In his lectures, Amit Choudhary, PhD, shows a slide where Cas9 is flanked by images depicting fire and the Internet. His point is not to situate this enzyme, which facilitates CRISPR-mediated genome editing, in the firmament of humanity’s greatest technological advances, but to offer a warning.

Amit Choudhary headshot
Amit Choudhary, PhD

“At the heart of any powerful technology is precision control,” says Choudhary, a researcher at Harvard Medical School. “Human civilization has benefited immensely because we were able to put precision control on the fire… and you can see the chaos that is happening with the lack of precision control over the Internet.”

His lab has developed a streamlined workflow for the discovery of chemical compounds that act as “off switches” for Cas9, allowing researchers to quickly pump the brakes once their desired genome modification is in place. Such control has become critical, as CRISPR rapidly hurtles from the laboratory into real-world deployment. For example, more than a dozen clinical trials1 are in the works worldwide that aim to exploit CRISPR’s targeted DNA-cleavage capabilities to repair disease-causing mutations in patients, or to reprogram immune cells to hunt and kill tumors. Other groups aspire to turn CRISPR loose in the wild, with plans for “gene drives” to genetically sterilize and thereby control populations of invasive rodents or disease-carrying mosquitoes.2

One of the hallmarks of CRISPR is the accuracy with which it introduces genome modifications. Cas9 forms a complex with a guide RNA strand, which can be designed to recognize and bind specific genomic sequences. Upon binding, Cas9 cleaves the adjacent DNA, enabling targeted deletion of unwanted sequences or creating the opportunity to insert new sequences. But beyond this point, active Cas9 becomes a liability. “Once it’s acted on its substrate, anything it’s doing after that is an off-target effect,” says Renee Wegrzyn, PhD, project manager of the Safe Genes initiative at the Defense Advanced Research Projects Agency (DARPA), which is funding efforts—including Choudhary’s—to control CRISPR.

Anti-CRISPR proteins

CRISPR initially evolved as a bacterial defense mechanism against bacteriophage viruses, and in 2012, researchers at the University of Toronto learned that some phages have evolved anti-CRISPR proteins that can protect viral DNA from being carved up. Since then the anti-CRISPR field has exploded, says Karen Maxwell, PhD, assistant professor at the University of Toronto and one of the co-authors on that 2012 study.3 “I think there’s about 45 published families of anti-CRISPR proteins,” she says, and notes that some of these have already been shown to inhibit genome editing in bacterial, yeast, and mammalian cells.

But proteins may not be optimal for CRISPR control in all scenarios, and Choudhary notes that small-molecule compounds are often a better fit for targeting proteins within the cell. “A typical small molecule is roughly 400 times smaller than Cas9, and so it can get in and out of cells just by passive diffusion and you don’t have to force it or specially deliver it,” he says. On the other hand, Cas9 is an extremely hard target for drug development—it binds to its guide RNA with an iron grip, and cuts DNA with a pair of unusual nuclease domains both of which must be disabled to prevent cutting. This difficulty has deterred many researchers—but Choudhary saw a challenge that was well-matched to his skillset.

Choudhary’s career has followed a winding multidisciplinary road, which began as an escape route from the limited career options he faced growing up in rural India. “Not many kids in my neighborhood went to college,” he says. “My early aspirations were being a cricketer or joining the army or working on a farm.” But his first exposure to the world of chemistry as an adolescent sparked a passion that would ultimately lead him to pursue a PhD in the U.S., where he embarked on research at the interface between chemistry, biology, and physics.

After finishing his PhD, he started to apply the insights gained from this work in the service of identifying therapeutic strategies for common diseases. In particular, he was intrigued by the potential of using CRISPR as a means for genetically manipulating islet cells to treat type 1 diabetes—but was also struck by the lack of chemical tools that might restrain CRISPR’s cutting within patients’ pancreases.

By relying on his expertise in chemical biology, Choudhary arrived at an efficient drug discovery strategy. First, his team homed in on a vulnerability in the CRISPR editing process. Rather than blocking the enzyme itself or struggling to break its pairing with the guide RNA, they focused on the interaction between Cas9 and the protospacer adjacent motif (PAM)—a DNA sequence element that the enzyme uses to decide where to cut. “It’s a really short sequence, and so the recognition doesn’t have to be very strong,” says Choudhary.

Reduce risk of misleading results

They then formulated a variety of assays to test the ability of different compounds to disrupt Cas9-PAM binding, with different readouts from each assay to reduce the risk of misleading results. “If you rely completeiy on fluorescence, for example, there are molecules that are already fluorescent and may produce false-positives,” he says. And rather than immediately testing many thousands of molecules in an open-ended fishing expedition, the researchers enacted a two-tiered approach. This entailed initial testing of selected subsets of molecules representing different structural “families,” followed by deeper investigation of those families that consistently inhibited Cas9-PAM binding in multiple different assays.

Choudhary and colleagues thus identified a pair of small molecules that could efficiently penetrate cells and efficiently and reversibly inhibit Cas9’s ability to cut genomic DNA without noticeable toxic effects. “It’s exciting to see this small-molecule activity,” says Maxwell. “It’s certainly a whole new way of blocking CRISPR-Cas9 activity.” These represent the first such inhibitors to be identified, but Choudhary notes that the true value of this work lies in the screening platform itself, which now gives investigators an effective workflow for not only identifying Cas9-modulating compounds, but also for exploring the mechanisms by which those compounds work. “Anyone with this kind of platform can screen 15,000 compounds in a week or less,” he says.

Wegrzyn sees such agents as a promising addition to the CRISPR control arsenal, especially in the context of therapeutic applications. “If these are orally available and could be delivered systemically, that would be an interesting solution,” she says. “They’ve shown inhibition in mammalian cell culture…now we’re wondering about the pharmacokinetic properties, and if they inhibit editing in vivo.” Small-molecule agents will not necessarily be a universal solution, and DARPA’s Safe Genes funding is also supporting a variety of other strategies for modulating genome editing, including an effort from CRISPR pioneer Jennifer Doudna, PhD, to isolate new anti-CRISPR proteins.

Maxwell notes that such proteins may be more unwieldy than small-molecule agents, but can also be genetically packaged with the CRISPR machinery so that each cell receiving the genome editing system also receives the means to control its activity. “That way, you can ensure tissue-specific expression of Cas9 by keeping it turned off in certain tissues,” she says.

Meanwhile, Choudhary’s group is continuing to push the capabilities of their discovery platform. For example, his team has recently adapted the platform to isolate compounds that do not merely disable Cas9 temporarily, but remove it completely from the cell by promoting rapid degradation—introducing a further level of protection against off-target effects.

He is also grappling with a surge of interest from the genome editing community in the wake of his group’s recent publication.  But he also sees this sharing as an important responsibility. Despite the disputes over patents and intellectual property rights that have captured the media’s attention in recent years, Choudhary notes that the CRISPR research community’s rapid progress has only been possible through a powerful spirit of collaboration. “In this field, things become outdated within months,” he says, “and in large part it is because we are sharing and having active discussions, identifying fault modes in the system and quickly correcting them.”


1. Callaway, Ewen. “Controversial CRISPR ‘gene drives’ tested in mammals for the first time.” Nature, July 6, 2018.
2. Bondy-Denomy, Joe, et al. Bacteriophage genes that inactivate the CRISPR/Cas bacterial immune system. Nature. vol 493, p. 429-432.

Previous articleMalaria Immune Protection in Mosquitoes Is a Study in Duality
Next articleShingles Vaccine Reduced Rash Outbreaks in Stem Cell Transplant Patients