Hematologic effects of recombinant murine granulocyte-macrophage colony-stimulating factor on the peripheral blood and bone marrow.

Recombinant murine granulocyte-macrophage colony-stimulating factor (GM-CSF) was noted to support rat bone marrow colony formation in vitro. The in vivo hematologic effects of a single intravenous injection of murine GM-CSF were therefore investigated. Doses of murine GM-CSF between 0.1 and 5 micrograms/rat caused an increasing leukocytosis that did not further increase with a dose of 25 micrograms/rat. In contrast, human GM-CSF at 25 micrograms/rat did not induce any significant peripheral hematologic effects. Murine GM-CSF induced peripheral neutrophilia and monocytosis, peaking between 4 and 8 hours and subsiding to baseline by 12 hours. Neutropenia and monocytopenia, which reached a nadir at 15 minutes, preceded the leukocytosis, suggesting that GM-CSF activates these leukocytes and causes transient intravascular margination. A mild lymphopenia occurred between 2 to 8 hours. The bone marrow at 6 hours after injection of GM-CSF demonstrated a variable and slight left-shifted myeloid hyperplasia most noticeable at the level of promyelocytes and myelocytes, suggesting a myeloproliferative effect. The marrow at 6 hours also demonstrated a decrease in mature neutrophils, documenting that the marrow contributes to the increased number of circulating neutrophils. Once-daily injection of GM-CSF for 7 days induced a repetitive daily neutrophilia of the same magnitude. The marrow after 1 week of injections did not show a generalized myeloid hyperplasia, but did show an increase in eosinophils and a decrease in lymphocytes. Granulocyte-macrophage colony-stimulating factor plus granulocyte colony-stimulating factor (G-CSF) have been reported to synergize in vitro in both mouse and human bone marrow colony assays. However GM-CSF plus G-CSF in vivo, administered as either a single injection or as daily injections for 1 week, were found in the present study to induce, at most, an additive effect on circulating numbers of neutrophils. It is concluded that murine GM-CSF will be useful in the rat model to study the in vivo hematoreconstitutive effects of GM-CSF alone and in combination with other hematologic growth factors. The relatively rapid kinetics and lesser magnitude of GM-CSF-induced neutrophilia and monocytosis, as compared to G-CSF and M-CSF, respectively, and the lesser myeloproliferative effect of GM-CSF in bone marrow smears, as compared to G-CSF, might be taken to suggest that GM-CSF's natural activity is predominantly as an inflammatory rather than a myeloproliferative factor.