Abstract

To study the role of acidic residues in modulation of NMDA receptors by spermine, we used site-directed mutagenesis of receptor subunits and voltage-clamp recording in Xenopus oocytes. Sixteen glutamate and aspartate residues, located in the first two thirds of the putative extracellular loop of the NR1A subunit, were individually mutated. This region of NR1A shows homology with bacterial amino acid binding proteins, a bacterial polyamine binding protein, and a bacterial spermidine acetyltransferase. Mutation of D669 to asparagine (D669N), alanine (D669A), or glutamate (D669E) abolished the "glycine-independent" form of spermine stimulation in heteromeric NR1A/NR2B receptors. These mutations also markedly reduced inhibition by ifenprodil and by protons at NR1A/NR2B receptors. Mutations at the equivalent position (D690) in NR1B, which contains the insert encoded by exon 5, reduced the pH sensitivity of NR1B/NR2B receptors. Thus, the effects of mutations at D669 are not prevented by the presence of exon 5, and the influence of exon 5 is not prevented by mutations at D669 (D690 in NR1B). Mutations at NR1A (D669) had little or no effect on the potencies of glutamate and glycine and did not alter voltage-dependent block by Mg2+ or the "glycine-dependent" form of spermine stimulation. Surprisingly, the D669N and D669A mutations, but not the D669E mutation, reduced voltage-dependent block by spermine at NR1A/NR2 receptors. Mutations in NR2B at a position (D668) equivalent to D669 did not alter spermine stimulation or sensitivity to pH and ifenprodil. However, mutations D668N and D668A but not D668E in NR2B reduced voltage-dependent block by spermine. Screening of the negative charges at NR1A(D669) and NR2B(D668) may be involved in voltage-dependent block by spermine. D669 in NR1A could form part of a binding site for polyamines and ifenprodil and/or part of the proton sensor of the NMDA receptor. Alternatively, this residue may be critical for coupling of modulators such as spermine, protons, and ifenprodil to channel gating.