Effect of fibroblast growth factor 2 on stromal cell-derived factor 1 production by bone marrow stromal cells and hematopoiesis.

BACKGROUND: Reduction of intramedullary hematopoiesis and the development of myelofibrosis and splenic hematopoiesis are frequent complications of clonal myeloid disorders that cause severe morbidity and death and present a therapeutic challenge. However, the pathogenesis of these complications is still unknown. We evaluated the effect of fibroblast growth factor 2 (FGF-2), the level of which is elevated in patients with clonal myeloid disorders, on bone marrow stromal cell expression of stromal cell-derived factor 1 (SDF-1), a chemokine that is essential for normal hematopoiesis.
METHODS: Reverse transcription-polymerase chain reaction analysis, immunoblot analysis, and enzyme-linked immunosorbent assays were used to examine effects of human recombinant FGF-2 exposure on SDF-1 expression in mouse stromal MS-5 and S-17 cells. Cocultures of human CD34-positive peripheral blood stem cells or mouse pre-B DW34 cells with mouse stromal cells were used to characterize the functional relevance of the effects of FGF-2 on SDF-1 expression. The in vivo hematologic effects of FGF-2 were determined by systemic administration to mice (n = 10). All statistical tests were two-sided.
RESULTS: FGF-2 reduced constitutive SDF-1 mRNA expression and secretion in stromal cells (SDF-1 levels in supernatants: MS-5 cells cultured for 3 days in medium only versus in medium with FGF-2, 95.4 ng/mL versus 22.2 ng/mL, difference = 73.2 ng/mL, 95% confidence interval [CI] = 60.52 to 85.87 ng/mL; P = .002, two-sided Student's t test; S-17 cultured in medium only versus in medium with FGF-2, 203.53 ng/mL versus 32.36 ng/mL, difference = 171.17 ng/mL, 95% CI = 161.8 to 180.6 ng/mL; P<.001). These effects of FGF-2 were reversible. FGF-2 compromised stromal cell support of the growth and survival of pre-B DW34 and myeloid lineage cells, and these effects were reversed in part by exogenous recombinant SDF-1alpha (rSDF-1alpha) (DW34 pre-B cells recovery on S-17 stromal cells, expressed as a percentage of DW34 cells recovered from medium only: with FGF-2 versus without FGF-2, 27.6% versus 100%, difference = 72.4%, 95% CI = 45.34% to 99.51%, P = .008; with FGF-2 plus rSDF1 versus with FGF-2 only, 60.3% versus 27.6%, difference = 32.7%, 95% CI = 9.35% to 56.08%, P = .034; fold increase in number of myeloid lineage cells after culture on S-17 stromal cells: with FGF-2 versus without FGF-2, 0.25-fold versus 3.8-fold, difference = 3.55-fold, 95% CI = 2.66- to 4.44-fold, P<.001; recovery of myeloid cells on S-17 stromal cells, expressed as a percentage of myeloid cells recovered from medium only: FGF-2 plus rSDF-1alpha versus FGF-2 only, 76.5% versus 32.4%, difference = 44.1%, 95% CI = 32.58% to 55.68%, P<.001). Administration of FGF-2 to mice reversibly reduced bone marrow levels of SDF-1 and cellularity and induced immature myeloid cell mobilization, extramedullary hematopoiesis, and splenomegaly.
CONCLUSIONS: Systemic administration of FGF-2 in mice disrupts normal bone marrow hematopoiesis in part through reduced expression of SDF-1. Thus, endogenous FGF-2 may represent a potential therapeutic target in clonal myeloid disorders characterized by bone marrow failure.