The same thing – gaining or losing information – can happen to a gene, because like sentences built of words, most genes come in pieces. For decades researchers have been trying to co-opt nature’s way of copying only some of a gene’s information into messenger RNA (mRNA) to bypass harmful mutations as if they are typos. The strategy, called exon skipping, is finally nearing the clinic.
GENES IN PIECES
Understanding exon skipping requires a short detour into the biology.
I was in grad school in February 1978 when Harvard’s Walter Gilbert published a short News & Views piece in Nature – “Why Genes in Pieces?”, that changed what everyone thought they knew. A gene’s information isn’t continuous, but is dispersed in exons alternating with seemingly meaningless introns.
Before an mRNA is translated into protein, enzymes snip out its introns at splice sites. If a mutation alters a splice site, results can be catastrophic. The mRNA can retain an intron or lose an exon, like a sentence with an extra or missing word, and the protein product is too long or too short. Some genes are normally spliced in different patterns, like creating different sentences from one long one, leading to versions of a protein adapted to different circumstances.
In the budding biotechnology of exon skipping, splice sites are chemically shielded in ways that enable an mRNA to form while ignoring a mutation, or altering how the RNA or protein folds. To do that, an “antisense” molecule binds to a specific sequence in the mRNA like a Velcro patch to cloth. The binding is based on the complementary base pairing (A with T or U, G with C) of nucleic acids. If a stretch of RNA is GCUA, the corresponding antisense sequence would mimic CGAU.
ALMOST READY FOR PRIME TIME
Antisense silencing of mRNAs isn’t new, and began in agriculture. I wrote “Building a Better Tomato” for High Technology magazine in 1985, and “Making Sense of Antisense” for BioScience in 1989. But this month’s special issue of Nucleic Acid Therapeutics, on exon skipping, indicates that the field has gone far beyond slow-ripening tomatoes.
A Guest Editorial by Annemieke Aartsma-Rus. PhD, in the Department of Human Genetics, Leiden University Medical Center, introduces the European Cooperation in Science and Technology (COST), with an action mandate to “overcome challenges through networking to allow clinical implementation of antisense-mediated exon skipping for as many rare disease patients as possible.” Just in time for Rare Disease Day next week.
COST will sponsor private workshops to discuss preclinical results, try to lower regulatory hurdles for testing treatments for extremely rare variants of rare diseases, standardize testing of the proteins that indicate whether exon skipping is working, and identify biomarkers from accessible body fluids to replace painful muscle biopsies and limited brain imaging.
The list of disease candidates for the technology is growing. Duchenne muscular dystrophy (DMD) is farthest along, with a phase 3 clinical trial just completed. Leiden University’s Dwi Kemaladewi shows how it works in this video, a winner of Science magazine’s “Dance Your PhD” contest from 2011. Other targets are a rare form of Alzheimer’s disease, spinal muscular atrophy, Leber congenital amaurosis (blindness) due to CEP290 mutation, and fibrodysplasia ossificans progressiva, which turns muscle to bone.
Three diseases represent the past, present, and future of exon-skipping: familial dysautonomia, DMD, and Huntington disease.
I first heard about exon skipping in the context of the very rare familial dysautonomia (FD), where it occurs naturally. FD is one of the “Jewish” genetic diseases – my friend’s daughter has it. The single-base mutation is at a splice site between the 20th intron and exon pair of a gene called IKBKAP.
FD affects neurons that control such autonomic functions as breathing, digesting, regulating temperature, sensory perception, and making tears. Some symptoms ebb and flow with crises of shaking cold and pervasive nausea and dizziness, while others, such as the need for tube feeding, are a constant for some people. A character in a 2010 novel by Lionel Shriver, So Much For That, shows what life can be like with FD.
But FD has a peculiarity that inspired the idea to exploit natural gene splicing– some cells ignore the mutation. They manufacture a normal protein, possible because the amino acid sequence information is still there, it’s just the splice site that the mutation alters. It’s a little like a lit indicator on a car’s dashboard signaling a problem that isn’t really there.
If cells that ignore the mutation and make IKBKAP protein happen to be in the parts of the brain that the disease affects, the child is lucky and may have a mild case – like my friend’s daughter, who tells her story in my human genetics textbook.
For FD patients not so lucky, cells from the brain and spinal cord skip the exon, while muscle, lung, liver, white blood cells, and glands – parts unaffected in the illness — produce normal-length proteins.
In 2003, discovery of the FD mutation and the tendency of some cells to ignore it inspired Drs. Berish Rubin, Sylvia Anderson, and their colleagues at Fordham University to screen compounds that might modulate levels of the protein, perhaps by affecting splicing. They found two –- a form of vitamin E (tocotrienol) and a green tea component (epigallocatechin gallate), and another team last year added phosphatidylserine. (I’m not a doc, so for info on these supplements, see FD Now.)
DMD – LEADER OF THE PACK
In FD, a skipped exon in brain cells causes the symptoms. In DMD, inducing exon skipping helps. Dutch biotech company Prosensa, with GlaxoSmithKline, has just completed a phase 3 clinical trial of drisapersen and Sarepta Therapeutics is nearly as far along in testing its candidate, eteplirsen The two approaches use different chemistries to create their “antisense oligonucleotides” (AONs), which bind specific splice sites.
The clinical trials have had recent ups and downs, mostly because tests of efficacy must go on for a long time to demonstrate that an intervention is actually having an effect, and that apparent progress isn’t just due to natural fluctuation of symptoms. A longer-term goal is to figure out a way to deliver the treatment to enough muscles to improve mobility and quality of life.
Despite these hurdles, the therapy itself is sheer genius.
In some boys with DMD, a mutation in exon 50 of this gargantuan 79-exon gene introduces a nonsense mutation, which is like a period appearing in the middle of a sentence, truncating it. The encoded protein, dystrophin, is too short to do its job of supporting a muscle cell membrane during a contraction.
But in the much milder Becker muscular dystrophy, a less disruptive mutation yields a slightly shortened but partially active dystrophin. Lifespan is normal. This means that some restoration of dystrophin function can, theoretically, perhaps help a boy with DMD, who faces a very limited future.
Indeed, when AONs bind near the nonsense mutation that causes DMD, the “spliceosome” that does the cutting dances right past the glitch and “restores the reading frame” (the sequence of base triplets that encode the amino acids). Some dystrophin is made. The effect is a little like ignoring a series of miskeys in a typed sentence.
But does the fix at the molecular level improve the performance of boys with DMD in a 6-minute test of walking ability? How far a boy can walk in this time may seem an arbitrary sort of measure, but it’s based on years of research into understanding the “natural history” of the disease. By identifying exactly which functions ebb and at what rate, the natural history provides the benchmarks against which a new treatment is assessed.
For DMD, the natural history is a sad series of losses. Craig McDonald, MD, a professor of pediatrics at the UC Davis School of Medicine, described it in a news conference Prosensa held in late 2013. “The time to stand predicts time to loss of standing, and loss of standing predicts the age at loss of ambulation. Age at loss ambulation predicts age at loss of self-feeding and need for assistance in ventilation. Inability to jump, hop, run, standing from the floor or form a chair, stair climbing, even the ability to stand in place, reach overhead, or put hands on a tabletop, are meaningful ambulatory milestones.”
The goal of the drugs that manipulate gene splicing is to slow this relentless course. And that’s why the observation that boys under age 7 given drisapersen for 96 weeks walked on average 49 meters more than those on placebo was good news. But the FDA sent the other drug back for further testing, because the 6-minute walk test is so variable that more time is required to see if the improvement persists. So stay tuned – both drug candidates seem well on the road to approval.
If FD provides exon skipping’s basis and DMD becomes first to hit the clinic, then maybe HD represents future success. A recent DNA Science post described the tragedy of HD in one young family.
HD is a classic “expanding triplet repeat” mutation — affected individuals have extra copies of the sequence CAG in exon 1 of the gene that encodes the protein huntingtin (Htt). The extra stuff tacked onto one end of the gene distorts how its mRNA folds, affecting splicing in a way that generates “toxic fragments” of protein.
Melvin Evers and co-workers, also from Leiden University, report in February’s Nucleic Acid Therapeutics how they deduced that skipping exon 12 (out of a total 67) of the HD gene might prevent release of the toxic fragments. So far the approach has worked in HD patients’ cells and in a mouse model of the disease. Because HD is a disorder of extra genetic material and currently has no treatment, gene silencing seems a logical approach.
The story of the harnessing of exon skipping provides a great example of the evolution of science to technology. From 1978’s “Why Genes in Pieces?” to little boys with muscular dystrophy gaining steps has taken 36 years. In the interim lay much hypothesizing about why most genes are patchworks. Now that we know that what was once thought to be exception is actually the norm, we can perhaps put mosaic genes in the context of being another manifestation of genome versatility.
Insight and application take time.