The cloning of the glutamate-gated ion channels of the brain revealed
an unexpected level of complexity: there are many different genes that
encode distinct subunits of the receptor/channel complex and an even
larger number of possible receptor subunit combinations. Many-nearly a
ll-of these gene products an expressed in the retina, and the question
s that we face today are: how are they used and why an there so many?
Answers to these questions will be found at several levels. At the lev
el of transcription, we have learned that different sets of subunits a
re expressed by different retinal neurons. Little is known about the t
ranscriptional control of these genes, so it remains to be determined
whether these patterns of expression reflect the need for different ge
ne products in different retinal neurons or whether these patterns of
expression reflect the functional constraints of gene expression. Anot
her level of complexity is caused by alternative splicing, and here we
report that at least four and possibly all eight of the different NMD
AR1 transcripts are present in the mouse retina. The consequences of t
his alternative splicing are poorly understood, but antibodies directe
d against the two different possible C-termini of NMDAR1 label many of
the same cell types. It is possible that these different splice varia
nts are combined to generate the channels. While immunohistochemistry
provides us with a glimpse of the subunit proteins, much remains to be
learned about their half-life within a retinal cell, their intracellu
lar trafficking, their regulation at the synapse, and the proteins ass
ociated with their cytoplasmic domains. An approach we have taken towa
rds studying the dynamic properties of receptor subunits has been to f
use them to the cDNA encoding the jellyfish Green Fluorescent Protein.
This makes it possible to follow functional subunits in transfected c
ells over time and to begin to measure the mobility of the protein. (C
) 1998 Published by Elsevier Science Ltd. All rights reserved.