NEW YORK – As CRISPR researchers develop new and better ways to edit the genome while leaving as few unintended consequences as possible behind, a team led by scientists at the University of California, San Francisco and MIT has developed a method that would create completely reversible gene edits.
In a recent study in Cell, UCSF's Luke Gilbert, MIT's Jonathan Weissman, and their colleagues described their method, called CRISPRoff — a programmable epigenome editor consisting of a single dead Cas9 fusion protein that establishes DNA methylation and repressive histone modifications. This transient CRISPRoff expression initiates highly specific repression of genes and DNA methylation that's maintained through cell division and differentiation of stem cells to neurons. In their experiments, they found that the epigenome editing was highly specific, with minimal off-target editing.
In order to reverse this effect, the researchers then engineered a switch they called CRISPRon, through which they used Cas9-mediated gene editing to inactivate DNMT1 — the main DNA methylation maintenance enzyme in mammalian cells — in cells where they had previously silenced specific genes. Post-DNMT1 knockout, 60 percent to 80 percent of cells demonstrated reactivated gene expression. Similarly, treatment of cells with a small-molecule inhibitor of DNMT1 showed reactivated expression of genes that had previously been silenced, demonstrating that depletion of DNA methylation was sufficient to reverse CRISPRoff-mediated gene silencing.
"If you want to fix a pathogenic mutation, then CRISPR is really enabling. But we felt that for many applications, you may not want to permanently mutate the genome," Gilbert said, explaining the method's genesis. "So, we were searching for ways to turn gene expression off or on, without manipulating the sequence of the genome and just manipulating the transcripts that are produced by a cell."
Medically, he noted, there could be many applications where patients might feel more comfortable with using genome editing if they know that their genes won't be permanently changed, in part based on the concept that gene therapy has already been in use for a variety of applications for more than a decade.
Gilbert further noted that while he, Weissman, and many other CRISPR researchers have been working with tools such as CRISPR interference (CRISPRi) that can downregulate gene expression rather than turning it off entirely, these tools are more awkward to work with from a therapeutic standpoint. While a regular CRISPR-Cas system uses a Cas nuclease to latch onto a gene a mutate it in some fashion to turn it off, CRISPRi uses deactivated Cas9, resulting in RNA-directed transcriptional control of the target region. In other words, it functions "almost like normal transcription factors within a cell, where you constitutively express proteins to target the gene, and then that turns the gene on or off," Gilbert explained. "One of the advantages of Cas9 is you can express it briefly and it'll make a change to the genome that's permanent, and carry it out for a long time. We were looking for ways to basically leverage the strengths of Cas9's permanence and durability, but also leverage this epigenetic editing feature of not having to permanently mutate a gene."
The dead Cas9 works as a programmable DNA binding element rather than as a programmable nuclease, Weissman added.
In terms of therapeutic applications for human beings, the technology has a lot of possible uses, the researchers believe. Before there was an Ebola vaccine, for example, they were working on CRISPRoff as a way to confer programable immunity for anyone who might be affected by the disease.
"If you have a virus where you know the receptor, you could use CRISPRoff to turn gene expression off," Gilbert said. For Ebola that receptor is a protein called NPC1. "We know if you turn NPC1 off in the liver, you're immune to Ebola. But you don't want to permanently mutate NPC1 because you cause cholesterol processing defects and lysosomal storage disorder phenotypes," he added. The idea they had, therefore, was to deliver CRISPRoff to the liver of healthcare workers traveling to Ebola hotspots so that they'd be completely immune to the disease while working with patients.
"And when they left the Ebola hotspot, to avoid detrimental effects of mutating or permanently silencing NPC1, then you could redeliver CRISPRon to restore gene expression and therefore not have any detrimental phenotypes from permanently losing that gene function," Gilbert added.
He further noted that the technology could even be used to modulate pain response. If someone were planning to have surgery, or recovering from an injury, CRISPRoff could be administered to shut down pain receptors for a short time. Once the patient recovered, the pain receptors could be turned back on. It would also help people avoid opioid pain killers.
Another example, according to Weissman, would be in the area of oncology. Cancer studies often reveal the presence of genes or gene mutations that lead to resistance to chemotherapy or radiotherapy. CRISPR is now being considered as a possible addition to some late-stage cancer patients' therapies as a way to knock out resistance genes and reawaken therapeutic response.
In May 2019, Christiana Care's Health System's Gene Editing Institute was preparing to file an investigational new drug application with the US Food and Drug Administration for a clinical trial protocol that would use CRISPR genome editing to improve the efficacy of chemotherapy for KRAS-positive non-small-cell lung cancer (NSCLC) patients. The protocol involved using CRISPR-Cas9 gene editing to knock down NRF2 in order to render patients more sensitive to chemotherapeutic agents.
Under a scenario using CRISPRoff, that gene's expression may only be off for the time it takes to administer the necessary cancer treatment. "You can imagine turning on or off genes in your intestine or in your blood stem cells," Weissman said. "The cells are more sensitive to radiation. But then after you have the radiotherapy, [the cells could return] to normal states so you don't have to worry about the long-term consequences of turning off the gene."
Weissman noted there may be some instances where CRISPRon isn't needed to turn gene expression back on. While conducting their experiments, the researchers noted that the gene silencing in certain loci would decay over a period of days or weeks, depending on the cell cycle turnover rate.
"If that can be tuned, we can now come in [with] one type of treatment and over the period of, say, weeks or months, it would naturally restore and you don't have to come in with the second," Weissman said.
That rate of decay would depend on the tissue in question and the dynamics of tissue turnover "will dictate how long these program changes last," Gilbert added. "In post-mitotic cells like muscle or neuron, these methyl marks in non-replicating cells may last for years and years. So, it depends on the cell type."
There are still many elements to CRISPRoff that have to be worked out and refined before it can be used in the clinic. As with any CRISPR system meant to be used as a therapy, delivery into the right cells at the right time is currently the principal challenge, Gilbert said. The researchers are also working on making the CRISPRoff complex smaller, and capable of targeting more than one loci at once, Weissman added.
But there's already been clear interest in commercializing the technology, he said. Both he and Gilbert, as well as a few other researchers who authored the paper, have already filed for patents on CRISPRoff and CRISPRon.
Indeed, Weissman said, the technology could have applications in cell therapy, and could even aid in the development of so-called off-the-shelf allogeneic CAR-T cells. The current procedure for making CAR Ts is expensive and time-consuming because it involves harvesting an individual's cells, engineering them, and re-administering them as a treatment. As of now, allogeneic CAR Ts could cause life-threatening graft-versus-host disease, and could be rejected by the host immune system.
Using CRISPRoff, however, Weissman envisions being able to edit allogeneic CAR Ts in ways that would camouflage them from an individual's immune system, while also adding safety controls that would allow a physician to turn the CAR Ts off, if needed. "It could make it a much more accessible treatment," he said. "You could have a safer and more universal cell therapy, and you can then do much more complicated engineering because you only have to do it once for many patients, as opposed to trying to do this complicated engineering in a bespoke way for each patient."
Overall, he added, what the study really shows is that cutting DNA and then repairing it is quite difficult. And while researchers have gotten better at avoiding off-target effects, so-called on-target off-targets — unintended consequences of on-targets editing — such as DNA damage response, large indels, and even chromothripsis, can still do damage to the genome.
"So, when you don't have to do that, therapeutically, there are lot of advantages," Weissman said. "Things like base editor and prime editor are examples of this, and we see CRISPRoff as a complement to this, which allows you to do epigenome editing from beginning to end, and to do it in a clean and controlled way."