Chronic granulomatous disease was for a long time seen only as an immunodeficient disorder, although its name reminds us all along of its granulomatous component. The exact nature of the granuloma started to be of interest to researchers in the recent years. Studies on patients provided us with the conviction that hyperinflammatory response in the colon is not caused by infections, but in the contrary responds well to immunomodulators. With the help of CGD animal models the mechanisms of hyperinflammation start to be unravelled. We demonstrated that hyperinflammation can be induced by fungal cell wall preparations. The major culprits are some of the main constituents of fungal cell walls, and in particular the curdlan. The resolution of the hyperinflammation caused by -glucan can take place in the wild-type but not in the CGD mice. This suggests that ROS play a role in the termination of the hyperinflammation. β-glucan-induced inflammation in CGD mice is associated with high amount of TNF,although this hyperinflammation is not TNF-dependent, as neither pharmacological inhibition with etanercept nor genetic inhibition did significantly dampen the exacerbated inflammation. In contrary, the involvement of dectin-1, the principal  glucan receptor, is a required step for the -glucan-induced CGD hyperinflammation. This thesis focuses on the apparently counterintuitive findings that NOX2 deficiency leads to a hyperinflammatory response. Since ROS are usually associated with inflammation, hyperinflammation in absence of ROS generation represents a paradox at first sight. A fine tuning in the balance of ROS production seems necessary as the ROSlow inflammatory response is also a mechanism for the host defence such as is the ROShigh inflammatory response.