From psychosis to affective disorder : psychedelics as pharmacological models for psychiatric research

Recent studies into the molecular, pharmacological and behavioral basis of psychotomimetics such as the glutamate N-methyl- D -aspartate receptor (NMDAR) antagonist ketamine and the mixed 5- hydroxytryptamine (5-HT) receptor agonist psilocybin in healthy human subjects suggest that both the glutamatergic and serotonergic system are implicated in the pathophysiolology of psychotic disorders such as schizophrenia. Specifically, administration of ketamine to healthy humans reproduces positive and negative symptoms as well as cognitive impairments that are seen in acute and chronic schizophrenia, while the mixed 5-HT2A/1A receptor agonist psilocybin engenders positive symptoms and cognitive deficits that resemble the symptoms seen in incipient or acute phases of schizophrenia. Moreover, the observation that both ketamine and psilocybin also disrupt sensory gating (e.g. PPI, P50) in healthy subjects comparable to that seen in schizophrenia has led to the widespread use of these drugs to provide models for identifying neurobiological factors that are crucial to the pathophysiology of schizophrenia and to the development of novel treatments. Along this line, it has repeatedly been demonstrated that schizophrenia patients also show a reduced mismatch-negativity (MMN) event-related potential (ERP). The MMN is interpreted as a prediction error signal during implicit perceptual learning. The processing of prediction errors is of considerable importance in regard to psychosis, because recent theories posit that aberrant encoding of prediction errors may underlie the expression of psychotic symptoms. Furthermore, the MMN (i.e. prediction error processing) depends critically on NMDAR-dependent synaptic plasticity. In fact ketamine disrupts MMN responses in healthy humans comparable to those observed in schizophrenia. Given that neuromodulatory transmitters like 5-HT are thought to be implicated in the regulation of NMDAR- dependent synaptic plasticity during prediction error processing, it is conceivable that psilocybin may also, though via a differential mechanism, affect the MMN expression in humans. Based on this background, we investigated in the present thesis (chapter 2) whether the encoding of prediction error (via the assessment of the MMN expression) is affected by S-ketamine or psilocybin and whether the encoding of prediction errors under placebo can be used to predict drug-induced symptoms. In brief, we found that S-ketamine, but not psilocybin, disrupted the processing of prediction errors as expressed by a disrupted MMN expression over fronto-central brain regions. Although both drugs produced positive-like symptoms, we found that only S-ketamine produced severe cognitive impairments, the extent of which significantly correlated with the processing of prediction errors under placebo. Our results suggest that the NMDAR, but not the 5-HT receptor system, is critically implicated in the processing of prediction errors during MMN generation as a form of implicit perceptual learning and that aberrant prediction error signaling contributes to the formation of cognitive impairment in this pharmacological model. Our results provide further insights into the pathophysiology of key cognitive symptoms of psychotic disorders and suggest that the assessment of the MMN expression in schizophrenia may allow detecting early phases of the illness and might also serve to assess the efficacy of novel pharmacological treatments, in particular of cognitive impairments. Moreover, in chapter 3 of this thesis we used a computational model-based approach to examine whether the known NMDAR-mediated reduction of MMN expression can explain the changes in the plasticity of glutamatergic long-range connections among hierarchically related auditory areas.