In vitro, EtOH acts as an allosteric inhibitor of the N-methyl-D-aspartate (NMDAR) class of ionotropic glutamate receptors at doses that induce behavioral intoxication (Lovinger et al., 1989;Masood et al., 1994). Chronic exposure to EtOH in vitro and in vivo leads to up-regulation of NMDAR (Carpenter-Hyland et al., 2004; Crabbe et al., 1991; Kumari and Ticku, 2000; Roberto et al., 2006; Smothers et al., 1997). Behaviorally, uncompetitive NMDAR antagonists such as ( + )-5-methyl-10,11-dihydro-SH-dibenzo[a,d]cyclohepten- 5,10-imine maleate (MK-801, dizocilpine) and ketamine substitute for relatively high doses of EtOH in rodent and monkey drug discrimination tests and mimic the subjective feeling of EtOH intoxication in humans (Grant, 1999; Hodge et al., 2006; Krystal et al., 2003; Rassnick et al., 1992). Furthermore, acute administration of NMDAR antagonists decreases EtOH self-administration and reward-related responses to EtOH in rats and mice (Boyce-Rustay and Cunningham, 2004; Krystal et al., 2003; McMillen et al., 2004; Rassnick et al., 1992). On the other hand, NMDAR antagonist treatment potentiates the acute depressant (sedative/hypnotic) effects of EtOH in rodents (Beleslin et al., 1997; Boyce-Rustay and Holmes, 2005; Silveri and Spear, 2002; Wilson et al., 1990). This latter finding is congruent with evidence from some but not all studies that the Long Sleep line of mice selected for relatively high sensitivity to EtOH’s depressant effects exhibit low MK-801-binding, diminished locomotor stimulant effects to MK-801 and attenuated NMDAR-mediated responses in hippocampus and striatum, as compared to Short Sleep mice (Daniell and Phillips, 1994; Hanania and Zahniser, 2002; Hanania et al., 2000, 2002; Musleh et al., 1996; Velardo et al., 1998; Wilson and Collins, 1996; Zahniser et al., 1999).

The molecular basis of these complex interactions between the NMDAR and EtOH is still not well understood. NMDAR are heteromeric complexes composed of an obligatory NR1 subunit together with NR2 (NR2A-D) and sometimes NR3 subunits (Schorge and Colquhoun, 2003). NMDAR subunit composition affects sensitivity to EtOH in vitro. Mutagenesis of NR1 can bidirectionally alter sensitivity to EtOH (Jin and Woodward, 2006; Ronald et al., 2001). NR2A- and NR2B-containing NMDAR are more sensitive to EtOH than NR2C- or NR2D-containing NMDAR (Masood et al., 1994; Mirshahi and Woodward, 1995). Previous studies have shown that NR2B-selective antagonists such as ifenprodil and Ro 25-6981 potentiate the depressant effects of EtOH in mice (Boyce-Rustay and Holmes, 2005; Malinowska et al., 1999; Yaka et al., 2003), while NR2A knockout (KO) mice exhibit normal acute responses to EtOH (Boyce-Rustay and Holmes, 2006a; Sato et al., 2006). Beyond these data, little is known about the relative contribution of specific NMDAR subunits to EtOH’s behavioral effects and interactions between NMDAR antagonists and EtOH on behavior (Woodward, 1999).

NMDAR signaling is closely linked to the L-alphaamino-3-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPA) GluR1 receptor subtype, particularly in the context of NMDAR-mediated neuroadaptation to (psychostimulant) drugs of abuse (Chao and Nestler, 2004; Wolf et al., 2004). Moreover, EtOH inhibits AMPA-related currents in cultured rat cortical and medial septal neurons (Fischer et al., 2003; Frye and Fincher, 2000) [although AMPA receptors appear to be less sensitive to EtOH than NMDARs (Costa et al., 2000; Lovinger et al., 1989)]. Recent studies have shown that KO mice lacking GluR1 display normal basal responses to EtOH (Cowen et al., 2003) but fail to respond to the locomotor hyperactivity-inducing effects of MK-801 (Wiedholz et al., 2008). The potential role of GluR1 in mediating MK-801 effects on EtOH intoxication has not been studied.

The aim of the present study was to further characterize the behavioral, genetic and molecular basis of NMDAR × EtOH interactions in vivo. We first characterized the effects of the uncompetitive NMDAR antagonist MK-801 on multiple measures of acute EtOH treatment (accelerating rotarod ataxia, hypothermia and sedation/hypnosis) in two genetically distinct inbred strains, C57BL/6J and 129/SvImJ (Payseur and Place, 2007). For comparison, the effects of another compound with NMDAR antagonist properties, phencyclidine, was also assessed. Next we tested whether functional inactivation of the NMDAR NR2A, NR2B or AMPA GluR1 subunits would alter MK-801’s effects on EtOH behaviors by testing whether these effects were altered following gene deletion of the NR2A or GluR1, or pharmacological blockade of NR2B. Results demonstrate that NMDA antagonism strongly potentiates acute sensitivity to EtOH in a behavior-specific manner, but does so in a manner that is either unaffected or, in one case, partially attenuated by loss of function of any one of three of the major glutamate receptor subunits.

In non-mutant mice, pretreatment with MK-801 markedly and dose-dependently increased both the sedative/hypnotic (sleep time) effects of 3 g/kg EtOH and the ataxic effects of 1.75 g/kg EtOH measured in the accelerating rotarod. BECs at recovery from the sedative/hypnotic effects of EtOH paralleled the prolonged sleep time produced by MK-801 treatment; such that BECs were lower at awakening in the longer-sleep MK-801-treated mice relative to mice treated with EtOH alone. (The same pattern of BECs was seen in the NR2A KO, NR2B antagonist and GluR1 KO experiments discussed below). This implies that the EtOH-potentiating effects of MK-801 were not simply an artifact of the drug interfering with EtOH clearance. Further discounting effects of MK-801 on EtOH metabolism, MK-801 treatment had no affect on the hypothermic effects of EtOH even though these effects were measured with the same dose and over the same timeframe as sedation/hypnosis. Thus, the first conclusion of the present study is that MK-801’s effects on EtOH were responsive-selective, with an effect on two behavioral measures of “intoxication” but no effect on a physiological measure.

MK-801 potentiation of EtOH’s depressant and ataxic effects was replicated in two genetically distinction inbred strains, C57BL/6J and 129S1. Replication across strains indicates that the MK-801 and EtOH interaction generalizable across diverse genetic backgrounds (although a fuller strain comparison would be needed to substantiate this). Moreover, given the greater sensitivity to the ataxic and depressant effects of EtOH per se in 129S1 relative to C57BL/6J, present data indicate that MK-801 is able to potentiate EtOH’s effects even against a high-sensitive baseline phenotype. Finally, both C57BL/6J and 129S1 mice showed increased sensitivity to EtOH’s sedative/hypnotic effects following treatment with another drug with NMDAR antagonist actions, phencycline. Although the magnitude of the interaction was less pronounced than for MK-801, as reported previously (Daniell, 1990), this shows that the potentiating effects of MK-801 extend to other NMDAR antagonists.

Present data confirm and extend earlier studies showing that treatment with MK-801, phencyclidine or with other NMDAR antagonists, including memantine, ketamine or 2-amino-5-phosphonovaleric acid, potentiates EtOH’s locomotor sedating and sedative/hypnotic but not hypothermic effects in rats (Beleslin et al., 1997; Brunet et al., 1985;Daniell, 1990; Danysz et al., 1992; Poch et al., 1990; Silveri and Spear, 2002, 2004), and various mouse strains including C57BL/6J, DBA/2J and the Long Sleep and Short Sleep lines (Goldberg et al., 1983; Kuribara, 1994; Maas et al., 2005; Meyer and Phillips, 2003a,b; Shen and Phillips, 1998; Wilson et al., 1990). Our findings are also consistent with reports that NMDAR antagonists, includingMK-801, augment the ataxic effect of EtOH in rats tested on the rotarod (McMillen et al., 2004), NSA mice tested on the horizontal wire test (Vanover, 1999) and DBA/2J mice tested on the grid test (Meyer and Phillips, 2003a). This consistency across different measures of ataxia is itself notable given evidence that these assays measure dissociable behaviors mediated by distinct molecular and genetic factors (Boyce-Rustay and Holmes, 2006a; Rustay et al., 2003a). Finally, although not tested in our study, MK-801 has also been shown to accentuate the locomotorstimulant effects of EtOH in DBA/2J mice (Meyer and Phillips, 2003a; Shen and Phillips, 1998). Thus, the present data further strengthen the literature demonstrating a marked potentiating effect of MK-801 and other NMDAR antagonists on a variety (but not all) of measures of EtOH’s acute intoxicating actions.

Behavioral indices of the magnitude of acute EtOH intoxication such as ataxia and sleep time are composite measures of initial sensitivity to EtOH’s effects and acute functional tolerance (AFT) to these effects (Erwin and Deitrich, 1996; Mellanby, 1919; Ponomarev and Crabbe, 2004; Tabakoff and Ritzmann, 1979). As with other forms of drug tolerance, AFT to EtOH’s behavioral actions is a neuroadaptive process. NMDARs are known to play a critical role in the neuroadaptive processes associated with exposure to other drugs of abuse (Hyman, 2005) and, more generally, to various forms of neural plasticity, including those thought to underlie learning (Malenka and Bear, 2004). MK-801 and other NMDAR antagonists (e.g., CGP 39551, ketamine) can block the development of tolerance to EtOH’s sedative/hypnotic and ataxic effects in rats and C57BL/6J mice measured over days (i.e., not AFT) (Karcz-Kubicha and Liljequist, 1995; Khanna et al., 1991, 1993, 1997; Szabo et al., 1994; Wu et al., 1993) [but see grid test ataxia in DBA/2J mice (Meyer and Phillips, 2003a)]. MK-801 also blocks the development of locomotorsensitization to EtOH in DBA/2J and Swiss mice, another form of neuroadaptation (Broadbent et al., 2003; Kotlinska et al., 2006; Meyer and Phillips, 2003a, 2007). Two studies that have explicitly examined AFT, have found that MK-801 and ketamine both inhibit the development AFT to EtOH’s ataxic (tilt test) or sedative/hypnotic effects in rats (Khanna et al., 2002; Silveri and Spear, 2004). We found that MK-801 treatment did not affect the initial latency to lose the righting reflex nor BECs at loss. While this could be construed as support for a lack of MK-801 on initial sensitivity and therefore more probably an effect on AFT, it is weak evidence of such a dissociation. Thus, the potential influence of MK-801 on AFT is not yet fully clear and this remains an important question for future study.

As noted in the Introduction, EtOH acts as an allosteric inhibitor of NMDAR-mediated neuronal responses at intoxicating doses in vitro (Lovinger et al., 1989; Masood et al., 1994). Woodward and colleagues among others have shown that, in vitro, the functional state of the NR1 subunit modulates EtOH effects on the NMDAR (Jin and Woodward, 2006; Ronald et al., 2001), while both the NR2A and NR2B subunits, but less so NR2C and NR2D, also determine sensitivity to EtOH (Masood et al., 1994; Mirshahi and Woodward, 1995). Replicating earlier results (Boyce-Rustay and Holmes, 2005, 2006a; Sato et al., 2006), gene knockout of NR2A did not affect the (rotarod) ataxic, sedative/hypnotic or hypothermic effects of EtOH on the same measures and EtOH doses that were clearly potentiated by MK-801. Pharmacological antagonism of NR2B with Ro 25-6981 in C57BL/6J mice produced only a modest increase in the sedative/hypnotic response to EtOH in one experiment (Ro 25-6981 alone) but not another (Ro 25-6981 × MK-801), and was without effect on EtOH’s ataxic and hypothermic effects. These relatively modest effects are similar to those obtained in earlier studies of Ro 25-6981 and other NR2B-selective antagonists in mice (Boyce-Rustay and Holmes, 2005; Malinowska et al., 1999; Yaka et al., 2003). Thus, taken together, these data demonstrate that functional inactivation of either NR2A- or NR2B-containing NMDARs alone is insufficient to mimic the potent EtOH-potentiating action of the subunit-non-selective NMDAR antagonist MK-801.

Inactivation of NR2A- or NR2B-containing NMDARs could have conceivably altered MK-801’s effects in various ways. If MK-801’s antagonist effects were dependent upon NR2A- or NR2B-containing NMDARs then inactivation of either subunit would have been predicted to diminish MK- 801’s effect on EtOH responses. Contrariwise, if MK-801’s antagonist action was independent of NR2A or NR2B then the combination of MK-801 blockade of NMDARs with inactivation of NR2A- or NR2B-containing receptors could have further potentiated EtOH’s behavioral effects. Data showed that NR2B antagonism via Ro 25-6981 treatment failed to affect MK-801’s EtOH-potentiating effects on any behavioral measure. On the other hand, although KO of NR2A also failed to alter MK-801’s effects on EtOH-induced ataxia or hypothermia, NR2A KO mice showed a general reduction in the sleep time response relative to WT controls. Because NR2A KO mice show normal sedative/hypnotic response to EtOH per se (this experiment, (Boyce-Rustay and Holmes, 2005, 2006a), this reduction appears to be most parsimoniously attributed to a partial reduction in the response to MK-801s potentiating effects. As such, this suggests that the NR2A subunit contributes to MK-801’s effects on at least this one specific measure of EtOH intoxication, and provide novel insight into the molecular basis of the MK-801 × EtOH interaction.

Notwithstanding this one difference, the overarching conclusion is that functional inactivation of NR2A nor NR2B is not sufficient to ablate the ability of a NMDAR antagonist to potentiate EtOH intoxication. There are a number of possible explanations for this. One is that NR2A- and NR2Bcontaining NMDARs both contribute to MK-801’s effects on EtOH behaviors, but that loss of any one is insufficient to markedly affect responses other than sedation/hypnosis. Alternatively, other NMDAR subunits may be more important than NR2A and NR2B for these effects. In this context, NR2C-, NR2D- and NR3A/NR3B-containing NMDAR all to some extent interact with EtOH in vitro (Masood et al., 1994; Mirshahi and Woodward, 1995; Smothers and Woodward, 2007) and represent possible candidates. However, their contribution to either MK-801’s or EtOH’s in vivo actions remains unclear (see Woodward (1999)). Another possibility is that MK-801s effects on EtOH sensitivity are in fact unrelated to the drug’s NMDAR antagonistic properties. Although seemingly unlikely, this cannot be fully excluded given the drug’s known effects on for example norepinephrine (Snell et al., 1988), dopamine (Seeman et al., 2005) and acetylcholine (Ramoa et al., 1990) function.

The present data show that MK-801’s effects on EtOH responses were normal in GluR1 KO mice [as were baseline EtOH behaviors, replicating (Cowen et al., 2003)]. NMDARs act in concert with AMPA receptors to mediate certain forms of behavior, notably learning and neuroadaptation to drugs of abuse (Chao and Nestler, 2004; Malenka and Bear, 2004; Wolf et al., 2004). Our data demonstrate that the MK- 801 × EtOH interaction is independent of functioning AMPA GluR1 subunits. Interestingly, this contrasts markedly with the recent demonstration that the same line of GluR1 KO mice fail to respond to the locomotor hyperactivity-inducing effects of MK-801 at doses similar to those tested in the current study (Wiedholz et al., 2008). Thus the contribution of GluR1 to MK-801’s effects on locomotor activity and EtOH behaviors are dissociable; likely reflecting the recruitment of different neural circuitry. More generally, the lack of GluR1 KO effects in the present study adds to the somewhat surprising conclusion that MK-801’s ability to potentiate EtOH responses is insensitive to loss of key molecular components mediating glutamate signaling. Thus, the mechanism by which MK-801 interacts with EtOH to affect behavioral intoxication remains to be elucidated. On a final note, while we found no evidence of statistical interactions between sex, genotype and MK-801 treatment, the possibility that larger sample sizes would reveal sex differences in the manner in which either GluR1 KO mice, or the other inbred or mutant lines tested, respond to MK-801-potentiation of EtOH’s effects should not be ruled out.