The percentage of interneurons in the dorsal lateral geniculate nucleus of the cat and observations on several variables that affect the sensitivity of horseradish peroxidase as a retrograde marker.

Ten cats ranging in age from 4 weeks postnatal to adult received large bilateral injections of horseradish peroxidase (HRP) into cortical areas 17 and 18. In one cat additional unilateral injections of HRP were made into the lateral suprasylvian visual areas (PMLS). The purpose of these injections was to label relay cells in lamina A of the dorsal lateral geniculate nucleus (LGN), in order to distinguish them from neurons that could not be labeled retrogradely. Several factors thought to influence the effectiveness of HRP as a retrograde marker were varied in an effort to label as many relay cells as possible. These factors included the (1) rate and duration of HRP injections; (2) volume and concentration of HRP injected; (3) addition of L-alpha-lysophosphatidylcholine or dimethyl sulfoxide to the injected HRP; and (4) aldehyde and buffers used for fixation. In all experiments DAB (3,3'-diaminobenzidine tetrahydrochloride) was used as the chromogen, either alone or with the addition of cobalt chloride, nickel, and cobalt salts, or cobalt-glucose oxidase. In 1-micrometer plastic sections, the influence of each of the above factors and DAB methods was determined by measuring the percentage of unlabeled neurons and the cytoplasmic HRP grain density of cells that were labeled. Our results show that approximately 22% of the neurons in lamina A of the LGN remain unlabeled following injections of HRP into areas 17 and 18 alone or combined with injections into PMLS. The percentage of unlabeled cells is similar at each of the ages that we studied and is not affected significantly by any of the factors that were varied or DAB methods that were used. Cross-sectional area measurements show that unlabeled cells tend to be among the smallest neurons in lamina A. Regardless of age, the mean size of labeled neurons was about twice that of unlabeled cells. However, we found only a weak correlation between the size of a labeled cell and the cytoplasmic density of HRP grains. Thus it is unlikely that small cell body size alone can account for the unlabeled cells in lamina A, since small neurons can be as effective in transporting HRP retrogradely as large neurons. We therefore conclude that there is a distinct population of small neurons in lamina A of the LGN that do not project to cortex. Although we cannot rule out the possibility that these cells project subcortically, we believe that it is reasonable to regard them as interneurons.