The opium poppy may soon meet its match. Researchers in the United States and Canada report today that they are closing in on a long-standing goal of engineering a complex suite of genes into yeast that would allow the microbes to synthesize morphine, codeine, and other medicines that have been harvested from poppies since before written history began. The new work holds out the prospect of being able to cheaply and easily produce widely used medicines with new capabilities and fewer side effects. At the same time, policy specialists worry that the new yeast strains could allow narcotics dealers to convert sugar to morphine or heroin as easily as beer enthusiasts create homebrews today.
“There really is potential for screwing things up,” says Kenneth Oye, a biotech policy expert at the Massachusetts Institute of Technology in Cambridge. “If you get the integrated pathway for one-pot synthesis of glucose to morphine, that’s not controllable if it gets out. You better darn well get on top of it before that happens,” says Oye, who offers several ideas for increasing oversight of the new biotechnology in a commentary released online today in Nature.
Morphine, heroin, and other opiates produced from poppies already wreak plenty of havoc. Some 16 million people worldwide use the drugs illegally. In the United States alone, nearly 14,000 people died from overdoses of heroin and other opiate pain relievers between 2010 and 2012, according to data compiled from 28 states by the U.S. Centers for Disease Control and Prevention. Oye says the concern is that those numbers could skyrocket if dealers and users can brew their own drugs.
Opiates belong to a class of compounds called benzylisoquinoline alkaloids (BIAs), which together with related families of molecules contains some 2500 known compounds. In addition to morphine, these include thebaine, a precursor to the pain relievers oxycodone and hydrocodone, as well as commonly used antispasmodic compounds, antibiotics, and anticancer agents. BIAs are complex, multiringed structures that are difficult and expensive to synthesize in a lab. Medicinal chemists have long sought an easier and cheaper route to making these compounds, in hopes that they might find new medicines. Health professionals have also sought versions that pose fewer side effects, such as risks of suppressed breathing and addiction that come with morphine. But so far engineering opium poppies to produce new compounds has proven difficult.
“Plants have slow growth cycles, so it’s hard to fully explore all the possible chemicals that can be made from the BIA pathway,” says William DeLoache, a Ph.D. bioengineering student at the University of California, Berkeley, and lead author of the new work on engineering yeast. “Moving the BIA pathway to microbes dramatically reduces the cost of drug discovery. We can manipulate and tune the DNA of the yeast and quickly test the results.”
A long path
Efforts to insert the BIA pathway into yeast have been under way for the better part of a decade. But it’s a major challenge, says Vincent Martin, a microbiologist at Concordia University in Montreal, Canada, whose lab has been working on the project since 2009. Engineering yeast to produce morphine, Martin notes, requires adding genes to produce enzymes that carry out a chain of 15 separate chemical transformations. By contrast, one of synthetic biology’s greatest successes to date—the synthesis of the antimalarial drug artemisinin—required giving yeast the genes to carry out just five chemical steps.
In reengineering yeast to make BIAs, researchers typically divide the project up into two parts. In the first part, researchers splice in genes for enzymes that convert the amino acid tyrosine into an intermediate compound called S-reticuline; this step creates a key branching point that can lead to the synthesis of many different BIA compounds. One trail leads to morphine and codeine, while others lead to antibiotics and anticancer compounds. To create morphine, S-reticuline is first converted to a very closely related compound called R-reticuline, which is then transformed into thebaine and ultimately to morphine.
Last year, researchers led by Christina Smolke, a synthetic biologist at Stanford University in Palo Alto, California, reported that they had given yeast the enzymes needed to carry out the thebaine to morphine steps at the end of the second part of the pathway. And last month, Martin and colleagues reported in PLOS ONE that they had engineered yeast to complete all of the second-half steps moving from R-reticuline to morphine.
Meanwhile, the first part of the pathway has been harder to pull off in yeast. Going from glucose to tyrosine is easy: Yeast do that naturally. In 2011, researchers in Japan reported that they got the complete first half of the pathway to work in Escherichia coli bacteria, transforming tyrosine to S-reticuline. But to date that set of steps hasn’t worked well in yeast. The biggest roadblock has been the first step: converting tyrosine into a compound called L-Dopa. When the gene that directs that step is engineered into yeast, the bacterial enzyme works poorly at best, says Pamela Peralta-Yahya, a synthetic biologist at the Georgia Institute of Technology in Atlanta.
But John Dueber, a bioengineer at Berkeley; DeLoache; and their colleagues caught a break when they were working on a separate project to see whether L-Dopa was present in certain cells. The found that an enzyme, called DOPA dioxygenase, converted L-Dopa into a yellowish pigment. They quickly realized that they could use this enzyme as a color sensor to detect whether any other enzyme was able to convert tyrosine to L-Dopa.
Next, they teamed up with Martin and his Concordia colleagues. The group tested an enzyme from sugar beets called a tyrosine hydroxylase. That beet enzyme was able to convert tyrosine to L-Dopa in yeast, in the process turning the petri dish yellow (see image, above), the team reports today in Nature Chemical Biology. They were able to increase the L-Dopa output by nearly threefold by randomly mutating versions of the enzyme and using their biosensor to track those that worked the best.
“It’s very nice work,” Smolke says. For now, she adds, the bacteria still produce higher yields of L-Dopa than the yeast. “But it sets the stage for being able to integrate these pathways in one organism,” Smolke says.
For now, the only missing step is being able to convert S-reticuline into R-reticuline, which links the first and second halves of the full pathway. But apparently that’s close at hand as well. Researchers at the University of Calgary have posted an abstract of a Ph.D. dissertation online that says they’ve identified a plant enzyme that carries out this S to R conversion, though the work has yet to be published. Once it is, researchers will be able to insert the gene in yeast, completing the full glucose to morphine pathway. “I think it’s doable within 2 to 3 years,” Dueber says. “This area is moving much faster than we thought.”
Social and legal concerns
Given this speed, about a year ago Dueber and Martin reached out to Oye, asking if he would be willing to explore ideas for how the scientific community can prevent engineered yeast from exacerbating the illegal drug trade. In their Nature commentary, Oye and colleagues make several recommendations. First, they suggest that companies that now synthesize and distribute long stretches of DNA should consider carefully reviewing requests for the genes that code for key drugmaking components, and block suspicious requests. Such companies already undertake a similar process for gene sequences involved in microbes that could be used as bioweapons, and voluntarily report requests for such genes to law enforcement agencies.
Other possible measures would be to require researchers to engineer morphine producing yeast strains so that they also produce toxins unwanted by homebrewers, or inserting genetic watermarks into the strains to make them easier to track in the event that the strains fall into the hands of outsiders.
Although such measures may help deter criminals, any watermarking or toxinmaking genes could also be removed by a trained and skilled microbiologist, Martin notes. One other option for slowing the spread of the technology would be to request that journals not provide the full genetic details of any organisms that can complete the full transformation. (Dueber and Martin say they haven’t yet received any request to omit data.)
In the end, a new technology for producing morphine could have a profound impact on law enforcement agencies. But for now, agencies such as the FBI “aren’t recommending any specific regulatory measures,” says Edward You, a supervisory special agent for the FBI’s Weapons of Mass Destruction Directorate’s Biological Countermeasures Unit in Washington, D.C. But he says the FBI is already part of an interagency working group, which includes representatives from the National Institutes of Health and other research funding organizations, that is considering ways to keep modified yeast strands out of the hands of illicit drugmakers. And ongoing engagement with scientists and policy analysts “will definitely facilitate those discussions,” You says. “There is a window of opportunity here to negotiate the security issues.”
Still, some researchers are concerned that hype over fears of homebrewed heroin could cause a harmful regulatory backlash. “I do believe that a thoughtful discussion of risks, opportunities, and regulatory needs is important with this technology,” Smolke says. However, she says she believes Oye’s commentary, for one, was “inflammatory.” The new technology could, in the long run, bring improvements over the existing poppy-driven drug trade and all the social ills that it brings, she and others note. Lab-derived drugs, for example, could be easier to nations to regulate, and reduce the environmental damage, social unrest, and violence associated with plant-derived drugs.
Smolke also emphasizes that researchers remain a considerable distance from putting together the full chain of chemical transformations needed for yeast to make morphine. And if and when that occurs, the organisms will still make only vanishingly small amounts of the drug. “In fact, it is more likely that a person could more easily access morphine by dumping a bunch of poppy seeds in their homebrew (or tea),” she says.
The question is, how long will this remain the case?