Skip to main content
Premium Trial:

Request an Annual Quote

With RNAi Screening Tech, Startup Recursion Aims to Repurpose Drugs for Rare Diseases


NEW YORK (GenomeWeb) – While RNAi-based screens are routinely used to gain insights into biological pathways or to uncover novel therapeutic targets, a Salt Lake City-based startup is hoping that the technology can help find new uses for existing small-molecule drugs, particularly in the area of rare diseases.

Called Recursion Pharmaceuticals, the company spun out of the University of Utah in November 2013 to develop and commercialize a drug-repurposing approach originating in the lab of researcher Dean Li.

According to Recursion Co-founder and CEO Christopher Gibson, Li's interests include a rare brain condition known as cerebral cavernous malformations (CCM), which are characterized by the expansion of capillaries in the brain that can lead to seizures, paralysis, or cerebral hemorrhaging.

The genetic form of CCM, he told Gene Silencing News, involves mutations or deletions in one of three genes called CCM1, CCM2, and CCM3. As a graduate student in Li's lab, Gibson realized that identifying and validating a new chemical entity that could be used to treat CCM was a multi-year process that wouldn't fit in with his academic timeline.

"We wanted to find a way to quickly get treatment to patients with CCM," he said, so "we decided to pursue drug repositioning."

To do so, Li and his lab conducted an RNAi screen in which siRNAs targeting CCM2 were transfected into human cells and screened. But rather than look at just one or two parameters, "we were measuring several hundred parameters in each cell and … looking for changes that were disease-specific" using open-source cell image-analysis software called CellProfiler, Gibson explained.

"We're basically making a human cellular model of the disease using RNAi," he said. Using the cells that best reflected the phenotype of CCM, Gibson and Li then conducted a chemical suppressor screen with a library of existing small-molecule drugs, ultimately identifying two that appeared to rescue the disease phenotype.

This work is nearing publication in Circulation, Gibson said, and work continues in Li's lab on the undisclosed drug candidates.

Buoyed by this success, the two scientists picked up the rights to the technology and created Recursion, which is using the same approach for other diseases. Gibson noted that Recursion's technology is only appropriate for genetic diseases, but he believes there may be as many as 2,000 that the company could model and test drugs against, many of which are rare conditions for which little research has been conducted.

However, it was clear early on that additional work on the technology was needed and the company spent its first eight months quietly optimizing it.

"We made lots of mistakes" in the initial CCM project, he said. "We were not as aware as we should have been … of all the off-target effects we were probably quantifying. We ended up getting lucky — what the computer picked out turned out to be mostly on-target effects. But we were naive to that issue when we actually did the screen."

To address this issue, Recursion has started to use siRNAs against multiple targets for each gene it is investigating.

"The probability of each of those … knocking down the gene is pretty high assuming you follow the protocol correctly," he said. However, "the probability of the off-target effects for each of those being the same is pretty low.

"We essentially … use three or four different RNAi [molecules] for each disease we want to model," he continued. "We then read out hundreds or thousands of parameters from our high-content assay … and look for those that are changing significantly and in the same direction for two or three of those RNAi" molecules. Other changes that aren't consistent across the different siRNAs are assumed to be off-target effects.

Recursion also created and is continually improving its own software to build upon CellProfiler, and is working to expand the library of chemical compounds it tests. Currently, the firm has about 2,000 drugs that are either US Food and Drug Administration-approved and now off patent or have gone through some clinical testing but never made it to the market. Through alliances with pharmaceutical companies, Recursion hopes to eventually add hundreds more.

And it is through such partnerships that the company sees its biggest commercial opportunity.

"If there is already a generic drug on the market and we find a new use for it, we will publish that," Gibson said. "But that's not very useful for us from a business standpoint. We're not going to be in the business of suing generic drugmakers."

As such, the company wants to focus on shelved drug candidates — those that have proven to be bioactive and perhaps even safe in early human trials, but never made it to the market — and run them through its assay platform to see if they may be effective for a condition that they hadn't previously been tested for.

"We can't guarantee that we're going to find a new indication for [a pharma partner] that is in the exact area that they like to work, but it's very likely for a portion of the drugs we get our hands on, we'll be able to find a new indication," Gibson said. "That's especially compelling for the rare disease space because most of these diseases aren't fully understood, so it may be that a drug that hits a well-known target will work very well for this disease, but nobody knows that's the case and nobody's made that association."

As talks with potential partners proceed, Gibson said that Recursion is running screens in conditions that, like CCM, are monogenic, loss-of-function diseases such as cystic fibrosis and sickle cell anemia. The company is also poised to receive a National Institutes of Health grant that will enable it to look for phenotypes in 2,000 different disease models.

Success with these projects will hopefully provide proof of concept that can accelerate the company's negotiations with interested pharmaceutical firms, he said.

The Scan

Billions for Antivirals

The US is putting $3.2 billion toward a program to develop antivirals to treat COVID-19 in its early stages, the Wall Street Journal reports.

NFT of the Web

Tim Berners-Lee, who developed the World Wide Web, is auctioning its original source code as a non-fungible token, Reuters reports.

23andMe on the Nasdaq

23andMe's shares rose more than 20 percent following its merger with a special purpose acquisition company, as GenomeWeb has reported.

Science Papers Present GWAS of Brain Structure, System for Controlled Gene Transfer

In Science this week: genome-wide association study ties variants to white matter stricture in the brain, and more.