When saving for retirement, people are advised to adjust investments as they age, sacrificing the potential for robust growth for protection against loss of assets. New work shows how neural stem cells adopt a similar strategy, trading regenerative capacity for protection against cancer.

In newborn mice, blood-forming cells (haematopoietic stem cells, HSCs) rely on a transcription factor known as Sox17 for self-renewal, but adult HSCs rely on a different transcription factor, Bmi-1. Sean Morrison and his colleagues at the University of Michigan in Ann Arbor wanted to know if similar shifts occurred with age in neural stem cells. They and others had already shown that one reason blood and other stem cells decline with age is because of increased expression of a tumour repressor gene called p16Ink4a.

A full-genome screen in blood cells showed that Hmga2, a small chromatin-associated protein, was the only protein that was highly expressed in HSCs compared to other blood cells and whose expression was lower in old adult mice than in young adult mice. Further investigation found a similar age-related expression pattern in neural stem cells.

Next, the researchers began a series of in vitro and in vivo experiments using mice that could not make Hmga2. (These mice have stunted growth and small brains but do not die early.) Compared with wild-type mice, young Hmga2-deficient mice had fewer stem cells and less self-renewal within the central and peripheral nervous systems; however, this difference disappeared when old Hmga2-deficient mice were compared with wild type1. Indeed, Hmga2 expression could not be detected in older wild-type mice. Experiments in neurospheres (cell clusters neural stem cells can form in culture) showed that as Hmga2 expression rose, expression of p16Ink4a and a related protein, p19Arf, fell. But what causes Hmga2 expression to decline with age? The researchers suggest this is due to the product of another gene, let-7b, which encodes a microRNA (miRNA), whose expression increases with age and which binds to Hmga2 mRNA. Stem-cell self-renewal was greater in neurospheres expressing a version of the mutant Hmga2 gene whose transcript that could not bind the let-7b miRNA.

“As Hmga2 is turned off in ageing stem cells by increasing let-7b expression, this allows Ink4a and Arf to be expressed,” explains Morrison. “The take-home message of the paper is that we have identified an entire pathway of genes that change in expression with age in stem cells.” This, in turn, reveals a complex mechanism of how stem cells trade the tendency to age as a way to ward off cancer.

Morrison's team is now investigating exactly how HmgA regulates the tumour suppressors p16Ink4a and p19Arf, but Jan van Deursen at the Mayo Clinic in Rochester, Minnesota,who is also investigating the role of p16Ink4a in cell senescence, says the mechanism is likely to become even more complex. Because HmgA2 alters chromatin, one would expect its effects to be more global than just affecting a handful of genes, he says. Furthermore, the genomic screen that identified HmgA2 was based on HSCs, he notes, “so additional mechanisms employed by other stem cells are entirely possible.”