Science Editor

Second only to cystic fibrosis, the commonest lethal recessive, autosomal genetic disease among Caucasians is spinal muscular atrophy. Its severest form takes the lives of babies halfway through their first year. Life ceases because their wasting muscles render breathing impossible. SMA happens only once in 10,000 births, and currently twice that number of Americans have some form of the malady.

"In just about every genetic disease for which there is a gene with multiple exons," observed molecular biologist Adrian Krainer at the Cold Spring Harbor Laboratory in New York, "at least some of the mutations cause exon skipping. That's the case in SMA patients."

Krainer is senior author of a research paper in Nature Structural Biology, released online Jan. 13, 2003, to be published in the journal's February issue. Its title: "Correction of disease-associated exon skipping by synthetic exon-specific activators."

"In this article," Krainer told BioWorld Today, "we report on the design of a new class of synthetic molecules to promote the inclusion of exon fragments that are skipped as a result of mutations in the gene. We designed molecules that force the inclusion of these exons. This is a new kind of synthetic molecule that is partly peptide and partly nucleic acid-like. It captures the essential properties of a natural splicing factor, and distills those properties, but condenses them into a molecule that's much smaller.

"One of the very interesting genetic diseases that we discuss in this paper," Krainer continued, "is spinal muscular atrophy. It's not so much a mutation in its SMN1 gene that causes SMA exon skipping, but the fact that the patients have a second gene almost like the normal version of the gene that's defective in this disease. But it doesn't splice correctly. So the hope is to be able to use these kinds of synthetic molecules to correct the splicing in that second gene, so it can really function therapeutically.

"There are probably other diseases in which some very common alleles occur with high frequency in patients with a particular disorder," he added. "Some of those alleles are also causing exon skipping. For example, cystic fibrosis, breast cancer, thalassemia, muscular dystrophy. Many types of mutations can alter those genes, but only a subset of them are causing exon skipping."

Molecular Patch Could Repair Crippled Gene

"For some of them, restoring exon inclusion might be sufficient to restore the normal protein. So it depends on the initial nucleotide chain that causes skipping. Some of those changes don't even alter the amino acid sequence. So if we brought the exon back it would make protein indistinguishable from that of the wild type.

"During splicing," Krainer explained, "the genes that are copied into RNA make up exons and introns. The natural process of splicing removes the introns, so the exons may join to one another. But occasionally, individual exons may be skipped by the splicing machinery. They're left out, so their message - used for translating protein - is missing a segment. That sometimes happens naturally. But in our case it's a mutation in a gene, which quite often causes inappropriate splicing, and one or more exons are left out at the final message.

"The natural proteins whose function we're emulating," Krainer recounted, "have an RNA-binding domain and a splicing-activation domain. So what we've done is use a minimal version of the splicing-activation domain. That's a piece of the protein's synthetic peptide but it's the same sequence. The targeting domain is different. Instead of using a protein motif to bind RNA, we made an antisense molecule, which binds RNA by Watson-Crick base pairing. We had that covalently attached to the peptide.

"At the moment," Krainer continued, "our experiments have been done in a cell-free system, using extracts from cells that carry out in vitro splicing. The next step will be to deliver these compounds into cells. That's what we're currently working on. There are modifications we can make to support spontaneous uptake, or we can administer them with liposomes or other carrier molecules to facilitate the uptake.

"We're proceeding by steps," Krainer went on, "so let's say we take spinal muscular atrophy as the disease model of interest. But no matter what we're going to do, getting into cells is the first hurdle. If we can deliver the compound into cells, then we can demonstrate that it's pretty much having the expected effect on splicing in vivo. We would use cultured human cell lines established from various patients. That's only a question of what splicing are we going to look at. We've done the first proof of principle in the test tube, and the next step would be in cell culture.

"But to be therapeutically useful," he pointed out, "obviously we have to go further. That would be, for example, making a mouse model. There is a mouse model for SMA that other labs have developed. In that particular disease our objective might be to have this novel compound go absolutely everywhere in the body, but where we really need it is in the spinal cord. So we have to find compounds that traverse the blood-brain barrier. That might be a significant hurdle, which we have to work on. At that point one can think of enrolling patients in clinical trials. It would take years to get there."

Spinal Disease Model Suggests Other Ailments

"In other diseases, the relevant target may be more accessible," Krainer observed. "After cell culture we have to do some experiments where the targets are more manageable - such as peripheral blood cells."

Krainer and his paper's lead author, Luca Cartegni, are first inventors on pending patent applications covering their synthetic molecules. "They claim the chimeric effector molecules and downstream applications," the Cold Spring Harbor's Technology Transfer Office states. Its spokesman said, "Companies have expressed interest in licensing the invention, and should contact us."

"Chimeric," Krainer explained, "refers to the fact that part of the synthetic molecule is a peptide, and part nucleic acid. We want to optimize these first-generation compounds. They're effective but if we play with different parameters we could probably make them more so. And then," he concluded, "a critical step will be to work on the in vivo delivery."