|Home | About | Journals | Submit | Contact Us | Français|
Duchenne muscular dystrophy (DMD) is a fatal muscle-wasting disease characterized by dystrophin deficiency from mutations in the dystrophin gene. Antisense oligonucleotide (AO)-mediated exon skipping targets restoration of the dystrophin reading frame to allow production of an internally deleted dystrophin protein with functional benefit for DMD patients who have out-of-frame deletions. After accelerated US approval of eteplirsen (Exondys 51), which targets dystrophin exon 51 for skipping, efforts are now focused on targeting other exons. For improved clinical benefits, this strategy requires more studies of the delivery method and modification of nucleic acids. We studied a nucleotide with a 2′-O,4′-C-ethylene-bridged nucleic acid (ENA), which shows high nuclease resistance and high affinity for complementary RNA strands. Here, we describe the process of developing a 2′-O-methyl RNA(2′-OMeRNA)/ENA chimera AO to induce dystrophin exon 45 skipping. One 18-mer 2′-OMeRNA/ENA chimera (AO85) had the most potent activity for inducing exon 45 skipping in cultured myotubes. AO85 was administered to mdx mice without significant side effects. AO85 transfection into cultured myotubes from 13 DMD patients induced exon 45 skipping in all samples at different levels and dystrophin expression in 11 patients. These results suggest the possible efficacy of AO-mediated exon skipping changes in individual patients and highlight the 2′-OMeRNA/ENA chimera AO as a potential fundamental treatment for DMD.
Duchenne muscular dystrophy (DMD) (OMIM 310200) is the most common inherited muscle disease in childhood, affecting more than one in every 3500 live-born boys. DMD is caused by mutations in the dystrophin gene and characterized by dystrophin deficiency in muscles. It is a progressive muscle-wasting disease and usually fatal in the third or fourth decades of life. Although general medical treatments and physical therapy have slowly increased life expectancy [1,2], establishment of a fundamental treatment that leads to dystrophin expression has long been the ultimate goal in DMD research.
To recover dystrophin expression in DMD, gene replacement therapy is a logically appropriate strategy, but more studies are needed to form the basis for clinical translation [3,4]. As an alternative, induction of exon skipping with antisense oligonucleotides (AOs) that convert out-of-frame dystrophin mRNA into in-frame sequences has been proposed . Exonic deletion mutations in the dystrophin gene cause not only DMD but also Becker muscular dystrophy (BMD), a milder progressive muscle-wasting disease. A translational reading frame rule explains the difference between DMD and BMD : Out-of-frame exon deletion mutations create premature stop codons that result in the more severe DMD phenotype because of a total lack of the dystrophin protein. In contrast, exon deletion mutations that maintain the original reading frame in the mRNA lead to the milder BMD phenotype because a mutated, but still functional, dystrophin protein can be expressed from the mRNA. Therefore, exon skipping allowing one exon to be omitted from the mutated dystrophin mRNA can be used to restore the reading frame and produce the internally deleted dystrophin . Compared with gene replacement therapy, the benefits of exon skipping with AOs include the ability to use the endogenous gene and straightforward chemical synthesis of AOs.
We demonstrated the first clinical study of exon skipping therapy using a phosphorothioate-oligonucleotide against dystrophin exon 19 and showed successful induction of exon skipping and dystrophin expression in skeletal muscle in a DMD patient (Patient 0) . However, the effectiveness was not enough to improve his motor function. This study has demonstrated a proof of concept as an AO-mediated exon skipping to treat DMD . Subsequently, in order to develop successful AO-mediated exon skipping therapy, various kinds of modified AOs have been chemically synthesized, with some of the most prominent being phosphorodiamidate morpholino oligonucleotides (PMOs), 2′-O-methyl (2′-OMe) AOs, locked nucleic acid (LNA), peptide nucleic acid (PNA), and tricycle-DNA (tcDNA) [10,11]. Furthermore, chimera AOs such as anhydrohexitol nucleic acid (HNA)/2′-OMe phosphorothioate (2′OMePS), cyclohexenyl nucleic acid (CeNA)/2′OMePS, altritol nucleic acid (ANA)/2′OMePS, and morpholino nucleic acid (MNA)/2′OMePS have been investigated [12,13]. Among these AOs, two AOs comprising different monomers—PMOs  and 2′-OMePS —were developed to induce skipping of dystrophin exon 51 and have undergone clinical trials. These trials yielded good results, enabling improvements in ambulation in DMD patients [16,17,18,19]. One PMO oligonucleotide, eteplirsen-mediated (Sarepta Therapeutics Inc., Cambridge, MA, USA) exon skipping therapy led to a slower rate of decline in ambulation compared to historical controls . After a long heated discussion , eteplirsen has received accelerated approval from the US Food and Drug Administration (FDA) . At the same time FDA stated that a clinical benefit of eteplirsen has not been established . There is controversy as to whether it is really beneficial [23,24,25].
Now, AO-mediated exon skipping therapy is reaching the final stage to provide a treatment for DMD patients. Efforts are focused on targeting other exons [4,26]. However, it is true that this therapy can stand further improvement in modification of nucleic acids, their delivery method, and evaluation methods . To improve clinical benefits, more studies are needed.
After establishing the proof of concept as AO-mediated exon skipping therapy, we have continuously developed AOs comprising a modified nucleotide with an ethylene bridge between oxygen at the 2′-position and carbon at the 4′-position of ribose (2′-O,4′-C-ethylene-bridged nucleic acid; ENA®, Daiichi Sankyo Co., Ltd., Tokyo, Japan), which is highly nuclease resistant and shows a high affinity for complementary RNA strands . Here, we describe the progress in developing the 2′-O-methyl RNA(2′-OMeRNA)/ENA chimera AO to induce dystrophin exon 45 skipping.
To improve the efficacy of nucleic acids, several modified nucleic acids have been created to provide higher nuclease resistance and high affinity for complementary sequences. ENA with an ethylene bridge between the oxygen and carbon of ribose (Figure 1) is thermodynamically stable and highly nuclease resistant [28,29] and has a high affinity for complementary RNA strands [30,31]. In fact, the ability of an AO consisting of the 2′-OMeRNA/ENA chimera to induce exon 19 skipping is more than 40 times stronger than that of the conventional phosphorothioate backbone oligonucleotides, making it a promising candidate for use as a low-toxicity, high-affinity oligonucleotide in the long-term treatment of DMD . Furthermore, a 2′-OMeRNA/ENA chimera against dystrophin exon 41 encoding a nonsense mutation has been shown to induce efficient skipping of the mutated exon 41 . Therefore, we decided to use 2′-OMeRNA/ENA chimera oligonucleotides for exon skipping.
Exon skipping therapy is mutation specific. To develop the broad therapeutic applicability of exon skipping therapy, we focused on induction of exon 45 skipping because it can be applied for treatment in approximately 9% of DMD patients [19,34]. The best AO to induce exon 45 skipping was identified by a trial-and-error procedure as described previously . 2′-OMeRNA/ENA chimera AOs consisting of 2′-O-methyl RNA with phosphodiester backbone, and ENA residues (at cytosines and thymines or at both the 5’ and 3’-ends) were employed. As the first screening for the identification of AOs against exon 45, five 15-mer AOs (AO32–36) to cover the splicing enhancer sequence of exon 45 were designed (Figure 2a) and examined for their ability to induce exon 45 skipping. Among five AOs, only one, AO33, induced exon 45 skipping (Figure 2b). To identify a more suitable AO, another set of 18-mer AOs (AO85–87) neighboring AO33 was designed (Figure 2a). Although all three AOs induced exon 45 skipping, only AO85 did so perfectly (Figure 2c). Therefore, AO85 was selected as the optimal AO for exon 45 skipping.
The half-maximum effective concentration of AO85 was determined in a cell-free splicing system and found to be 58 nM . To determine the stability of AO85, it was incubated with human and mouse blood. The mass peak corresponding specifically to the unchanged form of the full-length AO85 was monitored by LC-MS/MS system. The peak area ratio of AO85 to the internal standard (IS) remained stable for 2 and 8 h in human and mouse blood, respectively (Figure 3), demonstrating that AO85 remained as the full-length unchanged form over the time course of the incubation in human and mouse blood. Consistent with previous studies [28,29] on other ENA-AOs, AO85 was found to be stable in blood.
In our previous reports AO85 at 400 nM caused a 30% decrease in cell viability in MTT assay using fibroblasts . To examine the toxicity of AO85, we performed a single dose-escalation study using mdx mice, an animal model of DMD (Table 1). The mdx mice received intravenous administration of AO85 at doses of 0, 5, 15, and 50 mg/kg. They were observed for 14 days, and no abnormalities, including in body weight change, were found. Autopsy examination at Day 14 revealed no apparent abnormal pathological change in mdx mice treated with 0, 5, and 15 mg/kg. However, four of five mice treated with 50 mg/kg of AO85 showed renal enlargement, and two of five showed cysts on the renal surface. These findings suggested renal damage caused by high-dose AO85.
Next, we performed a repeated-administration toxicity study using mdx mice (Table 2). Intravenous AO85 injection at doses of 0, 0.5, and 5.0 mg/kg was repeated four times at one-week intervals in five mdx mice. Over four weeks, one mouse receiving 0 mg/kg died because of varicella; no other animals died. No abnormal finding was observed regarding the general condition, weight change, food intake, blood examination, and pathological findings. This result suggested that repeated administration of AO85 is safe.
To confirm the dose-dependent effect of AO85, it was transfected into cultured myotubes from a DMD patient with a deletion mutation of exon 46–51. RT-PCR amplification of mRNA in cultured myotubes treated at the concentration over 20 nM revealed a shorter product lacking exon 45, which was not observed in cultured myotubes at the concentrations lower than 10 nM (Figure 4a). At the concentrations over 100 nM, 100% of the dystrophin mRNA showed exon 45 skipping (Figure 4b).
Next, we analyzed dystrophin expression in cultured myotubes from the same patient with a deletion mutation of exons 46–51 (Figure 5). Before treatment with AO, immunohistochemical staining showed no dystrophin signal. However, at 7 days after transfection with AO85 (50 nM), immunohistochemical staining disclosed antibody-reacting materials in myotubes treated with AO85. These results revealed that AO85 transfection could induce exon 45 skipping and lead to dystrophin production in DMD myotubes.
We examined the ability of AO85 to induce exon 45 skipping and dystrophin expression in myotubes from 13 DMD patients carrying six different deletion mutations, such as deletions of exon 44, 46–47, 46–48, 46–49, 46–51, or 46–53. In each case, exon 45 skipping was induced by AO85, as expected; however, the skipping efficiency was different from patient to patient (Table 3). In myotubes from three patients with the exon 46–51 deletion, more than 80% of the dystrophin mRNA was exon skipping product while less than 43% of it was in myotubes carrying the exon 46–47 deletion. Dystrophin expression was seen in all but two patients. Unexpectedly, in myotubes with the exon 46–51 deletion, dystrophin expression was not apparent even though 82% showed exon 45 skipping in the mRNA. On the other hand, in myotubes with exon 46–47 deletion, dystrophin expression was apparent in spite of a low rate of exon 45 skipping in the mRNA.
Here, we have shown the process of developing the 2′-OMeRNA/ENA chimera AO against exon 45 and identified one suitable candidate, AO85. AO85 was effective in inducing exon 45 skipping and proved to be safe. These findings paved the way for clinical trials. In one clinical study, in fact, AO85 was shown safe and effective in increasing six-minute walk distance when it was injected intravenously at a dose of 0.5mg/kg/week . Currently, one 2′-OMeRNA/ENA chimera AO against dystrophin exon 45 (DS-5141b) is under clinical trial in Japan (JapicCTI No: 153072, http://www.clinicaltrials.jp/user/cte_main.jsp). For induction of exon 45 skipping, another AO consisting of PMO (SRP-4045, Sarepta Therapeutics) is now in clinical trial (https://clinicaltrials.gov).
Previous clinical trials of AO-mediated exon skipping have suggested a risk of renal impairment such as proteinuria . Although single high-dose AO85 administration induced slight renal change, repeated administration of AO85 at the appropriate clinical dose (0.5 mg/kg) revealed no toxicity. AO85 was considered not toxic to the kidney.
Theoretically, exon 45 skipping efficiency should be equal in all affordable patients. However, our current investigation showed the surprising result of differences in exon skipping efficiency among the patients. Even in same-age patients, efficiency varied, suggesting that age is not a determining factor in efficacy differences. We also found a difference in exon skipping efficacy between patients with the 46–51 deletion (82%–100%) and those with the 46–47 deletion (30%–43%). This difference suggested a relation between the efficiency of exon skipping and the breakpoint sequence of the deletion mutation , which requires further study for confirmation.
In-frame exon 45–46 deletions result in a severe DMD phenotype . Although these deletions are the outliers from the reading frame rule, the mechanism that produces the outlier is not known. Currently, the discrepancy between levels of exon skipping and dystrophin expression is not explainable. A limitation of this study was that quantitative evaluation was not performed, and more research is needed to clarify the mechanism that results in the discrepancy between exon skipping and dystrophin expression.
The cost of eteplirsen treatment was estimated $300,000–$400,000 in a year for one patient . To reduce the cost, it is important to develop more efficient AO by either improving drug delivery  or using other modified nucleotides. In our case we employed ENA. In the current scheme the dose is only 0.5 mg/kg , while 30–50 mg/kg in eteplirsen . Therefore, it is expected that the cost for treatment would be reduced using 2′-OMeRNA/ENA chimera. Efficient targeting of eteplirsen to cardiac muscle remains significant challenges . Our result suggests that the delivery to cardiac muscle of 2′-OMeRNA/ENA chimera AO should be evaluated.
One 2′-OMeRNA/ENA chimera AO that induces exon 45 skipping was identified and shown to be non-toxic and effective. Although a discrepancy in levels between exon skipping and dystrophin expression was observed among samples, we expect that AO85 can be applied for AO-mediated exon skipping therapy as a fundamental treatment for DMD.
This work was supported in part by Translational Research Grant from NEDO, Intramural Research Grant for Neurological and Psychiatric Disorders of NCNP, Grant-in-Aid for Exploratory Research and Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science, and Health and Labour Sciences Research Grant for Research on Psychiatric and Neurological Diseases and Mental Health.
M. Matsuo and Y. Takeshima are advisors for the Daiichi-Sankyo Co. Ltd., Tokyo. Other authors declare no conflict of interest.