Coronal slices of midbrain from adult rats, maintained in an interface chamber at 33-35 degrees C, were used to investigate projections from the dorsal raphe nucleus to the dorsolateral periaqueductal gray matter. Extracellular recordings were made from neurons in the dorsolateral periaqueductal gray matter. The majority of cells were silent but activity could readily be induced by iontophoretic application of D,L-homocysteic acid from the recording pipette. Neuronal perikarya in the dorsal raphe nucleus were activated either by topical application of 15-100 nl droplets of 1 pr 10 mM D,L-homocysteic acid or by iontophoresis into the dorsal raphe nucleus of 0.1 M D,L-homocysteic acid from a five-barrelled glass micropipette. Both forms of stimulation evoked increases or decreases in firing in neurons in the dorsolateral periaqueductal gray matter. The excitatory responses were evoked only from stimulation sites in the dorsal subnucleus of the dorsal raphe nucleus, whilst inhibitory responses could be evoked by stimulating throughout the nucleus. 5-Hydroxytryptamine was applied by iontophoresis to 55 neurons in the dorsolateral periaqueductal gray matter. Of these, 89% were inhibited whilst the remaining 11% were excited. No unresponsive cells were found. In a second series of experiments anterograde transport of biocytin was used to study projections from the dorsal raphe nucleus to the periaqueductal gray matter in vitro. Following injections of pellets of biocytin into the dorsal subnucleus of the dorsal raphe nucleus, two types of labelled axons could be distinguished which had either smooth or beaded profiles. Both types of axon could be followed into the region of the dorsolateral periaqueductal gray matter, from which recordings were made in the electrophysiological experiments. In contrast, following deposits of biocytin into the ventral half of the periaqueductal gray matter, very few labelled axons were found in the dorsolateral periaqueductal gray matter, although the ventrolateral periaqueductal gray matter was heavily labelled. We conclude that both excitatory and inhibitory projections from the dorsal raphe nucleus to the dorsolateral periaqueductal gray matter are present in the coronal midbrain slice preparation. Excitatory projections originate from the dorsal subnucleus of the dorsal raphe nucleus and may be mediated by a direct pathway. The inhibitory effects evoked from the ventral half of the dorsal raphe nucleus are probably mediated indirectly, perhaps by activation of inhibitory interneurons in the ventrolateral periaqueductal gray matter.
