The developmental expression of calcium (Ca2+)-permeable alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and kainate receptors in cultured neocortical neurons was evaluated by using cobalt uptake, a histochemical method that identifies cells expressing Ca2+-permeable, non-N-methyl-D-aspartate (non-NMDA) receptors. At a concentration of 500 microM, AMPA was found to stimulate cobalt uptake only late in development, resulting in staining of 2.7%+/-0.3% of the neurons maintained in culture for 12 days in vitro (DIV). When AMPA receptor desensitization was blocked with 50 microM cyclothiazide, the developmental profile of cobalt uptake mediated by 25 microM AMPA changed dramatically. The cobalt staining now appeared in young cultures (5 DIV), and the percentage of stained cells increased from 3.4%+/-0.2% at 5 DIV to 21.7%+/-1.6% at 12 DIV. The effect of 200 microM kainate was similar to that seen with 25 microM AMPA plus 50 microM cyclothiazide, resulting in 17.7%+/-0.3% stained neurons at 12 DIV. The cobalt uptake was specific to AMPA and kainate receptors because NMDA receptors and voltage-gated calcium channels were found not to mediate any cobalt staining. In addition, 10 microM 6-nitro-7-sulphamoylbenzo-[f]-quinoxaline-2,3-dione (NBQX) was able to prevent all staining at 5 and 8 DIV and most of the staining at 12 DIV, indicating that the non-NMDA ionotropic glutamate receptors are involved in cobalt uptake into the neurons. The AMPA receptor-selective antagonist GYKI 53655 was used to differentiate between cobalt influx through AMPA- or kainate-preferring receptors. After pretreatment with concanavalin A (con A), an inhibitor of kainate receptor desensitization, cobalt uptake was assessed after stimulation by 200 microM kainate in the presence of 25 microM GYKI 53655. No cobalt staining was observed under these conditions, indicating that most if not all of the cobalt influx induced by kainate was mediated through AMPA receptor channels.