WALNUT CREEK, Calif. (GenomeWeb News) – Sometimes project ideas come in the mail. Joseph DeRisi, a professor at the University of California, San Francisco, received a letter and a picture in the mail asking him to look into a disease affecting snakes. Based on this tip, DeRisi and his team used a metagenomic-based approach to uncover novel arenaviruses infecting snakes, an effort he outlined during a presentation at the US Department of Energy Joint Genome Institute's eighth annual User Meeting.
The disease, called Inclusion Body Disease, or IBD, is a transmissible disease in which snakes waste away. The disease, as its name implies, is marked by large eosinophilic cytoplasmic inclusions in the cells of infected snakes. The cause of IBD was unknown, but such inclusions led to suspicions that the disease was viral in nature. IDB is highly contagious among snakes — DeRisi noted that if one snake in a zoo or other collection came down with the disease, the zookeepers would cull the snakes housed with the ill one and any others that had been in contact with it to try to save the rest of the snake population.
The woman who wrote to DiRisi had a boa constrictor called Larry, and she was worried that she'd lose this pet snake as she had her previous ones to IBD.
By the time that DiRisi began to look into the project the California Academy of Sciences was experiencing an IBD outbreak. His team received six samples from the snakes being culled there. Five of those samples came from snakes with confirmed IBD, while one did not, and DeRisi's team was not told which samples were which. Those snakes included both boa constrictors and annulated tree boas.
They then "sequenced the heck out of" these samples using the HiSeq 2000, DeRisi said, generating about 6 million sequences per tissue — they examined RNA from brain, gastrointestinal, heart, kidney, lung, and liver tissue samples.
Some findings from this project have been published in mBio.
They then put that dataset through some data reduction paces, subtracting out the host sequences to get to the exogenous cause of the disease. However, there wasn't a snake genome available. The academy, though, had a separately housed, healthy boa constrictor called Balthazar who was used to generate a snake genome. In collaboration with Illumina, DeRisi's team generated mate-pair and paired-end reads from Balthazar.
Around the same time, Assemblathon 2 was gearing up for its second round in which researcher-participants were to assemble three genomes within three months. DeRisi suggested that the red-tailed boa be sequenced. "Just pulling it out of the sky," as he said at the meeting. The boa, along with the Malawi cichlid and the common pet parakeet were part of the Assemblathon contest.
With that assembly in hand, DeRisi and his team filtered the data from the sick snakes and found only one category that hit against the viral database — arenavirus.
Arenaviruses are ambisense, bipartite RNA viruses that are primarily found in rodents. "They are freaky little viruses," DeRisi said. When they do infect humans, arenaviruses can cause hemorrhagic fevers such as Lassa fever.
Here, the researchers identified three novel arenaviruses viruses, now called CAS virus, found in annulated tree boas, and Golden Gate virus and Collierville virus, found in boa constrictors.
To assemble their arenavirus reads, the researchers turned to an assembler developed in their lab. The assembler, called PRICE, for paired-read iterative contig extention, is a targeted assembler for metagenomic data and was published this week in G3: Genes, Genomes, and Genetics. In that paper, DeRisi and two of his colleagues write that "the motivation for PRICE was to facilitate the assembly of individual genomes of interest using complex, shotgun, metagenomic datasets, derived from virtually any origin, be it clinical or environmental."
Now assembled, these new arenaviruses appeared quite divergent from other, known arenaviruses as they formed their own clade compared to the others. In fact, they shared characteristics with another set of RNA viruses, the filoviruses.
DeRisi and his team examined envelope glycoproteins that the putative IBD viruses encoded. They did not match up with arenaviruses, but with filoviruses, which are famous for including the virus causing Ebola.
DeRisi added, though, that at least one paper in the literature suggested that arenaviruses and filoviruses might share a common RNA virus ancestor. The snake viruses' Z proteins also differed from what is typically found in an areanvirus.
Still, arenaviruses may not necessarily be the cause of IBD, but DeRisi argued that evidence is mounting. He noted that the virus was found in five of the six samples they received from the California Academy of Sciences, and the one that did not contain the virus was revealed to be from the snake without the disease. In addition, the CAS samples from infected snakes responded to viral antibodies.
To support the idea that the arenaviruses do lead to IBD, DeRisi and his team set about to establish a snake cell line — they tried a few available ones, but were unable to grow the virus. They issued a call to local vets, looking for boas with cancer.
Kidney cells from a snake called Julia were used to establish a cell line that DeRisi said is just about immortal, and these cells could support viral growth, allowing the researchers to grow the CAS virus and develop antibodies. They found that antibodies aggregate in infected cells, and near the inclusions.
DeRisi added that they have examined many more snakes — he and a colleague appeared on a herpetology podcast, and after that they began to receive many samples in the mail. They now have 40 isolates, which have been sequenced. They have never found the virus in healthy snakes, and those with IBD have the virus.
In response to a question from the audience, DeRisi said that the boa constrictor Larry was still healthy and did not have IBD.