What came first, cells or viruses?

Do humans really mark the pinnacle of evolution, or do viruses? While we’ve evolved along a pathway of ever-increasing complexity, viruses have streamlined, successfully jettisoning all but a handful of essential genes, research published in Science Advances in September suggests.

Gustavo Caetano-Anolles and his colleagues at the University of Illinois reached this conclusion after pioneering a new way to map the microbial family tree. Viruses did not evolve first, they found. Instead, viruses and bacteria both descended from an ancient cellular life form. But while – like humans –  bacteria evolved to become more complex, viruses became simpler.

Today, viruses are so small and simple, they can’t even replicate on their own. Viruses carry only the essential genetic information they need to be able to slip inside a host cell and coax it into making new copies of the virus. The influenza virus, for instance, has a mere 14 protein-coding genes. Because viruses are usually so basic, many biologists didn’t think they could even be classified as a life form.

But just over a decade ago, our view of viruses began to shift. French scientists who were examining a mystery microbe that looked like a bacterium, but was genetically quite different to bacteria, realised they’d discovered a giant virus. They named this bacteria look-alike the “mimicking microbe,” or “mimivirus”.

And the mimivirus wasn’t only physically large. They showed that it carried more than 1,000 genes – a huge genome for a virus, just a few hundred genes smaller than some bacteria. Several giant viruses have been discovered since, with pandoraviruses packing around 1,100 genes.

The genetic complexity of these monster microbes reawakened interest in a longstanding question about viruses – when did they first evolve? Were viruses an evolutionary stepping stone to more complex cellular life? Or did they spring up later? The question is thorny. Being made of a few short strands of DNA or RNA wrapped in a soft protein shell, viruses don’t fossilise. And without a fossil record to study, it has been almost impossible to untangle their lineage.

To try to unpick the question of virus evolution, Caetano-Anolles developed a new way to reconstruct the microbial family tree, and retrace bacteria and viruses back to their origins.

Scientists usually create evolutionary family trees, or “phylogenetic trees,” by comparing genes between species. The more genes two organisms have in common, the more closely they’re related. But this technique only lets you rewind a million years or so. Any further and the DNA has mutated so much it’s impossible to see the similarities between species.

Caetano-Anolles wanted to go back to the beginnings of life on Earth – around 3.5 billion years ago. So instead of comparing genes, his team compared the shape, or “folds,” of proteins. Proteins are high-precision molecular machines – if you change their shape, you disrupt their function. While life can tolerate a continual gentle drift in the genetic code, protein shape is critical and therefore evolves much more slowly. Retracing protein shape “takes us as far back as we can possibly hope to go,” says Michael Charleston, a computational biologist at the University of Tasmania.

The researchers developed algorithms to compare the protein shapes of 3,460 viruses and 1,620 cells. They found that 442 protein folds were shared between cells and viruses, but 66 folds were unique to viruses.

To make sense of the data, the team arranged the protein folds into a tree that grew a new ‘branch’ every time a new type of protein fold evolved. Wherever possible, the team used fossil evidence to put an approximate date on the budding of specific branches. For example, one particular protein fold was first seen in cyanobacteria (blue-green algae), and later appeared in all its descendants. By comparing when cyanobacteria first appeared in the fossil record (2.1 billion years ago) to when its offspring later emerged, they could establish this particular fold appeared around 2 billion years ago.

According to Caetano-Anolles’s microbial family tree, viruses are ancient – but they were not the first form of life. In fact, his family tree suggests viruses and bacteria share a common ancestor – a fully functioning, self-replicating cell that lived around 3.4 billion years ago, shortly after life first emerged on the planet. From this cell, bacteria have evolved in the direction of increasing complexity, while viruses have gradually shed genes they found they didn’t need – until they could no longer even reproduce on their own.

A key step in the virus evolutionary journey seems to have come about around 1.5 billion years ago – that’s the age at which the team estimated the 66 virus-specific protein folds came on the scene. These changes are to proteins in the virus’ outer coat – the machinery viruses use to break into host cells.

For Charleston, the remarkable point about the study is how many proteins today’s bacteria and viruses have in common. “I’m blown away by the idea that these protein folds are so conserved,” he says. “That is really quite strong evidence that they have a common ancestor way back.”

Today, it’s tempting to think of viruses as mere pests. But “they are not agents of destruction,” Caetano-Anolles says. Life on Earth would look very different without our viral co-inhabitants. “We wouldn’t be here without them,” says James Shapiro, a University of Chicago microbiologist. For example, researchers speculate that more than 100 million years ago a viral infection in a primitive mammal uploaded a gene that helped the placenta evolve. Syncytin is a protein viruses use to fuse cells together in order to hop from one host cell to the next. In mammals it fuses placenta cells with the uterus, allowing the foetus to draw nutrients from its mother.

And as for the more abstract question of whether viruses qualify as life, Caetano-Anolles argues that if viruses are descended from living cells, they’re still alive now – but in a unique way: when viruses infect a cell, that reunion forms a complete living system.

John Mattick, molecular biologist and director of Sydney’s Garvan Institute, agrees. “People say viruses aren’t free-living. But that’s a philosophical question – are we free-living?” he asks. “We can’t live  without plants. Life is an interconnected system.”

Related reading: New imaging reveals hidden details of viruses