Light and electron microscopic evidence of transneuronal labeling with WGA-HRP to trace somatosensory pathways to the thalamus

Abstract

Horseradish peroxidase conjugated to wheat-germ lectin is being used with increasing frequency as an anterograde label to trace pathways in the nervous system, owing to the sensitivity of the method and ease of use. However, it has been suggested that horseradish peroxidase conjugated to wheat-germ lectin may be transneuronally transported, thus affecting the ease of interpretation of the results. The present study used the projections of the dorsal column nuclei and spinal cord to the thalamus as a model system to determine whether transneuronal transport could be demonstrated and whether the degree of such transport was related to the size of the injection site. Light microscopic observation of sections incubated with tetramethyl benzidine after large injections (1 microL of a 10% solution of horseradish-peroxidase-conjugated wheat-germ lectin in water) in the dorsal column nuclei demonstrated the presence of labeled neurons in the nucleus reticularis thalami, which is not known to receive afferents from or project to these nuclei. The electron microscopic study, although based upon the use of the chromogen benzidine dihydrochloride, less sensitive than tetramethyl benzidine, revealed the existence of labeled neurons in the thalamic ventrobasal complex. This is unlikely to be due to retrograde labeling and is therefore interpreted as a result of transneuronal, perhaps transsynaptic, transport. Glial and perivascular cells also contained granules of reaction product in some cases. Smaller injections (100 nL) in the dorsal column nuclei, on the other hand, did not produce this apparent transneuronal labeling. After small injections (100 nL) in the spinal cord, anterograde labeling was observed mainly in the thalamic ventrobasal complex in the rat, and in the posterior group in the cat, and the nuclei centralis lateralis and submedius in both species, as has been described in numerous other studies. After large injections, additional labeled areas were observed in the posterior intralaminar region (parafascicular-center median complex), in the medial thalamus (nuclei reuniens, rhomboid and paraventricular), and in the cat, in the ventroposterolateral nucleus. In the rat, experiments were performed in which a kainic acid injection was made to induce neuronal loss in the nucleus reticularis gigantocellularis of the medulla, which is a relay of the spinoreticulothalamic pathway, known to project to some of these thalamic areas.(ABSTRACT TRUNCATED AT 400 WORDS)