Reactive oxygen species (ROS) have been implicated in the pathogenesis of fibrosis. However, it remains unclear which ROS is the major cause. We hypothesize that superoxide elicits specific toxicity to human lung fibroblasts and plays an important role in the development of pulmonary fibrosis. In this study, superoxide generated from xanthine and xanthine oxidase activated lung fibroblasts by increasing the release of TGF-beta1 and collagen. This was associated with increased levels of intracellular superoxide. SOD and tempol, by scavenging respectively extracellular and intracellular superoxide, prevented the activation of fibroblasts induced by exposure to exogenous superoxide, whereas catalase did not. Moreover, hydrogen peroxide did not activate fibroblasts. Apparently, superoxide rather than hydrogen peroxide is involved in the regulation of TGF-beta1 and collagen release in lung fibroblasts. The chloride channel blocker, DIDS, inhibited the increase of intracellular superoxide levels induced by exogenous superoxide and consequently prevented the activation of fibroblasts. This suggests that the cellular influx of superoxide through chloride channels is essential for superoxide-induced activation of fibroblasts. ERK1/2 and p38 MAPKs are involved in the intracellular pathway leading to superoxide-induced fibroblasts activation. Superoxide possesses until now undiscovered specific pro-fibrotic properties in human lung fibroblasts. This takes place via the cellular influx of superoxide through chloride channels rather than via the formation of hydrogen peroxide.