Retroactive epidemiology

Study of tadpoles preserved in zoology museums identifies fungus responsible for die-off registered nearly four decades ago

Guilherme Becker / Unesp Specimens in the Unicamp collectionGuilherme Becker / Unesp

A mass die-off of frogs affected some regions of Brazil in the 1980s, without anyone ever knowing the cause.  Some species became rare while others disappeared from a particular location or even completely.  “We estimate that at least 65 species were affected, leading to 49 local extinctions and the disappearance of 15 species,” calculates biologist Luís Felipe Toledo, a professor at the University of Campinas (Unicamp). The setting for that amphibian disaster was mainly the stretch of Atlantic Forest between the Brazilian states of Paraná and Espírito Santo, and included areas where the forest is still intact–such as the Boraceia Biological Station and Itatiaia National Park, thus at first blush excluding the habitual villain (deforestation) from the list of potential suspects.  Nearly 40 years later, in an article published in the journal Proceedings of the Royal Society B in February 2017, a trio of biologists, which includes Toledo, has implicated the fungus Batrachochytrium dendrobatidis, or chytrid, as the culprit.

The conclusion demanded patience and persistence on the part of biologist Tamilie Carvalho during her master’s studies under Toledo’s supervision. Over the course of six months, Carvalho visited 10 zoology museums in six Brazilian states, but she had no time for tours. “I practically slept at the museums, breathing in formaldehyde,” she says jokingly.  And for good reason: during the period, she used a magnifying glass to examine the mouths of 33,000 tadpoles stored in bottles of formaldehyde.  She learned tricks for streamlining her work and inhaling fewer of the nausea-inducing gases: rinsing the tadpole in water and setting it on a paper towel under the magnifying glass before lightly pressing the animal’s mouth with tweezers in order to observe the structures rich in keratin, the protein the fungus attacks.  All her practical experience got her to where she was able to examine one tadpole per minute, more or less.  Apparently, that is the fastest and least expensive way to diagnose chytrid infection, which causes normally black skin to turn white. “I examined tadpoles from 13 different families so I always had to study the morphology of healthy specimens to know if the spots might be normal,” says Carvalho.  They never were.

Concentrating the diagnoses on tadpoles stored in zoology museums rather than on adult frogs was one of the differences in the study that allowed Carvalho to reach what she considered a stunning outcome: understanding what happened to cause the puzzling declines of the 1980s.  The adult frogs that had been the focus of previous studies died quickly when infected and rarely made it to the zoology museums.  This is due to damage to the keratin, which causes anomalies in the frogs’ hydric balance, leading to respiratory and cardiac problems. Tadpoles are more resistant, but end up eating less and going through metamorphosis at a smaller size.  For purposes of the study, what is important is that the infected tadpoles were captured before the fungus led to their death.  And this is why they are found in the collections, thus allowing analysis of the rate of infection for practically the entire country, in areas that represent every biome (see map).

Luís Felipe Toledo / Unicamp Extinct in Boraceia, Cycloramphus boraceiensis is resistant in IlhabelaLuís Felipe Toledo / Unicamp

The study analyzed animals collected between 1930 and 2015, and in that period, the fungus was primarily detected in the Atlantic Forest: nearly 17% of the tadpoles analyzed in that biome were infected, but the cases were concentrated in the South and Southeastern regions.  A spatio-temporal statistical analysis allowed scientists to detect the coincidence between the presence of chytrid fungus and a good many of the registered extinctions or population declines in an area of the Atlantic Forest that stretches from Rio de Janeiro to Paraná and includes Itatiaia, in Rio de Janeiro State, and Boraceia, in São Paulo State. For biologist Guilherme Becker, a postdoctoral researcher at the Rio Claro campus of the State University of São Paulo (Unesp) and the author most involved in the statistical portion of the study, this strong correlation seems to indicate that the fungus could really have been the cause of the die-off.

The reality that appeared in the results throws Becker’s previous study into question. It had stated that laboratory experiments had found signs that high biodiversity would hinder transmission of the disease (see Pesquisa FAPESP Issue nº 226). “While the experiments indicate the protective effect of biodiversity, field studies point to a higher risk for diseases in areas with natural vegetation and high amphibian diversity,” he explains, pointing out that these findings are still considered to be speculative.  The higher the amount of biodiversity, the more complicated is the network of ecological relations and the greater the probability that amphibians that never come close to water would come in contact with aquatic species that transmit the disease.  Waterways are where chytrid fungus lives.

Lethal companion
The results are also surprising because the fungus has been found in that area for more than a century, to the point that some specialists such as zoologist Célio Haddad, a professor at Unesp Rio Claro, consider it to be part of the ecosystem (see Pesquisa FAPESP Issue nº 220). The perception, based on previous studies, was that amphibians were able to survive the disease in Brazil, unlike the mass extinction that occurred in other countries, such as Costa Rica.

But what makes the case of Brazil different is not just the alleged resistance of the amphibians.  The domestic lineage of the chytrid fungus, apparently exclusive to Brazil, may be less lethal.  One hypothesis is that what was responsible for the extinctions was a particularly virulent lineage of global chytrid fungus that landed in Brazil in the late 1970s, perhaps having hitched a ride on bullfrogs imported for raising in captivity.  Beyond that, the forest strain mixed with the local strain.  “Chytrid fungus is asexual all over the world,” Toledo says.  “But here, it apparently is able to reproduce sexually and form hybrids that may be more aggressive.”

In preliminary experiments, Becker had already indicated that the combination exclusive to Brazil is the most effective in killing frogs.  “They do not have an immunological response to that variant,” he explains.  One limitation in diagnosing the museum specimens up to now has been the fact that hybrids were only recently discovered and it has not yet been possible to complete a genetic characterization of the 1980s epidemic.

According to Haddad, who in 1979 began his undergraduate studies, embarking the next year on his career with frogs, it was only years later that the disappearance of some species in Itatiaia became a topic of interest. “When I went there, I wasn’t able to find them anymore,” he recalls.  Speculation about the causes involved human activity, such as the highway or pollution associated with Proálcool, the National Alcohol Program.

Guilherme Becker / Unesp Holoaden bradei, eliminated by the chytrid fungus in the 1980s, lived only in ItatiaiaGuilherme Becker / Unesp

But what looms large is the possibility that the effect was more global in nature, since in the 1980s, population declines were also observed in Central and North America as well as in Australia.  Climate change may also be a part of this equation, thus maximizing the presence of the fungus. “The chart of average global temperatures indicated that they were slowly increasing up to 1979, when they showed a huge jump,” Becker noted. His hypothesis is that the frogs that inhabited the higher elevations, those most affected by the chytrid fungus, have little capacity to adapt to temperature changes. “Those amphibians are adapted to climates that are consistently colder.” Becker is now beginning experiments on animals that come from different altitudes to see how they respond to the infection at different temperatures.

The study indicates that what is critical in Brazil is not the presence or absence of the fungus.  “In the other countries, the frogs died when chytrid fungus arrived,” Toledo says.  “Here, they died when there was an increase in its prevalence.” Or else something changes and this is why it is necessary to monitor the areas where it is found.  To some degree, the study by Carvalho, Becker and Toledo corroborates known risk factors for the disease: high-altitude areas with complex topography that are rainy and see cool temperatures.  But still to be investigated are factors such as pollution or deforestation that increase the amphibians’ vulnerability and therefore, mortality.  “We need to study variables that allow us to predict the presence of the chytrid fungus,” says the Unicamp biologist.  “The implications of the findings are huge for the preservation of the amphibians in Brazil,” Becker adds.  “Now we know that the fungus could return to threaten our biodiversity as it did in the 1980s.”

Carvalho, who has now begun doctoral studies at Unicamp, is trying to unlock precisely those factors and mechanisms behind the vulnerability to the fungus.  In her opinion, this study goes well beyond disease and frogs.  It is about using the disease to understand ecological mechanisms.  To confirm the extinctions, Haddad is involved in a project to analyze the DNA extracted from rivers.  “We may be unable to find the frogs, but if they are there, their DNA is in the water,” he says.

Communication and sensory systems of the anurans of the Atlantic Forest (nº 14/23388-7); Grant Mechanism Regular Research Grant; Principal Investigator Luís Felipe Toledo (Unicamp); Investment R$139,637.65.

Scientific article
CARVALHO, T. et al. Historical amphibian declines and extinctions in Brazil linked to chytridiomycosis. Proceedings of the Royal Society B. V. 284, No. 1848. February 8, 2017