Science Editor

Though its excesses are certainly less flamboyant than those of the financial markets, science, too, has its bubbles - and its contrarians.

Antisense oligonucleotides - single-stranded DNA molecules that silence genes by targeting messenger RNA (mRNA) - were once biotech's next great hope. But first-generation antisense molecules proved disappointing in the clinic. And these days, despite the fact that second-generation technology has solved some of the early snags and clinicaltrials.gov lists nearly 70 clinical trials under the keyword "antisense," exuberance - whether irrational or not - seems to be reserved for RNA interference (RNAi) approaches.

In the realm of basic science, RNAi has been a hugely important discovery in its own right. As a laboratory method, it is easier, and more reliable, than antisense DNA for targeting mRNA.

But Alan Gewirtz, a professor of hematology and oncology at the University of Pennsylvania School of Medicine, thinks that in the therapeutic arena, it is too early to proclaim siRNA as a big advance over antisense DNA. In the Sept. 16, 2008 issue of the Proceedings of the National Academy of Sciences, he and his colleagues report surprising findings about the relative merits of siRNA and antisense oligonucleotides.

Like antisense oligonucleotides, short interfering RNAs or siRNAs also target mRNA, with double-stranded RNA molecules instead of single-stranded DNA. But "it takes far fewer tries to get an siRNA molecule that is workable" for silencing a gene than an antisense DNA molecule, Gewirtz told BioWorld Today. "And so, everybody is just rushing into this." But Gewirtz thinks the bloom will come off the siRNA just as surely as it came off of its antisense cousin: "As there is more and more experience with the RNA molecules, there are common issues to both."

In their paper, Gewirtz and his colleagues directly compared antisense oligonucleotides to siRNAs targeting the same structure: A 182 base pair mRNA for a part of firefly luciferase. By adding chemical changes at defined sites, the scientists also tested the effect of making the target more or less accessible to the inhibitors.

The conclusion, Gewirtz said, was that "in a cell-free system, the DNA molecules seem to be a lot better" at silencing its target. "Against a known structure, in the absence of the cell around it, they seemed to be a lot less finicky."

In addition to being less persnickety, antisense oligonucleotides were three times faster at getting rid of mRNA, at least initially, though Gewirtz said that ultimately, siRNAs catch up.

However, when the inhibitors were tested in cells, siRNAs were more efficient than antisense molecules at reducing gene expression; while antisense DNA reduced gene expression by 35 percent, siRNAs were able to reduce it by more than twice that, at 78 percent.

Gewirtz said that antisense's reduced performance in cells is due to two factors. "I personally think that there is the issue of structure, and then there is the issue of things that are attached to it - proteins and RNA," he said.

The siRNA, he added, "appears to have evolved as a mechanism for dealing with this problem in... live cells, at least more often than that poor old naked DNA molecule." Part of that solution may be the RISC complex - a complex of several proteins that help siRNA find and cut its target.

"If we can uncover the RNA" by removing the proteins associated with it, Gewirtz said, "or figure out a RISC equivalent for the DNA, our data suggest that it will perform better because its kinetics of target cleavage are superior."

Gewirtz did not contest the suggestion that such removal of attachments would prove to be a tall order in the clinic, but put it into perspective: "You can bet... that siRNA is going to be a tall order also," he said. "Once you start introducing anything into a cell, then you've got to start dealing with the biology."

And antisense DNA molecules, he and his colleagues note in their paper, have "certain theoretical advantages" with respect to some of the roadblocks to turning post-transcriptional gene silencing into a clinically useful strategy.

Asked what specifically those advantages are, Gewirtz responded "Ease of manufacturing, for one thing." DNA is more robust that RNA; "you can make DNA in your basement!" Gewirtz declared. And while bargain-basement DNA therapeutics might be taking things a bit far, it is true that DNA is, in general, is easier to synthesize and store than its RNA cousin.

Additionally, siRNAs are more complex than antisense oligonucleotides, which adds to their cost: "The more complex the molecule, and the more - forgive me - tarted up it is, the more expensive it is," Gewirtz said.

In the long run, Gewirtz is sanguine about the prospects of posttranscriptional gene silencing. "I'm an optimist," he said." I'm an oncologist - I believe that everything can be cured, if you have the right tools." But to get those tools, he added, it will be necessary to address the many technical problems that antisense oligonucleotides and siRNA have in common.

"One thing that has really eluded people is the ability to deliver" inhibitory oligonucleotides, Gewirtz said. "That's a major logjam - you're not going anywhere until you can assemble these and deliver them."

"First get it in, and the rest will follow."