The N-methyl-D-aspartate receptor subunit NR3A : cloning, expression and interacting proteins

Abstract

This thesis focuses on the subunit NR3A, which together with subunits NR1, NR2A-NR2D and NR3B builds the excitatory glutamate receptor N-methyl-D-aspartate (NMDA). The receptor is most likely a tetramer, containing at least one NR1, probably two NR2 and sometimes one NR3 subunit. NR3A is highly expressed during development and present at moderate levels in the adult brain. Both NR3A and NR3B have an attenuating effect on NMDA receptor currents, when incorporated into the receptor complex.

In study I we cloned and sequenced the human NR3A subunit, which at the time only had been described in rat. We found the homology between the rat and human sequences to be high, and most potential phosphorylation and glycosylation sites to be conserved. In contrast to rat tissue, we did not detect the longer splice variant of NR3A in the human central nervous system. We found NR3A mRNA to be expressed both in the developing human brain and spinal cord, with prominent staining of the cortical plate, ventricular zone and the dorsal half of the spinal cord as well as the neuroepithelial layer surrounding the central canal. Interestingly, the neuroepithelial layer expressed very low levels of NR1 mRNA, suggesting that NR3A might have replaced one of the NR1 subunits in NMDA receptors in this area. In study II we investigated the occurrence of NR3A mRNA and protein in adult human brain, its association with the other NMDA receptor subunits and the subunits solubility. We found NR3A mRNA and protein to be expressed in adult human brain, especially in layers II/III and V of cerebral cortex and in thalamus and pons. Expression was low or undetectable in caudate nucleus, claustrum, cerebellum and spinal cord. Further we found NR3A to be associated with NR1, NR2A and NR2B in adult human cortex, and to some extent in the spinal cord. NR3A showed a different solubility profile to the other NMDA receptor subunits, being extracted with milder detergents. Both NR1 and NR3A were found as monomers, dimers and tetramers, as well as in larger protein complexes. In contrast, NR2 subunits were only found in tetramers and in larger protein complexes. This probably reflects the presence of an intracellular pool of unassembled NR1 and NR3A subunits.

To learn more about NR3A and its function we screened a fetal human brain cDNA library for proteins interacting with NR3A. Among a number of potentially interesting candidates we choose MAP1S/C19ORF5 and MAP1B for further analysis. The results are presented in study III and IV respectively. We found MAP1S to be highly expressed throughout brain and spinal cord, predominantly in neurons. MAP1S-EGFP over-expressed in cultured hippocampal cells, localizedboth to dendritic shafts and filopodia. Colocalization of MAP1S-EGFP with beta-tubulin III immunoreactivity was prominent, with synapsin and PSD95 immunoreactivity occasional and with NR3A immunoreactivity frequent in dendritic shafts and sparse in filopodia. Judging from the subcellular distribution of MAP1S and NR3A their interaction might be important for transport through shafts or regulation of NR3A-containing receptors in spines. The distribution of MAP1B immunoreactivity resembled that of MAP1S-EGFP, with prominent overlap with beta-tubulin III immunoreactivity, sparse colocalization with synapsin and SAP102 immunoreactivity, and frequent colocalization with NR3A immunoreactivity in dendritic shafts and infrequent colocalization in spines. To address the function of the NR3A-MAP1B interaction, we investigated the expression and distribution of NR3A in MAP1B deficient mice. These mice expressed increased NR3A and decreased NR1 levels compared to wild type, suggesting that NMDA receptors in the MAP1B deficient mice might have an altered subunit composition. NR3A was equally distributed to filopodia in neurons from MAP1B deficient and wild type mice, indicating that MAP1B is not essential for transport of NR3A-containing NMDA receptors to synaptic sites. Instead the interaction might involve regulating the distribution of NR3A-containing receptors between intracellular pools and the cell surface, as well as the distribution between synaptic and extrasynaptic sites.