Nature Isn’t Simple (A Bitter Pill-Part 4)
Posted by Richard Conniff on April 9, 2012
A plant sample—even something as basic as coffee or tea–may contain thousands of compounds. In the lab, chemists “fractionate” plant samples, breaking them down into crude extracts. But they still ended up with hundreds of compounds in each of their 1536 test wells. “This is where the wheels fell off of this thing,” says Paul Armond, a plant cell biologist who spent 30 years in drug development at Pfizer. “High-throughput screeners hated these samples. They didn’t want to have anything to do with them because even if you got a hit in one of these fractionated samples, you didn’t know which of the hundred compounds in the test well was the active one.” A high-throughput screener’s job is to test as many compounds and get as many hits as possible, and natural compounds just seemed to clog the pipeline.
Even if they managed to isolate an active compound from a plant, says Armond, “it would be, from the organic chemists’ point of view, some ugly compound, this big, giant molecule that no chemist could ever possibly synthesize. They‘d said, ‘What am I supposed to do with this?’” When a compound seems promising, the usual next step is to “add things to it, take things away, rearrange things, and find where the important parts of the molecule are and where the not-so-important parts are.” Through the magic of combinatorial chemistry, researchers can target the molecule more carefully, or weed out unwanted side effects. But if a compound from a natural product is too complex to synthesize in the first place, “then you can’t do any of those things.”
One final obstacle made natural products problematic: Getting enough of the desired compound can be difficult because living things normally vary by season or site—and sometimes disappear completely. It happened to another NCI research team in the late 1980s. When they got a promising hit for an anti-HIV compound from a tree in Sarawak, Malaysia, researchers hurried back to collect more samples. But someone had cut down the only known tree. After a frantic search, the only other evidence of the species they could find was a 100-year-old specimen in the Singapore Botanical Garden. Chemists eventually figured out how to synthesize the compound, and Calanolide A is now an experimental treatment for HIV patients.
By contrast, products of combinatorial chemistry are wonderfully simple. That may, however, be their only advantage: The combinatorial compound libraries researchers worked with in the early years were so badly flawed, according to Christopher Lipinski, a drug development guru who spent most of his career at Pfizer, that the industry would have been more productive if it had “stored them in giant dumpsters.” Even now, after tens of billions of dollars and 25 years of research, combinatorial chemistry and high-throughput screening have put only a single completely new FDA-approved compound into the marketplace. The new methodologies can thus seem a bit like the drunk who searches for his keys under a lamppost, not because that’s where he dropped them, but because the light is better there.
(to be continued)