Carbon monoxide (CO) is an anti-inflammatory molecule at low concentrations. It has been shown that this gas attenuates the inflammatory response, prevents acute lung and liver injury and limits vascular damage associated with ischemia-reperfusion following organ transplantation. The mechanism of its action however remains partially unknown and some ongoing studies show differential effects depending on cell type and the nature of the stress. In this study we evaluated the ability of CO to induce reactive oxygen species (ROS) production in macrophages, once these cells play a pivotal role during inflammation. We exposed RAW 264.7 macrophages to CO (500 p.p.m.) and the generation of ROS was evaluated by flow cytometry over time using the fluorescent probe dihidrorhodamine (DHR). We observed a transient time-dependent increase in DHR fluorescence, suggesting that CO can induce ROS in these cells. Another probe for ROS, H₂DCF-DA, was used and the induction of ROS was shown to be blocked in the presence of N-acetyl cystein (NAC), an antioxidant molecule, reinforcing the idea that CO induces ROS. We elected LPS-induced TNF- production as the readout for the CO anti-inflammatory effects. Inhibition of LPS-induced TNF- by CO was reversed in macrophages that were pretreated with PEG-modified superoxide dismutase and catalase prior to CO and LPS, showing that the anti-inflammatory effect of CO relies on its ability to increase ROS in these cells.  Our next hypothesis was that CO would affect hemeproteins present in the electron transport chain (ETC) of mitochondria, which could explain the generation of ROS elicited by this gas. To confirm this, we used ETC inhibitors and observed that pre-incubation of the macrophages with antimycin A (AA), a selective inhibitor of complex III, resulted in a substantial loss of CO-induced ROS production. We measured the activity of cytochrome c oxidase in RAW 264.7 macrophages exposed to CO and we observed an inhibition of this enzyme. This event was reproduced either by direct incubation with CO or by overnight incubation with heme, suggesting that inhibition of complex IV of the ETC could be the mechanism by which CO increased ROS production. Taken together, our results suggest that the mitochondria is one of the cellular targets of CO and that partial inhibition of respiration, and the consequent increase in ROS, seems to have a paramount importance in conferring the anti-inflammatory properties of this gas.