SAA has been shown to have potential proinflammatory properties in inflammatory diseases such as atherosclerosis. These include induction of tumor necrosis factor α, interleukin-6, and monocyte chemoattractant protein 1 in vitro. However, concern has been raised that these effects might be due to use of recombinant SAA with low level of endotoxin contaminants or its non-native forms. Therefore, physiological relevance has not been fully elucidated. In this study, we investigated the role of SAA in the production of inflammatory cytokines. Stimulation of mouse monocyte J774 cells with lipid-poor recombinant human SAA and purified SAA derived from cardiac surgery patients, but not ApoA-I and ApoA-II, elicited pro-inflammatory cytokines like granulocyte colony stimulating factor (G-CSF). However, HDL-associated SAA failed to stimulate production of these cytokines. Using neutralizing antibodies against toll like receptor (TLR) 2 and 4, we could evaluate that TLR 2 is responsible for G-CSF production by lipid-poor SAA. To confirm these data in vivo, we expressed mouse SAA in SAA deficient C57BL/6 mice using an adenoviral vector. G-CSF was identically expressed in SAA-Adenoviral infected mice as well as in control null-Adenoviral mice at the early time points (4–8h) and could not be detected in plasma 24h after infection when plasma SAA levels were maximally elevated, indicating that adenoviral vector rather than SAA affected G-CSF levels. Taken together, our findings suggest that lipid-poor SAA, but not HDL-associated SAA, stimulates G-CSF production and this stimulation is mediated through TLR 2 in J774 cells. However, its physiological role in vivo remains ambiguous.