, 1997). Moreover, in experimental animals, NMDA antagonists elevate extracellular glutamate ( Moghaddam and Javitt, ERK activity inhibition 2012) and induce hypermetabolism in cerebral

cortex as detected by CBV-fMRI ( Gozzi et al., 2008) at doses that induce behavioral and neurochemical abnormalities homologous with schizophrenia ( Bickel and Javitt, 2009; Jentsch and Roth, 1999; Moghaddam et al., 1997; Mouri et al., 2007; Pinault, 2008). To test the hypothesis that excess glutamate drives hippocampal subregional hypermetabolism and atrophy in psychosis, we used administration of the NMDA antagonist ketamine in mice, where invasive technologies can be deployed to precisely map extracellular glutamate within the hippocampal circuit. First, we used the same MRI tools applied to human patients to show that that ketamine administration in rodents replicated the specific anatomical pattern of hippocampal dysfunction in schizophrenia. We then conducted a series of experiments utilizing an in vivo glutamate biosensor http://www.selleckchem.com/products/z-vad-fmk.html method (Hu et al., 1994) in multiple hippocampal subregions to show that ketamine causes a regionally preferential increase in extracellular glutamate, and that a glutamate-reducing agent ameliorates this increase as well as MRI-detected hippocampal hypermetabolism and atrophy. Finally, we performed postmortem analyses to determine histopathological correlates of the neuroimaging phenotypes. Twenty-five subjects at clinical high-risk for

a psychotic disorder were imaged while experiencing prodromal symptoms, and then 20 were reimaged at clinical follow-up totaling 80% of the original sample. One patient Terminal deoxynucleotidyl transferase in the progressor group, and four patients in the nonprogressor group were lost to brain imaging follow-up, although none were lost to clinical follow-up. The average follow-up period and interscan interval was 2.4 years. During this time, 10 of the 25 subjects progressed to psychosis, and 15 subjects did not develop a psychotic disorder (see Tables S1 and S2 available online). Subjects were comparable on baseline

demographic variables including age (t23 = 0.7, p = 0.48), sex (Fisher’s exact p = 1.0), years education (t22 = 0, p = 1) and follow-up interval (t18 = 0.8, p = 0.76). Subjects were imaged at baseline and at follow-up using the steady-state gadolinium-enhanced fMRI technique, as previously described (Moreno et al., 2007). Based on our previous findings (Schobel et al., 2009b) we focused on the anterior aspect of the hippocampal formation. For each subject, mean CBV was measured from the following hippocampal subregions—entorhinal cortex (EC), dentate gyrus (DG), CA3, CA1, and subiculum (SUB). A repeated-measures analysis of variance with subregion (EC, DG, CA3, CA1, and SUB) and side (left, right) as within-subject factors and outcome (progression to psychosis versus non-progression) as between-subjects factor revealed a subregion by progression to psychosis interaction (F4,18 = 3.1, p = 0.