Though the average person musters about 10 billion white blood cells at any one time to fight off opportunistic pathogens, the half-life of neutrophils, which make up about 65 percent of the cells, is only about eight hours. To keep up the body’s defenses, according to a study in the Jan. 26 Cell by Gary Gilliland and colleagues, nature has evolved a molecular belt, suspenders, and elastic waistband for supporting the blood cell contingent.

This triple redundancy is provided courtesy of the FoxO family of transcription factors. First author Zuzana Tothova, an MD–PhD student, and coworkers found that while any one of FoxO1, FoxO3, or FoxO4 is sufficient for maintenance of hematopoietic stem cells in mice, absence of all three leads to a sudden surge, then a rapid depletion of these pluripotent cells. “These transcription factors are absolutely essential for normal life span and longevity of hematopoietic stem cells. Without them, the cells essentially flame out, and you would run out of blood in a matter of months,” said Gilliland, who is a Howard Hughes investigator and HMS professor of medicine at Brigham and Women’s Hospital.

The findings confirm what was long suspected but unproven, that FoxOs play an essential role in the development of blood cell lineages. But there is an unexpected twist. Tothova was able to reverse the FoxO-deficient phenotype by simply feeding mice a diet rich in antioxidants.

A Conditional KO

The FoxO transcription factors are involved in the regulation of basic cellular processes such as cell division and apoptosis. Their expression in cells of the hematopoietic lineage has been well documented. In fact, FoxOs may be important for preventing certain blood cancers because many leukemia and lymphoma oncogenes inactivate the transcription factors. Research has shown, however, that knocking out FoxOs individually has no significant effect on the hematopoietic system. Thinking this might be because the remaining FoxOs can compensate for the loss, Tothova and Gilliland, in collaboration with Ronald DePinho, HMS professor of medicine at Dana–Farber Cancer Institute, decided to study the hematopoietic system in conditional knockout mice that had been created in the DePinho lab.

The researchers engineered the animals with an inducible recombination system that excises FoxOs 1, 3, and 4 (the other family member, FoxO6, expressed only in the central nervous system, is left unscathed). By using a Cre recombinase driven by an interferon-dependent promoter, Tothova and colleagues were positioned to induce excision of Lox-flanked FoxO genes in adult animals. Administration of polyinosine-polycytidylic acid, which mimics a viral infection, drives an interferon surge that induces the recombinase and excises the genes. The system has the added benefit that gene knockouts are restricted to cells and tissues that express interferon, such as the blood and immune cells.

The Damage Done

When Tothova examined the mice four weeks after interferon activation, she found that the hematopoietic stem cells suffered most from loss of the transcription factors. Even though cells derived from the stem cells, such as common myeloid progenitors and megakaryocyte–erythrocyte progenitors, normally express the three FoxOs, there were just as many of these progenitors in FoxO-deficient mice as in wild type. In contrast, the numbers of long- and short-term hematopoietic stem cells were reduced eight- and three-fold, respectively. In addition, FoxO-deficient hematopoietic stem cells had abnormal cell cycle regulation with many more of the cells actively cycling than usual. While this could be viewed as potentially beneficial—more stem cells might mean more progenitors—the cells also exhibited increased rates of apoptosis. The increased death rate appeared to scupper any benefit derived from increased cycling of stem cells.

But it was the reactive oxygen connection that turned out to be most interesting. Tothova noticed that it was only FoxO-deficient hematopoietic stem cells that had a significant increase in reactive oxygen species (ROS). “That this was restricted to the stem cells suggested that there might be a mechanistic link between the elevated ROS and the cell cycle and apoptosis abnormalities,” said Tothova. To test this theory, she administered the antioxidant N-acetyl-L-cysteine to the mice for five weeks following FoxO-gene excision. The antioxidant completely reversed the FoxO-deficient phenotype—the HSC compartment size, cell cycling, and apoptosis in the cells returned to normal. “That was the cleverest part of this whole story, that Zuzana saw the mechanistic link between reactive oxygen and the stem cell phenotype,” said Gilliland.

Another Layer of Management

But it was this finding that was initially the most puzzling to the researchers. Gilliland explained that cells spawned by hematopoietic stem cells, such as myeloid progenitors, generate peripheral blood cells that protect the body against infection by killing bacteria and other pathogens with reactive oxygen species. So how do you derive “professional ROS-generating cells,” as Gilliland calls them, from stem cells that seem particularly sensitive to reactive oxygen? The easiest explanation might be that FoxOs, which can induce ROS scavengers like superoxide dismutase and catalase, are turned off in neutrophils and other myeloid cells; however, FoxOs are expressed throughout the hematopoietic lineage. Instead, Tothova found that in differentiated myeloid cells, there is a FoxO-independent induction of an entirely new suite of ROS-managing proteins.

Gilliland, whose major focus is leukemias and lymphomas, finds this aspect of the work particularly exciting. Because FoxOs are inactive in many blood cancers—in fact, in the same issue of Cell, DePinho’s group reports that FoxO deficiency leads to tumors in certain cell lineages—cancer stem cells may be exquisitely sensitive to reactive oxygen. Some leukemia oncogenes, for example, inactivate FoxO family members and presumably impair the cancer’s ability to manage the toxic effects of reactive oxygen species. “We might be able to target that Achilles heel in various ways, taking advantage of the fact that cancer stem cells are probably teetering on the edge of survival because they cannot manage reactive oxygen. We don’t know the answer to that yet,” said Gilliland, “but we are very excited about the possibility.” These findings also suggest that antioxidants could play a beneficial role in longevity of adult tissue stem cells. Although much work needs to be done to further explore this possibility, this hypothesis could have important implications for strategies that focus on tissue regeneration from adult stem cells.