Understanding the language of reproduction.

Chemical conversations between developing mammalian egg cells and the cells that surround them are conducted in a language scientists don't understand and through channels they couldn't identify-until now. By deleting a single gene from mice, researchers have identified one such channel and may at last be on the way to understanding the dialogue that helps an immature egg cell develop.

The gene the researchers deleted encodes connexin 37, a protein found in ovaries. Connexin 37 seems to link the developing eggs, or oocytes, and the surrounding cells, called granulosa cells. Female mice lacking the gene are infertile, report David L. Paul of Harvard Medical School in Boston and his colleagues in the Feb. 6 Nature.

The finding confirms previous suspicions that connexin 37 forms a channel through which ions, metabolites, and signaling molecules can pass.

Oocytes and granulosa cells together make up the follicles of a normal ovary. As a follicle develops, the number of granulosa cells increases. The oocyte grows and undergoes the biochemical changes that lead to maturity. It also begins meiosis, the process that gives it a single, rather than double, set of chromosomes. When the mature egg cell is expelled from the ovary, granulosa cells change into the steroid-secreting cells needed to support a developing embryo.

"When connexin 37 is absent, the whole process falls apart," says Paul. Neither the granulosa cells nor the oocyte grows properly, and the oocyte fails to start meiosis. The granulosa cells still change into steroid-secreting cells, but the oocyte disappears.

The steps of the failed developmental process offer clues to the messages that the connexin 37 channels convey.

"We believe the granulosa cells may send at least two types of messages to the oocyte," Paul says. A positive signal tells the egg to grow and prepare for meiosis, he argues, but there's also a negative signal that delays completion of meiosis until ovulation has occurred.

A message also travels the other way, from the oocyte to the granulosa cells, Paul says. This message says, "keep on being a regular granulosa cell; don't differentiate into a steroid secretor."

To understand the process better, scientists need to identify the chemicals that make up the signals being passed through the channels, Paul says.

John Eppig, a reproductive biologist at Jackson Laboratories in Bar Harbor, Maine, calls the study "exciting" but cautions that scientists don't think connexin 37 carries all the communications between the oocyte and granulosa cells.

"This is an area where we are really just beginning to get a feel for how important and how complex the communication is between cells," Eppig says.

A possible next step to pinpointing the actions of the signals, Eppig says, is to remove the oocyte from mice lacking connexin 37. Normal oocytes removed from a follicle mature spontaneously, as if they had already received the positive developmental signal. If an oocyte removed from a connexin 37-deficient mouse failed to mature, it would suggest that the protein is necessary for that signal to be effective.