Objective. Even though there are clinical trials assessing granulocyte-macrophage colony-stimulating factor (GM-CSF) blockade in rheumatoid arthritis (RA), questions remain as to how GM-CSF acts as a proinflammatory cytokine. The aims of this study on the regulation of arthritis progression by GM-CSF were to determine the source of the GM-CSF, whether there are systemic effects, the changes in synovial tissue leukocyte populations, and the arthritis model dependence on GM-CSF.Methods. Bone marrow chimeras were used to determine the source of GM-CSF required for the development of collagen-induced arthritis (CIA). The K/BxN serum-transfer model of arthritis was tested in GM-CSF ؊/؊ mice and using anti-GM-CSF monoclonal antibodies. Cell populations from arthritic mice were assessed by differential staining and flow cytometry.Results. In the CIA model, GM-CSF produced by bone marrow-derived cells was required for arthritis development. GM-CSF blockade, while ameliorating the development of CIA, was found to have systemic effects, limiting the increase in circulating Ly-6C high monocytes and neutrophils. GM-CSF blockade led to fewer synovial macrophages (both Ly-6C high and Ly-6C low ), neutrophils, and lymphocytes. In the absence of GM-CSF, K/BxN serum-transfer arthritis initially developed normally; however, the numbers of Ly-6C high monocytes and synovial macrophages (both Ly-6C high and Ly-6C low ) were again reduced, along with the peak disease severity and maintenance.Conclusion. GM-CSF is a key player in two arthritis models, participating in interactions between hemopoietic cells, both locally and systemically, to control myeloid cell numbers as well as presumably to "activate" them. These results could be useful for the analysis of current clinical trials targeting GM-CSF in patients with RA.Granulocyte-macrophage colony-stimulating factor (GM-CSF) was initially discovered by its role in the differentiation of hemopoietic precursor cells into mature granulocytes and macrophages (1); however, it can also affect mature cell function and may be considered a proinflammatory cytokine (2-4). In this context, clinical trials are in progress testing this concept in rheumatoid arthritis (RA) (4).GM-CSF is produced in vitro by a number of different cell types following exposure to various stimuli (4-8). A so-called CSF network has been proposed in which GM-CSF and other CSFs form part of a cytokinemediated hemopoietic cell-tissue cell interaction that drives chronic inflammatory reactions, for example, in joints (9). In support of this concept, several models of disease, including arthritis (10-12), multiple sclerosis Professor Hamilton has received consulting fees from MorphoSys AG and CSL, Ltd. (less than $10,000 a year). The University of Melbourne has licensed to MorphoSys AG, Germany, patented technology relating to therapeutically targeting granulocyte-macrophage colony-stimulating factor, for which licensing fees and milestone payments have been made.