Effects of pharmacological inhibition of the centrally-projecting Edinger-Westphal nucleus on ethanol-induced conditioned place preference and body temperature.

Abstract

Alcohol use disorder is a chronic disease characterized in part by repeated relapsing events. Exposure to environmental stimuli or cues that have previously been associated with the effects of alcohol can promote relapse through the triggering of craving for alcohol. Therefore, identifying and characterizing neuronal populations that may regulate these associations is of the upmost importance. Previous studies have implicated the centrally-projecting Edinger Westphal nucleus (EWcp) in this process, as the EWcp is both sensitive to, and can regulate alcohol intake. To date however, it is unclear if the EWcp is involved in the formation or expression of these alcohol-cue associations. As such, the present studies examined the involvement of the EWcp in male DBA/2J mice in the acquisition and expression of place preference for an alcohol-paired cue using the conditioned place preference (CPP) procedure. Pharmacological inhibition of the EWcp via the GABAA and GABAB receptor agonists muscimol and baclofen did not affect either the acquisition or the expression of CPP. Follow up studies did find however, that pharmacological inhibition of the EWcp increased body temperature and prevented alcohol-induced increases in c-Fos expression in the EWcp. When considered in light of previous studies, the present results indicate that the EWcp may be involved in the regulation of alcohol self-administration, and not conditioned alcohol-seeking. Additionally, the present studies provide further evidence for the involvement of the EWcp in thermoregulation and help elucidate the molecular mechanism by which alcohol increases c-Fos in the EWcp.

Introduction

Alcohol use disorder (AUD) is a chronic disease characterized by compulsive alcohol (ethanol, EtOH) intake, negative emotional affect during periods of abstinence, and repeated relapse events (American Psychiatric Association, 2013). Through Pavlovian conditioning, environmental stimuli or cues that are temporally proximal to the intoxicating effects of EtOH can become associated with EtOH, and can subsequently trigger conditioned responses (Chaudhri, Sahuque, & Janak, 2008; Cunningham & Noble, 1992; Duncan, Alici, & Woodward, 2000; Morales, Varlinskaya, & Spear, 2012; Remedios, Woods, Tardif, Janak, & Chaudhri, 2014). Of note, exposure to these cues is believed to promote relapse, even following prolonged periods of abstinence (Ciccocioppo, Angeletti, & Weiss, 2001; Ciccocioppo, Lin, Martin-Fardon, & Weiss, 2003; Nie & Janak, 2003). Indeed, in abstinent AUD patients, exposure to olfactory cues elicits a number of physiological and psychological responses, including elevated heart rate, salivation, desire to drink, and withdrawal symptoms (Pomerleau, Fertig, Baker, & Cooney, 1983; Staiger & White, 1991).

In rodents, the conditioned place preference (CPP) model has been readily used to assess the rewarding and aversive effects of drugs (Liu, Le Foll, Liu, Wang, & Lu, 2008; Tzschentke, 1998, 2007). In this model, a context acts as a conditioned stimulus (CS+), and through repeated pairings with an unconditioned stimulus (such as EtOH), strong preference for the CS+ develops. In the absence of EtOH (such as during a preference test), an animal will approach and/or prefer the context previously paired with EtOH. Although some debate exists as to why this approach and/or preference occurs, it has been hypothesized that the context has come to predict the rewarding effects of EtOH, and thus the animals spends more time approaching and/or preferring the context (Bardo & Bevins, 2000; Cunningham, Groblewski, & Voorhees, 2011). Following this logic, this approach and/or preference can be interpreted as an EtOH-seeking behavior. Importantly, the nature of this form of classical conditioning allows for separate examinations of how distinct brains regions may be involved in the acquisition, expression, and/or extinction of these cue-drug associations. Pharmacological methods have been used in combination with the place preference model to assess the involvement of numerous other brain regions in EtOH-cue associated behaviors, including the bed nucleus of the stria terminalis (BNST) (Pina, Young, Ryabinin, & Cunningham, 2015; Pina & Cunningham, 2017), the ventral tegmental area (VTA) (Bechtholt & Cunningham, 2005), and the medial prefrontal cortex (mPFC) (Groblewski, Ryabinin, & Cunningham, 2012). Although these studies have all provided critical insight into the neurobiology of EtOH-seeking behaviors, additional work examining how other brain regions may be involved is essential. One area in particular, the centrally-projecting Edinger-Westphal nucleus (EWcp), has been implicated in the regulation of voluntary EtOH intake, and may therefore also be involved in conditioned-EtOH-seeking behaviors.

The EWcp is a midbrain region situated along the midline in the perioculomotor area, between the VTA and the dorsal raphe nucleus (DRN). In contrast to the cholinergic parasympathetic neurons that form the pre-ganglionic EW (EWpg), the EWcp is the main source of urocortin 1 (Ucn1) (Kozicz, Yanaihara, & Arimura, 1998; Wong et al., 1996), a neuropeptide belonging to the corticotropin-releasing factor (CRF) family, within the brain. Previous work has implicated the CRF system as being involved in the regulation of ethanol intake, particularly through its role in mediating negative affective states during withdrawal (Becker, 2012; Menzaghi et al., 1994; Phillips, Reed, & Pastor, 2015; Zorrilla, Heilig, de Wit, & Shaham, 2013). Given that Ucn1 has greater binding affinities to both CRF1 and CRF2 receptors than CRF itself (Hsu & Hsueh, 2001; Vaughan et al., 1995), Ucn1 within the EWcp has been hypothesized to be involved in the regulation of EtOH intake and preference. The potential involvement of the EWcp in EtOH intake and preference was supported by early reports indicating that the EWcp is sensitive to EtOH administration. Indeed, experimenter-administered injections of EtOH result in increased c-Fos expression in the EWcp of rats (Ryabinin, Criado, Henriksen, Bloom, & Wilson, 1997; Spangler, Cote, Anacker, Mark, & Ryabinin, 2009) and mice (Bachtell & Ryabinin, 2001; Bachtell, Tsivkovskaia, & Ryabinin, 2002a; Turek & Ryabinin, 2005b). Furthermore, the EWcp is one of two brain regions (along with the central nucleus of the amygdala) that show increased c-Fos expression following an EtOH injection when mice are under isoflurane anesthesia (Smith, Li, Cote, & Ryabinin, 2016). Voluntary oral intake of EtOH has also been shown to preferentially increase levels of c-Fos in the EWcp in a number of rodent species, including mice (Bachtell, Wang, Freeman, Risinger, & Ryabinin, 1999; Ryabinin, Bachtell, Freeman, & Risinger, 2001; Ryabinin, Galvan-Rosas, Bachtell, & Risinger, 2003; Ryabinin, Wang, Freeman, & Risinger, 1999), rats (Topple, Hunt, & McGregor, 1998), and prairie voles (Anacker, Loftis, Kaur, & Ryabinin, 2011). In addition, the EWcp is involved in the regulation of EtOH intake, as electrolytic lesions (Bachtell, Weitemier, & Ryabinin, 2004; Giardino, Cocking, Kaur, Cunningham, & Ryabinin, 2011), as well as viral knockdown of the Ucn1 within the EWcp (Giardino et al., 2017), decrease EtOH intake and preference.

To date, only one study has investigated the role of the EWcp in conditioned EtOH-seeking behaviors. Using a Ucn1-KO transgenic line, Giardino et al. examined the involvement of Ucn1 within the EWcp in EtOH-CPP (Giardino et al., 2011). When compared to WT littermates, Ucn1-KO mice displayed no preference for the EtOH-paired context, indicating that Ucn1 is necessary for place preference. The use of a transgenic KO mouse line however, does not allow for one to distinguish between Ucn1’s involvement in the various phases of CPP, as mice are lacking Ucn1 throughout development and during the entire procedure. As such, we sought to pharmacologically inhibit the EWcp during the acquisition and expression phases of conditioning, in order determine its role in either the formation or expression of conditioned EtOH-seeking behaviors.

Multiple lines of evidence have also implicated the EWcp as being involved in thermoregulation. Specifically, acute, but not repeated exposure to warm (34º C) or cold (10º C) ambient temperatures increased the number of c-Fos positive cells in the EWcp (Bachtell, Tsivkovskaia, & Ryabinin, 2003). Furthermore, Ucn1 expression in the EWcp has been found to correlate with body temperature, as well with levels of EtOH-induced hypothermia in mice (Bachtell, Tsivkovskaia, & Ryabinin, 2002b). Lastly, Bachtell and colleagues found that electrolytic lesioning of the EWcp blunts the hypothermic response following an injection of EtOH (Bachtell et al., 2004). Given these data, we also sought to determine the effect of pharmacological inhibition of the EWcp on body temperature both in the presence and absence of EtOH. Lastly, as it known that EtOH administration increases c-Fos expression in the EWcp, we examined if inhibition of the EWcp via muscimol and baclofen (M+B) would prevent this EtOH-induced increase in c-Fos.

Materials and Methods

General Methods

Animals

Male, eight-week old DBA/2J mice (The Jackson Laboratory, CA) were housed in groups of four and were allowed to habituate to the animal room for one week prior to undergoing stereotaxic surgeries. Mice were kept on a 12:12 h light-dark cycle with lights on at 7:00 AM. After recovering from surgery (see below) mice were housed two to a cage for the remainder of all experiments. Standard rodent chow and water were available ad libitum throughout the duration of the study. All procedures were approved by the Oregon Health & Science University IACUC.

Drugs

Ethanol (20% v/v) was prepared from a 95% stock solution in 0.9% sterile saline, and was injected intraperitoneally (IP) at a dose of either 2 g/kg (volume, 12.5 mL/kg) or 3 g/kg (volume, 18.75 mL/kg). Using concentrations based on previous work in our lab (Pina et al., 2015), the GABA_A and GABA_B agonists muscimol (0.1 mM, Sigma-Aldrich, MO) and baclofen (1.0 mM, Sigma-Aldrich, MO) were dissolved in 0.9% saline and a cocktail (muscimol + baclofen, M+B) of the two drugs (100 nL) was microinfused into the EWcp over the course of 60 s. Injectors were then left in place for an additional 30 s in order for the drugs to completely diffuse into the EWcp.

Stereotaxic Surgery and Cannula Implantation

For all experiments, mice were anesthetized with 5% isoflurane delivered in oxygen via a precision vaporizer (Datex Ohmeda, WI), and subsequently secured into a stereotaxic frame (Kopf Instruments, CA). Once in the stereotaxic frame, mice were given a subcutaneous injection of the non-steroidal anti-inflammatory drug carprofen (0.5 mg/kg, 10 ml/kg) and maintained under 1–2% isoflurane anesthesia. The EWcp was targeted using coordinates (A/P −3.45 mm from bregma, M/L −1.20 mm, D/V −3.5 mm) based on previous studies in the Ryabinin lab (Giardino et al., 2017) and on the standard mouse brain atlas (Paxinos and Franklin, 2001). A single burr hole was drilled 3.45 mm from bregma, and 1.2 mm from the midline. In order to avoid the sagittal sinus, a single guide cannula (2.5 mm, 26 ga, Plastics One, VA) was implanted 2.0 mm above the EWcp at a 20-degree angle. Guide cannulae were held in place using Durelon carboxylate cement (3M, MN) anchored with stainless steel screws. Custom dummy cannulae (2.5 mm, 32 ga, Plastics One, VA) were then lowered and placed into the guide cannulae so as to prevent clogging. Following the surgery, mice were given additional daily subcutaneous injections of carprofen (0.5 mg/kg, 10 ml/kg) for 3 days, and were allowed to recover for 4–7 days before the start of an experiment.

Piercing

In all experiments, 24–48 h prior to the first intracranial infusion, a 3.5 mm stylet was lowered into the EWcp in order to minimize any behavioral effects associated with the initial lowering of an internal cannula (Gremel & Cunningham, 2009). For Exp. 1 and 2, this was done 24 h before the habituation and preference test, respectively. For Exp. 3, this was done during the habituation session.

Conditioning Apparatus

Conditioning was conducted using an apparatus described in detail by Cunningham and colleagues (Cunningham, Gremel, & Groblewski, 2006). Briefly, eight conditioning boxes (30 × 15 × 15 cm) were individually enclosed in larger, well-ventilated chambers (Coulbourn Instruments, Model E10–20, dimensions: 56.1 × 46 × 39.4 cm) in which sound and light were both attenuated. The conditioning boxes were each equipped with six sets of infrared photodetectors (5 cm apart, 2.2 cm above the floor) allowing for real-time activity measurements and the detection of an animal’s position. The conditioning floors used here consisted of two interchangeable halves that were either a grid or a hole pattern. Grid floors were made up of 2.3 mm stainless steel rods, mounted to acrylic sides in 6.4 mm intervals. Hole floors were made from stainless steel sheets perforated with 6.4 mm holes in a staggered manner (9.5 mm apart). These floors have been used extensively in our laboratory, and numerous studies have shown that DBA/2J mice will develop robust place conditioning using these two floor types, while showing an equal preference for both floors initially (Cunningham et al., 2003).

Experiment 1: Inhibition of the EWcp during EtOH CPP Acquisition

General Procedure

Exp. 1 was conducted to examine the effect of inactivation of the EWcp (via M+B) on CPP acquisition. Following recovery from surgery, mice were assigned to M+B (n = 24) or vehicle groups (n =24), as well as to GRID+ (G+) or GRID− (G−) subgroups (n = 12/subgroup). Mice in the G+ subgroup received 2 g/kg EtOH [CS+] while on the grid floor, and the hole floor was paired with saline [CS−]). Conversely, mice in the G− subgroup received EtOH while on the hole floor, and saline was paired with the grid floor. All mice were exposed to an unbiased, one-compartment conditioning procedure that consisted of three distinct phases: habituation (one session), conditioning (two CS+ sessions, two CS− sessions), and preference test (one preference test).

Conditioning Procedure

In Exp. 1, habituation occurred 24 h after the piercing session and consisted of a single session in which mice were handled and gently scruffed for 90 s as they would be during a microinjection session, but nothing was microinfused. Mice were then immediately returned to their home cage, and 30 min later, mice were given a single IP injection of saline and placed in the apparatus on white paper floor for 5 min. This habituation session was conducted to familiarize the animals to both the microinjection and CPP procedure, as well as to the CPP apparatus itself.

As the goal of Exp. 1 was to determine the effect of EWcp inhibition on the acquisition of EtOH CPP, M+B or vehicle was microinfused into the EWcp only during this phase. Previous work in our lab has shown that the handling associated with microinfusion, as well as the amount of time between microinfusion and the start of a conditioning trial can impede the development of place preference in control animals (Young, Dreumont, & Cunningham, 2014). Based on those studies, mice received a microinfusion of either M+B or vehicle 30 min prior to CS+ trials. During CS− trials, 30 min before the conditioning trial, mice were exposed to a sham infusion procedure, in which a dummy microinjector (2.5 mm, 32 ga) was inserted into the guide cannulae for an equivalent period of time, but no fluid was infused. The order in which mice were exposed to each trial type (CS+ or CS−) was counterbalanced between treatment and conditioning subgroups.

Twenty-four h after the last conditioning session, mice were weighed and injected with saline immediately prior to being placed in the apparatus for a 30-min preference test on split floors. The positioning of each floor during testing (left vs. right) was counterbalanced within treatment and conditioning subgroups. Immediately after the preference test, mice were sacrificed, and brains were collected for placement verification.

Experiment 2: Inhibition of the EWcp during EtOH CPP expression

General Procedure

Exp. 2 examined the effect of EWcp inhibition on CPP expression during a preference test. As with Exp. 1, following recovery from surgery, mice were assigned to M+B (n = 28) or Vehicle (n = 28) groups, as well as to G+ and G− subgroups (n = 14/subgroup). The one-compartment unbiased procedure consisted of a single 5-min habituation session, eight 5-min conditioning sessions (four CS+, 4 CS−) and a single 30-min preference test.

Conditioning Procedure

During habituation, mice were weighed and injected with saline immediately prior to being placed on white-paper floors for a single 5-min session. During conditioning, mice were weighed and given a single injection of either 2 g/kg EtOH (CS+ days) or saline (CS− days) immediately before being placed in the apparatus on the appropriate floor. Twenty-four h after the last conditioning session, mice were gently scruffed and received a single 100 nL microinfusion of either M+B or saline over the course of 60 sec, with an additional 30 sec allocated to complete diffusion of the drug or vehicle. Previous work in our lab has shown that handling immediately prior to preference testing has no effect on CPP expression (Bechtholt, Gremel, & Cunningham, 2004), and as such mice received a saline injection and were placed in the apparatus immediately after the microinfusion for a 30 min test on a floor made up of both floor types. Following the preference test, mice were euthanized, and brains were collected for placement verification.

Experiment 3: Ethanol-induced hypothermia and c-Fos in the EWcp

General Procedure

Following recovery from surgery, mice were assigned to either the M+B or vehicle group, as well as to either EtOH or saline subgroups. Mice were then exposed to a single habituation day in which mice were gently scruffed, pierced, and handled for 90 s so as to habituate all animals to the type and duration of handling that mice are exposed to during microinfusions.

Experiment 3: Inhibition of the EWcp and ethanol-induced hypothermia

Exp. 3 examined the effect of acute inactivation of the EWcp on ethanol-induced hypothermia using parameters based on previous studies (Bachtell et al., 2004). Following the habituation session, baseline rectal temperatures were assessed, immediately followed by a 100 nL microinfusion of either M+B (n = 12) or vehicle (n = 12) into the EWcp. Thirty min later, rectal temperatures were once again assessed, followed by an IP injection of either EtOH (3 g/kg, 18.75 mL/kg) or saline. Rectal temperatures were subsequently measured 15 and 30 min after the EtOH injection (45 and 60 min after microinfusion). Mice were then euthanized, and brains were collected for placement verification.

Experiment 4: Inhibition of the EWcp and ethanol-induced c-Fos in the EWcp

Exp. 4 was conducted to determine if inhibition of the EWcp via M+B was sufficient to prevent the EtOH-induced increase in c-Fos expression reported in previous studies (Bachtell et al., 2002, Smith et al., 2016). Twenty-four h following the habituation session, mice received a single 100 nl microinfusion into the EWcp of either M+B (n = 6) or vehicle (n = 4), followed by an IP injection of EtOH (2 g/kg, 12.5 mL/kg) 30 min later. Ninety min after the microinfusion, mice were sacrificed, and brains were collected for c-Fos immunohistochemistry.

Histology

Brain Extraction and Processing

In all experiments, mice were sacrificed via CO₂ inhalation, brains were extracted, post-fixed for 24 h in 2% paraformaldehyde/phosphate-buffered saline (PBS) and cryopreserved in 20–30% sucrose/PBS. Brains were sliced at 30 μm and were processed for placement verification and c-Fos immunohistochemistry.

Verification of Cannula Placement

For all experiments, coronal sections containing the EWcp (−3.16 mm to −4.0 mm from bregma) were collected and stained with 0.1% thionin in order to verify cannula placements.

c-Fos Immunohistochemistry

For each animal, 5–7 sections encompassing the EWcp were washed three times in PBS and then incubated with 0.3% peroxide in PBS. Sections were then blocked in 5% normal goat serum in PBS and 0.3% Triton X-100, followed by an overnight incubation with a 1:15000 rabbit polyclonal c-Fos antibody (Sigma-Aldrich, MO) in PBS/Triton X-100 and 0.1% bovine serum albumin. The following day, sections were incubated in a biotinylated anti-rabbit secondary. Finally, a Vectastain ABC Kit (Vector, CA) and a metal enhanced DAB Kit (Thermo Scientific, MA) were used. Slices were mounted, dehydrated and then coverslipped.

Statistical Analysis

Activity data for Exp. 1 and 2 were averaged across trials. For both experiments, activity was analyzed via a two-way ANOVA with Treatment (M+B vs. Vehicle) as a between-group factor and Trial Type (CS+ vs. CS−) as a within-group factor. Place preference was defined as a significant difference in time spent on the grid floor between conditioning subgroups. Time spent on the grid floor was analyzed using a two-way ANOVA with Treatment (M+B vs. Vehicle) and Conditioning Subgroup (G+ vs. G−) as between-group factors. Locomotor activity during the preference test was analyzed using a one-way ANOVA using Treatment as the factor. For Exp. 3, temperature measurements were converted into a change-from-baseline value, and these data were analyzed using a two-way ANOVA for the first time point (30 min after mice received the microinfusion), with Microinjection (M+B vs. Vehicle) and Treatment (EtOH vs. Saline) as between-group factors. The two time points after the IP injection of either EtOH or saline were analyzed via a three-way repeated measures (RM) ANOVA with Time as the within-subjects factor, and Microinjection, and Treatment as the between-subjects factors. For Exp. 4, a single c-Fos cell-count value was calculated for each animal by averaging the cell counts across the 5–7 sections that encompassed the EWcp in each animal. Thus, each animal had a single c-Fos cell count value for the EWcp. C-Fos data were analyzed using a one-way ANOVA with Microinjection (M+B vs. M+B Miss vs. Vehicle) as the between group factor. F statistics, p values, effect sizes (eta squared or partial eta squared values, where appropriate) and 95% confidence intervals (CI) of differences are reported within the results sections of each experiment.

Results

Experiment 1: Inhibition of the EWcp during CPP acquisition

Cannula Placement

Fig. 1C depicts the placements of the microinfusion injectors within the EWcp, based on thionin staining conducted following the conclusion of the experiment. Thirteen mice were excluded from the analysis due to lost headcap, while five mice were excluded for an incorrect placement of the injector. The final sizes for conditioning subgroups ranged from 6–8 per subgroup. Exact group numbers are depicted within individual bar graphs.

Fig. 1. Procedural timeline and successful cannulation of the EWcp in Exp. 1.

A cannula was placed above the EWcp, and following recovery, mice received either 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or saline prior to CS+ trials. A) Procedural timeline for Exp. 1. B) Brightfield image of thionin staining used to confirm cannula placement location. C) Cannula location placements for mice included in the study. Scale bar denotes 500 μM.

Conditioning Activity

Mean conditioning activity rates collapsed across the two conditioning trials of each type are depicted in Fig. 2A. Pharmacological inhibition of the EWcp during CS+ trials throughout conditioning did not alter the stimulatory effects of ethanol, nor did it alter general locomotion during CS- trials. Indeed, mice in both groups were significantly more active during CS+ trials, compared to CS− trials, and the differences in activity between trial types did not differ between groups. A two-way ANOVA (Trial Type x Treatment) confirmed these observations, as only a significant main effect of Trial Type was detected [F(1, 56) = 138.0, p < .0001, η² = .71, 95% CI (−114.8, −81.36)]. Neither the main effect of Treatment (p = .71) nor the Trial Type x Treatment interaction (p = .61) was significant.

Fig. 2. Pharmacological Inhibition of the EWcp does not alter the acquisition of EtOH-CPP.

Following recovery from stereotaxic surgery and cannula implantation, mice received a single microinfusion of 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or vehicle into the EWcp prior to CS+ trials. A) Mice were more active during CS+ trials than during CS-trials, and no differences in activity were detected between M+B and vehicle groups. B) A history of M+B did not alter locomotor activity during the preference test. C). Significant place preference was detected in both the M+B and vehicle groups, but there was no effect of M+B on CPP.

Preference Test

Fig. 2B–C depict the mean activity counts and times spent on the grid floor during the preference test. As can be seen, pharmacological inhibition of the EWcp during the acquisition phase did not have any effect development of place preference, or on locomotor activity during the preference test. No significant differences in test activity rates between mice treated with vehicle and M+B during conditioning were detected during the test. A two-way ANOVA (Conditioning Subgroup X Treatment) revealed a main effect of Conditioning Subgroup [F(1, 26) = 36.6, p < .001, η² = .55, 95% CI (16.52, 33.51)], indicating that mice in the G+ subgroup spent significantly more time on the grid floor than G− mice, regardless of the treatment they received during conditioning. The main effect of Treatment (p = .68) and the Conditioning Subgroup X Treatment interaction (p = .13) were not significant.

Experiment 2: Inhibition of the EWcp during CPP expression

Cannula Placement

Microinfusion injector placements within the EWcp are shown in Fig. 3C. In Exp. 2, 20 mice were excluded from the final analysis for lost headcap (n = 9), histological error (n = 2), or for incorrect placement of the injector (n = 9). Final sizes for conditioning subgroups ranged from 8–10 per subgroup. Exact group numbers are depicted within individual bar graphs.

Fig. 3. Procedural timeline and successful cannulation of the EWcp in Exp 2.

A cannula was placed above the EWcp, and following recovery, mice received either 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or saline immediately prior to the preference test. A) Procedural timeline for Exp. 1. B) Brightfield image of thionin staining used to confirm cannula placement location. C) Cannula location placements for mice included in the study. Scale bar denotes 500 μM.

Conditioning Activity

Mean conditioning activity rates collapsed across the four conditioning trials of each type are shown in Fig. 4A. Mice were significantly more active on CS+ trials, compared to CS− trials. Additionally, mice assigned to receive either M+B or saline during the preference test did not differ in their sensitivity to the stimulatory effects of ethanol during conditioning. Indeed, a two-way ANOVA (Trial Type x Treatment) revealed a main effect of Trial Type [F(1, 68) = 648.1, p < .0001, η² = .90, 95% CI (−137.8, −117.7)], but not of Treatment (p = .59). No significant Trial Type x Treatment interaction was detected (p = .67).

Fig. 4. Pharmacological Inhibition of the EWcp does not alter the expression of EtOH-CPP.

Following recovery from stereotaxic surgery and cannula implantation, mice received a single microinfusion of 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or vehicle into the EWcp prior to the place preference test. A) Mice were significantly more active during CS+ trials, compared to CS- trials. No differences were detected between groups assigned to receive M+B or vehicle during the preference test. B). Pharmacological inhibition of the EWcp did not alter locomotor activity during the preference test. C. Significant place preference was detected in both the M+B- and vehicle-treated groups, but inhibition of the EWcp did not alter the expression of CPP.

Preference Test

Mean activity counts as well as mean time spent on the grid floor during the 30-min preference test are presented in Fig. 4B–C. As can be seen, inhibition of the EWcp during preference testing had no effect on locomotor activity or on CPP expression. Indeed, the saline and M+B groups both displayed significant place preference that did not differ from one another. In support of this, a two-way ANOVA (Conditioning Subgroup X Treatment) revealed a main effect of Conditioning Subgroup [F (1, 32) = 37.0, p < .001, η² = .51, 95% CI (19.78, 39.71)], demonstrating that the amount of time spent on the grid floor between G+ and G− subgroups was significantly different in both treatment groups. The main effect of Treatment (p = .41) and the Conditioning Subgroup X Treatment interaction (p = .19) were not significant. Lastly, activity rates between the two groups did not differ significantly during the 30-min preference test indicating that inhibition of the EWcp during expression testing did not alter locomotor activity.

Experiment 3: Inhibition of the EWcp and ethanol-induced hypothermia

Cannula Placement

Microinfusion injector locations within the EWcp are shown in Fig. 5A. Three mice were excluded from the analysis due either to lost headcap (n =1) or for incorrect placement of the injector (n = 2). Final group sizes ranged from 5–6 per group. Exact group numbers are depicted within individual bar graphs in Fig. 5C–D.

Fig. 5. Pharmacological inhibition of the EWcp increases body temperature.

Following recovery from stereotaxic surgery and cannula implantation, mice received a single microinfusion of 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or vehicle into the EWcp, 30 min prior to receiving a single IP injection of either 3 g/kg EtOH or saline. Rectal temperatures were measured 30, 45, and 60 min post microinfusion. A) Cannula location placements for mice included in the study. B) Brightfield image of thionin staining used to confirm cannula placement location. C) No differences in baseline temperature were detected between groups. D) Microinfusion of M+B increased body temperature significantly in mice given a saline injection. Scale bar denotes 500 μM.

Body Temperature

As there were no differences in baseline rectal temperature (Fig. 5C) between groups, data for the three time points are presented as change-from-baseline values in Fig. 5D. As can be seen, 30-min after mice received the microinfusion, inhibition of the EWcp via M+B increased body temperature significantly (30 min post MI). When mice were then given an IP injection of 3 g/kg EtOH, body temperatures significantly decreased, compared to mice that received a saline injection (15 min Post IP Inj). In support of this, a two-way ANOVA found that there were no significant differences in baseline temperatures between groups (Microinfusion x Treatment), as no significant main effects or interactions were detected. In contrast, a two-way ANOVA (Microinfusion x Treatment) for temperature 30 min after the microinfusion revealed a significant main effect of Microinfusion, demonstrating that inhibition of the EWcp significantly increased body temperature when compared to vehicle [F(1, 17) = 7.3, p < .05, η² = .28, 95% CI (−0.98, −0.12)]. When the two time points after the IP injection were analyzing using a three-way RM ANOVA (Time x Microinfusion x Treatment) a main effect of Treatment was detected [F(1, 17) = 60.9, p < .001, η_p² = .78], confirming that EtOH significantly decreased body temperature. Additionally, a significant Time x Microinfusion x Treatment three-way interaction was detected [F(1, 17) = 4.5, p < .05, η_p² = 0.21] . In order to better understand this interaction, individual two-way ANOVAs were conducted for each of the two post-IP injection time points (15 min post IP inj, 30 min post IP inj). These ANOVAs showed main effect of Treatment 15 [F(1, 17) = 42.2, p < .0001, η² = .69, 95% CI (1.87, 3.67)] and 30 min [F(1, 17) = 59.7, p < .0001, η² = .73, 95% CI (2.26, 3.96)] after the IP injection, again indicating that the decrease in body temperature in EtOH-treated mice was significant. Furthermore, a significant Microinfusion x Treatment interaction was detected 30 min [F(1, 17) = 6.7, p < .05, η² = .08, 95% CI Vehicle/Saline vs. Vehicle/EtOH (0.70, 3.44), 95% CI M+B/Saline vs M+B/EtOH (2.73, 5.59)], but not 15 min after the EtOH injection. A Bonferroni post-hoc analysis for this interaction revealed that although EtOH significantly decreased body temperature in both microinfused groups, the difference in temperature between EtOH- and saline-treated mice was greater in M+B-treated animals, than in mice that received a vehicle microinfusion. No other significant main effects or interactions were detected at any of the time points.

Experiment 4: Inhibition of the EWcp and ethanol-induced c-Fos

Cannula Placement

One mouse was excluded from the analysis due to a lost headcap. The three mice with incorrect injector placements were included in the analysis as negative controls (M+B Miss). Final group numbers are shown within Fig. 6D.

Fig. 6. Pharmacological inhibition of the EWcp attenuates EtOH-induced c-Fos expression.

Following recovery from stereotaxic surgery and cannula implantation, 100 nL of Muscimol (0.1 mM) + Baclofen (1.0 mM) (M+B) or saline were microinfused into the EWcp 30 min prior to an EtOH injection (2 g/kg IP). A). Representative photomicrograph of cannulae placement and c-Fos expression in vehicle-treated mice. B.) Representative photomicrograph of cannulae placement and c-Fos expression in M+B-treated mice with incorrect cannulae placement. C) Representative photomicrograph of cannulae placement and c-Fos expression in M+B-treated mice D.) Pretreatment with M+B prevents the EtOH-induced increase in c-Fos expression in the EWcp. Scale bar denotes 200 μM.

c-Fos Expression in the EWcp following an EtOH injection

Expression of c-Fos was detected as nuclear immunoreactivity in the EWcp, as expected for a transcription factor. Representative brightfield photographs illustrating c-Fos immunoreactivity are presented in Fig. 6A–C. The mean number of c-Fos positive cells in the EWcp in mice that received a microinfusion of either M+B or vehicle prior to an EtOH injection are presented in Fig. 6D. As can be seen, inhibition of the EWcp via M+B 15 min prior to a 2 g/kg IP EtOH injection reduced the number of c-Fos positive cells in the EWcp. Importantly, this effect was not seen in mice with incorrect injector placements (M+B Miss). In support of this, a one-way ANOVA revealed a main effect of Treatment [F(2, 6) = 119.5, p < .0001, η² = 0.98]. A Bonferroni post-hoc analysis revealed that the M+B group was significantly different than both the Vehicle [p < .0001, 95% CI (18.56, 28.89)], and the M+B Miss [p < .0001, 95% CI (15.96, 26.29)] group, but that the Vehicle and M+B Miss groups were not significantly different [p = .52, 95% CI (−2.563, 7.766)].

Discussion

The studies presented here demonstrate that intra-EWcp microinfusion of the GABAA and GABAB agonists muscimol and baclofen (M+B) during the conditioning or preference test phases of CPP was not sufficient to alter place preference. Additionally, these studies show that inhibition of the EWcp increases body temperature, indicating that the EWcp may be involved in temperature regulation. Lastly, we found that microinfusion of M+B into the EWcp attenuated the increased expression of c-Fos following an ethanol injection.

To date, these are the first studies examining the effect of acute pharmacological inhibition of the EWcp on the acquisition and expression of EtOH-induced CPP. Here, we report that inhibition of the EWcp does not alter either the acquisition or the expression of EtOH-induced CPP. It could be theorized that inhibition of the EWcp during the acquisition phase did not alter the development of place preference because the 30-min delay between the microinfusion and the conditioning trial was too long, and as such lower concentrations of the drug were present in the EWcp during conditioning. The fact that we see decreased c-Fos expression in M+B-treated mice (Exp. 4) following an EtOH injection (using the same dose and timeline) suggests that this is not the case. In regard to the expression of CPP, it is possible that inhibition of the EWcp following 4 conditioning trials is simply not sufficient to alter place preference due to the fact that place preference is near-asymptotic after that many conditioning trials (Groblewski, Bax, & Cunningham, 2008). Future experiments inhibiting the EWcp during preference testing but following fewer trials could help elucidate this possibility. When considered in light of the numerous reports showing that the EWcp is involved in the regulation of voluntary EtOH intake our data suggest that the EWcp may not be involved in mediating conditioned responses to EtOH-paired cues. In support of this, when c-Fos expression across the brain was analyzed following exposure to EtOH or an EtOH-paired cue, the EWcp showed increased c-Fos expression after exposure to EtOH, but not following exposure to the EtOH-paired cue in the absence of EtOH (Hill, Ryabinin, & Cunningham, 2007). Interestingly, our data are somewhat at odds with previous work showing that mice lacking Ucn1 do not develop place preference for an EtOH-paired floor (Giardino et al., 2011). The use of different genotypes between studies, as well as the compensatory mechanisms associated with KO models make it difficult to compare these two findings. Even so, the potential involvement of the various neuropeptides expressed in the EWcp in conditioned behaviors, as well the disassociation between their involvement in the self-administration of EtOH, requires further attention moving forward.

Our finding that inhibition of the EWcp increases body temperature adds to previous literature that has implicated the EWcp in thermoregulation. Indeed, acute exposure to warm (34º C) or cold (10º C) ambient temperatures increases c-Fos expression in the EWcp (Bachtell et al., 2003). Furthermore, in the male F2 offspring of C57BL/6J and DBA/2J mice, mice with higher Ucn1 expression in the EWcp had higher baseline body temperatures and showed a stronger hypothermic response to EtOH (Bachtell et al., 2002b). Lastly, electrolytic lesions of the EWcp blunt the hypothermic response following an injection of EtOH (Bachtell et al., 2004). Given the known Ucn1-positive projections to the DRN, and the known involvement of DRN in thermoregulation (Ginefri-Gayet & Gayet, 1993; Hale, Dady, Evans, & Lowry, 2011; Ishiwata et al., 2016), one is tempted to hypothesize that the hyperthermic results presented here are mediated through an EWcp-DRN mechanism. This specific pathway has previously been shown to be involved in modulating body temperature, as intra-DRN microinfusion of Ucn1 significantly decreases body temperature (Turek & Ryabinin, 2005a). Furthermore, it has been hypothesized that EtOH-induced hypothermia may be mediated by inhibitory effects of Ucn1 on the DRN (Turek & Ryabinin, 2005b). Thus, our finding that acute inhibition of the EWcp increases body temperature, in combination with the hypothermic effects of Ucn1 in the DRN, suggests that the EWcp may be acting to regulate body temperature by modulating neuronal activity in the DRN. Our finding that an EtOH injection resulted in a strong hypothermic response even in the M+B group adds wide range of literature showing a similar effect of EtOH in other rodents (Lomax, Bajorek, Chesarek, & Chaffee, 1980) and mammals (Murphy & Lipton, 1983). Interestingly, we found a greater difference in body temperature between EtOH- and saline-treated mice that had M+B microinfused into the EWcp, compared to those that received a vehicle microinfusion. It is possible that this finding is simply due to the fact that M+B increased body temperature, and therefore resulted in a greater difference between EtOH and saline treated mice. It is also possible however, that inhibition of the EWcp may potentiate EtOH-mediated changes in temperature. Subsequent studies in which the EWcp is inhibited and mice are exposed to a body temperature challenge, such as a water bath immersion (Mou, Wilgenburg, Lee, & Hallenbeck, 2013), may provide additional insight into the effects of EWcp inhibition on temperature regulation.

Robust increases in c-Fos expression in the EWcp following administration of EtOH have been reported numerous times, across various models and species (Anacker et al., 2011; Bachtell et al., 1999; Ryabinin et al., 2003, 1999; Sharpe, Tsivkovskaia, & Ryabinin, 2005; Smith et al., 2016; Walcott & Ryabinin, 2017). Here, we report that microinfusion of GABAA and GABAB agonists into the EWcp prior to an ethanol injection prevents this increase in c-Fos expression. The GABAA receptor agonist muscimol has been shown to decrease the firing activity of neurons in the locus coeruleus (Jin, Cui, Zhong, Jin, & Jiang, 2013), as well as in midbrain regions, including the DRN (Kim et al., 2018) and the VTA (Theile, Morikawa, Gonzales, & Morrisett, 2011). Similarly, decreases in cell firing rates, as well as in burst activity have been observed when the GABAB receptor agonist baclofen has been applied to cells in the substantia nigra (Engberg, Kling-Petersen, & Nissbrandt, 1993; Erhardt, Andersson, Nissbrandt, & Engberg, 1998) and VTA (Chen, Phillips, Minton, & Sher, 2005), as well as direct hyperpolarization of neurons when applied in the DRN (Chieng & Christie, 1995). Given these findings, we hypothesized that microinfusion of a combination of muscimol and baclofen would have similar effects on neurons in the EWcp. Previous work has suggested that increased c-Fos expression in the EWcp following an EtOH injection is mediated through activation of the MEK1–2 and ERK1–2 pathway (Bachtell et al., 2002a). As it is known that Ca²⁺ influx, either through voltage-dependent ion channels or receptor-mediated Ca²⁺ channels, activates ERKs through its actions on the Ras pathway (Finkbeiner & Greenberg, 1996; Grewal, York, & Stork, 1999; Rosen, Ginty, Weber, & Greenberg, 1994), GABAA and GABAB receptor activation in the EWcp via M+B could prevent c-Fos expression by preventing Ca²⁺ influx. Somewhat in contrast to our findings, previous work had led to the hypothesis that GABAA receptor activation is necessary for increased c-Fos expression in EWcp neurons, as inhibition of GABAA receptors prevented EtOH-induced c-Fos expression in the EWcp (Bachtell et al., 2002a). Furthermore, GABAA positive modulators also result in increased c-Fos expression in the EWcp, although at lower levels than EtOH (Bachtell et al., 2002a). Importantly however, both the GABAA inhibitors and positive modulators were administered systemically via an IP injection in this previous study. That systemic inhibition of GABAA receptors prevents EtOH-induced c-Fos expression, in a similar manner as intra-EWcp inhibition via muscimol and baclofen suggests that EtOH’s ability to increase c-Fos in the EWcp must be mediated by upstream regions. Systemic inhibition of GABAA receptors may prevent EtOH’s actions in brain regions that project to the EWcp, thus preventing the EtOH-induced c-Fos expression.

As the goal of this experiment was to determine if inhibition of the EWcp was sufficient to prevent EtOH-induced c-Fos expression in this area, we did not include an additional control group that received a saline injection instead of EtOH. The EWcp shows minimal c-Fos expression following a saline injection (Hill et al., 2007), even when this is done in a novel context (Ryabinin et al., 1997). Additionally, restraint stress and LPS do not induce c-Fos expression in the EWcp, indicating that c-Fos expression in this area is preferentially sensitive to EtOH and several other drugs of abuse (Turek & Ryabinin, 2005a). The lack of c-Fos expression following a saline injection would therefore not allow for the observation of M+B-mediated changes in c-Fos expression. Although c-Fos is readily used as a marker of neuronal activity, it is important to note that there are several limitations associated with its use. First, it is possible that a stimulus activates a neuron without c-Fos being expressed (Figueiredo, Bodie, Tauchi, Dolgas, & Herman, 2003). Similarly, neuronal depolarization and an increase in the firing rate of a neuron may not be enough to induce c-Fos in every type of neuron (Luckman, Dyball, & Leng, 1994). Additionally, c-Fos expression in a neuron cannot be used to indicate if this activation is through the direct or indirect effect of a stimulus. Lastly, c-Fos expression can be non-specific to the stimulus as it has a rather low temporal resolution (Morgan, Cohen, Hempstead, & Curran, 1987). Future work in which electrophysiological recordings are obtained from EWcp neurons may provide further information on the interactions between EtOH and M+B in these neurons.

In summary, the present studies demonstrate that pharmacological inhibition of the EWcp is not sufficient to alter EtOH-induced CPP, suggesting that the EWcp may play a more selective role in regulating voluntary EtOH intake, and not conditioned EtOH-seeking behaviors. Additionally, our data provide further evidence for the EWcp’s involvement in thermoregulation, as inhibition via a microinfusion of M+B significantly increased body temperature. Lastly, we report that inhibition of the EWcp prevents EtOH-induced increased in c-Fos expression, further elucidating the mechanisms that may mediate EtOH’s effects in the EWcp.

Highlights.

• Centrally-projecting Edinger-Westphal nucleus inhibition doesn’t alter place preference

• Centrally-projecting Edinger-Westphal is involved in thermoregulation

• c-Fos expression is altered following Edinger-Westphal nucleus inhibition