North TE, Goessling W, Walkley CR, et al. . Nature. 2007;447:1007-11.

A chemical genetic screen for compounds affecting hematopoietic stem cell (HSC) survival/expansion has revealed that prostaglandin E2 promotes hematopoietic stem cell self renewal, and that inhibitors of PGE2, such as indomethacin, decrease functional HSC.

This study demonstrates the power of large-scale biologic screening to reveal new therapeutic approaches. The researchers screened more than 2,000 compounds including known U.S. Food and Drug Administration (FDA)-approved drugs. Zebrafish were used because they 1) are transparent, allowing visualization of developing blood cells; 2) mature quickly with blood cells visualized within 1.5 days of fertilization; and 3) can survive without blood, thus allowing compounds inhibiting HSC maintenance or differentiation to be assessed. The investigators previously showed that a line of flattened endothelial cells with a small cluster of hematopoietic cells can be visualized by probing for cells that co-express the zebrafish homologues of Runx1 (AML1) and cMyb. Zebrafish embryos were individually exposed to the compounds and HSC assessed by visualization of runx1+cmyb+ cells. Eighty compounds either increased or decreased HSC numbers, 10 of which were known to affect the prostaglandin pathway. On closer analysis, chemicals enhancing PGE2 synthesis increased HSC numbers, and those blocking prostaglandin synthesis decreased stem cell numbers. Data were confirmed by functional analyses of zebrafish HSC. Similar data were obtained for murine cells; PGE2 increased differentiation of murine embryonic stem cells to hematopoietic progenitors as assessed by colony-forming units and increased the frequency of functional HSC in competitive bone marrow transplant studies. Conversely, indomethacin and cox2 inhibitors decreased mouse hematopoietic stem and progenitor cells.

PGE2 and prostaglandin synthesis inhibitors have been known to affect hematopoiesis in human and mouse cells. Similar studies were reported previously; however, at that time, the data could not be completely interpreted to be affecting HSC because functional HSC assays were not yet well established. Addition of PGE to human bone marrow cells increased the number of CFU-GM and the proportion of cycling CFU-GM. It was proposed that PGE may have dual effects on hematopoiesis, limiting proliferation of CFU-GM at higher concentrations and promoting CFU-GM differentiation from earlier stem cells at low as well as high concentrations1 . In other studies, PGE enhanced proliferation of BFU-E and CFU-GEMM2 . In vivo administration of PGE2 in mice resulted in a decrease in CFU-GM, which, based on these new studies, could represent an effect on HSC self-renewal.

The data reported by North and colleagues are very important as they suggest that BM engraftment may be delayed by PGE2 inhibitors. Clinicians have routinely avoided prescribing NSAIDs or aspirin during engraftment or BM recovery due to the functional inhibition of already low numbers of platelets, and these new data suggest that specific cox2 inhibitors may need to be avoided as well. The data also suggest that stable forms of PGE2 or similar agents may enhance BM recovery following conventional-dose or high-dose chemotherapy. However, we need to take into consideration previous data showing that sustained administration of 16, 16 dimethyl PGE2 reduces restoration of hematopoiesis following cyclophosphamide administration3 . Additional studies will be required before we know for certain how to utilize the prostaglandin pathway for therapeutic manipulation of hematopoiesis.

Competing Interests

Drs. Krause and Pelus indicated no relevant conflicts of interest.