Defining the disturbance in cortical glutamate and GABA function in psychosis, its origins and consequences

Schizophrenia is a common disorder which typically begins in the late teens and twenties. Often there is a period (the prodrome) of gradual decline in motivation, interest and sociability before the acute onset of psychotic symptoms such as hearing voices and having threatening paranoid beliefs. The psychotic symptoms usually respond quite well to antipspychotic drugs but there are often residual symptoms when the psychosis has died down and patients are left with a degree of apathy that leads to a poor quality of life. We do not have medications that reverse or prevent these residual symptoms. Finding better drugs is difficult because we do not know what the underlying brain changes are; if we did we could develop drugs targeted on the process and reverse the illness or prevent the prodrome progressing to psychosis.

Increasing evidence from brain imaging studies suggest that subtle changes to the grey matter of the brain are occurring in the prodrome that continue into the acute phase. There is much enthusiasm for the idea that chemical messengers in the grey matter (neurotransmitters) called glutamate and GABA are somehow bound up with the process of becoming psychotic and with the residual state. Much of the interest comes from the effect of drugs like phencyclidine that can induce a state like a psychosis. However, to really know whether there is something wrong with glutamate and GABA we need to measure its release and activity in living people. It is possible to measure these chemicals using a technique called magnetic resonance spectroscopy (MRS). At high magnetic field strengths, the different compounds can be clearly separated and measured. A related technique allows us to go a step further and measure how much glutamate neurones are actually releasing. This method has never been applied to a brain condition. We will use these spectroscopic methods to give a decisive yes or no to the question of whether glutamate and GABA are abnormal, either early on the illnesses or in those with more than 10 years of illness.

We also want to know what might cause glutamate/GABA abnormalities. There is a good case that some form of inflammatory response may be involved in acute psychosis that dies down having left some mild damage that accounts for the residual symptoms. We will check this using the most sensitive Positron Emission Tomography (PET) camera in the country. It detects tracers that bind to inflammatory cells in the brain and this is clearly seen in diseases such as Parkinson's disease. We might find that glutamate problems were present in those with PET evidence of inflammation. Or it might be that they are independent risk factors.

Finally we want to know whether the glutamate/GABA changes actually produce symptoms and how they might do this. We can use magneto-encephalography (MEG) to detect tiny magnetic fields that brain cells induce outside the head when they fire. We are beginning to understand that different parts of the grey matter communicate with one another by firing in step to produce waves of activity. This has revealed that different networks do different jobs in the brain such as focussing attention or remembering things. Glutamate and GABA keep cells firing in step with each other and so abnormalities in these neurotransmitters may produce symptoms by affecting how networks operate.

Measuring MRS, PET and MEG together in the same people would be ideal but very demanding. We have devised a series of overlapping pairs of tests that will enable us to finally settle whether glutamate and GABA are functioning abnormally in schizophrenia, whether inflammation is anything to do with the process and how symptoms might result. The results are potentially game-changing and could point the way to new drug treatments and re-invigorate the interest of industry in developing new treatments for schizophrenia.

This proposal will identify if dysregulation of glutamate release initiates and sustains the development of schizophrenia in humans. The role of excessive glutamate release and whether it results from oxidative stress or inflammation, or perturbs long and short range cortical networks in patients will be investigated. Regional brain measurements in patients with early onset and established illness and age-matched controls will compare aspects of glutamate function and mechanisms which may lead to disease. 13C and 1H magnetic resonance spectroscopy (MRS) will measure glutamate, glutamine, GABA and glutathione and the conversion rate between glutamate and glutamine. These will indicate: the balance between excitation (glutamate) and inhibition (GABA); the rate of neurotransmitter release (conversion); and the oxidative status of the brain
regions (glutathione). Studies of gamma oscillations using magneto-encephalography, already known to be modulated by GABA in normal brain, will establish whether cortical networks are disturbed in schizophrenia and linked to abnormal neurotransmitter function. Positron emission tomography (PET) will be used to assess the burden of inflammation, determining whether a postulated link to schizophrenia exists and whether this too is related to alterations in neurotransmitters and brain oxidative stress. This proposal should confirm or refute a role for glutamate in causing schizophrenia. If refuted, efforts can be concentrated on other targets, but if confirmed, this will open new avenues for the development of targeted psychological and pharmacological therapy and provide biomarkers to evaluate them with.

This proposal will deliver impact by meeting the criteria for this call, in the following ways:
-generating major new mechanistic insights into a human disease: The onset and development of schizophrenia
-understanding the role of glutamate and GABA in this process through investigation in humans in vivo
-addressing a substantial gap in understanding the role of glutamate in psychosis for which there is strong circumstantial but little experimental evidence in vivo
-engendering application to new therapeutic approaches directly targeting pathogenesis developed through reverse translation to more basic research.

This will be achieved through an ambitious, innovative and highly collaborative study: 3 novel techniques studied in 3 institutions across several disciplines. The approach is highly integrative, identifying quantitative changes in glutamate and GABA in psychosis and also understanding how they are caused and produce symptoms.

Impact on clinical research:
Definitive evidence for a role of glutamate and GABA in psychosis would have a game-changing impact on clinical research if: changes were associated with causal origins via inflammation or reactive oxygen species; evidence was provided of a neurophysiological-cognitive route to symptoms; and evidence was provided that the changes related to the stage of illness. Disease-associated findings in any one of these domains would be major novel findings.

Positive findings would stimulate further studies in schizophrenia, for example in high-risk groups to determine whether some of the measures present in early psychosis can be seen prior to its onset. This in turn would lead to studies to identify preventative therapies based on pathogenesis that do not carry the stigma or risk of current antipsychotic drugs. With more sensitive and specific imaging tools to assess glutamate function, relationships with dopamine function would become clear.

The techniques could also be of wider use in psychiatry and neurodegenerative disorders. The ability to measure GABA and glutamate, using widely available 3T scanners, and the application in patients of novel methods for measuring their release, would also stimulate clinical research in depression and bipolar disorder where the role of glutamate and GABA has not been resolved by static MRS measures. Similarly, MEG oscillatory biomarkers - direct measures of integrated activity within local networks of pyramidal cells and inhibitory interneurons - should provide evidence of the functional role of neurotransmitter release and the effects of disease on both local cortical excitation/inhibition balance and long-range network integrity/connectivity. This tight link between our novel biomarkers and the underlying functional neurobiology, coupled with our ability to link both to symptoms and perceptual/cognitive performance will provide a step-change in the provision of relevant clinical information.

Novel treatments, better outcomes, prevention, loss of stigma:
The pharmaceutical industry has dramatically reduced its efforts in mental health drug-development, mainly through a lack of understanding of the neurobiology of such diseases and a failure to identify new drug targets. The identification of a new disease mechanism will stimulate drug research from preclinical through to clinical stages of development. There are many ways in which glutamate activity can be modulated but finding a presynaptic glutamate or GABA pathogenesis would focus attention on release modulators whose target engagement could be detected in humans in vivo. The results might also focus interest on anti-oxidants and anti-inflammatory drugs. The dependent variables could be used to identify subgroups with different pathologies that might operate at different stages of illness. Using the measures for stratification could delineate subgroups responsive to novel drugs or to drugs whose benefits have been missed in previous un-stratified studies.