Information from Lay-Language Summaries is Embargoed Until the Conclusion of the Scientific Presentation
308—Biomarkers and Imaging in Schizophrenia
Monday, November 11, 2013, 8:00 am - 11:15 am
308.12: Dopamine, salience and the risk of psychosis
*T. T. WINTON-BROWN1, J. P. ROISER3, O. D. HOWES2, N. BUNZECK4, E. DUZEL5, S. KAPUR6, P. MCGUIRE2; 1Psychosis Studies, 2Inst. of Psychiatry, London, United Kingdom; 3UCL Inst. of Cognitive Neuroscience., London, United Kingdom; 4Univ. Med. Ctr. Hamburg-Eppendorf, Hamburg, Germany; 5Inst. of Cognitive Neurol. and Dementia Research, Otto-von-Guericke Univ. Magdeburg, Magdeburg, Germany; 6Inst. of Pscyhiatry, London, United Kingdom
Abstract Body: This study unpacks the concept of ‘salience’ as invoked by Kapur and others in linking dopamine dysregulation to psychotic symptoms1. Rather than relying on unidimensional reward based conceptions that have dominated studies so far it reconsiders the notion from a healthy salience processing point of view. It sets out to then test this model using fMRI and PET scanning in unmedicated subjects at high clinical risk for psychosis in the neural setting of a subcortical network derived from the MAM animal model of schizophrenia, that predicts abnormal hippocampal outputs drive striatal dopamine dysfunction in psychosis2. We developed a novel factorial fMRI task that utilizes visual probes of Novelty and Emotion alongside aspects of Reward, and operationalizes salience as the extent of modification provoked in reaction time, delayed recognition accuracy and fMRI activation. We found significant behavioral and fMRI main effects and interactions of each probe in a sample of 29 healthy controls, corroborating the multidimensional nature of salience. We then found several specific departures from this in a sample of 29 demographically matched participants with an At Risk Mental State for psychosis, specifically overactivations to reward in the ventral pallidum and midbrain, and altered interactions of emotion with reward in the ventral striatum and ACC-Insulae. Finally, in half of the sample we additionally obtained 18-F-DOPA PET scans to test the predictions of the MAM model. We found specific alterations in the relationships between ventral hippocampal activation to salient stimuli and striatal presynaptic dopamine synthesis in high risk subjects compared to controls. These were largely as predicted by the MAM model, and again most prominent in the domains of reward and emotion. These results resonate with the affectively laden altered motivational states seen in early psychosis, and with modern cognitive models that emphasize the importance of emotion in producing psychotic symptoms3. They broaden our understanding of normal salience processing, and add support to aberrant salience models of early psychosis, suggesting that abnormalities in salience processing are present prior to the onset of the first psychotic episode. They also support predictions from the MAM model of schizophrenia, that ventral hippocampal overdrive may provoke striatal hyperdopaminergia in psychosis. They therefore also point towards this upstream target for potential new treatments. 1.Kapur, S., Mizrahi, R., & Li, M. (2005). 2. Lisman, J. E., & Grace, A. A. (2005). 3. Freeman, D., & Garety, P. A. (2003).
Lay Language Summary: Why do we notice some things and ignore others? How does the brain know what is important and needs further attention and action what can be discarded? What would happen if this basic mechanism broke down so that irrelevant and minor things suddenly seemed vitally important? This could be a clue to what happens in psychosis, where sufferers feel confused and overwhelmed, and eventually experience a breakdown in their experience of reality, with unreliable perceptions like hallucinations, and hard to understand beliefs, like delusions. These were the questions asked in this study, which started by trying to develop an understanding of the basic mechanisms of ‘salience’ processing - how the brain detects and sorts important sensual stimuli from the rest. By using fMRI (functional Magnetic resonance Imaging) scanning we compared the healthy brain’s responses to three aspects of pictures- novelty, emotion and monetary reward - and measured reaction time to the pictures, and then how long they lingered in memory afterwards. These responses were collected, compared and then related to dopamine levels measured by PET (Positron Emission Tomography) scanning - dopamine (a key neurotransmitter in the brain) is an important part of how this mechanism works. As we expected each of these 3 aspects provoked brain, behavior and memory changes, and they interacted with each other, with emotional aspects being particularly salient and memorable. Responses in the hippocampus - a part of the brain important for context and memory - also related to dopamine measured with PET, suggesting that the dopamine levels may be influenced upstream by the hippocampus, as has already been shown in mice. We then compared these to responses in a group of subjects who were experiencing early psychotic symptoms. We found heightened responses to the reward aspects of salience and to pictures that combined reward and emotion. We also found that the hippocampus - dopamine relationship was altered, particularly for these reward and emotional aspects. This has important implications for our understanding and treatment of psychosis. They suggest our mechanism of sorting and prioritizing the world is disturbed in the early phases of psychosis, and that this implicates the neurotransmitter dopamine. It also suggests that the hippocampus may be a useful new target for regulating dopamine levels more effectively. This will provide the focus of our next studies aimed at discovering new treatments.
Neuroscience 2013 (43rd annual meeting of the Society for Neuroscience)Exit