expertise in each, and plans to engineer into the yeast and other beasts the genetic mutations common to those that cause rare diseases in humans. That’s only possible now because of a huge step forward in genetic engineering known in shorthand as “CRISPR/Cas9,” which in a few short years has taken off in biomedical research (and might even turn into a new form of gene therapy in coming years).
Putting chemicals into cells or organisms to see what happens—a phenotypic approach—is how the pharmaceutical industry began decades ago. But in the digital and genomic age, the industry has moved toward “rational” design of therapeutic molecules that fit biological targets, like fine-grained locksmithing.
“The idea of phenotype screening is a little bit of a challenge to wrap my head around,” says consultant Tucker, who was a medicinal chemist with Upjohn in the 1990s. “I like having a target, and if at all possible, a couple of X-ray structures of ligands related to my lead bound to said target. But when we look at the world of approved drugs, a remarkably large fraction of our current pharmacopeia came out of the phenotype approach. I think a reasonable argument can be made that we’ve taken the reductionist approach too far for too long.”
Tucker is helping Perlstein sort through the off-the-shelf chemical libraries to screen for “hits”—in other words, to see which potential drugs have interesting effects in the yeast, fly, worm, and fish models. The more activity across the model organisms, the better the clue that the drug might work against the human version of the disease. “It would certainly give us more confidence to go forward,” says Perlstein.
Even so, there would be a ton of work to do. Compounds with promising activity in the four organisms will then be tested in human cells derived from patients with the specific mutations—what Perlstein calls a “sanity check.” For this, Perlstein will first turn to another recent technological development: a public bank of stem cell lines funded in part by California’s regenerative medicine agency.
Eventually, the drugs that show promise would also be tested in mice as a safety check, a more traditional (and expensive) drug development step that Perlstein won’t be able to avoid. At that point, however, he would ideally have corporate partners helping to pay for the work.
He first must prove that his model works, says Shawver. The drugs screened in his organisms must have an effect—“reverse a phenotype,” as she puts it.
Once that happens, Perlstein’s business idea is to give companies rights to all the mutations within a single gene. For example, Niemann-Pick disease is caused by a mutation of the NPC1 gene. But to date, 200 variations of that mutation have been cataloged, Perlstein says. He won’t be able to go after all of them, but he’d like to find a range of treatments for at least “a set of mutations that capture the majority of patients out there” for any given disease.
Ideally, those sets would entice drug companies to license his lab’s program for an entire disease. “We’re a preclinical company. If we don’t have co-development partners, these compounds won’t go anywhere,” Perlstein says.
There’s one other twist in the unorthodox story: Perlstein has registered his startup as a “B-Corp,” or public benefit company, which puts social responsibility at the core of its mission. (Outdoor gear maker Patagonia was one of the first.) The designation gives management a legal shield against lawsuits if, at any point, shareholders want to move the company away from its socially responsible remit. It’s an untested principle, but it gives Perlstein yet another unusual idea to work with as he tries to expand the range of treatments for rare diseases.
