I've been reading science-fiction pretty much all my life. I usually go through "phases" when I hit on a particular author and read several books by the same person. Last year I was in my Greg Bear phase and I have read eight of his books. He is one of those writers who gets better with age: more recent his book, more I liked it.
His is also some of the hardest of hard sci-fi around. He must be a really smart guy - he has a blog, after all! Bear really thoroughly researches whatever area of physics, astronomy, geology or biology he needs for his next novel. At a book reading last year, several people in the audience voiced their difficulties in understanding the science parts of his novels, while they thoroughly enjoyed the characters and plot. As a biologists, I found the science easy to understand, yet riveting anyway.
After I read his last two novels, "Darwin's Radio" and "Darwin's Children" I decided to check on his science - it sounded very good, yet so fantastic at the same time. What I found was a surprise: the real science is really that fantastic! Greg only needed to add a very little twist in order to turn it from fiction into science-fiction. So, here is some of what I discovered (though I am a biologist, this is way out of my area of expertise, so assume this is a lay-person writing).
Retroviruses are, well, viruses - short chains of nucleotides that enter the cell of a host and incorporate into the DNA of the host. Viruses we usually think about, like flu or herpes, use the genetic machinery of the host cell to make many copies of themselves, including protective protein coats that allow them to leave the cell and go and infect other cells.
Endogenous retroviruses (ERVs) do not have the capability to leave the host cell and infect other cells (though some may produce particles that remain within the cell). It is thought that ERVs have lost that ability over millions of years of evolution (or is it parasite-host coevolution?) due to mutation and fragmentation of their sequences. They have been incorporated in our genomes for a very long time. It is estimated that as much as a third of our genome may consist of bits and pieces of old retroviruses - remnants of ancient viral infections.
ERVs have been found in many species of mammals and the absence/presence of particular ERVs, or mutations within various ERVs, are often beeing used to estimate times when two lineages split from each other. Endogenous retroviruses found in the human genome are designated as HERVs (Human ERV), in pig genome PERVs (Porcine ERV), etc.
What Greg Bear did in his novels is to allow one of the HERVs to (re)evolve the capability to leave the cell and organism and infect another organism (or fetus). While doing so, it affects the patterns of transcription of many other genes in the human genome during embryonic development, leading to developmental changes in a number of subtle anatomical, physiological and behavioral traits - changes large enough for the "virus children" to be considered a new species. The novels are particularly good at describing how the new race is being treated by the xenophobic society. Of course, Bear is a novelist, so the new traits he picked are those that make for a really good story. Those traits are not any more or less probable than any others he could have picked (e.g., high sensitivity of the vomero-nasal organ).
Most animal genomes appear to possess ERVs, yet mammals have a whole class of them not found in any other group of animals. Those appear to be a product of reverse transcription of cellular RNA. Why just mammals? Greg Bear points to a paper discussing a possible evolutionary explanation:
Viviparous mammals confront an immunological dilemma in that mammals which have highly adaptive immune systems fail to recognize their own allogenic embryos. The relationship of mammalian mother to her fetus resembles that of a parasite and host in that the fetus 'parasite' must be able to suppress the immune response of the 'host' mother in order to survive. As viviparous mammals are also noteworthy for having genomes that are highly infected with endogenous retroviruses and as retroviruses are generally immunosuppressive, the possible participation of endogenous retroviruses in the immunosuppression by the embryo was then considered. In addition, it was considered if such endogenous viruses might be more broadly involved in the evolution of their host and the resulting host genome that now appear to have many derivatives (such retrotransposons and as LINE elements) of such genomic viruses.
In many animals, endogenous retroviruses have been co-opted into new functions useful to the host. For instance, some parasitoid wasps inject so-called polydnaviruses into the egg or larva of the moth host. The virus serves to supress the immunity of the host, so that wasp eggs can survive within the host's body, hatch and the little wasps can eat their way out.
The paper quoted above proposes that placental mammals have co-opted one (or a small group of) HERV in building a connection between placenta and uterus in a way that does not compromise immunological isolation between mother and embryo. In a sense, this HERV allows the early embryo to implant into the wall of the uterus. Anti-HIV drug AZT, which is an inhibitor of reverse transcription prevents implantation of the fertilized egg, presumably by blocking the expression of the HERV.
The very-gently-stated bottom line of that paper - one that Greg Bear twists for literary purposes - is that without this particular virus that remained in our genomes since an infection millions of years ago, there would be no Placental Mammals!
Now, indulge me for a moment and let me go on a mental sci-fi trip for a moment. Let's go back 65 million years. Dinosaurs are suddenly extinct. Birds are flapping around and small mammals are hiding in the burrows. What happens next?
If the mammals did not receive this particular virus (or could not survive the infection), there would, perhaps, be no evolutionary origin of the placenta. Would we still have elephants and rhinos, or would this be a planet dominated by the giant platypus and kangaroos as big as T.rex, or perhaps enormous ravens and super-smart parrots?
Let me go onto a very thin limb here and propose that perhaps there would be elephants and rhinos but they would be very stupid...and there would be no great human civilization even if hominids arose at some point in history of Earth. Why?
How did dinosaurs get so big? They laid very small eggs, then grew at fantastic rates during early years of life. Such speedy growth required enormous amounts of energy. Every gram of food was converted into growth. The organ that is energetically most expensive to build and operate is the brain. In an animal that needs to be super-efficient about its energy, brain has to remain small, if enough individuals are to survive and reproduce for the species not to go extinct. Perhaps dinosaurs died out because they became too smart for their own good!
What does placenta do? There is no need to lay small eggs any more. Placenta allows the embryo to slow down its growth. It can grow slowly inside its mother for months (in larger mammals) and get pretty large before it has to start foraging for itself. It gets the luxury to, through some allometric and heterochronic changes, grow a bigger brain. The bigger brain, in turn, allows it to forage smarter. The mammals, once they acquired the placenta thanks to this HERV, are now free to get not just big, but also big-brained.
I know this is a hypothesis from a left field, but anyway, don't think of viruses only as bad guys that make you sneeze - a third of your genome is viral and who knows what aspect of you biology that you cherish so much may be directly traced to a remnant of a viral infection that Ducky the Platypus managed to survive a long time ago.