Stable expression of homomeric AMPA-selective glutamate receptors in BHK cells.

cDNAs encoding glutamate receptor glu1, glu2 (Q and R) or glu4 under control of a constitutively active metallothionine promoter, were transfected into baby hamster kidney cells. Following the addition of selection agent, transfectants expressing high levels of glutamate receptor as measured by [3H]alpha-amino-3-hydroxyl-5-methyl-isoxalazole-4-propionate (AMPA) binding, were selected for further studies. Using glutamate receptor antibodies, the receptor proteins were visualized in Western blotting as having a molecular weight of approximately 100 kDa. [3H]AMPA binding to the glutamate receptor expressing cell lines revealed that glu1, glu2 (Q), and glu4 receptors displayed a single site in Scatchard analysis with Kd values of 12, 15.7 and 21 nM, respectively. However, the Ca2+ impermeable variant of the glu2 receptor, glu2 (R) displayed a curvilinear Scatchard plot. Computer resolution suggested the presence of a high and low affinity state (KH = 2.9 nM; KL = 40.7 nM). The pharmacological profile of the [3H]AMPA binding to these recombinant receptors resembled the high affinity [3H]AMPA binding site in rat brain showing high affinity for glutamate, quisqualate, and medium affinity for 6-cyano-7-nitro-quinoxaline-2,3-dione, CNQX; 6,7-dinitro-quinoxaline-2,3-dione, DNQX; and 6-nitro-7-sulphanyl-benzo(f)quinoxaline-2,3,dione, NBQX. Kainate displayed low affinity and N-methyl-D-aspartate (NMDA), was inactive in inhibiting specific [3H]AMPA binding. These cell lines will prove to be important tools in the study of glutamate receptors.

The importance of two specific domains in ligand binding to the AMPA/kainate glutamate receptors GluR2 and GluR6.

Chimeric receptor subunits of the AMPA receptor subunit GluR2 and the kainate receptor subunit GluR6 were constructed and stably expressed in baby hamster kidney cells. By using Ca2+ imaging and radioligand binding, we demonstrated that substitution of a specific domain showing homology to a bacterial leucine-isoleucine-valine binding protein (LIVBP) had no effect on the affinities of the tested agonists, but decreased the affinities of the antagonists CNQX, DNQX, and NBQX. On the other hand, when the first of two domains showing homology to a bacterial glutamine binding protein (QBP) in GluR2 was substituted with the corresponding region from GluR6, the affinity of AMPA decreased sevenfold and the affinity of kainate increased fourfold, indicating the importance of this domain in binding of these agonists. In contrast to this, the affinities of quisqualate and domoate, two other agonists, were unchanged, indicating that a region located C-terminal to the QBP domain is also involved in agonist binding.

Stable expression of a functional GluR6 homomeric glutamate receptor channel in mammalian cells.

This study demonstrates the stable expression of a functional ionotropic glutamate receptor in a mammalian cell line of non-neuronal origin. The kainate-selective glutamate receptor GluR6 was constitutively expressed under the control of a metallothionein promoter. Clones were isolated expressing approximately 3 pmol of receptor per mg of protein. Functionality of the recombinant GluR6 was demonstrated both by electrophysiology and by Ca2+ imaging. Application of kainate to the GluR6-transfected cells activated an inward current response at a holding potential of -60 mV. The kainate concentration needed to evoke 50% of the maximal response (EC50) was calculated to be 0.82 +/- 0.39 microM. The current-voltage relationship was found to be almost linear, with a reversal potential of -2.5 +/- 4.8 mV. Application of kainate also resulted in an increase in the intracellular Ca2+ concentration measured by Ca2+ imaging. The pharmacological profile of [3H]kainate binding to the recombinant GluR6 resembled the high-affinity [3H]kainate binding sites in rat brain, showing high affinity for domoate (Ki = 5.1 +/- 3.0 nM) and kainate (Kd = 12.9 +/- 2.4 nM). No decrease in GluR6 expression level was observed over > 75 passages of the transfected cells. When domoate, a slowly desensitizing GluR6 agonist, was included in the growth medium for 3 weeks, the number of GluR6 binding sites decreased by 30%, indicating the importance of complete channel closure for stable expression.