A drug discovery bottleneck? How cheaper reagents could speed branched molecule synthesis

Stephanie Baum

scientific editor

Meet our editorial team
Behind our editorial process

Andrew Zinin

lead editor

Meet our editorial team
Behind our editorial process
Editors' notes

This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

peer-reviewed publication

trusted source

proofread

The GIST
Add as preferred source

When chemists design drug candidates, shape matters enormously. Many active pharmaceutical ingredients contain branched carbon structures—points where the molecular chain forks in a specific direction—that are critical to whether a molecule will bind to its biological target and whether it will be safe. The challenge is that the branched building blocks used to create these structures are not very abundant or commercially available. Now, scientists at Scripps Research have devised a new approach to building these branched molecular structures found in many medicines and materials: one that could make the early stages of drug discovery faster and more efficient.

The method, published in Science, overcomes a stubborn technical obstacle that has limited chemists' ability to assemble complex molecules from simple, inexpensive starting materials.

"This work solves a selectivity problem that challenged us for years," says Ryan Shenvi, professor at Scripps Research and senior author of the study. "We've now laid the groundwork to access iteratively branching materials that occur in metabolites, fragrances and drugs."

The branched building blocks needed to assemble branched molecular frameworks are far less common than simpler, straight-chain starting materials called alkenes: compounds built around a pair of carbon atoms connected by a double bond, which makes them reactive and easy to modify. Chemists working with more complex branched structures have typically had to create branched building blocks from commercially available alkenes in one to three extra preparation steps before the real synthesis work can begin.

One promising shortcut has been a chemical process called metal hydride hydrogen atom transfer (MHAT). This process often uses cobalt as a catalyst (a substance that speeds up a chemical reaction without being consumed) to guide alkene reactions directly toward these branched shapes, skipping the need for branched starting materials.

But the chemists added a twist: They tried to coax the alkene starting material to make a branched product that itself was also an alkene. For this, they required a second catalyst, nickel, and a second alkene starting material. This goal introduced a formidable selectivity challenge: how to coerce each (nickel and cobalt) to react with each alkene but not the product alkene.

Because they share the same core chemical feature, getting the catalysts to react the two starting materials with each other—but leave the product untouched—is a bit like asking someone to pick out one specific shoe from a rack of identical styles. Compounding this challenge, the chemical additives required to power the cobalt catalyst also inadvertently powered the nickel, sending the reaction in the wrong direction and producing a mix of unwanted products.

Finding the right combination

The Shenvi lab's solution was to replace those problematic additives, known as silanes—silicon-based compounds commonly used in chemistry to donate hydrogen atoms and drive reactions forward—with a pairing of manganese metal and a mild acid called lutidinium. Silanes are widely used in this type of chemistry but are expensive and generate waste that is difficult to recycle, making them unattractive at an industrial scale. Manganese and lutidinium are cheaper, easier to handle and more amenable to recycling.

"Even in the early discovery phases of drug development, where researchers aren't yet thinking about large-scale manufacturing, this matters," says Milo Smith, a postdoctoral researcher in the Shenvi lab and co-author of the study. "Our approach can get chemists to their target branched compounds up to four times faster than previous methods. That kind of time savings adds up quickly when a scientist is exploring many new molecules at once."

Together, manganese and lutidinium activate the cobalt catalyst while leaving the nickel undisturbed, a principle the team calls "metal hydride selection." The result is a reaction with an unusually high degree of control that consistently steers products toward the branched structures chemists want.

"We were surprised to find that manganese worked because it's a weaker reducing agent, meaning it delivers less chemical energy to drive the reaction," says Chunyu Li, a Scripps Research graduate student and co-first author of the study. "But we found that when paired with the right acid, it hits a very narrow chemical window, just enough to activate the cobalt without causing undesired outcomes with nickel."

Discover the latest in science, tech, and space with over 100,000 subscribers who rely on Phys.org for daily insights. Sign up for our free newsletter and get updates on breakthroughs, innovations, and research that matter—daily or weekly.

Subscribe

The approach proved remarkably versatile. The team used it to produce—in a single chemical reaction—more than 50 new branched compounds that would previously have been difficult to make, requiring multiple labor-intensive chemical steps. The reaction tolerated a wide range of molecular features, including alcohols, amines and other sensitive chemical groups that would have derailed many alternative approaches.

Critically, the branched molecules the reaction produces are not chemical dead ends. Because the product alkene is stable under the reaction conditions, it can be isolated and then reacted again under slightly different conditions, allowing chemists to keep building and modifying the structure in subsequent steps. This ability to iterate on bond formation significantly expands the range of complex molecular architectures accessible from a single starting point.

The team also showed that the technique has broader reach. The same manganese-lutidinium system improved a related process called hydroarylation, in which a hydrogen atom and a carbon-containing ring structure are joined to a molecule across a double bond, and enabled alkene isomerization, a reaction that repositions a double bond within a molecule to change its properties. Both findings suggest metal hydride selection could become a general tool across many areas of synthesis.

Shenvi says the lab is already exploring how broadly metal hydride selection can be applied.

"Many avenues have now opened: mechanistic inquiry, expansion of the reaction and elaboration of the products." says Shenvi. "The principles here are broadly applicable."

Publication details

Chunyu Li et al, Cross- and branched-selective hydroalkenylation by metal hydride selection, Science (2026). DOI: 10.1126/science.aeb2389

Journal information: Science

Provided by The Scripps Research Institute