Cortical feedback to the thalamus is selectively enhanced by nitric oxide.

The brain somehow merges visual information with the behavioral context in which it is being processed, a task that is often attributed to the cerebral cortex. We have identified a new role of the gaseous neurotransmitter, nitric oxide (NO), in the early selective enhancement of corticogeniculate communication that may participate in this process at the level of the thalamus. Visual information is dynamically gated through the thalamus by brainstem neurons that release acetylcholine and NO. Using in vitro electrophysiology, we characterized NO effects on excitatory postsynaptic potentials and currents (EPSCs) elicited from retinal and cortical pathways in the lateral geniculate nucleus of the ferret. NO selectively and reversibly increased cortically-evoked postsynaptic responses, and this effect was mimicked by cyclic guanosine 3',5'-monophosphate (cGMP). Conversely, NO inhibited retinally-evoked responses independently of cGMP. We demonstrated that these differential effects were specific to postsynaptic N-methyl-d-aspartate (NMDA) receptors by studying treatment effects on pharmacologically isolated EPSCs from each pathway. We propose that when brainstem activity is increased during behavioral arousal or rapid eye movement sleep, NO may increase the relative sensitivity of relay neurons to corticogeniculate feedback. The net effect of these changes in synaptic processing may be to selectively suppress peripheral information while unifying data carried by reentrant corticogeniculate loops with the behavioral context in which the visual information is processed.

The native T-type calcium current in relay neurons of the primate thalamus.

The generation of thalamic bursts depends upon calcium currents that flow through transiently open (T)-type calcium channels. In this study, we characterized the native T-type calcium current underlying thalamic burst responses in the macaque monkey. Current clamp recordings from lateral geniculate nucleus (LGN) slices showed characteristic burst responses when relay cells were depolarized from relatively hyperpolarized membrane potentials. These bursts could also be elicited by stimulation of excitatory synaptic inputs to LGN cells. Under voltage clamp conditions, the inactivation kinetics of native currents recorded from primate LGN neurons showed consistency with T-type currents recorded in other mammals and in expression systems. Real-time reverse transcriptase PCR performed on RNA isolated from the LGN (including tissues isolated from magnocellular and parvocellular laminae) detected voltage-dependent calcium channel (Ca(v)) 3.1, Ca(v) 3.2, and Ca(v) 3.3 channel transcripts. Ca(v) 3.1 occurred at relatively higher expression than other isoforms, consistent with in situ hybridization studies in rats, indicating that the molecular basis for burst firing in thalamocortical systems is an important conserved property of primate physiology. Since thalamic bursts have been observed during visual processing as well as in a number of CNS disorders, studies of the expression and modulation of these currents at multiple levels are critical for understanding their role in vision and for the discovery of new treatments for disruptions of thalamic rhythms.

Transneuronal retrograde transport of attenuated pseudorabies viruses within central visual pathways.

Pseudorabies virus (PRV) has been shown to be an effective transneuronal tracer within both the peripheral and the central nervous system. The only investigations of this virus in the visual system have examined anterograde transport of PRV from injection sites in the retina. In the present study, we injected attenuated forms of PRV into the primary visual cortex of both rats and cats to determine whether transneuronal retrograde infection would occur back to the retina. In rats, we made small injections into visual cortex of a strain of PRV (Bartha Blu) that contained a beta-galactosidase promoter insert. In cats, we injected PRV-M201 into area V1 of visual cortex. After a 2- to 4-day incubation period, we examined tissue from these animals for the presence of the beta-galactosidase marker (rats) or the virus itself (cats). Cortical PRV injections resulted in transneuronal retrograde infection of the lateral geniculate nucleus (LGN), thalamic reticular nucleus (TRN), and retina. PRV was retinotopically distributed in the pathway. In addition, double-labeling experiments in cats using an antibody against gamma-aminobutyric acid (GABA) were conducted to reveal PRV-labeled interneurons within the LGN and TRN. All TRN neurons were GABA+, as was a subset of LGN neurons. Only the subset of TRN neurons adjacent to the PRV-labeled sector of LGN was labeled with PRV. In addition, a subset of GABA+ interneurons in LGN was also labeled with PRV. We processed some tissue for electron microscopy to examine the morphology of the virus at various replication stages. No mature virions were detected in terminals from efferent pathways, although forms consistent with retrograde infection were encountered. We conclude that the PRV strains we have used produce a local infection that progresses primarily in the retrograde direction in the central visual pathways. The infection is transneuronal and viral replication maintains the intensity of the label throughout the chain of connected neurons, providing a means of examining detailed circuitry within the visual pathway.