Swimming in a sea of viruses


A new technique for sifting tiny viruses from seawater could help the hunt for human pathogens

Looking out of his laboratory window at the CRG, Òscar Fornas can see the sea. Far out on the blue waves, sailors are busily hauling their catch onto the decks of fishing boats, packed full of fish that are too large to slip through the holes in their nets. The smaller the holes in the nets, the smaller the creatures that can be trapped inside. But how small a net do you need to trap something as tiny as a microscopic virus?

In order to understand the world around us, biologists are interested in finding out how animals, plants and microbes – including bacteria and viruses – interact to create an ecosystem. And the oceans are the most fascinating and important ecosystems on the planet, with more significance for the earth’s climate than the rainforests.

“Most of the biological diversity on the planet lives in the sea,” says Fornas, who leads the Flow Cytometry Unit shared between the CRG and Pompeu Fabra University. “Seawater organisms such as plankton and bacteria fix around half the carbon dioxide in the atmosphere and generate huge amounts of oxygen. But many viruses infect them and could potentially destroy them.”

In fact, the ocean is teeming with viruses – just 1 millilitre of seawater can contain about 10 million viruses – but estimates suggest we only know the identity of around one per cent of them.

Hunt the virus

As DNA sequencing has become faster and cheaper, researchers are increasingly capturing genetic data to understand more about the diversity of species living in an ecosystem. But while it’s relatively easy to collect DNA from individual animals or plants, it’s much more difficult to pick out single microscopic bacteria or even smaller viruses.

The most popular technique for analysing microbial genomes in the wild is known as ‘meta-genomics’. This involves taking a sample from the environment, such as a scoop of soil or a cup of water, and purifying and sequencing a mixture of DNA from all the microbes living there. Clever computational techniques are then used to separate out the genomes of individual species, but virus genomes are extremely small and are likely to be lost amongst all the rest of the data from larger organisms.

Working together with colleagues in Barcelona, Alicante and the US, Fornas wanted to find a way of separating single viruses from this mixture of microbes so each one could be sequenced individually. This approach has been successful with single cells from animal tissues or tumours and bacteria from many different environments. But it had never been done with something as tiny as a virus – around a thousand times smaller than a typical human cell.

Scaling down

To sift the viruses from the saltwater Fornas used a popular lab technique called fluorescence activated cell sorting (FACS), which is often used to separate individual cells from a mixed population. Cells are usually labelled with a fluorescent dye and fed into a machine where carefully designed liquid currents send them one by one through a laser beam for analysis. The cells are then automatically sorted into different multi-well plates for subsequent testing.

Adjusting the FACS machine to cope with something as small as a virus was a big challenge. Although Fornas and his team could label the viruses using a dye that stains viral DNA, the fluorescence is very faint because the genome is so small. There are also many other tiny particles in seawater, from specks of trash to microscopic blobs extruded from larger cells, which can confuse the sorting machine. And the whole process has to be kept free from contaminants that might damage DNA and make it impossible to sequence.

Finding the right settings took many months of testing and troubleshooting. And there was also the arduous day-long cleaning process that had to be performed before every experiment, making sure that every scrap of potentially contaminating DNA was gone. Luckily, Fornas and his team had access to a handy source of ocean viruses, using samples from the Mediterranean lapping outside the lab to optimise the technique.

“Seawater is like soup,” says Fornas. “We were working close to the limits of our technology. But we adjusted the lasers, slowed down the flow in the machine, and kept on running tests until we were able to successfully separate individual viruses.”

From sea to saliva

As well as his local samples, Fornas also isolated viruses from seawater collected from the sea surface and as far as 4 km down in the depths of the Mediterranean and Atlantic. In total, the team sifted more than 2,000 virus-sized particles yielding 392 rough viral genome sequences. 44 were sent for further more detailed sequencing, and all of them turned out to be viruses that were previously completely unknown to science.

Furthermore, viruses that turn up many times in the collection process are likely to be more common, providing a rough readout of the relative abundance of different species in that part of the ocean. Although finding so many new viruses is impressive, Fornas sees this project as a proof of concept.

“This project showed we can pull together FACS and genomics for viruses,” he explains. “We now have a tool that we use to identify new viruses in other ecosystems such as swimming pools, lakes, drinking water and even body fluids – we have already shown that we can use our method to find viruses in saliva. We don’t know what viruses are out there that could be harmful to humans, but now we have a tool to find them.”

Reference work

Francisco Martinez-Hernandez, Oscar Fornas, Monica Lluesma Gomez, Benjamin Bolduc, Maria Jose de la Cruz Peña, Joaquín Martínez Martínez, Josefa Anton, Josep M. Gasol, Riccardo Rosselli, Francisco Rodriguez-Valera, Matthew B. Sullivan, Silvia G. Acinas & Manuel Martinez-Garcia.

“Single-virus genomics reveals hidden cosmopolitan and abundant viruses”

Nature Communications, 8:15892 (2017), doi:10.1038/ncomms15892