Accelerating Lead Compound Optimization

Scientists at The Scripps Research Institute (TSRI) in La Jolla, CA, and Pfizer's La Jolla Laboratories have devised a new, more efficient way for researchers in drug discovery and development to optimize their lead compounds.

Adding small, strained-ring structures to the overall design of lead compounds is a strategy increasingly favored by medicinal chemists. “There are tons of reasons to do so—to make drugs more absorbable, for instance, or better able to resist metabolic clearance,” says the study's senior author Phil Baran, PhD, a professor of chemistry at TSRI. Strained-ring structures improve a drug's properties because they're “tight balls of carbon atoms,” explains Michael Collins, a senior principal scientist at Pfizer, “which are much more metabolically stable. Straight carbon chains would be readily oxidized at multiple sites.”

However, the process of synthesizing these strained-ring structures is often tedious and finicky. Pfizer's preclinical investigations of a promising cancer drug candidate had to be abandoned when the project scientists realized that obtaining sufficient quantities of bicyclo[1.1.1]pentan-1-amine, a structure necessary for the drug's development, would be impossible. The traditional method of synthesis—dating back to 1970—required as many as five steps, involved toxic and corrosive reagents, and yielded only tens of milligrams. To evaluate its potential drug, Pfizer would need kilograms of bicyclo[1.1.1]pentan-1-amine.

Baran and Collins pooled the expertise of their respective laboratory groups and, in just a few months, figured out a solution to this 45-year-old problem. Bicyclo[1.1.1]pentan-1-amine is made by joining propellane—a five-carbon “tight ball” shaped like a propeller—to dibenzylamine, one of medicinal chemistry's basic building blocks. In devising their new synthetic route, the team exploited the innate reactivity of propellane's C–C bonds, which are under considerable strain from being arranged in rings at odd angles.

“Propellane is this spring-loaded structure that's pretty unhappy alone; it wants to relieve the strain by reacting with an amine,” Baran explains. “We found a way to rapidly spark bicyclo[1.1.1]pentan-1-amine's synthesis, by getting propellane's most strained C–C bond to interact directly with dibenzylamine.” The team's strategy, called strain-release amination, was further accelerated when they came up with a recipe for a stock solution of propellane, eliminating the need to isolate this highly volatile carbon structure each time.

Pfizer researchers have since ramped up bicyclo[1.1.1]pentan-1-amine production, using strain-release amination, and resurrected preclinical studies of their cancer drug candidate. Meanwhile, Baran and Collins are gratified that their new method is broadly applicable—it can be used to append a wide range of strained-ring structures, such as azetidine and cyclobutane, to amine-containing compounds. In experimenting with cyclobutane, the team has also observed that its carbon atoms can form covalent bonds, swiftly and selectively, with the amino acid cysteine. As such, cyclobutane and its derivatives could facilitate more robust protein labeling techniques, and drugs that home in more accurately on their targets.

“One cool thing about this collaboration is that we started out with a very focused problem, and evolved fresh ideas to pursue along the way,” Collins says. “It's a trademark of good research.”

To Baran, strain-release amination is an example of “rapid bench-to-bedside chemistry that will have a positive impact on medicine, because it allows chemists access to drug-optimizing structures that traditionally have been difficult to synthesize.”

“The power of this method is that it's disease-agnostic,” he adds. “It will help people working in multiple fields, including neuroscience, cardiovascular health, and, of course, oncology.” –Alissa Poh