A Blog by

Jumping DNA and the Evolution of Pregnancy

About a decade ago, Vincent Lynch emailed Frank Grutzner to ask for a tissue sample from a pregnant platypus. He got a polite brush-off instead.

Then, around eight years later, Grutzner got back in touch. His team had collected tissues from a platypus that had been killed by someone’s dog. They had some uterus. Did Lynch still want some?

“Hell yes!”

The platypus was the final critical part of a project that Lynch, now at the University of Chicago, had longed to do since he was a graduate student. He wanted to study the evolution of pregnancy in mammals, and specifically the genetic changes that transformed egg-laying creatures (like platypuses) into those that give birth to live young (like us).

The platypus enjoys a short pregnancy. Its embryo sits in the uterus for just 2-3 weeks, surrounded by a thin eggshell, and nourished by a primitive placenta. It then emerges as an egg. Marsupials, like kangaroos and koalas, also have short pregnancies. But mothers give birth to live young, which live in a pouch until they’re big enough. Other mammals—the placentals, or eutherians—keep their babies in the uterus for as long as possible, nourishing them through a complex placenta. Their pregnancies can be marathons—up to two years in an elephant.

The move from egg-laying to live-bearing was huge. Mammals had to go from holding a shell-covered embryo for weeks to nourishing one for months. To understand how they made the leap, Lynch compared 13 different animals, including egg-layers like the platypus, marsupials like the short-tailed opossum, and eutherians like the dog, cow, and armadillo. He catalogued all the genes that each species switches on in its uterus during pregnancy. He then compared these different sets to work out when mammals started (or stopped) using those genes during reproduction.

He found thousands of differences, many more than he anticipated. For example, hundreds of genes are involved in making eggshell minerals; they’re active in the uterus of a platypus but silent in those of other live-bearing mammals. Conversely, the marsupials and eutherians started activating hundreds of genes involved in suppressing the immune system, and in passing hormonal signals between the mother and foetus.

This all makes sense. A platypus embryo, during its brief stay in the uterus, is separated from its mother—and its mother’s immune system—by a shell. “It’s like the embryo has a cloak,” says Lynch. When mammals evolved live births, the cloak disappeared and a problem arose. Every foetus shares only half of its genes with its mother, so mum’s immune system should recognise this lump of growing tissue as a potential threat. To dispense with eggs, early marsupials and eutherians had to evolve ways of tamping down their immune responses, and only in the uterus. They also needed ways of exchanging signals with their embryos. “The foetus needs to say, Hey I’m here, and the mum needs to say, Oh, that’s okay,” says Lynch.

His study shows that they did so by repurposing a vast array of genes that already had roles in other organs, like the guts, brains, and bloodstream. But how? How does an animal deploy a gene—or thousands of genes—in a different organ?

The answer involves jumping DNA. Many bit of the genome can cut themselves away from the surrounding DNA and paste themselves in elsewhere. Others can copy themselves and insert the duplicates into new spots. These sequences are genomic parasites—they reproduce, often at the expense of their host. If they disrupt other genes when they land, they can cause cancer and other diseases. But sometimes, they settle somewhere useful.

Think of the jumping DNA as the infrared sensor in your television. The sensor recognises a stimulus—the signal from your remote control—and switches on the TV. Imagine that the sensor makes thousands of copies of itself, and somehow wires these into appliances all over your house. Now, when you press the remote control, your TV whirrs into life, but your lights also flicker on, your washing machine starts up, your computer boots, and your radio starts playing. By duplicating and spreading the sensor, you ensure that the same stimulus now turns on a multitude of things.

This is what happened during the evolution of pregnancy except there, the stimulus isn’t an infrared signal but a hormone called progesterone. In the ancestor of eutherian mammals, jumping DNA littered the genome with sequences that progesterone can recognise. They allowed this one hormone to switch on a vast array of new genes in the uterus. And they did so in a very short span of time by evolutionary standards—just a million years or so, by Lynch’s reckoning.

Craig Lowe from Stanford University says that, for decades, scientists have theorised that jumping DNA could do something like this, but Lynch has shown that they actually have. Lowe also suspects that other scientists will use the same methods to study the evolution of traits like pregnancy, which seem overwhelmingly complicated at first pass.

Indeed, that’s what motivated Lynch originally. He’s interested in how evolution produces radically new structures. “We don’t have a good understanding of how you get something entirely new,” he says. “It’s easy to imagine how you select upon an existing structure to get a slightly different one, like a hand into a flipper or a bat wing. But how do you get the limb to begin with?”

The answer almost certainly involved using existing genes in new and innovative ways. “Does that happen slowly and step-wise, or can you have broad, genome-wide changes that reorganise things in larger jumps? Our work suggests that the larger jumps are possible.”

Reference: Lynch, Nnamani, Kapusta, Brayer, Plaza, Mazur, Emera, Shehzad, Sheikh, Grutzner, Bauersachs, Graf, Young, Lieb, DeMayo, Feschotte & Wagner. 2015. Ancient Transposable Elements Transformed the Uterine Regulatory Landscape and Transcriptome during the Evolution of Mammalian Pregnancy. Cell Reports. http://dx.doi.org/10.1016/j.celrep.2014.12.052