Home / In Conversation: George Church

In Conversation: George Church

In Conversation: Q&A with George Church

More about George Church
Arc Fusion Talk
New York Times Profile by David Ewing Duncan
Church Lab Page

There are a few individuals among us who seem to be living a few years into the future. They seem to be already working on the next big thing before most people are aware the next big thing exists. They are researchers, entrepreneurs, artists and others whose imagination thrusts them ahead, and whose knack for getting things done makes what they imagine become real.

One of these future people is Harvard geneticist George Church. Arc Fusion spent some time with him recently to get caught up on what he is thinking about and doing in his lab at the Harvard Medical School in Boston.

Arc Fusion Magazine: Can you talk about what you're doing right now? What's up in your lab?

George Church: So as usual we're trying to develop disruptive and transformative technologies that can be used in multiple fields of biology – or science in general. Some of them are mostly biology independent, for example encoding information to DNA. That was an experiment I did with my own hands in 2012 but since then we've had a collaboration with Technicolor where we've been encoding movies [storing them as DNA code] and including one of the classic first generation hand-colorized movies that they spent a lot of money to restore. We're trying to start with archiving information of very precious digital sources. We're also trying to extend this work to include analog sources.

AFM: Can you explain what that is for people that don't know what you just said?

GC: So the problem with archiving is you can’t use your day-to-day disk drive. These are those things you want to store for a decade or more. The problem is that the longer you store it the more you have to transfer it onto new fresh media. You don't want to make mistakes because those mistakes will accumulate over multiple copying cycles. And you know some of these things have been archived for a century, including this movie that we did with Technicolor called Voyage to the Moon.

So DNA as storage [literally digitizing data into As, Cs, Ts and Gs] – has two advantages. One is that it's tiny so it's about a million times smaller than any other storage media. And it has a good longevity track record at 700,000 years. And third it has zero energy consumption during storage. So during those 700,000 years there was no energy source.

AFM: This was DNA preserved in ancient bones of various creatures?

GC: Well that was a horse but you know he was clearly imperfectly stored. If we would store it today we would probably put it sealed inside of tiny glass spheres without oxygen and without water.

AFM: What else are you working on?

GC: We're working on aging reversal by using gene therapy where we're harvesting a pretty vast and rigorous literature on things that either cause longevity or other desirable aging consequences in animals. Many of these transgenic or genetic from birth. We want to see if you can turn this into a gene therapy that you apply in an adult animal and then move it into humans. This seems to be a unique approach and we're doing about 48 different cocktails…

AFM: Are you using CRISPR-cas9 [a gene-editing method] for that?

GC: We are not. Our lab more and more is developing alternatives to CRISPR. Not that we think CRISPR is bad or anything - no need to sell all your Editas stock. [Church is a co-founder of Editas, a start-up developing CRISPR technology for gene therapy]. We are finding increasing numbers of alternatives to that.

AFM: You're making it even easier for junior high students to edit DNA…

GC: Absolutely. I mean the $50 entry fee for being able to make designer whatever you want just seemed way too high [He smiles].

AFM: Does this technology that is so inexpensive and relatively easy for gene editing bother you at all – that it will soon be easy enough for a high school student to do?

GC: Of course. Everything worries me. Maybe not bothered – worried. And I try to encourage everybody to worry. Not to panic, but to be concerned. And not just be concerned, but to seek solutions and to be concerned about those solutions, too. It seems like an infinite regress but it actually is the way that we develop safe technologies. By the way, a high school student just won an Intel award for a CRISPR project.

And there is a safety component in almost every field – and sometimes people don't notice them happening behind the scenes. But people are generally aware that cars have shoulder harnesses and airbags, and you have test crash dummies and so forth. So you really have to have a big industry doing safety.

AFM: Are you all doing that in your lab?

GC: I think we're one of the main safety engineering labs for molecular biology and synthetic biology. And almost every time we think of something that might be a few years out before it's even research grade, much less practical, we immediately start raising the alarm and doing something about it. We did this with gene drives in 2014. We did it with large scale synthesis of DNA in 2004.

We did it with protocols for human subjects research in 2005. We're among the first to even notice that there is a problem, and hence one of the first to come up with some kind of safety solution. And sometimes people think there is not a problem but there will be eventually.

AFM: That is good to hear. There are clearly negative repercussions for every technology, probably starting with the discovery of fire - there is a plus and a minus. And you've got to really guard against the minus.

GC: There is great book “Evolution Man: Or, How I Ate My Father” where they describe the discovery of fire and one of the first things they do is they start a forest fire and they burn down their whole village. So every technology has a potential downside. And the funny thing is the better you are at anticipating and preventing the less of a hero you. Because nobody knows that you did it right.

AFM: So what else are you working on?

GC: So we’ve got DNA data and aging reversal; we’re also using gene drives to eliminate malaria and other vector-borne diseases - not just insect or mosquito borne, but also mammal, like the white-footed mouse for Lyme disease.

AFM: By the way, will genetically modified mosquitoes work to rid us of malaria? I know it's been attempted before, but hasn’t really worked.

GC: Yes. There are already genetically altered mosquitoes that are released into the wild. They're typically sterile males. (Before making genetically altered males they used irradiated males). But the problem is they knock the population down maybe 80 percent but you still have 20 percent, which just pops the populations back up. The advantage of gene drives is that rather than the things you're introducing being dead ends [from a reproductive and evolutionary stand point], they are the beginning of a chain reaction. And they are super prolific in that every one of their offspring is affected by the same thing, which is typically some desirable [genetic] payload instead of having just half of their offspring have it. So if you put a billion mosquitoes out in a population of a trillion it's a drop in the bucket and eventually it vanishes. But if you put a billion out there each of and all of its offspring have the thing, pretty soon you have all trillion carrying the desired genetic payload.

AFM: Is there a downside to this technology? Are people talking about this?

GC: Oh this is widely discussed. We raised the alarm before we did any experiments in 2014 and in the same paper suggested a bunch of safety mechanisms to keep it from going from lab into the wild, and if it goes into the wild, how to reverse it.

This was discussed by many branches of government and nongovernmental organizations. The Gates Foundation is very aware of it and for the most part, the whole malaria community. Within a year, gene drives are now working in four different organisms, all four that have been tried and there has been a community document that we helped organize talking about safety guidelines for everybody rather than just one group saying hey this could conceivably work. Also the safety mechanisms that we proposed are now tested and they all work in at least one organism. But they need to be scaled up and tested on multiple organisms.

AFM: Craig Venter and other scientists recently published a paper that claims to have identified the minimal genome needed to sustain life. What do you think of that? How important is this?

GC: Well I think it's important and I was quoted as saying it's a tour de force. It's the culmination of 20 years of sustained effort on a particular project. It's not that often that you get a big group that manages to maintain a high level of spending and get to the goal that's set out. Usually its goals drift off in a 20-year period. So that's to be noted and applauded.

AFM: But…

GC: I think that Craig’s project could be described as applied science. We still have some basic science questions to answer. The core question is how do we figure out why there are some genes in this minimal genome that we don’t understand. Why are there? I think that was the main basic science question. And I don't think they contributed much to that. That is to say that there is a rich history of going back to pre-genomics where people would sequence the gene next to their favorite gene and say “oh it's unknown but it's in the same operon so maybe it's related function” or more generally they would hunt it down via crystallography, genetic and biochemical couplings, etc.

And the function of many, many previously unknown genes have been figured out using tools like these: what does it interact with? What is it near? What residues are involved?

AFM: Can you talk about the Genome Project “phase 2” that you’re helping to lead? An update on the original Human Genome Project that read the first human genome – and your own Personal Genome Project that sequenced lots of people over the past few years, and provided a platform to share DNA information?

GC: So I helped propose the first Human Genome Project in 1984, but it was set up to sequence one human at whatever cost, and I didn't like that aspect particularly. I wanted it to be a technology project that would bring down the cost, which eventually we got to towards the end (2005-2013). The Personal Genome Project wanted to demonstrate the value of sharing information both back to the individual and across the world. And I think that was intended not to be a production project but kind of an inspiration project - and I think that's had its impact over the last decade. It went from being exceptional to do that to being standard to do that.

GC: But the idea with HGP-write is to develop technology to enhance the synthesis and testing of large genomes. You can think of this as being: HGP step 1 - read all of the DNA in a person; HGP step 2 will write DNA in organisms. We have a collection of projects and papers that will demonstrate the value of this on a sort of one percent scale and then rapidly scale up to bring the cost down. So probably even at today's costs this could be less expensive than the original HGP 1. There is a lot of excitement about this even though it hasn't been announced yet.