The present article examines the anatomical organization of the dorsal telencephalon of two gymnotiform fish: Gymnotus sp. and Apteronotus leptorhynchus. These electric fish use elaborate electrical displays for agonistic and sexual communication. Our study emphasizes mainly pallial divisions: dorsolateral (DL), dorsodorsal (DD), and dorsocentral (DC), previously implicated in social learning dependent on electric signals. We found that the pallial cytoarchitectonics of gymnotiformes are similar to those reported for the commonly studied goldfish, except that DC is larger and better differentiated in gymnotiformes. We identified a new telencephalic region (Dx), located between DL and DC, and describe the morphological and some biochemical properties of its neurons. Most neurons in DL, DD, and DC are glutamatergic with spiny dendrites. However, the size of these cells as well as the orientation and extent of their dendrites vary systematically across these regions. In addition, both DD and DL contained numerous small GABAergic interneurons as well as well-developed GABAergic plexuses. One important and novel observation is that the dendrites of the spiny neurons within all three regions remain confined to their respective territories. We confirm that DL and DC express very high levels of NMDA receptor subunits as well as CaMKIIα, a key downstream effector of this receptor. In contrast, this enzyme is nearly absent in DD, while NMDA receptors are robustly expressed, suggesting different rules for synaptic plasticity across these regions. Remarkably, GABAergic pallial neurons do not express CaMKIIα, in agreement with previously reported results in the cortex of rats.

The acoustic startle response is a protective response, elicited by a sudden and intense acoustic stimulus. Facial and skeletal muscles are activated within a few milliseconds, leading to a whole body flinch in rodents 1 . Although startle responses are reflexive responses that can be reliably elicited, they are not stereotypic. They can be modulated by emotions such as fear (fear potentiated startle) and joy (joy attenuated startle), by non-associative learning processes such as habituation and sensitization, and by other sensory stimuli through sensory gating processes (prepulse inhibition), turning startle responses into an excellent tool for assessing emotions, learning, and sensory gating, for review see 2,3 . The primary pathway mediating startle responses is very short and well described, qualifying startle also as an excellent model for studying the underlying mechanisms for behavioural plasticity on a cellular/molecular level 3 .We here describe a method for assessing short-term habituation, long-term habituation and prepulse inhibition of acoustic startle responses in rodents. Habituation describes the decrease of the startle response magnitude upon repeated presentation of the same stimulus. Habituation within a testing session is called short-term habituation (STH) and is reversible upon a period of several minutes without stimulation. Habituation between testing sessions is called long-term habituation (LTH) 4 . Habituation is stimulus specific 5 . Prepulse inhibition is the attenuation of a startle response by a preceding non-startling sensory stimulus 6 . The interval between prepulse and startle stimulus can vary from 6 to up to 2000 ms. The prepulse can be any modality, however, acoustic prepulses are the most commonly used.Habituation is a form of non-associative learning. It can also be viewed as a form of sensory filtering, since it reduces the organisms' response to a non-threatening stimulus. Prepulse inhibition (PPI) was originally developed in human neuropsychiatric research as an operational measure for sensory gating 7

No abstract

Prepulse inhibition (PPI) of acoustic startle is an operational measure of sensorimotor gating, which is disrupted in schizophrenia. NMDA receptor (NMDAR) antagonist induced PPI disruption has become an important pharmacological model for schizophrenia; however, knowledge of the underlying mechanism remains incomplete. This study examines the role of NMDAR in the caudal pontine reticular nucleus (PnC) and the medial prefrontal cortex (mPFC) in NMDARs antagonist induced PPI deficits, as well as the NMDA receptor subtypes involved. We administered the NMDA antagonist MK-801 locally into the caudal pontine reticular formation (PnC), where the PPI mediating pathway converges with the primary startle pathway, and into the mPFC prior to behavioural testing. PnC microinjections had no effect on startle and PPI, whereas injections into the ventro-rostral part, but not into the dorso-caudal part of the mPFC, disrupted PPI. These effects could be mimicked by local injection of the NR2B subunit specific antagonist ifenprodil, whereas co-application of MK-801 and the mGluR2/3 agonist LY354740 had no effect on PPI. Moreover, PPI disruptions by systemically administered MK-801 could be reversed by local injections of LY354740 into the ventro-rostral mPFC, but not into the dorso-caudal mPFC. Our results indicate that NR2B subunit containing NMDARs in a specific subregion of the mPFC play a major role in PPI disruptions by systemic NMDAR antagonism. Our results further support the hypothesis that glutamate hyper-function in the mPFC is a main mechanism involved in sensory gating deficits induced by systemic MK-801, supporting the notion that this is an important mechanism in schizophrenia pathology.