NEW YORK (GenomeWeb) – The evolution of pregnancy, a distinctive trait in mammals, was driven by transposable elements that spurred genes to be turned on and off in uterine cells, marking a prime example of how genetic mechanisms contribute to the evolution of new traits, according to a study published today in Cell Reports.
The study, led by Vincent Lynch of the University of Chicago and Gunter Wagner of Yale University, reconstructed part of the evolutionary history of pregnancy by looking at gene expression profiles of uterine cells in multiple mammalian species and several non-mammalian ones.
"How we reproduce defines the kind of mammal that you are," Lynch told GenomeWeb. "All mammals have hair and they all have mammary glands, but they reproduce different ways." All mammals feature some sort of gestation period, from the platypus, whose eggs spend a period of several weeks developing in utero, to the elephant, whose gestation period can last almost two years. "We're starting to get a really good idea of how, mechanistically, things evolve into a major mammalian trait," Lynch said.
The transcriptome sequencing data were obtained with the Illumina Genome Analyzer II platform. The scientists collected new data from the dog, cow, horse, pig, armadillo, short-tailed opossum, and platypus, as well as human decidualized endometrial stromal cells. They also used existing data from Rhesus monkey and mouse endometria and used the chicken, lizard, and frog as outgroup species to compare the mammals to. With a complete data set of expression information for 19,641 genes among 14 species, they used parsimony to reconstruct the evolutionary history of pregnancy.
The researchers found that the evolution of pregnancy was driven partly by two processes: one where uterine cells stopped expressing certain genes and one where they started expressing existing ones they previously hadn't.
Genes that ceased to be expressed in the uterine cells were involved with ion transport, especially calcium and iron. Though still expressed in mammalian brain cells, they were no longer used in the uterus, where they appear to be important for making an egg shell.
On the flip side, genes where expression was gained were involved with cell signaling and regulating the immune system. "We think they're playing a role in maternal-fetal communication and suppressing the immune system in the presence of the fetus," Lynch said. "The mother's immune system should mount an immune response against the fetus. Evolution has found a way to get around that. How did mammals evolve the ability to keep this parasite in the uterus?"
While gene-by-gene accumulation is possible, Lynch and his co-authors showed that transposable elements could explain the gain of gene expression in mammalian uterine cells. "This is one of the first examples where we're starting to get at the basic genetic mechanisms for the evolution of a new trait," Lynch said.
Transposable elements, or transposons, are segments of DNA that have the ability to jump around the genome and insert themselves into new areas. Much of the genome, between 50 and 66 percent, according to Lynch, is derived from ancient transposable elements that are no longer moving around.
When those transposons jumped and inserted into new parts of the genome in ancient mammals, they brought regulatory information with them. "That information happened to say 'be expressed in the uterus,'" Lynch said. In total, the study found 194 distinct families of ancient mammalian transposable elements active in the decidual stromal cells that line the uterus during pregnancy.
"A lot of these transposons have binding sites for the progesterone receptor. All the genes nearby have the opportunity to be regulated by progesterone, where they ancestrally were not," Lynch said. "Because transposons have regulatory information they use for their own purposes, you can think of them as pre-existing switches that turn genes on and off." Because there are multiple copies of the same transposons throughout the genome, any gene they're nearby is going to get that same information. Each gene doesn't have to evolve its own way of responding to progesterone.
The ability to provide a mechanistic explanation of how evolution works on the molecular level was partly due to the relatively recent development of pregnancy. "There's still the history present in the genome. Given enough time all the changes get washed away. That information is present in the genome, so we can find it."
There's enough information left in the genome for scientists to order the changes that happened in the evolution of pregnancy.
"First, the uterus has to provide some way of giving resources to the fetus. Then the eggshell decreases in thickness, Lynch explained. When the eggshell finally disappears, the mother must evolve a way to communicate chemically with the fetus, which allows a longer pregnancy. "Once you have long pregnancy, you need some way for the mother to silence her immune response that should be directed against the fetus."
One of the most recent developments in the evolution of pregnancy remains to be explained: the degree to which the placenta burrows into the uterine wall, which has gone in two directions. "In everything with a hoof, that no longer happens. The placenta just grows until it touches the uterine wall and more or less stops." In animals like primates however, the placenta has become more invasive.
"It used to be thought that things with really invasive placentas got more nutrition, which could speed up fetal development" and perhaps allow humans to develop bigger brains, Lynch said. But whales and dolphins have large brains and their placentas aren't invasive.