March 15, 2022

Breaking Barriers in Gene Editing with Harvard Professor David Liu

Breaking Barriers in Gene Editing with Harvard Professor David Liu

Genome editing holds the potential to be a transformative new therapy, and the pace of progress is truly breathtaking.

Gene editing is the process by which alterations are made to DNA.

There are three major challenges: make precise edits at a chosen site, make edits that do not result in subsequent mutations, and have an editing process flexible enough to address the mutations which cause human disease.

This week we talk to Professor David Liu of Harvard University’s Department of Chemistry and Chemical Biology. They discuss the progress that has been made to overcome these challenges, following the development of the base editing and prime editing methods in his lab.

Theory and Practice is a presentation of GV and Google AI.

This season we'll dive deep into the languages of life through explorations of the "dark genome", genome editing, protein folding, the future of aging, and more.

Hosted by Anthony Philippakis (Venture Partner at GV) and Alex Wiltschko (Staff Research Scientist with Google AI), Theory and Practice opens the doors to the cutting edge of biology and computer science through conversations with leaders in the field.


Anthony  00:06

Hello, you're listening to Theory & Practice. I'm Anthony Philippakis. 


Alex  00:10

And I'm Alex Wiltschko. 


Anthony  00:12

In this series, we are focusing on the technologies and ideas that will be most impactful in the next 10 to 20 years. Gene editing is one such technology. It has a pace of change that is truly breathtaking. CRISPR cas9 broke open the field in 2014. Then came base editing in 2016, which enabled a new class of edits and holds the potential to be safer, as it does not break both DNA strands. More recently, prime editing arrived unexpectedly fast on its tails in 2019. This search and replace multifunctional gene editing tool can correct all forms of genetic errors across nearly all cell types. 


Alex  00:50

CRISPR cas9 gene editing has been likened to a pair of genetic scissors, base editing to a pencil, and prime editing to a word processor. From scissors and pencils to word processing shows the extent of the technological leap forward prime editing signifies. 


Anthony  01:07

There are countless genetic variants associated with rare diseases, common diseases and cancer. Genome editing holds the potential to be a transformative new therapy. Prime editing is especially promising because it can mend most mutations responsible for human genetic diseases. It can act on both dividing and nondividing cells, and it can correct multiple mutations at once.


Alex  01:28

Today, we're meeting the scientist whose lab was behind the discovery of two of the three main forms of gene editing. We’re not going to ask him explicitly to name his favorite child but we are going to focus on prime editing because of its versatility and breadth of action. Our guest is David Liu, professor of chemistry at Harvard University, a member of the Broad Institute, and an investigator at the Howard Hughes Medical Institute. He's been newly elected to the National Academy of Sciences, is founder of nine companies including Editas, Beam and Prime Medicine, and has a prolific academic publication record. In 2021, he was author on over 40 academic papers. Today we'll explore with David Liu, not just the science of gene editing, but how you create a successful lab and overcome imposter syndrome.


Anthony  02:25

David, welcome to Theory & Practice. 


David Liu  02:28

My pleasure, thank you for having me. 


Anthony  02:30

The focus of the conversation today will be about genome editing. And over the last few years, we've seen an amazing set of innovations in genome editing, starting with CRISPR cas9, then going into base editing, then Prime Editing. What is the main innovation that made each so transformative?


David Liu  02:48

So the era of modern gene editing as I see, it, at least, really began with the programmable nucleases, of which CRISPR cas9 is the most popular. But really, decades before CRISPR cas9 came onto the scene, of course, zinc finger nucleases,  TALE nucleases, and homing endonucleases were all used to make double stranded breaks that are cuts in the DNA double helix. And making these cuts stimulates the cell to change the sequence of the DNA at the cut site. Unfortunately, when we make a double stranded break, it's difficult to control the outcome of how the cell changes the DNA sequence at that targeted site. But every now and then the cell during the process of trying to rejoin the broken double helix, the cut chromosome will actually make a mistake. And the result will be the loss of a few bases or sometimes larger numbers of bases, or even the insertions of small numbers of bases. In most cases, we need to correct the gene rather than disrupt a gene in order to maximally benefit patients based on the genetics and the biology of the disease. And so that's really the problem that inspired us to develop base editing.  


Anthony 04:09

Ok got it - all gene editing up to base editing broke both strands of the DNA resulting in the cell trying to mend it - sometimes making mistakes at this stage. So; base editing, go on..


David Liu 04:26

So base editing. Instead of cutting the DNA double helix, base editors directly convert one DNA base pair to another DNA base pair in a predictable way at a specific site in the genome that you target.  It turns out that, as you might expect, the most common kinds of mutations that cause genetic diseases are simple single letter swaps, so called point mutations or snips. And therefore, base editors can make four types of single letter changes that end up installing or correcting a substantial fraction - roughly 30% - of known human pathogenic mutations. And those four types of single letter swaps are changing a C into a T, a T  into a C, and A into a G or a G into an A. Collectively, those four changes are called, the transition mutations. And as you might have seen the FDA granted Beam Therapeutics the first IND clearance to begin a clinical trial using a base editor, so that would be the first time that a base editor would be given to a human patient. So, base editors have demonstrated the advantages of directly making DNA changes at a precise site of our choosing, without breaking the DNA double helix. But base editors are mostly limited to the four kinds of single letter swaps that I just described. But, of course, there are many genetic diseases and many mutations that we would like to study in the lab that don't fall into those four single letter swaps.


Anthony  06:13

So base editing is a great step forward, and human trials are starting, which is incredibly exciting, but at the same time, it’s clear that we still need more versatile gene editing tools. We want to introduce more changes other than those substitutions you just mentioned.  And that’s where prime editing, that your lab also developed, comes in. How does prime editing work?


David  Liu:  So prime editing, like base editing, does not require cutting the DNA double helix. 

Prime editing uses a fundamentally different mechanism to install a wide variety of DNA changes of our choosing, including all possible single letter swaps, combinations of multiple DNA letter swaps, small insertions and small deletions. So that's sort of a brief overview of nucleases base editors and prime editors, which, at least as of now are sort of the three easily programmable and widely used technologies for making edits in human cells.


Alex  09:10

Oh, I want to just double click on some of the stuff that you said, just to see if I understand things. 


David Liu  09:16



Alex  09:17

It sounds like there's a growing number of tools in the genome editing toolbox, and they're becoming better all the time. The upsides and the downsides are becoming understood more deeply over time. And I can think of a lot of reasons why you might reach for prime editing and maybe even more so in the future as the kind of the pros improve and the cons diminish. I mean, it can be used on dividing and nondividing cells, it can correct more errors than other methods. It can cover a range of more nucleotides, like you mentioned. And you can, you know, maybe you can work on multiple mutations at once. What are the other reasons to reach for Prime editing at least today and maybe in the future?


David Liu  09:57

So I think if your goal is to make a transversion mutation, of which there are eight types in a therapeutically relevant cell, or to do a small insertion or deletion, and you want to do so at a targeted site in the human genome in a live cell, there really aren't other robust ways to make those changes that I'm aware of.


Alex  10:22

Other than prime editing?


David Liu  10:23

Other than prime editing. And you know that that could change, of course, and in fact, I think it's a sign of a healthy field that the technologies have continued to advance so that each of these challenges has been successively overcome. I mean, as of 2018, or even most of 2019, there wasn't a way at all, to effectively do what I just described in mammalian cells to make small, targeted insertions or deletions or transversion mutations. We had nucleases, and base editors, and that was it.


Alex  10:59

So today, what are the things that we cannot do that might be unlocked with the next most important functional advance in prime editing?


David Liu  11:09

Right, so I think the simplest answer, and certainly the most popular answer to your question, is a long-standing challenge in the gene editing and genome engineering field, which is to insert or replace large, that is, gene-sized pieces of DNA at targeted sites of our choosing in live mammalian cells, animals or prospectively in patients. That aspiration remains unconquered, at least in a way that satisfies, in my opinion, the three most important criteria: high efficiency, high specificity, and programmability, meaning you choose the site, you're not dependent on whether the site happens to have a sequence that fits the natural preferred substrate sequence of a transposase or retrotransposase or recombinase. Now we took, I think, a significant step towards that goal, we used prime editing to install recombinase substrate sites for a recombinase called Bxb1 which is one of the more efficient, well-established recombinases. So Bxb1 recombinase, will cut and paste DNA at sites called attP and attB, which are specific DNA sequences, several dozen base pairs in length. So what we did recently is use prime editing to very efficiently install those attP and attB sites into targeted sites in the human genome of our choosing. And then you can use the Bxb1 recombinase to insert thousands and thousands of base pairs of DNA. So the reason I think that solution, while attractive, still has some room for improvement is that the prime editing part of that two stage process is quite efficient, it will typically be 70/80 plus percent once you've optimized the prime editing, but the Bxb1 recombinase, consistent with the literature reports will typically work maybe 20, or 25%. So when you multiply them out, you don't get hyper efficient results, although I think, for certain kinds of gene therapy, even getting five or 10% incorporation of a correct wild type gene is thought to be therapeutic. So this approach could already prove useful for the study, or potentially even the treatment of certain kinds of genetic diseases where the whole gene is missing. 



So hold on. If the gene is missing, there is nothing to base edit…so what do you do?


David Liu: You really need to install the gene. And ideally, you don't just want to install it in a new place in the genome, because we know that where a gene exists in our genomes is a really important part of how that gene is regulated and its proper function. But ideally, you'd like to install the missing gene exactly where it is in people who don't have that genetic disease. So that's certainly one of the remaining challenges that many labs are pursuing. But there's a lot more work than just being able to make those mutations in human cells in a laboratory before we can really, you know, as a field fully live up to the potential of those capabilities. And that's where so much of the important work comes in: getting the efficiencies high, getting the specificities maximized, minimizing the frequency of byproducts or undesired effects on the cells. And, of course, pairing those editing technologies with delivery systems that are clinically relevant.


Anthony  15:22

You know, this is just fascinating, David, the progress has just been breathtaking over the last decade. You know, one of the things that I'd like to just double click on is actually delivering the genome editor to a tissue of interest. Am I right, that more and more, that's the biggest barrier to seeing these tools used clinically? And what are the approaches that you're most excited about for addressing that?


David Liu  15:45

I mean, I guess the way I look at it is: there are a number of challenges that all have to be solved, in order to achieve something that sounded even five years ago as audacious as, “Go into the genome at one site in a in a human patient, and directly correct the misspelling in their genome that causes some grievous genetic disease.” That situation affects hundreds of millions of people in the world, because collectively, the genetic diseases are actually not so rare in aggregate. So I view it as we need several challenges to be addressed: you need some kind of molecular machine that makes the desired change, you need to make that machine perform at a level that is sufficient to provide patients with a very favorable benefit to risk ratio, you need a way to deliver the machine into the relevant tissue, you need the biology to tell you in bulletproof, completely rigorous terms with as much certainty as possible, the relationship between the gene and the disease, and the relationship between the tissue and the disease. And then you need a way to produce and to offer to patients, all of these materials, in forms that are cost effective enough that they won't just be limited to a tiny cadre of patients who have the good fortune of being able to pay exorbitant amounts for these treatments. So there's a number of bottlenecks, and the way I see it is as we make progress in one bottleneck, of course, there's more pressure on the other bottlenecks to follow suit. And the delivery field is also I think, making a lot of progress. So we now have a variety of ex vivo and in vivo delivery methods, that, while still leaving most of the human body difficult to deliver into, has collectively made, in my opinion, enormous progress over the past decade. We can deliver now into hematopoietic stem cells, at least ex vivo, efficiently into the liver, into muscle cells, into the heart, into the brain, into the eye, into the inner ear. While each of those tissues, you know, I'd say the delivery solutions are not ideal, they're not complete, we have some solutions now that that are therapeutically relevant. And that does feel like a bit of a change from say, a decade or two ago.


Alex  18:30

From where I sit, which is not in the field, things are moving incredibly rapidly. And you're sitting at this this nexus of several different technologies from the basic science all the way to the delivery. And you and your lab have worked on absolutely foundational pieces of this story. And so if I can just ask you, how are you running this lab to be so productive? Like what's your philosophy when it comes to people? How do you how do you think about that?


David Liu  19:00

Yeah, well, thank you for your kind words. The answer to your question is just that I have the good fortune of working with amazing graduate students, postdocs and collaborators. It is unequivocally the answer to your question. And it really is about the human capital; it's the human ingenuity. It's the human dedication that that drives so much progress in any field. So you know, I've really learned to first love my research group, I love them like family, and to recruit people, not just on the basis of their apparent intelligence and their experience, their background, their knowledge base, but probably at least as important if not more important, their ability to work together, their emotional maturity, the extent of which they're passionate about what they do in a very genuine fundamental way. So to me, there's been a, an evolution in my own thinking of what types of team members make for the most effective team that has correlated with this period of productivity and probably not by coincidence. So I, you know, I think the key is always to value very highly, treasure, the amazing people who get up every day, and they'd rather do nothing more than advance their research project,


Alex  20:30

You say that you've learned this or that your thinking has evolved. And now you have this philosophy. But what led you to that?


David Liu  20:39

Yeah, so I've been a professor for 22 years now, all at one academic institution, Harvard, although my geographic adjacencies have changed, focusing initially mostly on Harvard's FAS campus, and now at the Broad Institute most of the time. But within those 22 years, you know, early on, I think, like a lot of young academics, I thought that a research team of people who were as smart and as driven as possible, should be the priority. And, you know, I think that's a naive perspective, probably a mistake for a couple of reasons. First, if you excessively prioritize people on the basis of their past accomplishments, their ambition, their CVs, so to speak, you frequently over emphasize the qualities of the past and under emphasize the qualities that determine how well, they'll work together with others in the future. And second, of course, if you select on such a narrow set of criteria, such as you know, have published papers in the past as an undergraduate, what are their GRE scores, what's their GPA, how famous is the school they come from, if you limit yourself to those very traditional criteria, you'll end up with a very homogeneous group, in all respects, not only in their background, but I think you'll tend to skew your group homogeneous in terms of gender, in terms of demographics. So I've been through stages where my group was more homogeneous or less homogeneous. And for me, it's much better to have a diverse group by all metrics, gender, demographics, background field that they come from, you know, big schools, small schools, and to have people who just work beautifully with others.


Alex  22:36

I'm just curious if you could rewind to before, like, what were the teaching moments or the teaching experiences that seems like your mind changed?


David Liu  22:45

You know, we all come into our professional life or our pre professional life in the case of graduate students with insecurities. I recently shared with my lab and I guess I'll now share with your listenership that, you know, like everybody, I was full of insecurities when I started as an assistant professor in 1999. In fact, since I just turned 26, when I started as an assistant professor, I probably had more insecurities than average. When I was about a third year professor, I was convinced that I wasn't going to make it, that I had a paper or grant or two rejected in a short period of time. And, and I was pretty convinced that I, you know, I really wasn't going to make it, I wasn't going to accomplish enough to earn tenure, practically speaking, but more importantly, I wasn't going to accomplish enough to live up to the expectations that I had for myself and that my department had invested so heavily in me. So I actually prepared a slide that I was going to walk into my chair's office at the time and, and present to say, you know, I realize this isn't working out. And I just want to plan with you the best way to exit my assistant professorship and do the least damage possible to this department. So this, you know, this all happened. I went through the slide, I thought about things in the meantime, we had some more successes research-wise. And you know, I guess I convinced myself or maybe with the discussions with others that maybe that was an overreaction, but I think it exemplifies the fact that insecurities, imposter syndrome - it's all very real; we all feel it. And you know, as I recently shared the story with my group to say the answer to imposter syndrome to insecurities, in my opinion, is to learn how others are successful. And the way you do that is by helping others be successful. So in a strange way, maybe even counterintuitive for some: to me, the best way to live up to your potential by addressing your insecurities is to learn how to be the most effective possible scientist, researcher, collaborator, team member. And that comes from first, helping others be as successful as they can be. You learn in the process how to be successful, and you know, others then naturally want to make you as successful as possible, because you've been so helpful to everyone else. So that's the kind of culture now that I try to perpetuate, in part because I've learned, you know, through a variety of experiences, including my own, that addressing insecurities and focusing on how to maximize everyone's ability to help each other succeed, is not only a much more pleasant way to run a group, but it's much more effective, at least for me.


Anthony  26:03

You know, it's really inspiring, David; thank you so much for sharing that with us. I think that a lot of the trainees, especially who listen to this podcast, will really find that to be deep words of wisdom.


David Liu  26:14

Well, I hope it helps others.


Alex  26:17

I think that it will. Science is a human enterprise. And I think you've articulated the humanity of it.


Anthony  26:23

So you know, to close up this episode, I'd like you to kind of look forward just a little bit. And, clearly, genome editing is a revolution, one of the great developments of our time. What do you think will ultimately be its impact on the treatment of disease? Well, classes of disease will be most transformed by it, which will be further off.


David Liu  26:43

I see it as the start of a great revolution. And not just the life sciences, but also in medicine and in the quality of our lives. It's easy to predict, I think, at this point that the first few steps in terms of societal impact will be the treatment of initially, a small number of monogenic diseases that are all quite serious, like sickle cell disease. But success in doing so will inspire broader applications of those technologies. I think most people at this point would agree that as long as we have a good understanding of the potential benefits and the potential risks at using gene editing to alleviate the enormous amount of human suffering that comes from rare genetic diseases that frequently will end the life early and cause severe deterioration of the quality of the life, is ethical. But, you know, it becomes more of a slippery slope, as I just taught my gene editing class at Harvard, when you begin to think about expanded applications of gene editing. So, should gene editing be used to install disease prevention alleles? What about installing converting alleles that give one a higher risk of Alzheimer's to ones that are neutral? APOE4 to APOE3? What about APOE4 to APOE2: APOE2 giving one a lower than normal risk of Alzheimer's? What about installing the so-called Icelandic mutation and amyloid precursor protein that less than 0.1% of Icelandic people and pretty much nobody else has this Alanine-673 threonine mutation in amyloid precursor protein substantially lowers your risk of Alzheimer's disease? What about installing mutations that treat genetic deafness? Many people in the deafness community believe that would cause somebody to miss out on a rich culture of being deaf. It's not so black and white as to what an appropriate use of gene editing would be. I think early on, it's clear, because all of the credible gene editing therapeutic applications are really aimed at treating serious genetic diseases. But their success will pave the way, on the one hand, for an exciting expansion of the application of these technologies, with potentially enormous benefits to society, but on the other hand, will also prompt unnecessary dialogue about where on the spectrum of human genetic disease treatment versus disease prevention versus human improvement. You know, what should be inbounds versus out-of-bounds?


Anthony  29:38

You know, that was wonderful. David, thank you so much for taking the time to come by and speak with us today. Really appreciate it.


Alex  29:45

Thank you, David. It's a great conversation. Appreciate it.


David Liu  29:48

It's a pleasure. I really enjoyed it and thank you for having me.



Alex  29:58

Huge thanks to Professor David Liu. We usually take time at the end of each episode in the spirit of regular in-person meetups in Boston many years ago to discuss a big problem, a nail and possible solutions, and hammers, inspired by what we just heard. So Anthony, what do you have is a hammer or nail this week? 


Anthony  30:16

Actually I'll do something a little unusual and break with our tradition, and actually talk about a hammer and a nail, and in particular, how the two come together for a really special opportunity.


Alex  30:29

Well, that's what we're all about. I mean, we talk about hammers and nails, but the whole point is the moment when they meet and actually solve real problems. I'm excited to hear what you have to share.


Anthony  30:39

Yeah, I mean, in this case, the hammers are not just one but two of the great ideas of the 21st century, which is to say complex trade, human genetics, and genome editing. And then the nail is the number one cause of death in the world: coronary heart disease. So you know, as we heard about in last season, when we talked to David Altshuler, the past, roughly 15 years had been this incredible time in human genetics, where for the first time, we've been able to find lots of variants that are associated with common diseases, like coronary heart disease, or diabetes, etc. Before that, we could find lots of variants that were associated with rare Mendelian syndromes that are caused by just one genetic variant like sickle cell disease, or cystic fibrosis. But the idea of being able to find variants that are associated with common diseases, which are multifactorial, is relatively new. Now, going into this intellectual journey, I think one of the things that everybody expected is that we would find genetic variants that increased your risk of disease. And it kind of makes sense, right, because there are a lot of ways to break something. And most mutations break something. And so that increases your risk of getting sick. But one of the huge surprises, which I don't think anybody expected, was that we actually start to find a good number of variants that actually protect you from disease. So instead of causing disease, they actually protect you, which is kind of amazing. 



So this is like a mutation in a gene, which, for all appearances, breaks the gene, but yet makes you healthier. 



Exactly. And you know, not all mutations necessarily break the genes, sometimes they can actually cause them to function better, things like that. But yes, they often do. And in fact, there's a very famous example, that kicked off this whole way of thinking. And that actually had a huge impact in in the pharmaceutical industry, and how we think about drug development writ large. And it was in a gene called PCSK9 that's involved in cholesterol metabolism. So just to do a little bit of basic genetics, we often talk about genetic mutations as either being gain of function or loss of function. So a gain of function mutations cause the protein to do something it doesn't normally do. And then loss of function mutations, cause it to break its function, and so it no longer works. And in this gene called PCSK9, we knew about it from patients who had gained a function, and they had unnaturally high cholesterol. So this was a very rare Mendelian syndrome, with people having very high cholesterol. Now, Helen Hobbs and Jonathan Cohen, who are two researchers at UT Southwestern, started studying large numbers of individuals in sequencing this whole gene in them. And it turns out that a good fraction have a loss of function mutation in PCSK9 that breaks its function. And they have lower cholesterol than expected


Alex  33:42

Wait, so this is a mutation in the same gene and some mutations make your bad cholesterol higher, and some other mutations make your bad cholesterol lower.


Anthony  33:53

That's exactly right. And so as they were studying it, they saw that the people who had lower bad cholesterol had much lower rates of coronary artery disease as well. And then they started looking at people who carry these mutations. And they said, we know it's good for cholesterol, does it do bad things elsewhere? And as far as we can tell, and this has been very extensively studied, the only thing that this gene seems to do is to raise your bad cholesterol.


Alex  34:22

So why do we have this gene in the first place if the only function that it has is to make us less healthy?


Anthony  34:28

No one knows. And I think it's a really big mystery. Presumably, at some point in human evolution, this gene served a function that was beneficial. But in our modern world with our modern diet, it's lost its utility. It's a really puzzling situation.


Alex  34:43

And so I guess now the question is, well, how do we get rid of this gene in more people?


Anthony  34:48

Well, and that is the story of what's going on since this discovery in 2005. And so first two pharmaceutical companies, Amgen and Regeneron said, “What if we make an antibody for the protein produced by this gene,” which basically kind of takes it out. And sure enough, they put those antibodies to clinical trials, they both were safe and effective, lowered cholesterol and in fact, even prevented coronary disease.


Alex  35:14

This is kind of like a cholesterol vaccine in a way.


Anthony  35:17

Yeah, well, unlike a vaccine, you have to give it many times. So these are, you know, medications that are given over and over again, whereas the vaccine, you know, you just give it once and your body makes your own antibodies, here, you're actually giving the antibody for it. And in fact, it really sparked a revolution across the pharmaceutical industry. Because as human genetics was progressing, we started to realize that actually, there are lots of genes, and lots of variants that are protective of disease. So in coronary artery disease, we now know of nine different genes that, when mutated, protect you from coronary disease. There are examples of diabetes and Alzheimer's. And so lots of people in the pharmaceutical industry start to say, “Okay, this should be the basis of how we find targets for common diseases; we should look for variants that protect you from disease, and then make a drug that imitates the effect of this variant.”


Alex  36:11

So is the idea that living among us are superhuman mutants that have a change in a gene, which makes them just at birth healthier? And maybe we can confer that benefit that they have to other people that don't have?


Anthony  36:25

Exactly right, exactly, you nailed it. And you know, of course, these people who carry these protected mutations, they're not protected from all diseases…


Alex  36:33

…just one that that mutation addresses.


Anthony  36:37

Exactly. Now, here's where it gets interesting in the story intercepts with genome editing. So a good friend, colleague, and mentor at the Broad Institute, of mine, Sekar Kathiresan, was actually the one who found many of the subsequent mutations that protect you from coronary artery disease. So he's a coronary disease geneticist and researcher. And he kind of said, well, instead of making a drug that imitates the effect of a human variant that we know is safe and effective, why not use genome editing in order to just introduce that variant. And there are a couple of reasons to say that this actually might be even better than trying to make a drug. You know, especially for common diseases, one of the big problems is compliance. Like, you know, even after you've had a heart attack, a large number of people won't go on and fill their prescriptions for the drugs that will prevent them from having a second heart attack.


Alex  37:30

Really. So a mild inconvenience is enough to kind of stop human nature from doing something that's self preserving. 


Anthony  37:41

You know, you're gonna be shocked to hear that humans are not always rational.


Alex  37:43

Yeah, I mean, sometimes I order Thai food, and I eat the whole thing, even though I meant to just eat half, you know, it's just being a part of the species, I guess.


Anthony  37:52

Exactly. No, you said it. So, no, but one of the problems with this approach is that in general, we're still in the infancy of our ability to deliver genome editing to the organs we care about. Now, here, again, this is where Sekar had a really nice insight, which is: the one organ that right now we have a fighting chance of delivering genome editing to is, the liver. Why is that what makes a liver special in this regard. So your liver is very involved in fats and processing fats. And there are these new delivery technologies called lipid nanoparticles. By coincidence, they're actually the basis of the Pfizer and Moderna vaccines for Covid as well. And you can put mRNA inside of these lipid nanoparticles and inject them, and then they go around the body and then make it to the liver, because they're fat, they kind of hold on to the liver. Now, in the case of Covid, vaccines, we give them subcutaneously, rather than your bloodstream, and so they just stay local. And also, we're doing them in much lower doses, but the principle is the same. And so going back to these mutations that prevent coronary disease, many of them are involved in cholesterol metabolism. So they're actually proteins that are produced by the liver. So Sekar thought, “Alright, can we actually inject lipid nanoparticles with genome editing technologies, and introduce these variants that we know are protective of coronary disease?” So amazingly, he actually started a company which GV was invested in, and I had the privilege of helping to midwife it into existence with Sekar and many other people. And they're not in humans yet. But they have really impressive monkey data, that they can actually deliver genome editing technologies, and then be able to lower cholesterol. And, you know, we'll see where this story goes. But I think it's just kind of an amazing thread of Human Genetics and gene editing technologies being brought to bear on what is probably the biggest public health burden there is. So you know, to me, that's a great example of bringing a hammer and nail together.


Alex  39:52

It sounds like it's still early days, but that sounds like an incredible kind of confluence of technology and biology and problem. And it seems like an obvious thing to do, like, we know this gene, when it's gone is better for you, and it doesn't harm you. So let's get rid of it as thoroughly and as cleanly and as safely as possible. But one thing that comes to mind and just thinking back on the conversation we just had with David is, where does it stop? Let me explain what I mean by that. So if there's a mutation, that, if you don't remove it, you'll die. Obviously, we need to do something about that, right. And I think that our conversation around Mila and N=1 trials is a really good example of that, like, if you can make a change that save someone's life, you should do it. And here, this is a step away from that, which is, if we remove this gene, your risk of heart disease goes down, but it's not necessarily deterministic. But then there's a slippery slope after that, that ends all the way in, “Jeez, I'd like, myself or my son or daughter to have a different color of eye,” or something like that. So how do you think about that spectrum for interventions in our own genome? 


Anthony  41:11

Yeah, I mean, I'll be honest, you know, you’d have to ask a wiser man than me where this ends. A couple of thoughts, however, that I think, are very interesting is that our social norms on what are appropriate therapeutic boundaries have evolved over time. And one of the ones that really comes to mind for me, is IVF, you know, in vitro, fertilization. And when that first happened, there were no shortage of people saying it was playing God, and no shortage of people that objected to it, and sort of had real kind of ethical qualms. At the same time, you know, I think now the number of people that undergo it, and we feel like this is a chance to give people that would otherwise not be able to have children, the opportunity to have children, or going back to genetic diseases, the ability to select embryos that don't have genetic disease before implantation. You know, again, still controversial issues. No question. I'm not trying to say that we as humanity are completely agreed. But the pendulum has certainly shifted relative to when that technology was first introduced. It's hard to know where this will head. But to me, I think, at least at a minimum, being able to use genome editing, to prevent disease, at least I personally don't have any qualms with it, I think there's a chance to really alleviate human suffering. I'd actually even on a story about this with Sekar, my friend who started Verve, this is a very kind of personal and mission driven thing for him as an example, his brother died when he was in his early 40s, suddenly from a coronary disease, you know, was actually running earlier in the day had no signs of being at risk, and suddenly had sudden cardiac arrest due to a heart attack. And, you know, you think about the ability to find people who are at risk, and then to be able to intervene on them before something tragic happens. I think this is one of the big opportunities of our time. 


Alex  43:11

Well, thanks for sharing that story. And it kind of is reminiscent to me also of a story that you shared about early detection of atrial fibrillation, you know, just knowing that there's a risk factor, identifying it early, and then that gives us the ability to prevent human suffering or death for the individual and suffering for the family. So I think there's a thread here and a deep and an important one, and the hammers are different and the nails are slightly different, but it seems part and parcel.


Anthony  43:38

Excellent. Well, thank you so much, Alex. 


Alex  43:40

Thanks, Anthony for sharing. 


Next episode we’ll be speaking to Dr Jay Bradner about the Future of Cancer Medicine. 


Then later in the series, we'll be discussing diverse subjects such as the cellular and molecular basis of aging, protein folding, and the future of psychiatry. If you've got any questions for us, or our guests, email or tweet at @GVTeam. We'd love to hear from you.


Anthony  44:16

This is a GV podcast in a Blanchard House production. Our science producer was Hilary Guite. Executive producer Rosie Pye, with music by Dalo. I'm Anthony Philippakis.


Alex  44:31

I'm Alex Wiltschko. 


Anthony  44:32

And this is Theory & Practice.

David LiuProfile Photo

David Liu

David R. Liu is the Richard Merkin Professor, Director of the Merkin Institute of Transformative Technologies in Healthcare, and Vice-Chair of the Faculty at the Broad Institute of Harvard and MIT; Thomas Dudley Cabot Professor of the Natural Sciences and Professor of Chemistry and Chemical Biology at Harvard University; and Howard Hughes Medical Institute Investigator.