A Grand Challenge for Genome Engineering: Virginia Cornish and Harris Wang on the GP-write Project
Department of Systems Biology bioengineer Harris Wang describes the goals of the Human Genome Project - Write (HGP-write), an international initiative to develop new technologies for synthesizing very large genomes from scratch.
In June 2016, a consortium of synthetic biologists, industry leaders, ethicists, and others published a proposal in Science calling for a coordinated effort to synthesize large genomes, including a complete human genome in cell lines. The organizers of the project, called GP-write (for work in model organisms and plants) or sometimes HGP-write (for work in human cell lines), envision it as a successor to the Human Genome Project (retroactively termed HGP-read), which 25 years ago promoted rapid advances in DNA sequencing technology. As the ability to read the genome became more efficient and less expensive, it in turn enabled a revolution in how we study biology and attempt to improve human health. Now, by coordinating the development of new technologies for writing DNA on a whole-genome scale, GP-write aims to have a similarly transformative impact.
Among the paper’s authors were Virginia Cornish and Harris Wang, two members of the Columbia University Department of Systems Biology whose contributions to the field of engineering biology have in part made the idea of writing large-scale DNA sequences imaginable. We spoke with them to learn more about what GP-write hopes to accomplish, its potential benefits, and how the effort is evolving.
How did GP-write start?
Virginia Cornish (VC): GP-write grew out of series of meetings that began in 2014 with the publication of a strategic plan for engineering biology by Nancy J. Kelley, formerly the founding director of the New York Genome Center (NYGC). After finishing her tenure at NYGC, several of us involved in the field sensed that there was a need for a new center for engineering biology that could launch a grand challenge project. Harris and I, along with Nancy, Jef Boeke at New York University, Andrew Hessel at Autodesk, and a handful of others engaged in engineering biology were there at the beginning of these discussions. At the same time, Jef Boeke and George Church were doing pioneering work in synthesizing genomes in yeast and E. coli. Eventually, an organizing meeting was held in October 2015 to discuss the project.
Department of Systems Biology faculty members Virginia Cornish and Harris Wang were among the first participants in organizing the GP-Write project.
Harris Wang (HW): As things have developed, it’s really grown into a collaborative international effort. Gene synthesis is a big deal not just in the New York or the United States, but also in Asia and increasingly in Europe. Considering the growth in the field we all thought that these various efforts should be linked together, on both the industrial and academic sides. That was the impetus behind two organizing meetings. The first was held in New York, where we talked about what the project might look like and what interesting science and technology could come out of it. This was followed up with another meeting at Harvard Medical School in May 2016 (click for video) that included more than 130 scientists, industry leaders, ethicists, and policy makers from several countries. In addition to those initial participants, the GP-write Consortium has now grown to include more than 66 additional scientists affiliated with 62 institutions in 14 countries.
Why synthesize a human genome?
HW: It’s well recognized that there has never been a grand challenge project in engineering biology. It’s kind of tautological to say so, but the benefit of deciding to go to the moon is that as you work toward this goal you actually figure out how to go to the moon. It gives you a reason to improve rocket science and learn how to sustain humans in space, for example. In a similar way, having this kind of grand challenge for engineering biology — whether in human cell lines or in other organisms — will focus the community’s attention on developing the right technologies, as well as thinking about the many important ethical, legal, and commercial implications that will need to be considered in addition to the science.
The difference between engineering biology and things like going to the moon or sequencing large genomes, however, is that design is so open ended. It’s easy to get stuck in a design cycle and to debate constantly about what you should be doing. So having a particular goal and set of design criteria to follow — such as, for example, programming cells to be virus resistant or to have more stable genomes — could help keep the engineering biology research community focused. A new Center of Excellence for Engineering Biology will also help to coordinate the effort, setting goals and timelines, and conducting reviews that will keep the project moving toward its larger objectives.
A key reason for the project is to improve technologies for synthesizing large strands of DNA. What would be some benefits of being able to do that?
VC: The idea of being able to build really big pieces of DNA is just a starting point. As a chemist I see engineering biology as the next step in our ability to build useful things. In the last century we learned how to build small molecules that were about 100 atoms in size, and this became the basis of the pharmaceutical industry, leading to the development of things like antibiotics and anticancer drugs. In the 1970s we gained the ability to clone a gene. This suddenly made it possible to make antibodies, which led to a whole new class of therapeutics. Now we’ve realized that we aren’t just limited to building bigger and bigger things in a test tube. By engineering larger and larger genomes, we can build networks of molecules that come together in complex environments like the cell to carry out new functions.
We aren’t limited to building bigger and bigger things in a test tube. By engineering genomes, we can build networks of molecules that carry out new functions.
Right now when people want to make an antibody, they make an expression vector and put it into bacteria or a mammalian cell. The point of using the mammalian cell is not the creation of the cell itself. It’s just a machine to churn out the antibody. But antibodies are finicky; they’re expensive to make and difficult to maintain. Now we’re saying why don’t we imagine that the modified cell itself is the end product.
To give one example, my lab has been working on developing a biosensor product where the sensor is a modified form of baker’s yeast. Rather than using antibodies as a dipstick test, we’re engineering yeast with receptors for biomolecules produced by different pathogens. When they come into contact with the pathogen, they change color. This provides a visible readout that could potentially be used and interpreted by non-experts for a range of public health purposes. The beauty of this kind of approach is that it is extremely cheap, since yeast can be stored and transported easily, requires no specialized equipment to produce, and reproduces all by itself. For a purpose like testing for cholera in remote parts of Africa, you could basically hand it out for free. This is just one example, but shows some opportunities that engineering biology offers.
How close are we right now to being able to synthesize a human genome?
HW: Whether we synthesize one human genome is in some ways beside the point. HGP-write is similar to HGP-read in the sense that today it’s not important that we sequenced the first human genome so much as that we now routinely sequence hundreds and thousands of genomes, in many different species. If we could start synthesizing DNA at the human genome scale, it would lead to a different type of technology platform than exists today. We don’t have one particular human genome that we want to synthesize; we want to figure out the kinks associated with how to do it so that in the future we could synthesize large DNA strands more quickly and efficiently.
Right now a lot of the debate within the group is focused on answering the question of what we are going to do. We currently don’t have the technological tools and knowledge base to prescriptively synthesize a human genome with any likelihood of success. This was similar to the situation faced by scientists when HGP-read was launched. So you have to draw a line in the sand that’s achievable and try to accomplish it.
VC: At the same time, though, anytime you have a transformative technology like genome engineering, it is going to affect human health, materials, and many other areas. This is a whole new scale at which we’re building things. And so what’s interesting to me about GP-write is that it gives us a chance to ask what we can we do with engineering biology that we can’t do by cloning one gene at time.
As the community works toward the goal of building longer strands, there are going to be breakthroughs in how cheaply you can build DNA. Eventually that will be very high impact. In the near term, however, we can still say that we’re better than we have been before at building DNA. So let’s take the technology as it stands now and see what we can do with it that’s new.
Why a human genome and not some other goal?
HW: I think synthesizing the human genome has some sort of universal resonance. The announcement coincides with the 25 th anniversary of the initial Human Genome Project, what we are calling HGP-read, so I think it’s intended to suggest both how far we’ve come and where we could imagine going in the future. It’s also intended to differentiate between reading something and actually writing something new — a more proactive endeavor that can lead directly to beneficial outcomes.
I’ve argued that maybe we don’t need to do the human genome, though we definitely need to do human-sized genomes. Another approach might be to synthesize a thousand microbial genomes — for example synthesizing the human microbiome — which for the price of one human genome could in some ways be more impactful. One could imagine doing a pilot project where you synthesize microbiomes of the 10 most prominent gut bacteria. That would be another direction that could lead to widely applicable technological breakthroughs for genome engineering in any species.
The idea of intentionally altering the human genome is something that makes many people uncomfortable. How is HGP-write handling ethical questions related to a project of this kind?
VC: Asking and discussing questions related to the ethical ramifications of engineering biology is an important part of what the group is doing. Even if we’re very limited in what the technology can do at this point, it’s important that the scientists tackle the ethical questions head on, early in the process. You have to have scientists talking to people in policy and other areas, and everyone has to educate themselves to make sure this is done responsibly.
There were two panels on ethics during the meeting in May, and from early on, ethicists have been important members of the group. Even at the first informal dinner party of 10 people where this idea first started to germinate, we had government and policy people involved. The group is interdisciplinary, and is taking the ethical considerations very seriously. We’re trying to get out ahead of them and to involve the broad range of people who need to be involved.
"We need to have responsible dialogue. HGP-write is meant to be the start of a dialogue, not its end."
HW: Even without a project like HGP-write, engineering biology will continue to develop. So if we look down the road several years when more powerful technologies will become feasible, an important part of the ethical discussion is to decide on what technical questions we need to answer. If someone were to engineer new function into human germ line cells (which is not part of HGP-write, I should emphasize), what are things that are doable, what things could have negative effects, and what are things that would be impactful in a useful way?
The ethical concerns are valid and need to be debated rigorously. At the same time, there were some criticisms of the May meeting that several of us thought were a bit unfounded, particularly because the only reason it was closed was because the paper in Science was still under embargo. Scientists need to be able to come together and figure out what we really want to talk about before opening the discussion up to a larger forum. We need to have responsible dialogue. HGP-write is meant to be the start of a dialogue, not its end.
In practical terms, how is GP-write going to work?
HW: At the most recent meeting, several scientists proposed pilot projects that would begin to serve as a reality check for what’s doable and not doable. This was also the approach used in the original Human Genome Project. There was a huge debate at the time about whether they should just sequence the exome — the sections of the genome that code for proteins — and not the whole genome. In the end they started with the 1% that was the human exome.
For HGP-write my lab put together a proposal to synthesize a prototrophic human genome; that is, a genome that is able to synthesize all of the compounds needed for growth. The idea comes from the fact that human genomes, and all mammals, are missing biosynthesis genes associated with 9 out of the 20 essential amino acids. That’s also true for a variety of vitamins, such as vitamin A, as well as certain fatty acids and small molecule metabolites that are necessary for our diet. The interesting scientific question is why we don’t have these pathways. They are only found in bacteria and plants, and so all of the world’s amino acids are derived from those two sets of organisms. They percolate up to us through the food chain.
Our approach is to ask what the technical challenges would be in doing metabolic engineering for human cell lines in this context. What technologies are needed to install pathways of this size and get them working? And what are the implications for cell physiology?
This could also potentially have relevance for human health, as there are lots of links between metabolites and cancer propagation. Dennis Vitkup, another member of the Department of Systems Biology here at Columbia, has been working on cancer metabolism, which got me thinking about this. Aberrant cancer cells basically make their own food to propagate. What would happen if normal cells did as well? This kind of project is also compelling because new technologies are starting to put synthesis at this scale within reach.
What kinds of obstacles does GP-write still face?
HW: GP-write will develop and change over time. We need to develop a road map and organizational timeline. The project will need to be organized in a manner that will coordinate the work of many institutions, as well as manage financial support from international governments. It has already received strong expressions of interest from some US agencies and several foreign governments, which has been encouraging, as this is something that you need a lot of support for, either through government funding or private investment. We also need to fill in the detail around the desired benefits spelled out on the project’s website, particularly for the pilot projects, on which work has already begun.
Irrespective of these challenges, GP-write will build on the knowledge and technological advances of HGP-read, and could be an equally transformative next step. Potential applications that might arise from GP-write could have significant impact on many global problems that we are currently facing, including human health and lifespan longevity.
— Interview by Chris Williams
Boeke JD, Church G, Hessel A, Kelley NJ, Arkin A, Cai Y, Carlson R, Chakravarti A, Cornish VW, Holt L, Isaacs FJ, Kuiken T, Lajoie M, Lessor T, Lunshof J, Maurano MT, Mitchell LA, Rine J, Rosser S, Sanjana NE, Silver PA, Valle D, Wang H, Way JC, Yang L. Genome Engineering: The Genome Project-Write. Science. 2016 Jul 8;353(6295):126-7.
The Center of Excellence for Engineering Biology
Visit the center's website for more information about GP-write as well as the latest news related to the project.
Video of the May 2016 GP-write meeting
View presentations from the May 2016 meeting of the GP-write project in their entirety. Topics discussed include an overview of the state of the art in engineering biology, presentations of 1% pilot projects, infrastructure development, industry engagement, and regulatory and ethical considerations.