Electrophysiological analysis of NMDA receptor subunit changes in the aging mouse cortex

Abstract

NMDA receptors play an important role in memory processes and plasticity in the brain. We have previously demonstrated a significant decrease in NMDARε2 subunit mRNA and protein with increasing age in the C57Bl/6 mouse frontal cortex. In the present study, two-electrode voltage clamp electrophysiology on Xenopus oocytes injected with total RNA harvested from the frontal cortex of young and old C57Bl mice was used to detect changes in receptor composition during aging. Ifenprodil concentration–response curves, magnesium current–voltage curves, and single channel conductances were determined for native receptors. In addition, ifenprodil and magnesium curves were generated for recombinant NMDA receptors of varying subunit ratios. Ifenprodil dose–response curves for all receptors were biphasic. The low affinity component of the curve increased slightly with age, while the high affinity population decreased, mimicking recombinant receptors with decreasing levels of ε2. A decrease in maximal current was also observed in aged animals with decreased levels of ε2, although single channel conductances were identical between young and old mice. In addition, an increase in sensitivity to magnesium was observed for receptors from older animals. Results are consistent with the interpretation that the ε2 subunit is reduced in older mouse frontal cortex. A change in NMDA receptor subunit composition could influence memory processes during aging.

Introduction

The N-methyl-d-aspartate (NMDA) receptor plays an integral role in central nervous system physiology. Activity-dependent modulation of NMDA channel function and associated calcium influx promotes synaptic development and modulation (Watkins, 1994), including processes essential to learning and memory (Bliss and Collingridge, 1993, Morris and Davis, 1994, Tsien et al., 1996). The functional receptor is hypothesized to be either a tetramer (Laube et al., 1998, Rosenmund et al., 1998) or pentamer (Premkumar and Auerbach, 1997) composed of NR1 (mouse ζ), NR2 (mouse ε), and possibly NR3 subunits. Only one NR1 subunit is known to exist, but at least eight splice variants have been reported (McBain and Mayer, 1994, Seeburg et al., 1994). Four NR2 subunits have been characterized (NR2A-D or ε1–ε4 (mouse)). Because the current studies were performed in mouse, but oligonucleotide primers were designed against rat subunit sequences, and because both rat and mouse studies are cited in the literature, both rat and mouse nomenclature will be used.

NR2 subunits confer a variety of properties on the functional NMDA receptor. NR2B (ε2) exhibits a greater affinity than NR2A (ε1) for glutamate (Kutsudwada et al., 1992; Priestly et al., 1995), glycine (Kutsudwada et al., 1992; Priestly et al., 1995), and a variety of antagonists (Whittemore et al., 1997, Tamiz et al., 1998, Chenard and Menniti, 1999), most notably ifenprodil (Williams, 1993, Williams et al., 1993, Gallagher et al., 1996). The gating kinetics of the NMDA receptor complex appear to be much faster when NR2A is present, as opposed to other receptor subunits (Monyer et al., 1992, Vicini et al., 1998). The NR2A and NR2B subunits are much more sensitive to magnesium block than NR2C (ε3) or NR2D (ε4; Kutsuwada et al., 1992; Monyer et al., 1992, Kuner and Schoepfer, 1996, Wagner and Leonard, 1996). NR2A and NR2B reportedly have similar conductances, but much smaller conductances have been reported for NR2C (Stern et al., 1992), NR2D (Wyllie et al., 1996, Cull-Candy et al., 1998), and NR3 (Das et al., 1998).

Multiple studies have characterized the ontogeny of NMDA receptors during the postnatal development of the rodent cortex (Watanabe et al., 1992, Sheng et al., 1994, Kew et al., 1998). Significant NR2A expression is observed 6–10 days after birth (Sheng et al., 1994), and continues to increase to adulthood (Kew et al., 1998). Neither the NR2A mRNA nor protein is detectable at birth in the rodent cortex (Sheng et al., 1994). The NR2B subunit is present at birth (Sheng et al., 1994, Kew et al., 1998), with very little (Kew et al., 1998) or no (Sheng et al., 1994) decrease in expression observed as the animal approaches adulthood. NR3 subunits are expressed at high levels prior to birth, but decrease postnatally. NR3 subunit expression is highly attenuated in the cortex of adult rodents (Ciabarra et al., 1995, Sucher et al., 1995).

Developmental changes in NMDA receptor function and pharmacology parallel the observed changes in subunit expression. During postnatal development, NMDA receptors become less sensitive to glycine (Kew et al., 1998). Excitatory post-synaptic currents exhibit a decrease in rise time and decay time (Kirson and Yaari, 1996, Flint et al., 1997). Experiments with the non-competitive, allosteric antagonist ifenprodil demonstrate a major change in the receptor population from a single, high affinity binding site to two populations, exhibiting both high and low affinity sites (Williams et al., 1993, Kew et al., 1998). Only the NR2B subunit confers a high affinity to ifenprodil, suggesting a change from a pure to a mixed subunit population during postnatal development.

Previous studies have shown that [³H]glutamate binding to the NMDA binding site is affected by aging in the frontal cortex of C57Bl mice (Magnusson and Cotman, 1993, Magnusson, 1995). Subsequent in situ hybridization studies demonstrated a significant decrease in ζ and ε2 mRNA in the frontal cortex of aged C57Bl mice, although changes in ε1 were not significant (Magnusson, 2000). In addition, a significant decline in the amount of ε2 subunit protein expression between 3 and 30 month old mice was seen with Western blot analysis (Kuehl-Kovarik et al., 1998). The frontal cortex is responsible for multiple functions, including memory and executive functions involved in working and reference memory (Kolb et al., 1983, Goldman-Rakic, 1990, Miotto et al., 1996, Owen et al., 1996). Spatial reference memory deficits have been detected in aged C57Bl mice and show a relationship to changes in binding to the NMDA receptor in the frontal cortex (Magnusson, 1998). The binding changes also showed correlations with changes in expression of ε2 mRNA (Magnusson, 2000), indicating that changes in NMDA receptor subunit composition could influence memory function during the aging process.

To further explore changes in NMDA receptor subunit composition in the C57Bl mouse frontal cortex with aging, we have used two-electrode voltage clamp and single channel analysis in Xenopus oocytes. Functional analysis of receptors from aged brain is extremely challenging, and the experiments performed describe a novel methodology for addressing this issue. Oocytes were injected with total RNA isolated from the brains of young or old mice. We used sensitivity to ifenprodil and to magnesium, as well as single channel analysis, to detect changes in NMDA receptor pharmacology and biophysics. In addition, these same techniques were performed on recombinant NMDA receptors of various stoichiometric ratios in order to attempt to mimic the effects of ifenprodil on a mixed NMDA receptor population.

Results from these studies were consistent with the findings of a decrease in the NMDARε2 subunit during aging in the frontal cortex of C57Bl/6 mice (Kuehl-Kovarik et al., 1998, Magnusson, 2000). Overexpression of the ε2 subunit has been shown to enhance memory processes in mice (Tang et al., 1999). Thus, changes in receptor subunit composition could be relevant to the memory changes observed in aged animals.