MMD Research: Blocking Harmful Interactions

Matthew Disney is an associate professor in the Department of Chemistry at the Scripps Research Institute in Jupiter, Fla. He has MDA support to design small molecules that target the toxic RNA that underlies myotonic muscular dystrophy (MMD, also known as DM).

MDA Medical and Science Editor Margaret Wahl talked with Matthew Disney about his work.

Background note: RNA is derived from DNA and is the chemical step between DNA and protein synthesis. In MMD, the RNA is abnormally expanded.

Q: Your background is a bit different from that of some of the other researchers working on myotonic dystrophy. Are you coming at this challenge not so much as an expert in MMD, but as a biochemist with a special interest in RNA?

A: Yes. I have a background in biochemistry with a focus on the physical properties of RNA.

Myotonic dystrophy is thought to be a disease in which abnormal RNA folding plays an important role, and I'm particularly interested in RNA folding and misfolding. My training in RNA biophysics has had a great influence on our efforts in developing myotonic dystrophy therapies.

Q: How has your background as a biochemist and biophysicist influenced your approach to developing myotonic dystrophy therapies?

A: We've been studying what types of folds the toxic RNA in myotonic dystrohpy adopts, and we've been comparing these folds to those adopted by other cellular RNA molecules that we do not want to target.

By using this information, we've leveraged our ability to design and synthesize small molecules that can potentially bind to just the toxic RNA folds seen in myotonic dystrophy.

Q: So, you're attempting to target only the toxic RNA repeats and not other RNA repeats?

A: Right. It remains to be seen if you can do this and how well you can do this. The approach that we've put forward is that our designed compounds specifically target the expanded repeats of [the chemical sequence] CUG in myotonic dystrophy type 1 and the expanded repeats of [the chemical sequence] CCUG in myotonic dystrophy type 2.

We're targeting multiples of these repeats in a row, not a single copy of the repeat or just a few copies of the repeat. We think that by only targeting many repeats in a row, the compounds should have fewer off-target effects in cells and — although this has to be seen down the road — fewer side effects.

Q: You mentioned type 2 as well as type 1 myotonic dystrophy. Are you working on both?

A: It's interesting that you ask that. When we first got interested in myotonic dystrophy drug targets, we designed compounds for type 2. However, there weren't adequate cell and animal models of MMD2 in which to test these, so we switched to type 1. But, yes, we're optimistic about type 2.

Q: Your goal in targeting the repeats is to block their interaction with other molecules in the cell?

A: That is our goal.

Q: So you're not trying to destroy the repeats or the whole RNA strand, as some other strategies are attempting to do?

A: Right. The reason we don't want to destroy the repeats is that the RNA attached to the repeat may have some physiological function. If you destroy the CUG repeats in type 1 myotonic dystrophy, for example, you may affect the DMPK messenger RNA [genetic instructions for the DMPK protein] that the CUG repeats are attached to.

We've demonstrated that when we target the CUG repeats in the DMPK instructions with our small molecules, our designer compounds, these instructions are no longer stuck in the cell nucleus. They can get out of the nucleus and into the cytoplasm of the cell, where they have to go for the DMPK protein to be made.

Q: Why do you think the RNA can get out of the nucleus once you've targeted it with one of your small molecules?

A: It's because the small molecules are binding to the CUG repeat RNA and stopping large proteins from sticking to them. It turns out that the protein sticking is what keeps that DMPK messenger RNA stuck in the nucleus. If we can dislodge the large proteins, then the messenger RNA can leave the nucleus.

Q: How do your small molecules compare to other small molecules in development to treat myotonic dystrophy?

A: Our compounds have RNA-binding modules that are separated by what we're calling spacers. We think by making small molecule with spacers, they will go to and recognize RNAs that fold up into a long hairpin shape and not the ones that don't. [It's believed that the CUG repeats in MMD1 cause the RNA to fold into a shape that looks like a hairpin.]

I think it remains to be seen how one can target these expanded repeats specifically with a small molecule. But if I were to bet I'd put my money on a modularly designed compound like the one we've designed that targets the expanded repeats in MMD1, as opposed to a traditional small molecule that might recognize just one repeat.

Q: What kinds of chemicals are you using to target CUG or CCUG repeats?

A: We've made derivatives of things like antibiotics, and we've made some other compounds that are similar to some known drugs. The bis-benzimidazoles are one class we're working on.

Q: Are those the ones you're calling H molecules?

A: Yes.

Q: What have you seen in the laboratory with these molecules?

A: In cells that have an MMD1-type defect, we've seen that the designed compounds are capable of correcting three key features of the disease: the binding of cellular proteins to the CUG repeats; the retention of DMPK RNA in the cell nucleus; and the creation of potentially cell-damaging clumps of RNA and proteins in the nucleus.

All three of those defects were corrected in cell models of MMD1. The cellular proteins were released from the CUG repeats, allowing them to do their normal jobs; the DMPK messenger RNA was able to leave the cell nucleus and be used for synthesis of DMPK protein; and there were fewer RNA-protein clumps.

Q: Have you done any experiments with these compounds in mice?

A: Yes. The animal studies were done in collaboration with Charles Thornton. We wouldn't have been able to do them without him.

Charles' lab injected our compounds into their MMD1 mice. These mice have expanded CUG repeats in the actin gene, not the DMPK gene.

We saw correction of the MMD1-associated defects in chloride ion channels and calcium ion channels, which are required for controlling muscle contraction and relaxation. We think that's because MBNL1 [muscleblind-like protein 1] and possibly other proteins, which are needed for correct synthesis of these channels, were freed from the CUG repeats to do their usual jobs.

So far, that's all we've looked for, and that's all we've seen, but we're extending those studies.

We're also going to see if oral dosing of these compounds in the mice is effective.

Q: Do you think you'll reach the heart with these compounds?

A: Yes.

Q: And the brain?

A: I'm not so sure. That remains to be seen. Here at Scripps, we have a distribution, metabolism and pharmacokinetics facility, where we can test compounds to see their distribution very quickly and potentially optimize them if need be.

These drug discovery facilities are a major advantage that we have here at Scripps. They set us apart from more traditional academic institutions and make us uniquely positioned to potentially advance compounds from the lab to the patient. 

Q: How does your strategy compare to, say, some of the antisense oligonucleotide strategies being developed for myotonic dystrophy?

A: The antisense strategies are much further along in development than the small molecules, and some have been shown to reduce myotonia [the prolonged contraction of muscles associated with myotonic dystrophy].

But the advantage of small molecules is that they're tried-and-true therapeutics. Most FDA-approved drugs are small molecules. There are only maybe one or two antisense oliogonucleotides that have been approved by the FDA [U.S. Food and Drug Administration].

Small molecules are easier to manufacture than antisense oligonucleotides, and they may get into targeted tissues and cells more easily.

Also, they can often be delivered orally. Aspirin, for instance, is a small molecule. It would be hard to envision an antisense molecule being orally available. It would be much easier to envision a small molecule being orally available.

And finally, antisense compounds are pretty costly to make. The thought is that a small molecule will be easier to make and easier to produce in large quantities, and therefore cheaper.

Q: How big is an antisense oligonucleotide compared to a small molecule of the type you're developing?

A: Antisense oligonucleotides are 10 to 50 times the size of a small molecule like ours.