And virus makes four

We may be infected — and protected — from our earliest days of development

Photo of Joanna Wysocka

The next time you start to feel special, keep in mind that much of your DNA isn’t even yours. In fact, your genome is littered with the ancient corpses of viral invaders of hundreds (or even millions) of years ago. Basically, each of us is just a giant junk heap.

If you find that dispiriting, here’s another bit of unsettling news: Some of these skeletons come back to life during very early human development. The viral DNA makes viral proteins, which assemble themselves into something that looks suspiciously like infectious viral particles.

“It’s both fascinating and a little creepy,” says Joanna Wysocka, PhD, Stanford associate professor of developmental biology and of chemical and systems biology. “We can’t say yet whether these viral particles can be infectious, but regardless of whether they are, viral proteins within a cell are rarely completely inert.”

Wysocka described the phenomenon in a paper published earlier this year in Nature. Graduate student Edward Grow was the study’s first author.

The finding raises questions as to who, or what, is really pulling the strings during human embryogenesis. Grow and Wysocka have found that these viral proteins are well-placed to manipulate some of the earliest steps in our development by affecting gene expression and even possibly protecting the embryo’s cells from further viral infection.

It’s unclear, however, whether we are watching an ongoing battle between viruses and humans or the outcome of an uneasy truce hashed out over tens of thousands of years of evolution.

“Does the virus selfishly benefit by switching itself on in these early embryonic cells?” wonders Grow. “Or is the embryo instead commandeering the viral proteins to protect itself? Can they both benefit? That’s possible, but we don’t really know.”

When genes jump

The researchers didn’t start out looking for reanimated zombie viruses. They were mostly interested in understanding the earliest stages of human embryonic development: how a newly fertilized egg no bigger than the period at the end of this sentence becomes a squalling newborn with limbs, hair, fingers and a hefty set of lungs with which to manipulate the emotions and actions of nearby parents.

Researchers have known for nearly six decades that nearly every cell in the human body contains 23 pairs of chromo­somes — stubby bundles of DNA strings that carry the instructions to make every protein in the body. This DNA makes up what’s known as our genome, and it’s kept within a special control center inside the cell called the nucleus. With a few exceptions, all cells in your body contain the same genome, but they use the encoded instructions to make different tissues and organs.

It’s somewhat like how a good cook can use a set of cookbooks and a well-stocked pantry to make an omelet, or a cake, or a succulent beef roast. Ingredients and timing are everything. All these foods are tasty, but you wouldn’t want a roast for dessert or a cake for the main course of your holiday dinner.

A cell relies on the sequential and coordinated expression of genes in the genome for its molecular recipes. Specific sequences in the DNA are copied into RNA in a process called transcription. For the most part, these RNA messages then leave the nucleus and travel to protein-making machinery called ribosomes in the cell’s cytoplasm. (Other RNA molecules perform regulatory functions that direct the expression of other genes in the genome.) At the ribosome, the protein is assembled and sent off on its merry way to direct the function or development of the cell.

Wysocka and Grow were interested in the regulatory mechanisms within an early embryo that control which genes are made into RNA and proteins, and at which times during development. They homed in on a mobile genetic element called a transposon. Transposons are short bits of DNA that in their active, mobile form can insert (and re-insert) themselves over and over into the host DNA, resulting in hundreds or even thousands of copies hiding in our genome. Altogether, they make up about half of our genome. However, with time they accumulate mutations and rarely encode protein-making instructions themselves. Recently, though, they have been shown to play an important regulatory role in the expression of nearby genes.

You may remember transposons as the “jumping genes” that garnered Barbara McClintock the Nobel Prize in Physiology or Medicine in 1983. Unfortunately, her discovery, made in the 1950s, was largely ignored for decades. Until the full effect of McClintock’s groundbreaking research was realized some 30 years later, transposons were considered to be genetic junk.

Think of transposon jumping as reorganizing or flipping the order of segmented pieces of track in a toy train set. Depending on the configuration or specific purpose of each section — does it have an off-shoot to direct the train off the route, or a stop signal to halt its progress? — the location and orientation of the newly inserted sections can matter a great deal. As McClintock showed, the transposon’s brand of genetic shenanigans can affect the color and pattern of kernels on a cob of corn.

But when Grow began to investigate when, how and which transposons were activated in human development, he found something surprising.

Stealth viruses

“There was something very interesting, and very specific, going on,” says Wysocka. “Those transposons that were particularly dynamic, activating at very specific times early in development, were made up of endogenous retroviruses.”

A retrovirus is a special category of transposon — one with nonhuman origins. About 8 percent of our genome is made up of these ancient retroviral sequences left behind during past infections hundreds of thousands of years ago. They’re referred to as endogenous because they are within our DNA, as opposed to the exogenous viruses we might contract from another human being.

In an active infection, retroviruses insert their genetic material into the genome of the host cell for later reactivation. In this stealth mode, the virus bides its time, taking advantage of cellular DNA replication to spread to each of an infected cell’s progeny every time the cell divides. HIV is one well-known example of a retrovirus that infects humans.

When a retrovirus infects a germ cell, which makes sperm and eggs, or infects a very early-stage embryo before the germ cells have arisen, the viral DNA is passed along to future generations. Over evolutionary time, these viral genomes typically become mutated and inactivated. One retrovirus, however, called HERVK, infected humans repeatedly until relatively recently — within about 200,000 years. Much of HERVK’s genome is still snuggled, intact, in each of our cells.

“HERVK is an interesting exception to most other endogenous retroviruses that infected primates like ourselves,” says Wysocka. “It’s evolutionarily younger and, in the case of more than 100 insertions into our genome, it has retained its protein-coding potential.”

Most of these viral sequences are inactive in mature cells, but recent research in other labs has shown that DNA from a retrovirus called HERVH, which is related to HERVK, is made into RNA at specific points in human embryonic development. This happens due to a phenomenon called hypomethylation, in which a cell sheds chemical tags called methyl groups that normally speckle its DNA. These methyl groups are a key way that a cell silences the expression of unnecessary genes. It wouldn’t do to have a skin cell suddenly start churning out digestive enzymes, for example.

“During development, there is a global wave of hypomethylation of the genome, which provides a window of opportunity for reactivation of previously silent elements,” says Wysocka.

This shedding of methyl groups is one way egg cells gain the developmental potential necessary to become all the tissues in the body. It also occurs sometimes in cancer cells, permitting them to attain new, potentially dangerous functions during the course of the disease. Furthermore, as the previous research showed, it releases recently acquired transposons and retroviruses for expression by the cellular machinery. Until now, however, it wasn’t known whether the cell actually used the RNA sequences made from ancient retroviral DNA to make viral proteins that could affect its function and development.

“We started looking at the transcription of these transposable elements early in development and found that there’s something very interesting and very specific going on. Different transposons are activated during different cellular states — so much so that it’s possible to identify the state a cell is in just by which transposons have been activated,” Wysocka says. “Endogenous retroviral elements,” like HERVK, “are particularly dynamic.”

Grow and Wysocka found that, in 3- to 4-day-old embryos, some HERVK viruses are transcribed into RNA. This viral activation coincides with the activation of other key human genes in the embryo. Then, researchers teamed up with Shawn Chavez, PhD, and Mark Wossidlo, PhD, two postdoctoral scientists from the lab of Renee Reijo Pera, PhD, a former Stanford professor of obstetrics and gynecology and former director of the Center for Human Embryonic Stem Cell Research and Education, to visualize viral proteins in the human embryos using antibodies labeled with fluorescent dyes. Finally, the researchers used electron microscopy to observe what appear to be intact viral particles in the human blastocyst, the hollow ball of cells that arises within five to six days after fertilization. They verified that the particles were made up of viral proteins by tagging them with gold-labeled antibodies. (Gold’s density appears as distinct black spots when viewed with an electron microscope.).

“When we looked at these human blastocysts, we saw they were packed full of viral proteins,” says Wysocka. “This was true for every blastocyst we looked at. Early human development clearly proceeds in the presence of viral proteins.”

‘Our “junk DNA,” including some viral genes, is recycled for development in the first few days and weeks of life. The question is, what is it doing there?’

When asked why these proteins haven’t been found before, Wysocka says, “I don’t know. People simply haven’t looked. Nobody really thought, ‘Well, let’s see if they actually make proteins in human embryos,’ even though such proteins have been seen before in germ cell tumors and cancer cells that transcriptionally reactivate HERVK.”

Although the viral proteins seem to assemble themselves into viral particles, it’s unclear whether they are capable of infecting other cells. That’s one thing the researchers would like to test. What they do know is that one of the proteins made by HERVK is a viral protein called Rec. When a cell is first infected by HERVK, Rec binds to viral RNA particles and escorts them to the ribosomes in the cells’ cytoplasm to be made into proteins. These viral proteins are then assembled into new viral particles, which are released to infect more cells.

Wysocka and Grow found that Rec affects the expression of more than just viral genes. In collaboration with the laboratory of Howard Chang, MD, PhD, a professor of dermatology at Stanford and a Howard Hughes Medical Institute investigator, they found it also binds to many RNAs made from human genes and affects the degree to which they interact with the cell’s ribosomes. Furthermore, its presence stimulates the cell to increase the amount of a surface-bound protein that protects it from subsequent viral invasion — a kind of molecular “get off my lawn” sign that firmly stakes a virus’ claim to the cell. Many questions remain, however, as to how long this protection may last, or what could be its purpose.

Tug of war

So, who’s in charge here? Us or the viruses? Or is there no longer any distinction? There’s certainly been plenty of evidence showing that humans are far from free operators when it comes to, well, pretty much anything. Our bodies are teeming masses of bacteria, viruses and even fungi that are collectively known as the microbiome. Many of these microorganisms, which are 10 times more numerous than our own cells, are essential to a healthy life, such as the gut bacteria that help us digest our food.

“What we’re learning now is that our ‘junk DNA,’ including some viral genes, is recycled for development in the first few days and weeks of life,” says Pera, who is now on the faculty of Montana State University. “The question is, what is it doing there?”

It’s not clear whether this sequence of events is the result of thousands of years of co-existence, a kind of evolutionary symbiosis, or if it represents an ongoing battle between humans and viruses. Regardless, it’s clear the fates of both virus and human cell are intertwined within days of conception.

Grow and Wysocka are now doubling down on their efforts to understand whether and how the viral transposons themselves affect the regulation of human gene expression. Do they enhance the expression of our own genes at particular times during development or in particular cells? More to the point: Who are the real cooks in our genomic kitchens?

“There is always a tug of war between a host and a virus,” says Wysocka. “Infection of our ancestors with HERVK was an accident of evolution. Regardless of whether the effect of this infection is positive or negative from the natural-selection standpoint, the virus clearly has an impact. We believe that the virus is likely to influence and fine tune many early developmental pathways specific to primates. They may even affect those that make us uniquely human.”

Author headshot

Krista Conger

Krista Conger is a Senior Science Writer in the Office of Communications. Email her at

Email the author