George Church: Changing The World One Nucleotide at a Time
Maybe it’s writer’s block. Maybe I’m star struck. Or maybe that cat finally got my tongue and is hiding somewhere contentedly licking its paws. Whatever the cause, I have been thinking about, but unable to actually write, this post for days.
It all started when a friend of mine, Stephen, a post-doc at Harvard Medical School, sent me a link to a Science paper from his neighbors in the lab of Dr. George Church. In this study, Church’s group used a variety of cutting edge tools to re-write the genetic code of the bacterium E. coli.
A little background…
Every organism on our planet has a genetic blueprint, or genome, made up of a unique sequence. This sequence is made from a four-letter alphabet of DNA nucleotides; A, C, G, and T (a.k.a. U). Within every cell is the machinery required to read this code, and a crucial part of that machinery is the ribosome. This molecular powerhouse reads our genetic code in three letter chunks called codons and generates proteins that conform to the genetic blueprint. The goal of Church’s paper was to change one of these codons in every gene in such a way that a free codon was created.
Why on earth would we want to change the genetic code of bacteria? For one, a number of technical applications require proteins that include synthetic amino acids. There are 21 naturally occurring amino acids, the building blocks of proteins, and each is added to a growing protein by the ribosome depending on the specific codon being read. The genetic code allows for certain mistakes, or mutations. If mistakes are introduced into an organism’s genome at any point in its life cycle this degeneracy means they are not always a death sentence. But as you can see from the table below, every codon is used by a cell, so the introduction of synthetic amino acids requires competition with naturally occuring ones.
Another exciting application for this research is the ability to generate bacteria that are resistant to viral infection; a major area of interest in a number of biotech and food-production industries. If bacterial cells are using a slightly different genetic code than viruses, the cellular intruder will not be able to hi-jack the host’s protein-making machinery because the viral genome would simply make no sense. You could think of it as handing someone who only speaks English a German novel. The words on the paper would be made up of the same characters, but they would have no idea what the words meant.
But back to George Church.
A few weeks ago Stephen’s pal Laura, Church’s assistant, offered to get me an interview with the man himself. A couple of emails later I had an appointment, a phone number, and a list of questions that I was informed were “fearless”. (Yes, that made me a little nervous.) I wanted to know about the paper I had just read, but I was also curious about the Church lab and how it came to be the powerhouse it is today. So that’s how I started the interview….
KP: So, the obvious question, what inspired you to start your own lab? Was it an innate desire to explore genetics, or something different?
GC: Well it’s not so much starting my own lab as starting my career. You know, you are basically starting your own lab right now, right?
KP: Me? No I’m still a graduate student….
GC: No but I mean, you are, whether you know it or not, starting your own lab right now. The question really is “why did I start my graduate career?” which lead more or less to where I am now. I was always interested in Math and Biology since I was very young, and I was always looking for some combination of the two, and the first combination that really worked was crystallography. I got exposed to automation: it was one of the few fields in biology at the time that was automated…actually I think it was the only field that was automated, period. And it struck me that there was an opportunity for reducing costs and increasing accuracy by automating the rest of biology. So one thing led to another and I realized I not only had to start a lab but a particular kind of lab, so that’s where we are.
KP: One application you suggest for this research is the engineering of virus-resistant organisms. This seems like a huge leap from the single-codon changes shown in the paper. Do you see this as a much further downstream application?
GC: Yeah there are additional steps that we have anticipated. We don’t know how much we need to change the code…clearly if we change the code radically enough it will be resistant not only to one virus but to every virus, past, present and future (except for synthetic ones). Right now we’re in the process of changing 15 codons. We’ve already done one, and we’re going to do another 14. I mean, already it looks as though all of them are pretty easy to change, but I’m sure we’ll eventually run into one that’s not so easy. And so rather than bash our heads against a brick wall we’ll just drop those. It’ll be a war of attrition eventually. We’ll see how many we get…I think three would probably be enough to make it multi-virus resistant.
KP: A central premise in this genetic recoding is that bacteria use two different release factors that recognize distinct termination codons. Could you speak to the evolutionary advantage of having two different type-one release factors in the first place?
GC: What you can think of is that a release factor is an honorary tRNA and each amino acid has one or more tRNA. Some have one and some have three. And they have anywhere from 1 to 6 codons. And if you think of the STOP function there are three codons, which is right in the middle of that range. To get six codons you typically need three tRNAs, to get four you need two and to get one or two codons you need one tRNA. So for release factors to recognize three stop codons you need two. Now why you have this redundancy is an example of how nature abhors a vacuum. Let’s imagine you were super-efficient and only had 22 codons; one for each amino acid and one for STOP. Every time you got a mutation in your gene that took it off the map (eg if you had a CCC for proline but you didn’t have a CCG which in our world codes for proline but in this parallel universe doesn’t code for anything) every time you got a C>G mutation you would have a completely dead gene, whereas in a redundant code where you have lots of synonyms that CCG will still code for proline and you’re still fine, or it could code for a closely related amino acid and you’re also still fine. In this case nature avoids the vacuum because a single mutation could drastically reduce the fitness of the cell.
KP: What motivated you to mess with the genetic code in bacteria? I mean, in the introduction to the paper you give the logical reasons, but really…was there at least an element of pure scientific curiosity?
GC: We try to have every project in the lab do three things. One is to be scientifically interesting, to push some philosophical boundary in science. The second is to be practically relevant to a societal problem like energy or agriculture and so forth. And thirdly to push a technological parameter, whether or not it’s relevant. So it’s basically philosophy, relevance, and edgy technology. So this had all those. The edgy technology was to accelerate evolution and to get better at being able to write genomes at low cost and with high throughput. So that’s the technology component. The practical one started out as being able to use non-standard amino acids efficiently. In order to use non-standard amino acids efficiently you need to have a free codon out of the 64, and in most industrial microorganisms all 64 codons are already doing something. They have machinery that is essential to the cell that has all sorts of applications within the cell. So to free it up, you have to eliminate or move those essential functions somewhere else. So in our case we had to move 314 TAGs to TAA, which thereby allowed us to delete the otherwise essential gene prfA, which encodes RF1. We were also interested in generating virus-resistant organisms and organisms that were unable to transmit usable DNA into their environment. And then the philosophical question was “can you do that”? Can you change the genetic code that dramatically and still have a viable organism? And secondly can you make an organism that’s multi-virus resistant?
The Personal Genome Project
THEN, a few days after the interview, I found this:
I know, I know, I must’ve been living under a rock for the last few years to miss the Personal Genome Project. (Well, I suppose if you replace the word “rock” with “the pressures of graduate school”, then yes, I had.)
The PGP is a fascinating project. It started in 2007 with ten volunteers (one of which is author and neuroscientist Stephen Pinker, who wrote a fantastic piece on the subject for the New York Times) who agreed to submit blood and saliva samples to have their entire genomes sequenced. They also agreed to extensive phenotypic (observable characteristics) evaluation, and for all of this data to be made publicly available on the Internet.
The ultimate goal of the project is to collect genomic data from 100,000 volunteers. However by starting out with 10 subjects a few critical questions can be addressed. The first is how feasible is this kind of endeavor? Sequencing a genome, while quicker and cheaper than it used to be, still costs thousands of dollars and takes weeks to months to complete. But demand drives down cost, and research improves technologies, so the hope is that as the PGP continues to flourish these obstacles will diminish.
And then there are ethical concerns. Will knowing you carry a certain disease-causing mutation change your life, either positively or negatively? Will it affect your insurance premiums? What if you don’t fully understand the impact of what your genetic counselor is telling you? And how will projects like this affect the future of the human race? Will we be tempted to select for smarter, healthier, prettier children? These concerns are being carefully addressed by the PGP. Volunteers are required to pass an entrance exam to ensure they can give informed consent to the procedure, and the psychological and financial implications are being closely monitored.
There are currently over 1,000 participants enrolled in the PGP. The prospects for the project are extremely exciting, not only for the individuals involved, who will learn the ultimate details of their bodies, but for the human race in general.
And I can’t think of a better man or a better lab to be in charge of such a venture. As George said,
“We work more on technology than your average lab does, and by technology I don’t just mean a small engineering tweak, but re-thinking how something can be done and going for orders of magnitude changes in costs or capabilities. And sometimes that requires a commitment to engineering a very polished product; basically something that’s commercially viable, rather than just getting to the point of duct tape and rubber bands and lobbing it over the fence and hoping some company picks it up on the other side.”
To learn more about the PGP head over to the website, www.personalgenomes.org and watch a few more videos about the participants, or volunteer yourself! To read the recent publication in Science, follow this link. And to voice your opinion on this or any of the other sequencing ventures out there (such as 23andMe) you can comment on this post in the space below.
Isaacs, F., Carr, P., Wang, H., Lajoie, M., Sterling, B., Kraal, L., Tolonen, A., Gianoulis, T., Goodman, D., Reppas, N., Emig, C., Bang, D., Hwang, S., Jewett, M., Jacobson, J., & Church, G. (2011). Precise Manipulation of Chromosomes in Vivo Enables Genome-Wide Codon Replacement Science, 333 (6040), 348-353 DOI: 10.1126/science.1205822