This is the second of three meeting reports from the recent annual meeting of the Society for Biological Psychiatry, held in Vancouver, British Columbia, Canada, 14-16 May 2009. We thank the Society and researcher C. Anthony Altar, NeuroDrug Consulting, Bethesda, Maryland, for his fine meeting missive.
6 July 2009. This plenary session on Saturday, chaired by Daniel Pine (NIMH), described how thin is not in when it comes to the neocortex and neuropsychiatric disorders, the relevance of this to cognitive decline in schizophrenia, and, finally, how GABA receptor activation may reverse it.
Bradley Peterson (Columbia University) described anatomical (MRI) and functional (EEG) biomarkers in children at high risk for major depressive disorder (MDD) and anxiety, based on a family history of depression in their parents and grandparents. These children have a deficiency in arousal and vigilance for perceiving social and emotional cues from the environment.
Left hemisphere cortical thinning correlated highly with summed anxiety + depression scores. A thinning throughout the right and left posterior neocortex in these children correlated highly with greater degrees of inattention, and worsening of visual memory for family settings, faces, and other social/emotional stimuli.
Right and left cortical thinning proved to be a good predictor for the development of anxiety + depression, and those with familial risk for depression show an even stronger association with cortical thinning. Thinning in the right posterior neocortex leads to inattention and visual memory problems, but not necessarily MDD/anxiety. In contrast, thinning in the left posterior neocortex is associated with MDD/anxiety and particularly poor recognition of social/emotional cues. These associations were strongest when both sides were thin.
Cameron Carter (UC Davis) suggested that cognition problems in schizophrenia are 1) early and persistent signs of the disease, 2) strong predictors of disability in schizophrenia, and 3) current drug treatments have little if any impact on cognitive disability. In his talk, Carter emphasized how cognitive deficits are amenable to investigation with non-invasive tools.
Using fMRI, Carter and colleagues studied attention, memory, and language disturbances. For example, schizophrenia patients make more errors on the Stroop test of detecting contextual mismatches between a stimulus and the correct response (e.g., the word "Blue" is the correct response but the word is colored orange). Patients show less activation of the dorsolateral prefrontal cortex (DLPFC) when making these errors than do control subjects. This is another aspect of neocortical hypofrontality in schizophrenia and is related to impaired cognition.
A meta-analysis of 41 fMRI studies conducted by Carter and his UC Davis colleague Michael Minzenberg shows that schizophrenia patients engage the DLPFC and anterior cingulate cortex (ACC) far less than healthy controls during working memory. This is also true in never-medicated, first-episode patients, who also showed far less recruitment of the DLPFC and the parietal lobes during cognitive task performance.
Abnormal mechanisms for dorsolateral hypofrontality in schizophrenia include diminished γ oscillations in clusters of DLPFC, possibly related to losses of the GABA interneurons that normally generate γ oscillations while regulating pyramidal neuron outflow. In support of this, postmortem findings by David Lewis, Francine Benes, and others show lower GAD67 mRNA, and GAT-1 transporter and protein in GABA interneurons, increases in the α2 subunit of the GABAA receptors, and decreased GAT-1 protein on pyramidal cells in schizophrenia. Their loss, and hypofunction of frontal cortical GABA interneurons, may contribute to the decreased γ oscillations in schizophrenia. Thus, γ oscillations may be a promising biomarker for schizophrenia when combined with therapeutic strategies that boost GABA function.
In line with this possibility, MK-0777, an allosteric potentiator of GABA signaling at GABAA receptors that contain the α2 subunit, or its placebo control, was given to patients. Four weeks treatment with MK-0777 augmented performance in most of the cognitive test categories that were evaluated. It will be of great importance to see what effect MK-0777 has on frontal γ oscillations in these patients.
Naomi Driesen (Yale University), described how the NMDA antagonists ketamine, MK-801, and 2-APV alter brain activation and functional connectivity to decrease cognition and activation of the PFC during maintenance of working memory in primates. While the amnestic properties of NMDA antagonism are well known, Driesen provided an AMPA/NMDA imbalance hypothesis for these actions. This hypothesis states that blocking the normal excitatory NMDA receptor activation of GABA neurons disinhibits cortical glutamatergic pyramidal neurons. The pyramidal neurons then release more glutamate to stimulate other cortical receptors for AMPA, while NMDA receptors remain blocked. This results in excessive AMPA signaling, and an AMPA/NMDA "imbalance" favoring AMPA signaling by glutamate on pyramidal neurons.
A spatial position, delayed memory task for four separate locations was given to normal human volunteers aged 21-44. Compared with a saline IV bolus and constant infusion, a subsequent ketamine bolus (0.23 mg/kg for 1 min) and infusion (0.58 mg/kg/hour) produced large neocortical activation in parietal, temporal, and some frontal areas, and compromised function on cognitive tasks. This high IV dose, though sub-anesthetic, also produced visual hallucinations, so this interesting paradigm of GABA hypofunction could mimic the proposed decrease in GABA inputs to cortical pyramidal neurons, with hallucinogenic and cognitive deficits arising, as in schizophrenia. Support for the AMPA/NMDA hypothesis is provided by the improved cognition in schizophrenia described by Carter for the allosteric GABAA potentiator, MK-0777.—C. Anthony Altar.