Learning To Ignore Your Viruses

A couple weeks ago I wrote about the 98,000 viruses that have permanently pasted their genes into our genome over the past 60 million years. What makes these viruses doubly fascinating is that scientists are making new discoveries about them all the time. Over at the open-access journal PLOS Pathogens, two new papers add some pieces to the puzzle of how these viruses get into our genomes, and how they affect our health along the way.

The first paper offers a striking portrait of a virus hopping species. Researchers at the Cleveland Clinic stumbled across the virus as they were studying prostate cancer that runs in families.

It turns out that some forms of prostate cancer are associated with mutations in a gene called RNASEL. Some studies indicate that one mutation in particular is behind 13% of all prostate cancer cases. If one copy of the gene is mutated, a man's chances of getting prostate cancer go up fifty percent. If both copies are mutated, his chances double. And yet--as is often the case--other studies on RNASEL failed to make this link. So the scientists wondered if the mutation was just one ingredient in the recipe for prostate cancer, and if some other factor in the environment might have to come into play as well.

A clue to what that factor may be comes from RNASEL itself. When it's not crippled by a mutation, the gene produces an enzyme that shreds virus genes (in particular, viruses that carry genes on single-stranded RNA as opposed to double-stranded DNA). Viruses are known to play a role in many cancers, triggering cells to replicate like mad. So the scientists wondered if the RNASEL mutation weakened defenses to a particular virus that could contribute to prostate cancer.

To find out, the scientists embarked on a virus hunt. They isolated cells from prostate tumors linked to the RNASEL mutation, and split them open. They placed the contents of the cells on a glass slide studded with 20,000 molecular probes, each tailored to snag a particular fragment of a virus gene. The probes on this so-called Virochip are based on genes of known viruses, but they can also snag genes from previously unknown viruses. That's because closely related viruses share relatively similar genes.

The scientists discovered virus genes. And not just any virus genes. These genes belonged to a new virus whose closest relatives are found in mice. The biggest genetic fragment the scientists extracted from the cancer cells is 96% identical to murine leukemia virus. It belongs to the remarkable group of viruses that become part of their host's genome (known as endogenous retroviruses). As I wrote in my earlier post, over the generations these viruses tend to lose their ability to make new copies of themselves and infect other hosts. But murine leukemia virus can still break out. Until now, the virus had only been found infecting rodents. Now, however, it turns out that it is infecting humans as well.

Exactly how the virus turned up in the cancer cells is a mystery. The scientists doubt that the people who developed these particular tumors all got the virus from direct contact with mice. Instead, they suggest that the virus originated in mice, acquired mutations that allowed it to infect other vertebrate animals, and then--some time in the past--infected some people. Then the virus began to spread from human to human. The precise connection between the virus and prostate cancer remains to be discovered. The virus turns out to infect not the cancerous prostate cells themselves but surrounding cells, called stromal cells. It's possible that infected stromal cells send out signals that cause other prostate cells to turn cancerous. In any case, it would explain the strange mismatch of studies on the link between RNASEL and prostate cancer. A mutation to the gene weakens a person's defenses, allowing the virus to infect successfully. If the virus is found among some people and not others, association studies will yield different results.

Will this mouse virus find a home within our species as well? It may take hundreds of thousands of years to find out. The virus will need to infect a sperm cell or an egg, so that it can spread from one generation to the next without having to infect a new host. After it makes that transition, only time will tell whether its genes spread through the human population or eventually reach a dead end. That's how the thousands of in-house viruses we carry have made the transition.

If the mouse virus actually does help cause prostate cancer, treating it may prove to be a major medical advance. Over 30,000 men die of prostate cancer annually, and so blocking the virus might save several thousand lives each year. And understanding these viral shifts may also help in the fight against a more devastating pathogen, namely HIV.

HIV is also a retrovirus, meaning that it inserts its genes into our own. But it is not a live-in virus. It primarily infects one class of white blood cells, and then spreads to other people through shared needles, sex, and other forms of contact. HIV leads to the collapse of the immune system, otherwise known as AIDS. Growing evidence suggests that it does so not by killing cells directly, as once thought, but by chronically overactivating the immune system. As the immune cells divide madly, they eventually start malfunctioning and even committing suicide.

In an opinion piece in PLOS Pathogens, Viktor Muller and Rob J. De Boer point out that most of HIV's cousins, which infect other primates, don't do anything of the sort. I've reproduced a tree they put together, showing the relationship of HIV-like viruses in apes and monkeys. (Go here for a closer view.) HIV, marked in red, is not a single lineage of viruses. One form, HIV-2, jumped from sooty mangabey monkeys into people several times. The more common form, HIV-1, descends from chimpanzee viruses, which have moved into humans many more times. As the tree shows, lots of primates get infected by their own HIV relatives, and this appears to have been going on for millions of years. But if you look at sooty mangabeys or some other monkey, you generally find abundant amounts of the virus without any sign of an overactive immune system. It's not that the virus carried by sooty mangabeys is weak. Scientists have injected it into other monkeys, and it has triggered a strong immune response. The blue arrows on the tree mark the rise of new virus strains in macaques that came from sooty mangabeys. This shift appears to have happened at primate research centers in the past few decades. In their new hosts, these viruses cause lots of nasty symptoms.

Muller and De Boer propose an intriguing hypothesis to explain all of this: perhaps apes and monkeys don't suffer ill effects from these viruses because they carry copies of the viruses in their own genome. After all, the authors point out, HIV's genes have been isolated in human sperm DNA, so these viruses clearly have the potential to make their way into a host genome. Muller and De Boer suggest that primate viruses got into their hosts' genome. The young primates then began making proteins from the virus, which their developing immune system recognized as part of their "self." When the primates then got infected with new copies of the virus, they didn't mount an attack or become overstimulated. The viruses infected the primate's immune cells, but they were only a minor burden to the primates compared to a collapsed immune system. Natural selection would have favored the primates who carried these in-house viruses, as those without them died from viral infections.

Muller and De Boer point out that scientists have created a similar kind of tolerance by injecting viruses into mice--specifically into the thymus, the finishing school for immune cells. It would be nice if Muller and De Boer could also point to the DNA of HIV-like viruses sitting in the genomes of primates. They did look at the published sequences and came up dry. But the absence of evidence in this case definitely does not mean the evidence of absence. Only a couple primate genomes have been sequenced so far, and it is possible that simple searches may miss virus genes that have become fragments and have acquired a lot of mutations. Muller and De Boer propose that scientists should closely examine the genes that are expressed in developing immune cells in a wide range of primates. Some of these genes may turn out to be similar to the genes of the HIV-like viruses that infect their species. Another way to the test the hypothesis is to run an experiment in which the tolerance to the virus is blocked. Theoretically, these harmless viruses should become as vicious as HIV.

It's cool but a little frightening to imagine if Muller and De Boer are on to something. It would mean that primates have not survived their own HIV epidemics by destroying the virus. Nor would it mean that the virus had become more benevolent, in order to spare its host. It would mean that they simply evolved to ignore the virus altogether. I'm not sure I would agree to gene therapy to insert HIV genes into the genome of my children to protect them from HIV. But it might turn out to be the best way to come to an evolutionary truce with the viruses.