Interleukin-6 (IL-6) is a pleiotropic cytokine involved in numerous physiological functions and it plays a critical role in acute, such as traumatic brain injury (TBI), and chronic inflammatory conditions, such as multiple sclerosis. In a mouse model of TBI such as cryolesion of the frontoparietal cortex, total Il6-deficient mice (IL-6 KO) showed a compromised inflammatory response compared to their controls. On the other hand, in experimental autoimmune encephalomyelitis (EAE), one of the most commonly used multiple sclerosis mouse models, IL-6 KO mice are resistant. IL-6 is produced by multiple peripheral and central nervous system cells such as neurons, astrocytes and microglia. However, the specific role of IL-6 derived from CNS in different neuroinflammatory contexts remains poorly understood. For this reason, we have generated astrocyte-, neuron- and microglia-derived Il6-deficient mice by crossing floxed Il6 mice with Cre recombinase mice, in which its expression is regulated by Glial fibrillary acidic protein (Gfap-Cre), Synapsin-1 (Syn1-Cre) and CX3C Chemokine Receptor-1 (Cx3cr1CreER) promoter, respectively. Then, mice were cryolesioned in frontoparietal cortex and/or immunized with myelin oligodendrocyte glycoprotein 35-55 (MOG35-55). Since the Cx3cr1CreER mouse is an inducible model, all derived animals were treated with tamoxifen, and the different experiments were performed 4 weeks after tamoxifen administration, ensuring that Il6 deficiency was specific to microglia. In addition, we have created and validated a new reversible conditional Il6 mouse (IL6-DIO-KO), which is characterized by a total Il6 deficiency and by the fact that Il6 expression can be reverted with cell-specific Cre recombinase. In this dissertation, we have reversed the microglial Il6 expression in the Il6-deficient mouse and studied its phenotype with the EAE model. Cryolesioned microglial Il6-deficient mice showed the transcriptional profile of the frontoparietal cortex dramatically altered one day after injury. However, they had a less altered glial response at 3 and 6 days post-injury, suggesting that compensatory mechanisms might be present. On the other hand, the Il6 deficiency from the different cellular sources of the nervous parenchyma did not prevent the EAE induction; however, the results revealed a cell-and sex-dependent effect. Astrocytic Il6 deficiency reduced clinical symptoms in females, while a trend was observed in those deficient in microglial Il6 during the disease peak. In contrast, the deficiency in astrocytic and microglial Il6 did not affect the clinical course in males. Moreover, Il6 deficiency in neurons did not alter disease development either in males or in females. In line with the phenotypic results, the inflammatory response was altered by astrocytic and microglial Il6 deficiency, but not by neuronal Il6 deficiency. Furthermore, the recovery of microglial Il6 in total Il6-deficient mice was sufficient to initiate milder paralyzing symptoms and trigger an inflammatory response against the MOG33-55 peptide. In conclusion, the results obtained in this dissertation suggest that IL-6 derived from CNS regulates neuroinflammatory processes in response to cryolesion and immunization with MOG33-55 peptide in a cell-dependent manner. For this reason, the study of IL-6 secreted by different cellular sources is essential to understand the functional complexity of IL-6.