Powerful Gene Editing Approach Offers the Promise of Correcting a Range of Rare Diseases
February 5, 2021
CRISPR is a powerful editing tool, but it works best as a way to knock out genes rather than correct them. New approaches to gene editing, though, are providing the promise of more effective tools for addressing the underlying drivers of monogenic diseases. A recent study in Nature of an approach known as base editing in a mouse model of the ultra-rare genetic condition progeria, a disease that causes premature aging, demonstrated the powerful potential of the approach. While CRISPR has been likened to scissors, base editing has been compared to the find-and-replace function of a word processor. We spoke to study leader David Liu, director of the Merkin Institute for Transformative Technologies at the Broad Institute, about base editing, how it works, and why it may offer the potential to treat a wide range of rare diseases. This episode is part of an occasional series on innovations in gene editing and gene therapy. David, thanks for joining us.
Daniel Levine: David, thanks for joining us.
David Liu: Sure. A pleasure.
Daniel Levine: We’re going to talk about progeria gene editing and your efforts to apply a relatively new type of gene editing to treat this rare and deadly condition. Let’s start with progeria for people, not for Melia who has the condition. What is it?
David Liu: Progeria is an infamous, fatal, progressive disease in which these children who have the disease age very rapidly, and experience a variety of problems and pass away typically around the age of 14.
Daniel Levine: How does the condition manifest itself and progress?
David Liu: So it is caused by a single one letter mistake and the genome, a single C is mutated to a T in just one copy of a gene called lamin A. And the consequences of that single misspelling that C to T mutation is that a toxic protein called progerin is made instead of the healthy protein called lamin A. And the consequence of this toxic progerin protein is that it damages the nucleus, in principle, any cell in the body. But the way that it manifests is commonly by causing symptoms of aging, even in children who are toddlers, and by eventually causing cardiovascular problems, including strokes and heart attacks that are a common cause of death in progeria children.
Daniel Levine: We’ve seen the recent approval of the first treatment for the condition, but the prognosis for patients remains rather grim. What makes the condition a particularly good candidate for a gene editing therapy?
David Liu: Right. So like many of the thousands of genetic diseases caused by mutations in our genome, progeria’s root cause is really this single letter misspelling in lamin A. The FDA approved drug Lonafarnib brings new hope to patients, but like virtually all genetic diseases, this treatment really is treating a symptom of the disease. It’s treating the fact that the protein is toxic because it has a greasy group called a farnesyl tail attached to it. Lonafarnib is a drug that helps impede the process of adding that greasy tail to a protein, but in the process of impeding the ability of progerin to receive this greasy tail, Lonafarnib will also inhibit the ability of any protein that is supposed to receive this greasy tail, even as part of its normal health promoting activity, to do so. So like many of the treatments that are available to patients with a genetic disease, these treatments, while they are important in improving the quality and the lifespan of patients with these diseases, they don’t really address the root cause of the disease, which in this case, again, is that single C to T change. These characteristics of progeria make it an ideal disease for treatment with a gene editing technology that can convert the T that causes the disease back to the C that people without progeria have at that position in the genome.
Daniel Levine: I think if people are familiar with gene editing, they’re familiar with CRISPR; you’re working on a form of gene editing that’s known as base editing. How does base editing differ from CRISPR?
David Liu: Right. CRISPR, as most people have heard of it, evolved in nature as a bacterial immune system. And the reason it serves as a bacterial immune system to protect bacteria from infection by viruses is because CRISPR cuts DNA. It evolved to cut and destroy the genes in viruses and doing so is an effective way to protect a bacteria from infection by a virus. So CRISPR Cas9 has been repurposed to serve as molecular scissors that can disrupt genes by targeting genes of our choosing instead of the viral genes targeted by bacteria. But the outcome of using CRISPR Cas9 protein is still the same in human cells as it is in bacteria, namely, if you cut a targeted gene with CRISPR Cas9, you disrupt the gene. If you cut a gene, you generally cause the cells to insert or delete mixtures of DNA letters at the cut site in an effort to repair the cut and the consequence most of the time of that cut will be disruption of the gene. So if you think about the mistake that causes progeria, simply disrupting that gene won’t necessarily benefit patients. Instead you really want to fix it, to convert the misspelled form of the gene that’s causing the disease and convert it back to the healthy form. And so we developed gene editing technology called base editing that achieves this goal. Base editors, like the original CRISPR system, can be very precisely targeted to a particular DNA sequence in the vast human genome. But instead of cutting the DNA, base editors directly rewrite one base pair into a different base pair, they actually rearrange the atoms in one DNA base to instead become a different base. And so we can directly convert that TA base pair that causes progeria back to a CG base pair–that is the sequence that people without progeria have at that position in the genome–using a base editor without cutting the DNA double helix, and therefore, without creating this uncontrolled mixture of gene byproducts that would otherwise just disrupt the targeted gene.
Daniel Levine: Is the expectation that base editing could replace CRISPR as a therapeutic approach? Is one type of gene editing better or more appropriate for a given condition than another?
David Liu: Types of gene editing tools have different niches. The CRISPR Cas9 form that I liken to molecular scissors can be quite useful if your goal is to disrupt a gene. And there are some diseases, including some clinical trials going on now where simply messing up a gene, simply disrupting it, can have a therapeutic benefit. But for most genetic diseases and for most mutations associated with genetic disease, in order to benefit the patient, we believe that you need to fix the broken gene and turn it back into a normal gene or into something that behaves like a normal gene rather than simply messing it up further. So I think you have to be a little bit lucky in the nature of your disease and in the genetics behind your disease for you to be able to use a gene disruption technology like cutting with CRISPR scissors to actually provide a benefit to a patient, but those cases do exist and they are an important part of this new generation of gene editing technologies. The other important point that I want to make is that the first classes of base editors used the DNA homing machine from CRISPR scissors. So CRISPR played an important role in their development because CRISPR enabled the first forms of base editors, at least to accurately target DNA, just without cutting as the consequence. Newer base editors have been developed in our lab and are being used widely now that don’t use CRISPR and that actually use proteins called tail repeaters to do the DNA targeting instead. And they have certain advantages like we can deliver these new CRISPR free base editors into the mitochondria, an important part of the cell that has previously been resistant to CRISPR because we can’t get the CRISPR guide RNAs–the important component of CRISPR that programs their DNA specificity–we can’t get those guide RNA into the mitochondria. But we can deliver these CRISPR freebase editors, which are entirely made of protein into the mitochondria. And that’s enabled us to make the first purposeful changes in the sequence of mitochondrial DNA.
Daniel Levine: Earlier this year, you published a study in Nature, which uses this approach to treat progeria in a mouse model. This is a collaboration you did with NIH Director Francis Collins. How did the collaboration come about?
David Liu: A lot of science comes about from some good luck and from the enthusiasm of a highly collaborative community. The progeria research community is a very passionate, very dedicated community that cares deeply about pursuing any new kind of treatment, because as you’ve pointed out, even all the current state-of-the-art treatments don’t offer a major increase in the lifespan to patients. So I was invited to give a talk at NIH on base editing. And we had unpublished results at that time of very new results, which normally I wouldn’t even present in a seminar on our use of base editing in a single pilot mouse, showing that we could indeed correct the mutation at the DNA level and that fixed the RNA problem and the protein problem associated with progeria. We didn’t have any data on the lifespan of the mice. We didn’t have any data on whether their cardiovascular systems were more healthy as a result. We just had this one mouse that we had injected one time with a base editor that we fit into two viruses and had shown that that pilot mouse showed a surprisingly high amount of correction in a variety of organs. So as is customary, before you give a talk at a institution, including the NIH, you meet with some of the scientists one-on-one before you give your seminar. And so the last meeting I had before my talk was scheduled was with Francis Collins, and I knew he was a progeria researcher. So, I shared with him these unpublished results that we had had success correcting the mutation directly back to the normal sequence and that we had even done so in one brave pilot mouse with some encouraging results and he got really excited and he encouraged me to present that early data in the talk. I remember going to the men’s room after my meeting with Francis and installing the slides into my talk in the bathroom and calling and texting my collaborators, which at the time didn’t include Francis Collins yet, . and ask would it be okay to present these results because Francis is really interested in them. And they enthusiastically said, yes, that would be fine. So I presented them and right after my talk Francis proposed that we collaborate, which was incredibly exciting and enabling to us because it turns out that the Collin’s lab has one of the world’s largest colonies of progeria mice. Francis’s lab made a very special mouse model of progeria in which every mouse that we used has two copies of the human mutated lamin A gene that causes progeria. So they have two copies each of the human progerin gene, and that makes it very useful as a foundation for a potential therapy, because you can directly target the same human gene that you would be targeting in a patient rather than targeting a mouse gene and then trying to translate your mouse targeting efforts into the human gene, knowing that the human and the mouse genes, while similar, differ in a number of ways.
Daniel Levine: This was an early study in mice that was published in Nature, but it produced some stunning results. What did it show?
David Liu: So we dosed mice with a single injection of our base editors encoded in an adeno-associated virus called AAV9, which is a clinically used FDA approved gene delivery vehicle, and a single dose of our base editor injected into the mice, into their bloodstream resulted in quite good editing. In many of the organs we looked at, say 20 to 60% editing, and correction at not just the DNA level, but the RNA and the protein level, but then things got really interesting when we started to look at more of the physiological impact of the treatment. We looked carefully at the pathology and the aorta, led by our collaborators at NIH who did an amazing job monitoring these mice and doing necropsies and histology on the aorta cross sections. And what they observed is that the aorta of six month old mice treated with our base editor looked statistically indistinguishable from normal, healthy, wild type mice with no progeria mutation, which was stunning in contrast at six months to the cross-section of the aorta of progeria mouse control that we injected with saline. Those control mice all showed the hallmark symptoms of progeria. They had lost many of the critical vascular, smooth muscle cells that are needed for a functional heart. And they had an accumulation of stiff non-pumping, fibrotic tissue called adventitial fibrosis that is also a hallmark of the disease. Then it was just a matter of waiting and observing these fairly large cohorts of mice. They all started out with 10 to 12 mice and seeing how long they would live. And what we saw is that, just as Francis’ lab has done many times before, the saline injected mice had a median lifespan of 215 days, about seven and a half months. In contrast, the mice that received the base editor injection, the single injection, had a median lifespan of 510 days. So they lived about two and a half times as long as their saline injected counterparts. And in fact, they reached a median lifespan that approached the start of old age in normal, healthy non-progeria mice. And as you might imagine, and I think we included videos that you can download from our paper, the sort of vitality and the general activity level, the quality of life, so to speak, of the treated mice was much, much better than that of the saline injected control mice. At their first birthdays, they looked pretty much like normal mice and even at 400, 500 days old they were generally in much better shape than we could have hoped for. Of course, all of the saline injected mice died around 200, 215 days, but these base editor treated mice were much better off. And that level of rescue of the disease in an animal model of progeria as measured by Aridol histology, by lifespan, and by general animal vitality had never before been observed to our knowledge. So we were incredibly excited by the results and frankly, we were surprised how well it worked.
Daniel Levine: What’s known about the safety, if anything, at this point?
David Liu: Well, there’s there’s still quite a lot of work that’s needed before we can know exactly what form of such a treatment might be ready for human patients. Mice, of course, are not humans. So there’s a certain leap of faith that’s needed to translate these results into a child with progeria. But because the results were so promising, we’re sort of taking two approaches: in one arm, we are advancing the treatment pretty much as we described to the next steps towards the clinic in partnership with our research collaborators and with Beam Therapeutics , the company I co-founded; but in the second arm, we’ve recognized that in the several years that have passed since we started the study, a number of improvements in base editing in the viruses that we use to deliver base editors have been reported. And so in the second arm, we’re trying to explore the use of all possible combinations of the latest base editors and the latest viruses to see if we can get an even better result. And then hopefully if the studies that might enable a clinical trial continue to look promising, we can bring in whichever of those look the best from the perspective of offering patients a better outcome. Hopefully, we can bring those into the clinic.
Daniel Levine: How broadly applicable might this be to other monogenic diseases?
David Liu: Well, that’s really one of the most exciting long-term implications the study and other studies that our lab and many other labs around the world have begun to report using base editors to correct animal models of human genetic disease. In the many cases where we have a pretty bulletproof understanding that a specific mutation causes a specific disease, and for those diseases where we know that the base editor is capable of directly reversing the mutation back to the normal sequence, or at least to a sequence that doesn’t cause disease, in those cases it is looking increasingly promising as a way to probe our ability to rescue the disease in an animal, and hopefully to provide a blueprint for how to do so in human patients. So while there is still a lot of work to be done, the field of gene editing and of delivery of these agents and of diagnosing and monitoring genetic disease, and understanding genetic disease in general. These fields are moving so quickly that it’s proven to be a bad bet to bet against the field.
Daniel Levine: And what’s the path forward to making this into a viable therapeutic approach?
David Liu: Well as I mentioned, there is a company, Beam Therapeutics that I co-founded along with my colleagues Keith Joung and Feng Zhang, to bring base editing to patients to cure, or at least treat genetic diseases, ideally with a one-time treatment. So Beam has been working very hard to do so. They have many programs that they’ve announced that range from diseases in the eye to the liver, to the blood, and many other organs. And they are a very vibrant and active and talented group of scientists and leaders who are really leading the charge to make sure that this research doesn’t simply result in publications, but also ends up benefiting patients in some way.
Daniel Levine: So how far might you expect to take this in your research lab and at what point might the handoff be to Beam?
David Liu: Once we have successful animal results, we or anybody really have successful results using base editors in animals to rescue an animal model of the human genetic disease, then I think very academic labs, and certainly my lab is not one of them, are actually positioned to be able to run a clinical trial and to do all of the incredibly important work that’s needed to make sure that experimental drug candidates can be administered to patients in ways that maximize their likelihood of benefiting the patient and minimize risks. So I think that’s a natural transition point to have companies begin to take the base editor and, and to try to bring it into patients. I think it’s worth sort of taking a step back and appreciating that even five years ago the prospect of performing chemistry on an individual base pair in the vast genome of an animal to correct a mutation that causes a devastating, fatal genetic disease with a one-time treatment using a laboratory engineered molecular machine just seemed like science fiction and maybe even unrealistic science fiction. You know, five years ago we were still working on the first base editor. And if you would ask me back then how long it would take before a base editor might be used to treat a disease like progeria in an animal to greatly extend its lifespan and rescue the disease at the DNA, RNA protein and vascular pathology levels, I would have probably guessed a decade or more. So it’s hard to predict exactly how quickly all of this will reach the clinic and will start to have a clinical impact. Again, I wouldn’t bet against the field, given the dedication and the talent of the thousands of researchers who are using these tools and who are contributing to the development of these fields.
Daniel Levine: David Liu director of the Merkin Institute for Transformative Technologies at the Broad Institute. David, thanks so much for your time today.
David Liu: Thanks for your interest and for having me on.
Thanks to Pfizer, Inc., Bluebird, and Novartis Gene Therapies for their support of this podcast, part of our Platforms of Hope: Advances in Gene Therapy and Gene Editing series.
Sign up for updates straight to your inbox.