WO2010150231A1 - Antisense oligonucleotides capable of inducing exon skipping and the use thereof as a medicament for the treatment of duchenne muscular dystrophy (dmd) - Google Patents

Abstract

The invention relates to the use of an antisense oligonucleotide directed against an exon sequence of the dystrophin gene for preparing a medicament capable of determining the skipping of said exon from the mRNA dystrophin transcript, thereby promoting the production of a functional or partially functional dystrophin protein in a patient carrying a small mutation in said exon. Such a use is characterised in that the exon is selected from the group consisting of exon 10, exon 16, exon 26, exon 33 and exon 34 of the dystrophin gene. Specific antisense oligonucleotides suitable to be used in the invention are also described.

Description

Antisense oligonucleotides capable of inducing exon skipping and the use thereof as a medicament for the treatment of Duchenne muscular dystrophy (DMD)

The present invention falls within the field of therapeutic treatments for Duchenne muscular dystrophy (DMD) by the exon skipping approach.

Nonsense mutations or those that induce frame-shifting in the dystrophin gene (DMD, MIM# 310377) result in Duchenne muscular dystrophy (DMD MIM# 310200), a serious neuromuscular disorder caused by the complete absence of a functional dystrophin. Instead, Becker muscular dystrophy (BMD, MDM# 310376), which is a less serious allelic disorder, is caused by in- frame mutations that result in a shorter yet functional protein. [Hoffman et al., 1987; Munich 1989]. The hypothesis of the reading frame explains the phenotype differences detectable in about 92% of the BMD/DMD cases [Aartsma-Rus et al., 2006; Koenig et al., 1989]. Even though most of the mutations, in the DMD gene consist in large rearrangements, the frequency of deletions and duplications (which are thought to be about 80-85%) has been undoubtedly overestimated in the past years. Recently, thanks to the improvement in diagnostic systems [Bennett et al., 2001; Buzin et al., 2005; Flanigan et al., 2003; Hofstra et al., 2004; Roberts et al., 1993, 1994; Tuffery-Giraud et al., 2004; Whittock et al., 1997], an increasing number of both nonsense mutations and small frame-shifting mutations has been described. Now it is estimated that point mutations account for nearly 30% of the dystrophin mutations [Aartsma-Rus et al., 2006; Deburgrave et al., 2007].

At this point, with the approach of the clinical trials specific for each of the mutations, the characterization of the dystrophin mutations has become necessary. For instance, only those patients who carry nonsense mutations are eligible for the I/IIa stages of the clinical trials performed by PTC Therapeutics, which are based on the stop codons read-through strategy [Welch et al., 2007].

Another mutation-specific approach is the exon skipping induced by antisense oligonucleotides (AONs), which has the aim of restoring the reading frame of the dystrophin transcripts. Among the various antisense types, the antisense oligonucleotides 2'OMePS AONs (2'-O-methyl- modified ribose molecules with a complete phosphorothioate backbone) targeting exon 51 have recently been used in a pilot study in 4 DMD patients, obtaining very encouraging results [van Deutekom et al., 2007].

Since most of the mutations that cause DMD are out-of-frame deletions clustered within the two major hotspot regions, the exon skipping approach has been nearly exclusively studied in this type of rearrangements, as with the skipping of a single exon theoretically 75% of the deletion mutations can be brought back in frame [Aartsma-Rus et al., 2004]. Only two nonsense mutations have been addressed with an in vitro approach by antisense oligonucleotides 2'OMePS AONs designed on the basis of the wild type exon sequence [Aartsma-Rus et al., 2003, 2004].

However, there is a non-negligible percentage of patients carrying small mutations within in- frame exons, which could be eligible for treatments based on the single exon skipping approach.

Thus, there is a need of verifying if such an approach can also be extended to this type of DMD gene mutations.

To that end, the inventors carried out studies on patients with Duchenne muscular dystrophy but without big deletions or insertions within the dystrophin gene.

More specifically, the inventors identified 54 unrelated patients with Duchenne muscular dystrophy (DMD) carrying small mutations within the DMD gene. Out of 24 patients carrying a mutation within an in-frame exon, the inventors selected 5 mutations to be modulated in 5 different exons. The inventors induced skipping of a single exon by using AONs designed based on the normal or wild type sequence or the mutated sequence, both in splicing assays in a cell- extract (cell-free) system and in myogenic cell cultures derived from 5 DMD patients. This way, it has been proved that the skipping efficiency of some of the mutated exons differed significantly from what previously reported in connection with the corresponding non-mutated or wild type exon.

Through the studies carried out by the inventors, which will be described in detail in the ex- perimental section that follows, it has been possible to establish that the exon skipping approach is actually applicable to small mutations within the dystrophin gene, responsible for the DMD phenotype.

Therefore, a first object of the present invention is the use of an antisense oligonucleotide directed against an exon sequence of the dystrophin gene for preparing a medicament capable of determining the skipping of said exon from the dystrophin transcript (mRNA), thereby promoting the production of a functional or partially functional dystrophin protein in a patient carrying a small mutation in said exon, characterised in that the exon is selected from the group consisting of exon 10, exon 16, exon 26, exon 33 and exon 34 of the dystrophin gene. Such a use is the object of claim 1 of the present patent application. The subordinate claims specify further characteristics of the use according to the invention and are an integral part of the specification.

A second object of the invention is an antisense oligonucleotide suitable to be used for the skipping of an exon within the dystrophin gene, including a small mutation responsible for the Du- chenne muscular dystrophy phenotype. The antisense oligonucleotides suitable to this end have been studied by the present inventors and the characteristics thereof are illustrated in Table I. The nucleotide sequences of the antisense oligonucleotides of the invention are provided in the Sequence Listing, wherein they are indicated as SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9 and SEQ ID NO: 10, respectively. Given their ability of causing the skipping of an exon having a small mutation capable of causing the Duchenne muscular dystrophy phenotype, the antisense oligonucleotides of the invention are particularly suitable to be used as a medicament, more particularly as a medicament for the therapeutic treatment of the Duchenne muscular dystrophy.

The invention is described in more detail in the following experimental section, referring to the appended drawings, wherein:

Figure 1 represents the in vitro splicing assay of the mutated exon 34 (c.4780delTins37) on cell extracts at incubation times of 1 - 2 - 3 hours, and the characterization of all of the fragments produced by sequence analysis. Splicing pBG exon34: splicing: A: 1127 transcript not subjected to splicing. B: transcript subjected to partial splicing, wherein the rabbit beta- globin exon 2 is linked to exon 34, while the intron between rabbit beta-globin exon 34 and ex- on 3 is maintained. C: mature splicing product consisting of rabbit beta-globin exon 2 - mutated exon 34 - rabbit beta-globin exon 3. D: aberrant splicing product due to the recognition of a cryptic intron splicing site. E: regular 280-nucleotide splicing product consisting of exon 2 connected to exon 3. F: exon 2 (224 nucleotides). Also shown are the chromatograms of the exon- exon splicings.

Figure 2 A represents the specific exon skipping by using 2OMePS AONs in the in vitro splicing assays on cell extracts. The antisense oligonucleotides wtAONs are designed based on the normal sequence, the antisense oligonucleotides msAONs are designed based on the mutated sequences. The 5 mutated exons (patients 3, 7, 18, 21 and 23) were analyzed by in vitro splicing in the presence of increasing concentrations of AONs (100, 200 and 500 nM). The products obtained in this in vitro assay are shown.

Figure 2 B is a graphical representation of the densitometric analysis of the exon skipping percentages obtained by using 100, 200 and 500 nM concentrations of all of the different AONs. The colors and the bars distinguish the skipping efficiency of each mutated exon with wtAON or msAON. The skipping behavior does not change significantly at the different AON concentrations used.

Figure 2 C is a graphical representation of the average skipping value, considering the results obtained with all of the increasing concentrations of AONs. The skipping efficiency differs significantly only for oligo wtAON on exon 33 and oligo msAON on exon 34.

Figure 3 A illustrates the RT-PCRs of the exon skipping induced by AONs in myogenic cells. Skipping of exons 10 (patient 3), 26 (patient 18) and 34 (patient 23) was detected by nested RT-PCR, whereas for the other two exons only a primary RT-PCR was performed. New and shorter transcripts were observed for all exons, with the exception of exon 10, when compared to control myotube cultures (NT). The sequence analysis confirmed the precise skipping of the target exons, with the exception of exon 10, in which skipping of exon 9 was confirmed. Skipping of exon 9, which occurs normally, was also observed in the control myotubes. M indicates the molecular weight marker, B: blank.

Figure 3 B is a graphical representation of the quantitative differences in the exclusion of the dystrophin exons 10, 16, 26, 33 and 34 from the mature transcript in Myo-D-transformed AON-treated fibroblasts derived from the selected patients, compared to untreated cells, calcu- lated by ESRA (Exon-Specific Real-Time Assays). The histograms represent the percentage of specific exon skipping of treated cells compared to untreated cells, by using wtAON (grey bars) or msAON (white bars). In patient 3 (c.l 132 1 135dup in exon 10), the ESRA assay, by using exons 6 and 16 as the references, showed exon 10 skipping levels of 2.4% with wtAON and 10.9% with msAON. On the contrary, patient 23 (c.4780delTins37 in exon 34) showed a skipping of 6.5% (wtAON exon 34) and of only 1.6% (msAON exon 34) by using exons 26 and 40 as the references. For patients 7 (c.l912delC in exon 16), 18 (c.3447_3448delinsTT in exon 26) and 21 (c.4565delT in exon 33), the skipping levels were comparable for wtAONs and msAONs. However, the skipping levels of exon 33 were very high (58.8% and 58.3%, respectively, by using exons 10 and 26 as the references), whereas the skipping levels of exon 16 were lower (13.2% and 15.8%, respectively, by using exons 10 and 20 as the references), and the skipping levels of exon 26 were hardly detectable (1.6% and 2.3% with exons 10 and 16 as the references).

Figure 4 illustrates the immunohistochemical analysis of dystrophin in cells from patients. A) The double staining with antibodies directed against the myosin heavy chain (green) and dystrophin (red) clearly shows restoration of dystrophin expression in all of the treated Myo-D- transformed fibroblasts. Even though most of the cells show a cytoplasmic staining of dystrophin, at a high magnification (B, d) dystrophin is seen to localize correctly at the myotube sar- colemma. wtAON: antisense designed based on the normal exon; msAON: antisense designed based on the mutation. B. Cells from patients untreated and treated with AONs become stained with the antibodies anti-myosin heavy chain and anti-dystrophin. No dystrophin signal is detected in untreated cells (a, b) stained with the polyclonal HC300 antibody, whereas a dystrophin signal is seen on the membrane of treated cells (c, d) after induction of exon skipping.

The following experimental section is provided merely by way of illustration and not limitation of the scope of the invention as defined in the appended claims.

EXPERIMENTAL SECTION

Materials and Methods Analysis of mutations

The DNA from 54 patients with Duchenne muscular dystrophy (DMD) was extracted from whole blood using a BioRobot QIAGEN (Universal System 8000), after having obtained the informed consent and that for diagnostic purposes. All of the patients were negative for deletions and duplications within the DMD gene, as checked by MLPA. Amplification of the single ex- ons was obtained by using 79 pairs of primers designed by the present inventors. The amplicons were sequenced by using an automated 3130 sequencer (Applied Biosystem) by following the manufacturer's instructions.

Constructs

The 5 selected mutated exons and the flanking intron sequences (see Table I) were amplified from the genomic DNA of the patients and were cloned into the vector pBG described in [Gua- landi 2003]. The forward primers of each amplicon were designed upstream of the consensus sequence of the branch site, while the reverse primers were located at least 100 nucleotides downstream of the splice donor site. The correct nucleotide composition of each construct was checked by sequencing.

AON design and synthesis

The design of the AONs on the normal sequences was performed on the basis of the highest ESE-Finder value and the probability of sites for SF2/ASF and SC35, paying attention to the absolute distance from the 3' splice site, as previously reported [Aartsma-Rus et al., 2005]. The AON sequences, the characteristics thereof and the selected parameters are shown in Table I.

The synthesized AONs contain a complete phosphorothioate backbone and ribose molecules modified with 2'-O-methyl (2OMePS AON). The synthesis of the oligonucleotides was carried out with a DNA/RNA synthesizer AKTATM oligopilot plus 10 (GE Healthcare) using the tri- tyl-on mode. The molecule was synthesized on a 2 μmol scale using a Primer Support 200 charged at 80 μmol/g (Amersham Biosciences). Commercial 2'-O-methyl phosphoramidites (Proligo) were dissolved at a nominal concentration of 50 mM in anhydrous acetonitrile (CH3CN) and activated with a 0.25 M solution of 5-(bis-

3,5-trifluoromethylphenyl)-lH-tetrazole (Proligo) in CH3CN. The final de-tritylation was obtained with a 0.1 M aqueous solution of NaOAc (pH 3.0). The DMT-protected de-tritylated crude oligonucleotides were purified with the liquid chromatography system AKTAbasic UPC with the 3ml column Amersham Biosciences Resource RPC eluted with a CH3CN gradient in

0.1 M TEAA (pH 8). The final oligonucleotides were dissolved in water and filtered through a short Dowex 50WX8 column (Na⁺ form, 100-200 mesh) to give, after lyophilization, 0.8 umol (40%) of the desired product. The purity of the complete desired product was assessed by MALDI-TOF MS, 31 P- NMR and RP-HPLC analysis.

In vitro splicing assay on cell extracts

Five in frame-mutated exons cloned in the splicing vector pBG were transcribed in vitro using T7 RNA polymerase and α32-rGTP and were purified from 5% acrylamide gels in order to perform in vitro splicing assays in HeLa nuclear extracts, as previously described [Gualandi et al., 2003]. All of the splicing products were subjected to reverse transcription with the High-capacity cDNA Reverse Transcription kit (Applied Biosystems). The cDNAs were amplified with the oligos Ex2/Ex3 designed based on the rabbit beta-globin sequences [Gualandi et al., 2003], cloned into the vector pCRϋ (TA Cloning Invitrogen) and sequenced on ABI Prism 3130 (Applied Biosystems) to verify the accuracy of the exon splicing process.

In order to induce exon skipping in the in vitro cell extract, 2OMePS AONs were used at various concentrations from 100 to 500 nm. Incubation was carried out for 3 hours. For the semi- quantification of the amount of skipped transcript, the amount of product which underwent splicing was compared to the total amount of product. It was assumed that the two transcripts (unspliced/spliced) represented a total of 100%.

Cultivation of myogenic cells and trans fection of AONs

Human primary fibroblasts of 5 patients (3, 7, 18, 21 and 23 in Table II) were isolated from a skin biopsy (obtained after informed consent for research purposes, Ethical Approval N. 9/2005) by explant cultures. The cells were grown in high-glucose DMEM (GIBCO) supplemented with 20% fetal bovine serum (GIBCO) and antibiotic/antimycotic solution (Sigma). Myogenesis was induced by infection with an Ad5-derived adenovirus vector, with EAl deletion, carrying the MyoD gene as previously described [Aartsma-Rus et al., 2003; Havenga et al., 2002; Roest PA et al., 1996]. All the experiments were done by using cells which had underwent no more than 4 passages. Myotubes obtained after 10-14 days of cultivation in differentiation medium (2% FBS) were transfected with the AONs [100 nM] in the presence of polyethyl- eneimine (ExGen500, MBI Fermentas) (2 μl per μg of AON) as the transfection reagent, according to the manufacturer's instructions [Van Deutekom et al., 2007].

Studies on the RNA of the myogenic cultures

48 hours after transfection, the total RNA was isolated from the myotube cultures (RNeasy Kit Qiagen) and subjected to reverse transcription into cDNA by using a random primer and the high-capacity cDNA reverse transcription kit (Applied Biosystems). The RT-PCR was performed on beta-actin, in order to check the cDNA synthesis, and on dystrophin by using primers complementary to many flanking exons, in order to rule out spontaneous multi-exon skipping. For exons 26 and 34, a nested PCR has been necessary to detect the specific exon skipping.

The PCRs were carried out with 5 cycles of 94°C (30s), 63°C (30s) and 72°C (30s), plus 30 cycles of 94°C (30s), 62°C (30s) and 72°C (30s). The PCR products were analyzed on 1.5% agarose gels. It should be noted that no evidence of a significant preference for amplification of shorter fragments was obtained in the RT-PCR. All of the amplified fragments were sequenced on ABI Prism 3130 (Applied Biosystems).

Exon-Specific Real-Time Assay (ESRA)

In order to precisely quantify the percentage of exon skipping, eight exon-specific real-time assays (ESRA) were set up, which were able to detect exons 6, 10, 16, 26, 33, 34, 40 and 70 of human dystrophin. These exons were selected because in human beings they are not involved in spontaneous alternative splicing events. The ESRA assays on exons 6, 10, 16, 26, 40 and 70 were used as a reference for quantifying the amount of dystrophin transcript compared to control myotubes, the percentage of physiological exon skipping and the percentage of induced ex- on skipping in treated cells compared to untreated cells (internal reference). These data were confirmed by using an external reference gene (beta-actin gene) on cDNA samples from each myotube culture (treated and control cells). All the ESRA assays were based on the TaqMan MGB technology and were designed with the PrimerExpress Applied Biosystems software. The amount of target sequences compared to internal references (represented by a non-mutated dystrophin exon) and compared to an appropriate endogenous control (beta-actin gene) was assessed by the CT comparative method in comparison with the untreated control (_Ct Method) (Applied Biosystems User Bullettin #2).

Immunohistochemical analysis

The cultures of treated myotubes were grown on coverslips and fixed in methanol at -20°C from 2 to 6 days post-transfection, depending on the survival rate of the myotubes. The samples were incubated for 30 minutes in phosphate buffered saline (PBS) (Euroclone). All the samples were stained with polyclonal anti-dystrophin antibodies (Santa Cruz Biotechnology), diluted 1 :40, washed with PBS (Sigma) and detected with a TRITC-conjugated secondary anti-rabbit antibody, all the samples were also stained with mouse monoclonal antibodies against desmin or myosin heavy chain (Novocastra Laboratories Ltd), diluted 1 :10 and 1 :60, respectively. After several washes with PBS, all of the samples were incubated with FITC-conjugated secondary anti-mouse antibodies (DAKO). The slides were mounted with an anti-fade mounting medium (Molecular Probe) and analyzed with a fluorescence Nikon Eclipse 80i microscope.

Results

Identification of the small mutations

A series of 54 patients with no deletion or duplication identified by MLPA were screened for small mutations. The identified mutations are reported in Table II. The mutation in patient 23 had been described previously [Tuffery-Giraud et al., 2004] and was included in the table because the patient's fibroblasts were used for modulation with AONs.

Pathogenic mutations were identified for each patient. Mutations include splicing, reading frame shift and nonsense mutations and 24 DMD patients exhibit small mutations in 16 different in-frame exons.

In order to confirm the applicability of exon skipping for small mutations in in-frame exons, 5 patients to be studied were selected [number 3 (c.l l32_1135dup in exon 10), number 7 (c.l912delC in exon 16), number 18 (c.3447_3448delinsTT in exon 26), number 21 (c.4565delT in exon 33) and number 23 (c.4780delTins37 in exon 34)].

AON design

AONs specific for the mutation (mutation-specific, msAONs) and AONs that did not cover the mutation (wild type, wtAONs) were designed for each exon. AONs targeted to exons 16 and 26 had been designed previously, but they covered the splice acceptor sites [Wilton et al., 2007]. The inventors decided to design and synthesize new AONs targeted to internal sequences in exons 16 and 26. For exon 34 (proved to be non-skippable), the inventors designed a new AON closer to the 3' splice site [Wilton et al., 2007], whereas the wtAON against exon 33 is 5 nucleotides shorter than the one reported previously [Wilton et al., 2007].

The ESE-finder software showed that the mutations in exon 10 and in exon 34 did not determine differences in the predicted SR protein binding sites. On the contrary, the mutated exon 16 shows two additional SR protein binding sites, caused by the mutation (among them, the highest score was for protein SC35). In exon 33, the mutation causes formation of a binding site for protein SC35. Mutation in exon 26 causes loss of two SR protein binding sites.

Considering these data and taking into account the fact that the object of the inventors was also to include mutation-specific AONs, it has been possible to design msAONs for exons 10, 16, 26 and 33 in accordance with the rules described above in the Materials and Methods section. Mutation in exon 34 entails the presence of a new stretch of 37 nucleotides, and the inventors decided to design the msAON so that it would be targeted to the inserted sequence. This stretch does not cause significant changes in the SR protein binding sites predicted by ESE-finder analysis.

Splicing assays on in vitro cell extracts

The 5 mutated exons (10, 16, 26, 33 and 34 from patients 3, 7, 18, 21 and 23 in Table II) were cloned into the splicing vector pBG [Gualandi et al., 2003] and analyzed in an in vitro splicing assay.

The splicing products were characterized and all the exon-exon splicings were sequenced (data not shown). In any case, the sequencing of the splicing products showed that the mutated dystrophin exons were spliced correctly between exons 2 and 3 of rabbit beta-globin (Figure 1). Modulation with antisense oligonucleotides 2OMePS msAONs and wtAONs was able to induce exon skipping in 4 constructs out of 5 in splicing assays on cell extracts at the increasing AON concentrations used (Figure 2A).

The semi-quantitative method used was based on the quantification of the amount of spliced transcript (including the exon) compared to the total amount of product (spliced and unspliced).

In most of the in vitro assays, the percentage of exon skipping was not dependent on the dose of AON (Figure 2B). Therefore, the average value of the skipping percentages was calculated (Figure 2C), which can be compared more easily with the AON efficiency in the patient's cells. The inventors found that the msAON was significantly more effective for the skipping of exon 34 (85.55% msAON versus 9.19% wtAON) and of exon 10 (25.63% msAON versus 14.83% wtAON), whereas the wtAON was more effective for the skipping of exon 33 (50.72% wtAON versus 19.93% msAON). No difference was observed in the skipping efficiencies obtained with the wild type and mutation-specific AONs for exon 16, whereas exon 26 resulted non-skippable in vitro (Figure 2C).

Transcription analysis in myogenic cells The same AONs tested in the in vitro splicing system were used in the cell cultures. The inventors were able to detect two products that represent the skipped and non-skipped transcripts, by primary RT-PCR (exons 16 and 33) and nested RT-PCR (exons 26 and 34) in patient-derived cell cultures (Figure 3A). On the contrary, no skipping of exon 10 was observed by primary RT- PCR or nested PCR, as previously noted (Figure 3A) [Wilton et al., 2007]. These experiments were highly reproducible.

The RT-PCR of treated and untreated cells did not show any unspecific splicing product, with the exception of cells carrying mutation in exon 10, in which a low-level physiological skipping of exon 9 was always observed. This event was also observed in untreated control muscle cells, as previously reported [Aartsma-Rus et al., 2005; Reiss J at al., 1994].

In order to precisely quantify the specific exon skipping level and to detect very low skipping quantities, an exon-specific real-time assay (ESRA) was used in all the treated and untreated cells of the patients. Quantification was performed by using both adjacent dystrophin exons and beta-actin as the reference transcripts and they were compared with the untreated cells. The skipping percentages are shown in Figure 3B.

In patient 3 (c.l 132 1135dup in exon 10), the ESRA assay with exons 6 and 16 as the references showed exon 10 skipping levels of 2.4% with wtAONs and 10.9% with msAONs. On the contrary, in patient 23 (c.4780delTins37 in exon 34), a skipping of 6.5% (wtAON exon 34) and only of 1.6% (msAON exon 34) was detected by using exons 26 and 40 as the references. For patients 7 (c.l912delC in exon 16), 18 (c.3447_3448delinsTT in exon 26) and 21 (c.4565delT in exon 33), the exon skipping levels were comparable by using msAONs and wtAONs. However, the skipping levels for exon 33 were very high (58.8% and 58.3%, respectively, by using exons 10 and 26 as the references), whereas the skipping levels for exon 16 were lower (13.2% and 15.8%, respectively, by using exons 10 and 20 as the references), and the skipping levels for exon 26 were hardly detectable (1.6% and 2.3% with exon 10 and exon 16 as the references). The ESRA assays, by assessing the skipping levels in comparison with many references, gave robustness to the data, allowing the inventors to detect the skipping of exon 10, not visible by RT-PCR, and allowing to identify a low level (less than 0.5%) of spontaneous skipping of the mutated exons in all the control cells.

In conclusion, for a few exons the differences in the efficiencies are not significant (exons 16, 26 and 33), whereas msAONs induce higher skipping levels in exon 10, and wtAONs are more efficient for the skipping of exon 34 (Figure 3B).

Immunohistochemical analysis

Immunohistochemical analysis of desmin (data not shown) and of the development myosin heavy chain (Figure 4A) was carried out to define the differentiation stage of the Myo-D- transfoimed fibroblasts.

The double staining with anti-dystrophin antibodies detected protein restoration in all the patient's cells treated with wtAON and msAON. The number of dystrophin-positive myotubes ranged greatly between 5 and 15%. Lower levels (5%) were detectable in exon 10-mutated and wtAON-treated cells, in exon 26-mutated cells treated both with wtAON and msAON, and in exon 34-mutated and msAON-treated cells. This corresponds to the lowest exon skipping levels observed with these AONs (Figure 3B). Dystrophin localized at the sarcolemma in all of the transfected myotubes (Figure 4).

DISCUSSION

The small mutations in the dystrophin gene are scattered all over the encoding sequence and for this reason the development of therapeutic approaches specific for each mutation represents an extremely complex challenge. Also the development and optimization of AONs such as to allow for an increase in the exon skipping levels represents a difficult task. Therefore, the object of the work done by the inventors was to induce an AON-mediated exon skipping for in-frame exons containing small mutations. The mutated exons must be studied in a specific way, as it is well known that small mutations producing premature stop codons in in-frame exons can occasionally result in skipping of the mutated exon replacing the dystrophin transcript in frame and thereby causing the phenotype BMD instead of the expected DMD phenotype, can create cryp- tic splice sites or can alter the exon recognition [Aartsma-Rus et al., 2006; Deburgrave et al., 2007].

As will be illustrated in detail in the following experimental section, the inventors tested an- tisense oligonucleotides 2'OMePS AONs designed to be targeted to normal sequences (wtAON) or - for the first time - to mutated sequences (msAON), in order to assess the skipping levels and compare the skipping efficiency obtained with the oligonucleotides msAONs and wtAONs.

The inventors took advantage of the fact that they identified a large group of patients with small mutations within the DMD gene. As previously indicated, 5 in-frame dystrophin exons (exons 10, 16, 26, 33, 34) were modulated, each containing different small mutations responsible for the DMD phenotype.

The study carried out on patients' cells allowed to prove that the modulation of these 5 exons with wtAON determines restoration of the dystrophin transcript reading frame. This brings about restoration of the dystrophin protein synthesis and the localization thereof at the sar- colemma. This result confirms the applicability of exon skipping to in-frame exons having small mutations [Aartsma-Rus et al., 2003, 2004].

Also 5 mutation-specific AONs (msAONs) were designed, which were used to induce exon skipping in myogenic cells.

The msAON for exon 10 induced a significantly higher level of specific exon skipping, as detected by ESRA. Instead, wtAON worked better for exon 34. These data are partially consistent with what is reported in the literature. The inventors confirmed that the parameters proposed for the design of the antisense oligonucleotides are effective and that the in-frame exons are sensitive to exon skipping [Aartsma-Rus et al., 2005; Wilton et al., 2007]. Mutation in exon 10 does not alter the composition of the exon in terms of splicing enhancer motifs and therefore the antisense oligonucleotides designed based on the wild type and mutated sequences (wtAON and msAON) would be expected to allow for obtaining a similar skipping level. Wilton and col- leagues [Wilton et al., 2007] found that the skipping of exon 10 can not be achieved per se, using a single AON, and that the skipping of exon 10 is always accompanied by skipping of exon 9. The inventors, by ESRA, could observe a considerable level (10.9%) of exon 10 skipping by using the mutation-specific antisense (msAON). One possible explanation is that the mutation modifies the conformation of exon 10, maybe increasing the skipping inclination mediated by the antisense oligonucleotides. Even if encouraging, the ESRA results obtained on exon 10 do not allow to determine if also the skipping of exon 9 occurs simultaneously. It is however to be pointed out that skipping of exon 9 is a very common physiological alternative splicing event and therefore it could be very difficult to prevent it from happening. Moreover, as exon 9 is in frame, very likely this exon skipping is not detrimental.

Mutation of exon 16 causes an increase in the value predicted by ESE Finder for SC35, which is one of the main parameters to be considered when designing AONs [Aartsma-Rus et al., 2005]. Therefore, a higher effectiveness of msAONs could be speculated, suggesting that point mutations that create SR protein binding sites are probably a good target. However, the skipping levels reached for this exon both with wtAON (13.2%) and msAON (15.8%) are much lower when compared with what reported previously. Again, the composition of the exon sequence could be important in the definition of the exon and thus for its skipping inclination. It is possible that when a mutation creates new or improved SR protein binding sites, this determines an increase in the recognition of the exon, thereby making exon skipping more difficult.

Exon 26 is considered a poorly skippable exon (ratio <10%) [Wilton et al., 2007]. Both wtAON and msAON used in the present invention induced very low skipping levels similar to each other (1.6% and 2.3%, respectively). This finding is again surprising, since the loss of the two SR protein binding sites caused by the mutation led to the prediction of a higher skipping inclination.

The inventors have been able to obtain exon 33 skipping very efficiently, as Wilton and colleagues [Wilton et al., 2007], irrespective of the AON used.

The insertion mutation in exon 34 is difficult to analyze because, even though it does not appear to modify any binding site for splicing factors, this long stretch can obviously alter the exon inclination to skipping in other ways. AONs for inducing skipping of exon 34 have not been identified yet [Wilton et al., 2007]. However, the inventors were able to obtain skipping of the mutated exon 34 in 6.4% of the dystrophin transcripts by using wtAON. One possible explanation is that the insertion/deletion mutation (insertion of 37 nucleotides and deletion of 1 nucleotide) increases the inclination of exon 34 to skipping by increasing the distance between the motifs recognized by SR proteins and other splicing proteins. Alternatively, modifying the RNA's secondary structure could affect the recognition of the mutated exon, even though the accessibility of the splicing sites is the same for the mutated and the wild type exon 34 (data not shown).

The inventors desire to emphasize the quantitative nature of the method used for determining the exon skipping level (ESRA). This method allowed the amount of skipped transcript to be quantified in a precise way, both compared to internal references, such as dystrophin exons not subjected to alternative splicing, and to an external reference gene (β-actin). Furthermore, the assay can detect a very low level of skipping induced by AON (exon 10) and of physiological skipping, not detectable by RT-PCR.

The inventors also performed splicing assays in systems based on cell extracts (cell-free systems) to reproduce the pre-mRNA splicing of the mutated dystrophin exons in a reporter construct. The splicing was reproduced successfully in the system based on cell extracts for all of the 5 dystrophin exons analyzed. However, modulation by AONs was successful only for exons 10, 16, 33 and 34. It is interesting to note that, for exons 10, 16 and 33, the msAON (exon 10) and the wtAONs (exons 16 and 33) always work better with the myogenic cells. Exon 26 was poorly skippable both in cells and cell extracts.

Instead, exon 34 resulted better skippable with the msAON in cell extracts than in myogenic cells.

These data are interesting in that they suggest that a splicing assay in cell extracts can be used to test the inclination of an exon to skipping. However, the inventors point out that - since the ex- on's composition, length and flanking intron regions are crucial for in vitro splicing [Eperon et al., 1988] - the long stretch included in the mutated exon 34 represents an obstacle to the optimal AON identification in the in vitro assay. These data allowed the inventors to conclude that in vitro splicing may enable to assess the modulation of the exon but not to accurately distinguish between mutation-specific AONs and wild type AONs, in case the small mutation alters greatly the composition of the exon. Therefore, this can not be considered as a general tool for optimizing AONs.

In conclusion, by using the antisense oligonucleotides wtAON and msAON directed against ex- ons 10, 16, 26, 33 and 34 of the dystrophin gene, the inventors were able to induce specific exon skipping of all of the 5 in frame-mutated exons in myogenic cells. This strongly suggests that the small mutations may affect the definition of the exon, thereby modifying its inclination to be skipped by an AON, which implies that the exon skipping approach for in frame exons with small mutations deserves careful examination, as the skipping efficiency may depend on the mutations. The inventors modulated exons with small mutations thus belonging to a genomic context wherein the dystrophin gene does not show rearrangements (the configuration of the exons and introns is intact). It is noteworthy that the AON-mediated skipping of these 5 exons never appeared to be accompanied by other unspecific splicing products, such as for instance the induction of a cryptic splicing. The only product of an alternative splicing event is represented by skipping of exon 9, which however was also observed in the control myotubes. It could be supposed (and this is very interesting) that fusion introns, derived by splicings subsequent to deletion or duplication, may affect the recognition of the adjacent exon and thus the splicing accuracy.

Despite the promising results obtained with the exon skipping approach, another crucial aspect must be considered: the possible effect caused by the absence of the skipping-target exons from the dystrophin protein. Isolated deletions of exons 10 and 34 are described, which cause a BMD phenotype or only a CK-high phenotype. The isolated deletion of exon 16 only causes a CK- high phenotype [Schwartz et al., 2007 and personal observation]. However, as to exons 26 and 33, the Leiden database does not report isolated deletions [White et al., 2006]. Therefore, phe- notypes resulting from the skipping of these exons are difficult to predict, even though this could also imply that these deletions have not yet been identified because they are asympto- matic. However, all these belong to the spectrin-like region (exon 10 up to repeat 1, exon 16 up to repeat 3, exon 26 up to repeats 7-8, exons 33 and 34 up to repeats 11 and 12) and thus, on the basis of the literature data, the skipping thereof is expected to have slight consequences, of the BMD type, on the function of the protein.

These data represent the first attempt to modulate small mutations within the dystrophin gene by using both wild type and mutation-specific AONs and highlight both the particular splicing characteristics of the mutated exons and the complexity in designing optimal AONs for a therapy based on exon skipping, in the perspective of a customized medicine.

Table I. Characteristics of the AONs

Table II. List of patients carrying small mutations ir i dystrophin exons

Type of

ID Mutation Exon Skipping mutation

1 c.713/716 delTT exon 8 (stop ex 8) frameshift

2 c. 676_678del exonδ deletion

3 exon 10 (stop ex frameshift skipping ex 10 c. l l32insdupCAGT H)

4 c.l292/1294InsG exon 1 1 frameshift

5 c.1331 + 1 G > T exonl l/Intl l splicing

6 c.l482+ldelG exonxl2/Intl2 splicing

7 c. l910/1912 delC exon 16 frameshift skipping ex 16

8 exon 17 (stop ex frameshift c.2141/2142 insT

17)

9 C.2302 C > T exon 19 stop

10 c.2474 G > A exon 20 stop

1 1 C.2791 G > T exon 21 stop

12 c.2947 O T exon 22 stop

13 c.2949+1 G > A exon22/Int22 splicing

14 c.2950-2 A to G Int22/exon23 splicing

15 c.2956 O T exon 23 stop skipping ex 23

16 exon 25 (stop ex frameshift skipping ex 25 c.3345/3347delA 26)

17 c.3433-1 G > A Int25/exon26 splicing

18 c.3447_3449 GG > exon 26 (stop ex stop skipping ex 26 TT 26)

19 C.4099 O T exon 30 stop skipping ex 30

20 C.4405 O T exon 32 stop skipping ex 32

21 exon 33 (stop ex frameshift skipping ex 33 c.4565delT 33)

22 C.4600 O T exon 33 stop skipping ex 33

23* c.4780delTins37 exon 34 stop skipping ex 34

* mutation in patient 23 had been previously described

REFERENCES

Aartsma-Rus A, Janson AA, Kaman WE, Bremmer-Bout M, den Dunnen JT, Baas F, van Ommen GJ, van Deutekom JC. 2003. Therapeutic antisense-induced exon skipping in cultured muscle cells from six different DMD patients. Hum MoI Genet. 12:907-14.

Aartsma-Rus A, Janson AA, Kaman WE, Bremmer-Bout M, van Ommen GJ, den Dunnen JT, van Deutekom JC. 2004. Antisense-induced multiexon skipping for Duchenne muscular dystrophy makes more sense. Am J Hum Genet. 74:83-92.

Aartsma-Rus A, De Winter CL, Janson AA, Kaman WE, Van Ommen GJ, Den Dunnen JT, Van Deutekom JC. 2005. Functional analysis of 1 14 exon-intemal AONs for targeted DMD exon skipping: indication for steric hindrance of SR protein binding sites. Oligonucleotides. 15:284-97.

Aartsma-Rus A, Van Deutekom JC, Fokkema IF, Van Ommen GJ, Den Dunnen JT. 2006. Entries in the Leiden Duchenne muscular dystrophy mutation database: an overview of mutation types and paradoxical cases that confirm the reading-frame rule. Muscle Nerve.34: 135-44-

Bennett RR, den Dunnen J, O'Brien KF, Darras BT, Kunkel LM. 2001. Detection of mutations in the dystrophin gene via automated DHPLC screening and direct sequencing. BMC Genet 2:17.

Buzin CH, Feng J, Yan J, Scaringe W, Liu Q, den Dunnen J, Mendell JR, Sommer SS. 2005. Mutation rates in the dystrophin gene: a hotspot of mutation at a CpG dinucleotide. Hum Mutat 25:177-188.

Deburgrave N, Daoud F, Llense S, Barbot JC, Recan D, Peccate C, Burghes AH, Beroud C, Garcia L, Kaplan JC, Chelly J, Leturcq F. 2007. Protein- and mRNA-based phenotype-genotype correlations in DMD/BMD with point mutations and molecular basis for BMD with nonsense and frameshift mutations in the DMD gene. Hum Mutat. 28:183-95.

Eperon LP, Graham IR, Griffiths AD, Eperon IC. 1988. Effects of RNA secondary structure on alternative splicing of pre-mRNA: is folding limited to a region behind the transcribing RNA polymerase? Cell. 54(3):393-401.

Flanigan KM, von Niederhausern A, Dunn DM, Alder J, Mendell JR, Weiss RB. 2003. Rapid direct sequence analysis of the dystrophin gene. Am J Hum Genet 72:931-939.

Gualandi F, Rimessi P, Cardazzo B, TofFolatti L, Dunckley MG, Calzolari E, Patamello T, Muntoni F, Ferlini A. 2003. Genomic definition of a pure intronic dystrophin deletion responsible for an XLDC splicing mutation: in vitro mimicking and antisense modulation of the splicing abnormality. Gene. 311:25-33.

Havenga MJ, Lemckert AA, Ophorst OJ, van Meijer M, Germeraad WT, Grimbergen J, van Den Doel MA, Vogels R, van Deutekom J, Janson AA, de Bruijn JD, Uytdehaag F, Quax PH, Logtenberg T, Mehtali M, Bout A. 2002. Exploiting the natural diversity in adenovirus tropism for therapy and prevention of disease. J Virol. 76:4612-20.

Hoffman EP, Brown RH Jr, Kunkel LM. 1987. Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell 51 :919-28. Hofstra RM, Mulder IM, Vossen R, de Koning-Gans PA, Kraak M, Ginjaar IB, van der Hout AH, Bakker E, Buys CH, van Ommen GJ, van Essen AJ, den Dunnen JT. 2004. DGGE-based wholegene mutation scanning of the dystrophin gene in Duchenne and Becker muscular dystrophy patients. Hum Mutat 23:57-66.

Koenig M, Beggs AH, Moyer M, Scherpf S, Heindrich K, Bettecken T, Meng G, Mϋller CR, Lindlδf M, Kaariainen H, et al. 1989. The molecular basis for Duchenne versus Becker muscular dystrophy: correlation of severity with type of deletion. Am J Hum Genet. 45:498-506. Monaco AP. 1989. Dystrophin, the protein product of the Duchenne/Becker muscular dystrophy gene. Trends Biochem Sci. 14:412-5.

Reiss J, Rininsland F. 1994. An explanation for the constitutive exon 9 cassette splicing of the DMD gene. Hum MoI Genet. 3:295-8.

Roberts RG, Passos-Bueno MR, Bobrow M, Vainzof M, Zatz M. 1993. Point mutation in a Becker muscular dystrophy patient. Hum MoI Genet 2: 75-77.

Roberts RG, Gardner RJ, Bobrow M. 1994. Searching for the 1 in 2,400,000: a review of dystrophin gene point mutations. Hum Mutat 4:1-1 1.

Roest PA, van der Tuijn AC, Ginjaar HB, Hoeben RC, Hoger-Vorst FB, Bakker E, den Dunnen JT, van Ommen GJ. 1996. Application of in vitro Myo-differentiation of non-muscle cells to enhance gene expression and facilitate analysis of muscle proteins. Neuromuscul Disord. 6:195- 202.

Schwartz M, Dunø M, Palle AL, Krag T, Vissing J. 2007. Deletion of exon 16 of the dystrophin gene is not associated with disease. Hum Mutat. 28:205.

Tuffery-Giraud S, Saquet C, Chambert S, Echenne B, Marie Cuisset J, Rivier F, Cossee M, Philippe C, Monnier N, Bieth E, Recan D, Voelckel MA, Perelman S, Lambert JC, Malcolm S, Claustres M. 2004. The role of muscle biopsy in analysis of the dystrophin gene in Duchenne muscular dystrophy: experience of a national referral centre. Neuromuscul Disord 14:650-658.

van Deutekom JC, Janson AA, Ginjaar IB, Frankhuizen WS, Aartsma-Rus A, Bremmer-Bout M, den Dunnen JT, Koop K, van der Kooi AJ, Goemans NM, de Kimpe SJ, Ekhart PF, Ven- neker EH, Platenburg GJ, Verschuuren JJ, van Ommen GJ. 2007. Local dystrophin restoration with antisense oligonucleotide PRO051. N Engl J Med. 357:2677-86.

Welch EM, Barton ER, Zhuo J, Tomizawa Y, Friesen WJ, Trifillis P, Paushkin S, Patel M, Trotta CR, Hwang S, Wilde RG, Karp G, Takasugi J, Chen G, Jones S, Ren H, Moon YC, Cor- son D, Turpoff AA, Campbell JA, Conn MM, Khan A, Almstead NG, Hedrick J, Mollin A, Ri- sher N, Weetall M, Yeh S, Branstrom AA, Colacino JM, Babiak J, Ju WD, Hirawat S, Nor- thcutt VJ, Miller LL, Spatrick P, He F, Kawana M, Feng H, Jacobson A, Peltz SW, Sweeney HL. 2007. PTC 124 targets genetic disorders caused by nonsense mutations. Nature. 447:87-91.

White SJ, den Dunnen JT. 2006. Copy number variation in the genome; the human DMD gene as an example. Cytogenet Genome Res. 115:240-6.

Whittock NV, Roberts RG, Mathew CG, Abbs SJ. 1997. Dystrophin point mutation screening using a multiplexed protein truncation test. Genet Test 1 : 115-123.

Wilton SD, Fall AM, Harding PL, McClorey G, Coleman C, Fletcher S. 2007. Antisense oligonucleo- tide-induced exon skipping across the human dystrophin gene transcript MoI Ther. 15:1288-96.

Claims

1. The use of an antisense oligonucleotide directed against an exon sequence of the dystrophin gene for preparing a medicament capable of determining the skipping of said exon from the dystrophin transcript (mRNA), thereby promoting the production of a functional or partially functional dystrophin protein in a patient carrying a small mutation in said exon, characterised in that the exon is selected from the group consisting of exon 10, exon 16, exon 26, exon 33 and exon 34 of the dystrophin gene.

2. The use according to claim 1, wherein the partially functional dystrophin protein is the protein produced by a Becker muscular dystrophy (BMD) patient or a protein that is functionally equivalent thereto.

3. The use according to claim 1 or 2, wherein the medicament is effective for the therapeutic treatment of a patient carrying a small mutation capable of causing the Duchenne muscular dystrophy phenotype.

4. The use according to any of claims 1 to 3, wherein the patient is a human patient.

5. The use according to any of claims 1 to 4, wherein the antisense oligonucleotide is a 2'-O- methyl-phosphorothioate antisense oligonucleotide (2OMEPS AON)-

6. The use according to any of claims 1 to 5, wherein the nucleotide sequence of the antisense oligonucleotide is selected from the group consisting of SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4 and SEQ ID NO:5.

7. The use according to any of claims 1 to 5, wherein the nucleotide sequence of the antisense oligonucleotide is selected from the group consisting of SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9 and SEQ ID NO: 10.

8. An antisense oligonucleotide, characterised in that its nucleotide sequence is selected from the group consisting of SEQ BD NO: 1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9 and SEQ ID NO: 10.

9. The antisense oligonucleotide according to claim 8, as a medicament.

10. The antisense oligonucleotide according to claim 8, as a medicament for the therapeutic treatment of the Duchenne muscular dystrophy.