An easier approach to recreate the powerful nerve-blocking molecule found in shellfish

LA JOLLA, CA—Chemists have long been fascinated and frustrated by saxitoxin: a molecule that causes temporary paralysis by blocking the electrical signals that nerve cells (neurons) use to activate muscle, and which accumulates in shellfish like clams, oysters and scallops. Although the way saxitoxin works has inspired interest in developing new anesthetics, extracting it from natural sources is neither scalable nor practical. Since its discovery, the molecule has defied practical laboratory synthesis, slowing the effort to create long-acting, highly targeted pain therapies inspired by its mechanism.

Now, scientists at Scripps Research, in collaboration with Merck, report a streamlined approach to synthesize saxitoxin and related molecules (known as analogs) in the lab. They combined a synthetic chemical route with an engineered enzyme to make the molecule in less than 10 steps, compared to 11–21 in earlier approaches. Reducing the total step count is a major advance in synthesis because each step typically lowers yield while increasing cost and time; a shorter route can deliver much more material with far fewer resources.

Published in Nature, the research team’s tactic relies on two methods. The first is radical cross-coupling: a technique that uses highly reactive molecular fragments, called radicals, to form bonds that are difficult or even impossible to create with traditional reactions. The second method is biocatalysis, where enzymes—proteins that speed up reactions—carry out chemical steps that would otherwise be difficult.

“In the past, chemists were stuck with expensive, time-consuming strategies that generated only tiny amounts of saxitoxin—but now, one student in a lab can make one gram in a week,” says co-senior author Professor Phil Baran, the Dr. Richard A. Lerner Endowed Chair at Scripps Research. “If a single person can do that, pharmaceutical companies with a fleet of chemists could easily make kilograms.”

But abundant supply alone doesn’t make the molecule a suitable therapy.

“Saxitoxin itself is far too potent to use directly,” explains Baran. “You need analogs that only act on the nerve pathways involved in pain.”

Chemically, saxitoxin belongs to a family of heteroatom-rich natural products: molecules with unusual structures and challenging chemical bonds. Earlier syntheses often stretched to more than a dozen steps, demanded specialized starting materials and yielded milligram quantities. For scientists wanting to study sodium channels—the protein targets of saxitoxin that regulate electrical signals in neurons and muscle cells—this was a major bottleneck.

The new platform addresses such hurdles. First, the research team engineered a fungal enzyme to produce large amounts of hydroxyproline, a simple amino acid derivative that serves as a key starting point for synthesis. Next, they used radical cross-coupling to join two common amino acids obtainable from commercial suppliers. This strategy shortened the synthetic sequence and created opportunities to build new analogs of saxitoxin.

The same radical cross-coupling method used here originated from earlier work involving Baran’s lab and Pfizer, and it was later applied in development of the COVID-19 antiviral Paxlovid. Findings from the current paper also build on a 2024 Science publication from Baran’s lab demonstrating how biocatalysis paired with radical cross-coupling can streamline the construction of complex, medically relevant molecules. With this framework in place, executing synthesis was the next obstacle.

Much of the synthesis effort for the recent study came from Yiheng (Anna) Li, a doctoral student in Baran’s lab and co-first author of the paper.

“The most rewarding part was solving long-standing synthetic challenges for complex natural products,” says Li. “Thanks to Scripps Research’s highly collaborative culture, we could quickly turn that success into functional biological insight.”

To assess the biological function of synthetic saxitoxin and its analogs, the team collaborated with Professor Marisa Roberto, the Paul and Cleo Schimmel Endowed Chair at Scripps Research and co-senior author of the study. Biological tests confirmed that the lab-made molecules behave as expected, providing tools for future studies that involve sodium channels. Dysfunction of these proteins is associated with chronic pain, epilepsy and other neurological conditions.

“We tested the biological activity of new saxitoxin-related compounds on neurons and found that they rapidly block critical sodium channels, reducing neural electrical activity,” says Roberto. “By identifying which compounds are biologically active, we spark new research questions that could lead to important clinical implications.”

This study not only verified how these synthetic molecules act in neurons, but it also achieved the first total synthesis of neosaxitoxin, a naturally occurring saxitoxin analog that’s been evaluated in clinical studies as a potential local anesthetic. Achieving reliable laboratory access to this molecule and others like it eliminates reliance on scarce natural sources.

Beyond saxitoxin, the new platform offers a generalizable entry point for studying similar complex molecules. However, continued progress will largely depend on financial support and collective efforts.

“We’re happy to share material and we’re open to collaboration,” says Baran. “Ultimately, this work provides democratized access to any saxitoxin analog that other researchers may want to explore for therapeutic application.”

In addition to Baran, Li and Roberto, authors of the study “Scalable total synthesis of saxitoxin and related natural products,” include Yinliang Guo, Sihan Chen, Yige Wu, Oscar Poll, Zhouyang Ren, Zhonglin Liu, Roman Vlkolinsky, Michal Bajo and Benjamin F. Cravatt of Scripps Research; and Christopher K. Prier and Kai-Jiong Xiao of Merck & Co., Inc.

This work was supported by funding from the National Institute of General Medical Sciences.