Their seeds had lain in the Iowa dirt all winter. Now, in the spring of 2002, the little stalks pushed up into cornsilk-yellow Midwest sunlight. They extended rough, narrow leaves as if stretching to see the view. Soybeans, soybeans everywhere. Whoops.

The problem was that the corn plants, accidentally seeded by the previous year’s crop, were no ordinary weeds. Each cell held a gene coding for an experimental vaccine. Each cell was a vaccine factory, churning out proteins designed to rev up the immune system, proteins never intended to enter the food supply. And now these little factories had escaped into a field of ordinary soybeans.

The plants’ discovery started a chain of events that bankrupted their grower, the Texas biotech start-up ProdiGene. The discovery inflamed activists worried about environmental escape of genetically engineered “frankenfoods.” And the rogue corn sprouts added to a long string of hurdles that have kept cheap, easy-to-use, plant-grown vaccines — a potential godsend for developing countries — from making it to market.

“In my heart of hearts, I still think plant biotechnology can be a manufacturing protocol to make very inexpensive, orally active vaccines for the developing world,” says Charles Arntzen, PhD, the plant biologist at the Biodesign Institute at Arizona State University who invented the corn that escaped. But getting his vaccines from the lab to the clinic has been far more difficult than Arntzen anticipated when he first thought of splicing vaccine genes into plants two decades ago.

“It’s been frustrating,” he says. Arntzen isn’t daunted by technical hurdles; despite his white hair, he still exudes passion for scientific discovery. It’s what happens when the fruits of his experiments escape into the real world — figuratively and literally jumping into the mire of soybean fields, regulatory catch-22s and bad PR — that bogs him down. “Those of us who are in science want to be developing products rather than dealing with the uncertainty of public communication,” he says.

Advocates for bioengineered plants are bumping hard against increasingly cautious regulators and growing public skepticism. For example, many European nations have mandated special labels for all foods containing genetically modified organisms. Scientists like Arntzen are struggling to navigate their role in this environment.

“Scientists and companies have made pretty strong statements that genetically engineered plants are safe, and won’t escape,” says Mildred Cho, PhD, associate director of the Stanford Center for Biomedical Ethics. “But they do escape, in ways we can’t control. So scientists’ credibility is shot. We’ve kind of dug ourselves into a hole.”

Getting out of the credibility hole may require a slam-dunk biotech plant: one with big benefits and low risks, a new rendering of plant vaccine technology that the public can embrace without worry.

Arntzen has his eye on a plant-grown cancer vaccine designed at Stanford. In some ways, the experimental cancer vaccine couldn’t be less similar to the products he most wants to develop: It’s not a prophylactic against infectious disease, not edible and not designed for the world’s poorest children. But if the Stanford cancer treatment makes plant biotechnology more palatable, it may smooth the way for Arntzen’s Third-World vaccines. At least, that’s what he hopes.

Vaccination programs have made huge strides since the early 20th century, contributing to plummeting mortality across the industrialized world. In the United States, the list of childhood vaccines has grown to include inoculations against 16 infectious diseases.

But vaccination spread unevenly. One-quarter of the world’s children still do not receive the six vaccines — for measles, polio, pertussis, diphtheria, tetanus and tuberculosis — mandated in 1974 by the World Health Organization. Seventy percent of children in sub-Saharan Africa aren’t fully vaccinated.

“The problem is that vaccination is relatively expensive, at least when compared with per-capita health expenses for third-tier countries,” says Yvonne Maldonado, MD, a professor of pediatrics at Stanford who studies vaccine delivery in the developing world. A significant portion of the cost covers clean needles, well-trained health-care workers and working refrigerators — expenses we hardly think about in the United States. In places where that infrastructure is missing, administering injected vaccines becomes a daunting task.

“Transporting a vaccine is like scooping a scoop of ice cream in California and delivering it to a remote village in Tanzania before it melts,” says Nicole King, spokesperson for the Global Alliance on Vaccines and Immunization, a public-private partnership of stakeholders in immunization including the World Health Organization and UNICEF.
“It really begs the question: Do we need to vaccinate in the traditional manner in these places?” says Maldonado. Like Arntzen, she wants to see new vaccines that can be given without needles. (While studying measles vaccination in Mexico, Maldonado and a colleague used to joke that “if we could put vaccines in Coke, we’d have it made.”) Arntzen’s idea of using plants to grow heat-stable oral vaccines could be very useful. But it isn’t the only option, she notes: Other scientists are testing immunizations delivered by inhalation and skin application, for instance.

“We need innovation, we need marketing, we need technical capacity,” says Maldonado.

How plant-grown vaccines work

Arntzen hit on the idea of leafy vaccine factories after engineering plants to produce insecticide-like proteins that thwarted hungry bugs. Why not use the same molecular tool kit, he thought, to elicit immune responses in plant-eating people?

It was an “aha” moment, the inspiration for a radical new way of making cheap vaccines. Immune-stimulating proteins derived from common pathogens would be incorporated into plants that people could eat like ordinary foods. A person would consume a few vaccine “meals” on a specific schedule, timed like an initial injected immunization and booster shots, and get immune benefits similar to those conferred by an injected vaccine. The biotech plants could be grown anywhere, and storing and delivering the vaccines they contained would be no more complicated than storing and delivering food. Plants have built-in mechanisms for stabilizing their proteins, so vaccine proteins grown in plants would be much sturdier than heat-sensitive, conventional vaccines.

In early experiments, Arntzen spliced into potato plants genes that coded for proteins from the surface of Norwalk virus, a common cause of acute gastroenteritis, which manifests as severe diarrhea and vomiting. His research team incorporated Norwalk genes into small, circular pieces of bacterial DNA called plasmids, and put the plasmids into a plant bacterium. They used the engineered bacterium to infect potato-plant cells with Norwalk genes, then grew the infected cells into mature plants that churned out Norwalk surface proteins. Mice fed the resulting raw potatoes developed antibodies to the virus. Later, the team tried the same gene-splicing technique with E. coli genes and fed the potatoes to humans, observing an increase in E. coli antibodies in the subjects’ intestines and blood.

“In principle, everything we tried early on worked,” Arntzen says.

His bioengineering technique produced fertile plants that had vaccine genes in every cell. After the initial gene-splicing procedure, new generations of vaccine-making potatoes could be grown like ordinary plants.

Excited, Arntzen proposed making vaccines for cholera, tetanus, diphtheria and hepatitis B, diseases for which the developed world already had effective injected vaccines. He pictured incorporating vaccine genes into plants such as banana and papaya — foods that are often grown in developing countries and are typically eaten raw.

Regulatory and ethical hurdles

And then, after several promising proof-of-concept experiments, Arntzen’s plant vaccines began hitting bumps. First, his team discovered it was impossible to ensure every engineered plant produced a uniform quantity of vaccine. Some plants cranked out a bit more, some less.

The researchers solved the problem by harvesting and freeze-drying their vaccine-producing plants, and mixing the freeze-dried plant parts together in large, homogenized batches. Each batch was tested for potency, then packaged in gelatin capsules that delivered uniform doses. This solution was less elegant than feeding a person a vaccine-containing banana harvested from a nearby tree, but it improved the product without jacking up the price. It was the sort of scientific solution Arntzen enjoyed working out.

The next step in getting plant-grown vaccines to market was running a series of clinical trials. The U.S. Food and Drug Administration gave the team permission to test an oral, plant-grown hepatitis B vaccine in people who had already received the injected vaccine. This type of study, a “boosting trial,” tested whether the oral vaccine boosted participants’ immune responses to hepatitis B beyond the immunity conferred by the injected vaccine alone.

The experiment worked. But then Arntzen’s team began encountering problems scientific ingenuity couldn’t resolve. The FDA indicated it wouldn’t allow a subsequent, more rigorous “priming trial” (giving the oral vaccine to people who received no vaccine previously) because they feared it could endanger study volunteers by lessening the effectiveness of later doses of injected vaccine. Unable to fully develop the hepatitis B vaccine in the United States, the researchers sought permission to run vaccine trials in a developing country where injected vaccine was unavailable. Getting regulatory approval for an overseas priming trial proved problematic — Arntzen’s team has been unable to obtain funding for the needed studies.

“There is a very strong ethic that we must first test vaccines on U.S. populations,” Arntzen says. Regulators were wary of medical trials conducted exclusively in developing countries because such trials “would be exploiting the poor and uninformed,” he says. Given that his ultimate goal was to help the poor, he believes the decision to evaluate the trial in the same way as a test of a new product for U.S. markets was misguided. “My personal feeling is that that’s an arrogance bordering on racism,” Arntzen says.

Regulators may indeed have taken too narrow a view of Arntzen’s proposed experiments, says Stanford bioethicist Cho. Running overseas trials of oral vaccines grown in plants would not be ethically equivalent to testing a new injected vaccine, she says. “The piece of evidence that this isn’t being done to exploit people is that it’s not being tested in the Third World for use in the U.S.,” says Cho. “That’s not the eventual market. We already have our vaccine.”

Around the same time, in the summer of 2001, another version of Arntzen’s vaccine technology was being field-tested in the United States. Three of Arntzen’s 17 plant biotechnology patents had been licensed by the Texas biotech company ProdiGene, which planned to develop and market the resulting products. Working independently from Arntzen’s research group, ProdiGene gained approval from the Department of Agriculture to grow test plots of corn engineered to contain a vaccine against transmissible gastroenteritis virus, a pig disease.

ProdiGene’s USDA permit carried restrictions designed to prevent its bioengineered plants from mixing with other crops nearby. The company was expected, for instance, to examine their test plots after harvest and promptly remove any “volunteer” sprouts that came up on their own from among the following season’s crop. However, in the summer of 2002, USDA inspectors found that ProdiGene failed to remove volunteers from two soybean fields the company had used the previous year. The USDA also found ProdiGene corn mixed into a 500-bushel load of soybeans. The USDA supervised the destruction of the contaminated soybeans and ordered the company to pay a $250,000 penalty and reimburse them another $2.7 million for the cost of destroying the soybeans. Then in 2007, ProdiGene was found responsible in a similar incident of rogue corn plants. The company was ordered to pay a second civil penalty, which bankrupted it. Additionally, ProdiGene and its primary owners were prohibited from ever again applying for a permit for field trials of genetically modified organisms. Another branch of Arntzen’s research had dead-ended.

“I would ask: What would the public think about the possibility of vaccines in their corn products?”

ProdiGene’s problems raised the ire of activist groups worried about the unintentional escape of genetically modified crops into the food supply. Jane Rissler, PhD, a senior scientist with the nonprofit Union of Concerned Scientists, points to the ProdiGene incidents to illustrate why her organization recommends a moratorium on growing pharmaceutical products, such as vaccines, in food crops raised outdoors. Corn and other food crops cross-pollinate easily, and the risk that vaccine genes could jump into nearby fields of food is too great, she says.

“I would ask: What would the public think about the possibility of vaccines in their corn products?” Rissler says. The Union of Concerned Scientists doesn’t think the USDA capable of enforcing an adequate regulatory program to ensure that bioengineered food crops are appropriately contained when grown outdoors, she says. “It is unwise, on the face of it, to do this work in food crops.”

Not everyone shares Rissler’s view that food crops shouldn’t be used to grow vaccines, however.

“I personally wouldn’t go that far because there are advantages to producing some of these proteins in food crops,” says Robert Peterson, PhD, an associate professor of land resources and environmental sciences who studies biological risk assessment at Montana State University. Growing vaccines in food plants may produce safer vaccines, Peterson says. The plants are safe to ingest, so vaccines grown in food plants require little purification. In contrast, vaccines grown in nonfood crops might become contaminated with plant toxins. In general, Peterson advocates evaluating the risks of raising plant-grown vaccines in open fields on a case-by-case basis.

And one representative of the USDA says the agency’s regulatory process is adequate.

“We have not had a case where any of these products [vaccines or other pharmaceuticals] have ended up in a food,” says John Cordts, a risk assessor with the USDA’s Biotechnology Regulatory Services. The fact that ProdiGene was caught is evidence the system works, he says. “We feel under current regulations, our amount of oversight is adequate to allow this work to continue outdoors.”

But Arntzen has grown weary of the negative attention his work attracted. His research group has stopped trying to produce vaccines in food plants, and is now working to engineer vaccines into tobacco. (They intend to purify the vaccines out of the bioengineered tobacco, removing all tobacco toxins from the vaccine preparation, and give the vaccines as a nasal spray.) The lab’s directional switch was driven partly by scientific advances, but public perceptions of their work also played a role. Asked if ProdiGene’s problems influenced his decision to steer away from his original concept of edible vaccines, Arntzen says, “Yes, it definitely did — not because growing vaccines in food plants was technically a problem but because it aroused such public concern. If you can find another way around the brick wall, there’s little point in trying to pound down the brick wall.”

And the dance between plant biologists, regulators and the public continues. Arntzen is optimistic about plant biotech advances like Stanford’s experimental cancer vaccine being developed by professor of oncology Ronald Levy, MD.

“I still tend to be a Pollyanna,” Arntzen says. “I’ve spent so much of my career on projects in the developing world.” Even though progress is frustratingly slow, he still sees advances in plant biotechnology as a solution for the developing world’s vaccine crisis.

Levy’s vaccine has two key characteristics that could help its public profile: It’s not grown in a food crop, and production will likely be restricted to greenhouses. Arntzen hopes acceptance of this product could pave the way for acceptance of inexpensive plant-grown vaccines for the world’s poor.