Abstract
DNA-based ASOs can form triple helixes on favorable genomic targets and interfere with transcription. However, they are mostly used to form heteroduplexes with a targeted RNA, which then becomes a substrate for degradation by RNase H. Other pathways for RNA degradation are triggered by RNA-based ASOs and involve RNase P or, in certain cases, Argonaute and the RNA-induced silencing complex (RISC). Steric-blocking ASOs prevent the interaction of factors responsible for mRNA processing, transport, or translation. For instance, splice-switching oligonucleotides (SSO) interfere with the proper formation of the spliceosome and modulate the inclusion of targeted exons in the mRNA.
Multiple chemistries for ASO synthesis are available, with a variety of modifications introduced into the basic DNA or RNA structure. Sugar backbones containing phosphorothioate links and 2′O-methyl or 2′O-methoxyl groups are resistant to nuclease degradation. Other modifications like those found in locked nucleic acid (LNA) or phosphorodiamidate morpholino oligomers (PMO) further enhance nuclease resistance and significantly increase duplex stability. Although they can be very advantageous for the pharmacokinetic properties of the ASO, these modifications may also be inhibitory to their activity, requiring the more sophisticated assembly of ASO with stretches of unmodified nucleotides (gapmers).
Today, ASOs represent an active area of innovative drug development with a pipeline of about 15 products in mid- to late-stage clinical trials for pathologies ranging from cancers and rheumatoid arthritis to Duchenne muscular dystrophy (DMD). The Food and Drug Administration first approved an ASO (Fomivirsen) in 1998 for the treatment of cytomegalovirus-induced retinitis in AIDS patients. This product is now discontinued, but a second ASO, mipomersen, was approved in January 2013 for an orphan indication. It is designed to target the apolipoprotein B-100 mRNA, and it reduces the levels of atherogenic lipoproteins in patients with familial hypercholesterolemia. Three reviews in this issue of Human Gene Therapy discuss the recent developments of ASOs for the treatment of inherited neuromuscular diseases, a complex group of highly debilitating rare conditions for which no treatment is available. Three of these diseases are the focus of these reviews, each of them exemplifying a different molecular strategy by which ASOs can restore functional levels of a missing protein or counteract the toxic effect of a mutant mRNA.
Koo and Wood explain how Duchenne muscular dystrophy, an X-linked disease due to mutations in the gene encoding dystrophin, may be treated with SSOs. Dystrophin is a modular protein with a large domain composed of spectrin-like repeats. In most patients, large deletions of the DMD gene remove one or several of the 79 exons and result in out-of-frame mRNAs. SSOs are used to induce the skipping of additional exons on the DMD pre-mRNA in order to reestablish a translational reading frame and the production of internally deleted but functional dystrophins. The approach is now being tested in Phase III trials.
Porensky and Burghes describe a different situation with spinal muscular atrophy (SMA), which is caused by mutations in the SMN1 gene and insufficient levels of SMN, a ubiquitous protein involved in the assembly of small nuclear ribonucleoproteins (snRNPs). A second copy of the gene—SMN2—exists but is not sufficiently expressed to compensate for the loss of SMN1. In fact, most SMN2 transcripts are defective because they exclude exon 7 due to the presence of a strong splicing silencer. Masking this silencer with steric block ASOs results in exon 7 inclusion, restores normal levels of SMN, and counteracts the deleterious effects of the SMN1 inactivation. Following spectacular preclinical results in animal models, ASO-mediated exon inclusion is now being tested in early clinical trials.
Possible treatments of myotonic dystrophy type I (DM1) based on ASOs are presented by Gao and Cooper. The disease is caused by an expansion of CUG triplets in the 3′ untranslated region of the DMPK gene. The highly structured CUG repeat directly or indirectly depletes the cell from key factors regulating alternative splicing. Hence, the presence of the expanded mRNA results in a toxic gain of function, with multisystemic pathological consequences. The administration of steric-blocking ASOs against the CUG repeat to a murine model of DM1, or to patient-derived myoblasts, is able to release the sequestered muscleblind-like (MBNL1) proteins and to correct myotonia. Alternatively, the injection of an RNAse H-active ASO to DM1 mice results in the degradation of the expanded mRNA, which, surprisingly, is still detectable after 1 year. These preclinical proofs of concept should lead to clinical trials in the near future.
Antisense approaches have already changed the game in the development of therapies for the three diseases reviewed here. Discussions around therapeutic strategies have now moved from how to be palliative to how to define the best conditions for disease stabilization and even symptom reversal. This is a great achievement for diseases that many considered hopeless 15 years ago. These complex diseases primarily affect skeletal muscles, heart, diaphragm, nerves, and brain, and the outstanding difficulty is to reach enough targets using reasonable doses of ASO. Amazingly, this has been shown to be possible in preclinical studies during which the appropriate combinations of ASO chemistry, carriers, and administration routes were defined.
Clinical development is most advanced in Duchenne muscular dystrophy, with three active clinical trials involving hundreds of patients—a rarely seen situation in genetic therapies for an inherited disorder. Even though rescuing dystrophin at functional levels in the skeletal muscle, diaphragm, heart, and brain is a very tall order, the recent clinical data with ASO indicate that it may not be so farfetched. Remarkably, ASO activity allowing dystophin rescue is now documented in patients receiving systemic injections. A few cases of toxicities associated with certain chemistries and/or sequences have been documented in preclinical models, but none have been seen in clinical trials so far.
The two other diseases represent an even more daunting challenge, because they are caused by deficiencies in housekeeping proteins involved in the complex tissue-specific regulation of alternative splicing. The primary manifestations are in the neuromuscular system, but they probably represent only the tip of the iceberg. Animal studies indicate that once they are treated, other organ defects often surface. Efficient treatments of these diseases are likely to require a widespread distribution of ASOs to multiple organs and a better understanding of the effect of the primary mutation on RNA splicing.
A significant part of the research and early development discussed in the reviews has been supported by patients' organizations. With the interest and support of large pharmaceutical companies, the global clinical development of ASOs is at an inflexion point where clinical trials are multiplying and marketing authorizations are starting to be granted. Whether any of these novel drugs will reach blockbuster status is a matter of speculation, but their coming of age in the three neuromuscular diseases discussed in these reviews is a vibrant illustration of how desperate situations can drive great science and innovation.