Stopping the Cane Toad

Jackie Pallister is standing in a stairwell naked and barefoot except for the terrycloth bathrobe she pulled from a hook outside her airlock. She shuffles down the cool tile floor and through a warren of secure laboratories. She steps into a small dressing room and sheds her robe before climbing through an airlock into a much warmer room, almost tropical, where she dons an orange jumpsuit and a pair of rubber boots. These are all precautionary measures that Pallister must take to keep a deadly virus from escaping the Australian Animal Health Laboratory (AAHL) here in the coastal town of Geelong, just south of Melbourne.

Jackie Pallister in the lab with young cane toads.

However, the virus here cannot hurt Pallister. It's not a threat to the rest of humanity either. In fact, this attenuated Ranavirus cannot kill the dime-sized toads bobbing up and down in the algae-lined tub where she has just tossed a handful of baby crickets. These are cane toads, one of Australia's most destructive invasive species. Advancing westward at 50 km a year, these toxic amphibians have been conquering cattle ponds and swamplands from Cairns to Darwin, threatening the survival of Australia's unique predators: from frog-eating goanna lizards to a cat-like marsupial known as a quoll. One day, Pallister, a researcher at Australia's Commonwealth Scientific and Industrial Research Organization (CSIRO), hopes to insert just the right gene into this Ranavirus to turn it into the perfect cane-toad assassin. At least, that is the dream.

"It's a never-ending battle," says Pallister, who is still euthanizing her toadlets with the slice of a scalpel in order to hunt for an immune response to her proof-of-principle viral treatments. Today, she scoops up more than a dozen toadlets from these tanks to the taunts of music piped through the emergency public-address system - part of a 24-hour broadcast that is supposed to add a little bit of cheer to these windowless labs. Pallister has been stripping off her street clothes to come to work in this building for the last six years, and CSIRO has been pursuing viral biocontrol of the cane toad since the early 1990s. "There is no easy solution," says Pallister. "If there was, it would already be done."

Meanwhile, competing strategies for controlling the cane toad in both the high-tech and low-tech realm are gaining favor among scientists and policymakers, and an unconvincing outcome to CSIRO's latest trials of a modified Ranavirus could spell the end of what some believe to be a multimillion dollar boondoggle. Rick Shine, a herpetologist at the University of Sydney who has studied the ecological impact of cane toads, is one such skeptic. "I think it's vanishingly unlikely that the Australian public would allow genetically constructed microorganisms that are designed to kill [toads] loose in ecosystems," he says. If and when a release actually takes place, it may be too late to reverse the damage caused by 75 years of Bufo marinus in Australia.

The invasion of the cane toad is itself a parable of biocontrol gone wrong. In the early 1930s, the grub of the native greyback beetle was plaguing Australia's sugar plantations. Reginald Mungomery, a 33-year-old entomologist at Queensland's Bureau of Sugar Experiment Stations, had heard about the South American toad's success at controlling other beetle species in Hawaii and Puerto Rico, and he longed to bring it to Australia. He recognized that the Australian beetle would be a formidable adversary, because it spends most of its time in trees and out of the toad's reach. Even so, Mungomery's biggest environmental consideration was the possible nuisance of the toad's high-pitched trill. "Their call is not objectionable," he reasoned in 1934, "and certainly not as loud nor as shrill as that of many species of frogs which are indigenous to Australia." One year later, he traveled to Hawaii to collect 102 cane toads for Queensland's first breeding colony.

The cane toad is anything but cute. Its skin is sallow and warty, hanging off its body like an oversized Halloween costume. The toad will proudly urinate on you if you try to pick it up. A full-grown cane toad can weigh nearly 1 kg at the extreme, and while they soon demonstrated a prodigious appetite for native insects, they did not control the greyback cane grub. Nevertheless, Mungomery was undaunted and continued to lobby for additional releases of the toad throughout cane country, because cane farmers had yet to discover an alternative. The toads were soon pumping out more than 30,000 eggs each breeding season, and they spread westward, largely unmonitored, for the next 50 years.

The cane toad does more than devour insect prey and out-compete native wildlife. When injured, toads secrete a milky blend of neurotoxins, called bufotoxin, from the parotid glands behind their heads. Australian fauna evolved in isolation over an estimated 45 million years and, consequently, lacks defenses against the bufotoxins produced by toads around the world. When the cane toad hopped through Kakadu National Park, a World Heritage Site, in 2004, they turned waterholes into ghost towns. Freshwater crocodiles and water monitors that munched on the toads vanished. Populations of the northern quoll have been decimated, and researchers have already established several populations on off-shore islands in hopes of staving off species-wide extinction. The toad is also a backyard menace; hundreds of the beasts can be found congregating under porch lights in parts of Queensland. A golden retriever that playfully catches one will end up convulsing on its back; death occurs by respiratory arrest.

Current and projected range of invasive cane toads in Australia. Cane toads now inhabit approximately 1.2 million km2 of Australia (dark orange). The predicted distribution of cane toads (light orange) based on annual maximum and minimum temperature, and other factors. The future range area of cane toads predicted by this model (2.0 million km2) is almost triple the projections from the most recent model (0.7 million km2) based on its native limits ( Proc R Soc B, 274:1414, 2007).

Recent studies have shown that the toads are now moving westward at a rate of as much as 50 km a year. Shine's group has found that the benefit for a male colonizing a new waterhole is so great that advancing toads are evolving longer legs, which allows them to hop westward more quickly. He has also found that some are hopping quickly enough that they leave their parasites behind, while others are moving so far and fast that they are developing arthritis in their overtaxed joints.

Some observers thought toads would slow as they moved into drier regions of western Australia, but Lee Scott-Virtue, cofounder of the Kimberley Toadbusters in Kununurra (see Sidebar, "Toadbusting"), now believes that toads have learned to absorb moisture by burying themselves in cow manure, enabling them to withstand drier climates. Although the extent of their ecological impact is still a matter of debate, a 2007 study uses global warming predictions to show that, with a few gaps in their distribution, cane toads may eventually encircle the continent.

A waterhole, a typical gathering spot for the toads.

Louis Pasteur was among the first to propose biocontrol for Australia's other hopping invader, the European rabbit (Oryctolagus cuniculus). Although his scheme to release chicken cholera in 1888 was not enacted, Australia released its first rabbit biocontrol agent, the myxoma virus, in 1950. Combined with the calicivirus, which was inadvertently released in 1995, Tony Peacock, the CEO of Australian's Invasive Species Cooperative Research Center, estimates that biocontrol has saved Australian farmers and ranchers $4.5 billion in damages. "Australia is pretty much the only place where biocontrol has worked on a vertebrate," he says. The geographic isolation that exacerbates Australia's pests has also proven to be a boon for biocontrol efforts. Foreign species may be able to propagate once released from ecological pressures in their homeland, but their evolutionary distinctness leaves open the possibility of fighting back by identifying foreign diseases that are unlikely to jeopardize native wildlife.

Unlike rabbits, which eat vegetable crops and livestock forage, cane toads have yet to make a dent in the country's rural economy, so interest in stopping them developed only with the country's growing ecological awareness. In the 1980s, cane toad paperweights could be readily purchased in Queensland souvenir shops, but few seemed to notice, or protest, when the toad crossed into the Northern Territory in 1982.

The country was galvanized, in part, by the 1988 release of the documentary Cane Toads: An Unnatural History - a lighthearted look at the cane toad problem that featured one angry driver intent on squishing as many of the buggers as possible. In July of 1989, a journalist wrote in Sydney's Sun Herald that while CSIRO had spent millions of dollars studying the lactation of marsupials, the agency "has never spent a cent on cane toad research - nor does it have any plans to do so." She pointed out that one meagerly funded research project was coming to a close without plans for renewal. That same meager project would ultimately spark this 20-year quest to kill the cane toad with a virus.

Since 1987, Rick Speare, a parasitologist at James Cook University in Townsville, had been hunting for native infectious agents that might be used for biocontrol of the cane toad. Speare's studies showed that cane toads played host to a menagerie of pathogenic bacteria, fungi, protozoa, and tapeworms, yet none of these agents caused serious disease. His project came to a close in 1990, but that same year, the federal government gave CSIRO $1.5 million AUD (about $1 million US) to continue the hunt for another three years.

CSIRO researcher Alex Hyatt, Speare's collaborator and an expert at identifying viruses using an electron microscope, went to Venezuela to establish collaboration with the country's national research institute. On an early trip, researchers collected 206 cane toads from both the eastern and western slopes of the Andes, and took samples of their livers, lungs, kidneys, and spleens. These specimens were shipped back to Australia where Hyatt was able to isolate and identify a half-dozen viruses in the diverse iridovirus family, including a few in the genus Ranavirus, which is specific to frogs and fish. ¹ "Everyone was very excited about that," he says, because it meant that there might be a pathogen that would kill the cane toad and only the cane toad. In 1993, CSIRO received another $2 million AUD ($1.4 million US) for four more years of research.

A solution seemed to be near. Hyatt ran some trials at AAHL and found that the viruses killed 100% of their cane toad tadpoles, which is the most susceptible stage in an amphibian's life cycle. "That made everyone more excited," he says, although he was already harboring reservations that the foreign Ranavirus could infect native frogs as well. So, before the project's funding elapsed in 1997, Hyatt's group tested the virus with tadpoles from the white-lipped tree frog, an icon from Australia's rainy north. It killed them, too.

This final experiment was a major letdown for the effort, although there was an upshot: Through this work, Hyatt and his international collaborators were the first to recognize that the chytrid fungus was playing a role in declines of amphibians around the world. ² Hyatt's group still ships their diagnostic test for the fungus to laboratories around the world.

Around the same time that Hyatt was working in Venezuela, Speare isolated a new virus from the ornate burrowing frog living along the Bohle River in Queensland. As a native Ranavirus, it soon caught the interest of Hyatt's group. The Bohle virus is a large double-stranded virus, making it relatively easy to manipulate and insert foreign DNA. It wasn't specific to cane toads, of course, but perhaps the ideal virus didn't exist. If the team couldn't find a virus, they realized, they might have to build one. In the late 1980s, crop scientists had successfully tweaked species-specific baculoviruses so that they would survive longer in the field and kill their insect hosts more quickly. The Australian plan to create species-specificity de novo and to unleash their creation in a vertebrate population was unprecedented.

More than three years would pass before the project received renewed political and financial support. During that time, the cane toad pressed on. By early 2001, the colonizing front had crossed into the southern end of Kakadu National Park's 36,000 square kilometers of swamp-laden savannah, where Aboriginal people have inhabited continuously. When the cane toad arrived, residents were still supplementing their diet with traditional "bush tucker," ranging from file snakes to goanna lizards. Today, goannas are no longer on the menu.

The tile walls of Jackie Pallister's virology laboratory at AAHL are decked with posters of wild places from around the globe. Pallister, who was born in England, researched a vaccine for feline infectious peritonitis at the former Upjohn Company in Kalamazoo, Mich., before moving to Australia. "I've been to some of those places in Western Australia," she says, referring to the rugged Kimberley region still untouched by cane toads. "I've camped on these rock shelves where there are so many frogs you can hardly move. ... To lose that kind of diversity is not something I would like to see happen."

Pallister joined the cane toad project in 2001, after the Minister for the Environment and Heritage, Senator Robert Hill, resuscitated it with a promised $1 million AUD over two years ($600,000 US). Pallister began leading the construction of the recombinant viruses. Tony Robinson at the Black Mountain Laboratory in Canberra led a separate team that started doling out organic lettuce to their tadpoles ("taddies" in Pallister's parlance) and focusing their science on toad genetics. The two groups collaborated early on to identify the best method of attacking cane toads at the tadpole stage. They hit upon an elegant 1969 Science paper by George Maniatis and colleagues, then at the Massachusetts Institute of Technology. ³

Using North American bullfrog tadpoles, Maniatis had found that he could elicit a deadly immune response during metamorphosis by inoculating tadpoles with adult hemoglobin. "There are a lot of critical processes going on at that stage," says CSIRO scientist Damien Halliday. "If you tip the balance, they could be in real trouble." Some early experiments with inoculating toads after a fixed number of days in their life cycle seemed to indicate that the toads were not making the switch to adult globin.

This experiment would prove less useful than the cane toad researchers had hoped, because a bullfrog in North America is quite different from a cane toad in Australia. Bullfrogs have large tadpoles that take as long as two months to metamorphose, whereas cane toads undergo metamorphosis at a small size in just more than a month. The team also found that tadpoles in the same cohort were developing at wildly different rates. "When you start looking into metamorphosis," says Halliday, "it's a huge topic, and we started learning things for the first time."

At CSIRO's Black Mountain Laboratory, Halliday pulls up a slide with a side-by-side comparison of a tadpole and an adult frog, their intestines dyed fluorescent green. During metamorphosis, he explains, as he traces the gut with his index finger, the amphibian digestive tract shortens by almost 70%, transforming from a series of coiled loops into the well-defined stomach and intestine that characterizes the adults. If the team could stick a monkey wrench in that process, they might be able to stop the cane toad.

This is just one of the targets that Halliday, an expert in immunology and genetics who joined Robinson's team in 2004, has been examining. Over the last several years, he has built a subtracted cDNA library and used microarrays to identify differences in gene expression in adults and tadpoles, allowing him to build "a top-ten list" of targets. He also has been considering a rational gene target - a target generated a priori based on the toad's natural history. "If you're targeting an invasive species, you look at what's different about it or why it's doing well," he says. "The cane toad does well because it's toxic and has no natural predators."

The cane toad toxin blocks a sodium-potassium pump, so Halliday knew that the species must have evolved a specialized region of its ATPase that made it immune to its own toxin. If Halliday could silence this ATPase with RNA interference, the cane toads would, in effect, poison themselves. In addition, because this region of the genome is specific to toads, Australia's native frogs would be safe. His preliminary experiments in 2005 showed that it was possible to silence the ATPase, but he needed to wait for his gene-expression data to make sure that the gene continues to be expressed throughout development.

In the meantime, Pallister's team was honing the Bohle virus into the perfect vessel for its customized genetic package. ⁴ Raising the pathogen in monkey kidney cells, they selected for a less virulent strain, but they needed to prove that this strain could still get inside toad cells, a key step in making their biological control specific to cane toads. The molecular machinery of the virus had to remain functional, but its contents should be no longer pathogenic to amphibians. Later, the team would be able insert a genetic sequence that would target only toads. The challenge at this stage was to prove that the harmless virus was actually doing its job.

Cane toads press their way into Australia's arid west during the yearly monsoon season, but as the waters recede not all of the hopping migrants can find a moist refuge. This cane toad, however, was likely hit by a car and died in a dry stream bed.

When Pallister began this phase in 2002, conventional PCR was still not sensitive enough to give her an answer. She would bathe tadpoles in various concentrations of the attenuated virus and later challenge them with the deadly wild type to look for evidence of an immune response. Conventional PCR, however, could identify the presence of the wild-type virus in only 40% of her negative controls, when the virus should have been present in all of them. Without establishing that baseline, she was unable to say whether the attenuated virus had, indeed, infected the toads. Only four years later in October 2006, with the advent of a new TaqMan gene-expression assay, she was able to prove that the harmless virus was working. "That was the first time we could show a virus had gone in and done something biological," she says.

The next - and hardest step - will be to put together Halliday's targets and Pallister's virus to create the world's first cane toad-specific assassin. But, just as the team is beginning to see a light at the end of the tunnel, support has all but evaporated outside the walls of CSIRO. "It's fairly well known that I'm not a great fan of that work," says Peacock of the Invasive Species Cooperative Research Center. "My concern is the virus is just not specific to the cane toad, and therefore I cannot for the life of me see it being released." He says his own office is looking for strategies that could come to fruition in the next two to 10 years, and he is skeptical that viral biocontrol falls into that time frame. Even CSIRO scientists admit that they are working on a long-term solution.

Beyond the scientific challenges, CSIRO does not exactly have a clean record when it comes to containing foreign pathogens. In 1995, the agency had to explain to the Australian public how the calicivirus escaped from Wardang Island where it was being field tested with the rabbit population. No one knows for sure, but CSIRO suggested it could have hitched a ride on birds or insects, while others have suggested ranchers must have trapped infected rabbits and carried them to the mainland. In the end, the virus turned out to be a success, but its premature escape also means that more political barriers lay in store for the next biocontrol agent, particularly a genetically modified one. Even if the agency can prove that the virus will not wipe out native frogs, it is just a salty swim to Papua New Guinea and the rest of Asia, where toads are native. The chytrid fungus has already racked amphibians around the world, and one more pathogen could push many populations over the edge.

So, in trying to save one ecosystem, CSIRO could be endangering many more. That's the worry of herpetologist Rick Shine, who says that the threat of the cane toad does not warrant the gamble of introducing a virus. "It's difficult to imagine the level of evidence required to convince skeptics it won't move across species boundaries," he says. He rightly points out that not a single organism has become extinct due to the cane toad; it has caused local extinctions for only a small group of predators that consider the animals as prey. Shine's team has shown that some snakes, such as the red-bellied black snake, seem to be evolving relatively smaller heads, which causes them to prey on smaller, less toxic cane toads. Some species of birds and rodents have also learned how to eat the toads safely by flipping them over and feasting on their organs.

Only scattered efforts have documented the cane toad's effects, and no one has looked into smaller vertebrates or insects. Ross Alford, an ecologist at James Cook University who has studied cane toads independently and in collaboration with Shine, says: "All in all, toads do not seem to be the complete environmental catastrophe that some people believe, but they certainly have had documented effects on a variety of native species and systems, some of which may prove to be quite important once they are better studied."

Shine says he certainly does not mean to play down the importance of losses in the predator community or the chance that further impacts will be discovered; he is simply unconvinced that viral biocontrol is worth the risk. He has been championing simple, less radical measures to mitigate the cane toad problem, measures that stem from a basic understanding of cane toad behavior. "People believe toads are these invulnerable invasion machines that we can't stop without incredible technology," he says. "We've done the basic biology, and cane toads have a bunch of Achilles' heels." For instance, he says that cane toads will not breed in a pond that lacks a muddy bank, so manipulating pond characteristics could provide a way to reduce or concentrate toads in certain areas. Researchers at James Cook University are looking into the biochemical ecology of the toads. Injured tadpoles, for example, produce a pheromone that triggers other tadpoles to accelerate metamorphosis and emerge from ponds at a smaller size. Other pheromones might be used to improve trapping methods.

Having invested more than 15 years in the project, Hyatt has little patience for the criticism leveled by Peacock and Shine. "How on earth are cane toads going to be controlled without an infectious organism?" Hyatt says. Cane toads are spread out across the country's most remote areas, and local solutions are never going to succeed, he adds. "It's very easy to criticize," he says, "It's harder to come forth with solutions."

Other researchers have proposed more outrageous gambits, such as flooding the population with sterile male toads or introducing a "daughterless" gene into the cane toad population to skew the sex ratio. According to the CSIRO team, the chance that its virus will mutate and wipe out the world's amphibians is far less likely than the chance that the virus will peter out over time or that it will fail to put a dent in cane toad populations.

Last March, following Pallister's recent success with her challenge experiments, the two teams met in Geelong, Victoria, to finalize the plan for their next batch of viruses. Halliday's gene-expression data had arrived, and he was disappointed to find that the toads were downregulating the ATPase on their own and still surviving. He is not sure what it means biologically, but for the project it meant that knocking down ATPase with RNA interference was no longer a straightforward strategy. Instead, the team elected two gut proteins that undergo major changes in expression during metamorphosis as part of the gut's remodeling process. All they need now is a few more months for Pallister to get these genes into her viruses, and they might just have an answer.

Meanwhile, the cane toad marches on.