In a step that may someday help lead to new treatments for infertility, scientists produced live mice by using sperm-like cells derived from embryonic stem cells.

Seven mice resulted from the work, six living long enough to reach adulthood. However, the animals were unusually small or large and died within five months of birth, apparently because they lacked normal controls over gene activity.

Still, the work establishes a method for learning more about how sperm is formed and perhaps for finding new treatments for male infertility, researchers said.

Scientists must learn more about the sperm production process "to get to the root of infertility," the study's lead author, Karim Nayernia of the University of Newcastle upon Tyne in England, said in a statement.

Such knowledge could help scientists find a treatment for infertile men who can't produce mature sperm, although "this is several years down the line," he said.

The work is reported in this month's issue of the journal Developmental Cell.

Nayernia and colleagues started with mouse embryonic stem cells, highly versatile cells that can develop into any of the specialized cells of the body. In a laboratory, they used genetic and chemical techniques to derive cells that resembled precursors to sperm.

They injected these cells into 210 mouse eggs. That resulted in 65 embryos that were transplanted into female mice. Seven births followed.

That result is a significant advance, said Mary Ann Handel, who studies sperm formation and male infertility at the Jackson Laboratory in Bar Harbor, Maine.

Previously, other researchers had gotten as far as injecting such cells into eggs and producing early embryos, she said. But nobody else has reported live births, which demonstrates the potential of the injected cells, she said.

The new results "raise interesting questions for future research on how sperm cells are formed," she said.

John Gearhart, a stem cell expert at Johns Hopkins University, called the experiment "a nice illustration of the extraordinary power of embryonic stem cells." He said it's not yet clear how the mouse work would translate to human cells.