Reading some accounts of synthetic biology, it seems that the ability to manipulate life is limited only by imagination. Researchers may soon be able to program cells to produce large amounts of biofuels from renewable resources, or sense the presence of toxins, or release precise amounts of insulin when the body needs it — all visions influenced by biology We can extend genetic engineering to be more like the engineering of any hardware.
From characterization of parts to design and construction of systems, there are challenges every step of the way. “There are a lot of biological factors that get in the way of engineering,” says Christina Agapakis, a graduate student working on synthetic biology at Harvard Medical School in Boston, Massachusetts. But difficult biology isn’t enough to stop practitioners in the field, who are already grappling with five key challenges.
◆Many parts are undefined
A biological part can be anything from a DNA sequence that codes for a specific protein to a promoter (a sequence that promotes gene expression). The problem is that many parts are not well characterized. They are not always tested to show that they work, and even when they are, their performance can change with different cell types or different laboratory conditions.
For example, the Standard Biological Parts Registry at MIT in Cambridge has more than 5,000 parts available for order but does not guarantee their quality, said director Randy Rettberg. Most were sent in by undergraduate students competing in the International Genetically Engineered Machines (iGEM) competition, an annual event started in 2004. In competitions, students design synthetic biological systems using parts from a “kit” or developing new parts. But many competitors don’t have the time to thoroughly characterize parts.
◆The process is unpredictable
Even if you know what each part does, the parts may not work as expected when put together, Keasling said. Synthetic biologists often get stuck in a laborious trial-and-error process, unlike the more predictable design procedures of other modern engineering disciplines.
“We’re still like the Wright brothers, putting wood and paper together,” says Luis Serrano, a systems biologist at the Center for Genomic Regulation in Barcelona, Spain. “You fly one thing and it crashes. You try another thing and maybe it flies a little better.”
Jim Collins, a bioengineer at the University of Massachusetts in Boston, and his colleagues failed several times while implementing a system called a toggle switch in yeast. His lab set up one about a decade ago in E. coli: The team wanted the cells to express a gene — call it gene A — and then used chemical signals to prompt them to turn off A and express another gene, B. But the cells refused to continue expressing B; they always switched back to expressing A. The problem, Collins says, is that the promoters controlling the two genes are imbalanced, so A overpowers B. He said it took about three years of tweaking the system to make it work. Computer modeling can help reduce this guesswork.
◆Complexity is difficult to control
As circuits get larger, the process of building and testing them becomes more arduous. The system developed by Keasling’s team, which uses about a dozen genes to produce precursors of the antimalarial compound artemisinin in microorganisms, may be the most cited success story in the field. Keasling estimates it has taken about 150 person-years of work, including discovering the genes involved in the pathway and developing or improving parts to control their expression. For example, researchers had to test many part variations before finding a structure that sufficiently increased the production of enzymes needed to consume toxic intermediate molecules.
At Berkeley, synthetic biologist J. Christopher Anderson and colleagues are developing a system that lets bacteria do the job. Modified E. coli cells, called “assembly” cells, are being equipped with enzymes that can cut and stitch together parts of DNA. Other E. coli cells designed to act as “selector” cells will pick out intact products from the remainder. The team plans to use virus-like particles called phagemids to transport DNA from the assembler to select cells. Anderson said the system could reduce the time required for the assembly phase of a BioBrick from two days to three hours.
◆Many parts are incompatible
Once constructed and placed into cells, synthetic gene circuits can have unexpected effects on their hosts. Chris Voigt, a synthetic biologist at the University of California, San Francisco, encountered this problem in 2003 while working as a postdoc at Berkeley. Voigt assembled parts of genes mostly from Bacillus subtilis into a switch system that is supposed to turn on the expression of certain genes in response to chemical stimuli. He wanted to study the system independently of other genetic networks in B. subtilis, so he put the circuit into E. coli — but it didn’t work.
“You look under the microscope and the cells are sick,” Voigt said. “One day it would do one thing, and another day it would do another.” He eventually saw in the literature that one part was greatly disrupting E. coli’s natural gene expression. “There’s nothing wrong with the design,” he said. “There’s just some incompatibility.”
◆Mutability crashes the system
Synthetic biologists must also ensure that the system operates reliably. Molecular activity within cells is prone to random fluctuations or noise. Changes in growing conditions can also affect behavior. In the long run, randomly generated genetic mutations can completely disrupt function.
Michael Elowitz, a synthetic biologist at the California Institute of Technology in Pasadena, observed cells’ stochastic abilities about a decade ago when his team built a genetic oscillator. The system consists of three genes whose interactions cause the production of fluorescent proteins to rise and fall, causing cells to flash. However, not all cells respond in the same way. Some are brighter, some dimmer; some blink faster, others slower; some cells skip a cycle entirely.
Elowitz said the differences could arise for a variety of reasons. Cells can express genes in bursts rather than steadily. Cells may also contain varying amounts of mRNA and protein production machinery, such as polymerases and ribosomes. Furthermore, the copy number of genetic circuits in cells fluctuates over time.
Despite the challenges, synthetic biologists are making progress. Researchers recently developed a device that allows E. coli bacteria to count events, such as the number of times they divide, and detect light and dark edges. Some systems have evolved from bacteria to more complex cells.
A new center for synthetic biology has been established at Imperial College London, and a program at Harvard University’s Institute for Biologically Inspired Engineering in Boston has recently been launched. Fussenegger says it’s time for synthetic biologists to develop more practical applications. He said. “Now it needs to deliver the product.”