What Kind of Artificial Life is in our Future?
by Alan S. Brown
Scientists are reengineering simple organisms and trying to build new cells from scratch. This could revolutionize our relationship with the natural world.
The Author
The Researchers
SCIENTISTS ARE RAPIDLY MOVING TOWARD CREATING new cells, the basic unit of life, unlike any found in nature. Some researchers are already modifying the genetic code of existing cells beyond recognition. Others have launched ambitious projects to try to build living cells from non-living biomolecules.
Their goal is to understand what distinguishes life from a mere sequence of biochemical reactions and also to probe the mechanisms that support life. They hope to use this knowledge to create new cells than can replace injured ones, withstand the impact of climate change, or produce more effective medicines and greener fuels.
A roundtable hosted by The Kavli Foundation brought together three pacesetters in this rapidly emerging field. We asked them about a future in which humans program cells the way we now program computers. We also discussed the science behind their work, the promise and risk of synthetic organisms, and—ultimately—what it means to be alive.
The participants were:
-
JEF BOEKE—is a molecular biologist, geneticist, and founder of the Langone Health’s Institute for Systems Genetics at New York University. He leads the 11-university Synthetic Yeast 2.0 consortium, which aims to learn how to modify the yeast genome to produce medicines, liquid fuels and valuable chemicals. Boeke is also a director of the Genome Project-Write, which is developing new ways to synthesize the genomes of microbes, plants and animals.
-
MARILEEN DOGTEROM—is a biophysicist and chair of the Department of Bionanoscience at the Kavli Institute of Nanoscience at the Delft University of Technology, which seeks to apply nanoscience to the study of biology. She heads The Netherlands’ 10-year, $25-million Building a Synthetic Cell (BaSyC) program, which aims to build a living synthetic cell by starting with non-living biomolecules.
- PAMELA SILVER—is a professor of biochemistry and systems biology at Harvard Medical School, a member of the Wyss Institute for Biologically Inspired Engineering, and a member of the Kavli Institute for Bionano Science & Technology at Harvard University. Silver is a co-inventor of the “bionic leaf,” which uses genetically modified bacteria to convert sunlight, carbon dioxide and water into liquid fuels, fertilizers, medicines and chemicals.
The following is an edited transcript of their roundtable discussion. The participants have been provided the opportunity to amend or edit their remarks.
The Kavli Foundation: When we talk about artificial life today, we are talking about synthetic cells. Professor Dogterom, how do you even begin to make a cell?
Marileen Dogterom: There are two ways to think about making a synthetic cell. The “top-down” approach begins with something that is already alive. Then researchers modify its genome—its DNA—so extensively that it becomes a different kind of cell entirely. For example, the Venter Institute in the United States started with a live bacterium and eliminated every gene that is not essential to life. The resulting cell has only 473 genes, about half the number initially found in the original bacterium. It also has a very different set of behaviors. So the genetics and behavior of these new bacteria are unique and not found in nature.
I’m leading a project called Building a Synthetic Cell (BaSyC), which instead takes a “bottom-up” approach. We are starting with natural biomolecules and trying to combine them into the biological structures and systems of living cells. We are still a decade or more from realizing this vision, and do not expect our initial synthetic cells to do everything a natural cell can do. But if we succeed, our cells will be able to do at least three essential life-sustaining tasks: First, convert food into energy to make complex molecules out of simpler ones; second, use DNA to transfer the information needed to build proteins that keep the cell alive; and third, grow and reproduce.
Jef Boeke: The analogy I use to explain the difference between the two approaches involves a computer. A computer needs hardware and software—a set of instructions—to make it work. Marileen’s project is trying to build a computer—her synthetic cell—by throwing together all the hardware and all the software needed to perform the basic functions of life.
Pam [Silver] and I are starting with the existing hardware—a living cell—and reprograming its software—the DNA—to give it a different set of instructions. That is a really, really important distinction between our two approaches. We always start with a living cell and put new DNA code into it. That’s why I don’t call our cells “synthetic cells.” They are really cells with a synthetic genome.
TKF: Why create synthetic cells or other types of artificial life? Could they change our lives?
Pamela Silver: Yes, they could, and in many possible ways. I’ll give you some examples. My team engineered gut bacteria to produce a medicine when they sense inflammation in the gut, so they could be a way to deliver medicine. We’ve also engineered naturally photosynthetic bacteria to turn carbon dioxide, a greenhouse gas, into plastic precursors and other useful molecules. We call this the bionic leaf, and there’s a lot of interest in using it to do green chemistry.
But I think food plants are an even more important target. Humanity has a big problem: feeding a world of 10 billion people. I want to help solve that problem by changing plant genomes so that they thrive in a wider range of climates.
Cells are really living mosaics, made up of different, specialized components or systems. There’s one for generating energy, and another for storing and organizing genetic information. As we understand more and more about how those systems work within cells, we will be able to modify or replace them in predictable ways. This enables us to do useful things, like the ones I just mentioned, while also adding to our fundamental knowledge of how cells work.
Boeke: I agree with Pam. My research consortium works with yeast, and we want to modify it to do similar things, like produce useful chemicals, medicines, and even foods.
Dogterom: Building a cell from scratch is a more basic research project, and our main motivation is a bit different. We want to learn about real cells by trying to build one. The result doesn’
t have to look like something that is really alive. But it does have to teach us about how the dif
ferent components of a cell communicate to make things like growth and reproduction happen.
Silver: So, you see this as basic research as opposed to something that has a direct application?
Dogterom: Yes, though what we discover will have many potential uses. Synthetic cells may one day act as microbial factories to produce chemicals, medicines or even drug delivery systems. We may learn to engineer cells differently than we do now and use those cells for entirely new purposes. I think those options are both open.
TKF: Let’s go back to something we were discussing a few moments ago. You’re changing the genetic instructions in cells like yeast. How does that approach differ from the type of gene editing that researchers and biotech firms have been doing for 20 years?
Silver: In the past, researchers knew enough about DNA to insert a single gene into a cell that enabled it to assemble various biochemicals into, say, insulin. That worked really well. But fast-forward to today, and we know a lot more about biology. We can synthesize and test DNA so much more quickly and cheaply than before. We’re using this knowledge and technology to introduce multiple genes into a cell, enough to change their biological processes or systems.
The work of Stanford University’s Christina Smolke is a good example. Her goal is to synthesize painkillers that are less addictive than those we have today. She studied plants that make opioids and teased out how they do it. Then she put together a system of 21 different genes from yeast, plants, bacteria and even a rodent. When she inserted them into yeast, an organism that doesn’t normally produce opioids, it began making a new synthetic painkiller. That is where we’re going.
Boeke: Right. There’s nothing magically new about what we’re doing. It’s just a gigantic scale-up, and that has a lot of power.
TKF: Professor Boeke, you use a technique to modify yeast that essentially speeds up evolution in the laboratory. How do you direct a natural process like evolution to get the results you want?
Boeke: We developed a method to insert short bits of DNA into yeast genes. This allows us to build a million different variations of a specific gene quickly. It’s like shuffling a deck of cards, but instead of a million different hands, we make a million different genes. We can change the cards’ order, delete cards, and even cheat and put 10 jokers in the deck. If we did that with cards, some of those hands would give you the best possible outcomes for a game. With genes, depending on the biological problem you’re trying to solve or optimize, you might get several different pathways to a solution. And when you study how each of those pathways work, you can learn a lot about how the genome is wired.
Why Yeast?
Yeast has become a valuable model for studying artificial life because researchers know a great deal about it.
In fact, yeast and humanity have been intertwined since the earliest days of civilization, when people learned to use it to make bread, beer and wine. Today, yeast is still used to make food—and also everything from pharmaceuticals and flavor enhancers to biofuels, lubricants, and industrial enzymes and chemicals. It is also used to clean up the environment, for example, to remove toxic metals from mining waste.
“There are probably 1,000 different yeast labs in the world, and there are people who have studied yeast their whole life,” said Jef Boeke, who leads the Synthetic Yeast 2.0 consortium, an international effort to build a new yeast genome. “There are few organisms for which this kind of knowledge is available.”
Boeke and his collaborators have drawn on this knowledge to refine their plans for merging yeast’s 16 chromosomes into two “super-sized” chromosomes, and to design their experiments.
“I think our failure rate would have been much higher if we tried to do this for a random organism,” he said.
TKF: Professor Silver, have you used the type of high-speed discovery methods Professor Boeke just described?
Silver: Not often, but I think they represent a real transition in genomics. There are people entering this field who do not have a deep understanding of biology. Yet their ability to do high-speed biology—and Jef is probably laughing at this—perhaps transcends the need to know a lot of detailed biology. Instead, if they can shuffle a million genes and they have a good screening procedure, that might enable them to find the results for which they are looking.
But I still think there’s value in knowing some biology. I’ve been doing biology for a billion years. I draw from that knowledge to design things, and they often work pretty well.
Boeke: I couldn’t agree more with that statement. It is true that we’ve scaled up and sped up DNA synthesis in lots of ways. Still, before touching a single yeast in the lab, my colleagues at Johns Hopkins University, Professors Srinivasan Chandrasegaran and Joel Bader, and I spent about eight months debating about what we wanted to build and what that design would look like. My lab looked at every gene in great detail. Fortunately, yeast is a commercially important organism and has been studied extensively, so we had a lot of fundamental science to build on. Otherwise, our failure rate would have been much higher.
TKF: This is very different from what you do, Professor Dogterom. Can you tell us more about how you are trying to build a cell from scratch?
Dogterom: Like Jef, we start with a high-level approach to what we want to build. In the cell, that means starting with proteins. That is because proteins perform most of the work in the cells, from breaking down food to orchestrating the division of cells when they reproduce.
We have come a long way in understanding how individual proteins behave. Now, we are just starting to link these proteins into more complex systems that reproduce a cell’s basic biological functions. But we don’t fully understand the mechanisms by which they do that work. For example, in what sequence should biochemical reactions occur? Do we need some kind of three-dimensional organization within a cell to turn on the machine?
The next challenge is creating those proteins within a cell. The instructions for building proteins are encoded in DNA. So, to build a synthetic cell, we need to learn how to build DNA that encodes all the right proteins—and that also produces those proteins at the right time and the right place to carry out all the different functions done by a living cell.
TKF: Clearly, synthetic life opens the door to a clearer understanding of biology and to a range of practical applications. But what about the risks? What checks and balances are you putting in place to ensure synthetic life won’t harm the environment?
Boeke: We have ways to contain our synthetic creations. For example, biologists have known for a long time that if we delete some of the genes needed to make basic building blocks of life, the resulting organisms cannot compete in the wild. We work with these “constrained” strains in our lab, and so does pretty much everyone else in the field.
Another thing we could do is change the cell’s genome so that its survival depends on a small molecule not found in nature that only the researcher can provide. That change is called a genomic safeguard, and it prevents the synthetic organism from multiplying outside the lab.
TKF: Even so, some yeast are used in factories to produce chemicals and drugs. Couldn’t those strains get out and mate with wild yeast and potentially introduce undesirable traits?
Boeke: We recently published a paper in Nature that shows how synthetic genomes could actually solve that problem. We work with yeast, which has 16 chromosomes. We succeeded in our goal to string those chromosomes together and make yeast with two large “super chromosomes.” That yeast would have all the properties of normal brewer’s or baker’s yeast, but it would be unable to interbreed with wild yeast because of the vast difference in chromosome numbers. So, we have at least three different strategies to contain yeast.
TKF: Dr. Dogterom, do the “bottom up” synthetic cells you aim to create pose similar threats to nature?
Dogterom: All these protective strategies would work equally well with synthetic cells built from the bottom up. And in a cell that we build from scratch, there are even more possibilities of control. But these cells are unlikely to out-compete anything in nature for a long time, so it will be much less of an issue.
Boeke: It’s a problem you wish you had, right?
Dogterom: Exactly. Bottom-up synthetic cells will not have the level of sophistication to outcompete real cells in nature. Nor do I think we need to worry about mutations. Natural cells mutate all the time, yet competition between all different life forms tends to keep things in balance. The same will hold true for synthetic cells—if they ever become sophisticated enough to mutate.
Of course, you can also turn that argument around: once you have more control, you can build things that may be do harm on purpose. But I think that’s true for any application of biotechnology. Just as you can create cells that cannot be invaded by viruses, you could design viruses that bypass those defenses. So, there’s always a flip side to any of these arguments.
Marileen Dogterom leads The Netherlands’ Building a Synthetic Cell research program, a 10-year effort to build a living cell from biomolecules on up. (Credit: Graham Johnson, Building a Synthetic Cell)
TKF: So far, we have been talking about synthetic cells. Do you expect to see synthetic organisms that contain multiple cells in your lifetimes?
Silver: Probably not, though it depends on your definition of multicellular. I think that there already are cases where people have created artificial syncytia, which are groups of cells whose members have specialized behaviors and that work together to perform a task. That’s the basis for engineering artificial tissues and things like that.
TKF: What about a true multicellular organism, like a slime mold that consists of about 1,000 cells?
Boeke: Some people argue that yeast is a multicellular organism because some cells specialize. So, certainly, from that perspective, it’s doable. A synthetic mouse genome is not that far down the road, either. Right now, doing a whole mouse chromosome or something like that would take a lot longer and cost a lot more money. But I can imagine seeing that in my lifetime.
TKF: So, finally, what is the end goal? What do each of you hope you will be working on 10 years from now?
Boeke: I’ve been doing a lot of reading on the immune system and believe that it is the final frontier of complexity in the human body. I would hope that we would have learned something about reprogramming therapeutically useful types of human cells. Some of this is already being done on a small scale, and it’s creating a revolution in cancer therapy. But I’m imagining doing it on a substantially larger scale. I think massive genomic engineering will play a role in teaching us how the immune system can control cancer.
Dogterom: As I said earlier, we’re doing this to learn the basic principles of natural life. We want to learn the rules, so we understand how this chemistry takes place in time and space to make cellular life possible.
Silver: I’m hoping to see the conversion of the chemical industry to completely biological ways of producing any molecule on demand, as well as a fast-paced revolution in biomedicine and therapeutics.
TKF: In just 10 years?
Silver: Oh, yes, I think in 10 years. It’s already happening. I mean, Jef just said he wants to reengineer the immune system. Just the fact that we can make statements like this now because we have the tools to do it is pretty amazing.