Ian Lipkin, a virus hunter from Columbia University, along with fellow professors Thomas Briese and Amit Kapoor have designed a new system, known as VirCapSeq-VERT, that they claim can detect any known human virus in a blood sample.
“When people analyze samples from people who are ill, they have some idea in mind. This is probably an enterovirus, or maybe it's a herpesvirues. They then do a specific assay for that particular agent. They don't usually have the capacity to look broadly.”
The team recently received a blood sample from the National Institutes of Health. They were taken from a man who had received a bone-marrow transplant and had fallen mysteriously ill, with evidence of severely inflamed blood vessels.A few years back, working on a similar case, Lipkin discovered a new polyomavirus -- part of a family that can cause disease in people with compromised immune systems -- so he jumped at the chance; perhaps this was his chance to discover yet another virus.
Lipkin's team ran the sample through a system that they had devised to detect human viruses, and found that the man was infected with dengue virus. In hindsight, that made sense—he had recently returned from Vietnam, where dengue is prevalent. The team wasn't looking for dengue virus, but it still showed up -- and as regular blood tests can find if a particular virus is present or not in a sample, their new system could save thousand of hours in the lab, trying to guess the right virus to test for, and thousands of lives.
“It wasn't what we anticipated, but we didn't have to make a priori decisions about what we planned to find,” Lipkin says.
In the 120 or so years since viruses were first discovered, our ability to identify them, and diagnose the diseases they cause, has improved enormously. But even the most cutting-edge of techniques have limitations. Sequencing technologies allow scientists to unambiguously decipher the genetic material of viruses in a sample, but they suffer from poor sensitivity -- sometimes they just miss detecting the genes they're looking for, as they are usually swamped by the genetic material of heir host.
One way to solve this is to use a method of DNA amplification known as PCR. This solves the problem by making lots and lots of copies of the analyzed genes beforehand. It allows for exquisitely sensitive results but it's also hard to do in bulk, and you need to have some idea of what you're looking for in the first place.
VirCapSeq-VERT takes the best features of these techniques, and the team threw a few improvements of their own in the mix. They compare the method to a massive exercise in fishing for viruses. The hooks are made of distinctive stretches of DNA from the genomes of every group of virus known to affect humans and other vertebrates, that the team identified and synthesized -- netting them two million such virus-hooks, each tailored to catch a specific virus. If you throw the hooks in a blood sample, wiggle them a bit, then take them out and sequence everything that they caught, you get the full genome of every virus in the sample.
The team tested the system using tissue samples spiked with genes from many infamous viruses, including those responsible for Ebola, dengue, flu, and MERS. They also tried analyzing a nasal swab from a patient and a stool sample from a bat. VirCapSeq-VERT successfully identified all the viruses in these samples, even when they were present at miniscule amounts.
And since the technique offers up the full genomes of the virus it detects, it's impossible to fool.
“If you get a genome, you know what it is. It's unequivocal,” says Lipkin. “It also allows you to find mutations that would circumvent traditional diagnostics, or that might would affect resistance to drugs or vaccines.”
And it can analyze many patients at once. To do that, they fuse a unique barcode sequence to the viral DNA from each individual sample, before mixing them together and running them through VirCapSeq-VERT. After the system does its thing, the team can check the barcodes to work out which sample each virus came from.
“We can do up to 21 at a time, which makes it financially viable,” says Lipkin.
Finally, the system is flexible. It should pull out any virus that's even a modest fit to the baited hooks, which includes mutant strains and, potentially, previously unknown viruses.
“We probably won't find anything entirely new, but we'll be able to find anything that's within 60 percent similarity to something represented in our library,” says Lipkin, whose team has discovered over 700 viruses. Already, they have used VirCapSeq-VERT to identify a new virus in tilapia, an economically important fish that is being increasingly farmed in aquaculture.
“The results look very promising,” says Nick Loman from the University of Birmingham, who was not involved in the study.
“Going forward, we can combine techniques like this with portable sequencing and have a diagnostic device which provides incredibly rich data for clinicians and epidemiologists. Ultimately what we would like is an entirely unbiased method that captured all pathogens—known and unknown—with exquisite sensitivity.”