Do captured viral genes make human pregnancies possible?
The placenta, a tissue created by the developing embryo while it's still a ball of cells, mediates this intrusion. As the physical interface between mother and child, a placenta helps prevent the maternal immune system from rejecting the baby as foreign tissue. Early in pregnancy, it also secretes hormones that prepare the uterine lining to receive the fetus. Then, placental cells burrow into the uterus, where they burst maternal blood vessels to create pools of blood from which a fetus draws sustenance and into which it deposits wastes. Later, placental hormones ready the fetus for delivery.
"In many ways, the placenta is the SCUBA system for the fetus, while at the same time being the Houston Control Center guiding the mother through pregnancy," reproductive biologist Harvey J. Kliman of Yale University School of Medicine in New Haven, Conn., notes.
Despite the tissue's sophistication, evolution created the placenta only within the past 100 million years. Earlier, all animals laid eggs, which forced offspring to develop quickly and independently of their mothers. Biologists view placental pregnancy as a major advance because it allows a mother to nourish and protect an embryo longer.
"We know that the longer the gestation, the greater potential there is for fetal neurological development," notes placental biologist James C. Keith of the biotech firm Genetics Institute in Cambridge, Mass.
As with many evolutionary adaptations, the origins of the placenta remain shrouded in mystery. That hasn't kept biologists from speculating, however. One remarkable conjecture concerns viral genes that became embedded in the genomes of the forerunners of mammals. These captive genes, the theory holds, may have facilitated the placenta's evolution.
This provocative scenario recently got its first solid dose of supporting data. At least two viral genes captured by the human genome millions of years ago encode proteins that, according to new experiments, may guide the development of the human placenta. The data on one of the proteins remain controversial, but work with the other protein strongly suggests that it fuses placental cells. This finding may explain the creation of the unusual but crucial barrier linking the blood supplies of mother and fetus.
Beyond shedding light on placental evolution, these viral genes could provide insight into pregnancies that go awry. Investigators are already examining whether the genes are defective in women who have had problem pregnancies.
The story of the placenta and its viruses begins with some suspicious black-and-white photos. In the 1970s, investigators focused powerful electron microscopes on placental tissue from mice, cats, baboons, and people. The resulting images unexpectedly revealed viruslike particles budding from apparently healthy cells.
The particles turned out to be retroviruses, a viral family that includes the AIDS virus. Like other viruses, retroviruses must sneak their genetic material into cells and command a cell's internal machinery to replicate themselves. Unlike their viral relatives, retroviruses have adopted a strategy for permanently inserting a copy of their genes into the genomes of the cells they invade.
Throughout the 1970s and 1980s, notes Erik Larsson of the Uppsala University in Sweden, investigators continued to gather evidence that the placenta is a hot spot for retroviral activity. They also began to theorize about why. Larsson and other biologists saw a few obvious possibilities.
As the AIDS virus so tragically confirms, most retroviruses have a knack for suppressing a host's immune system. Consequently, the fetus may use retroviral proteins to help prevent the mother from rejecting it as foreign. After all, half the fetus' genes come from the father.
"You just need to express [the retroviral genes] for a rather short time because the placenta produces several strong immunosuppressive proteins that are not related to retroviruses," says Larsson.
Another theory drew upon the unusual anatomy of the placenta. As it invades the uterine wall, this tissue develops fingerlike projections that penetrate pools of maternal blood. These fingers extend from a single thin cell, called the syncytiotrophoblast or the placental syncytium, which contains multiple nuclei.
Such merged cells are rare in the human body, and scientists have wondered whether proteins from retroviruses create this cell layer. Viral surface molecules known as envelope proteins fuse certain viruses to a cell as the first step in infecting it. The AIDS virus, HIV, even appears to use its envelope proteins to fuse immune cells to each other (SN: 12/19&26/98, p. 391).
It's hard to imagine that women get infected with a retrovirus every time they become pregnant. Where else could the placental retroviral components come from? From within. Surprisingly large portions of mammalian genomes, perhaps as much as 1 percent in people, consist of so-called endogenous retroviruses.
These remnants of ancient infections result from instances in which a retrovirus happened to slip inside a host's sperm or egg cell. Any embryo then conceived would have the genes of the virus embedded within the DNA of all its cells. Descendants of that organism would also contain those viral genes.
The human genome has more than a thousand sites where a retrovirus has inserted its genes. Because of the ravages of evolutionary time, most of these genes consist of quiescent, tattered DNA.
Investigators have found that a few of the endogenous retroviruses remain relatively intact. In some cases, the viral genes still produce retroviruses. In other cases, a single gene, often the one encoding the protein that makes up the viral envelope, persists in a functioning state. One explanation for the enduring integrity of such genes, investigators speculate, is that the host cell has co-opted the viral proteins for some vital function.
Unfortunately, admits Larsson, investigators who suggest a placental role for endogenous retroviral proteins have failed to amass much evidence. "We've never been able to prove it," he says.
Biologists have shown, for example, that the envelope proteins of most retroviruses share a common fragment that curbs immune cell activity in test-tube experiments. Yet they haven't shown that any envelope protein produced by an endogenous retrovirus plays that role in the placenta.
Over the past decade, an endogenous retrovirus named ERV-3 has drawn most of the attention of researchers trying to establish a role for viral proteins within the placenta. ERV-3 doesn't produce complete retroviral particles because most of its genes harbor debilitating mutations. Its envelope gene, however, remains well preserved and active in a few human tissues, including the placental syncytium. Indeed, ERV-3's envelope protein makes up as much as 0.1 percent of all the protein in this microscopic sheet of fused cells (SN: 9/2/95, p.151).
Since ERV-3's envelope protein harbors the characteristic immunosuppressive domain, scientists were quick to suggest that it was part of the fetal defense against a mother's immune cells. Last year, Neal S. Rote of Wright State University in Dayton, Ohio, and his colleagues reported that the ERV-3 protein may play other roles as well.
In their experiments, described in the January 1999 PLACENTA, they added the gene for the ERV-3 envelope protein to laboratory cultures of trophoblasts, cellular precursors of several placental layers, including the syncytium. The inserted gene drove the trophoblasts, called BeWo cells, to slow their growth, alter their shape, and begin producing human chorionic gonadotropin--the hormone that home kits detect in urine to indicate pregnancy.
All those changes reflect what happens when trophoblasts start to form the placenta. In Rote's view, these findings represent the first strong evidence that an endogenous retroviral protein has an essential role in the placenta.
The ERV-3 story remains murky, however. In 1998, before Rote's paper came out, Nathalie de Parseval and Thierry Heidmann of the Institute Gustave Roussy in Villejuif, France, had delivered an apparent knockout blow to speculation about ERV-3. They documented that about 1 percent of the people they examined have a mutation in both their copies of the ERV-3 envelope gene. The defect would severely shorten the resulting protein. Indeed, the protein that these people make doesn't include the immunosuppressive domain or the part thought to mediate cell fusion in the syncytium.
While many researchers interpreted de Parseval and Heidmann's data as a clear dismissal of ERV-3's importance to the human placenta, Rote holds fast. He speculates that the short form of the ERV-3 envelope protein retains some of its active elements and still has the effects on placental cells that his group observed. Although the people with the ERV-3 envelope mutations are healthy today, it's not certain that they resulted from problemfree pregnancies, he adds.
Finally, Rote argues that the body often builds in redundancies for important functions. Even if a person has a defective ERV-3 envelope gene, he says, "there might be a backup system. There might be multiple endogenous retroviruses performing the same functions."
Although ERV-3's possible placental roles continue to draw debate, another human endogenous retrovirus has popped up to steal the spotlight. Several years ago, hoping to identify molecules that have commercial potential as drugs, Genetics Institute launched an intensive search for genes encoding novel proteins secreted by human cells.
As part of that effort, the company identified a fragment of a gene active in human testes. The firm's scientists would ultimately find that the gene encoded a protein that sits on the surface of testes cells rather than being secreted. Before that, however, they determined the full DNA sequence of the gene and deduced the amino acid sequence of the protein it encoded. They then scanned a database of previously discovered proteins for similar molecules. Their search found an unexpected match: An envelope protein from a baboon endogenous retrovirus bore a significant, though far from complete, resemblance to their protein.
"That's what originally peaked our curiosity," recalls John M. McCoy, who recently moved from Genetics Institute to its Cambridge neighbor Biogen.
Curious about why human testes would make a retroviral-envelope protein, McCoy, Keith, Sha Mi, and their other Genetics Institute colleagues took a closer look at where else in the human body the protein's gene is turned on, or expressed. The gene was much more active in the human placenta than in the testes. "It wasn't expressed anywhere else as far as we could see," says McCoy.
The scientists next documented that the envelope gene functions primarily in the trophoblast cells that fuse into the placental syncytium. Because they knew that other viral-envelope proteins fuse cells, the group set out to explore whether their protein performs such a function.
It does. In a series of experiments, the scientists showed that adding their gene to nonplacental cells causes them to fuse as if they were forming a syncytium. Many n Cells engineered this way would nurtures even join to liposomes, sacs composed only of the fatty molecules that make up a cell's membrane.
The investigators also determined that when they treated BeWo cells with a chemical that triggers trophoblasts to fuse, the envelope gene dramatically increased its expression. Moreover, chemically treated BeWo cells that were exposed to antibodies to neutralize the gene's protein fused much less often than cells not subjected to the antibodies. To reflect the retroviral protein's putative role in forming the placental syncytium, the company named it syncytin.
In an independent search for new human endogenous retroviruses, Francois Mallet and his colleagues at the Ecole Normale Superieure in Lyon, France, also identified the gene for syncytin. In the April JOURNAL OF VIROLOGY, they confirm that placental cells make the protein and that it can fuse cells in the test tube.
The investigators at Genetics Institute and in France caution that their research hasn't proved that mammals use syncytin in their placentas. Indeed, there's a wrinkle in that notion. While people and certain primates possess the gene for syncytin, scientists haven't found the retroviral gene in mice or other mammals that also have placentas.
That's not discouraging to McCoy, who suggests that other mammals may have co-opted the envelope proteins of different endogenous retroviruses. "It's not out of the realm of possibility that this happened more than once" during mammalian evolution, he says.
Still, verifying syncytin's role in mammalian placentas poses a challenge. If mice had a similar gene, biologists could simply deactivate it in fertilized mouse eggs and observe whether placentas develop and support healthy offspring. Researchers obviously can't perform such experiments on people. Rather, to disprove that the protein has an essential role in the placenta, investigators are trying to find people who were born healthy despite mutations in the gene for syncytin.
Meanwhile, Genetics Institute has started to compare the production of syncytin in healthy and abnormal placental tissue. Some of the latter comes from women who during pregnancy developed preeclampsia--dangerously high blood pressure thought to be caused by defective placental development.
Academic scientists will no doubt continue to investigate the role of the endogenous retroviral protein, but Genetics Institute may stay interested in syncytin only if the researchers there find that it's altered or missing in infertility or problem pregnancies.
"If it isn't involved in a pathology, it's an interesting footnote, as far as a biotech company is concerned. If it is, then things get a little bit more exciting," says McCoy.
At the very least, the company has already offered the most compelling evidence to date for the seemingly outrageous idea that human and some primate embryos depend on ancient viral genes to form their life-endowing link with their mothers.
"There are multiple endogenous retroviruses turned on in the placenta. It wouldn't surprise me if there were more that actually contribute to placental function," says Rote. "This might be the tip of the iceberg."