(1): 35-44.

Masuda, A., H. S. Andersen, et al. (2012). "CUGBP1 and MBNL1 preferentially bind to 3'

UTRs and facilitate mRNA decay." Sci Rep 2: 209.

Miller, J. W., C. R. Urbinati, et al. (2000). "Recruitment of human muscleblind proteins to

(CUG)(n) expansions associated with myotonic dystrophy." EMBO J 19(17): 4439-48. Mouisel, E., B. Blondet, et al. (2006). "Outcome of acetylcholinesterase deficiency for neuromuscular functioning." Neurosci Res 55(4): 389-96.

Mulders, S. A., W. J. van den Broek, et al. (2009). "Triplet-repeat oligonucleotide-mediated reversal of RNA toxicity in myotonic dystrophy." Proc Natl Acad Sci U S A 106(33):

13915-20.

Nakamori, M., K. Sobczak, et al. (2013). "Splicing biomarkers of disease severity in myotonic dystrophy." Ann Neurol.

Pascual, M., M. Vicente, et al. (2006). "The Muscleblind family of proteins: an emerging class of regulators of developmentally programmed alternative splicing."

Differentiation 74(2-3): 65-80.

Ranum, L. P. and T. A. Cooper (2006). "RNA-mediated neuromuscular disorders." Annu Rev

Neurosci 29: 259-77.

Rau, F., F. Freyermuth, et al. (2011). "Misregulation of miR-1 processing is associated with heart defects in myotonic dystrophy." Nat Struct Mol Biol 18(7): 840-5.

Rivera, V. M., G. P. Gao, et al. (2005). "Long-term pharmacologically regulated expression of erythropoietin in primates following AAV-mediated gene transfer." Blood 105(4):

1424-30.

Savkur, R. S., A. V. Philips, et al. (2001). "Aberrant regulation of insulin receptor alternative splicing is associated with insulin resistance in myotonic dystrophy." Nat Genet 29(1): 40-7. Sobczak, K., T. M. Wheeler, et al. (2012). "RNA interference targeting CUG repeats in a mouse model of myotonic dystrophy." Mol Ther 21(2): 380-7.

Taneja, K. L., M. McCurrach, et al. (1995). "Foci of trinucleotide repeat transcripts in nuclei of myotonic dystrophy cells and tissues." J Cell Biol 128(6): 995-1002.

Tran, H., N. Gourrier, et al. (2011). "Analysis of exonic regions involved in nuclear localization, splicing activity, and dimerization of Muscleblind-like-1 iso forms." J

Biol Chem 286(18): 16435-46.

Tsilfidis, C, A. E. MacKenzie, et al. (1992). "Correlation between CTG trinucleotide repeat length and frequency of severe congenital myotonic dystrophy." Nat Genet 1(3): 192- 5.

Wang, E. T., N. A. Cody, et al. (2012). "Transcriptome-wide regulation of pre-mRNA splicing and mRNA localization by muscleblind proteins." Cell 150(4): 710-24.

Warf, M. B., M. Nakamori, et al. (2009). "Pentamidine reverses the splicing defects associated with myotonic dystrophy." Proc Natl Acad Sci U S A 106(44): 18551-6. Wheeler, T. M., A. J. Leger, et al. (2012). "Targeting nuclear RNA for in vivo correction of myotonic dystrophy." Nature 488(7409): 111-5.

Wheeler, T. M., J. D. Lueck, et al. (2007). "Correction of ClC-1 splicing eliminates chloride channelopathy and myotonia in mouse models of myotonic dystrophy." J Clin Invest

117(12): 3952-7.

Wheeler, T. M., K. Sobczak, et al. (2009). "Reversal of RNA dominance by displacement of protein sequestered on triplet repeat RNA." Science 325(5938): 336-9.