Animals and housing
Female Sprague-Dawley rats obtained from Harlan (Indianapolis, IN) initially weighing 253±3 g were housed individually in polycarbonate cages with Sani Chips (Harlan) bedding. They were not monitored for estrous cyclicity. They were housed in a vivarium with a reverse 12:12 light dark cycle (lights off @ 0800) and an ambient temperature maintained at 23±2 °C. Purina 5001 Rodent Chow and water was available ad libitum at all times. Rats were randomly divided into two groups for use in “jello shot” self-administration or gavage experiments and weighed at least once a week. The experimental protocol was approved by the University of Florida Institutional Animal Care and Use Committee and the procedures were in compliance with the National Institute of Health Guide for Care and Use of Laboratory Animals (Publication No. 85-23, revised 1985).
The “jello shots” were made from 10% ethanol (w/w), 10% Polycose® (Abbott Laboratories, Abbott Park, IL) and 0.25% gelatin (Knox brand, Kraft Foods, Northfield, IL) in water, and solidified overnight in a refrigerator in small glass jars. Under refrigerated and sealed conditions, ethanol levels in the gelatin remained stable at 10% for at least one week. Additionally, we determined that ethanol content of the jars did not decrease significantly when left open at room temperature for up to three hours. After 24 hrs, opened at room temperature, the top layer of gel would lose approximately 30% of ethanol content.
In the first group of rats, the jars were suspended in holders from the sides of the cage (see ) for 24 hrs for the first two days of exposure, then for 6 hrs for two days, then 3 hrs for two days and finally for 1 hr for about 2 weeks (see ). This one-hour free access period occurred between 10 AM and noon each day, including weekends. The jars were weighed prior to and after the allotted period of time in the cage and the difference was calculated. Cages and jars were inspected for evidence of spillage or contamination but evidence for this was negligible. To estimate individual ethanol doses, g of “jello shot” consumed was transformed into g of ethanol/kg body weight.
A graphic representation of the “jello shot” procedure.
Figure 2 Rats received free access to 10% ethanol in Polycose gelatin for 24 hrs on the first and second days, 6 hrs on the third and fourth days, 3 hrs on the fifth and sixth days and 1 hour on all subsequent days. Stable levels of self-administration were achieved (more ...)
At least one week prior to surgery to implant guide cannulas, each rat was accustomed to eating the “jello shot” under microdialysis conditions (designated rl1, rl2, rl3 etc in ). This consisted of removing the rat’s cage from the housing rack and placing it on a table in the vivarium next to two other rat-occupied cages and with a red light shining on it. An experimenter was seated nearby and tapped lightly on the cage every 10 minutes to accustom the rats to periodic mild disturbances. After 2 hrs, animals were given 1-hr free access to the “jello shot”. They were then returned to their regular housing rack. This adaptation procedure occurred daily until surgery. Rats did not receive “jello shot” access on the day of surgery. After surgery, each rat was allowed to recover for 2 additional days before daily ethanol access was restarted under microdialysis conditions as described above except that rats were also tethered to a swivel in the top of the cage by a spring attached to a clip embedded in the dental cement on their skulls. Habituation to the tethering procedure continued for at least five daily sessions (indicated by ps1, ps2 etc in ). The dialysis experiment occurred 7–11 days after surgery only when levels of consumption had stabilized.
Figure 3 Mean ethanol consumption (on each of nine days) after environmental and surgical disruption. “Home” indicates mean consumption for the 2 days preceding the first day of ethanol access in the microdialysis cage. “rl1, rl2, rl3” (more ...)
A second group of female rats (ethanol-naïve) had no prior exposure to ethanol or habituation to the test procedure although they were weighed weekly and handled on a daily basis.
Surgery and Microdialysis
On the day of surgery, each rat was anesthetized with ketamine/xylazine (87 and 13 mg/kg respectively) and placed in a stereotaxic apparatus for surgical implantation of a unilateral guide cannula (21 gauge 8mm long; Plastics One, Roanoke, VA). Cannulae were implanted to end within 1 mm above the NAc using the following flat skull coordinates from bregma: +1.6 anterior, +1.7 lateral, −6.2 ventral. The guide cannula was secured to the skull with dental cement anchored by two stainless steel screws.
Microdialysis probes (o.d. 270 μm; active length 2 mm; cellulose membrane, 13,000 MWCO) were constructed by the method of Pettit and Justice (1991)
. The probe was connected to a single channel swivel (Instech, Plymouth Meeting, PA) and perfused with artificial cerebrospinal fluid (aCSF; 145 mM NaCl, 2.8 mM KCl, 1.2 mM MgCl2
, 1.2 mM CaCl2
, 1.55 mM Na2
, 0.45 mM NaH2
, pH 7.4) at 1 μl/min (Harvard Apparatus, South Nattick MA). Recovery of ethanol by each dialysis probe was determined at the start of each experiment, before probe implantation, by immersing the probe in 10 mM ethanol in aCSF (see below). After this, the dialysis probe was inserted into the guide cannula while the rat was briefly sedated with halothane.
In the rats trained to self-administer ethanol via the “jello shot” technique, one 10-min basal dialysate sample was collected 2 hr after probe implantation followed by the 1-hr gelatin access period during which 6 dialysate samples were taken. An additional 6–12 samples were taken after ethanol was removed. The behavior of the animal during the 1-hr access period of the microdialysis experiment was observed to allow for analysis of the time course of consumption. Each 30-s interval was scored as a “1” if the animal displayed ingestive behavior (licking, nibbling of gelatin) at any time within that interval, or as a “0” otherwise. Scores were averaged within each 5-min period of gelatin access for each rat.
The ethanol-naïve rats received ethanol via intragastric gavage during the microdialysis experiment. Two hrs after probe insertion and immediately after the basal sample was taken, ethanol (0.5 g/kg) was gavaged in either water or in the gelatin vehicle and microdialysis samples were collected as described above. Each rat received both treatments, in a counterbalanced fashion, at least 3 hrs apart after brain ethanol had reached undetectable levels for at least 60 minutes.
The ethanol content of each dialysate sample was determined by the alcohol dehydrogenase assay (see Kristoffersen & Smith-Kielland 2005
) adapted for small volume microdialysate samples. Dialysate samples (10 μl) were stored in 100 μl 0.6 M glycine buffer (pH 9.2) on ice until all samples were collected. A standard curve was prepared containing 10 μl of 0, 0.32, 0.63, 1.25, 2.5, 5, 10 and 20 mM ethanol added to 100 μl glycine buffer both before and after sample collection, to ensure that ethanol evaporation was minimal during the 3–4 hr dialysis experiment. After all samples were collected, 100 μl of enzyme solution (0.88 mg NAD and 0.29 mg ADH per ml of 0.6 M glycine buffer) were added to each sample and then all samples were incubated at 37°C in a shaking water bath for 20 minutes. Samples were then kept on ice for at least 5 minutes, before transfer to a 96-well plate (Costar, Corning NY) on ice. Absorbance was read at 340 nm using a Synergy HT platereader (Bio-Tek, Winooski VT). Absorbance was converted to mM ethanol by the use of a polynomial equation since at these low volumes and concentrations, the standard curve was not linear. Fits were generally better than R2
= 0.98, and there was less than 10% error when the absorbance of standard curve values were recalculated back to mM concentrations.
Prior to implantation of each probe into an animal’s guide cannula, extraction efficiency of that probe was determined by measuring the amount of ethanol in two 10-min dialysate samples when the probe was placed in a 10 mM ethanol solution stirred at 37°C. The extraction efficiency was the ratio of the dialysate concentration and the actual concentration of the 10 mM standard. All subsequent in vivo dialysate samples were corrected by dividing the measured concentration by the extraction efficiency.
In separate experiments (N = 6), we determined the appropriateness of this in vitro calibration method by comparing extraction efficiency using different concentration of ethanol standards (1.25, 2.5, 5, 10 and 20 mM). The data were plotted as ethanol gain to probe versus outside solution concentration and the slope of this line was called the Ed for diffusion into the probe under in vitro conditions. Similarly, in separate experiments (N=6) we determined the Ed for diffusion of ethanol out of the probe, both in vivo and in vitro, as described by Robinson et al (2000)
. An initial dialysate sample was taken while the probe was perfused with aCSF. Then the perfusate was switched to either 2.5, 5, 10, or 20 mM ethanol in aCSF. After at least 10-min equilibration time, duplicate dialysate samples were collected for 10 min each. The perfusate was then switched in a random order to another concentration. The slope obtained from linear regression of the data indicated the in vitro delivery Ed. For three of these probes, this procedure was repeated 2 hrs after the probe was inserted into a guide cannula placed in nucleus accumbens as described above. The data were plotted as ethanol loss from probe versus perfusate concentration. The slope obtained from linear regression of the data indicated the in vitro and in vivo Ed for delivery of ethanol, respectively.
Dialysate ethanol levels in each 10-min sample were converted to mM concentration and then corrected for efficiency using the extraction efficiency determined by the in vitro calibration with a 10 mM standard. The area under the curve (AUC) was calculated for each rat by summing the mM values calculated over time. The maximal ethanol concentration (Cmax) and the time to maximal concentration (Tmax) were determined visually from the graphs of dialysate concentration over time. ANOVA or paired T-test was used to analyze changes in consumption levels (g ethanol/kg body weight) as well as resulting mM brain ethanol concentrations over time, AUC, Tmax and Cmax values. Paired T-test was used to analyze differences in Ed, AUC, Tmax and Cmax values.
The day-by-day intakes of the jello shot while access was tapered from 24 to 1 hr per day are shown in . On the first day of 24-hr access, average consumption was over 7 g/kg ethanol with only 3 out of 14 rats consuming less than 3.0 g/kg per day. On the second day of 24-hr access, no rat ate less than 3.0 g/kg per day and average consumption was increased to over 8 g/kg/day. Rats consumed progressively less ethanol as the access time per day decreased (from 24 to 6 to 3 to 1 hr) until reaching an average consumption of 1.2±0.2 g/kg ethanol during the days in which they had 1-hr access to the “jello shots”. When these data were analyzed using a one-way ANOVA, there was a significant effect of Days (F(17,221) = 49.2; p < 0.001), which was entirely due to a significant decrease in consumption during the first 8 days of jello shot access (F(7,91) = 40.9; p < 0.001). There was no significant effect of days during the last 10 days of consumption before starting the habituation procedures (F(9,117) = 1.6; NS).
When presentation of the 10% ethanol “jello shot” took place in the novel environment of the microdialysis set-up for their 1-hr gelatin access period (), consumption was initially decreased about 60% but resumed previous levels within a couple of days. Similarly, when “jello shot” presentation was resumed in the dialysis cage post-surgery, with the addition of the tethering procedure, there was a 60% decrease in mean ethanol consumption that recovered within a few days. Even so, there was significantly lower consumption during the microdialysis experiment () compared to prior consumption either before surgery or on after surgery. These observations were supported by a significant effect of days (F(7,91) = 7.4; p < 0.001), which was due to significant decreases in consumption on rl1, rl2, ps1 and the day of the dialysis experiment relative to the home cage consumption. The behavioral observations for all rats during the dialysis experiment indicate that most of the gel was consumed in the first 10–15 minutes, after which the rats remained inactive for the rest of the hour ().
Figure 4 Time course of “jello shot” Consumption. Mean consumption scores during the microdialysis experiment was scored on 30-s intervals for the 60-min ethanol access period. Behavior consistent with consumption (chewing, licking) was assigned (more ...)
Prior to each microdialysis experiment described below, extraction efficiency of the probe used in each experiment was determined in vitro using 10 mM ethanol as a standard. The average extraction efficiency for the 25 microdialysis experiments was 0.35±0.02. In separate experiments (N = 6), we determined the appropriateness of this in vitro calibration by determining Ed into and out of the probe under in vitro and in vivo conditions. We found that extraction efficiency of ethanol into the probe under in vitro conditions did not differ significantly (ranging from 0.31 to 0.40) regardless of the concentration of ethanol (1.25, 2.5, 5, 10 and 20 mM) in the tube, with no relationship between efficiency and ethanol concentration. When the data were plotted as ethanol gain to probe versus outside solution concentration (), the slope obtained from linear regression of the data (Ed, in vitro in) was 0.35±0.04 (n = 6) which was not significantly different from the extraction efficiency calculated for the 10 mM standard. We next determined the Ed for diffusion of ethanol out of the probe, both in vivo and in vitro (). The Ed for diffusion out of the probe under in vitro conditions (Ed in vitro out) was 0.53±0.03 (n = 6) while Ed for diffusion out of the probe under in vivo conditions (Ed in vivo out) was 0.34±0.03 (n = 3). The in vitro Ed ratio (Ed diffusionIN/Ed diffusionOUT) was 0.61±0.04, thus, the Ed values for in vitro diffusion into and out of the probe were significantly different (T-test, p < 0.01) as were the Ed values for diffusion out of the probe in vivo vs in vitro (p < 0.01). However, there was no difference in the Ed values for diffusion out of the probe under in vivo conditions and diffusion into the probe under in vitro conditions (p = 0.88).
Figure 5 The effect of diffusion direction on ethanol Ed into and out of microdialysis probes under in vitro and in vivo conditions. The data are plotted as the diffusion of ethanol across the probe membrane versus the ethanol concentration from where it is diffusing (more ...)
Brain ethanol levels during the 1-hr free access to “jello shot” are shown in . Based on each rat’s consumption during the dialysis experiment, rats were divided into those consuming more than 0.7 g/kg (high consumers; mean intake = 0.9±0.1 g/kg ethanol, N = 7) and less than 0.7 g/kg (low consumers; mean intake = 0.4±0.1, N = 7). The respective Cmax values were 8.2±2.1 and 3.7±0.7 mM and these occurred for both groups between 20 and 30 min after the “jello shots” were introduced into the cage. When individual AUCs were calculated, there was a significant correlation between dose and resulting brain ethanol levels ().
Figure 6 Panel A. Brain ethanol levels determined every 10 minutes during and after gelatin access. The majority of ethanol-containing gelatin was consumed within 10 minutes. For rats consuming less than 0.7 g/kg ethanol (Low; N = 7), the AUC was 16.08±4.0. (more ...)
When a separate group of ethanol-naïve rats received gavage of 0.5 g/kg ethanol, (), brain ethanol levels were lower by about 20% when the ethanol was delivered via a gelatin vehicle versus a water vehicle. ANOVA of the mM concentrations in each sample over time indicated a significant main effect of vehicle (F(1,7) = 6.3, p < 0.05) and time (F(12,84) = 49.1, p < 0.001). The AUC for ethanol in gelatin was also about 20% lower than the AUC for ethanol in water (paired T-test, p < 0.05). Cmax values were significantly decreased by about 20% (7.0±0.9 mM for gelatin vs 9.1 ± 0.9 mM for water, paired T-test, p < 0.05). There was no difference in Tmax values (26.3±2.8 min for gelatin vs 22.5±1.7 for water, paired T-test, p = 0.76).
Figure 7 Brain ethanol levels after intragastric gavage of 0.5 g/kg ethanol either in “jello shot” or in water. Ethanol levels were decreased about 20% by the gelatin vehicle when data were analyzed over time and for calculated AUCs (gelatin = (more ...)