Alcohol xxx (2015) 1e9

Contents lists available at ScienceDirect

Alcohol journal homepage: http://www.alcoholjournal.org/

Chronobiology of ethanol: Animal models Alan M. Rosenwasser a, b, c, * a

Department of Psychology, University of Maine, USA School of Biology and Ecology, University of Maine, USA c Graduate School of Biomedical Science and Engineering, University of Maine, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 March 2015 Received in revised form 10 April 2015 Accepted 10 April 2015

Clinical and epidemiological observations have revealed that alcohol abuse and alcoholism are associated with widespread disruptions in sleep and other circadian biological rhythms. As with other psychiatric disorders, animal models have been very useful in efforts to better understand the cause and effect relationships underlying the largely correlative human data. This review summarizes the experimental findings indicating bidirectional interactions between alcohol (ethanol) consumption and the circadian timing system, emphasizing behavioral studies conducted in the author’s laboratory. Together with convergent evidence from multiple laboratories, the work summarized here establishes that ethanol intake (or administration) alters fundamental properties of the underlying circadian pacemaker. In turn, circadian disruption induced by either environmental or genetic manipulations can alter voluntary ethanol intake. These reciprocal interactions may create a vicious cycle that contributes to the downward spiral of alcohol and drug addiction. In the future, such studies may lead to the development of chronobiologically based interventions to prevent relapse and effectively mitigate some of the societal burden associated with such disorders. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Alcohol Ethanol Addiction Circadian Chronobiology Animal models

Introduction Chronic alcoholism is associated with dramatic disruptions in sleep and other circadian biological rhythms (e.g., Brower, 2001; Conroy et al., 2012; Kuhlwein, Hauger, & Irwin, 2003). These disruptions can sometimes persist over extended periods of abstinence, and appear to increase the risk of relapse (Brower, 2003; Landolt & Gillin, 2001). Similarly, chronobiological disruption may be a risk factor for excessive alcohol intake in non-dependent populations, such as adolescents, shift-workers, and frequent transmeridian travelers (Rogers & Reilly, 2002; Trinkoff & Storr, 1998; see also Hasler, Soehner, & Clark, 2015). Together, these clinical and epidemiological observations suggest that linkages between circadian disruption and excessive drinking reflect bidirectional causal interactions (Danel & Touitou, 2004; Hasler, Smith, Cousins, & Bootzin, 2012; Rosenwasser, 2001; Spanagel, Rosenwasser, Schumann, & Sarkar, 2005). Such reciprocal interactions could lead to a vicious cycle phenomenon and possibly contribute to the downward spiral from alcohol use to abuse to addiction. * Corresponding author. Department of Psychology, University of Maine, 5742 Little Hall, Orono, ME 04469, USA. Tel.: þ1 207 581 2035; fax: þ1 207 581 6128. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.alcohol.2015.04.001 0741-8329/Ó 2015 Elsevier Inc. All rights reserved.

Of course, it is very difficult to infer causal relationships from clinical and epidemiological data, and even more difficult to isolate and identify controlling variables. Thus, as is the case for other psychiatric disorders, animal models have proven to be extremely useful in this regard. As described in this review, substantial data have now accumulated to indicate that alcohol (ethanol) alters fundamental properties of the circadian clock. In turn, disruptions or alterations in the circadian system induced by either environmental or genetic manipulations can alter voluntary ethanol intake in experimental animals. This review summarizes the experimental findings supporting these conclusions, emphasizing studies performed in the author’s laboratory. In the first two sections of the review, I describe the evidence that alcohol consumption and/or administration alters both the endogenous free-running period of the circadian clock as well as its phase-shifting response to environmental light signals. These studies demonstrate that alcohol acts directly on the circadian clock system to affect circadian rhythms. Next, I discuss the evidence that manipulation of the circadian clock via exposure to atypical environmental lighting regimens can alter voluntary alcohol intake. While the results of such studies are complex and include some inconsistent findings, they do demonstrate that chronobiological “stressors” may influence alcohol consumption. Finally, I review the evidence for reciprocal genetic linkages

2

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

between the circadian rhythmicity and alcohol intake, and show that circadian clock genes influence alcohol drinking, while in turn, genes associated with ethanol responsiveness contribute to circadian regulation. Effects of chronic ethanol treatment on free-running circadian period My laboratory initially became interested in the chronobiology of ethanol because of a fortuitous observation made in the course of an experiment conducted to determine the effects of neonatal treatment with the serotonin-selective tricyclic antidepressant, clomipramine, on free-running circadian rhythms in adult Wistar rats (Dwyer & Rosenwasser, 1998). In this and other experiments, we frequently examine free-running period because it is thought to reflect a quantitative property of the underlying circadian pacemaker that can be evaluated at the level of physiology and behavior. Neonatal clomipramine treatment was known to produce a spectrum of affective-behavioral and neurobiological effects in adulthood, including increasing ethanol preference in 2-bottle freechoice drinking tests. As a manipulation check, we offered the animals free access to 10% v/v ethanol and plain water for several weeks under constant darkness while monitoring free-running circadian drinking rhythms via lickometers. Both male and female rats showed significant shortening of free-running period during ethanol access, and a gradual return toward baseline after termination of ethanol access. At that time, only one previously published study had examined the effects of ethanol on circadian rhythms in a whole-animal model (Mistlberger & Nadeau, 1992). In that experiment, male hamsters were housed in running-wheel cages and maintained on either 28% ethanol or plain water as their sole drinking fluid. While no effects of ethanol treatment were seen under steady-state lightedark (LD) entrainment or during re-entrainment following LD phase-shifts, ethanol treatment significantly lengthened freerunning circadian period under constant dim light, opposite in direction to the effect observed by Dwyer and Rosenwasser (1998). While these two studies differed in choice of species, housing conditions, light intensities, ethanol concentrations, and ethanol access conditions, it has continued to be the case that the reported effects of ethanol on free-running period are subtle and somewhat variable between and within experiments. As discussed elsewhere, similar variability also characterizes the reported effects of antidepressants and other mood-altering drugs on free-running period (Klemfuss & Kripke, 1994; Rosenwasser, 2001; Wollnik, 1992). In a subsequent experiment (Rosenwasser, Fecteau, & Logan, 2005), we maintained male Long-Evans rats in running-wheel cages and provided either plain water or 10% or 20% ethanol as the sole drinking fluid. Free-running period was evaluated before, during, and following 3e5 weeks of ethanol treatment. While most animals showed discernible alterations in free-running period during ethanol treatment relative to baseline conditions, responses were idiosyncratic and included clear examples of both period shortening and period lengthening. Further, ethanol-induced period change was significantly correlated with individual differences in baseline period, such that rats with the shortest baseline periods displayed the most robust period lengthening under ethanol. Finally, when ethanol treatment was terminated, most animals in the 10% ethanol treatment group showed a return toward baseline periods, as in Dwyer and Rosenwasser (1998), but many animals in the 20% ethanol group actually showed an exacerbation of the original ethanol effect during ethanol “withdrawal.” In a later experiment, we examined the effects of both freechoice and forced intake of 10% ethanol on free-running circadian period in male C57BL/6 mice housed in running wheels

Fig. 1. Mean (SEM) free-running circadian period in male C57BL/6 mice housed in continuous darkness. Water-only mice were maintained on plain water and served as controls; free-choice ETOH mice were allowed continuous access to plain water and 10% v/v ethanol solution via separate drinking tubes; forced ETOH mice had 10% ethanol solution as their only drinking fluid. B indicates a 3-week baseline during which all groups had only water available; E1eE7 indicate successive 3-week epochs during which ethanol was presented to the relevant groups while the controls were continued on plain water only. Modified from Seggio et al., 2009.

(Seggio, Fixaris, Reed, Logan, & Rosenwasser, 2009). This experiment employed a between-groups design, in which the forced- and free-choice ethanol groups were maintained on ethanol for 21 weeks, while a control group was maintained on plain water. Forced ethanol intake resulted in a persistent shortening of free-running period relative to controls, but no significant effect was seen in the free-choice group (Fig. 1). Most recently, we monitored free-running circadian period under long-term ethanol access in male and female rats of the selectively bred high-drinking P (“Preferring”) and HAD2 (“High Alcohol Drinking, replicate 2”) lines (Rosenwasser, McCulley, & Fecteau, 2014). In this experiment, animals were maintained on either continuous free-choice access to 10% ethanol or on an intermittent ethanol-access schedule consisting of two weeks of access alternating with two weeks of ethanol deprivation. While this study revealed subtle effects of sex, genotype, and access schedule, the major effect was that animals displayed shortening of free-running period during ethanol access. While animals consumed considerable amounts of ethanol via their drinking water in the studies just discussed, there is little evidence that intoxication or dependence occur under such conditions, even in high-drinking genotypes. In order to focus on possible effects of ethanol dependence and withdrawal on circadian period, we assessed circadian activity rhythms in male C57BL/6 and C3H/He mice following chronic-intermittent exposure to ethanol vapor using a protocol known to induce signs of dependence (Logan, McCulley, Seggio, & Rosenwasser, 2012; Logan, Seggio, Robinson, Richard, & Rosenwasser, 2010). In these experiments, we employed both 4-day and 16-day vapor exposure regimens, but significant shortening of free-running period was seen only in C3H mice, and only following the longer exposure regimen. Taken together, these experiments indicate that chronic ethanol exposure can modify free-running circadian period. Nevertheless, these effects are typically modest, and may vary by species and strain. Further, while most of our studies revealed ethanol-induced shortening of circadian period, examples of period lengthening were also observed (Rosenwasser, Fecteau, & Logan, 2005). Finally, it appears that the most consistent effects are seen during and following ethanol treatments that expose animals to levels of

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

ethanol beyond those typically achieved in free-choice preference drinking. Thus, such effects could underlie in part the chronobiological disruptions seen in human alcoholism. Effects of chronic and acute ethanol treatment on light-induced circadian phase shifting Like free-running circadian period, the phase-shifting response to an acute photic stimulus (light “pulse”) provides an opportunity to infer the quantitative properties of the underlying circadian pacemaker through behavioral-level measurements. In contrast to studies of free-running period, however, analysis of ethanol’s effects on photic phase-shifting have consistently revealed a reduced pacemaker response to light pulses, both during chronic ethanol exposure and following acute ethanol pretreatment. My laboratory has examined the effects of chronic ethanol intake on photic phase shifting in rats, hamsters, and mice (Rosenwasser, Logan, & Fecteau, 2005; Seggio et al., 2009; Seggio, Logan, & Rosenwasser, 2007). These experiments all employed a very similar design, in which individual animals were tested repeatedly for phase-shift responses to brief light pulses at different circadian phases using the well-established Aschoff Type 2 protocol. In this protocol, phase-shift stimuli are presented on day 1 of constant darkness, immediately following stable lightedark entrainment (an illustrative example of the protocol is shown in Fig. 2). Light pulses were presented during both early and late subjective night (ZT 14 or 15 and ZT 21, respectively; by convention, ZT 0 is defined as the time of lights-on under the prior lightedark cycle), and phase-shift responses were compared to those seen in a no-pulse control condition.

Fig. 2. Representative rasterestyle activity records illustrating the Aschoff Type 2 phase-shift protocol in a single control and a single ethanol-drinking mouse. In these records, time of day is represented along the Xeaxis, and successive days are represented from top to bottom along the Yeaxis. For each 10-min epoch, a bar is plotted with size proportional to the amount of activity occurring in that time bin, resulting in a graphic representation of the daily activity distribution. In this protocol, animals are maintained under stable lightedark (LD) entrainment for at least one week, after which conditions are changed to constant darkness (DD) for a minimum of one week. At various times after the beginning of DD, animals are exposed to a single brief light pulse (yellow star). The time of the light pulse is referred to as “Zeitgeber Time” (ZT), and by convention, the time of the preceding light-to-dark transition is referred to as ZT 12. Thus, in these examples, the light pulse was presented 3 h after the preceding light-to-dark transition, at ZT 15. Phase shifts are evaluated by fitting a straight line to the last 7 activity onsets preceding the light pulse (red lines) and another line to the 7 activity onsets following the light pulse (blue line), and then comparing the phase of these two lines on the day after the light pulse. In these examples, animals showed phase delays (i.e., activity onsets following the light pulse occurred later than would have been predicted by the pre-pulse line), as expected for light pulses given at ZT 15. At other pulse times (e.g., ZT 21), phase advances may be seen (not shown). Note also that the ethanol animal in these examples displayed a smaller light-induced phase delay than the control mouse. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Modified from Seggio et al., 2009.

3

In our initial study (Rosenwasser, Logan, et al., 2005), male LongEvans rats were maintained in running-wheel cages with 20% ethanol or plain water as their sole drinking fluid. Unfortunately, however, variability in daily activity onsets prevented the accurate assessment of circadian phase, and we instead relied on measurement of a correlated variable, light-induced period change. While significant period responses were not seen in either ethanolexposed or control animals following light pulses given at ZT 14, control animals but not ethanol-treated animals showed significant period shortening following light pulses at ZT 21. Thus, chronic ethanol intake appeared to block the circadian pacemaker’s response to a brief light pulse during late subjective night. In a subsequent study (Seggio et al., 2007), male Syrian hamsters were maintained on free-choice access to 25% ethanol solution or plain water. Hamsters are known to show very high levels of voluntary ethanol intake, and consume approximately 80% of their daily fluid intake from the ethanol bottle at this concentration. Relative to a no-pulse control condition, control animals showed the expected phase delays to light pulses at ZT 14 and phase advances to pulses at ZT 21. In contrast, however, while ethanolexposed animals showed normal phase-delays at ZT 14, phase advances at ZT 21 were completely blocked by chronic ethanol intake, as was seen in rats (Fig. 3). In addition, very similar effects were reported by others (Ruby, Brager, DePaul, Prosser, & Glass, 2009; see Prosser & Glass, 2015), who found attenuated phase shifting during late subjective night following chronic forced ethanol intake in hamsters. Finally, we examined the effects of chronic free-choice and forced 10% ethanol intake in male C57BL/6 mice (Seggio et al., 2009). Similar to the findings described above for free-running period, no ethanol effects were seen during free-choice drinking, but forced ethanol intake significantly attenuated the phaseshifting effects of light pulses. In contrast to our studies with rats and hamsters, however, ethanol attenuated light-induced phase shifting only at ZT 15 and not at ZT 21 in mice (Fig. 4). It must be noted here that rats and hamsters typically show larger phase-shift responses to light pulses during late subjective night, while mice typically show more robust phase-shift responses during early subjective night. Thus, all three species showed ethanol-induced attenuation of photic phase shifting at their more

Fig. 3. Mean (þSEM) circadian phase shifts in male Syrian hamsters induced by 15-min, 50-lux light pulses delivered during either early (ZT 14) or late (ZT 21) subjective night using the Aschoff Type 2 protocol, as described fully in the caption to Fig. 2. In this experiment, an additional group of animals (“No-Pulse”) was tested to control for possible phase-shifting effects of transfer from LD to DD. Individual animals either had continuous free-choice ethanol access or served as water-only controls. For other details, see text. Modified from Seggio et al., 2007.

4

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

administered either via intraperitoneal injection or directly to the suprachiasmatic nucleus (SCN), site of the mammalian circadian pacemaker (hamster: Ruby, Prosser, DePaul, Roberts, & Glass, 2009; mouse: Brager, Ruby, Prosser, & Glass, 2011). Further, acute ethanol application to the SCN in slice culture attenuates the phase-shifting effect of glutamate treatment, an established in vitro probe for the light entrainment pathway (Prosser, Mangrum, & Glass, 2008). Additional details on these studies can be found in Prosser and Glass (2015). To summarize, chronic and acute ethanol treatments consistently attenuate the phase-shifting effects of light on the circadian pacemaker. Along with the alterations in pacemaker period described in the previous section, such effects could also contribute to the chronobiological disruptions associated with clinical alcoholism.

1.5

Phase shift (h)

1.0

Control Free-choice EtOH

0.5

0.0

-0.5

-1.0

-1.5

ZT15

ZT21

1.5

Phase shift (h)

1.0

Control Forced EtOH

0.5

0.0

-0.5

-1.0

*

-1.5

ZT15

ZT21

Fig. 4. Mean (þSEM) circadian phase shifts in male C57BL/6 mice induced by 15-min, 50-lux light pulses delivered during either early (ZT 15) or late (ZT 21) subjective night using the Aschoff Type 2 protocol, as described fully in the caption to Fig. 2. In separate experiments, water-only control mice were compared either to mice with free-choice ethanol access (top panel) or to mice with ethanol solution as their only available drinking fluid (bottom panel). For other details, see text. Modified from Seggio et al., 2009.

light-responsive but not at their less light-responsive phase. Again, similar effects were also seen by others (Brager, Ruby, Prosser, & Glass, 2010; see Prosser & Glass, 2015), who found attenuation of photic phase shifting during early subjective night following chronic forced ethanol intake in mice. The attenuation of light-induced circadian phase shifting seen during chronic ethanol intake could reflect ethanol-induced neuroadaptive changes in the photic entrainment pathway. If so, one might expect that such attenuation would persist e at least temporarily e after discontinuation of ethanol intake. While this question has not been well studied, there is limited evidence to indicate that attenuation of photic phase-shifting persists for at least a few days after termination of ethanol availability in both hamsters and mice (Brager et al., 2010; Seggio et al., 2009), and that phase-shifting may in fact be potentiated in ethanol-withdrawn hamsters (Ruby, Brager, et al., 2009). Similar attenuation of light-induced phase-shifting has also been observed in response to acute ethanol pre-treatment,

Effects of environmental lighting on voluntary ethanol intake During entrainment to an environmental lightedark (LD) cycle, the circadian pacemaker maintains a stable phase relationship to the LD cycle (external synchrony), while multiple rhythmic processes within the organism maintain appropriate internal phase relationships to one another (internal synchrony). Recent research indicates that disruption of these phase relationships can promote pathophysiological changes, potentially leading to disease. While much of this research has focused on medical diseases such as diabetes and cancer, there is also considerable evidence that internal and/or external circadian desynchrony are implicated in psychiatric disorders, especially depression (for recent reviews, see McClung, 2013; Schnell, Albrecht, & Sandrelli, 2014). Perhaps the most direct way to induce circadian disruption and desynchrony is via manipulation of the environmental lighting regimen. Several different manipulations have been employed, including constant darkness (DD) and constant light (LL), non-24-h LD cycles, long and/or short photoperiods, and repeated LD phase shifts mimicking the light exposure patterns associated with jet lag or shift work (often referred to as “shift-lag” in the animal literature). Indeed, recent studies indicate that such non-standard lighting schedules can alter affective behavior in a variety of animal models. While a detailed analysis of these effects is beyond the scope of the present paper, the reader is referred to several recent reviews of this literature (Bedrosian & Nelson, 2013; Landgraf, McCarthy, & Welsh, 2014b; LeGates, Fernandez, & Hattar, 2014; Stephenson, Schroder, Bertschy, & Bourgin, 2012). Instead, this section will describe the rather more limited literature on the effects of atypical lighting schedules on voluntary ethanol intake. Early studies reported increases in voluntary ethanol intake under DD relative to standard LD entrainment in both rats (Geller, 1971) and hamsters (Geller & Hartmann, 1977; Reiter, Blum, Wallace, & Merritt, 1974), while studies of pinealectomized and melatonin-treated animals suggested that these effects could be due to DD-induced increases in melatonin secretion (Burke & Kramer, 1974; Reiter, Blum, Wallace, & Merritt, 1973; Reiter et al., 1974). In contrast, however, Goodwin, Amir, and Amit (1999) found reduced ethanol intake in rats under both DD and LL. More recently, Trujillo, Do, Grahame, Roberts, and Gorman (2011) also reported reduced ethanol intake under DD in selectively bred HAP2 and LAP2 (“High Alcohol Preferring” and “Low Alcohol Preferring”, replicate 2) mice, while two recent studies failed to detect significant effects of LL on ethanol intake in hamsters (Hammer, Ruby, Brager, Prosser, & Glass, 2010; Ruby et al., 2014). Although the results of Goodwin et al. (1999) and Trujillo et al. (2011) appear inconsistent with earlier studies of ethanol intake in DD, these findings do resemble recent studies showing that both DD and LL

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

can induce depression-like behaviors (e.g., LeGates et al., 2014). We have recently examined the effects of DD and LL on voluntary ethanol intake in several inbred mouse strains. In these experiments, male C57BL/6 and C3H/He mice showed reduced ethanol intake under both DD and LL relative to LD entrainment, while DBA/2 mice showed reduced intake in DD but not in LL (Rosenwasser & Fixaris, 2013; Rosenwasser & Fixaris, unpublished data). In another experiment, we found reduced ethanol intake under both DD and LL in male and female mice of the genetically heterogeneous WSC-2 line, originally derived from an 8-way cross among inbred strains (unpublished data). Thus, our data in mice are generally consistent with the study in rats by Goodwin et al. (1999), as well as with studies showing similar depressogenic effects of DD and LL (LeGates et al., 2014). It is not at all clear, however, how to account for the similarity of the effects seen under both DD and LL. Thus, while LL is known to engender circadian disruption and desynchrony, robust and coherent free-running rhythmicity is typically seen under DD. In addition, while LL is known to suppress melatonin secretion, this is not seen in DD. Further arguing against a critical role of melatonin, reduced ethanol intake under DD and LL is seen in both “melatonindeficient” C57BL/6 mice and “melatonin-proficient” C3H/He mice. Finally, DD and LL typically have opposite effects on photoperiodic responses such as those of the reproductive and immune systems. One possibility is that these effects are due to the lack of stable circadian entrainment that occurs under both DD and LL conditions. Another reasonable hypothesis is that housing under either DD or LL is generally stressful, relative to standard LD conditions. Of course, it is also possible that the effects of DD and LL on ethanol intake (and affective behaviors more generally) are mediated via partially or entirely distinct mechanisms. For example, DD could affect ethanol intake via a photoperiodic mechanism, while LL could produce superficially similar effects on ethanol intake via its ability to disrupt circadian rhythms. In addition to studies employing DD and/or LL, a few experiments have examined the effects of photoperiod duration within a 24-h LD cycle on voluntary ethanol intake. In general, “winter-like” short photoperiods result in increased depression- and anxiety-like behaviors while “summer-like” long photoperiods may engender antidepressant-like responses, relative to standard LD 12:12 regimens (LeGates et al., 2014). Nevertheless, there are several exceptions to this generalization, especially among different mouse lines (Flaisher-Grinberg, Gampetro, Kronfeld-Schor, & Einat, 2011). Millard and Dole (1983) conducted the first study of photoperiod effects on voluntary ethanol intake. In their study, C57BL/6 mice were maintained either under a standard LD 12:12 cycle or under LD 6:6, and had plain water, 10% and 20% ethanol freely available from separate drinking tubes. By monitoring fluid intake via lickometers, these authors were able to determine the circadian licking pattern for all three fluids. Under LD 12:12, mice showed typical circadian patterns of fluid intake, with high levels of licking during the dark phase and very low levels during the light phase. Under LD 6:6, mice showed high levels of fluid intake during only one of the two daily 6eh dark periods; thus, their circadian system “interpreted” the LD 6:6 cycle as though it were a long-photoperiod, LD 18:6 lighting cycle. Further, intake of both 10% and 20% ethanol was greatly attenuated under the LD 6:6 schedule relative to LD 12:12. These important observations show that photoperiod-related effects on voluntary ethanol intake may be mediated by differences in circadian entrainment pattern, rather than by differences in total daily light exposure per se (which was identical under both lighting regimens). My laboratory recently examined voluntary ethanol intake in male C57BL/6 and C3H/He mice housed under either short (LD 6:18) or long (LD 18:6) photoperiods. In one experiment (Rosenwasser,

5

Fixaris, Crabbe, Brooks, & Ascheid, 2012), we found significantly increased ethanol intake in C3H/He mice under long photoperiods, but failed to detect an effect of photoperiod in C57BL/6 mice. In a second experiment, however (Rosenwasser et al., in press), we did find significantly increased ethanol intake in C57BL/6 mice under short photoperiods, opposite to what we observed in C3H/He mice. Thus, our results appear to be generally consistent with those of Millard and Dole (1983), as well as with recent studies showing atypical increases in anxiety and depression-like behaviors in C3H/ He mice under long photoperiods (Becker, Bilkei-Gorzo, Michel, & Zimmer, 2010; Kopp et al., 1999). The results also reinforce the idea that dramatic strain differences in affective responses to photoperiod may be seen in mice. In an additional experiment, we found identical photoperiod effects on ethanol intake under both normal LD cycles and under so-called “skeleton photoperiods,” in which daily light exposure is limited to two brief light pulses occurring at dawn and dusk, while Trujillo, Roberts, and Gorman (2009) also reported similar levels of ethanol intake under regular and skeleton photoperiods. Together with the report of Millard and Dole (1983), these findings indicate further that photoperiod effects on affective behavior, as is true of the more well-studied photoperiodic regulation of reproductive state, are mediated by changes in circadian entrainment pattern rather than by light exposure per se (Goldman, 2001). Finally, a small number of experiments have examined the effects of shift-lag lighting schedules on voluntary ethanol intake. In the first such report, Gauvin et al. (1997) found transient increases in ethanol intake following single LD cycle phase shifts and sustained increases in ethanol intake during repeated phase shifts in rats. Further, more substantial effects on ethanol intake were seen following acute phase advances than after phase delays, mirroring the well-known differential effects of advances and delays on circadian re-adaptation time. These findings suggested that circadian disruption associated with re-adaptation to shifted LD cycles could lead to increased ethanol intake. Our lab conducted two subsequent experiments examining voluntary ethanol intake during exposure to repeated 6-h LD phase advances in male and female rats (Clark, Fixaris, Belanger, & Rosenwasser, 2007; Rosenwasser, Clark, Fixaris, Belanger, & Foster, 2010). In marked contrast to the results of Gauvin et al. (1997), however, we found that rats exposed to shift-lag lighting generally displayed reduced ethanol intake relative to animals maintained under steady-state LD 12:12 conditions (Fig. 5). While our studies differed in several ways from those of Gauvin et al. (1997) (e.g., rat strains, specific lighting schedules), one potentially important difference is that Gauvin et al. (1997) used limited daily access to ethanol while we employed continuous 24-h preference drinking. More recently, we failed to find any effect of shift-lag lighting in either C57BL/6 or DBA/2 inbred mice (Rosenwasser & Fixaris, 2013), while Trujillo et al. (2011) found no effect of maintenance under either 22-h or 26-h LD cycles on ethanol intake in C57BL/6 mice. Taken together, these results indicate that non-standard lighting regimens, including LL and DD, long and short photoperiods, and shift-lag lighting regimens, can alter voluntary ethanol intake in experimental animals. These effects may reflect widely reported light-induced changes in depression and anxiety-like behaviors in response to similar manipulations. However, it must be stated that this literature includes many conflicting findings, some of which may be accounted for by species and strain differences and/or by numerous protocol variations across experiments. Further, the relative contributions of multiple factors including circadian disruption, photoperiodism, and the so-called “direct” affective-behavioral effects of light and darkness remain to be elucidated.

6

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

Fig. 5. Mean (þSEM) free-choice ethanol intake solution with plain water concurrently available from a separate drinking tube in female (F) and male (M) Fischer (FISCH) and Lewis (LEW) rats. Control animals were maintained under a lightedark (LD) 12:12 cycle throughout the experiment, while Shift-Lag animals were exposed to a series of six, 6-h LD phase advances at 3-week intervals. Modified from Rosenwasser et al., 2010.

Genetic linkages between the circadian system and ethanol consumption At the cellular level, circadian clock function reflects the activity of so-called circadian clock genes and their protein products. These molecules compose a set of interlocking feedback loops in which clock proteins translocate to the cell nucleus, alter transcription, and thereby regulate their own subsequent rhythmic expression. This molecular mechanism operates not only within circadian pacemaker tissues such as the SCN, but is also widely expressed in diverse cell types throughout the brain and body, and increasing evidence indicates that local disruption of this mechanism can result in tissue-specific pathophysiology. For the present discussion, it is highly pertinent that the molecular circadian clock is expressed within the mesolimbic dopamine “reward” pathway as well as in stress- and anxiety-related neurocircuitry (Mendoza & Challet, 2014; Parekh, Ozburn, & McClung, 2015; Webb, Lehman, & Coolen, 2015). Thus, clock gene mutations can alter rewardand stress-related behaviors including depression, anxiety, and drug use (for recent reviews, see Landgraf, McCarthy, & Welsh, 2014a; Logan, Williams, & McClung, 2014; McClung, 2013; Parekh et al., 2015; Rosenwasser, 2010; Schnell et al., 2014). Clock genes and proteins show a remarkable degree of evolutionary conservation, and indeed, the first studies demonstrating that clock gene mutations could alter behavioral responses to drugs of abuse (in this case, cocaine sensitization) were conducted in flies (Drosophila; Andretic, Chaney, & Hirsh, 1999). More recently, Pohl et al. (2013) presented similar results implicating clock genes in Drosophila ethanol tolerance, while Ahmad, Steinmetz, Bussey, Possidente, and Seggio (2013) showed that allelic variations in the Drosophila Per gene dramatically affect the circadian clock response to ethanol treatment in this species (see also Nascimento, Carlson, Amaral, Logan, & Seggio, 2015). To my knowledge, however, the effects of clock gene variations on ethanol reward or ethanol preference have not yet been examined in the fly model. Increased voluntary ethanol intake has been found in mice bearing mutant alleles of several clock genes, including Per1, Per2, and Clock (Brager, Prosser, & Glass, 2011; Gamsby et al., 2013;

Ozburn et al., 2013; Spanagel, Pendyala, et al., 2005; Spanagel, Rosenwasser, et al., 2005). Reciprocally, chronic ethanol intake alters clock gene expression patterns in reward- and stress-related brain areas and in the SCN itself (Chen, Kuhn, Advis, & Sarkar, 2004; Ozburn et al., 2013). While circadian clock genes may alter physiological processes via cellular effects distinct from their role in the molecular clock mechanism (Rosenwasser, 2010), effects on voluntary ethanol intake seem likely to be secondary to the loss of coherent cellular rhythmicity. This conclusion is based on the fact that the Per genes are generally considered to compose the negative limb of the clock loop while Clock is a key mediator of the positive limb; thus, the Per and Clock mutations tend to have opposite effects on free-running period (with Per mutations shortening and Clock mutations lengthening free-running circadian period). Physiologically, the effects of clock gene mutations on ethanol intake have been related to alterations in glutamatergic (Spanagel, Pendyala, et al., 2005; Spanagel, Rosenwasser, et al., 2005) and dopaminergic (Ozburn et al., 2013) neurotransmission, ethanol metabolism (Gamsby et al., 2013), and the daily drinking pattern (Brager, Prosser, et al., 2011). Genetic linkages between ethanol drinking and the circadian system have also been revealed by studies of circadian rhythms in animals bred for differential ethanol responsiveness. Thus, Hofstetter, Grahame, and Mayeda (2003) reported that selectively bred High Alcohol Preferring (HAP) mice displayed shorter freerunning circadian period in DD relative to the Low Alcohol Preferring (LAP) line (note that these original HAP and LAP lines are now referred to as HAP-1 and LAP-1 due to the subsequent derivation of a second pair of HAP/LAP lines, referred to as HAP-2 and LAP-2). While a recent study failed to replicate this finding in the HAP-2 and LAP-2 mice, HAP-2 mice did display shorter periods when maintained on chronic ethanol availability (Trujillo et al., 2011). My laboratory has performed very similar experiments in the selectively bred P (Preferring) and NP (Non-Preferring), as well as in the HAD-2 (High Alcohol Drinking, Replicate 2) and LAD2 (Low Alcohol Drinking, Replicate 2) rat lines. Importantly, these two line pairs were selected based on essentially identical phenotypic criteria, but using genetically dissimilar progenitor stocks. We found that P rats showed shorter free-running period than NP rats when maintained in LL, but not in DD, while HAD-2 rats displayed shorter freerunning periods in both DD and LL (Fig. 6; Rosenwasser, Fecteau, Logan, Reed, et al., 2005). Thus, despite differences in environmental requirements, high-alcohol drinking mice and rats consistently displayed shorter free-running period than low-drinking animals. Finally, we recently performed circadian phenotyping experiments on mouse lines selected for ethanol-related traits other than 24-h preference drinking. In one experiment, we compared selectively bred High Drinking-in-the-Dark, replicate 1 (HDID-1) mice to their genetically heterogeneous HS/Npt progenitor stock (McCulley, Ascheid, Crabbe, & Rosenwasser, 2013). HDID-1 mice were selected for high levels of blood ethanol in the Drinking-in-the-Dark (DID) protocol, in which 20% ethanol solution is provided as the only drinking fluid for a brief (2e4 h) exposure period each day (Crabbe et al., 2009). Many animals (and most HDID-1 mice) will achieve intoxicating blood alcohol levels in this test, which is thus considered to provide a valid model of binge-like drinking. Despite this susceptibility to binge-like drinking, however, HDID-1 mice show little evidence for excessive voluntary drinking under 24-h access conditions (Crabbe, Spence, Brown, & Metten, 2011; Rosenwasser et al., 2012), nor do they differ from controls in their ethanol withdrawal sensitivity (Crabbe et al., 2012). We found that HDID-1 mice display shortening of free-running period relative to unselected HS/Npt controls under LL, but not in DD. In a second experiment (McCulley et al., 2013), we compared circadian

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

ethanol vapor exposure (Crabbe & Phillips, 1993). While WSP-2 and WSR-2 mice were initially reported to display differences in voluntary ethanol consumption (Kosobud, Bodor, & Crabbe, 1988), this difference seems to have been eliminated during the subsequent maintenance of the lines (Ford et al., 2011; Rosenwasser et al., 2012). Despite their reported lack of excessive voluntary ethanol intake, we found that WSR-2 mice displayed shorter free-running period in DD than did WSP-2 mice, suggesting that genes influencing multiple ethanol-related traits also contribute to circadian clock function. While the identity of these genes is not known, alleles related to GABAergic and/or glutamatergic signaling are obvious candidates, since these systems play a critical role in circadian regulation and since GABA and glutamate receptors represent the primary targets for ethanol action within the central nervous system.

Free-running period (h)

26.5 26.0

P NP

25.5 25.0 24.5 24.0 23.5

DD1

LL

DD2

Free-running period (h)

26.5 26.0

HAD LAD

25.5 25.0 24.5 24.0 23.5

DD1

7

LL

DD2

Fig. 6. Mean (þSEM) free-running period in male alcohol-Preferring (P) and NonPreferring (NP) rats (top panel) and in male High Alcohol Drinking (HAD) and Low Alcohol Drinking rats (bottom panel). Animals were maintained successively in continuous darkness (DD1) for 23 days followed by continuous light (LL) for 82 days and finally returned to continuous darkness (DD2) for an additional 44 days. Modified from Rosenwasser, Fecteau, Logan, Reed, et al., 2005.

phenotype in Withdrawal Seizure Prone and Withdrawal Seizure Resistant (WSP-2 and WSR-2, respectively) mice. These animals were differentially selected for high and low severity of handlinginduced convulsions following induction of dependence via

Fig. 7. Illustration summarizing the data reviewed in this paper. Alcohol (ethanol) disrupts sleep and alters circadian clock properties, especially when ethanol intake is high. These effects are seen at the behavioral, physiological, and molecular levels of clock function. In turn, environmental and genetic manipulations that disrupt circadian clock function are capable of altering voluntary alcohol consumption, although these effects are somewhat less well-characterized and less consistent.

Conclusion As summarized in cartoon form in Fig. 7, the work described in this review indicates that reciprocal interactions occur between the circadian system and ethanol consumption at both physiological and genetic levels of analysis. Thus, both chronic ethanol consumption and acute ethanol administration are capable of affecting fundamental properties of the circadian pacemaker, including its free-running period and responsiveness to photic stimulation. These results are mediated in part via direct pharmacological effects on the circadian pacemaker, and while more work needs to be done, are probably due to interactions with SCN GABAergic and glutamatergic receptors (see Prosser & Glass, 2015). In addition, chronic ethanol exposure alters clock gene expression within the SCN and within the brain’s reward and stress pathways (see Parekh et al., 2015), and these effects contribute to the affective and addictive properties of ethanol as they do for other drugs of abuse. While there has been virtually no parametric analysis of the dosage requirements for these effects, it does appear that high levels of ethanol exposure may be required to produce significant and enduring alterations in clock function. For example, we found that, despite their typically high levels of voluntary intake, C57BL/6 mice displayed alterations in free-running circadian period only under conditions of forced ethanol intake (Seggio et al., 2009). Similarly, C3H mice displayed alterations in circadian period following a 16day but not following a 4-day ethanol vapor exposure protocol (Logan et al., 2012). Of course, it should also be noted that chronobiological disruption has typically been observed in human alcohol abusers and/or alcoholics, and there is little or no evidence that circadian disruption occurs as a consequence of moderate, social drinking. In turn, environmental and genetic manipulations of the circadian system alter voluntary ethanol intake, potentially creating a vicious cycle-like phenomenon. Unfortunately, the effects of environmental lighting manipulations on voluntary ethanol intake are far from consistent, and most studies appear to show that animals actually decrease rather than increase their intake under “exotic” or unstable lighting regimens. Of course, there is a substantial literature employing more traditional forms of laboratory stressors also showing reduced rather than increased voluntary drinking under stress (Becker, Lopez, & Doremus-Fitzwater, 2011), supporting the idea that altered lighting environments may indeed function as a form of chronobiological stressor. These observations are not as surprising as they might seem initially, given that most animal drinking studies examine non-dependent, and presumably rewarddriven, ethanol intake, and it is widely appreciated that stress can interfere with reward seeking. It will therefore be important for future studies to explore the effects of chronobiological manipulations in animal models of excessive or dependent drinking.

8

A.M. Rosenwasser / Alcohol xxx (2015) 1e9

Despite the generally inconsistent literature on environmental circadian disruption, there is rather consistent evidence that genetic disruption of the cellular circadian clock mechanism increases ethanol intake. As described above, increased drinking has been reported in mice with mutations of both the clock gene and the per genes. While these mutations share the ability to disrupt the normal display of circadian rhythms, they do so via nonredundant functions within the molecular clock loop, making it highly likely that the observed effects on ethanol (and other drug) intake are secondary to clock dysfunction. Remarkably, clock genes have also been linked to excessive drinking in human populations (see Partonen, 2015; Perreau-Lenz & Spanagel, 2015). Finally, we have also shown that rats and mice selectively bred for differential ethanol intake and/or withdrawal severity consistently display alterations in circadian phenotype, although the specific nature of these alterations appears to vary somewhat from one model to another. The genes underlying these presumably multigenic effects have not been identified, but it appears likely that they could include genes regulating glutamatergic and GABAergic neurotransmission, and could also possibly include clock genes as well. Whatever the genetic mechanisms, these results indicate that comorbidity of sleep and circadian disruption with alcohol and drug abuse could occur because common underlying gene sets regulate these distinct behaviors. In conclusion, animal studies support the hypothesis that the chronobiological disruption commonly seen in association with alcohol and drug use disorders reflects bidirectional causal influences at both physiological and genetic levels. It is hoped that a better understanding of these relationships may lead eventually to novel chronobiologically based interventions that may help mitigate relapse risk and possibly reduce the overall societal burden of drug and alcohol abuse.

References Ahmad, S. T., Steinmetz, S. B., Bussey, H. M., Possidente, B., & Seggio, J. A. (2013). Larval ethanol exposure alters free-running circadian rhythm and per locus transcription in adult D. melanogaster period mutants. Behavioural Brain Research, 241, 50e55. Andretic, R., Chaney, S., & Hirsh, J. (1999). Requirement of circadian genes for cocaine sensitization in Drosophila. Science, 285, 1066e1068. Becker, A., Bilkei-Gorzo, A., Michel, K., & Zimmer, A. (2010). Exposure of mice to long-light: a new animal model to study depression. European Neuropsychopharmacology, 20, 802e812. Becker, H. C., Lopez, M. F., & Doremus-Fitzwater, T. L. (2011). Effects of stress on alcohol drinking: a review of animal studies. Psychopharmacology, 218, 131e156. Bedrosian, T. A., & Nelson, R. J. (2013). Influence of the modern light environment on mood. Molecular Psychiatry, 18, 751e757. Brager, A. J., Prosser, R. A., & Glass, J. D. (2011). Circadian and acamprosate modulation of elevated ethanol drinking in mPer2 clock gene mutant mice. Chronobiology International, 28, 664e672. Brager, A. J., Ruby, C. L., Prosser, R. A., & Glass, J. D. (2010). Chronic ethanol disrupts circadian photic entrainment and daily locomotor activity in the mouse. Alcoholism: Clinical and Experimental Research, 34, 1266e1273. Brager, A. J., Ruby, C. L., Prosser, R. A., & Glass, J. D. (2011). Acute ethanol disrupts photic and serotonergic circadian clock phase-resetting in the mouse. Alcoholism: Clinical and Experimental Research, 35, 1467e1474. Brower, K. J. (2001). Alcohol’s effects on sleep in alcoholics. Alcohol Research & Health, 25, 110e125. Brower, K. J. (2003). Insomnia, alcoholism and relapse. Sleep Medicine Reviews, 7, 523e539. Burke, L. P., & Kramer, S. Z. (1974). Effects of photoperiod, melatonin and pinealectomy on ethanol consumption in rats. Pharmacology, Biochemistry, and Behavior, 2, 459e463. Chen, C. P., Kuhn, P., Advis, J. P., & Sarkar, D. K. (2004). Chronic ethanol consumption impairs the circadian rhythm of pro-opiomelanocortin and period genes mRNA expression in the hypothalamus of the male rat. Journal of Neurochemistry, 88, 1547e1554. Clark, J. W., Fixaris, M. C., Belanger, G. V., & Rosenwasser, A. M. (2007). Repeated light-dark phase shifts modulate voluntary ethanol intake in male and female high alcohol-drinking (HAD1) rats. Alcoholism: Clinical and Experimental Research, 31, 1699e1706.

Conroy, D. A., Hairston, I. S., Arendt, J. T., Hoffmann, R. F., Armitage, R., & Brower, K. J. (2012). Dim light melatonin onset in alcohol-dependent men and women compared with healthy controls. Chronobiology International, 29, 35e42. Crabbe, J. C., Colville, A. M., Kruse, L. C., Cameron, A. J., Spence, S. E., Schlumbohm, J. P., et al. (2012). Ethanol tolerance and withdrawal severity in high drinking in the dark selectively bred mice. Alcoholism: Clinical and Experimental Research, 36, 1152e1161. Crabbe, J. C., Metten, P., Rhodes, J. S., Yu, C. H., Brown, L. L., Phillips, T. J., et al. (2009). A line of mice selected for high blood ethanol concentrations shows drinking in the dark to intoxication. Biological Psychiatry, 65, 662e670. Crabbe, J. C., & Phillips, T. J. (1993). Selective breeding for alcohol withdrawal severity. Behavior Genetics, 23, 171e177. Crabbe, J. C., Spence, S. E., Brown, L. L., & Metten, P. (2011). Alcohol preference drinking in a mouse line selectively bred for high drinking in the dark. Alcohol, 45, 427e440. Danel, T., & Touitou, Y. (2004). Chronobiology of alcohol: from chronokinetics to alcohol-related alterations of the circadian system. Chronobiology International, 21, 923e935. Dwyer, S. M., & Rosenwasser, A. M. (1998). Neonatal clomipramine treatment, alcohol intake and circadian rhythms in rats. Psychopharmacology (Berl), 138, 176e183. Flaisher-Grinberg, S., Gampetro, D. R., Kronfeld-Schor, N., & Einat, H. (2011). Inconsistent effects of photoperiod manipulations in tests for affective-like changes in mice: implications for the selection of appropriate model animals. Behavioural Pharmacology, 22, 23e30. Ford, M. M., Fretwell, A. M., Anacker, A. M., Crabbe, J. C., Mark, G. P., & Finn, D. A. (2011). The influence of selection for ethanol withdrawal severity on traits associated with ethanol self-administration and reinforcement. Alcoholism: Clinical and Experimental Research, 35, 326e337. Gamsby, J. J., Templeton, E. L., Bonvini, L. A., Wang, W., Loros, J. J., Dunlap, J. C., et al. (2013). The circadian Per1 and Per2 genes influence alcohol intake, reinforcement, and blood alcohol levels. Behavioural Brain Research, 249, 15e21. Gauvin, D. V., Baird, T. J., Vanecek, S. A., Briscoe, R. J., Vallett, M., & Holloway, F. A. (1997). Effects of time-of-day and photoperiod phase shifts on voluntary ethanol consumption in rats. Alcoholism: Clinical and Experimental Research, 21, 817e825. Geller, I. (1971). Ethanol preference in the rat as a function of photoperiod. Science, 173, 456e459. Geller, I., & Hartmann, R. J. (1977). Alteration of ethanol preference in hamsters: effects of photoperiod and 5-hydroxytryptophan. Advances in Experimental Medicine and Biology, 85B, 223e233. Goldman, B. D. (2001). Mammalian photoperiodic system: formal properties and neuroendocrine mechanisms of photoperiodic time measurement. Journal of Biological Rhythms, 16, 283e301. Goodwin, F. L., Amir, S., & Amit, Z. (1999). Environmental lighting has a selective influence on ethanol intake in rats. Physiology & Behavior, 66, 323e328. Hammer, S. B., Ruby, C. L., Brager, A. J., Prosser, R. A., & Glass, J. D. (2010). Environmental modulation of alcohol intake in hamsters: effects of wheel running and constant light exposure. Alcoholism: Clinical and Experimental Research, 34, 1651e1658. Hasler, B. P., Smith, L. J., Cousins, J. C., & Bootzin, R. R. (2012). Circadian rhythms, sleep and substance abuse. Sleep Medicine Reviews, 16, 67e81. Hasler, B. P., Soehner, A. M., & Clark, D. B. (2015). Sleep and circadian contributions to adolescent alcohol use disorder. Alcohol, 44, 239e244. Hofstetter, J. R., Grahame, N. J., & Mayeda, A. R. (2003). Circadian activity rhythms in high-alcohol-preferring and low-alcohol-preferring mice. Alcohol, 30, 81e85. Klemfuss, H., & Kripke, D. F. (1994). Antidepressant and depressogenic drugs lack consistent effects on hamster circadian rhythms. Psychiatry Research, 53, 173e184. Kopp, C., Vogel, E., Rettori, M. C., Delagrange, P., Renard, P., Lesieur, D., et al. (1999). Regulation of emotional behaviour by day length in mice: implication of melatonin. Behavioural Pharmacology, 10, 747e752. Kosobud, A., Bodor, A. S., & Crabbe, J. C. (1988). Voluntary consumption of ethanol in WSP, WSC and WSR selectively bred mouse lines. Pharmacology, Biochemistry, and Behavior, 29, 601e607. Kuhlwein, E., Hauger, R. L., & Irwin, M. R. (2003). Abnormal nocturnal melatonin secretion and disordered sleep in abstinent alcoholics. Biological Psychiatry, 54, 1437e1443. Landgraf, D., McCarthy, M. J., & Welsh, D. K. (2014a). Circadian clock and stress interactions in the molecular biology of psychiatric disorders. Current Psychiatry Reports, 16, 483. Landgraf, D., McCarthy, M. J., & Welsh, D. K. (2014b). The role of the circadian clock in animal models of mood disorders. Behavioral Neuroscience, 128, 344e359. Landolt, H. P., & Gillin, J. C. (2001). Sleep abnormalities during abstinence in alcoholdependent patients. Aetiology and management. CNS Drugs, 15, 413e425. LeGates, T. A., Fernandez, D. C., & Hattar, S. (2014). Light as a central modulator of circadian rhythms, sleep and affect. Nature Reviews. Neuroscience, 15, 443e454. Logan, R. W., McCulley, W. D., 3rd, Seggio, J. A., & Rosenwasser, A. M. (2012). Effects of withdrawal from chronic intermittent ethanol vapor on the level and circadian periodicity of running-wheel activity in C57BL/6J and C3H/HeJ mice. Alcoholism: Clinical and Experimental Research, 36, 467e476. Logan, R. W., Seggio, J. A., Robinson, S. L., Richard, G. R., & Rosenwasser, A. M. (2010). Circadian wheel-running activity during withdrawal from chronic intermittent ethanol exposure in mice. Alcohol, 44, 239e244.

A.M. Rosenwasser / Alcohol xxx (2015) 1e9 Logan, R. W., Williams, W. P., 3rd, & McClung, C. A. (2014). Circadian rhythms and addiction: mechanistic insights and future directions. Behavioral Neuroscience, 128, 387e412. McClung, C. A. (2013). How might circadian rhythms control mood? Let me count the ways.. Biological Psychiatry, 74, 242e249. McCulley, W. D., 3rd, Ascheid, S., Crabbe, J. C., & Rosenwasser, A. M. (2013). Selective breeding for ethanol-related traits alters circadian phenotype. Alcohol, 47, 187e194. Mendoza, J., & Challet, E. (2014). Circadian insights into dopamine mechanisms. Neuroscience, 282C, 230e242. Millard, W. J., & Dole, V. P. (1983). Intake of water and ethanol by C57BL mice: effect of an altered light-dark schedule. Pharmacology, Biochemistry, and Behavior, 18, 281e284. Mistlberger, R. E., & Nadeau, J. (1992). Ethanol and circadian rhythms in the Syrian hamster: effects on entrained phase, reentrainment rate, and period. Pharmacology, Biochemistry, and Behavior, 43, 159e165. Nascimento, N. F., Carlson, K. N., Amaral, D. N., Logan, R. W., & Seggio, J. A. (2015). Alcohol and lithium have opposing effects on the period and phase of the behavioral free-running activity rhythm. Alcohol, 44, 367e376. Ozburn, A. R., Falcon, E., Mukherjee, S., Gillman, A., Arey, R., Spencer, S., et al. (2013). The role of clock in ethanol-related behaviors. Neuropsychopharmacology, 38, 2393e2400. Parekh, P. K., Ozburn, A. R., & McClung, C. A. (2015). Circadian clock genes: effects on dopamine, reward and addiction. Alcohol, 44, 341e349. Partonen, T. (2015). Clock genes in human alcohol abuse and comorbid conditions. Alcohol, 44, 359e365. Perreau-Lenz, S., & Spanagel, R. (2015). Clock genes X stress X reward interactions in alcohol and substance use disorders. Alcohol, 44, 351e357. Pohl, J. B., Ghezzi, A., Lew, L. K., Robles, R. B., Cormack, L., & Atkinson, N. S. (2013). Circadian genes differentially affect tolerance to ethanol in Drosophila. Alcoholism: Clinical and Experimental Research, 37, 1862e1871. Prosser, R. A., & Glass, J. D. (2015). Assessing ethanol’s actions in the suprachiasmatic circadian clockusing in vivo and in vitro approaches. Alcohol, 44, 321e339. Prosser, R. A., Mangrum, C. A., & Glass, J. D. (2008). Acute ethanol modulates glutamatergic and serotonergic phase shifts of the mouse circadian clock in vitro. Neuroscience, 152, 837e848. Reiter, R. J., Blum, K., Wallace, J. E., & Merritt, J. H. (1973). Effect of the pineal gland on alcohol consumption by congenitally blind male rats. Quarterly Journal of Studies on Alcohol, 34, 937e939. Reiter, R. J., Blum, K., Wallace, J. E., & Merritt, J. H. (1974). Pineal gland: evidence for an influence on ethanol preference in male Syrian hamsters. Comparative Biochemistry and Physiology. A, Comparative Physiology, 47, 11e16. Rogers, H. L., & Reilly, S. M. (2002). A survey of the health experiences of international business travelers. Part one e physiological aspects. AAOHN Journal, 50, 449e459. Rosenwasser, A. M. (2001). Alcohol, antidepressants, and circadian rhythms. Human and animal models. Alcohol Research & Health, 25, 126e135. Rosenwasser, A. M. (2010). Circadian clock genes: non-circadian roles in sleep, addiction, and psychiatric disorders? Neuroscience and Biobehavioral Reviews, 34, 1249e1255. Rosenwasser, A. M., Clark, J. W., Fixaris, M. C., Belanger, G. V., & Foster, J. A. (2010). Effects of repeated light-dark phase shifts on voluntary ethanol and water intake in male and female Fischer and Lewis rats. Alcohol, 44, 229e237. Rosenwasser, A. M., Fecteau, M. E., & Logan, R. W. (2005). Effects of ethanol intake and ethanol withdrawal on free-running circadian activity rhythms in rats. Physiology & Behavior, 84, 537e542.

9

Rosenwasser, A. M., Fecteau, M. E., Logan, R. W., Reed, J. D., Cotter, S. J., & Seggio, J. A. (2005). Circadian activity rhythms in selectively bred ethanol-preferring and nonpreferring rats. Alcohol, 36, 69e81. Rosenwasser, A. M., & Fixaris, M. C. (2013). Chronobiology of alcohol: studies in C57BL/6J and DBA/2J inbred mice. Physiology & Behavior, 110e111, 140e147. Rosenwasser, A. M., Fixaris, M. C., Crabbe, J. C., Brooks, P. C., & Ascheid, S. (2012). Escalation of intake under intermittent ethanol access in diverse mouse genotypes. Addiction Biology, 18, 496e507. Rosenwasser, A. M., Logan, R. W., & Fecteau, M. E. (2005). Chronic ethanol intake alters circadian period-responses to brief light pulses in rats. Chronobiology International, 22, 227e236. Rosenwasser, A. M., McCulley, W. D., 3rd, & Fecteau, M. (2014). Circadian activity rhythms and voluntary ethanol intake in male and female ethanol-preferring rats: effects of long-term ethanol access. Alcohol, 48, 647e655. Ruby, C. L., Brager, A. J., DePaul, M. A., Prosser, R. A., & Glass, J. D. (2009). Chronic ethanol attenuates circadian photic phase resetting and alters nocturnal activity patterns in the hamster. American Journal of Physiology. Regulatory, Integrative and Comparative Physiology, 297, R729eR737. Ruby, C. L., Prosser, R. A., DePaul, M. A., Roberts, R. J., & Glass, J. D. (2009). Acute ethanol impairs photic and nonphotic circadian phase resetting in the Syrian hamster. American Journal of Physiology. Regulatory, Integrative and Comparative Physiology, 296, R411eR418. Ruby, C. L., Vadnie, C. A., Hinton, D. J., Abulseoud, O. A., Walker, D. L., O’Connor, K. M., et al. (2014). Adenosinergic regulation of striatal clock gene expression and ethanol intake during constant light. Neuropsychopharmacology, 39, 2432e2440. Schnell, A., Albrecht, U., & Sandrelli, F. (2014). Rhythm and mood: relationships between the circadian clock and mood-related behavior. Behavioral Neuroscience, 128, 326e343. Seggio, J. A., Fixaris, M. C., Reed, J. D., Logan, R. W., & Rosenwasser, A. M. (2009). Chronic ethanol intake alters circadian phase shifting and free-running period in mice. Journal of Biological Rhythms, 24, 304e312. Seggio, J. A., Logan, R. W., & Rosenwasser, A. M. (2007). Chronic ethanol intake modulates photic and non-photic circadian phase responses in the Syrian hamster. Pharmacology, Biochemistry, and Behavior, 87, 297e305. Spanagel, R., Pendyala, G., Abarca, C., Zghoul, T., Sanchis-Segura, C., Magnone, M. C., et al. (2005). The clock gene Per2 influences the glutamatergic system and modulates alcohol consumption. Nature Medicine, 11, 35e42. Spanagel, R., Rosenwasser, A. M., Schumann, G., & Sarkar, D. K. (2005). Alcohol consumption and the body’s biological clock. Alcoholism: Clinical and Experimental Research, 29, 1550e1557. Stephenson, K. M., Schroder, C. M., Bertschy, G., & Bourgin, P. (2012). Complex interaction of circadian and non-circadian effects of light on mood: shedding new light on an old story. Sleep Medicine Reviews, 16, 445e454. Trinkoff, A. M., & Storr, C. L. (1998). Work schedule characteristics and substance use in nurses. American Journal of Industrial Medicine, 34, 266e271. Trujillo, J. L., Do, D. T., Grahame, N. J., Roberts, A. J., & Gorman, M. R. (2011). Ethanol consumption in mice: relationships with circadian period and entrainment. Alcohol, 45, 147e159. Trujillo, J. L., Roberts, A. J., & Gorman, M. R. (2009). Circadian timing of ethanol exposure exerts enduring effects on subsequent ad libitum consumption in C57 mice. Alcoholism: Clinical and Experimental Research, 33, 1286e1293. Webb, I. C., Lehman, M. N., & Coolen, L. M. (2015). Diurnal and circadian regulation of reward-related neurophysiology and behavior. Physiology & Behavior, 143, 58e69. Wollnik, F. (1992). Effects of chronic administration and withdrawal of antidepressant agents on circadian activity rhythms in rats. Pharmacology, Biochemistry, and Behavior, 43, 549e561.