In vitro eye-blink reflex model: role of excitatory amino acids and labeling of network activity with sulforhodamine.

Evidence is presented suggesting that the neural correlate of the eye-blink reflex can be evoked in an in vitro brainstem-cerebellum preparation from the turtle by using electrical rather than natural stimulation of cranial nerve inputs. Abducens nerve discharge is triggered by brief electrical stimulation of the ipsilateral trigeminal nerve. This discharge corresponds behaviorally to EMG recordings of extraocular muscles and eye retraction recorded in situ, in a reduced preparation. The abducens nerve discharge has two components: a short-duration response having a latency of 3-6 ms and a duration of 50-100 ms, followed by a long-duration component having a latency of 12-20 ms and a duration of several seconds. The long-duration component of the reflex is blocked by the NMDA receptor antagonist APV, while both reflex responses are blocked by the non-NMDA glutamate receptor antagonist CNQX. To visualize the spatial distribution of activity during the abducens nerve reflex, bath application of the activity-dependent dye sulforhodamine was used. During reflex activity, neurons in the ipsilateral trigeminal nucleus, principal abducens nucleus, and presumed interneurons ventrolateral to the principal abducens nucleus, labeled with the dye, in addition to areas in the raphe nucleus and reticular formation. In conditions where the long-duration component of the reflex was suppressed, sulforhodamine label was absent in the principal abducens nucleus and in the caudal brainstem. From these data it is hypothesized that the region of interneurons and the accessory abducens nucleus participate in the short-duration component of the reflex. This response is mediated by non-NMDA receptors. The principal abducens nucleus is postulated to contribute also to the short-duration portion of the reflex, but is primarily involved in the generation of the long-duration component. This component of the reflex is mediated principally by NMDA receptors. Sustained reflex activity is further postulated to originate from recurrent excitation in pathways within the caudal brainstem, particularly the reticular formation. This interpretation is consistent with the observed patterns of sulforhodamine label, the effects of local microinjections of APV, and the elimination of sustained activity when the caudal brainstem is transected. These data have implications for pathways that may underlie conditioning of the eye-blink response.