This application responds to PAR-22-170 Building in vivo preclinical assays of circuit engagement for
application in therapeutic development. The project will optimize non-human primate (NHP) and rodent analogs
of human auditory mismatch negativity (MMN) for use in preclinical development programs. Deficits in MMN
generation have been extensively described in schizophrenia (Sz) and shown to correlate with cognitive
impairment and poor functional outcome in established Sz, and in conversion to Sz among individuals at clinical
high risk (CHR), demonstrating clinical relevance. MMN may also decrease in amplitude during the initial stages
of Sz, providing a target for development of treatments that may prevent neurodegeneration during initial stages
of the disorder. MMN indexes the integrity of early auditory processing (EAP) such as delayed tone matching
abilities, which are critical for processes such as auditory emotion recognition, verbal memory, and phonological
reading ability. Deficits in these processes, in turn, significantly predict outcome. In addition, local circuit
abnormalities that give rise to MMN impairments in auditory cortex may be present across brain regions. Insights
gleaned from investigation and remediation of MMN deficits may therefore be relevant across cortical regions. In
humans, MMN generation is inhibited reliably by N-methyl-D-aspartate receptors (NMDAR), suggesting
involvement of underlying glutamatergic mechanisms and local excitatory/inhibitory (E/I) balance. In spectral
analyses, MMN shows primary power within the theta frequency range, suggesting additional involvement of
somatostatin (SOM)-type GABA interneurons. MMN-like activity has been demonstrated in both NHP and
rodents, and in both species shows similar spectral content and NMDAR sensitivity to human MMN. The present
project will 1) further optimize these measures for use in early-stage drug development, while 2) also evaluating
their sensitivity and selectivity to compounds that do (e.g. NMDAR antagonists) and do not (e.g. 5-HT2A
antagonists) affect their generation in humans. These measures will then 3) be used to test specific local and
distributed circuit computational models to permit refinement in use of MMN paradigms across humans, NHP and
rodents. At the distributed network level, mechanistic testing will seek to refine emergent prediction error (PE)
based theories of MMN generation, which posit specific feed-forward and feed-back information flow among
primary auditory cortex (A1), superior temporal gyrus (STG) and inferior frontal gyrus (IFG), and to develop
spectral signatures of this information flow. At the local circuit level, mechanistic testing will assess the relative
contributions of specific interneuron populations using cell-specific genetic manipulation. Several compounds that
target NMDAR via the glycine/D-serine modulatory site (e.g. iclepertin, luvadaxisat) have shown promise for
treatment of Cognitive Impairment Associated with Sz (CIAS). The present pipeline will enable identification and
validation of additional targets within the E/I circuit and development of additional approaches for enhancement
of glutamatergic function and restoration of E/I balance across neuropsychiatric disorders.
