NEW YORK – In medicine, it's not only important to know which compound to administer to treat a given condition, but also the correct dose. CRISPR researchers are finding that if the gene editing technology is to be successful as a therapeutic platform, they'll also have to find a way to dose patients correctly.
But for CRISPR, gauging the right "dose" is often a matter of knowing how long it takes after a CRISPR complex has been introduced to a cell for the editing activity to become inefficient, rather than how much of the complex to administer. A successful CRISPR-based treatment could theoretically depend on being able to turn off the editing activity once that optimal time has been reached.
Knowing when to turn CRISPR's editing activity off is also a matter of safety — an uncontrolled CRISPR reaction can lead to off-target effects and chromosomal translocations.
One way to turn off the editing activity of CRISPR-Cas enzymes may be the anti-CRISPR proteins that viral phages have developed to defend themselves against bacterial CRISPR systems. But the introduction of an anti-CRISPR to the CRISPR editing complex means an additional physical component is being introduced into the cell beyond a typical ribonucleoprotein (RNP). That can require additional engineering, and can increase the size of the overall complex that has to eventually be introduced into the cell.
To address this problem, some researchers have begun to work on artificial off switches. In a recent paper published in Nature Communications, scientists from genome engineering company Synthego described an on/off switch called CRISPRoff that allows for titratable levels of editing efficiency and spatial patterning via a light-induced degradation of the single-guide RNA (sgRNA).
In other words, rather than targeting an off switch to the Cas enzyme itself, Synthego has instead engineered a guide RNA that itself contains a light-activated off switch.
"People are obviously very interested in improving the accuracy of their editing. So, the goal is to be able to provide patients with the safest possible approach," said Synthego CSO Robert Deans. "People are looking to maximize that editing efficiency and eliminate off-target effects [by looking] at different CRISPR enzyme complexes, or evolving enzymes so that they can edit very, very rapidly, so you get a very fast CRISPR response. We've seen some neglect in that space in not looking at the guide as a target for modulating and controlling activity. What we're doing is building in chemical moieties on the guide molecule itself."
In their study, the researchers described their development of CRISPRoff sgRNAs, which are chemically synthesized to undergo cleavage in response to UV light. To develop a universal system, they tested a variety of sgRNA molecules, in which nucleotides at various positions along the backbone were replaced with photocleavable residues. The sgRNAs fragmented upon exposure to broad-spectrum light. The researchers eventually developed what they called dual-breakage sgRNAs (DBsgRNA) — these guides could be cleaved at either one of two sites upon exposure to light and stopping the editing activity.
DBsgRNAs maintained in vitro editing activity comparable to standard sgRNAs when untreated and intact. When the researchers tested the ability of CRISPRoff to modulate genome editing events within human cells, they found that the degree of editing was significantly reduced in cells treated with DBsgRNAs compared to standard sgRNAs.
Further, the researchers were able to fine tune the level of gene editing within a population using DBsgRNAs. They transfected HEK293 cells with DBsgRNA targeting DNMT1 and illuminated a distinct cell sample, one time each, every two hours for two days. After 48 hours, they isolated genomic DNA from all the samples and analyzed them for the presence of indels. They found that they could indeed titrate the editing levels within cell populations by modulating when they irradiated each sample after transfecting them with the DBsgRNAs.
To test the universal effectiveness of the CRISPRoff system, the researchers then created a panel of standard sgRNAs and DBsgRNAs targeting a variety of chromosomes and genomic regions. They found that all but two intact DBsgRNAs formed double-strand breaks at a similar frequency as standard sgRNAs and generated a similar indel profile across all targets. The majority of DBsgRNAs also showed a decrease in editing efficiency when illuminated four hours after transfection compared to cells from the same transfection that remained in the dark.
Importantly, the researchers said, CRISPRoff reduced rates of off-target editing. They ran an experiment that included an sgRNA known to have an off-target site in an essential gene. When DBsgRNAs of this guide were used in conjunction with irradiation, a greater proportion of cells survived, potentially due to an increase in the ratio of on- to off-target events, while maintaining editing efficiency.
"Off-target events occur at a lower kinetic rate and they're less specific, [while] on-target, accurate hits occur very, very quickly because of full specificity with the guide," Deans said. "What you're trying to do is to maximize the dose of enzyme you use to try to get that hit really quickly. And then in the background, your off-targets will start accumulating. And that's why you want to immediately shut off."
With that in mind, the researchers created sgRNAs that had significant levels of off-target editing at one or two sites within the genome. They rationalized that they could maximize the ratio between on- and off-target editing by illuminating DBsgRNAs at an optimal time point after transfection. They transfected independent pools of cells with seven unique sgRNAs and exposed the pools to light at four, eight, 16, 24, or 48 hours after transfection. Following illumination, the degree of editing at many off-target sites plateaued, demonstrating that inactivating DBsgRNAs slowed down off-target editing. By illuminating DBsgRNAs at discrete times after transfection, the investigators found that they were able to modulate and maximize the on- to off-target cutting ratios.
The Synthego researchers did acknowledge that CRISPRoff is currently limited when it comes to in vivo applications — UV light, after all, has low penetrance through tissues. But they added that the development of different chemistries could extend the range of the photocleavable molecules, such as with two-photon cleavage systems. Further, because CRISPRoff makes modifications to the backbone of the sgRNA, it could be compatible with other technologies, such as sgRNA modifications to activate gene editing, or Cas9 modifications to enhance on-target specificity. Such enhancements could, theoretically, make the CRISPRoff system a workable tool for both in vitro and in vivo control of CRISPR technologies in the future, they added.
Deans believes CRISPRoff's design gives it the potential to work in many different ways. For example, he said, because the DBsgRNAs can receive different light signals, they have the potential to be turned off and then back on. Synthego researchers have done experiments in the lab where they've started with inactivated DBsgRNAs, turned them on with a light signal, and then turned them off again.
The design could also allow for multiplexing, in which different Cas enzymes are attached to different DBsgRNAs within the same CRISPR complex, and then turned off with different light signals at different times, depending on the purpose of each enzyme.
"You've got these dimensions that you can play with and build a matrix of alternatives for very tight [editing] control," Deans said.
Eventually, once the technology has progressed far enough to be useful in a clinical setting, he anticipates that it could be used to precisely position a CRISPR therapy in the body.
"[CRISPRoff] lets you do spatial positioning. So, you could even think about tissue architecture in vivo where you can actually use the light to make sure the editing occurs in a tissue-specific pattern — so, things like repairing meniscal tissue or something," he said. "Now, you could actually control the editing, against the architecture of the tissue."
Joseph Bondy-Denomy, a principal investigator at the University of California, San Francisco, and an expert on naturally occurring anti-CRISPRs, believes that although there may be potential for CRISPRoff as a therapeutic tool down the road, it's most useful immediately as a research tool for investigators who are looking to better understand how and when off-target events start to occur in different cell types and cell lines after they're transfected with a CRISPR-Cas9 complex.
"I think where this would be most useful is that one could easily define for a given guide how long they want it to be active for, how long it takes to achieve the on-targeting that they desire, and when the off-targeting starts to creep up over time. That would be the ideal window where having a very simple off-switch like light would be applied," he said.
As part of a therapeutic platform, he added, the technology could be used to generate edited T cells ex vivo, which could then be injected into a patient to treat a disease, for example.
For now, the DBsgRNAs seem to have the most utility for in vitro and ex vivo applications, Bondy-Denomy said. The results shown in the Nature Communications paper would have to be reproduced by other researchers and the DBsgRNAs must be put through their paces by other labs before their full functionality and limitations can truly be assessed. As the Synthego researchers acknowledged, the current iteration of CRISPRoff will likely not work in vivo, given the difficulties of passing light through human tissue, he added. Alternatives such as anti-CRISPRs, Cas9 variants that have been engineered to fall apart on their own after a specific amount of time, or drugs that target Cas9 for degradation will be more useful for patients right now.
"But if it really is a robust and easy-to-adopt tool for others, I think it's a beautiful in vitro application," he said. "The elegance of the work is that it doesn't require more pieces. It doesn't require a drug, it doesn't require an anti-CRISPR protein, it just requires light. So, broadly, as a research tool, it could be very useful."