Blame the blood-brain barrier for the difficulties of neurological drug discovery.

"You have this barrier that protects the brain," D. Martin Watterson told BioWorld Today. And the chemical properties that allow compounds to penetrate the blood-brain barrier "overlap the properties that make compounds go through the liver in first-pass metabolism."

In other words, the better a compound is at getting to the brain, the better the body is at getting rid of it, leaving the intrepid drug discoverer where he or she started. The net result is that neuroscience drug discovery has an even higher failure rate than drug discovery for other indications.

But Watterson and his colleagues at Northwestern University's Center for Drug Discovery and Chemical Biology in Chicago have a new compound in hand that might be able to beat the odds. In the Jan. 11, 2006, issue of Journal of Neuroscience, they reported on synthesizing a molecule that was able to selectively inhibit brain inflammation caused by activated glia, reversing both molecular and behavioral signs of Alzheimer's disease in mice.

The molecule was identified by an approach named "de novo compound discovery." Rather than screening large numbers of compounds for activity and then attempting to make them pharmacologically acceptable, de novo compound discovery "starts with a chemotype that actually has no activity, but that has properties that make it amenable to being drug-like," Watterson explained. "Then you diversify that compound and screen until you have activity."

Using that approach, the scientists found 188WH "in under two years, with a small amount of people and a shoestring budget," Watterson said. "I like it because most people said it couldn't be done."

He also attributed the success of the approach to the fact that it is "a novel chemical platform integrated with an old-fashioned biological screen," which assures early on that the biological activity of a potential drug is relevant to the disease being targeted.

In cell culture, the scientists found that 188WH could block inflammatory cytokine production, and was metabolized by the liver at a pace that would allow it to accumulate in the body at a level that was likely to be effective with minimal toxicity.

In a series of animal experiments, the team first determined that 188WH was able to cross the blood-brain barrier with comparative ease, and then tested 188WH's effects on inflammation, synaptic dysfunction in the hippocampus and behavior.

When mice were treated with 188WH after infusion of pathological amyloid-beta peptide, the drug suppressed the increase of several glial-produced inflammatory cytokines activated by amyloid-beta, at doses that did not affect peripheral inflammatory responses. 188WH had no effect on the formation of amyloid plaques - clumps of amyloid-beta peptide that accumulate in Alzheimer's patients - suggesting the drug acts specifically by suppressing brain inflammation.

In the hippocampus, a key brain structure for memory, it also prevented the loss of two proteins that are decreased in Alzheimer's patients. Finally, when the animals were tested in a maze, 188WH reversed the memory deficits caused by amyloid-beta infusions.

Watterson noted that "we can do [behavioral testing] in the same animal as doing the biochemical endpoints, which is nice." In general, both the biochemical and the behavioral parameters could be used in the clinic, as well, increasing the odds of success.

Why - or how - the compound acts specifically on glia is not clear. Watterson said that based on previous work, it is reasonable to assume that 188WH affects gene expression via modifying protein phosphorylation, but there is no data yet to support that idea. The compound may well act on several pathways.

"Most good drugs are not single-molecular target drugs," Watterson said, because most diseases are not caused by a single dysfunction. "That doesn't make a simple story for the public and the press, but that's reality."

188WH is exclusively licensed to Toronto-based firm NeuroMedix Inc., which began trading Monday on the TSX Venture Exchange, and plans to enter the clinic with an analog of 188WH in the third quarter.