Mammals Made By Viruses

By Carl Zimmer | February 14, 2012 11:48 pm

If not for a virus, none of us would ever be born.

In 2000, a team of Boston scientists discovered a peculiar gene in the human genome. It encoded a protein made only by cells in the placenta. They called it syncytin.

The cells that made syncytin were located only where the placenta made contact with the uterus. They fuse together to create a single cellular layer, called the syncytiotrophoblast, which is essential to a fetus for drawing nutrients from its mother. The scientists discovered that in order to fuse together, the cells must first make syncytin.

What made syncytin peculiar was that it was not a human gene. It bore all the hallmarks of a gene from a virus.

Viruses have insinuated themselves into the genome of our ancestors for hundreds of millions of years. They typically have gotten there by infecting eggs or sperm, inserting their own DNA into ours. There are 100,000 known fragments of viruses in the human genome,  making up over 8% of our DNA. Most of this virus DNA has been hit by so many mutations that it’s nothing but baggage our species carries along from one generation to the next. Yet there are some viral genes that still make proteins in our bodies. Syncytin appeared to be a hugely important one to our own biology. Originally, syncytin allowed viruses to fuse host cells together so they could spread from one cell to another. Now the protein allowed babies to fuse to their mothers.

It turned out that syncytin was not unique to humans. Chimpanzees had the same virus gene at the same spot in their genome. So did gorillas. So did monkeys. What’s more, the gene was strikingly similar from one species to the next. The best way to explain this pattern was that the virus that gave us syncytin infected a common ancestor of primates, and it carried out an important function that has been favored ever since by natural selection. Later, the French virologist Thierry Heidmann  and his colleagues discovered a second version of syncytin in humans and other primates, and dubbed them syncytin 1 and syncytin 2. Both virus proteins seemed to be important to our well-being. In pre-eclampsia, which gives pregnant women dangerously high blood pressure, levels of both syncytin 1 and syncytin 2 drop dramatically. Syncytin 2 also performs another viral trick to help its human master: it helps tamp down the mother’s immune system so she doesn’t attack her baby as a hunk of foreign tissue.

In 2005, Heidmann and his colleagues realized that syncytins were not just for primates. While surveying the mouse genome, they discovered two syncytin genes (these known as A and B), which were also produced in the same part of the placenta. This discovery allowed the scientists to test once and for all how important syncytin was to mammals. They shut down the syncytin A gene in mouse embryos and discovered they died after about 11 days because they couldn’t form their syncytiotrophoblast. So clearly this virus mattered enormously to its permanent host.

Despite their name, however, the primate and mouse syncytins didn’t have a common history. Syncytin 1 and 2 come from entirely different viruses than syncytin A and B. And the syncytin story got even more intricate in 2009, when Heidmann discovered yet another syncytin gene–from an entirely different virus–in rabbits. While they found this additional syncytin (known as syncytin-Ory1) in a couple different species of rabbits, they couldn’t find it in the close relative of rabbits, the pika. So their own placenta-helping virus must have infected the ancestors of rabbits less than 30 million years ago.

Now Heidmann has found yet another virus lurking in the ancient history of mammals. This one is in dogs and cats–along with pandas and hyenas and all the other mammals that belong to the so-called carnivoran branch of the mammal tree. In every carnivoran they’ve looked at, they find the same syncytin gene, which they named syncytin-Car1. In every species it is strikingly similar, suggesting that it’s experienced strong natural selection for an important function for millions of years. But it’s missing from the closest living relative of carnivorans, the pangolins. The diagram here, from the authors, shows how they see this evolution having unfolded. After the ancestors of carnivorans split from other mammals 85 million years ago, they got infected with a virus which eventually came to be essential for their placenta.

The big picture that’s now emerging is quite amazing. Viruses have rained down on mammals, and on at least six occasions, they’ve gotten snagged in their hosts and started carrying out the same function: building placentas. The complete story will have to wait until scientists have searched every placental mammal for syncytins from viruses. But in the meantime there is something interesting to consider. Some mammals that scientists have yet to investigate, such as pigs and horses, don’t have the open layer of cells in their placenta like we do. Scientists have come up with all sorts of explanations for why that may be, mainly by looking for differences in the biology of each kind of mammals. But the answer may be simpler: the ancestors of pigs and horses might never have gotten sick with the right virus.

(For more information on our inner viruses, see this 2010 story I wrote for the New York Times and my book from last year, A Planet of Viruses.)

[Top image: Leonardo da Vinci’s sketch of a human fetus. From Universal Leonardo]


Comments (52)

  1. Fascinating article… thanks!

  2. What a remarkable discovery

  3. Daniel

    “What made syncytin peculiar was that it was not a human gene. It bore all the hallmarks of a gene from a virus.”

    While it may have originated from a virus, I can’t see how it is not a human gene, being present in and essential for humans.

  4. Daniel

    “While they found this additional syncytin (known as syncytin-Ory1) in a couple different species of rabbits, they couldn’t find it in the close relative of rabbits, the pika. So their own placenta-helping virus must have infected the ancestors of rabbits less than 30 million years ago.”

    Maybe, but couldn’t the syncytin gene have been present in both rabbits and pika in the past, but pika lost it after developing an alternative?

  5. Daniel

    Very interesting article. Thanks.
    It sounds like horses and pigs will be found to lack the syncytin gene.

  6. RA Jensen

    Sperm and egg mutations may have multiple causes. An important study was published a few months ago that examined the frequency of sperm mutations in workers at a benzene manufacturing plant in China. The study recruited 30 workers who had worked at the benzene manufacturing plant for more than a year and divided the workers into three groups, a low exposure group, a moderate exposure group and a high exposure group. The study included a control group of 11 unexposed workers from the same town.

    Every participant in all four groups was found to have de novo sperm mutations including 1p36 sperm mutations. The frequency of the sperm mutations was lowest, but present, in the unexposed group, higher in the low exposed group, higher still in the moderate exposed group and highest in the high exposed group. The 1p36 deletion syndrome is present in 1 in 5,000 to 10,000 newborns:

    The 1p36 deletion syndrome is also associated with co-occurring autism:

    This is the first study that has demonstrated a direct connection between a specific de novo sperm mutation (1p36 deletion), a specific severe genetic syndrome (1p36 deletion syndrome) and a specific environmental pathogen (benzene). Benzene, because of its high octane number, is an important component in the production of refined gasoline and diesel fuels.

    For a further in-depth discussion visit:

    Marchetti F, Eskanazi B, Weldon RH et al (2011). Occupational exposure to benzene and chromosomal structural aberrations in the sperm of Chinese men. Environ Health Perspect

  7. I wonder why, if it is clearly so important, did this not happen in the common ancestor of all placental mammals? Why occur multiple times in the evolution of the individual groups? It seems strange to me.

  8. jemand

    Well there are different depths to which placentas in various species entwine with the mother. I know humans are on the more extreme end… what about the other species with these virus proteins? Do they also have more extreme connections between fetus and mother?

  9. Steve Carballeira

    Interesting research. Do you know if any marsupials have been tested for the viral gene?

  10. Bobby LaVesh

    Evolution can not occur without God producing viruses to direct it. God controls all life on earth through viruses.

    Kidding- but you know that’s what *they* will say. Creationism died. Intelligent Design is dying. Intelligent Virus will be the next wave.

  11. Chuck Pell

    It seems odd that syncytin isn’t fundamental to and present in all placental mammals.  Either the multiple invasions are apparent in only those placentals that survived virally induced modification of the basic process, or maybe placental mammals are not monophyletic.  That would be a very cool discovery, eh?

  12. Chuck–this work doesn’t challenge the copious evidence that the mammal placenta only evolved once. It suggests that viruses modified a simple ancestral placenta in many lineages. While it’s not fundamental to the original placenta, syncytin has become essential for the placenta, as shown by studies such as the mouse gene-knockout experiments.

  13. We are all genetically modified organisms

  14. “The concept that viruses might play a fundamental role in the evolution of the complexity of cellular life, as here proposed, may seem novel to many, especially to evolutionary biologists (Villarreal, 2004, p. 310).”

    Your article reminded me of past discussion on the evolutionary psychology yahoo group that resulted from my mention of the 1532 genes and interactions among them in the context of sexual reproduction that requires the mammalian placenta (Lynch, Leclerc, May, & Wagner, 2011). I learned about the likely involvement of viruses when I was asked about the role of pheromones in species specific communication as might occur in a new human species—detailed by Greg Bear in 1999/2003 in two of his science fiction novels (see Bear, 2004). But his questions, presentation to the American Philosophical Society, and comments led me to LP Villarreal (2004), and other more recent works that contain background information on the involvement of viruses in the creation of new species (Villarreal, 2009; Villarreal & Witzany, 2009). These works make it more clear that “…viruses may well be the unseen creator that most likely did contribute to making us human (Villarreal, 2004, p. 322).”

    Did they do this by epigenetically altering food acquisition behavior, pheromone production, and thus social behavior? Perhaps viruses are calibrating all mechanisms for both individual survival and for speciation across all species. If so, shall we credit them for all of Creation?

    Bear, G. (2004). When Genes Go Walkabout. Proceedings of the American Philosophical Society, 148(3), 324-331.
    Lynch, V. J., Leclerc, R. D., May, G., & Wagner, G. P. (2011). Transposon-mediated rewiring of gene regulatory networks contributed to the evolution of pregnancy in mammals. Nat Genet, 43(11), 1154-1159.
    Villarreal, L. P. (2004). Can Viruses Make Us Human? Proceedings of the American Philosophical Society 148(3), 296-323.
    Villarreal, L. P. (2009). Origin of group identity: viruses, addiction and cooperation. New York: Springer.
    Villarreal, L. P., & Witzany, G. (2009). Viruses are essential agents within the roots and stem of the tree of life. J Theor Biol, 262(4), 698-710.

  15. reader

    Article error: the virus uses it not to fuse host cells together but to fuse itself to a potential host cell

  16. Kudos! This story is utterly fascinating! With advances in molecular tools, I cannot wait to see what the future has in store for us! The linear, hierarchical view of evolution is unraveling at an accelerating rate! Horizontal gene transfer, transposable elements, and the like will likely be credited for having drastic consequences to current evolutionary theory!

    My only question is this: What will emerge as the new metaphor if we no longer have a tree of relatedness, but rather, something more intertwined?

  17. Alison Hunter

    Really cool story, thanks!
    I think some of the confusion arises from the wording ” the same function: building placentas” . Maybe that should be altered to indicate it does not build the whole placenta, but alters the way that the placenta interacts with maternal tissue, as Carl & jemand indicate in their responses.

  18. Joe Ififa

    God really is a genius !!He planned for everything in this universe.
    Nothing happens by chance

  19. zmil


    “Article error: the virus uses it not to fuse host cells together but to fuse itself to a potential host cell.”

    There are a number of viruses that induce syncytia formation in their hosts, for example, respiratory syncytial virus (naturally) and HIV. I believe some use this as a way of spreading from cell to cell with less contact with the immune system. I have seen speculation that this is an important pathway for HIV in some tissues.

  20. One of THEM

    One of THEM (see #10) wants to know what the supposed virus infected. If the gene is essential to survival, didn’t the original host already exist. Leading me to suspect that “bore the hallmarks” is not actually proof of viral origin.
    Please watch the flames. I am not being sarcastic, only asking questions.

  21. Michael

    I’m a bit confused here. The opening paragraph states “If not for a virus, none of us would ever be born.” In the concluding paragraph we hear “Some mammals […] don’t have the open layer of cells in their placenta like we do.”

    If this is the case isn’t it possible that we could still have been born if we had evolved placentas without the open layer of cells i.e. without the virus gene?

    [CZ: The mouse experiments demonstrate that once mammals have virus-derived syncytin genes for a few million years, those genes become essential. If we were pigs without syncytin genes, then yes, we might still be here. But we wouldn’t be blogging.]

  22. The question remains: Did endogenous retroviruses cause placental development in the skink: Trachylepis ivensi? Or, simply put, did viruses make lizards, too? Did God make the viruses that made lizards and mammals, or did they Create themselves? And how did the novel cell types evolve that are required for sexual reproduction at its advent in yeasts? Did viruses cause sexual reproduction in unicellular organisms?

    Okay, I can count; that’s more than one question that remains.

  23. Since 99% of the genetic material, as well as 90% of the cells in a human are microbes, it is likely that microbes fit somewhere in this picture.

  24. mohamed

    I would like to know if there is any possibility to any virus to be fused with another one, to be infected by other virus?
    As we know the infection concept is just a conscious term or scientific term to us, since it causes us to be sick [unwanted health condition] due to our reaction with micro-organisms such as viruses. If viruses are independent [not related to us] organisms, how would they got their compatible receptors to ours on our cells?
    In Naica mine, a mine is best known for its extraordinary selenite crystals in the Mexican state of Chihuahua, virologists have discovered millions of viruses in many samples from underground water and crystals. The temperature in such a mine is nearly 50 degrees and the humidity is nearly 100%, after sucking out the emerging water in the cave, roughly for million years ago.
    Maybe such these conditions are suitable to form the basic molecules for any living cells such as amino acids which are the basic building blocks of viruses. What if more than aboriginal or primitive viruses or separated molecules fused to form a single genetic material to generate new generation(s) or advanced virus(s)? Consequently, forming more advanced cells to form the aboriginal creatures such as bacteria, abiogenesis?

  25. Dr. Rolf

    My atheist brother desperately says there can’t be a God because he made bacteria and virus that attack humans. That argument is gone.

  26. jack cheetham

    it sort of begs the question: what came first, the primitive primate or the syncytin, was the primitive primate gestating in some way other than in a placenta? or…was the syncytin replacing an earlier protein involved in placental attachment to the uterus…i could go on, but i think i made my point…

  27. Greg Bear’s novels came immediately to my mind as well.

    “They typically have gotten there by infecting eggs or sperm, inserting their own DNA into ours.”

    My understanding of how germline viruses work (and there are some documented cases out infecting people as we speak, e.g., in Latin America), is rather more involved than your explanation, at least in males. While eggs are created early and then doled out in women, reducing their mutagenic succeptibility, sperm are routinely created and replaced and a germline virus doesn’t have to hit a man close to the time of infection to be effective.

    My understanding is that a germline virus will generally change the genome of the infected person throughout their body in many or all of their cells, and that this in turn gets reflected in sperm production, rather than infecting sperm cells directly.

    I’m not a virologist or a geneticist and I am willing to stand corrected. I also recognize that the post tries to sum up a complex biological process in a single sentence and employs weasel words to make clear that the mechanism that it describes isn’t a universal rule. But, if the process really is more complicated, it probably deserves a footnote at least disclaiming the idea that it is really that simple.

  28. Thanks! I had a lot of information here, useful for my work. But, I am asking the opportunity for debating a different approach about these phenomena, based in the viewpoint of Matrix/DNA Theory.

    The existence of virus, the origins of placenta and mammalian reproductive apparatus are explained by the Matrix formula that was present in the state of the world before biogenesis. Hence, this information here about syncytin, the fuses, already was predicted by this theory 30 years ago, as proved by copyrights.

    What’s virus? There is a universal natural formula for all natural systems and viruses are the performer of systemic function number 5 at cells systems (See the formula at my article “Nós, Mamíferos Fomos Feitos Tambem Por Virus! Mamma mia! (Portuguese – fevereiro | 17 | 2012). Google it.

    How was the origin of placenta and mammalian reproductive apparatus? It is explained in my article “From Reptile to Mammals: A Heroic Act? (English – November 13th, 2011)”

    There is a controversial point between the two theories. Matrix suggests that the original genes that later produced syncytin already were registered in the primordial DNA and they are there, at the junk DNA. But they were prohibited to express because the evolution of biological systems obeys the same chronology ordered by the formula and the formula first makes system laying eggs out, only in a later phase the system keeps the egg within. So, in parallel was evolving the virus which DNA had several mutations at the point that some genes are not re-cognoscible when faced with its similar genes at the living beings’ junk DNA.

    There is this possibility: the living beings genes produced the viral genes which had mutations and went back to living beings DNA when evolution determined that it was time for their expression.
    Viruses, as performers of F5, are supposed to be cells organelles. This function build tools for reproduction and perpetuation of species, then it appears only at intervals in the cell system. Viruses are the performers of male functions, the cells tool as their spermatozoon. The cause evolution had keep this viral organelle outside the cell system is because they have the bias to reproduce by re-cycling closed systems. Then, the virus keeps orbiting around the female apparatus (in this case, the uterus) and mixing at ingredients and cells in that region they produces syncytin.

    Everything equal when the Matrix formula was building the galaxies building blocks: uterus is the black hole at the galaxy nucleus, placenta is the formation of events’ horizon with stellar dust. Fetuses are seeds of a new star and virus are the comets that make the fecundation, fusing the black hole with the external Universe, which, in this biological case, is the mother’s body.
    This is a different theory, maybe has a lot of wrong things in their models, but his ability to make predictions has been incomparable with others theories.

  29. Besides the debate among different theories, there is this practical and urgent case related to these viruses:

    Wikipedia: Clinical significance

    HERV-W has been associated with multiple sclerosis and schizophrenia in humans. (HERV-W_7q21.2 provirus ancestral Env polyprotein also known as Env-W or enverin or syncytin is a protein that in humans is encoded by the ERVWE1 gene.

    Then, my job just now is searching information everywhere about these diseases and trying to understand it from the Matrix/DNA formula viewpoint.

  30. Torbjörn Larsson, OM

    Viral proteins are the best! (Well, at least sometimes.)


    What will emerge as the new metaphor if we no longer have a tree of relatedness, but rather, something more intertwined?

    Those who argue that want to say bush.

    But the fact is that if it happened on a genome level, we couldn’t resolve trees as robustly as we do. Even if biologists at times have to use models that acknowledge the sometime lateral transfer of pieces of genomes, such as matrix methods.

    Also fascinating how many crackpots crawled out of them cracks:


    If you believe there is no randomness, I have this card game I can interest you in.

    In other words, you propose the utterly impossible.


    The “original host” wasn’t a host, it didn’t use the viral gene because it wasn’t around, yet the host-to-be was viable then. That is why it is called evolution.


    Too much gods, sex and viruses in your questions to make sense.

    #28, # 34:

    [enough said]


    Are you kidding? One good outcome can never explain away the religious problem of the existence of utter misery. Your gods have to take responsibility for what you claim they do, as adults do.

  31. John B.

    When I started this article I expected to learn that a single viral insertion in the ancestor of all placental mammals had somehow made the placenta possible, but it’s obviously a lot more complex than that. But if the viral syncytin genes are so essential to placental function, how did the ancestors of the various lineages carrying these genes make do without them? And how do the placental mammals that lack these genes manage to survive? Another way to look at it is that these genes are not (or rather were not originally) essential at all, but as perfect parasites have coerced their hosts into requiring them in order to ensure their own replication.

    [CZ: Not all placental mammals have a syncytiotrophoblast layer in their placenta. That feature appears to have evolved several times–with the help of several viruses. Once it did, it became essential to those lineages’ development. Hence the lethal effects of knocking out the virus gene in mice.]

  32. To everybody,

    These guys are NOT looking at viruses or remnants there of. They are studying transposable elements. Commonly thought of as remnants of ancient virus integrations, these genetic elements turn out to be part of the regulatory genome. Transposable genetic switches, they regulate the transcriptome and are the basis to understand the origin of variation (i.e. differential gene expression). The idea that these elements are remnants of RNA viruses is non-sensical. It is the other way around: RNA viruses have their origin in the genome, to be precise in a particular type of Transposable Element known as ERV. At last, we know where viruses originate: In the genome from TEs!

    For more information on the novel hypothesis that TE’s are in fact variation-inducing genetic elements (VIGEs) and the predecessors of RNA viruses have a look here (scroll down all the way down and open my papers:


    [CZ: To everybody, um, yes, these guys are looking at viruses. Endogenous retroviruses are clearly viruses. When they are only recently integrated into host genomes, they can still produce infective viral particles. And scientists have been able to reconstruct infective viral particles from mutated virus sequences, resurrecting viruses after millions of years. These viruses belong to the same group of viruses as HIV and retroviruses that cause various cancers. There are a huge number of papers on this subject–go to PubMed and type in “human endogenous retroviruses” and start reading.

    This commenter tells us that all this research is “non-sensical,” and claims that these genetic sequences in genomes are the origin of viruses, rather than the other way around. He points us to a series of papers published in something called the Journal of Creation. (See for yourself.) As far as I can tell, those papers do not address the huge amount of research in recent years in which scientists have reconstructed the evolutionary history of viruses as they enter host genomes and then mutate in the descendants of their hosts. Here’s a Loom post on one such study.]

  33. No, Mr Zimmer, I did not tell you that the research is non-sensical…what is non-sensical is regarding these pieces of regulatory DNA as the remnants of viruses, while we do not have a single clue where viruses originate. Apparently, RNA viruses simply originate in the genome: from TEs. For the Rous Sarcoma Virus you can simple follow the whole “evolution”: from VIGE via HERV-K to RSV. HERV-K only had to take a small part of the SRC gene from the genome and it became an oncovirus. That is the most parsimonous explanation for the existance of this RNA viruses. In my opinion, this is a general mechanims to account for the origin of RNA virusen. Hence, viruses are secondary mistakes and not original design. To top it up, this view would solve the RNA virus paradox.


  34. Plants have transposons (e.g. Bs, Cin & mu-elements) to induce variation, fungi have transposable TY-elements to induce variation, insects have transposons like gypsy, copia and P-elements to induce variation, bacteria have IS-elements to induce variation, mammals have ERVs, LINEs and SINEs to induce variation in offspring. These mobile genetic elements are all remnants of ancient virus invasions? Of course not! The induction of variation in offspring is preprogrammed. It may have been highly specific and controlled, but now degenerate.

  35. Dear Mr Zimmer,

    Thanks for spreading “the VIGE–>RNA-virus meme” on twitter.

    Question to you: where do viruses have their origin. I mean, before they start to evolve (change) they must have their origin somewhere, isn’t it?

    Some believe they come from space. The better solution is: they come from the genome. They have their origin in transposable elements knowns as ERVs. Look into the genetics of HIV, influenza and other RNA viruses. You will observe that they have the usual core genes (gag and pol) like the harmless VIGE-family known as ERVs.

    In addition, the RNA viruses have acquired genes from the genome in which they reside. HIV has acquired a piece of the CCR5 ligand (to be exact: a piece of the interleukin-8 gene) to enter immune cells. Influenza acquired the neuramidase gene, which is an essential gene found in the genome of mammals and functions to modificy proteins.

    In my opinion, evolutionists have turned the world upside down. They believe that viruses made mammals, but the opposite is true. Viruses destroy mammals. The origin of RNA viruses can be found in harmless TEs (type: ERV), which are particularly good at putting genes in new regulatory contexts (to generate novel variation in offspring). Hence, RNA viruses are derailed VIGEs.


  36. Jeff Clemente

    Many would think such a title would be surprising. We find this article interesting because a virus, something commonly believed to be detrimental to our health, is actually something that has survived in the human race for millennia. Many associate the word “virus” with the rhinovirus (common cold) or HIV (Human immunodeficiency virus), which are very detrimental. But, as this article points out, syncytin, which allows host cells to stick to one another, can be beneficial when present in the right places. Evolution simply favors beneficial traits, and this trait certainly helped developing fetus’ so it stayed in the human gene pool.

  37. Ancient virus integrations…???. Did you guys ever hear about the RNA virus paradox? Maybe you should look into it? Remnants of viruses that make up the major part of the observed genomes is just main stream ignorance that is not going to bring us anywhere. The idea that RNA viruses originate from ERVs and not the other way around solves the paradox.

    Read my papers on the evolutionary informatics labs and you will understand the origin of RNA viruses and the diseases they spawn.



Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

The Loom

A blog about life, past and future. Written by DISCOVER contributing editor and columnist Carl Zimmer.

About Carl Zimmer

Carl Zimmer writes about science regularly for The New York Times and magazines such as DISCOVER, which also hosts his blog, The LoomHe is the author of 12 books, the most recent of which is Science Ink: Tattoos of the Science Obsessed.


See More

Collapse bottom bar