One of the unwelcome effects of menopause is loss of bone density, or osteoporosis, which also affects patients with HIV and is a side effect of treatments for diseases such as prostate cancer and leukaemia. The protein RANKL, discovered only 12 years ago, quickly became a target for treatments against osteoporosis because it was shown to be a key regulator of bone physiology. In August, two phase III clinical trials showed that a therapy directed against RANKL reduced bone fractures in postmenopausal women and in men with prostate cancer1,2.

Work leading up to these trials included research by Josef Penninger, a functional geneticist and the founding scientific director of the Institute of Molecular Biotechnology of the Austrian Academy of Sciences in Vienna. His group was the first to demonstrate in vivo that RANKL stimulates the development of cells called osteoclasts, which break down bone tissue3. But RANKL and its receptor RANK are also produced in the central nervous system, where their functions were unknown. “It was a complete black box,” says Penninger. On page 505, his group reports the surprising finding that, in the brain, RANKL is a key regulator of body temperature.

This more recent work had its origins in what Penninger admits was something of a fishing expedition. Shortly after joining Penninger's lab in 2005, postdoc Reiko Hanada proposed injecting RANKL into the brains of rats, “to see what would happen”. What happened was that the rats became inactive and they became hot to the touch. The scientists realized that the animals were suffering from a severe fever.

To investigate this puzzling response, Hanada, Penninger and their colleagues had to study thermoregulation in rodents. This complex homeostatic mechanism incorporates information from a range of inputs, and is tuned to account for factors such as normal fluctuations in body temperature that occur when, for example, an animal's activity level changes. The authors also had to engineer ways to knock out expression of the Rankl gene selectively in neurons and astrocytes — the brain's two main cell types.

“We had to be really thorough, because this was so entirely new. And if you come up with something really novel, you'd better do your controls — as many as possible,” Penninger says with a laugh. “This paper probably breaks my record — 26 supplemental figures.”

Their experiments slowly revealed a pathway in the brain that starts with proinflammatory molecules called cytokines turning on RANKL. The same cytokines also activate RANKL in bone cells, but when the protein binds to its receptor in astrocytes in the central region of the brain, it initiates a protein cascade involving cyclooxygenase-2 and prostaglandin E2 to generate a fever response.

The ability to delve into such uncharted research waters fits in with his institute's philosophy, Penninger says. “Our vision is to hire great people with interesting ideas and provide them with a candy store of facilities where they can just jump around from technology to technology, to allow them to play with an idea.”