Introduction

An individual’s risk of developing alcoholism may be associated with sensitivity to the initial behavioral and subjective effects of alcohol (ethanol). For example, some researchers have suggested that persons at risk for the development of alcoholism are more sensitive to the stimulant-like subjective effects of ethanol, while less sensitive to its ataxic effects (de Wit et al. 1987; Holdstock et al. 2000; Newlin and Thomson 1991; Schuckit 1984, 1994). Further, repeated administration of ethanol can result in increased sensitivity to the stimulant-like subjective effects of ethanol, and this increased sensitivity occurs more readily in persons at risk for the development of alcoholism (Newlin and Thomson 1991). While subjective effects are not easily measured in animal models of ethanol sensitivity, ethanol has stimulant, ataxic, and sedative effects on simple behaviors involving locomotor activity in rodents (Breese et al. 1985; Crabbe 1983; Erickson and Kochhar 1985). In some rodent strains, repeated treatment with ethanol has been shown to increase sensitivity to the locomotor stimulant effects of ethanol, compared to ethanol-naïve animals (Hoshaw and Lewis 2001; Phillips et al. 1994; Pillai et al. 1998). This phenomenon, known as behavioral sensitization, is thought to be a behavioral index of the neuroadaptive response to repeated ethanol treatment. While several experiments have dissociated the neural mechanisms underlying sensitization from those underlying addiction (Ben-Shahar et al. 2005; Cunningham et al. 2002; Sanchez et al. 1996; Wise and Munn 1993), some researchers have suggested that the sensitizing effects of several abused drugs, including ethanol, are related to their reinforcing effects (Bowers et al. 2004; Chester and Cunningham 1999; Ferrario and Robinson 2007; Grahame et al. 2000; Kalivas et al. 1998; Lessov et al. 2001a, b; Robinson and Berridge 1993; Samaha et al. 2002; Tindell et al. 2005).

Ethanol is an allosteric inhibitor of the N-methyl-d-aspartate (NMDA) receptor (Lovinger et al. 1989; Wright et al. 1996) and has direct modulatory effects on other ion channels as well (El-Fakahany et al. 1983; Messing et al. 1986). Chronic ethanol treatment has been shown to alter sensitivity to NMDA-mediated responses (Crabbe et al. 1991; Gulya et al. 1991; Nagy et al. 2004; Roberto et al. 2006; Sheela Rani and Ticku 2006), suggesting that changes in the NMDA receptor number or function may be a compensatory response to ethanol inhibition of these receptors. Ethanol-induced stimulation is thought to be related to its effects on several neurotransmitter systems, including those modulated by NMDA receptors (Meyer and Phillips 2003b; Phillips and Shen 1996; Shen and Phillips 1998). The NMDA receptor antagonist MK-801 has been shown to alter sensitization to both ethanol (Broadbent and Weitemier 1999; Camarini et al. 2000) and other abused drugs (Trujillo 2002; Vanderschuren and Kalivas 2000). However, some researchers have argued that MK-801 affects the development of sensitization by altering the acute locomotor and/or stimulus effects of the sensitizing drug, rather than blocking the neuroadaptations that occur during sensitization to the drug alone (Meyer and Phillips 2003a; Tzschentke and Schmidt 2000).

Cross-sensitization paradigms provide an alternate approach to studying the neural mechanisms of behavioral sensitization. In such a paradigm, sensitized and nonsensitized animals are compared for their responses to novel pharmacological agents, which are chosen based on their ability to modulate the neurochemical systems associated with behavioral sensitization. Ethanol-sensitized rodents have been shown to display enhanced responses to cocaine, amphetamine, and morphine, compared to their nonsensitized counterparts (Itzhak and Martin 1999; Lessov and Phillips 2003; Manley and Little 1997). These examples of cross-sensitization suggest that the neural systems involved in determining sensitivity to the novel drugs are altered upon repeated ethanol treatment. For example, Itzhak and Martin (1999) found that bidirectional cross-sensitization between ethanol and cocaine was related to upregulation of the dopamine transporter, suggesting that alterations of dopaminergic systems are involved in the neuroadaptations to these drugs. The advantage of cross-sensitization studies is that ethanol is never given in combination with other drugs, which avoids some interpretational complications caused by drug interactions or additive drug effects (Kotlinska et al. 2006; Meyer and Phillips 2003a; Shen and Phillips 1998).

We conducted several cross-sensitization studies to examine whether ethanol sensitization would result in altered sensitivity to various NMDA receptor ligands. To this end, we measured the behavioral responses to two NMDA receptor antagonists (MK-801 and ketamine) and to NMDA (the prototypical NMDA agonist) in an inbred strain (DBA/2J) and a heterogeneous stock (HS) of mice. Because ethanol is an allosteric inhibitor of NMDA receptors and some previous findings have suggested that NMDA receptors are necessary for and are altered during ethanol sensitization, we hypothesized that ethanol sensitization would result in cross-sensitization to the NMDA receptor antagonists. We also examined the phenotypic correlations between the magnitude of ethanol sensitization and the locomotor response to MK-801 and ketamine and the convulsant response to NMDA in HS mice. We predicted a positive correlation between the magnitude of ethanol sensitization and the level of response to the NMDA receptor antagonists, and a negative correlation between ethanol sensitization and the response to NMDA.

Materials and methods

Subjects

We chose two genotypes of mice for these experiments. Male DBA/2J mice were chosen because they have been shown to be particularly sensitive to the acute and sensitizing effects of ethanol (Dudek and Phillips 1983; Phillips et al. 1994). These mice were purchased from The Jackson Laboratory (Bar Harbor, ME) at 4–6 weeks of age and allowed to acclimate to the Portland Veterans Affairs Medical Center (VAMC) animal care colony room for at least 1 week before testing began. HS mice are a genetically heterogeneous population of mice created from an intercross of eight inbred strains (McClearn and Kakihana 1981). The population maintained in Portland was originally used as a control group for a selective breeding experiment conducted by Crabbe et al. (1983). These mice are bred and maintained in the Portland VAMC animal care colony, and were chosen (1) to assess the possible influence of genetic background on our cross-sensitization results and (2) because individuals of this population are variable in their magnitude of sensitization to ethanol and are thus useful for investigating phenotypic correlations between this trait and other responses. Both male and female HS mice were used.

All mice were housed in groups of 2–5 in 28 × 18 × 13 (l × w × h)-cm cages with corn-cob bedding. Food (Purina Laboratory Rodent Chow; Purina Mills, St. Louis, MO) and water were suspended from wire-top lids and were available at all times except during the test sessions. Room temperature was maintained between 20 and 22°C. Mice were aged 45 to 74 days and weighed 18.7 to 31.5 g at the beginning of testing. Testing occurred between 08:00 hours and 16:00 hours (the colony lights were on from 06:00 hours to 18:00 hours). All procedures were performed in accordance with the Institutional Animal Care and Use Committee and National Institutes of Health guidelines for the care and use of laboratory animals. Experiments were designed in such a way as to minimize suffering and utilize the smallest number of animals possible.

Drugs

All drugs were prepared in 0.9% physiological saline (Baxter Healthcare, Deerfield, IL), except for NMDA (Sigma, St. Louis, MO), which was prepared in distilled water, according to Finn and Crabbe (1999). Ethanol (Pharmco Products, Brookfield, CT) was diluted from 100% to a final concentration of 20% (v/v). Mice were injected intraperitoneally (i.p.) with 2 or 2.5 g/kg ethanol. These two doses were chosen from previous studies indicating that they have significant stimulant effects (Dudek and Phillips 1983) and produce robust sensitization to ethanol (Lessov and Phillips 2003; Phillips et al. 1994). MK-801 and ketamine (Sigma) were injected i.p. in volumes of 10 ml/kg. The 0.25-mg/kg dose of MK-801 and 10- and 20-mg/kg doses of ketamine were chosen from previous studies to induce modest to robust stimulation in DBA/2J mice (Meyer and Phillips 2003b) and because other laboratories have found cross-sensitization between other drugs of abuse and this dose of MK-801 (Biala and Weglinska 2004; Iijima et al. 1996; Ito et al. 2006; Itzhak and Martin 1999; Vanderschuren et al. 1997; Xu and Domino 1999). Control groups were injected with 0.9% physiological saline. NMDA (8 mg/ml; Sigma) was infused via the lateral tail vein at a rate of 0.5 ml/min. This procedure was adapted from previous studies (Finn and Crabbe 1999), which used this procedure to determine sensitivity to NMDA-induced convulsions in several strains of mice.

Behavioral tests and equipment

Activity monitors

Mice were tested in 40 × 40 × 30 (l ×w × h)-cm clear acrylic plastic boxes, covered by plastic lids with 0.64-cm-diameter holes for ventilation. These boxes were placed in automated activity monitors (Accuscan Instruments, Columbus, OH), which consisted of eight pairs of intersecting infrared photobeams, located 2 cm above the cage floor. The occlusion of these photobeams was used to calculate the distance traveled (in cm) by a mouse in 5-min time bins for 20–60-min test sessions. The activity monitors were housed in individual, opaque sound attenuation chambers (Flair Plastics, Portland, OR), each illuminated by a 15-W fluorescent bulb and ventilated by a fan that also provided background noise.

NMDA-induced convulsions Because the locomotor response to systemic NMDA is not well characterized in mice, we chose to measure sensitivity to NMDA-induced convulsions as our measure of ethanol-induced changes in NMDA receptor-mediated systems. Mice were restrained in small plastic cylinders (5 cm diameter). NMDA (8 mg/ml) infusions were performed using a butterfly needle connected to a Cole-Parmer (Vernon Hills, IL) infusion pump. The tail was placed briefly in warm water to dilate the lateral tail vein, after which the needle was inserted into the vein. NMDA was infused at a rate of 0.5 ml/min, and the latencies to running/bouncing clonus (whole-body clonus, including running and jumps) and tonic hind limb extension (extreme rigidity, with forelimbs and hind limbs extended caudally) were measured (Finn and Crabbe 1999). These latencies were converted to a cumulative dose of NMDA by calculating the dose of NMDA infused (in mg) per kilogram body weight for each mouse, providing a calculation of the effective dose to produce each seizure type in each mouse.

Behavioral procedures

All experiments consisted of an ethanol treatment phase followed by a challenge phase. The treatment phase was similar for all experiments. On days in which a behavioral test was scheduled, mice were moved to the testing room 45–60 min before testing began, to allow habituation to the testing room. On days 1–3, mice were weighed and injected with saline before being placed immediately into the activity monitors for 20 or 60 min, depending on the experiment. These days were included to provide habituation to the testing procedures, and day 3 data were used as a measure of baseline activity. On day 4, mice were given 2-g/kg ethanol or saline injections, depending on their group assignment, and placed immediately in the testing chamber. This day provided a measure of the acute response to ethanol. Mice were not tested on days 5–14 but continued to receive saline or ethanol treatments in the colony room. Ethanol-treated DBA/2J mice received 2 g/kg on days 5–14, while HS mice received 2.5 g/kg ethanol. In our experience, the 2.5-g/kg pretreatment dose produces robust sensitization in HS mice (Lessov and Phillips 2003; Lessov et al. 2001b), and the 2-g/kg pretreatment dose works well in DBA/2J mice in our and other laboratories (Broadbent et al. 2003; Phillips et al. 1994). Finally, on day 15, mice were given 2-g/kg ethanol or saline injections and placed immediately in the testing chamber. This test day provided a within-group measure of the sensitized response to ethanol. The challenge phases after the induction of ethanol-induced sensitization and group assignments for each experiment are described below and summarized in Table 1.

Table 1 Experimental design for experiments 1–6
Full size table

Experiment 1: response to MK-801 and ketamine in DBA/2J mice

The purpose of this experiment was to determine whether cross-sensitization to MK-801 and ketamine would be seen in ethanol-sensitized DBA/2J mice. Two groups of mice were given repeated saline or ethanol (2 g/kg) injections during the treatment phase. On day 15, ethanol-treated groups were given 2 g/kg ethanol, and saline-treated groups were given saline injections before being placed into the activity monitors. The same subjects were tested on days 16, 17, and 18, during which all mice were injected with 0.25 mg/kg MK-801, 2 g/kg ethanol, and then 20 mg/kg ketamine, respectively, before being placed into the activity monitors for 60 min. This test duration was chosen because the peak locomotor response to MK-801 occurs between 20 and 40 min after an i.p. injection and to standardize the test duration across days (see Figs. 1 and 2). For this and subsequent experiments, the additional test for ethanol sensitization on day 17 was included to verify that mice were still sensitized after the drug challenge on day 16. For the ethanol and ketamine challenges, subjects were left in the activity boxes for the entire 60-min test period, although only the peak response to these drugs (within the first 20 min) are presented.

Fig. 1
figure 1

Response to MK-801 and ketamine in ethanol-sensitized vs saline-treated DBA/2J mice. a Distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates significantly enhanced response to 2 g/kg ethanol on day 15, compared to that on day 4 (p < 0.01). b Time course of the response to 0.25 mg/kg MK-801 in repeated ethanol- and saline-treated mice. c Time course of the response to 2 g/kg ethanol on day 17. Asterisk indicates that repeated ethanol-treated mice had a larger response to ethanol (p < 0.01), compared to saline treated mice. d Time course of the response to 20 mg/kg ketamine on day 18. There were no statistically significant differences between the repeated saline and ethanol groups in the time course of responses to ketamine, ethanol, or MK-801 (b, c, and d, respectively). All values are mean ± SEM. Ethanol- and saline-treated groups contained 12 and 10 mice, respectively

Full size image
Fig. 2
figure 2

Response to MK-801 in repeated ethanol- and saline-treated HS mice. a Distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates a significantly enhanced response to ethanol on day 15, compared to that on day 4 (p < 0.01). b Time course of the response to 0.25 mg/kg MK-801 in repeated ethanol- and saline- treated mice. c Time course of the response to 2 g/kg ethanol on day 17. Asterisk indicates that repeated ethanol-treated mice had a larger response to ethanol (p < 0.05), compared to saline-treated mice. There were no statistically significant differences between the repeated saline and ethanol groups in the time course of responses to MK-801 or ethanol (b and c, respectively). All values are mean ± SEM. Each group contained 18 mice

Full size image

Experiments 2 and 3: response to MK-801 and ketaminein HS mice

Cross-sensitization to MK-801 and ketamine was examined in ethanol-sensitized HS mice in two additional experiments. Because a limitation of experiment 1 is that the response to ketamine on day 18 may have been influenced by prior MK-801 treatment, these experiments tested the responses to MK-801 and ketamine in separate groups of mice. In both experiments, two groups of mice received either saline or 2 g/kg ethanol on day 4, followed by either saline or 2.5 g/kg ethanol on days 5–14, respectively. On day 15, ethanol-treated groups were given 2 g/kg ethanol, and saline-treated groups were given saline injections before being placed into the activity monitors. In experiment 2, the same repeated ethanol- and saline-treated mice were then tested for their responses to 0.25 mg/kg of MK-801 on day 16 and 2 g/kg ethanol on day 17. The test duration was 60 min on all days of this experiment. In experiment 3, mice were challenged with 10 or 20 mg/kg of ketamine on day 16 and 2 g/kg of ethanol on day 17. The test duration was 20 min on all days of this experiment, which corresponds to the peak locomotor effects of ketamine and ethanol.

Experiment 4: correlational analysis of ethanol sensitization and the response to MK-801 and ketamine in HS mice

Another method for exploring cross-sensitization is through phenotypic correlation. In such a paradigm, the magnitude of sensitization to one drug in a sample of individuals is examined for correlation with the magnitude of response to a novel compound or other neurological measure. This experiment was designed to determine whether magnitude of ethanol-induced sensitization would predict magnitude of locomotor response to ketamine and MK-801. After receiving 2-g/kg ethanol injections followed by locomotor activity testing on day 4, HS mice received repeated 2.5-g/kg ethanol injections during the treatment phase, and then these same subjects were tested with ethanol (2 g/kg), MK-801 (0.25 mg/kg), ethanol (2 g/kg), and finally ketamine (20 mg/kg) on days 15, 16, 17, and 18, respectively. The test duration in this experiment was 60 min. For the ethanol and ketamine challenges, subjects were left in the activity boxes for the entire 60-min test period, although only the peak response to these drugs (within the first 20 min) are presented. Sensitization was defined as an increase in the response to ethanol after repeated ethanol administration (i.e., day 15 minus day 4 and day 17 minus day 4).

Experiment 5: response to NMDA in DBA/2J mice

This experiment was designed to determine whether ethanol-sensitized DBA/2J mice would display altered sensitivity to NMDA, using the convulsant response to NMDA tail-vein infusions. During the treatment phase of this experiment, DBA/2J mice were repeatedly treated with 2 g/kg of ethanol or saline. On day 16, NMDA (8 mg/ml) was administered via the tail vein, and the variables described under “NMDA-induced convulsions” were calculated. The test duration on days 1–15 was 20 min.

Experiment 6: response to NMDA in HS mice

This experiment was designed to determine whether magnitude of ethanol sensitization would predict NMDA sensitivity. HS mice received 2 g/kg of ethanol on day 4, followed by repeated 2.5-g/kg ethanol injections during the treatment phase. On day 15, the sensitized response to 2 g/kg of ethanol was measured (i.e., day 15 − day 4). On day 16, the convulsant response to tail-vein infusions of NMDA was measured as in experiment 5.

Statistics

For the activity data, the dependent variable was distance traveled in centimeters. Analysis of variance was conducted with repeated measures when appropriate, followed by simple main effects (SME) analysis for significant two-way interactions and planned mean comparisons when appropriate. For correlations, Pearson product-moment correlations were calculated between the acute ethanol response, ethanol sensitization, and the responses to MK-801 and ketamine. For the NMDA-induced convulsions, differences between the ethanol- and saline-treated groups were assessed using the Student’s t test. Correlations between the acute ethanol response, ethanol sensitization, and convulsant responses to NMDA were also calculated. Correlational analyses were not performed in experiments using inbred DBA/2J mice because variation among individuals must be largely attributable to nongenetic causes. All ethanol and ketamine data used in analyses were for the initial 20-min test period, when the stimulant effects of the drug are most robust. MK-801 data for the entire 60 min were used in all analyses, as this drug has a longer duration of effect.

Results

Experiment 1

As expected, DBA/2J mice exhibited significant locomotor sensitization to ethanol. A significant Treatment × Day interaction (F [4, 96] = 22.7; p < 0.01) indicated that the pattern of locomotor activity differed between the saline and ethanol treatment groups during the treatment phase (Fig. 1a). SME analysis followed by planned comparisons indicated that the response to 2 g/kg ethanol was significantly enhanced in ethanol-treated mice on day 15, compared to that on day 4. The activity levels of the repeatedly saline-treated group were not significantly different on these days. On the MK-801 challenge day 16 (Fig. 1b), there was no effect of treatment on the response to MK-801. On day 17 (Fig. 1c), when mice were challenged with ethanol, the response to ethanol was lower than on day 15 in ethanol-sensitized mice (~11,000 cm in 20 min on day 15 vs ~9,000 cm on day 17). The acute ethanol response of the repeated saline group mice was also lower when tested on day 17 compared to the acute response of the repeated ethanol group on day 4 (~5,000 cm on day 17 vs ~7,000 cm on day 4). However, a significant effect of treatment (F [1, 24] = 11.9; p < 0.01) on day 17 indicated that the repeated ethanol group had an elevated response to 2 g/kg ethanol compared to the repeated saline group, verifying the persistence of ethanol sensitization after the MK-801 challenge. On day 18, when mice were challenged with ketamine, there were no differences between the repeatedly saline- and ethanol-treated groups (Fig. 1d). Thus, cross-sensitization to MK-801 or ketamine was not seen in ethanol-sensitized DBA/2J mice.

Experiment 2

Significant behavioral sensitization was also seen in HS mice. This was supported by a significant Treatment × Day interaction (F [4, 136] = 49.5; p < 0.01) followed by SME analyses and planned comparisons, which indicated that the response to ethanol was significantly enhanced on day 15, compared to that on day 4 (Fig. 2a). Consistent with experiment 1, there was no difference between the repeatedly saline- and ethanol-treated mice in response to MK-801. However, a significant effect of Treatment (F [1, 34] = 4.3; p < 0.05) on day 17 (Fig. 2c) supported the persistence of ethanol sensitization 1 day after the MK-801 challenge. Thus, neither ethanol-sensitized DBA/2J mice nor HS mice exhibited cross-sensitization to MK-801.

Experiment 3

Again, HS mice exhibited behavioral sensitization to ethanol. A significant Treatment × Day interaction (F [4, 136] = 56.6; p < 0.01) was associated with a significantly enhanced response to ethanol on day 15, compared to that on day 4 (Fig. 3a). On the challenge day 16 (Fig. 3b), a significant effect of Dose (F [1, 32] = 19.7; p < 0.01) reflected the dose-dependent effects of ketamine on locomotion. However, a significant effect of Treatment (F [1, 32] = 9.2; p < 0.01) was associated with a diminished response to both doses of ketamine in ethanol-treated mice. In addition, a significant Treatment × Time interaction (F [3, 96] = 7.4; p < 0.01) followed by SME analysis and planned comparisons indicated that this diminished response occurred primarily during the last three time points of the session. This suggests that repeated ethanol injections resulted in tolerance, not cross-sensitization, to the locomotor stimulant effects of ketamine. Challenge with ethanol on day 17 (Fig. 3c) showed persistence of behavioral sensitization to ethanol 1 day after the ketamine challenge. This was supported by a significant effect of Treatment (F [1, 32] = 4.6; p < 0.05), which was due to a larger response to ethanol in the ethanol-treated groups compared to the saline-treated mice on day 17. A Treatment × Time interaction (F [3, 102] = 13.4, p < 0.01) followed by SME analysis and planned comparisons indicated that this enhanced response occurred during the first two time points of the session. Again, there was no evidence for cross-sensitization to ketamine in ethanol-sensitized mice.

Fig. 3
figure 3

Response to ketamine in repeated ethanol- and saline-treated HS mice. a Distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates a significantly enhanced response to ethanol on day 15, compared to that on day 4 (p < 0.01). b Time course of the response to 10 and 20 mg/kg ketamine in repeated ethanol- and saline-treated mice on day 16. Double asterisks indicate a significantly reduced response to ketamine during the last three time points of the session in mice treated with ethanol during the treatment phase (p < 0.01). c The time course of the response to 2 g/kg ethanol on day 17. Asterisk indicates that repeated ethanol-treated mice had a larger response to ethanol (p < 0.05), compared to saline-treated mice at the first two time points. All values are mean ± SEM. Each treatment group contained 18 mice, which were further divided into groups of nine for the two doses of ketamine

Full size image

Experiment 4

A significant effect of Day (F [4, 140] = 123.2; p < 0.01) followed by planned comparisons indicated that as a group, HS mice exhibited significant locomotor sensitization to ethanol (Fig. 4a). Correlations of individual animal scores for acute ethanol response, magnitude of ethanol sensitization, response to ketamine, and response to MK-801 are shown in Table 2. Because the relationship between the measures of ethanol sensitization and the responses to MK-801 and ketamine were most relevant to the study hypotheses, only these relationships are plotted in Fig. 4. The acute response to ethanol was positively correlated with the responses to MK-801 and ketamine. Further, the responses to MK-801 and ketamine were significantly correlated. The acute response to ethanol was modestly but significantly negatively correlated with the first measure of ethanol sensitization (day 15 minus day 4) but not with the second measure of sensitization (day 17 minus day 4). However, the measures of ethanol sensitization showed a stronger relationship (Fig. 4b), indicating that individual sensitization scores were related on the 2 days and that ethanol sensitization was not disrupted by MK-801 treatment on day 16. There were no significant correlations between either measure of ethanol sensitization and the responses to either MK-801 (Fig. 4b) or ketamine (Fig. 4d). This suggests that sensitivity to MK-801 or to ketamine is not related to the magnitude of ethanol sensitization in HS mice.

Table 2 Correlations between the acute response to ethanol, ethanol sensitization and the locomotor responses to MK-801 and ketamine
Full size table
Fig. 4
figure 4

Correlations between ethanol sensitization, response to MK-801 and response to ketamine in HS mice. a Mean (±SEM) distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates a significantly enhanced response to ethanol on day 15, compared to that on day 4 (p < 0.01). b The magnitude of ethanol sensitization (day 15 minus day 4) is not associated with the response to 0.25 g/kg MK-801 (day 16 minus day 3). c The two measures of ethanol sensitization (day 15 minus day 4, day 17 minus day 4), were significantly correlated (r = 0.70; p < 0.01). d The magnitude of ethanol sensitization (day 15 minus day 4) is not associated with the response to 20 mg/kg ketamine (day 18 minus day 3). There were 36 mice in this experiment

Full size image

Experiment 5

Similar to the results for the locomotor activity studies, measurement of sensitivity to the convulsant effects of NMDA provided no support for an alteration in NMDA receptor systems in ethanol-sensitized mice. DBA/2J mice were sensitized to ethanol as indicated by a significant Treatment × Day interaction (F [4, 76] = 47.6; p < 0.01), which was associated with an enhanced response to ethanol on day 15, compared to that on day 4 (Fig. 5a). There was no difference between the repeated ethanol- and saline-treated groups in the dose of NMDA required to induce running/bouncing clonus or tonic hind limb extension, indicating that ethanol sensitization did not alter the response to NMDA (Fig. 5b).

Fig. 5
figure 5

Response to NMDA in repeated ethanol- and saline-treated DBA/2J mice. a Mean (±SEM) distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates a significantly enhanced response to ethanol on day 15, compared to that on day 4 (p < 0.01). b Mean (±SEM) cumulative dose of NMDA (mg/kg) required to induce running/bouncing clonus and tonic hind limb extension on day 16. Ethanol- and saline-treated groups contained 11 and 10 mice, respectively (seven were lost because of procedural errors on day 16)

Full size image

Experiment 6

Correlations between magnitude of ethanol sensitization, latency to running/bouncing clonus, and latency to tonic hind limb extension in HS mice are shown in Table 3 and Fig. 6. The two measures of NMDA sensitivity were strongly correlated with each other (r = 0.97, p < 0.01) but not with magnitude of ethanol sensitization. This suggests that the magnitude of ethanol sensitization is not associated with sensitivity to NMDA.

Table 3 Correlations between ethanol sensitization and the convulsant response to NMDA
Full size table
Fig. 6
figure 6

Correlations between ethanol sensitization and sensitivity to the convulsant effects of NMDA in repeated ethanol-treated HS mice. a Mean (±SEM) distance traveled over the first 20 min of the test session during the treatment days. Dagger indicates a significantly enhanced response to ethanol on day 15, compared to that on day 4 (p < 0.01). b On day 16, the two measures of NMDA sensitivity were highly correlated (r = 0.97; p < 0.01). c, d The magnitude of ethanol sensitization was not correlated with either measure of NMDA sensitivity on day 16. There were 33 mice in this experiment (three were lost because of procedural errors on day 16)

Full size image

Discussion

Ethanol is a noncompetitive antagonist at NMDA receptors (Lovinger et al. 1989), and this property may be important for ethanol-induced locomotion. In support of this, we found a positive association between the acute locomotor response to ethanol and the responses to MK-801 and ketamine (experiment 4). Previous data supporting a genetic correlation between sensitivity to MK-801, ketamine, and ethanol (Meyer and Phillips 2003b; Velardo et al. 1998) led us to speculate that changes in NMDA receptors may be one neuroadaptive change that occurs during ethanol sensitization. These studies found no evidence for cross-sensitization to MK-801, ketamine, or NMDA using two genotypes of mice and two experimental approaches. Instead, contrary to our hypotheses, repeated ethanol injections resulted in tolerance to the stimulant effects of ketamine in HS mice (experiment 3). In general, these studies suggest that ethanol sensitization does not result in increases in sensitivity to the doses of NMDA receptor ligands used in these studies.

Few studies have investigated cross-sensitization between ethanol and NMDA receptor ligands. However, one study (Quadros et al. 2002) reported that the magnitude of sensitization was positively associated with the magnitude of the acute stimulant response to 0.25 mg/kg MK-801 in male Swiss mice and that ethanol sensitization was negatively associated with radiolabeled NMDA receptors in certain brain areas. We did not observe this correlation in HS mice in the current study; there was no association between the magnitude of ethanol sensitization and the response to MK-801. The involvement of NMDA receptors in ethanol sensitization may depend on the specifics of these experiments, including the species of mice used (Swiss vs HS), the dose of ethanol used (2.4 vs 2 or 2.5 g/kg), and the length of treatment (21 vs 12 days).

Other studies have reported that MK-801 and several non-NMDA glutamate receptor antagonists blocked the expression of the sensitized response to ethanol in DBA/2J mice and Swiss mice (Broadbent et al. 2003; Kotlinska et al. 2006). In the current study, we did not find any evidence of cross-sensitization to NMDA receptor antagonists in DBA/2J or HS mice. The reduction in the behavioral expression of ethanol sensitization by MK-801 that has been reported may be due to the interaction of these drugs with the acute response to ethanol, as the NMDA antagonists that were effective in attenuating ethanol sensitization in those studies also interfered with the acute response to ethanol. Subsequent research has supported this conclusion (Meyer and Phillips 2003a).

Our negative results suggest that alterations in non-NMDA glutamate receptors or other neurotransmitter systems are involved in ethanol-induced behavioral sensitization. For example, metabotropic glutamate receptor antagonists decreased the expression of ethanol sensitization without affecting the acute response (Kotlinska et al. 2006). A study from our laboratory (Meyer et al. 2005) found that mice sensitized to ethanol did not display cross-sensitization to positive modulators of the γ-aminobutyric acid (GABAA) receptor, and a study by Broadbent et al. (2003) found that the development of ethanol sensitization was not affected by coadministration of a GABAA receptor agonist. Although pharmacological blockade studies such as that of Broadbent et al. (2003) examine the neural mechanisms involved in the development of ethanol sensitization, while cross-sensitization studies assess the mechanisms involved in the expression of ethanol sensitization, both studies came to the same conclusion. In addition, several studies have reported cross-sensitization between ethanol and cocaine, amphetamine, and morphine (Itzhak and Martin 1999; Lessov and Phillips 2003; Manley and Little 1997; Nestby et al. 1997), suggesting that dopamine and opioid systems may be altered during ethanol sensitization.

One drawback of these experiments is the limited range of doses tested. It is possible that ethanol sensitization caused altered sensitivity to other doses of MK-801 that were not tested. For example, it is possible that a ceiling effect, in which increased responses to 0.25 mg/kg MK-801 could not be observed because of the limitations of our measurements, may have precluded the observation of an increased response to MK-801. The dose used in these studies was chosen because it produces robust stimulation and because other researchers have demonstrated changes in sensitivity to this dose of MK-801 after repeated treatment with MK-801 itself and other drugs (Biala and Weglinska 2004; Iijima et al. 1996; Ito et al. 2006; Itzhak and Martin 1999; Vanderschuren et al. 1997; Xu and Domino 1999). Further, we partially addressed this issue by testing multiple doses of ketamine and still found no cross-sensitization. In fact, we obtained evidence for reduced sensitivity to ketamine in the repeated ethanol group mice, suggesting that a ceiling effect may not have been an issue for the MK-801 study. In addition, the tail-vein infusion procedure is a cumulative dose–response procedure that is sensitive to changes in NMDA’s seizure-inducing activity (Becker et al. 1998; Kosobud and Crabbe 1993). In these studies, no changes in sensitivity to NMDA receptor activation were observed.

A limitation of the current study may be that the changes in NMDA receptor-mediated processes are not detectable by these behavioral assays. For example, ethanol sensitization may alter the subunit composition of the NMDA receptor in a manner that affects their sensitivity to ethanol specifically and does not affect the behavioral pharmacology of NMDA antagonists or agonists. This could potentially be caused by an alteration of the ethanol modulatory site that occurs independently of MK-801-, ketamine-, and NMDA-binding sites. However, we did measure a significant reduction in locomotor response to ketamine in HS mice receiving repeated ethanol treatments compared to those receiving repeated saline injections (Fig. 3). The reduced sensitivity to ketamine in experiment 3 suggests that repeated ethanol decreases sensitivity to ketamine, which could be due to ethanol-induced changes in NMDA receptor number or function. However, because sensitivity to ketamine was not associated with the magnitude of ethanol-induced sensitization (experiment 4), such changes are unlikely to be key mechanisms that underlie sensitization to ethanol. In addition, repeated ethanol treatment may have caused changes in MK-801-induced ataxia, jumping, and rearing behavior that interfered with potential increases in forward locomotion. However, differences in MK-801-induced jumping and rearing behavior between repeated ethanol- and repeated saline-treated mice were not consistently found (measured by occlusion of sensors placed at 6 cm above the chamber floor; data not shown). While we did not measure ataxia in the current studies, previous studies in our laboratory have demonstrated that tolerance occurs to MK-801-induced ataxia, as well as MK-801-induced decreases in rearing (Meyer and Phillips 2003a). Therefore, increases in competing behaviors induced by MK-801 are unlikely explanations for the current findings.

Because NMDA receptor antagonists likely do not have effects on NMDA receptor activity, we conducted two experiments, one in DBA/2J and one in HS mice, measuring behavioral sensitivity to NMDA. NMDA receptor activation is required for NMDA-induced convulsant behavior. We hypothesized that this behavioral assay would be sensitive to changes in NMDA receptor function that occur during ethanol treatment. Repeated treatment with an NMDA receptor antagonist such as ethanol may lead to a compensatory upregulation in receptor function or density. As mentioned, Quadros et al. (2002) found such an upregulation, but it occurred only in mice that did not develop sensitization. Finn and Crabbe (1999) found that mice chronically exposed to ethanol vapor were more sensitive to the convulsant effects of NMDA, compared to unexposed animals. We did not find this to be true for mice sensitized to ethanol by repeated injection. However, a limitation of the current study is that running/bouncing clonus and tonic hind limb extension depend on hindbrain circuitry, while other seizure components (myclonal twitch and face and forelimb clonus) are dependent on forebrain circuitry (Gale 1988). Unfortunately, myoclonal twitch and face and forelimb clonus are not induced by NMDA infusions, so the possibility still remains that ethanol sensitization may cause changes in NMDA receptor function specific to forebrain circuitry. Forebrain areas that may influence the locomotor response to ethanol include areas of the mesolimbic dopamine system and the extended amygdala (Demarest et al. 1998; Phillips and Shen 1996).

While there was no relationship between ethanol sensitization and the response to NMDA receptor ligands in this study, we did observe a significant negative correlation between the acute response to ethanol and the magnitude of ethanol sensitization on day 15. This could be due to a ceiling effect; mice with very large acute responses to ethanol may be unable to increase their response to ethanol above a certain level. The correlations calculated here are phenotypic correlations, and the contribution of environmental and genetic variation to these correlations is unknown. However, we have found no genetic correlation between the acute and sensitized responses to ethanol using a panel of recombinant inbred strains (Phillips et al. 1995).

In summary, these results do not support a model of ethanol sensitization in which alterations in NMDA receptors result in cross-sensitization to the behavioral effects of other NMDA receptor ligands. To further rule out a role for NMDA receptor-mediated processes in the expression of ethanol sensitization, receptor alterations specific to the interactions of ethanol with the NMDA receptor during ethanol sensitization should be examined.