On neuroinflammation in psychotic disorders

Abstract

Neuroinflammation has been implicated in several psychiatric conditions. Based on experimental studies, the glia-derived tryptophan metabolite kynurenic acid (KYNA) may be especially relevant for positive psychotic symptoms and impaired executive functioning.

The first two studies (I and II) of this thesis translate these experimental findings to bipolar disorder patients. Rooted in a genome-wide association study against cerebrospinal fluid (CSF) concentrations of KYNA, we used a multi-pronged approach and linked the identified genetic risk marker not only with other CSF biomarkers in bipolar patients, but also with specific symptoms such as delusions, hallucinations, and impaired executive functioning. We conducted cell studies, postmortem analyses, and clinical association studies that together suggested a sorting nexin 7 driven activation of caspase-8/IL-1ß as a mechanism underlying increased CSF concentration of KYNA in psychotic bipolar patients. Caspase-1 is activated by the purinergic ionotropic receptor P2X7R. This receptor is suggested to be internalized by a G-protein coupled receptor kinase 3 (GRK-3)-dependent mechanism. Decreased protein levels of GRK-3 have been observed in postmortem studies of schizophrenia and psychotic bipolar patients.

In study III, we used a mouse with a targeted deletion for GRK3. These mice displayed impaired P2X7R internalization, increased brain levels of IL-1ß, increased immunoreactivity for the astrocytic marker glial fibrillary acidic protein (GFAP), a more pronounced accumulation of hippocampal KYNA, as well as an accentuated dopaminergic response to amphetamine. In behavior models, these animals displayed disrupted pre-pulse inhibition, as well as impaired contextual fear conditioning with spared cue-specific fear conditioning. Taken together, these findings suggest that the GRK3-/-mouse is a novel genetic animal model of schizophrenia that may prove useful in exploring the actions of the emerging immunomodulatory drugs in psychotic disorders.

Study IV was a validation study aiming at defining an algorithm that identifies bipolar disorder patients in Swedish national registries as accurate as possible, a pre-requisite for study V. In the last study V of this thesis, we studied the association between psychotic disorders and rheumatoid arthritis (RA). We found that the previously reported inverse association is likely to have been confounded by underreporting and/or underdiagnosis of RA. However, an inverse association between schizophrenia and seronegative RA may be real, tentatively due to shared genetic underpinnings involving glia functioning.

In conclusion, this thesis suggests an important role of glial mechanisms in the pathophysiology of the two main psychiatric disorders, schizophrenia and bipolar disorder. Our results add to growing evidence that cytokine and kynurenine metabolite signaling is tied to psychotic and cognitive symptoms. These findings open up for novel drug targets and imply that contemporary glia research can provide a rewarding foundation for investigations into pathology of psychiatric disorders.