Metabotropic glutamate receptors (mGluRs) on relay cells of the lateral geniculate nucleus appear to be activated exclusively by cortical inputs. We thus sought to manipulate these receptors in an effort to gain insight into the possible role of the corticogeniculate pathway. We used in vivo recording and pharmacological techniques in cats to activate or inactivate these receptors on geniculate neurons while analyzing their response properties. 2. Iontophoretic application of the mGluR agonist 1-amino-cyclopentane-1,3-dicarboxylic acid (ACPD) to X and Y cells in the geniculate A laminae diminished or abolished burst activity characteristic of low-threshold Ca2+ spikes. This was accompanied by pronounced changes in the visual response, including a decrease in signal detectability as measured with receiver operating characteristic curves. 3. ACPD effects appear specific to mGluRs, because a specific antagonist of ionotropic glutamate receptors (iGluRs) failed to affect the ACPD-evoked responses, and antagonists of ACPD failed to affect iGluR-mediated responses. We found that 3,5-dihydroxyphenylglycine, an agonist reported to be specific for phosphatidylinositol (PI)-linked mGluRs, had effects similar to those of ACPD, implying that these effects are mediated by PI-coupled mGluRs. Furthermore, antagonists reported to be effective against PI-linked mGluRs were effective in antagonizing the ACPD-mediated effects, and substances reported to be agonists to mGluRs coupled to the adenosine 3',5'-cyclic monophosphate cascade did not affect neuronal responses on their own. These data, when added to our preliminary anatomic data, indicate that the receptor responsible for the observed effects may be mGluR1, or a functionally equivalent mGluR. 4. Activation of mGluRs produces changes in geniculate relay cell activity consistent with depolarization of these cells seen during in vitro studies. Such membrane depolarization has been shown to control the activation state of a voltage-dependent Ca2+ conductance, and this, in turn, determines whether the relay cell fires in tonic or burst mode. Our data show that application of ACPD produces a shift in response mode from burst to tonic. Because response mode is an important characteristic of the geniculate relay and because the activation state of certain mGluRs, which helps determine response mode may be controlled by corticogeniculate input, we conclude that an important function of this input is to provide a visuotopically discrete transition from burst to tonic response mode.
