Mammals lost their egg yolk genes after acquiring genes for milk proteins, according to a study published yesterday in PLoS Biology. The results pinpoint an important step in how mammals evolved, the authors say.

Lactation is "what makes us mammals, basically," said Henrik Kaessmann, who led the study. "Using egg yolk genes as markers, we found a unique way to put a timeframe on how key transitions in mammals occurred."

There are three types of mammals: true placental mammals, marsupials and monotremes. Though each type nourishes its young in a different way, they all use milk to some extent, and their eggs have far less yolk than their reptilian and bird-like ancestors. But in the evolution of mammals, there's a longstanding "chicken and egg" question, or rather, a milk and egg question: What came first in the mammalian lineage — genes involved in lactation, or the loss of genes for making egg proteins? Now researchers at the University of Lausanne in Switzerland have cracked the problem.

Kaessmann and his colleagues compared the sequences of genes encoding vitellogenin, an essential egg yolk protein, in the different mammalian lineages. They found that the three genes present in the mammalian ancestor were progressively lost in all lineages except for the monotremes, which have retained one working gene. These primitive mammals, which include platypuses and echidnas, lactate, yet lay small, parchment-shelled eggs. So the presence of both functional and non-functional vitellogenin genes is consistent with this intermediate reproductive state, said Kaessmann.

The researchers also found that all three groups of mammals shared major milk resource genes, called caseins, indicating that these genes arose over 200 million years ago, before the split of the mammalian lineages. Putting findings from both sets of genes together, Kaessmann argues that lactation in the common mammalian ancestor, followed by the emergence of placentas in some mammals, probably allowed for the loss of yolk-nourishment in mammals.

"Everything makes sense," said Jay Storz of the University of Nebraska in Lincoln, who was not involved in the research. He told The Scientist he was "surprised that the progressive loss of the yolk genes coincided so nicely with the origins of lactation and placental development." Nigel Finn of the University of Bergen in Norway agreed that the evidence was "quite convincing," but he thinks the picture of mammalian evolution is still incomplete. The transition to mammalian reproduction was not just about nutrition, he argues, mammals had to overcome a water problem as well, and the study does not address the evolution of amniotic fluid, which nourishes the embryo and keeps it moist. "Yolk is a luxury, but water is essential," he said.