Introduction

Animal models of self-administration have a long, productive history in alcohol research. However, the amounts of alcohol actually self-administered are usually relatively modest. That is, voluntary oral consumption of ethanol (EtOH)-containing solutions rarely resulted in substantial blood EtOH concentrations (BECs) unless experimental manipulations such as weight reduction, schedule-induced polydipsia, post-prandial tests, or secondary conditioning procedures were undertaken (e.g., Falk 1961; Lester 1961; Freund 1969; Falk et al. 1972; Ogata et al. 1972; Middaugh and Kelley 1999; Mittleman et al. 2003). While prolonged access to EtOH, or periods of fluid or EtOH deprivations, or their combination have occasionally been shown to yield high levels of voluntary self-administration, in almost all cases, rodents do not self-administer alcohol to the degree that they will develop substantial, intoxicating, blood alcohol levels and then continue to self-administer at these levels. That is, animals experienced with alcohol self-administration and/or dependent on alcohol will increase their intake to a modest degree for a relatively short period of time after a period of alcohol deprivation (Hölter et al. 2000; Rodd-Hendricks et al. 2001; Rodd et al. 2003; Serra et al. 2003; Vengeliene et al. 2003) or during withdrawal (Schulteis et al. 1996; Roberts et al. 2000a; Becker et al., unpublished data), but the magnitude and architecture of the response does not convincingly display either gross excess or obvious loss of control.

Most of the studies examining self-administration of EtOH in rodents have used the oral route. One widely used method is two-bottle preference drinking, in which an animal is given a choice between a bottle containing a dilute EtOH solution (often 10% v/v in tap water) and a bottle containing tap water. Several studies have demonstrated that the selectively bred alcohol-preferring (P) rats (reviewed in Li et al. 1993; McBride and Li 1998) and high alcohol preference (HAP) mice (Grahame et al. 1999) as well as the C57BL/6 (B6) inbred strain (Lê et al. 1994; Middaugh et al. 1999, 2003; Finn et al. 2004) will orally consume substantial doses of alcohol during 1–2 h limited access sessions (i.e. BEC ≥50 mg%). Samson (1986) pioneered a sucrose fading operant self-administration procedure that produced reliable responding for substantial doses of alcohol in a 30-min session in rats (e.g. Samson 1986; Weiss et al. 1990) and mice (Elmer et al. 1987; Risinger et al. 1998; Middaugh and Kelley 1999; Roberts et al. 2000b). Notably, the BECs achieved with these procedures rarely approached 100 mg%, which appears to be necessary to guarantee visible signs of intoxication in most genotypes (Crabbe et al. 1994).

Early work by Belknap et al. (1978) indicated that restricting fluid intake to 90 min per day could provide the motivation for a mouse to consume an intoxicating dose of EtOH in a period of 10 min. With this procedure, B6 mice could consume a dose of EtOH as high as 2.5 g/kg during the first EtOH session without causing a decrease in consumption during the second EtOH session. However, consumption of higher EtOH doses (i.e. 3.2 g/kg) produced a reduction in EtOH intake during the second test. While mice were visibly intoxicated following consumption of both EtOH doses, only the 3.2 g/kg dose of EtOH appeared to produce a taste aversion.

Therefore, our goal was to determine whether the fluid restriction procedure could be optimized to yield stable and high EtOH intake in mice (i.e. EtOH dose ≥2 g/kg in 30 min; BECs ≥100 mg%). That is, could conditions be adjusted so that mice would consume a high EtOH dose without engendering a taste aversion in subsequent EtOH sessions? First, we used the procedure described in Belknap et al. (1978) to determine whether B6 mice could overcome their putative aversion for the 3.2 g/kg EtOH dose with additional pairings of EtOH and water and to test genetically heterogeneous mice (i.e. WSC). The rationale for testing genetically heterogeneous mice was to determine if this procedure could produce high EtOH intake in a strain that does not normally exhibit EtOH preference. Second, we characterized the fluid restriction procedure in male and female mice from both genotypes that were on a schedule of 3 h fluid/day. Finally, we determined whether decreasing the period of overall fluid restriction to the point where animals were no longer fluid restricted but rather, drinking on a schedule, could maintain consistent high consumption of EtOH.

Materials and methods

Subjects

Studies were conducted in drug-naive male and female mice. B6 mice (Jackson Laboratory, Bar Harbor, Maine, USA) were purchased at 6 weeks of age, while WSC mice were bred at the Portland VA Medical Center Veterinary Medical Unit. For a minimum of 1 week, the animals were separated by sex, housed four to a cage and allowed to acclimate to the temperature- and humidity-controlled environment and to a 12-h light/dark cycle (lights on 0600 hours) at 26±1°C. In all studies, mice had ad libitum access to food, but the period of fluid access varied with the experiment (details below and in Table 1). Body weights were measured every day. All procedures were carried out in accordance with the United States Public Health Service National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and the principles of laboratory animal care (http://www.nap.edu/reading room/books/labrats/), and were approved by the local Institutional Animal Care and Use Committee.

Table 1 Experimental procedures
Full size table

General procedures

Unless noted, mice had access to one tube that contained either an EtOH solution or water during the period of fluid access. Mice were approximately 8 weeks old when they were individually housed in standard polycarbonate shoebox cages with ground corncob bedding. Fluids were presented in one 25 ml graduated cylinder placed on the stainless steel cage top. For the 2 days following individual housing, tap water was available ad libitum in one tube. At 5 p.m. on the second day of individual housing, the water tube was removed, which began the period of fluid restriction. On each subsequent day, mice had access to the water tube for a designated period of time beginning at 8–9 a.m., which varied depending on the experiment (Table 1). Every 3rd or 4th day (depending on the study), the mice initially had access to a tube containing EtOH (5, 7, or 10% v/v solution in tap water; Pharmco Products, Brookfield, Conn., USA) for either 10 min or 30 min at 8–9 a.m., followed by their designated access to water to complete the access period. Depending on the experiment, mice received a minimum of four sessions with the EtOH solution. In some studies, a retro-orbital or tail blood sample (20 μl) was taken either immediately or at 30 min following the EtOH session on day 12 and analyzed by gas chromatography for BEC (Roach and Creaven 1968). Daily fluid consumption was measured each day by recording the level of the meniscus on the graduations of the drinking tube. EtOH and water tubes placed on two empty cages allowed for the measurement of leakage and evaporation from the tubes. Average volume depleted from these tubes was subtracted from the individual drinking volumes each day.

Data analysis

The dependent measures were BEC, volume of EtOH or water consumed, EtOH dose (g/kg) consumed, and body weight. In one experiment, two bottles were offered; one bottle contained an EtOH solution and one bottle contained tap water. Here, an additional variable, preference ratio was calculated as the volume of EtOH consumed (ml) divided by the total volume (ml, EtOH+water) consumed. Depending on the experiment, the effect of the period of fluid access, the concentration of the EtOH solution, and the sex and genotype of the animal on these dependent measures was analyzed by ANOVA with day as a repeated measures factor. When appropriate, simple main effects analysis, followed by post hoc comparisons were conducted. Correlation analyses assessed the relationship between EtOH dose and BEC.

Experiment 1: effect of daily fluid access of 1.5 h per day on limited access EtOH consumption in female B6 and WSC mice

The goal of this study was to extend the findings of Belknap et al. (1978) in B6 mice and to test genetically heterogeneous mice. Mice were treated as described above, with the exception that the animals received the EtOH solution every 4th day, and had retro-orbital blood samples taken following each EtOH session. All groups had 90 min access to water on days 1–3. On day 4, the WSC mice had access to a 7% EtOH solution (n=10) for 10 min, while separate groups of B6 mice had access to a single bottle containing either a 7% or 10% EtOH solution (n=10/group). A retro-orbital blood sample was taken at 40 min following access to the EtOH solution (i.e. 30 min after the termination of the EtOH access), and then the mice had 80 min of access to the water tube. Since we had noticed that the animals were visibly intoxicated after the initial EtOH session, the water tube was available for an additional 2 h to ensure that the animals drank enough fluid to remain hydrated. This regimen was repeated a total of 4 times.

Experiment 2: effect of daily fluid access of 3 h per day on limited access EtOH consumption in male and female B6 and WSC mice

The rationale for this study was to offer the EtOH solution every 3rd day and to have the EtOH solution available for 30 min in male and female B6 and WSC mice. The results of experiment 1 suggested that the pattern of drinking differed when mice were offered the EtOH solution on the first versus fourth session. Since additional studies determined that EtOH intake was more stable across repeated trials of EtOH access when the EtOH session was increased from 10 min to 30 min (data not shown), the period of EtOH availability was increased to 30 min. Separate groups of male (n=10/group) and female (n=9–10/group) B6 mice had access to either a 5% or 10% EtOH solution, while male (n=8) and female (n=10) WSC mice had access to the 5% EtOH solution. Mice were treated as described under General procedures. Every 3rd day, mice had 30 min access to an EtOH solution (5% or 10%), with an orbital blood sample taken immediately after the fourth EtOH session on day 12 to assess BEC.

Experiment 3: effect of daily fluid access of 3 h per day on EtOH preference in female WSC mice

The purpose of this experiment was to determine whether the procedure utilized in experiment 2 would alter EtOH preference in WSC mice. Since B6 mice typically exhibit high EtOH preference, they were not tested. Three independent groups of female WSC mice were tested: group 1 (scheduled fluid access+single preference test, n=15) was treated identically to that in experiment 2, with the exception that a fifth EtOH session was added. However, the fifth EtOH session was a two-bottle (i.e. EtOH versus water), rather than a single bottle test. Group 2 (scheduled fluid access+multiple preference tests, n=15) had 3 h fluid per day, but each EtOH session was a two-bottle test. Group 3 (continuous fluid access+preference tests, n=11) had unlimited access to a bottle containing a 5% EtOH solution versus a bottle containing tap water for 15 days. These animals were not fluid deprived and served as the “control” group to demonstrate the preference that this genotype exhibited for a 5% EtOH solution. A retro-orbital blood sample was taken on day 15 from all mice, immediately following the fifth EtOH session for groups 1 and 2 and at a comparable time for group 3.

Experiment 4: effect of increasing fluid availability on EtOH consumption in male and female B6 and WSC mice

The purpose of this experiment was to determine whether increasing fluid availability so that animals were no longer considered “fluid restricted” could maintain stable high EtOH intake. Animals had access to a 5% EtOH solution for 30 min every third day. Initially, all groups of mice had access to 4 h fluid per day for 9 days (n=10/group). Beginning on day 10, the period of fluid access was increased by 2 h, and then further increased by 2 h after each subsequent EtOH session, until fluid availability was 10 h per day (i.e. 6 h fluid/day for days 10–12, 8 h fluid/day for days 13–15, and 10 h fluid/day for days 16–21).

Results

Experiment 1: effect of daily fluid access of 1.5 h per day on limited access EtOH consumption in female B6 and WSC mice

With the first exposure to the EtOH solution on day 4, the average EtOH dose consumed in 10 min ranged from 2.6 g/kg to 4.0 g/kg, depending on the genotype and concentration of the EtOH solution (Fig. 1a). Mean BEC ranged from 1.7 mg/ml to 2.6 mg/ml (Fig. 1b), and individual BECs ranged from 1.5 mg/ml to 3.4 mg/ml (i.e. 150–340 mg%; Fig. 1c). The majority of BECs were above 2 mg/ml, with the mice exhibiting visible signs of intoxication (e.g., listing from side to side while walking). On the second exposure to EtOH, consumption was significantly reduced in all groups. However, consumption increased in some animals with the third exposure to EtOH, suggesting that animals might overcome their putative taste aversion for the consumption of a high EtOH dose with repeated pairings. For instance, during the fourth EtOH session, the B6 mice drinking the 10% EtOH solution consumed an average dose of 2.2 g/kg, achieving an average BEC of 1.35 mg/ml.

Fig. 1
figure 1

Effect of daily fluid access of 1.5 h per day on a EtOH dose consumed, b BEC, and c the correlation between BEC and EtOH dose consumed in female B6 and WSC mice. Mice had 10 min access to either a 7% or 10% EtOH solution every 4th day. Values represent the mean±SEM for 9–10 (g/kg) or 7–9 (BEC) per group. Note the difference in y-axes. Data in c are collapsed across all genotypes/EtOH concentration groups, and the best-fit regression line is shown separately for each day

Full size image

The findings stated above are supported by the following analyses. There was a significant main effect of genotype/EtOH concentration [F(2,28)=7.93, P<0.005] and time [F(3,87)=25.06, P<0.001] on EtOH dose consumed, although the interaction was not significant. Similar results were found for the analyses conducted on BEC. In addition to the main effect of genotype/EtOH concentration [F(2,23)=12.88, P<0.005] and time [F(3,72)=50.22, P<0.001], there was a trend for an interaction between main effects.

BEC was significantly positively correlated with the EtOH dose consumed for each EtOH session (Fig. 1c): day 4 (r=0.89, n=30, P<0.001), day 8 (r=0.72, n=25, P<0.001), day 12 (r=0.89, n=28, P<0.001) and day 16 (r=0.62, n=29, P<0.005).

The 1st day that the animals were on the 90-min period of fluid access, mean water consumption ranged from 1.4 ml to 1.6 ml. From day 2 onward, all groups of mice drank an average of 2.2–3.5 ml water during the period of fluid access. In our experience, a mouse that is not fluid deprived will drink approximately 5 ml water during a 24-h period. Thus, mice were drinking 44–70% of the total amount of fluid that they would normally consume over 24 h during the 90-min period of fluid access. Importantly, the mice were able to maintain their body weight. At the beginning of the experiment, mean±SEM body weight for each group of mice was 19.1±0.3 g (WSC, 7%), 20.4±0.7 g (B6, 7%), and 19.6±0.5 g (B6, 10%). These values did not differ significantly. Body weight dropped approximately 10% during the first 4 days of the experiment, and recovered to baseline values within the next 4 days (not shown).

Experiment 2: effect of daily fluid access of 3 h per day on limited access EtOH consumption in male and female B6 and WSC mice

With the first exposure to the EtOH solution, the average dose of EtOH consumed exceeded 2 g/kg in five of the six groups of mice (Fig. 2a). In most cases, the mice continued to drink EtOH doses that exceeded 2 g/kg during each subsequent 30 min EtOH session. Notably, during the third and fourth periods of EtOH access, the average EtOH dose consumed exceeded 2 g/kg in all groups of mice. There was a main effect of genotype/sex/EtOH concentration [F(5,51)=3.168, P<0.02] and time [F(3,153)=4.199, P<0.01] on the dose of EtOH consumed. However, the interaction between main effects was not significant. Overall, these results suggest that adjusting the concentration of the EtOH solution and period of EtOH access produced high stable EtOH intake in both B6 and WSC mice, regardless of whether females or males were studied.

Fig. 2
figure 2

Scheduling fluid availability to 3 h per day produced consistent high a EtOH consumption and b BEC in male and female B6 and WSC mice. Our definition of high intake in the mouse was consumption of a 2 g/kg dose of EtOH in 30 min (see dashed line on a) and resultant BEC ≥100 mg%. Notably, in this and subsequent experiments, there was no lasting taste aversion on subsequent EtOH sessions, as was seen in the first experiment (see Fig. 1). Mice had 30-min access to either a 5% or 10% EtOH solution every 3rd day. BEC was assessed after the EtOH session on day 12. Values represent the mean±SEM for 9–10 (g/kg) or 8–10 (BEC) per group, with the exception of the WSC males where the n=8 for the g/kg data and n=7 for the BEC data. Closed symbols and dashed lines portray data in female mice, while open symbols and solid lines depict data in male mice. Note the difference in y-axes

Full size image

The dose of EtOH consumed on day 12 produced average BECs that exceeded 1 mg/ml in all groups (Fig. 2b) and that did not differ significantly. Individual BECs ranged from 0.6 mg/ml to 2.34 mg/ml (not shown). BEC was significantly positively correlated with the dose of EtOH consumed only in the WSC mice (r=0.51, n=17, P=0.03; not shown). Individual values in the B6 mice consuming the 10% EtOH solution were much more variable than in the B6 mice consuming the 5% EtOH solution (not shown), suggesting that EtOH consumption was more consistent in mice consuming a 5% EtOH solution.

The first day that the animals were on the 3-h period of fluid access, mean water consumption ranged from 1.5 ml to 2.3 ml. From day 2 onward, all groups of B6 mice drank an average of 2.2–3.2 ml water during the period of fluid access, while the WSC mice drank an average of 3.0–4.3 ml of water. These values suggest that B6 mice were consuming 44–64%, and WSC mice were consuming 60–86%, of the total amount of fluid that they would normally consume over 24 h during the 3-h period of fluid access. As in experiment 1, the mice were able to maintain their body weight. At the beginning of the experiment, mean±SEM body weight for each group of mice was 21.7±0.3 g (B6 male, 5%), 21.8±0.3 g (B6 male, 10%), 18.0±0.4 g (B6 female, 5%), 17.5±0.4 g (B6 female, 10%), 27.8±0.8 g (WSC male, 5%) and 24.6±0.8 g (WSC female, 5%). Baseline body weight was significantly higher in male than in female mice [F(1,53)=59.62, P<0.001] and in WSC than in B6 mice [F(1,53)=197.66, P<0.001]. Body weight dropped between 10% and 12% during the first 4 days of the experiment, and recovered to baseline values within the next 4 days (not shown).

Experiment 3: effect of daily fluid access of 3 h day on EtOH preference ratio in female WSC mice

EtOH intake and preference ratio were very low in the WSC animals with unlimited access to a bottle containing a 5% EtOH solution versus a water bottle (i.e. group 3). Initially, the average dose of EtOH consumed by these animals was 1.4 g/kg, but EtOH intake fluctuated from 0.4 g/kg to 0.7 g/kg throughout most of the experiment (Fig. 3a). Preference ratio ranged from 0.02 to 0.05 (Fig. 3b), indicating that the continuous fluid access group were drinking only 2–5% of their total fluid intake from the EtOH bottle. Consistent with the low EtOH intake, the BECs were negligible when assessed upon completion of the study on day 15 (Fig. 4a).

Fig. 3
figure 3

Scheduling fluid availability to 3 h per day in WSC mice. The a EtOH dose and b preference ratio was significantly higher in groups 1 and 2 than in group 3. Group designations are as follows: group 1 (scheduled fluid access+single preference test), group 2 (scheduled fluid access+multiple preference tests), and group 3 (continuous fluid access+preference tests). Values represent the mean±SEM for 15 (groups 1 and 2) or 11 (group 3) female WSC mice per group. Note the difference in y-axes and that group 1 only had a preference test on day 15. +P=0.10, *P<0.05 versus respective value in group 3, Tukey’s post hoc test

Full size image
Fig. 4
figure 4

a The effect of scheduled fluid access (groups 1 and 2) versus continuous fluid access (group 3) on BEC on day 15 and b the correlation between BEC and EtOH dose consumed in female WSC mice. In a, the mean±SEM is shown for the mice depicted in Fig. 3, whereas the individual values are shown in b. Notably, EtOH intake (g/kg) exceeded 1.5 g/kg in 20% of the mice during the first two-bottle test (group 1), while EtOH dose exceeded 1.5 g/kg in 33% of the mice during the fifth two-bottle test (group 2)

Full size image

The scheduled fluid access+multiple preference test group (group 2), which had a choice between a 5% EtOH bottle and a water bottle for each EtOH session, consumed doses of EtOH in 30 min that were not significantly different from that consumed by group 3 over 24 h. EtOH intake ranged from 0.45 g/kg to 0.85 g/kg across the five EtOH sessions (Fig. 3a), with preference ratio ranging from 0.10 to 0.28 (Fig. 3b). However, even though the group 2 animals were not exhibiting preference for the EtOH bottle (i.e. preference ratio >0.5), the preference ratio in this scheduled fluid access group was significantly higher than in the continuous fluid access group (group 3). This conclusion is supported by the significant main effect of group [F(1,24)=61.92, P<0.001] and group by time interaction [F(4,96)=2.91, P=0.026] on preference ratio.

The scheduled fluid access+single preference test group (group 1), which had a choice between the EtOH and water bottle only for the fifth EtOH session (i.e. all other EtOH sessions were single bottle, 5% EtOH), consumed high EtOH doses when they were offered a single EtOH bottle during the EtOH session (Fig. 3a, and see Fig. 2a, WSC female, for comparison). When the animals had a choice between the EtOH and water bottle, EtOH intake dropped significantly and was comparable to that of group 2 (schedule fluid access+multiple preference test). The EtOH dose consumed was significantly influenced by group [F(2,38)=105.02, P<0.001] and time [F(4,152)=6.78, P< 0.001], and there was a significant interaction between main effects [F(8,152)=8.47, P<0.001]. Preference ratio on day 15 differed significantly among groups [F(2,38)=3.39, P<0.05]. Post hoc analyses indicated that day 15 preference ratio in group 1 was comparable to that of group 2 (Fig. 3b) and tended to be higher than that of group 3 (continuous fluid access). Day 15 preference ratio in group 2 was significantly higher than that of group 3 (P<0.05).

Mean BEC in the scheduled fluid access groups (groups 1 and 2) on day 15 were 0.26 mg/ml and 0.22 mg/ml, respectively (Fig. 4a); these values did not differ significantly. Notably, the EtOH dose in 20% of the group 1 mice exceeded 1.5 g/kg, resulting in a BEC of ≥1 mg/ml (Fig. 4b). Likewise, the EtOH dose in 33% of the group 2 mice exceeded 1.5 g/kg, but the resultant BEC in these animals was a bit more variable and ranged from 0.26 mg/ml to 1.0 mg/ml. BEC was significantly positively correlated with the EtOH dose consumed in both group 1 (r=0.975, n=15, P<0.001) and group 2 (r=0.88, n=15, P<0.001) mice, although both these correlations depend substantially on the high drinking of 20–33% of the mice.

Total fluid consumption in the continuous fluid access group (group 3) ranged from 6.3 ml to 7.0 ml per day. The first 2 days that the scheduled fluid access groups (groups 1 and 2) were on the 3 h period of fluid access, the mean water intake was 2.5–2.7 ml. From day 3 onward, the average fluid intake was 3.0–4.9 ml, representing 48–78% of the fluid consumed by the continuous fluid access group (group 3). As with the earlier studies, groups 1 and 2 were able to maintain their body weights. At the beginning of the experiment, mean±SEM body weights were 22.6±0.6 g (group 1), 22.8±0.5 g (group 2), and 21.4±0.5 g (group 3). These values did not differ significantly. For groups 1 and 2 (scheduled fluid access), body weight dropped 14–15% during the first few days of the experiment and returned to baseline within the next 5 days (not shown). Even though group 3 had unlimited access to fluid, these animals also were individually housed, so body weight in this group increased only 4% across time.

Experiment 4: effect of increasing fluid availability on EtOH consumption in male and female B6 and WSC mice

With the first exposure to the EtOH solution, the average EtOH dose consumed was approximately 2 g/kg. During each subsequent 30-min EtOH session, EtOH intake was >2 g/kg in B6 mice and ∼2 g/kg in WSC mice (Fig. 5). The fact that EtOH intake remained the same or increased during the second EtOH session after 10 h of fluid access provides evidence for the stability of this high EtOH consumption. EtOH intake was significantly higher in B6 than in WSC mice [F(1,36)=19.22, P<0.001] and did fluctuate significantly over time [F(6,216)=3.59, P<0.005]. The interactions between time and sex [F(6,216)=2.49, P<0.05] and between time and strain [F(6,216)=6.15, P<0.001] also were significant. These interactions were likely due to the fact that there was an overall increase in EtOH intake in B6 mice across EtOH sessions, while intake remained at approximately 2 g/kg in WSC mice. Although BEC was not assessed in this study, a separate study determined that consumption of a comparable EtOH dose on day 12 produced average BECs that ranged from 1.3 mg/ml to 1.5 mg/ml (not shown).

Fig. 5
figure 5

Increasing fluid availability to 10 h per day produced consistent high EtOH intake in male and female B6 and WSC mice. Initially, all groups of mice had access to 4 h fluid per day for 9 days (n=10/group). Beginning on day 10, the period of fluid access was increased by 2 h, and then further increased by 2 h after each subsequent EtOH session, until fluid availability was 10 h per day (i.e. 6 h fluid/day for days 10–12, 8 h fluid/day for days 13–15, and 10 h fluid/day for days 16–21). Values represent the mean±SEM for ten mice per group. Closed symbols and dashed lines portray data in female mice, while open symbols and solid lines depict data in male mice

Full size image

Water intake initially ranged from 2.5 ml to 3.5 ml during the 4-h period of fluid access and increased as the period of fluid access increased so that intake ranged from 4.5 ml to 5.3 ml during the 10 h periods of fluid access. In this study, mice also were able to maintain their body weight. The mean±SEM baseline body weights for each group were 23.0±0.4 g (B6 male), 17.3±0.2 g (B6 female), 27.0±0.4 g (WSC male) and 21.3±0.5 g (WSC female). These initial body weights were significantly higher in WSC than in B6 mice [F(1,36)=104.92, P<0.001] and were significantly higher in male than in female mice [F(1,36)=217.9, P<0.001]. While body weight dropped 11–13% during the first 2 days of the experiment, weights recovered to baseline values within the next 7 days of the experiment (not shown).

Discussion

The present experiments examined the effect of scheduling periods of fluid availability on EtOH intake in an inbred strain that naturally exhibits preference for an EtOH solution (i.e. B6 mice) under two-bottle choice conditions and in genetically heterogeneous mice (i.e. WSC mice). By manipulating the concentration of the EtOH solution, the period of EtOH access, as well as the total fluid availability, most animals consumed an EtOH dose that was ≥2 g/kg in a 30-min EtOH session, and which produced BECs ≥100 mg% (i.e. our definition of high EtOH intake in the mouse). Importantly, increasing fluid availability per day, so that animals were conditioned to drink on a schedule rather than being fluid restricted, did not significantly reduce EtOH intake. Also, consumption of these high EtOH doses did not produce a taste aversion during subsequent EtOH sessions. Thus, the present procedure of scheduling periods of fluid availability per day produced high, consistent EtOH intake in both B6 and WSC mice.

Since existing animal models cannot model an entire disorder as complex as alcoholism, a useful strategy is to develop many partial models, each of which attempts to mimic a subset of the disorder (e.g. McClearn 1979; Falk and Tang 1988; Cunningham et al. 2000; Crabbe 2002). Our focus in the present studies has been on an aspect of alcoholism that has been difficult to model in rodents, namely whether an animal will consume alcohol uncontrollably. B6 mice will voluntarily consume a 2 g/kg dose of EtOH during daily 2 h limited access sessions under a two-bottle choice paradigm (i.e. 10% EtOH bottle versus water bottle), when animals were not fluid deprived and drinking was measured during the dark phase of the circadian cycle or just prior to lights off. However, the BECs achieved after the 2 h limited access drinking session ranged from 50 mg% to 60 mg% (Lê et al. 1994; Middaugh et al. 2003; Finn et al. 2004), concentrations that were substantial but probably did not produce visible signs of intoxication. Likewise, mice that were selectively bred for high alcohol preference during continuous access two-bottle preference tests with 10% EtOH versus water (HAP mice) also voluntarily consumed an average EtOH dose of 1.4 g/kg in 30 min (Grahame and Grose 2003) or 3 g/kg in 2 h limited access sessions (Grahame et al. 1999), when drinking was measured during the dark phase. BECs in the HAP mice averaged 60 mg% after the 30-min EtOH session, but were lower and more variable after the 2 h EtOH session. The increased variability in BEC following the 2-h limited access EtOH session could be due to the pattern of drinking, since mice consuming the majority of their EtOH dose during the first 30 min of the 2-h session would have more time to metabolize the alcohol prior to the assessment of BEC. Early work in P rats, which found that the variability in BEC increased as a function of the time following alcohol consumption (Murphy et al. 1986), is consistent with this assumption. Thus, rodents exhibiting high EtOH preference, either innately or because their genotype has been manipulated via selective breeding, can consume high EtOH doses during the dark phase of the circadian rhythm, but the resultant BEC is rarely one that would produce visible signs of intoxication.

In contrast, the present findings indicate that mice consistently drank high EtOH doses in a 30-min session so that the resulting BEC was ≥100 mg%, and which was found in separate studies to produce visible signs of ataxia, measured by decreased performance on the rotarod, screen test and balance beam (Cronise et al., unpublished data). Overall, the present findings suggest that animals will consume high, intoxicating doses of EtOH each time that they are offered the EtOH solution, when periods of fluid availability have been scheduled.

The physiological and behavioral consequences of deprivation versus restriction protocols have recently been reviewed (Toth and Gardiner 2000). Whereas the term “deprivation” implies total denial of food or water and generally is described in terms of the interval during which food or water is withheld from the animal, the term “restriction” implies a limitation on ad libitum intake, as opposed to total denial, and typically is described in terms of either the amount of food or water provided daily or the amount of time that the animal has access to food or water. While different species vary in how much and how fast they drink after deprivation, most of the water consumption that occurs during the first hour of access actually takes place during the initial 10–30 min (e.g. Rolls et al. 1980). Importantly, this amount of drinking reverses the hematologic changes that develop with water deprivation; for example, plasma osmolality and protein values are high after 24 h of water deprivation, but return to normal levels within about 20 min of drinking (Rolls et al. 1980).

With regard to “restriction” schedules of fluid availability, daily clinical examination and comprehensive postmortem evaluation revealed no abnormalities in rats that were chronically maintained on a 21-h restriction schedule (Hughes et al. 1994). Similarly, results from our laboratory indicate that control groups of mice maintained on a schedule of 3-h water/day did not exhibit any change in motor performance throughout the restriction schedule when compared with their baseline performance, measured by the rotarod, balance beam and screen test (Cronise et al., unpublished data). While limited availability of water is a common homeostatic challenge, animals are well adapted physiologically to accommodate this situation in that homeostatic mechanisms automatically regulate their urine output in accordance with their current state of hydration. Importantly, chronic water restriction is generally not associated with marked adrenal activation (Johnson and Levine 1973; Gray et al. 1978; Heiderstadt et al. 2000), since glucocorticoid elevations during dehydration might cause diuretic effects that would counter fluid conservation mechanisms. Collectively, the evidence suggests that the mice undergoing various periods of fluid restriction in the present studies would have adapted to the restriction schedule and would not be exhibiting any detrimental physiological or behavioral effects.

Since fluid intake can show characteristics of a rhythm entrained with feeding (Toth and Gardiner 2000), the fluid restriction utilized in the present studies could cause food intake reductions despite the ad libitum presence of food. The daily body weight data suggest that there might have been a transient decrease in food intake, based on the initial drop in body weight at the beginning of each experiment. Notably, body weight approached baseline fairly rapidly, and animals were able to maintain baseline body weight throughout the remainder of each experiment.

Certain research protocols commonly employ food or water restriction to motivate the animal to learn or perform a task that is totally unrelated to the deprivation, with the challenge being to achieve a balance between the severity of the imposed restriction schedule and the need to motivate learning or performance (Toth and Gardiner 2000). An important part of the training is teaching the animal that access to the reward is context dependent: either work is required for all access, or access is available for only a limited period of time each day. Early work and recent studies indicate that motivational states related to hunger and thirst could influence ethanol consumption of rodents (Dole and Gentry 1984; Elmer et al. 1986, 1987; Samson et al. 1988; Middaugh and Kelley 1999; Middaugh et al. 1999). While we cannot know the animal’s motivation to consume EtOH in the present studies, they easily learned that access to fluid was available for a specific period of time each day and consumed intoxicating doses of EtOH during each period of EtOH access. Although absolute preference for EtOH over water was not observed with a two-bottle test in the groups of WSC mice with scheduled fluid access, EtOH intake and BEC was high in 20–33% of these animals, when they had a choice between an EtOH and water bottle (see Fig. 4b). Thus, scheduling periods of fluid availability produced high EtOH intake in all mice during a single bottle test and in a subset of the animals during a two-bottle test.

In summary, animals easily learn to live under scheduled access to food or water, and the present procedure of scheduling periods of fluid availability per day produced high, stable EtOH intake in an inbred strain that exhibits EtOH preference (i.e. B6) and in genetically heterogeneous mice that do not exhibit EtOH preference (i.e. WSC). EtOH intake was high in all mice during a single bottle test and remained high in a subset of the animals during a two-bottle test. These high EtOH doses were shown in additional studies (Cronise et al., unpublished data) to produce visible signs of ataxia, indicating that the mice were consuming intoxicating doses of EtOH during each period of EtOH access. Additionally, high, stable EtOH intake was maintained as the fluid availability was increased to the point where animals were not considered “fluid restricted.” Correlational analyses in WSC mice from the final study also suggest that mice with high EtOH intake initially may transition to stable high intake, since the EtOH dose consumed on day 3 was significantly positively correlated with EtOH intake on day 21 (r=0.55, n=20, P=0.01). While this procedure may not distinguish between high versus low alcohol preferring genotypes, since the parameters have been adjusted so that non-preferring genotypes will drink high doses of EtOH, it has the potential to provide important information on mechanisms underlying high EtOH intake.