Fig. 2. Differences in neurotransmission in primate V1 vs. dlPFC.

a Neurons in primate V1 exhibit classic neurotransmission, relying heavily on glutamate stimulation of AMPAR with only a minor contribution of NMDAR. The rapid kinetics of AMPARs are appropriate for the accurate encoding of a sensory event. b The stimulus-evoked firing of V1 neurons in monkeys depends on AMPAR, as even low doses of an AMPAR antagonist (CNQX) greatly reduce firing. In contrast, equivalent doses of an NMDAR–NR2B antagonist Ro25-6981 (Ro) have little effect on V1 neuronal firing. c In contrast to V1, layer III dlPFC delay cells rely on NMDAR (NR2A and NR2B) with only subtle reliance on AMPA. The permissive depolarization of the PSD is instead contributed by acetylcholine via Nic-α7R (shown) and muscarinic M1R (not shown), both of which reside within the glutamatergic PSD [33, 34]. It is of interest that the Nic-α7R fluxes the most calcium of all cholinergic receptors, and that M1R can increase internal calcium release via Gq signaling. These data suggest that calcium entry from NMDAR–NR2B, as well as from Nic-α7R and M1R signaling, may maintain sufficient calcium near the PSD to maintain depolarization and allow persistent firing. M1R also enhance neuronal firing by closing KCNQ “m” channels, some of which are localized in the glutamate PSD [34]. These powerful cholinergic actions in dlPFC contrast with more subtle effects in V1, where nicotinic receptors are restricted to presynaptic thalamic inputs in layer IV, and to interneurons [180], and acetylcholine has a more classic, modulatory role, e.g., enhancing responses during attention [181, 182]. d Delay-related firing of dlPFC neurons in monkey depend on NMDAR, as even low doses of an NMDAR–NR2B antagonist such as Ro greatly reduce firing. In contrast, equivalent doses of an AMPAR antagonist CNQX have little effect on delay cell firing. b and d were adapted with permission from Yang et al. [30].