- Email: [email protected]
Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats
Accepted Manuscript Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats Giancarlo Colombo, Carla Lobina, Irene Lorrai, Carla Acciaro, Paola Maccioni, Gian Luigi Gessa PII:
S0741-8329(17)30227-6
DOI:
10.1016/j.alcohol.2017.04.002
Reference:
ALC 6703
To appear in:
Alcohol
Received Date: 2 February 2017 Revised Date:
5 April 2017
Accepted Date: 5 April 2017
Please cite this article as: Colombo G., Lobina C., Lorrai I., Acciaro C., Maccioni P. & Gessa G.L., Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats, Alcohol (2017), doi: 10.1016/j.alcohol.2017.04.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats
RI PT
Giancarlo Colombo a, Carla Lobina a, Irene Lorrai a, Carla Acciaro a, Paola Maccioni a, Gian Luigi Gessa a a
EP
TE D
M AN U
SC
Neuroscience Institute, Section of Cagliari, National Research Council of Italy, S.S. 554, km. 4,500, I-09042 Monserrato (CA), Italy
AC C
Corresponding author: Giancarlo Colombo Neuroscience Institute, Section of Cagliari National Research Council of Italy S.S. 554, km. 4,500 I-09042 Monserrato (CA) Italy Telephone: +39 070 6754342 Fax: +39 070 6754320 E-mail: [email protected]
ACCEPTED MANUSCRIPT
Abstract Previous studies suggested that exposure of Sardinian alcohol-preferring (sP) rats to daily drinking sessions of 1 h, during the dark phase of the light/dark cycle,
RI PT
with multiple alcohol concentrations, and unpredictable access to alcohol, resulted in exceptionally high intakes of alcohol when the drinking session occurred over the
last hours of the dark phase. Additionally, higher levels of anxiety-related behaviors were observed at the 12th, rather than 1st, hour of the dark phase, suggesting that
SC
uncertainty of time of alcohol access and expectation of alcohol availability produced an emotional “distress”. The present study was designed to provide pharmacological
M AN U
support to the hypothesis that high alcohol intake under this drinking procedure is secondary to exacerbation of the anxiety-like state of sP rats. To this end, sP rats were initially exposed to daily 1-h drinking sessions during the dark phase and with multiple alcohol concentrations (0%, 10%, 20%, and 30%; v/v); time of alcohol
TE D
exposure was changed each day and was unpredictable to rats. Rats were then treated acutely with non-sedative doses of diazepam (0, 1, 2, and 3 mg/kg; intraperitoneally [i.p.]) before two drinking sessions occurring at the 1st and
EP
12th hour of the dark phase, respectively. Treatment with diazepam was ineffective at the 1st hour; conversely, it selectively reduced alcohol intake (up to 50% at the
AC C
dose of 3 mg/kg) at the 12th hour. The preferential effectiveness of diazepam in reducing alcohol intake when the drinking session occurred at the 12th hour of the dark phase is consistent with the hypothesis that uncertainty of time of alcohol access and expectation of alcohol availability generated an emotional “distress” that rats counterbalanced with high alcohol drinking; the results of the present study are interpreted as the anxiolytic effects of diazepam substituting for those of alcohol, resulting in the observed reduction in alcohol intake.
ACCEPTED MANUSCRIPT
Introduction Recent research found that selectively bred, Sardinian alcohol-preferring (sP) rats displayed alcohol binge-like drinking when exposed to daily drinking sessions of
RI PT
limited duration, during the dark phase of the light/dark cycle, with multiple alcohol concentrations, and – importantly – unpredictable access to alcohol (Colombo et al., 2014; 2015). More specifically, unpredictable exposure to daily 1-h drinking sessions led to intakes of alcohol exceeding 2 g/kg when the drinking session occurred during
SC
the later hours of the dark phase. These large intakes of alcohol a) were more than double those recorded when the drinking session occurred during the first hours of
M AN U
the dark phase (with intermediate values in the in-between hours), b) resulted in blood alcohol levels as high as 100 mg%, and c) produced severe motor impairment (sign of alcohol-induced intoxication) (Colombo et al., 2014, 2015). Sensitivity of alcohol drinking to time schedule generalized to operant self-administration of
TE D
alcohol: sP rats with a “history” of lever responding for multiple alcohol concentrations, in daily 1-h self-administration sessions and unpredictable time schedules, displayed stronger reinforcing and motivational properties of alcohol at
EP
the end, rather than the beginning, of the dark phase (Maccioni et al., 2015). We hypothesized that the exceptionally high intakes of alcohol recorded when
AC C
the drinking session occurred over the final hours of the dark phase were secondary to an exacerbation of the inherent anxiety-like state of sP rats (Colombo et al., 2015). In other words, we hypothesized that a) the unpredictable schedule of alcohol access, together with the expectation of alcohol availability, produced an emotional “distress” and b) rats coped with this progressively increasing, negative affective state by consuming progressively larger amounts of alcohol, likely seeking its anxiolytic effects (Colombo et al., 2015). Accordingly, sP rats with a “history” of repeated, unpredictable exposures to daily alcohol drinking sessions, subsequently
ACCEPTED MANUSCRIPT
exposed to the Social Interaction test [a validated experimental procedure to measure anxiety-related behaviors in rats (File, Kenny, & Ouagazzal, 1998; File, Lippa, Beer, & Lippa, 2004)], displayed higher levels of anxiety-related behaviors at
RI PT
the 12th, rather than 1st, hour of the dark phase (Colombo et al., 2015). If this “anxiety” hypothesis is correct, an anxiolytic compound would be
expected to effectively reduce alcohol intake in sP rats exposed to this peculiar,
SC
“anxiogenic” drinking regimen. Accordingly, the anxiolytic compound should be
preferentially effective at the last, rather than first, hours of the dark phase. In an
M AN U
attempt to address this research question, and provide pharmacological proof of this working hypothesis, the present study tested the effect of acute treatment with diazepam (a reference anxiolytic compound in preclinical behavioral pharmacology) on alcohol intake at the 1st and 12th hour of the dark phase in sP rats exposed to daily, 1-h drinking sessions, with unpredictable and concurrent access to 0%, 10%,
TE D
20%, and 30% (v/v) alcohol.
Part of the relevance of the present study may reside in the highly frequent co-occurrence of anxiety and alcohol-use disorder (see Hasin, Stinson, Ogburn, &
EP
Grant, 2007). It has indeed been estimated that up to 30% of individuals with alcohol-use disorder also suffer from comorbid anxiety disorders (see Vorspan,
AC C
Mehtelli, Dupuy, Bloch, & Lépine, 2015); among anxiety-disorder patients, up to 10% develop an alcohol-use disorder and up to 50% consume alcohol to self-medicate their anxiety symptoms, with a high risk of converting their alcohol consumption into alcohol dependence (see Vorspan et al., 2015). An animal model of anxiety-induced excessive alcohol intake, up to intoxication, may therefore constitute an appropriate tool for investigations of the neurobiological and pharmacological bases of this comorbidity.
ACCEPTED MANUSCRIPT
Materials and Methods The experimental procedures employed in the present study were in accordance with European Directive no. 2010/63/EU and subsequent Italian
RI PT
Legislative Decree no. 26, March 4, 2014, Law on the “Protection of animals used for scientific purposes”. Animals
Male sP rats (n = 32) from the 91st generation were used. Rats were singly
SC
housed in standard plastic cages with wood chip bedding; single housing started at the age of approximately 45 days. The animal facility was under an inverted 12:12-h
M AN U
light/dark cycle (lights on at 8:00 PM) at a constant temperature of 22 ± 2 °C and relative humidity of approximately 60%. Regular food pellets (Harlan, San Pietro al Natisone, Italy) and water were always available in the home cage. Starting from the first day of single-cage housing, rats were extensively
TE D
habituated to handling and intraperitoneal injections. Inverted screen apparatus
The apparatus was custom-made; its assembly was inspired by the inverted
EP
screen for motor ataxia in mice initially described by Coughenour, Mclean, & Parker (1977). The apparatus consisted of a rectangular (120 mm × 300 mm) wire mesh
AC C
screen; the wire had a diameter of 3 mm. The entire screen was surrounded by 120-mm high plastic walls to prevent rats from attempting to climb onto the other side. The screen was mounted on a metal rod, connected to an engine that rotated the screen 90° (from horizontal to vertical) over 5 sec at constant speed. The apparatus was positioned on a lab bench such that, when the screen was in a vertical position, the low edge of the screen was placed 30 cm from the bench top. A thick foam pad was positioned below the screen to cushion the rat fall.
ACCEPTED MANUSCRIPT
Experimental procedure Schematic representation of the experimental procedure is given in Fig. 1. At the age of approximately 60 days, rats were exposed to the home cage,
RI PT
4-bottle choice regimen between water and 10%, 20%, and 30% (v/v) alcohol with unlimited access (24 h/day) for 12 consecutive days (Phase 1). A 24-h period of
alcohol withdrawal, with water as the sole fluid available, was interposed between Phases 1 and 2.
SC
Rats were then exposed to the above 4-bottle choice regimen, with access
limited to 1 h/day during the dark phase, for 15 consecutive days (Phase 2). It was
M AN U
conceived that rats might be initially disoriented by the abrupt and unexpected removal of the alcohol bottles at the end of the drinking session, and thus require time to adapt their alcohol-drinking behavior to this new regimen. Accordingly, the first three drinking sessions of Phase 2 (Sessions A–C in Fig. 1) were considered as
TE D
“acclimatization” sessions, and data were excluded from analysis. Time of access to alcohol was established semi-randomly so that, over the 12 drinking sessions following the “acclimatization” sessions, all 12 hours of the dark phase were tested.
EP
The sequence used was: 3rd, 8th, 11th, 2nd, 9th, 6th, 10th, 4th, 1st, 12th, 5th, and 7th hour; the “acclimatization” sessions occurred at the 6th, 2nd, and 10th hour.
AC C
Water was always available (24 h/day). Tests with diazepam occurred in two different drinking sessions. The first test
session was conducted on the day immediately after the last drinking session of Phase 2 and during the 1st hour of the dark phase. The second test session was conducted 7 days later and during the 12th hour of the dark phase. The in-between phase (Phase 3) was made up of 6 consecutive drinking sessions, the timing of which was changed daily and established semi-randomly from odd hours of the dark phase (sequence used: 5th, 11th, 3rd, 9th, 1st, and 7th hour). The marked
ACCEPTED MANUSCRIPT
consistency of sensitivity of alcohol drinking to time schedule, resulting in virtually completely superimposable intakes of alcohol even after repetition of several consecutive phases of alcohol drinking sessions with time of access to alcohol
RI PT
changing daily and being unpredictable to rats (this laboratory, unpublished results), allowed us to test diazepam in two separate drinking sessions occurring 1 week
apart; it was indeed ruled out that Phase 3 could alter basal alcohol drinking at the
SC
second test session with diazepam. In both test sessions with diazepam, rats were
divided into four groups of n = 8, matched for alcohol intake and schedule sensitivity
M AN U
over the preceding phase; additionally, in the second test, no rat was treated with the same diazepam dose received in the first test. Diazepam (injectable Valium®, Roche, Monza, Italy) was dissolved in a 1:1 mixture of saline and polyethylene glycol 600 and injected intraperitoneally (injection volume: 2 mL/kg) at the doses of 0, 1, 2, and 3 mg/kg, 30 min before the start of the drinking session.
TE D
After the second test session with diazepam, rats were exposed to 6 additional consecutive drinking sessions (Phase 4), the timing of which was changed daily and established semi-randomly from even hours of the dark phase (sequence
EP
used: 10th, 4th, 12th, 2nd, 6th, and 8th). The following day was devoted to testing the possible motor-incoordinating effects of diazepam at the inverted screen.
AC C
Specifically, rats were divided into four groups of n = 8, matched for alcohol intake and schedule sensitivity over Phase 4; additionally, no rat was treated with the same diazepam doses received in the two drinking sessions. Rats were treated with 0, 1, 2, and 3 mg/kg diazepam. Diazepam was dissolved and injected as described above. Thirty minutes after diazepam administration, rats were exposed to the inverted screen test. Specifically, each rat was placed individually in the center of the wire mesh screen. Immediately after rat placement, the screen was rotated from
ACCEPTED MANUSCRIPT
horizontal to vertical and then kept vertical. The time when the rat fell off the screen was recorded. If the rat did not fall off, it was removed after 120 sec. Rats were exposed to two different trials, occurring one after the other with no time interval in
RI PT
between. Tests at the inverted screen took place between the 6th and 7th hour of the dark phase; no drinking session was planned on the day in which the test at the inverted screen was conducted.
SC
In all drinking phases (Phases 1–4) and in the two test sessions with
diazepam, bottles were refilled daily with fresh solution and their positions changed
M AN U
randomly and daily to prevent development of position preference. Measured variables and statistical analysis
In all drinking phases (Phases 1–4) and in the two test sessions with diazepam, the measured variables were alcohol and water intake. These were monitored by weighing the bottles with a 0.01-g accuracy a) every day immediately
TE D
before the start of the dark phase in Phase 1, and b) immediately before and immediately after each daily drinking session in Phases 2–4 as well as in the two test sessions with diazepam. Possible fluid spillage was calculated by using multiple
EP
bottles filled with the different alcohol concentrations and positioned in empty cages interspersed in the cage racks; mean spilt volumes were subtracted before data
AC C
analysis. Alcohol and water intake were expressed in g/kg pure alcohol and mL/kg water, respectively. In the test sessions with diazepam, food intake was also recorded. Food pellets were weighed with a 0.01-g accuracy immediately before and immediately after the session. Food intake was expressed in g/kg. Data on alcohol and water intake over the 12 days of Phase 1 were analyzed by separate 1-way ANOVAs with repeated measures. Data on alcohol and water intake over the 12 drinking sessions of Phase 2, 6 drinking sessions of Phase 3, and 6 drinking
ACCEPTED MANUSCRIPT
sessions of Phase 4 were a) analyzed by separate 1-way ANOVAs with repeated measures and b) exposed to separate regression analyses (mean alcohol intake vs. time of the drinking session) and calculation of the Pearson correlation coefficient.
RI PT
Data on alcohol, water, and food intake from the test sessions with diazepam were analyzed by separate 2-way [time (1st or 12th hour); treatment (diazepam dose)] ANOVAs, followed by the Tukey’s test for post hoc comparisons.
SC
In the inverted screen test, the measured variable was the time (expressed in sec) spent on the screen before falling off. Rats that did not fall off were assigned the
M AN U
value of 120 sec. The average of values recorded in the two trials provided the value assigned to each rat. Data were statistically analyzed by 1-way ANOVA. Results Alcohol drinking in Phases 1–4
Phase 1 – ANOVA indicated highly significant differences in daily alcohol
TE D
intake during the 12 days of Phase 1 [F(11,341) = 15.06, p < 0.0001] (Fig. 2, left panel). All rats rapidly acquired alcohol-drinking behavior, as indicated by a mean daily intake of alcohol as high as approximately 3.8 g/kg on the initial day of exposure to
EP
the 4-bottle choice regimen; on the following days mean daily alcohol intake rose progressively, averaging around 6 g/kg over the last three days. ANOVA indicated
AC C
highly significant differences in daily water intake during the 12 days of Phase 1 [F(11,341) = 6.82, p < 0.0001] (Fig. 2, right panel). Mean daily water intake decreased progressively over the period. Phase 2 – ANOVA indicated highly significant differences in alcohol intake
during the 12 drinking sessions of Phase 2 [F(11,341) = 50.81, p < 0.0001] (Fig. 3, left panel). Specifically, mean alcohol intake varied largely, from a minimum value of 0.65 ± 0.05 g/kg, when the drinking session occurred during the 2nd hour of the dark
ACCEPTED MANUSCRIPT
phase, to a maximum value of 2.42 ± 0.09 g/kg, when the drinking session occurred during the 10th hour of the dark phase. Mean alcohol intake was highly positively correlated with time of access to alcohol (r = 0.906, slope = 0.127 ± 0.019,
RI PT
intercept = 0.696 ± 0.139, p < 0.0001). Mean water intake was negligible (<1 mL/kg) in each drinking session.
Phase 3 – ANOVA indicated highly significant differences in alcohol intake
SC
during the 6 drinking sessions of Phase 3 [F(5,155) = 76.67, p < 0.0001] (Fig. 3, center panel). Specifically, mean alcohol intake varied largely, from a minimum value of
M AN U
0.78 ± 0.05 g/kg, when the drinking session occurred during the 3rd hour of the dark phase, to a maximum value of 2.35 ± 0.09 g/kg, when the drinking session occurred during the 11th hour of the dark phase. Mean alcohol intake was positively correlated with time of access to alcohol (r = 0.972, slope = 0.156 ± 0.019, intercept = 0.538 ± 0.130, p < 0.005). Mean water intake was negligible (<1 mL/kg) in
TE D
each drinking session.
Phase 4 – ANOVA indicated highly significant differences in alcohol intake during the 6 drinking sessions of Phase 4 [F(5,155) = 38.51, p < 0.0001] (Fig. 3, right
EP
panel). Specifically, mean alcohol intake varied largely, from a minimum value of 0.84 ± 0.06 g/kg, when the drinking session occurred during the 2nd hour of the dark
AC C
phase, to a maximum value of 2.14 ± 0.10 g/kg, when the drinking session occurred during the 10th hour of the dark phase. Mean alcohol intake was positively correlated with time of access to alcohol (r = 0.870, slope = 0.104 ± 0.030, intercept = 0.859 ± 0.230, p < 0.05). Mean water intake was negligible (<1 mL/kg) in each drinking session. Effect of treatment with diazepam on alcohol drinking
ACCEPTED MANUSCRIPT
ANOVA indicated a highly significant effect of time of the drinking session (1st or 12th hour) [F(1,56) = 68.96, p < 0.0001], a significant effect of treatment with diazepam [F(3,56) = 3.99, p < 0.05], and a significant “time × treatment” interaction
RI PT
[F(3,56) = 5.53, p < 0.005] on alcohol intake (Fig. 4). In vehicle-treated rats, alcohol intake differed by approximately 3-fold when the drinking session occurred during the 1st (0.92 ± 0.18 g/kg) and 12th (2.72 ± 0.36 g/kg) hour of the dark phase
SC
(p < 0.0001, Tukey’s test). Post hoc analysis indicated that no dose of diazepam
altered alcohol intake when the drinking session occurred during the 1st hour of the
M AN U
dark phase (Fig. 4, left panel). Conversely, when the drinking session occurred during the 12th hour of the dark phase, treatment with 2 and 3 mg/kg diazepam resulted, in comparison to vehicle treatment, in statistically significant reductions in alcohol intake; the magnitude of these reductions averaged approximately 35% (p < 0.05, Tukey’s test) and 50% (p < 0.0005, Tukey’s test), respectively (Fig. 4, right
TE D
panel).
ANOVA indicated a significant effect of time of the drinking session [F(1,56) = 5.33, p < 0.05], no effect of treatment with diazepam [F(3,56) = 1.01,
EP
p > 0.05], and no “time × treatment” interaction [F(3,56) = 1.40, p > 0.05] on water intake (Table 1). ANOVA indicated no effect of time of the drinking session
AC C
[F(1,56) = 0.22, p > 0.05], no effect of treatment with diazepam [F(3,56) = 0.55, p > 0.05], and no “time × treatment” interaction [F(3,56) = 1.23, p > 0.05] on food intake (Table 1).
Effect of treatment with diazepam on motor performance Treatment with diazepam failed to alter rat motor-performance, as indicated by the lack of statistical difference – among the four rat groups – in time spent on the inverted screen [F(3,28) = 0.08, p > 0.05] (Table 2).
ACCEPTED MANUSCRIPT
Discussion Data on alcohol intake from Phases 2–4 confirmed the sensitivity to time schedule of alcohol drinking in sP rats exposed to daily 1-h drinking sessions, during
RI PT
the dark phase of the light/dark cycle, with concurrent availability of multiple alcohol concentrations, and unpredictable access to alcohol. Alcohol intake highly positively correlated with the time of alcohol access, and increased progressively from an
average of 0.6–0.8 g/kg, when the drinking session occurred during the first hours, to
SC
an average of 2.1–2.4 g/kg, when the drinking session occurred during the later
hours. Similar data were collected in the two test sessions with diazepam, in which
M AN U
alcohol intake in vehicle-treated (control) rats averaged approximately 0.9 and 2.7 g/kg at the 1st and 12th hour of the dark phase, respectively. Together, these data closely replicate those collected in previous studies (Colombo et al., 2014; 2015) and confirm that this experimental procedure may result in sP rats consuming
TE D
exceptionally high amounts of alcohol [meeting the criteria posed for binge-like drinking in animal models (Bell, Rodd, Lumeng, Murphy, & McBride, 2006)]. The similarity of values of alcohol intake recorded in Phases 2, 3, and 4 (occurring
EP
sequentially, one after another) confirmed that – once developed – sensitivity of alcohol drinking to time schedule remained stable over time.
AC C
The experiment testing diazepam provided some interesting information. First, treatment with diazepam was totally ineffective on alcohol intake when the drinking session occurred at the 1st hour of the dark phase: no tested dose of diazepam altered alcohol intake. The tested dose range of diazepam (1–3 mg/kg; i.p.) was likely adequate and sufficiently wide, as higher doses would have resulted in sedative and motor-incapacitating effects, affecting the regular drinking rates of rats; additionally, it virtually overlapped with those used in several previous studies investigating the effect of diazepam treatment on alcohol-motivated behaviors in rats
ACCEPTED MANUSCRIPT
(e.g., Daoustet et al., 1987; Janak, Redfern, & Samson, 1998; Rimondini, Sommer, & Heilig, 2002; Shelton & Balster, 1997). Our initial investigation on the drinking procedure employed in the present
RI PT
study found that alcohol intake in sP rats was modestly sensitive to unpredictability of time of alcohol access when the drinking session occurred during the first hours of the dark phase: in this time period, alcohol intake was indeed found to be highly
SC
similar under “unpredictable” and “fixed” conditions (the latter comprising daily drinking sessions taking place constantly at the 1st hour of the dark phase)
M AN U
(Colombo et al., 2014). Thus, alcohol intake at the 1st hour of the dark phase is minimally influenced by unpredictability of time of alcohol access and can be equated to that recorded under more standard, or basic, conditions of experimental alcohol drinking in rats. Based on this premise, the observed failure of diazepam treatment to alter alcohol drinking in sP rats exposed to alcohol at the 1st hour of the
TE D
dark phase was not entirely unexpected, as acute treatment with dose ranges of diazepam highly comparable to that used in the present study has repeatedly been reported to be ineffective on alcohol intake and lever-responding for alcohol in rats
EP
exposed to conventional procedures of alcohol drinking and operant, oral alcohol self-administration (e.g., Daoust et al., 1987; Janak et al, 1998; Rimondini et al.,
AC C
2002; Shelton & Balster, 1997). Conversely, treatment with diazepam effectively reduced the exceptionally
high alcohol intake recorded when the drinking session occurred during the 12th hour of the dark phase. In comparison to the vehicle-treated rat group, alcohol intake decreased by approximately 25%, 35%, and 50% in the rat groups treated with 1, 2, and 3 mg/kg diazepam, respectively (although only reduction exerted by the two highest doses of diazepam reached statistical significance at post hoc
ACCEPTED MANUSCRIPT
analysis). Notably, the reducing effect of diazepam was selective for alcohol intake, as treatment with diazepam did not affect water and food intake. While data on water intake might be of limited usefulness (water intake was extremely low, making its
RI PT
analysis virtually meaningless in terms of evaluation of the selectivity of diazepam action), data on food intake clearly demonstrate that feeding was totally unaffected by diazepam treatment. Additionally, treatment with diazepam failed to alter rat
SC
performance at a highly challenging task (in terms of muscle strength and motor
coordination), such as hanging or moving on the inverted screen. These data confer
M AN U
specificity to the reducing effect of diazepam on alcohol intake, as they suggest that no sedative or motor-incoordinating effect of diazepam hampered the regular rates of drinking and feeding.
The preferential effectiveness of diazepam in reducing alcohol intake when the drinking session occurred at the 12th, rather than 1st, hour of the dark phase
TE D
provides support to the working hypothesis of the present study: uncertainty of time of alcohol access and expectation of alcohol availability, i.e., the peculiar features of the procedure of alcohol drinking used in the present study, added an emotional
EP
“distress” to the inherent, basal anxiety-like state of sP rats (see Colombo, Lobina, Carai, & Gessa, 2006); sP rats apparently cope with this exacerbated, negative
AC C
affective state by consuming larger amounts of alcohol, likely seeking its anxiolytic effects. Accordingly, we hypothesize that the anxiolytic effects of diazepam substituted for those of alcohol, resulting in a less urgent need to consume alcohol and, in turn, in the observed, remarkable reduction in alcohol intake. To summarize, the results of the diazepam experiment (present study) complement those of the recent Social Interaction experiment (Colombo et al., 2015) demonstrating that sP rats with a “history” of unpredictable exposures to daily
ACCEPTED MANUSCRIPT
alcohol drinking sessions displayed more anxiety-related behaviors over the later rather than first hours of the dark phase. Taken together, these results provide an “anxiety”-founded hypothesis to the neurobiological bases of sensitivity to time
RI PT
schedule and excessive alcohol intake in sP rats exposed to this new experimental procedure of alcohol drinking. Combination of inherently “anxious” sP rats with this “anxiogenic” procedure of alcohol drinking apparently opens up a rat model of
SC
potential utility for investigations of comorbid anxiety and alcohol-use disorder. Acknowledgments
M AN U
The authors are grateful to Ms. Anne Farmer for language editing of the manuscript and Mr. Alessandro Capra for technical support. Conflict of Interest None. Funding
TE D
This research did not receive any specific grant from funding agencies in the
AC C
EP
public, commercial, or not-for-profit sectors.
ACCEPTED MANUSCRIPT
References Bell, R. L., Rodd, Z. A., Lumeng, L., Murphy, J. M., & McBride, W. J. (2006). The alcohol-preferring P rat and animal models of excessive alcohol drinking. Addiction Biology, 11, 270–288. doi: 10.1111/j.1369-1600.2005.00029.x
RI PT
Colombo, G., Lobina, C., Carai, M. A., & Gessa, G. L. (2006). Phenotypic characterization of genetically selected Sardinian alcohol-preferring (sP) and -non-preferring (sNP) rats. Addiction Biology, 11, 324–338. doi: 10.1111/j.1369-1600.2006.00031.x
Colombo, G., Lobina, C., Maccioni, P., Carai, M. A.., Lorrai, I., Zaru, A., et al. (2015). Anxiety-like behaviors at the end of the nocturnal period in sP rats with a “history” of unpredictable, limited access to alcohol. Alcohol, 49, 707–712. doi: 10.1016/j.alcohol.2015.04.010
SC
Colombo, G., Maccioni, P., Acciaro, C., Lobina, C., Loi, B., Zaru, A., et al. (2014). Binge drinking in alcohol-preferring sP rats at the end of the nocturnal period. Alcohol, 48, 301–311. doi: 10.1016/j.alcohol.2014.02.001
M AN U
Coughenour, L. L., Mclean, J. R., & Parker, R. B. (1977). A new device for the rapid measurement of impaired motor function in mice. Pharmacology, Biochemistry, and Behavior, 6, 351–353. Daoust, M., Saligaut, C., Lhuintre, J. P., Moore, N., Flipo, J. L., & Boismare, F. (1987). GABA transmission, but not benzodiazepine receptor stimulation, modulates ethanol intake by rats. Alcohol, 4, 469–472.
TE D
File, S. E., Kenny, P. J., & Ouagazzal, A. M. (1998). Bimodal modulation by nicotine of anxiety in the social interaction test: role of the dorsal hippocampus. Behavioral Neuroscience, 112, 1423–1429. File, S. E., Lippa, A. S., Beer, B., & Lippa, M. T. (2004). Animal tests of anxiety. Current Protocols in Neuroscience, 8, 8.3. doi: 10.1002/0471142301.ns0803s26
EP
Hasin, D. S., Stinson, F. S., Ogburn, E., & Grant, B. F. (2007). Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Archives of General Psychiatry, 64, 830–842. doi: 10.1001/archpsyc.64.7.830
AC C
Janak, P. H., Redfern, J. E., & Samson, H. H. (1998). The reinforcing effects of ethanol are altered by the endogenous neurosteroid, allopregnanolone. Alcoholism: Clinical and Experimental Research, 22, 1106–1112. Maccioni, P., Lorrai, I., Marras, M. F., Contini, A., Capra, A., Piras, G., et al. (2015). Elevated reinforcing and motivational properties of alcohol at the end of the nocturnal period in sP rats. Psychopharmacology, 232, 3585–3595. doi: 10.1007/s00213-015-4010-2 Rimondini, R., Sommer, W., & Heilig, M. (2002). Effects of tiagabine and diazepam on operant ethanol self-administration in the rat. Journal of Studies on Alcohol, 63, 100–106. Shelton, K. L., & Balster, R. L. (1997). Effects of gamma-aminobutyric acid agonists and N-methyl-D-aspartate antagonists on a multiple schedule of ethanol and
ACCEPTED MANUSCRIPT
saccharin self-administration in rats. The Journal of Pharmacology and Experimental Therapeutics, 280, 1250–1260.
AC C
EP
TE D
M AN U
SC
RI PT
Vorspan, F., Mehtelli, W., Dupuy, G., Bloch, V., & Lépine, J. P. (2015). Anxiety and substance use disorders: co-occurrence and clinical issues. Current Psychiatry Reports, 17, 4. doi: 10.1007/s11920-014-0544-y
ACCEPTED MANUSCRIPT
Figure captions Fig. 1. Schematic representation of the experimental design.
RI PT
Fig. 2. Alcohol (left panel) and water (right panel) intake in Sardinian alcoholpreferring (sP) rats exposed to the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen, with unlimited access (24 h/day), for 12 consecutive days. Each point is the mean ± SEM of n = 32 rats.
SC
Fig. 3. Alcohol intake in Sardinian alcohol-preferring (sP) rats exposed to 12 (Phase 2; left panel), 6 (Phase 3; center panel), and 6 (Phase 4; right panel) consecutive daily 1-h drinking sessions under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen. Drinking sessions occurred during one of the 12 hours of the dark phase of the daily light/dark cycle; time of the drinking session was changed daily in a semi-random order and was unpredictable to rats. Each point is the mean ± SEM of n = 32 rats.
AC C
EP
TE D
M AN U
Fig. 4. Effect of acute treatment with different doses of diazepam on alcohol intake in Sardinian alcohol-preferring (sP) rats exposed to 1-h drinking sessions, under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen with unpredictable access, conducted at the 1st (left panel) and 12th (right panel) hour of the dark phase. Diazepam was administered intraperitoneally 30 min before the start of the drinking session. Each bar is the mean ± SEM of n = 8 rats. : p < 0.005 and : p < 0.0001 in comparison to the rat group treated with the same diazepam dose at the 1st hour (Tukey’s test); §: p < 0.05 and §§: p < 0.0005 in comparison to the rat group treated with vehicle (0 mg/kg diazepam) at the 12th hour (Tukey’s test).
ACCEPTED MANUSCRIPT
Table 1 – Effect of acute treatment with different doses of diazepam on water and food intake in Sardinian alcohol-preferring (sP) rats exposed to 1-hour drinking sessions, under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs water” choice regimen with
RI PT
unpredictable access, conducted at the 1st and 12th hour of the dark phase.
Diazepam (mg/kg)
Food intake (g/kg)
1
1st Hour
0.29 ± 0.17
2.11 ± 1.30
12th Hour
0.04 ± 0.03
0.06 ±0.04
1st Hour
3.54 ± 0.66
4.30 ± 0.91
12th Hour
4.82 ± 1.19
5.11 ± 1.09
2
3
3.42 ± 1.72
1.29 ± 1.09
0.04 ±0.03
1.03 ± 0.86
7.16 ± 1.50
4.88 ± 1.18
4.18 ± 1.42
4.13 ± 1.56
SC
Water intake (ml/kg)
0
M AN U
Diazepam was administered intraperitoneally 30 min before the start of the drinking
AC C
EP
TE D
session. Each value is the mean ± SEM of n=8 rats.
1
ACCEPTED MANUSCRIPT
Table 2 – Effect of acute treatment with different doses of diazepam on motor performance, evaluated by the inverted screen test, in Sardinian alcohol-preferring (sP) rats previously exposed to several drinking sessions of 1 hour, under the 4-bottle “alcohol
RI PT
(10%, 20%, and 30%, v/v) vs water” choice regimen and unpredictable access.
Diazepam (mg/kg) 0 Time (in s) spent on the inverted screen before falling off
1
79.3 ± 9.5
2
3
83.5 ± 10.2 76.8 ± 14.7 75.6 ± 14.7
SC
On the test day, no drinking session took place. Thirty min after the intraperitoneal administration of diazepam, each rat was placed individually in the center of the wire mesh
M AN U
screen. The screen was rotated 90° (from horizontal to vertical) over 5 s at constant speed. The time (in s) when the rat fell off the screen was recorded. If the rat did not fall off, it was removed after 120 s and assigned the value of 120 s. Rats were exposed to two different trials, occurring one after the other with no time interval in between. The average of values recorded in the two trials provided the value assigned to each rat. Each value is
AC C
EP
TE D
the mean ± SEM of n=8 rats.
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights • Alcohol drinking in sP rats was highly sensitive to time schedule of alcohol access. • Exceptionally high alcohol intakes were recorded at the last hours of the dark phase. • Alcohol expectation produced an emotional “distress” (anxiety-related behaviors).
RI PT
• The anxiolytic compound, diazepam, effectively reduced excessive alcohol intake. • The anxiolytic effects of diazepam likely substituted for those of alcohol.
AC C
EP
TE D
M AN U
SC
Keywords: time schedule of alcohol drinking; limited and unpredictable access to alcohol; experimental model of binge drinking; anxiety-like behaviors; diazepam; Sardinian alcohol-preferring (sP) rats
S0741-8329(17)30227-6
DOI:
10.1016/j.alcohol.2017.04.002
Reference:
ALC 6703
To appear in:
Alcohol
Received Date: 2 February 2017 Revised Date:
5 April 2017
Accepted Date: 5 April 2017
Please cite this article as: Colombo G., Lobina C., Lorrai I., Acciaro C., Maccioni P. & Gessa G.L., Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats, Alcohol (2017), doi: 10.1016/j.alcohol.2017.04.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Binge drinking and anxiety at the end of the nocturnal period in alcohol-preferring sP rats
RI PT
Giancarlo Colombo a, Carla Lobina a, Irene Lorrai a, Carla Acciaro a, Paola Maccioni a, Gian Luigi Gessa a a
EP
TE D
M AN U
SC
Neuroscience Institute, Section of Cagliari, National Research Council of Italy, S.S. 554, km. 4,500, I-09042 Monserrato (CA), Italy
AC C
Corresponding author: Giancarlo Colombo Neuroscience Institute, Section of Cagliari National Research Council of Italy S.S. 554, km. 4,500 I-09042 Monserrato (CA) Italy Telephone: +39 070 6754342 Fax: +39 070 6754320 E-mail: [email protected]
ACCEPTED MANUSCRIPT
Abstract Previous studies suggested that exposure of Sardinian alcohol-preferring (sP) rats to daily drinking sessions of 1 h, during the dark phase of the light/dark cycle,
RI PT
with multiple alcohol concentrations, and unpredictable access to alcohol, resulted in exceptionally high intakes of alcohol when the drinking session occurred over the
last hours of the dark phase. Additionally, higher levels of anxiety-related behaviors were observed at the 12th, rather than 1st, hour of the dark phase, suggesting that
SC
uncertainty of time of alcohol access and expectation of alcohol availability produced an emotional “distress”. The present study was designed to provide pharmacological
M AN U
support to the hypothesis that high alcohol intake under this drinking procedure is secondary to exacerbation of the anxiety-like state of sP rats. To this end, sP rats were initially exposed to daily 1-h drinking sessions during the dark phase and with multiple alcohol concentrations (0%, 10%, 20%, and 30%; v/v); time of alcohol
TE D
exposure was changed each day and was unpredictable to rats. Rats were then treated acutely with non-sedative doses of diazepam (0, 1, 2, and 3 mg/kg; intraperitoneally [i.p.]) before two drinking sessions occurring at the 1st and
EP
12th hour of the dark phase, respectively. Treatment with diazepam was ineffective at the 1st hour; conversely, it selectively reduced alcohol intake (up to 50% at the
AC C
dose of 3 mg/kg) at the 12th hour. The preferential effectiveness of diazepam in reducing alcohol intake when the drinking session occurred at the 12th hour of the dark phase is consistent with the hypothesis that uncertainty of time of alcohol access and expectation of alcohol availability generated an emotional “distress” that rats counterbalanced with high alcohol drinking; the results of the present study are interpreted as the anxiolytic effects of diazepam substituting for those of alcohol, resulting in the observed reduction in alcohol intake.
ACCEPTED MANUSCRIPT
Introduction Recent research found that selectively bred, Sardinian alcohol-preferring (sP) rats displayed alcohol binge-like drinking when exposed to daily drinking sessions of
RI PT
limited duration, during the dark phase of the light/dark cycle, with multiple alcohol concentrations, and – importantly – unpredictable access to alcohol (Colombo et al., 2014; 2015). More specifically, unpredictable exposure to daily 1-h drinking sessions led to intakes of alcohol exceeding 2 g/kg when the drinking session occurred during
SC
the later hours of the dark phase. These large intakes of alcohol a) were more than double those recorded when the drinking session occurred during the first hours of
M AN U
the dark phase (with intermediate values in the in-between hours), b) resulted in blood alcohol levels as high as 100 mg%, and c) produced severe motor impairment (sign of alcohol-induced intoxication) (Colombo et al., 2014, 2015). Sensitivity of alcohol drinking to time schedule generalized to operant self-administration of
TE D
alcohol: sP rats with a “history” of lever responding for multiple alcohol concentrations, in daily 1-h self-administration sessions and unpredictable time schedules, displayed stronger reinforcing and motivational properties of alcohol at
EP
the end, rather than the beginning, of the dark phase (Maccioni et al., 2015). We hypothesized that the exceptionally high intakes of alcohol recorded when
AC C
the drinking session occurred over the final hours of the dark phase were secondary to an exacerbation of the inherent anxiety-like state of sP rats (Colombo et al., 2015). In other words, we hypothesized that a) the unpredictable schedule of alcohol access, together with the expectation of alcohol availability, produced an emotional “distress” and b) rats coped with this progressively increasing, negative affective state by consuming progressively larger amounts of alcohol, likely seeking its anxiolytic effects (Colombo et al., 2015). Accordingly, sP rats with a “history” of repeated, unpredictable exposures to daily alcohol drinking sessions, subsequently
ACCEPTED MANUSCRIPT
exposed to the Social Interaction test [a validated experimental procedure to measure anxiety-related behaviors in rats (File, Kenny, & Ouagazzal, 1998; File, Lippa, Beer, & Lippa, 2004)], displayed higher levels of anxiety-related behaviors at
RI PT
the 12th, rather than 1st, hour of the dark phase (Colombo et al., 2015). If this “anxiety” hypothesis is correct, an anxiolytic compound would be
expected to effectively reduce alcohol intake in sP rats exposed to this peculiar,
SC
“anxiogenic” drinking regimen. Accordingly, the anxiolytic compound should be
preferentially effective at the last, rather than first, hours of the dark phase. In an
M AN U
attempt to address this research question, and provide pharmacological proof of this working hypothesis, the present study tested the effect of acute treatment with diazepam (a reference anxiolytic compound in preclinical behavioral pharmacology) on alcohol intake at the 1st and 12th hour of the dark phase in sP rats exposed to daily, 1-h drinking sessions, with unpredictable and concurrent access to 0%, 10%,
TE D
20%, and 30% (v/v) alcohol.
Part of the relevance of the present study may reside in the highly frequent co-occurrence of anxiety and alcohol-use disorder (see Hasin, Stinson, Ogburn, &
EP
Grant, 2007). It has indeed been estimated that up to 30% of individuals with alcohol-use disorder also suffer from comorbid anxiety disorders (see Vorspan,
AC C
Mehtelli, Dupuy, Bloch, & Lépine, 2015); among anxiety-disorder patients, up to 10% develop an alcohol-use disorder and up to 50% consume alcohol to self-medicate their anxiety symptoms, with a high risk of converting their alcohol consumption into alcohol dependence (see Vorspan et al., 2015). An animal model of anxiety-induced excessive alcohol intake, up to intoxication, may therefore constitute an appropriate tool for investigations of the neurobiological and pharmacological bases of this comorbidity.
ACCEPTED MANUSCRIPT
Materials and Methods The experimental procedures employed in the present study were in accordance with European Directive no. 2010/63/EU and subsequent Italian
RI PT
Legislative Decree no. 26, March 4, 2014, Law on the “Protection of animals used for scientific purposes”. Animals
Male sP rats (n = 32) from the 91st generation were used. Rats were singly
SC
housed in standard plastic cages with wood chip bedding; single housing started at the age of approximately 45 days. The animal facility was under an inverted 12:12-h
M AN U
light/dark cycle (lights on at 8:00 PM) at a constant temperature of 22 ± 2 °C and relative humidity of approximately 60%. Regular food pellets (Harlan, San Pietro al Natisone, Italy) and water were always available in the home cage. Starting from the first day of single-cage housing, rats were extensively
TE D
habituated to handling and intraperitoneal injections. Inverted screen apparatus
The apparatus was custom-made; its assembly was inspired by the inverted
EP
screen for motor ataxia in mice initially described by Coughenour, Mclean, & Parker (1977). The apparatus consisted of a rectangular (120 mm × 300 mm) wire mesh
AC C
screen; the wire had a diameter of 3 mm. The entire screen was surrounded by 120-mm high plastic walls to prevent rats from attempting to climb onto the other side. The screen was mounted on a metal rod, connected to an engine that rotated the screen 90° (from horizontal to vertical) over 5 sec at constant speed. The apparatus was positioned on a lab bench such that, when the screen was in a vertical position, the low edge of the screen was placed 30 cm from the bench top. A thick foam pad was positioned below the screen to cushion the rat fall.
ACCEPTED MANUSCRIPT
Experimental procedure Schematic representation of the experimental procedure is given in Fig. 1. At the age of approximately 60 days, rats were exposed to the home cage,
RI PT
4-bottle choice regimen between water and 10%, 20%, and 30% (v/v) alcohol with unlimited access (24 h/day) for 12 consecutive days (Phase 1). A 24-h period of
alcohol withdrawal, with water as the sole fluid available, was interposed between Phases 1 and 2.
SC
Rats were then exposed to the above 4-bottle choice regimen, with access
limited to 1 h/day during the dark phase, for 15 consecutive days (Phase 2). It was
M AN U
conceived that rats might be initially disoriented by the abrupt and unexpected removal of the alcohol bottles at the end of the drinking session, and thus require time to adapt their alcohol-drinking behavior to this new regimen. Accordingly, the first three drinking sessions of Phase 2 (Sessions A–C in Fig. 1) were considered as
TE D
“acclimatization” sessions, and data were excluded from analysis. Time of access to alcohol was established semi-randomly so that, over the 12 drinking sessions following the “acclimatization” sessions, all 12 hours of the dark phase were tested.
EP
The sequence used was: 3rd, 8th, 11th, 2nd, 9th, 6th, 10th, 4th, 1st, 12th, 5th, and 7th hour; the “acclimatization” sessions occurred at the 6th, 2nd, and 10th hour.
AC C
Water was always available (24 h/day). Tests with diazepam occurred in two different drinking sessions. The first test
session was conducted on the day immediately after the last drinking session of Phase 2 and during the 1st hour of the dark phase. The second test session was conducted 7 days later and during the 12th hour of the dark phase. The in-between phase (Phase 3) was made up of 6 consecutive drinking sessions, the timing of which was changed daily and established semi-randomly from odd hours of the dark phase (sequence used: 5th, 11th, 3rd, 9th, 1st, and 7th hour). The marked
ACCEPTED MANUSCRIPT
consistency of sensitivity of alcohol drinking to time schedule, resulting in virtually completely superimposable intakes of alcohol even after repetition of several consecutive phases of alcohol drinking sessions with time of access to alcohol
RI PT
changing daily and being unpredictable to rats (this laboratory, unpublished results), allowed us to test diazepam in two separate drinking sessions occurring 1 week
apart; it was indeed ruled out that Phase 3 could alter basal alcohol drinking at the
SC
second test session with diazepam. In both test sessions with diazepam, rats were
divided into four groups of n = 8, matched for alcohol intake and schedule sensitivity
M AN U
over the preceding phase; additionally, in the second test, no rat was treated with the same diazepam dose received in the first test. Diazepam (injectable Valium®, Roche, Monza, Italy) was dissolved in a 1:1 mixture of saline and polyethylene glycol 600 and injected intraperitoneally (injection volume: 2 mL/kg) at the doses of 0, 1, 2, and 3 mg/kg, 30 min before the start of the drinking session.
TE D
After the second test session with diazepam, rats were exposed to 6 additional consecutive drinking sessions (Phase 4), the timing of which was changed daily and established semi-randomly from even hours of the dark phase (sequence
EP
used: 10th, 4th, 12th, 2nd, 6th, and 8th). The following day was devoted to testing the possible motor-incoordinating effects of diazepam at the inverted screen.
AC C
Specifically, rats were divided into four groups of n = 8, matched for alcohol intake and schedule sensitivity over Phase 4; additionally, no rat was treated with the same diazepam doses received in the two drinking sessions. Rats were treated with 0, 1, 2, and 3 mg/kg diazepam. Diazepam was dissolved and injected as described above. Thirty minutes after diazepam administration, rats were exposed to the inverted screen test. Specifically, each rat was placed individually in the center of the wire mesh screen. Immediately after rat placement, the screen was rotated from
ACCEPTED MANUSCRIPT
horizontal to vertical and then kept vertical. The time when the rat fell off the screen was recorded. If the rat did not fall off, it was removed after 120 sec. Rats were exposed to two different trials, occurring one after the other with no time interval in
RI PT
between. Tests at the inverted screen took place between the 6th and 7th hour of the dark phase; no drinking session was planned on the day in which the test at the inverted screen was conducted.
SC
In all drinking phases (Phases 1–4) and in the two test sessions with
diazepam, bottles were refilled daily with fresh solution and their positions changed
M AN U
randomly and daily to prevent development of position preference. Measured variables and statistical analysis
In all drinking phases (Phases 1–4) and in the two test sessions with diazepam, the measured variables were alcohol and water intake. These were monitored by weighing the bottles with a 0.01-g accuracy a) every day immediately
TE D
before the start of the dark phase in Phase 1, and b) immediately before and immediately after each daily drinking session in Phases 2–4 as well as in the two test sessions with diazepam. Possible fluid spillage was calculated by using multiple
EP
bottles filled with the different alcohol concentrations and positioned in empty cages interspersed in the cage racks; mean spilt volumes were subtracted before data
AC C
analysis. Alcohol and water intake were expressed in g/kg pure alcohol and mL/kg water, respectively. In the test sessions with diazepam, food intake was also recorded. Food pellets were weighed with a 0.01-g accuracy immediately before and immediately after the session. Food intake was expressed in g/kg. Data on alcohol and water intake over the 12 days of Phase 1 were analyzed by separate 1-way ANOVAs with repeated measures. Data on alcohol and water intake over the 12 drinking sessions of Phase 2, 6 drinking sessions of Phase 3, and 6 drinking
ACCEPTED MANUSCRIPT
sessions of Phase 4 were a) analyzed by separate 1-way ANOVAs with repeated measures and b) exposed to separate regression analyses (mean alcohol intake vs. time of the drinking session) and calculation of the Pearson correlation coefficient.
RI PT
Data on alcohol, water, and food intake from the test sessions with diazepam were analyzed by separate 2-way [time (1st or 12th hour); treatment (diazepam dose)] ANOVAs, followed by the Tukey’s test for post hoc comparisons.
SC
In the inverted screen test, the measured variable was the time (expressed in sec) spent on the screen before falling off. Rats that did not fall off were assigned the
M AN U
value of 120 sec. The average of values recorded in the two trials provided the value assigned to each rat. Data were statistically analyzed by 1-way ANOVA. Results Alcohol drinking in Phases 1–4
Phase 1 – ANOVA indicated highly significant differences in daily alcohol
TE D
intake during the 12 days of Phase 1 [F(11,341) = 15.06, p < 0.0001] (Fig. 2, left panel). All rats rapidly acquired alcohol-drinking behavior, as indicated by a mean daily intake of alcohol as high as approximately 3.8 g/kg on the initial day of exposure to
EP
the 4-bottle choice regimen; on the following days mean daily alcohol intake rose progressively, averaging around 6 g/kg over the last three days. ANOVA indicated
AC C
highly significant differences in daily water intake during the 12 days of Phase 1 [F(11,341) = 6.82, p < 0.0001] (Fig. 2, right panel). Mean daily water intake decreased progressively over the period. Phase 2 – ANOVA indicated highly significant differences in alcohol intake
during the 12 drinking sessions of Phase 2 [F(11,341) = 50.81, p < 0.0001] (Fig. 3, left panel). Specifically, mean alcohol intake varied largely, from a minimum value of 0.65 ± 0.05 g/kg, when the drinking session occurred during the 2nd hour of the dark
ACCEPTED MANUSCRIPT
phase, to a maximum value of 2.42 ± 0.09 g/kg, when the drinking session occurred during the 10th hour of the dark phase. Mean alcohol intake was highly positively correlated with time of access to alcohol (r = 0.906, slope = 0.127 ± 0.019,
RI PT
intercept = 0.696 ± 0.139, p < 0.0001). Mean water intake was negligible (<1 mL/kg) in each drinking session.
Phase 3 – ANOVA indicated highly significant differences in alcohol intake
SC
during the 6 drinking sessions of Phase 3 [F(5,155) = 76.67, p < 0.0001] (Fig. 3, center panel). Specifically, mean alcohol intake varied largely, from a minimum value of
M AN U
0.78 ± 0.05 g/kg, when the drinking session occurred during the 3rd hour of the dark phase, to a maximum value of 2.35 ± 0.09 g/kg, when the drinking session occurred during the 11th hour of the dark phase. Mean alcohol intake was positively correlated with time of access to alcohol (r = 0.972, slope = 0.156 ± 0.019, intercept = 0.538 ± 0.130, p < 0.005). Mean water intake was negligible (<1 mL/kg) in
TE D
each drinking session.
Phase 4 – ANOVA indicated highly significant differences in alcohol intake during the 6 drinking sessions of Phase 4 [F(5,155) = 38.51, p < 0.0001] (Fig. 3, right
EP
panel). Specifically, mean alcohol intake varied largely, from a minimum value of 0.84 ± 0.06 g/kg, when the drinking session occurred during the 2nd hour of the dark
AC C
phase, to a maximum value of 2.14 ± 0.10 g/kg, when the drinking session occurred during the 10th hour of the dark phase. Mean alcohol intake was positively correlated with time of access to alcohol (r = 0.870, slope = 0.104 ± 0.030, intercept = 0.859 ± 0.230, p < 0.05). Mean water intake was negligible (<1 mL/kg) in each drinking session. Effect of treatment with diazepam on alcohol drinking
ACCEPTED MANUSCRIPT
ANOVA indicated a highly significant effect of time of the drinking session (1st or 12th hour) [F(1,56) = 68.96, p < 0.0001], a significant effect of treatment with diazepam [F(3,56) = 3.99, p < 0.05], and a significant “time × treatment” interaction
RI PT
[F(3,56) = 5.53, p < 0.005] on alcohol intake (Fig. 4). In vehicle-treated rats, alcohol intake differed by approximately 3-fold when the drinking session occurred during the 1st (0.92 ± 0.18 g/kg) and 12th (2.72 ± 0.36 g/kg) hour of the dark phase
SC
(p < 0.0001, Tukey’s test). Post hoc analysis indicated that no dose of diazepam
altered alcohol intake when the drinking session occurred during the 1st hour of the
M AN U
dark phase (Fig. 4, left panel). Conversely, when the drinking session occurred during the 12th hour of the dark phase, treatment with 2 and 3 mg/kg diazepam resulted, in comparison to vehicle treatment, in statistically significant reductions in alcohol intake; the magnitude of these reductions averaged approximately 35% (p < 0.05, Tukey’s test) and 50% (p < 0.0005, Tukey’s test), respectively (Fig. 4, right
TE D
panel).
ANOVA indicated a significant effect of time of the drinking session [F(1,56) = 5.33, p < 0.05], no effect of treatment with diazepam [F(3,56) = 1.01,
EP
p > 0.05], and no “time × treatment” interaction [F(3,56) = 1.40, p > 0.05] on water intake (Table 1). ANOVA indicated no effect of time of the drinking session
AC C
[F(1,56) = 0.22, p > 0.05], no effect of treatment with diazepam [F(3,56) = 0.55, p > 0.05], and no “time × treatment” interaction [F(3,56) = 1.23, p > 0.05] on food intake (Table 1).
Effect of treatment with diazepam on motor performance Treatment with diazepam failed to alter rat motor-performance, as indicated by the lack of statistical difference – among the four rat groups – in time spent on the inverted screen [F(3,28) = 0.08, p > 0.05] (Table 2).
ACCEPTED MANUSCRIPT
Discussion Data on alcohol intake from Phases 2–4 confirmed the sensitivity to time schedule of alcohol drinking in sP rats exposed to daily 1-h drinking sessions, during
RI PT
the dark phase of the light/dark cycle, with concurrent availability of multiple alcohol concentrations, and unpredictable access to alcohol. Alcohol intake highly positively correlated with the time of alcohol access, and increased progressively from an
average of 0.6–0.8 g/kg, when the drinking session occurred during the first hours, to
SC
an average of 2.1–2.4 g/kg, when the drinking session occurred during the later
hours. Similar data were collected in the two test sessions with diazepam, in which
M AN U
alcohol intake in vehicle-treated (control) rats averaged approximately 0.9 and 2.7 g/kg at the 1st and 12th hour of the dark phase, respectively. Together, these data closely replicate those collected in previous studies (Colombo et al., 2014; 2015) and confirm that this experimental procedure may result in sP rats consuming
TE D
exceptionally high amounts of alcohol [meeting the criteria posed for binge-like drinking in animal models (Bell, Rodd, Lumeng, Murphy, & McBride, 2006)]. The similarity of values of alcohol intake recorded in Phases 2, 3, and 4 (occurring
EP
sequentially, one after another) confirmed that – once developed – sensitivity of alcohol drinking to time schedule remained stable over time.
AC C
The experiment testing diazepam provided some interesting information. First, treatment with diazepam was totally ineffective on alcohol intake when the drinking session occurred at the 1st hour of the dark phase: no tested dose of diazepam altered alcohol intake. The tested dose range of diazepam (1–3 mg/kg; i.p.) was likely adequate and sufficiently wide, as higher doses would have resulted in sedative and motor-incapacitating effects, affecting the regular drinking rates of rats; additionally, it virtually overlapped with those used in several previous studies investigating the effect of diazepam treatment on alcohol-motivated behaviors in rats
ACCEPTED MANUSCRIPT
(e.g., Daoustet et al., 1987; Janak, Redfern, & Samson, 1998; Rimondini, Sommer, & Heilig, 2002; Shelton & Balster, 1997). Our initial investigation on the drinking procedure employed in the present
RI PT
study found that alcohol intake in sP rats was modestly sensitive to unpredictability of time of alcohol access when the drinking session occurred during the first hours of the dark phase: in this time period, alcohol intake was indeed found to be highly
SC
similar under “unpredictable” and “fixed” conditions (the latter comprising daily drinking sessions taking place constantly at the 1st hour of the dark phase)
M AN U
(Colombo et al., 2014). Thus, alcohol intake at the 1st hour of the dark phase is minimally influenced by unpredictability of time of alcohol access and can be equated to that recorded under more standard, or basic, conditions of experimental alcohol drinking in rats. Based on this premise, the observed failure of diazepam treatment to alter alcohol drinking in sP rats exposed to alcohol at the 1st hour of the
TE D
dark phase was not entirely unexpected, as acute treatment with dose ranges of diazepam highly comparable to that used in the present study has repeatedly been reported to be ineffective on alcohol intake and lever-responding for alcohol in rats
EP
exposed to conventional procedures of alcohol drinking and operant, oral alcohol self-administration (e.g., Daoust et al., 1987; Janak et al, 1998; Rimondini et al.,
AC C
2002; Shelton & Balster, 1997). Conversely, treatment with diazepam effectively reduced the exceptionally
high alcohol intake recorded when the drinking session occurred during the 12th hour of the dark phase. In comparison to the vehicle-treated rat group, alcohol intake decreased by approximately 25%, 35%, and 50% in the rat groups treated with 1, 2, and 3 mg/kg diazepam, respectively (although only reduction exerted by the two highest doses of diazepam reached statistical significance at post hoc
ACCEPTED MANUSCRIPT
analysis). Notably, the reducing effect of diazepam was selective for alcohol intake, as treatment with diazepam did not affect water and food intake. While data on water intake might be of limited usefulness (water intake was extremely low, making its
RI PT
analysis virtually meaningless in terms of evaluation of the selectivity of diazepam action), data on food intake clearly demonstrate that feeding was totally unaffected by diazepam treatment. Additionally, treatment with diazepam failed to alter rat
SC
performance at a highly challenging task (in terms of muscle strength and motor
coordination), such as hanging or moving on the inverted screen. These data confer
M AN U
specificity to the reducing effect of diazepam on alcohol intake, as they suggest that no sedative or motor-incoordinating effect of diazepam hampered the regular rates of drinking and feeding.
The preferential effectiveness of diazepam in reducing alcohol intake when the drinking session occurred at the 12th, rather than 1st, hour of the dark phase
TE D
provides support to the working hypothesis of the present study: uncertainty of time of alcohol access and expectation of alcohol availability, i.e., the peculiar features of the procedure of alcohol drinking used in the present study, added an emotional
EP
“distress” to the inherent, basal anxiety-like state of sP rats (see Colombo, Lobina, Carai, & Gessa, 2006); sP rats apparently cope with this exacerbated, negative
AC C
affective state by consuming larger amounts of alcohol, likely seeking its anxiolytic effects. Accordingly, we hypothesize that the anxiolytic effects of diazepam substituted for those of alcohol, resulting in a less urgent need to consume alcohol and, in turn, in the observed, remarkable reduction in alcohol intake. To summarize, the results of the diazepam experiment (present study) complement those of the recent Social Interaction experiment (Colombo et al., 2015) demonstrating that sP rats with a “history” of unpredictable exposures to daily
ACCEPTED MANUSCRIPT
alcohol drinking sessions displayed more anxiety-related behaviors over the later rather than first hours of the dark phase. Taken together, these results provide an “anxiety”-founded hypothesis to the neurobiological bases of sensitivity to time
RI PT
schedule and excessive alcohol intake in sP rats exposed to this new experimental procedure of alcohol drinking. Combination of inherently “anxious” sP rats with this “anxiogenic” procedure of alcohol drinking apparently opens up a rat model of
SC
potential utility for investigations of comorbid anxiety and alcohol-use disorder. Acknowledgments
M AN U
The authors are grateful to Ms. Anne Farmer for language editing of the manuscript and Mr. Alessandro Capra for technical support. Conflict of Interest None. Funding
TE D
This research did not receive any specific grant from funding agencies in the
AC C
EP
public, commercial, or not-for-profit sectors.
ACCEPTED MANUSCRIPT
References Bell, R. L., Rodd, Z. A., Lumeng, L., Murphy, J. M., & McBride, W. J. (2006). The alcohol-preferring P rat and animal models of excessive alcohol drinking. Addiction Biology, 11, 270–288. doi: 10.1111/j.1369-1600.2005.00029.x
RI PT
Colombo, G., Lobina, C., Carai, M. A., & Gessa, G. L. (2006). Phenotypic characterization of genetically selected Sardinian alcohol-preferring (sP) and -non-preferring (sNP) rats. Addiction Biology, 11, 324–338. doi: 10.1111/j.1369-1600.2006.00031.x
Colombo, G., Lobina, C., Maccioni, P., Carai, M. A.., Lorrai, I., Zaru, A., et al. (2015). Anxiety-like behaviors at the end of the nocturnal period in sP rats with a “history” of unpredictable, limited access to alcohol. Alcohol, 49, 707–712. doi: 10.1016/j.alcohol.2015.04.010
SC
Colombo, G., Maccioni, P., Acciaro, C., Lobina, C., Loi, B., Zaru, A., et al. (2014). Binge drinking in alcohol-preferring sP rats at the end of the nocturnal period. Alcohol, 48, 301–311. doi: 10.1016/j.alcohol.2014.02.001
M AN U
Coughenour, L. L., Mclean, J. R., & Parker, R. B. (1977). A new device for the rapid measurement of impaired motor function in mice. Pharmacology, Biochemistry, and Behavior, 6, 351–353. Daoust, M., Saligaut, C., Lhuintre, J. P., Moore, N., Flipo, J. L., & Boismare, F. (1987). GABA transmission, but not benzodiazepine receptor stimulation, modulates ethanol intake by rats. Alcohol, 4, 469–472.
TE D
File, S. E., Kenny, P. J., & Ouagazzal, A. M. (1998). Bimodal modulation by nicotine of anxiety in the social interaction test: role of the dorsal hippocampus. Behavioral Neuroscience, 112, 1423–1429. File, S. E., Lippa, A. S., Beer, B., & Lippa, M. T. (2004). Animal tests of anxiety. Current Protocols in Neuroscience, 8, 8.3. doi: 10.1002/0471142301.ns0803s26
EP
Hasin, D. S., Stinson, F. S., Ogburn, E., & Grant, B. F. (2007). Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Archives of General Psychiatry, 64, 830–842. doi: 10.1001/archpsyc.64.7.830
AC C
Janak, P. H., Redfern, J. E., & Samson, H. H. (1998). The reinforcing effects of ethanol are altered by the endogenous neurosteroid, allopregnanolone. Alcoholism: Clinical and Experimental Research, 22, 1106–1112. Maccioni, P., Lorrai, I., Marras, M. F., Contini, A., Capra, A., Piras, G., et al. (2015). Elevated reinforcing and motivational properties of alcohol at the end of the nocturnal period in sP rats. Psychopharmacology, 232, 3585–3595. doi: 10.1007/s00213-015-4010-2 Rimondini, R., Sommer, W., & Heilig, M. (2002). Effects of tiagabine and diazepam on operant ethanol self-administration in the rat. Journal of Studies on Alcohol, 63, 100–106. Shelton, K. L., & Balster, R. L. (1997). Effects of gamma-aminobutyric acid agonists and N-methyl-D-aspartate antagonists on a multiple schedule of ethanol and
ACCEPTED MANUSCRIPT
saccharin self-administration in rats. The Journal of Pharmacology and Experimental Therapeutics, 280, 1250–1260.
AC C
EP
TE D
M AN U
SC
RI PT
Vorspan, F., Mehtelli, W., Dupuy, G., Bloch, V., & Lépine, J. P. (2015). Anxiety and substance use disorders: co-occurrence and clinical issues. Current Psychiatry Reports, 17, 4. doi: 10.1007/s11920-014-0544-y
ACCEPTED MANUSCRIPT
Figure captions Fig. 1. Schematic representation of the experimental design.
RI PT
Fig. 2. Alcohol (left panel) and water (right panel) intake in Sardinian alcoholpreferring (sP) rats exposed to the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen, with unlimited access (24 h/day), for 12 consecutive days. Each point is the mean ± SEM of n = 32 rats.
SC
Fig. 3. Alcohol intake in Sardinian alcohol-preferring (sP) rats exposed to 12 (Phase 2; left panel), 6 (Phase 3; center panel), and 6 (Phase 4; right panel) consecutive daily 1-h drinking sessions under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen. Drinking sessions occurred during one of the 12 hours of the dark phase of the daily light/dark cycle; time of the drinking session was changed daily in a semi-random order and was unpredictable to rats. Each point is the mean ± SEM of n = 32 rats.
AC C
EP
TE D
M AN U
Fig. 4. Effect of acute treatment with different doses of diazepam on alcohol intake in Sardinian alcohol-preferring (sP) rats exposed to 1-h drinking sessions, under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs. water” choice regimen with unpredictable access, conducted at the 1st (left panel) and 12th (right panel) hour of the dark phase. Diazepam was administered intraperitoneally 30 min before the start of the drinking session. Each bar is the mean ± SEM of n = 8 rats. : p < 0.005 and : p < 0.0001 in comparison to the rat group treated with the same diazepam dose at the 1st hour (Tukey’s test); §: p < 0.05 and §§: p < 0.0005 in comparison to the rat group treated with vehicle (0 mg/kg diazepam) at the 12th hour (Tukey’s test).
ACCEPTED MANUSCRIPT
Table 1 – Effect of acute treatment with different doses of diazepam on water and food intake in Sardinian alcohol-preferring (sP) rats exposed to 1-hour drinking sessions, under the 4-bottle “alcohol (10%, 20%, and 30%; v/v) vs water” choice regimen with
RI PT
unpredictable access, conducted at the 1st and 12th hour of the dark phase.
Diazepam (mg/kg)
Food intake (g/kg)
1
1st Hour
0.29 ± 0.17
2.11 ± 1.30
12th Hour
0.04 ± 0.03
0.06 ±0.04
1st Hour
3.54 ± 0.66
4.30 ± 0.91
12th Hour
4.82 ± 1.19
5.11 ± 1.09
2
3
3.42 ± 1.72
1.29 ± 1.09
0.04 ±0.03
1.03 ± 0.86
7.16 ± 1.50
4.88 ± 1.18
4.18 ± 1.42
4.13 ± 1.56
SC
Water intake (ml/kg)
0
M AN U
Diazepam was administered intraperitoneally 30 min before the start of the drinking
AC C
EP
TE D
session. Each value is the mean ± SEM of n=8 rats.
1
ACCEPTED MANUSCRIPT
Table 2 – Effect of acute treatment with different doses of diazepam on motor performance, evaluated by the inverted screen test, in Sardinian alcohol-preferring (sP) rats previously exposed to several drinking sessions of 1 hour, under the 4-bottle “alcohol
RI PT
(10%, 20%, and 30%, v/v) vs water” choice regimen and unpredictable access.
Diazepam (mg/kg) 0 Time (in s) spent on the inverted screen before falling off
1
79.3 ± 9.5
2
3
83.5 ± 10.2 76.8 ± 14.7 75.6 ± 14.7
SC
On the test day, no drinking session took place. Thirty min after the intraperitoneal administration of diazepam, each rat was placed individually in the center of the wire mesh
M AN U
screen. The screen was rotated 90° (from horizontal to vertical) over 5 s at constant speed. The time (in s) when the rat fell off the screen was recorded. If the rat did not fall off, it was removed after 120 s and assigned the value of 120 s. Rats were exposed to two different trials, occurring one after the other with no time interval in between. The average of values recorded in the two trials provided the value assigned to each rat. Each value is
AC C
EP
TE D
the mean ± SEM of n=8 rats.
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights • Alcohol drinking in sP rats was highly sensitive to time schedule of alcohol access. • Exceptionally high alcohol intakes were recorded at the last hours of the dark phase. • Alcohol expectation produced an emotional “distress” (anxiety-related behaviors).
RI PT
• The anxiolytic compound, diazepam, effectively reduced excessive alcohol intake. • The anxiolytic effects of diazepam likely substituted for those of alcohol.
AC C
EP
TE D
M AN U
SC
Keywords: time schedule of alcohol drinking; limited and unpredictable access to alcohol; experimental model of binge drinking; anxiety-like behaviors; diazepam; Sardinian alcohol-preferring (sP) rats