Professor Andrew McMahon and his research group discovered that the gene Wnt-4 helps control normal female sexual development. Photo by Rose Lincoln.

It's an old idea: embryos develop as females until and unless they receive specific genetic instructions to become males. Now, biologists at Harvard University have proved this to be wrong, at least for mice.

"It's really nonsense to imagine that making a female is a passive event," says Andrew McMahon, professor of molecular and cellular biology.

Females have two sex chromosomes, called "X" chromosomes. Males have an X and a Y chromosome. Many people thought that if an embryo didn't develop a Y chromosome, it became female by default.

"It makes no sense to think that making a female reproductive system requires no female input," McMahon comments. However, he never imagined the idea would be disproved in his laboratory, and by accident.

McMahon's team of researchers was investigating how a gene called Wnt-4 is involved in kidney development. The gonads lie close to where kidneys form in both mouse and human embryos. While examining kidneys, Seppo Vainio, a postdoctoral fellow, spotted some abnormalities in mice with mutant Wnt (pronounced "went") genes. It appeared to him as if the females were developing male sex organs.

Vainio speculated that mutant Wnt-4 genes were masculinizing the females. McMahon, his boss, wouldn't buy it.

"I was very skeptical," McMahon recalls. "But I let him pursue the idea."

McMahon remained skeptical and Vainio continued the pursuit for three years. Finally, Vainio proved his point: an embryo must possess a properly functioning Wnt-4 gene to become a female. Without it, she does not develop a normal oviduct, uterus, or vagina; instead, she develops male structures such as sperm ducts.

But is the same true for humans? "Everything about the regulation of sexual development that has been looked at in mice translates pretty well to humans," McMahon answers. "All mammals probably share the same genetic requirements for reproductive development."

Vainio, McMahon, and their colleagues reported their conclusions in the Feb. 4 issue of the journal Nature.

Starting Out Male and Female

Mice and humans start out having both male and female reproductive systems, and both have master genes that guide their development. In females, the gene is called Dax-1; in males, it's SRY, a gene on the Y chromosome.

"It looks like a battle occurs between the male and the female gene," McMahon says. "The winner determines sex."

Wnt-4 is a leading general on the female side. When it functions normally, the gene suppresses the male sex system by preventing production of the hormone testosterone. At the same time, Wnt-4 initiates development of the Müllerian duct, which gives rise to the oviduct, uterus, and upper vagina. Apparently, the gene also plays a role in egg development; biologists aren't totally sure how this occurs.

As McMahon's team discovered, a mutated Wnt-4 interferes with this process and masculinizes the female.

Since both sexes start out with the rudiments of male and female reproductive systems, one system must be eliminated. Males release a hormone called Müllerian Inhibiting Substance (MIS), which destroys the female duct. Males also make testosterone to promote differentiation of the rudimentary Wolffian duct into sperm ducts (epididymis and vas deferens) and seminal vesicles.

In normal females, lack of MIS allows development of female organs, while the absence of testosterone leads to degeneration of the male parts.

"These scenarios are undoubtedly oversimplified," McMahon admits. "Wnt-4 by itself does not turn a testicle into an ovary." There are other genes involved, principally the testis-forming gene on the Y chromosome. Also, a number of genes act like Wnt-4 to regulate development of different reproductive structures. For example, last year McMahon's group found that destruction of the Müllerian duct in males, and development of the uterus in females, also depend on a cousin gene known as Wnt-7a.

In mice, this sex-determining activity occurs between 11.5 days after fertilizationand birth, usually at 19 days. In humans, it starts during the first three months of pregnancy.

McMahon did not expect to find that Wnt-4 plays a key role in sex determination. He was studying the part this gene plays in making nephrons -- filter cells that remove wastes from the kidneys. "It was a complete surprise to find Wnt-4 involved in two completely separate and independent functions in the body," McMahon comments.

He will continue working on Wnt-4's role in kidney construction and to experiment with Wnt-4 in male reproduction. McMahon wants to determine if this gene will turn off male development by halting the production of testosterone.

If sex can be juggled by adding or subtracting key genes in mice, would it be possible to engineer human embryos so they are born as males or females? "No, not with Wnt genes," McMahon answers emphatically. "These genes affect development in only part of the reproductive system, the internal part. Externally, mice still develop male or female genitalia."

In other words, an individual may have male external organs, like a penis, together with rudimentary female internal organs. A female may have rudimentary sperm ducts.

"Such individuals," notes McMahon, "are almost always infertile."