Martin Burke says his synthesis machine could turbocharge chemists' ability to create natural products.

CHRIS BROWN

Billion-dollar project would synthesize hundreds of thousands of molecules in search of new medicines

Martin Burke is a tad envious. A chemist at the University of Illinois in Urbana, Burke has watched funding agencies back major research initiatives in other fields. Biologists pulled in billions of dollars to decipher the human genome, and physicists persuaded governments to fund the gargantuan Large Hadron Collider, which discovered the Higgs boson. Meanwhile chemists, divided among dozens of research areas, often wind up fighting for existing funds.

Burke wants to change that. At the American Chemical Society (ACS) meeting here earlier this month, he proposed that chemists rally around an initiative to synthesize most of the hundreds of thousands of known organic natural products: the diverse small molecules made by microbes, plants, and animals. "It would be a moon mission for our field," Burke says. The effort, which would harness an automated synthesis machine he and his colleagues developed to snap together molecules from a set of premade building blocks, could cost $1 billion and take 20 years, Burke estimates. But the idea captivates at least some in the field. "Assuming it's a robust technology, I would have to think it would be revolutionary," says John Reed, the global head of pharma research and early development at Roche in Basel, Switzerland. "Even if it only allowed you to make half the compounds, it strikes me as worthy."

Natural products have countless uses in modern society. They make up more than half of all medicines, as well as dyes, diagnostic probes, perfumes, sweeteners, lotions, and so on. "There's probably not a home on the planet that has not been impacted by natural products," Burke says.

But discovering, isolating, and testing new natural products is slow, painstaking work. Take bryostatins, a family of 20 natural products first isolated in 1976 from spongelike marine creatures called bryozoans. Bryostatins have shown potential for treating Alzheimer's disease and HIV, and demand has skyrocketed. Yet chemists must mash up 14 tons of bryozoans to produce just 18 grams of bryostatin-1. Synthesizing new bryostatins is equally hard, each one requiring dozens of chemical steps.

Burke thinks there is a better way. Two years ago, he and colleagues unveiled a machine that can link a variety of building blocks Lego-wise to create thousands of natural product compounds and their structural relatives. Now he says the approach can be scaled up. Molecular biologists have already automated the synthesis of short strands of DNA, proteins, and sugar chains, revolutionizing biomedicine. Burke argues that doing the same for natural products "could have major positive implications not just for chemistry, but for society."

Two years ago, Burke estimated that assembling 75% of natural products with his machine would take some 5000 different building blocks, compared with just four for DNA—a challenging number for chemical suppliers to make and stock. But now, Burke told the ACS meeting, the problem looks more manageable. His lab recently teamed up with that of Jeffrey Skolnick, a computational biologist at the Georgia Institute of Technology in Atlanta. They surveyed the literature on natural products, counted 282,487 compounds, and mapped all their structures. Skolnick's team then designed an algorithm to break each one into fragments, snapping only single bonds between carbon atoms—the kind of bonds Burke's machine can reassemble. Then the researchers asked the computer how many unique fragments it would take to reconstruct the library. It turned out that just 1400 building blocks would suffice to synthesize 75% of all natural product "chemical space," which includes related compounds not made by any organism. "This suggests it's a bounded, solvable problem," Burke says.

"It's a profound idea," says Mukund Chorghade, president of THINQ Pharma in Mumbai, India, who believes it would be a boon for drug discovery because it could provide untold numbers of lead compounds for developing new treatments. But not everyone is sold. "Marty is a visionary," says Larry Overman, a synthetic organic chemist at the University of California, Irvine. However, he says, natural product molecules are vastly more complex in structure than biopolymers such as DNA and proteins, and whether automated synthesizers could reproduce that complexity isn't clear.

An even harder problem could be securing funding for a large-scale effort to make the building blocks, assemble them into intermediate structures, and fold and tailor those to produce the final shapes. Bob Lees, head of the division at the National Institute of General Medical Sciences (NIGMS) in Bethesda, Maryland, that supports much of the organic chemical synthesis research in the United States, notes that in recent years NIGMS has devoted less funding to large-scale projects and more to research by single investigators. And last month, the Trump administration's initial budget request for 2018 proposed cutting the budget for the National Institutes of Health, the parent organization of NIGMS, by nearly 20%. Whatever its merits, Burke's cornucopian vision could face a steep uphill climb to become reality.