Intermittent ethanol access increases sensitivity to social defeat stress

Abstract

Background:

Comorbidity between alcoholism and depression is extremely common. Recent evidence supports a relationship between alcohol exposure and stress sensitivity, an underlying factor in the development of depression. Our lab has recently shown that chronic alcohol gavage increases sensitivity to social defeat stress (SDS). However, the effects of voluntary alcohol consumption, resulting from protocols such as intermittent ethanol access (IEA), on defeat stress sensitivity have yet to be elucidated.

Methods:

We first assessed the effects of 4 weeks of IEA to 20% alcohol on sensitivity to subthreshold SDS exposure. Next, to examine neuroinflammatory mechanisms, we analyzed gene expression of inhibitor of NFkB (IkB) following IEA or chronic alcohol exposure (10 days of 3.0g/kg alcohol via intragastric gavage). Then, we quantified NFkB activation via β-galactosidase immunohistochemistry following IEA or chronic alcohol gavage in NFkB-LacZ mice.

Results:

IEA-exposed mice displayed an increase in sensitivity to subthreshold SDS compared to water-drinking controls. We also found that IkB gene expression was decreased in the nucleus accumbens (NAC) and amygdala (AMY) following IEA but was not altered following chronic alcohol gavage. Finally, we observed increased NFkB activity in the central amygdala (CEA), basolateral amygdala (BLA), and medial amygdala (MEA) after IEA, and increased NFkB activity solely in the CEA following chronic alcohol gavage.

Conclusions:

These findings further corroborate that prior alcohol exposure, in this case intermittent voluntary consumption, can impact development of depressive-like behavior by altering stress sensitivity. Furthermore, our results suggest the CEA as a potential mediator of alcohol’s effects on stress sensitivity, as NFkB was activated in this region following both IEA and chronic alcohol gavage. Thus, this study provides novel insight on alterations in the NFkB pathway and identifies specific regions to target future in experiments assessing the functional role of NFkB in these processes.

Introduction

A substantial portion of alcohol-dependent individuals also meet clinical criteria for depression (27.9%) or anxiety (36.9%)(Kessler et al., 1996). Epidemiological data have indicated that the presence of alcohol use disorder (AUD) and severity of AUD symptoms predicts onset of depressive disorders (Boschloo et al., 2012, Boden and Fergusson, 2011), and previous alcohol misuse in young adulthood associates with increased rates of major depressive disorder (MDD) (Fergusson et al., 2013, Boden and Fergusson, 2011). Comorbid depression in detoxified alcohol abusers has been linked to increased risk for relapse (Driessen et al., 2001), as co-occurrence of these disorders can lead to increased symptom severity and impairment (Prior et al., 2017). As such, comorbidity of AUD and MDD presents a significant health burden within our society. The circuitry and cellular mechanisms intertwining these disorders must be further understood in order to develop effective therapeutics for patients displaying this comorbidity.

The relationship between chronic alcohol exposure and alterations in sensitivity to subsequent stressors has been extensively studied. Clinically, alcohol withdrawal can induce a depressed mood and negative affect that can persist into abstinence and increases risk of relapse (Heilig et al., 2010, Sinha et al., 2009, Koob, 2003, Greenfield et al., 1998, Witkiewitz and Villarroel, 2009, Koob, 2015, Heilig and Koob, 2007). In animal models, exposure to chronic alcohol results in increased reactivity in various paradigms that assess stress sensitivity and anxiety-like behavior, including the Vogel test of punished drinking (Sommer et al., 2008), elevated plus maze (Valdez et al., 2003, Valdez et al., 2002, Perez and De Biasi, 2015), social interaction (SI) following restraint stress (Breese et al., 2005), open field test (Perez and De Biasi, 2015), marble burying (Perez and De Biasi, 2015), and fear extinction (Holmes et al., 2012). In addition to increasing stress-sensitivity, a history of alcohol exposure also potentiates the ability of stress exposure to increase alcohol consumption and self-administration (Sommer et al., 2008, Lopez et al., 2016, Anderson et al., 2016a, Anderson et al., 2016b, Becker, 2012, Griffin et al., 2014, Valdez et al., 2002).

Our lab and others use the social defeat stress (SDS) model to assess the bidirectional relationship of alcohol exposure and stress sensitivity. We have previously found that exposure to chronic alcohol gavage increases sensitivity to SDS (Nelson et al., 2018). This relationship is bidirectional, as exposure to chronic SDS can impact drinking behaviors, including consumption (Hwa et al., 2016, Newman et al., 2018a, Newman et al., 2018b, Newman et al., 2018c, Norman et al., 2015, Croft et al., 2005, Karlsson et al., 2017, Dong et al., 2011, Nelson et al., 2018), conditioned place preference (Macedo et al., 2018), self-administration (Rodriguez-Arias et al., 2016, Caldwell and Riccio, 2010), and motivation to seek alcohol (Rodriguez-Arias et al., 2016). While we have examined the effect of experimenter-delivered chronic alcohol on sensitivity to SDS, it is important to determine if voluntary alcohol consumption could have a similar effect in mice. To address this question, we used intermittent ethanol access (IEA), which is a limited two-bottle choice (2BC) access model that results in elevated alcohol consumption compared to that observed on a continuous access schedule (Rosenwasser et al., 2013, Crabbe et al., 2011, Wise, 1973, Simms et al., 2008). When given IEA access, C57BL6/J mice reach consumption levels stabilizing around 20g/kg/day, reach blood alcohol levels above 100mg/dl, and display higher ethanol preference compared to mice on continuous access 2BC (Hwa et al., 2011). As such, IEA results in consumption patterns which may be more similar to drinking behaviors observed in human alcohol abusers (Rosenwasser et al., 2013).

To examine a cellular mechanism that could underlie the relationship between alcohol exposure and stress sensitivity, we assessed the effects of alcohol exposure on activation of the transcription factor nuclear factor light chain enhancer of activated B cells (NFkB). Under baseline conditions, NFkB subunit dimers are bound in the cytosol by inhibitor of NFkB (IkB)(Nennig and Schank, 2017). A protein kinase known as IkB kinase (IKK) phosphorylates IkB, tagging it for proteasomal degradation and releasing the NFkB subunit dimers to translocate to the nucleus where they act as a transcription factor for a variety of genes involved in inflammation and various other processes (Nennig and Schank, 2017). Several studies have examined the influence of the NFkB pathway on alcohol-related behaviors including continuous 2BC (Truitt et al., 2016), drinking-in-the-dark (Truitt et al., 2016), and conditioned place preference (Nennig et al., 2017), as well as the immune-related effects following long-term ethanol self-administration (Helms et al., 2012). Clinically, a single nucleotide polymorphism in the NFKB1 gene, which encodes one of the NFkB subunits, associates with alcohol dependence (Edenberg et al., 2008), and the NFkB system is dysregulated following chronic alcohol abuse (Okvist et al., 2007, Yakovleva et al., 2011).

The NFkB pathway has also been implicated in the behavioral and molecular outcomes following chronic SDS (Christoffel et al., 2011, Christoffel et al., 2012). Specifically, protein levels of IKK, IkB, and phosphorylated IkB are increased in the nucleus accumbens (NAC) of mice sensitive to SDS. Additionally, infusion of an IKK dominant-negative virus into the NAC of susceptible mice resulted in a reversal of defeat-induced behavioral phenotypes, whereas infusion of a constitutively active form of IKK into the accumbens increased sensitivity to SDS exposure (Christoffel et al., 2012).

In the present study we aimed to assess the effects of IEA on stress sensitivity and NFkB pathway activity. First, we examined the effects of IEA on sensitivity to subthreshold SDS. Then, we then analyzed gene expression of IkB and regionally specific NFkB activation following IEA. Last, we assessed gene expression of IkB and regionally specific NFkB activation following chronic alcohol gavage to determine if NFkB is activated in similar regions as those affected by IEA. We hypothesized that IEA would increase sensitivity to social stress, and that activity of the NFkB pathway would be increased following both IEA and chronic alcohol gavage.

Materials & Methods

Animals:

Male C57BL6/J mice (8 weeks of age, Jackson Laboratory, Bar Harbor, ME) were used in Experiments 1, 2, and 4. Male NFkB-LacZ mice (8 weeks of age) bred in house on a C57BL6/J background were used for Experiments 3 and 5. NFkB-LacZ mice express a lacZ transgene under the direction of an NFkB-regulated promoter. Thus, wherever NFkB is activated, activity-dependent β-galactosidase (β-gal) will be expressed in that cellular population (Bhakar et al., 2002, Russo et al., 2009, Nennig et al., 2017). Retired breeder male CD-1 mice (4–5 months of age, Charles River, Wilmington, MA) were used as SDS aggressors in Experiment 1. Due to sex specific differences in territorial aggression and the SDS protocol being replicated and highly validated using male C57BL6/J mice, only male subjects were used. Mice were allowed 1 week of habituation to the UGA College of Veterinary Medicine vivarium before experiments began. Mice were housed in a 12hr light cycle (on at 1:00 and off at 13:00). Food and water were available ad libitum. All procedures were in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of the University of Georgia.

Drugs:

For continuous access 2-bottle choice (2BC) in Experiments 2 and 3, 95% ethanol (Decon Labs, Inc., King of Prussia, PA) was diluted to a 20% v/v solution in tap water. For chronic alcohol gavage in Experiments 4 and 5, 95% ethanol was diluted to a 25% v/v solution in tap water.

IEA:

One week prior to the start of IEA, mice were singly housed. Mice were then presented with two bottles in their homecage, one containing water and one containing a 20% ethanol solution for 24 hours on Monday, Wednesday, and Friday for a total of 4 weeks. On the remaining days of the week (Tuesday, Thursday, Saturday, and Sunday), the alcohol bottle was removed and replaced with a bottle containing water. Control mice were also singly housed but had access to water only throughout the 4-week period. Bottles were weighed at the same time each day and g/kg consumption was calculated based on each mouse’s body weight.

Chronic EtOH exposure:

Mice were intragastrically gavaged with 3.0g/kg ethanol (25% ethanol in tap water) for a total of ten days. Control mice were gavaged with tap water. Gavages occurred at the same time each day throughout the exposure period. This method of exposure was chosen because it allows for tight experimenter control compared to voluntary access.

Blood Ethanol Concentration (BEC) analysis: (IEA):

Mice underwent intermittent access ethanol two-bottle choice (IEA) as described above for two weeks (six sessions). Ethanol (20% v/v) was introduced Monday, Wednesday, Friday and water only was available on the remaining days. On the 6^th IEA session, mice were sacrificed 2 hours into the dark cycle via live decapitation. (Gavage): Mice were intragastrically gavaged with 3g/kg ethanol (25%, v/v diluted in tap water) and sacrificed via live decapitation 30 minutes later. Trunk blood was collected for BEC analysis. Samples were collected in heparinized microcentrifuge tubes and spun at 10,000 x g. Serum was then transferred to a new microcentrifuge tube and stored at 4° until analysis. Samples were analyzed using an Analox AM1 Alcohol Analyzer (Analox Instruments, Stourbridge, England, UK) according to the manufacturer’s instructions and were quantified in g/dl.

Subthreshold SDS:

Subthreshold SDS is a modification of the chronic SDS protocol described by Golden et al. 2011. Briefly, chronic SDS is a major preclinical model of depression that results in depressive-like behavior in rodents, including anhedonia, social avoidance, weight loss, and immune suppression (Golden et al., 2011, Jasnow et al., 2001, Krishnan et al., 2007). These behavioral and physiological outcomes mirror depressive symptoms at the clinical level, as three DSM-5 criteria for depression include anhedonia, weight alterations, and a marked reduction in normal social functioning (APA, 2013). The SDS model involves the use of aggressive male CD-1 retired breeder mice that are housed on one side of a large cage that is separated in half by a clear, perforated divider. SDS involves repeated exposure to brief physical defeat sessions by a novel CD-1 mouse each day of the 10-day protocol, after which defeated mice are moved to the other side of the divider and housed overnight next to the aggressor it just encountered. As such, SDS consists of brief physical stress and prolonged emotional stress. In the subthreshold protocol, mice are exposed to 3 brief defeats on a single day, and are not housed next to an aggressive animal. Because the full 10 day SDS exposure results in a majority of animals showing stress sensitivity, subthreshold SDS is a valuable method for detecting increases in stress sensitivity because it reduces the risk of a ceiling effect.

Prior to subthreshold SDS, CD-1 mice were screened for aggressive behavior by placing a screener C57BL6/J mouse (not used in any of the experiments) into the home cage of the CD-1 mouse for 180 seconds for 4 consecutive days. Aggressors were selected based on the following criteria: the CD-1 mouse must initiate an attack in at least two consecutive sessions and the latency to first display aggressive behavior must be less than 60 seconds. For subthreshold SDS, mice were exposed to 3 5-minute defeat sessions separated by 15 minutes each. During each defeat session, mice were placed into the homecage of a novel male CD-1 mouse that passed aggression screening. Between defeat sessions, mice were returned to their homecage to recover until the next defeat session started 15 minutes later. Defeats occurred at the end of the light cycle in order to visualize any injuries. If blood was drawn, the defeat session ended early and the time at which the session ended was noted.

SI Test:

To assess depressive-like behavior, mice are tested for social interaction (SI) 24 hours after the final defeat session (Golden et al., 2011). The SI test occurred at the beginning of the dark cycle approximately 24 hours after subthreshold SDS exposure. This test consisted of 2 150-second trials separated by 30 seconds, the first without a novel CD-1 target mouse present, and the second with a target mouse present in an enclosure in the predetermined interaction zone. Time spent in the interaction zone during each trial is recorded and divided (time in the interaction zone with the target mouse present divided by time when the target was absent) to obtain the SI ratio. Two phenotypic subpopulations arise following exposure to the chronic SDS protocol in mice: those with an SI ratio less than 1.0 and are thus “susceptible” to SDS, and those “resilient” to this stressor that display SI ratios over 1.0 (Golden et al., 2011, Krishnan et al., 2007). Following the chronic SDS protocol, typically 70% of mice are of the susceptible phenotype, while 30% are considered resilient to this stressor and behavior similarly to nonstressed controls. The one day subthreshold SDS protocol does not decrease social interaction in naive mice, thus allowing for the detection of increases in sensitivity following specific interventions, such as IEA or chronic alcohol gavage (Golden et al., 2011). SI tests were scored by an observer blind to the experimental group of each mouse.

qPCR:

Mice were sacrificed 48-hours after the last IEA drinking session or chronic alcohol treatment via rapid decapitation and gross dissections were taken of the amygdala (AMY), NAC, and dorsal striatum. Tissue was snap frozen on isopentane and stored at −80°C in RNAase free tubes. Total RNA was extracted and reversed transcribed using a first-strand cDNA synthesis kit (Invitrogen, Carlsbad, CA, USA). qPCR reactions were run in triplicate using specific FAM-labeled TaqMan probes (Gapdh Mm99999915_g1, IKB Mm00456853_m1; Applied Biosystems, Foster City, CA, USA) and ran on an Applied Biosystems QuantStudio6 machine. Reactions were normalized to endogenous control Gapdh and expressed as fold change from control. For IEA, 3 NAC samples (1 IEA, 2 water) and 1 IEA AMY sample were excluded due to insufficient sample quality to conduct qPCR. For chronic alcohol gavage, 1 water NAC sample and 3 AMY samples (2 alcohol, 1 water) were excluded for the same reason.

Immunohistochemistry:

NFkB-LacZ reporter mice were used for immunohistochemistry. Forty-eight hours after the last IEA drinking session or chronic alcohol gavage treatment, mice were overdosed with ketamine/xylazine and were transcardially perfused with 4% paraformaldehyde. Brains were postfixed overnight in 4% paraformaldehyde, transferred to 10% sucrose for one hour and then 30% sucrose until the tissue sank, and then frozen quickly on powdered dry ice. Thirty micron sections were collected in a freezing cryostat (Leica, Buffalo Grove, IL) and floating sections were frozen in cryopreservant at −20°C. After washing, β-galactosidase expression was visualized using X-gal (Roche Diagnostics, Germany) immunohistochemistry to examine NFkB activation. Briefly, after 3 10-minute PBS washes, sections were incubated overnight at 37° in X-gal reaction buffer (X-gal diluted 1:25 in warmed X-gal dilution buffer containing 100mM sodium phosphate dibasic (Fisher Scientific, Fair Lawn, NJ, USA), 100mM sodium chloride (Fisher Scientific, Fair Lawn, NJ, USA), 5mM ethylene glycol-bis(2-aminoethylether)-N,N,N’,N’-tetraacetic acid (EGTA) (Sigma-Aldrich, St. Louis, MO, USA), 2mM magnesium chloride (Sigma-Aldrich, St. Louis, MO, USA), 5mM potassium hexacyanoferrate(III) (Sigma-Aldrich, St. Louis, MO, USA), 5mM potassium hexacyanoferrate(II) trihydrate (Sigma-Aldrich, St. Louis, MO, USA) in 0.2% Triton X-100 (Sigma-Aldrich, St. Louis, MO, USA)). The next day, sections are washed in PBS 3 times for 10 minutes after which sections were mounted on gelatin slides. Neutral red solution (Sigma-Aldrich, St. Louis, MO, USA) was used prior to the ethanol dehydration series as a counter stain to improve visualization of landmarks for imaging. Slides were coverslipped and images were taken on a Zeiss Axioscope A1. For NAC shell, images from 6 fields (3 per side) were taken at 40x magnification. For NAC core, images from 4 fields (2 per side) were taken at 40x magnification. For the central amygdala (CeA), basolateral amygdala (BLA), and medial amygdala (MEA), images from 2 fields (1 per side) were taken at 20x magnification. X-gal positive cell counts were quantified using ImageJ software and were performed by an investigator blind to the experimental conditions. For each region, the average cell count across the total number of frames taken was used as the dependent variable for comparisons.

Experimental timelines:

A summary timeline for each experiment in this study can be found in Figure 1 and specific protocols are described below. A withdrawal period of 48 hours was chosen for all experiments. This timepoint is at the end of acute withdrawal in rodents, and while the animals may been in emotional withdrawal, the physical withdrawal symptoms and motor abnormalities that may impair a stress response have most likely subsided(Heilig et al., 2010). In Experiment 1, mice were exposed to 4 weeks of IEA. Forty-eight hours after the last drinking session, mice were exposed to subthreshold SDS and assessed for SI 24 hours later. In Experiments 2 and 4, mice were exposed to 4 weeks of IEA (Experiment 2) or chronic alcohol exposure (Experiment 4) and were sacrificed 48 hours after the last drinking session for qPCR. InExperiments 3 and 5, NFkB-LacZ mice were exposed to 4 weeks of IEA (Experiment 3) or chronic alcohol exposure (Experiment 5) and were perfused 48 hours after the last drinking session for immunohistochemistry.

Figure 1.

Experimental timelines for Experiments 1–5.

Statistical Analyses:

Statistical analyses were preformed using Prism GraphPad software. SI ratio, gene expression, and cell count analyses were compared using an unpaired t-test.

Results

Experiment 1: IEA increases sensitivity to subthreshold SDS

As we have previously determined that experimenter-delivered chronic alcohol gavage decreases SI behavior following exposure to subthreshold SDS (Nelson et al., 2018), we first assessed the effects of voluntary alcohol access via IEA on stress sensitivity. Mice (IEA: n=14, water: n=13) were allowed 4 weeks of IEA to 20% ethanol and displayed high levels of consumption (Figure 2A) and preference (Figure 2B). Mice were exposed to subthreshold SDS 48 hours after the final drinking period. We chose this timepoint of stress exposure based on our previous findings using chronic alcohol gavage (Nelson et al., 2018). Control mice had access to water only throughout the 4 weeks. IEA-exposed mice displayed significantly decreased SI following subthreshold SDS compared to water-drinking controls (t(25)=2.53, p=0.018, Figure 2C). No significant correlations were observed between the average consumption or preference displayed in the final week of drinking and SI score (data not shown).

Figure 2. IEA increases sensitivity to subthreshold SDS.

Mice were allowed 4 weeks of IEA to 20% ethanol or water only, and IEA mice displayed high levels of EtOH consumption (A) and preference (B). Forty-eight hours after the last drinking period, mice were exposed to subthreshold SDS and SI was tested 24 hours later. IEA-exposed mice displayed decrease social interaction compared water drinking controls (C). *p˂0.05 compared to control group.

Experiment 2: IEA decreases IkB expression in the AMY and NAC

Next, we were interested in how components of the NFkB signaling pathway, particularly IkB, are altered following IEA. Mice (n=12/group) were exposed to IEA or water for 4 weeks and were sacrificed 48 hours following the final alcohol access period, at which time the NAC and AMY were dissected and prepared for qPCR. Following 4 weeks of IEA, IkB gene expression was significantly decreased in the NAC (t(19)=2.87, p=0.01, Figure 3A) and the AMY (t(21)=3.35, p=0.003, Figure 3B). No significant correlations were observed between the average consumption or preference displayed in the final week of drinking and NAC or AMY IkB gene expression (data not shown).

Figure 3. IEA decreases IkB expression in the AMY and NAC.

Mice were allowed 4 weeks of IEA to 20% ethanol or water only. Forty-eight hours after the last drinking period, mice were sacrificed and the NAC and AMY were dissected and processed for qPCR. IEA-exposed mice displayed decreased IkB gene expression in both the NAC (A) and AMY (B) compared to water drinking controls. **p˂0.01 compared to control group.

Experiment 3: IEA increases NFkB expression in specific subregions of the amygdala

While assessing IkB gene expression can provide indirect insight on how NFkB regulation is altered following these alcohol exposure paradigms, this was performed in homogenates encompassing the entire NAC and AMY. As this method lacks regional specificity, we next assessed NFkB activation in our NFkB-LacZ reporter mice by assessing β-gal expression using X-gal immunohistochemistry. This transgenic mouse provides a functional readout of NFkB activation with greater neuroanatomical specificity. Mice (IEA: n=6, water: n=4) were exposed to 4 weeks of IEA or water and were sacrificed 48-hours following the final drinking session. While no impact of IEA on NFkB activation in the NAC shell (t(8)=1.29, p=0.23, Figure 4A–B) or NAC core (t(7)=0.95, p=0.38, Figure 4C–D) was observed, a significant increase in X-gal positive cells was observed in the CEA (t(6)=8.95, p=0.0001, Figure 4E–F) and BLA (t(7)=2.88, p=0.024, Figure 4G–H). The MEA showed a similar change in X-gal positive cells, and was nearly significant compared to water controls (t(8)=2.28, p=0.052, Figure 4I–J).

Figure 4. IEA increases NFkB expression in subregions of the amygdala.

Mice were allowed 4 weeks of IEA to 20% ethanol or water only. Forty-eight hours after the last drinking period, mice were perfused and stained for X-gal expression as a readout of NFkB activity. While IEA-exposed mice did not display differences from water drinking controls in the NAC shell (A/B) or NAC core (C/D), an increase was observed in the CEA (E/F), BLA (G/H), and MEA (I/J). ***p˂0.001, * p˂0.05, and ^#p=0.052 compared to control group. Scale bars at 100μm.

Experiment 4: Chronic EtOH does not alter IkB expression in the AMY or NAC

Due to similar effects of IEA and chronic alcohol gavage on stress sensitivity, as shown in experiment 1 above and our previously published work (Nelson et al., 2018), respectively, we were also interested in how IkB gene expression was altered following chronic alcohol gavage. Mice (n=9/group) were exposed to 10 days of chronic alcohol (3.0g/kg) via intragastric gavage. Forty-eight hours after the final alcohol gavage, mice were sacrificed and the NAC and AMY were dissected. Chronic alcohol exposure did not alter IkB gene expression in the NAC (t(15)=1.40), p=0.18, Figure 5A) or AMY (t(13)=0.18, p=0.86, Figure 5B).

Figure 5. Chronic EtOH does not alter IkB expression in the AMY or NAC.

Mice were intragastrically gavaged with 3.0g/kg ethanol or water for 10 days. Forty-eight hours after the gavage, mice were sacrificed and the NAC and AMY was dissection and processed for qPCR. No differences in IkB expression was observed in the NAC (A) or AMY (B) compared to water-gavaged controls.

Experiment 5: Chronic EtOH increases NFkB expression specifically in the CEA

Lastly, we assessed the effects of chronic alcohol gavage on NFkB activation using X-gal immunoreactivity in our NFkB-LacZ mice. Mice (n=4–6) were exposed to 10 days of chronic alcohol gavage and were sacrificed 48-hours following the final alcohol treatment. Chronic alcohol gavage did not impact the number of X-gal positive cells in the NAC shell (t(9)=0.87, p=0.41, Figure 6A–B) or NAC core (t(9)=1.17, p=0.27, Figure 6C–D). However, it did significantly increase the number of X-gal positive cells in the CEA (t(8)=2.31, p=0.0497, Figure 6E–F). This effect was specific to this subregion of the amygdala, as no differences were observed in the BLA (t(8)=0.13, p=0.90, Figure 6G–H), or MEA (t(8)=0.08, p=0.94, Figure 6I–J).

Figure 6. Chronic EtOH increases NFkB expression in the CEA.

Forty-eight hours after the last alcohol or water gavage, mice were perfused and stained for β-gal expression as a readout of NFkB activity. No differences in X-gal positive cells were observed in the NAC shell (A/B) or NAC core (C/D). In the amygdala, chronic alcohol-exposed mice displayed increased X-gal positive cells in the CEA (E/F), but not the BLA (G/H) or MEA (I/J). *p˂0.05 compared to control group. Scale bars at 100μm.

BEC analysis:

To determine the approximate BEC to which animals were exposed in our experiments, we measured the BECs obtained following either two weeks of IEA or 3g/kg ethanol gavage. For the IEA analysis, mice (n=8) were allowed 2 weeks of IEA to 20% ethanol and BECs were analyzed two hours into the 6^th IEA session. At this timepoint, mice obtained BECs of 39.43g/dl ± 12.45g/dl (mean ± SEM). For the gavage analysis, mice (n=6) were intragastrically gavaged with 3.0g/kg ethanol and BECs were analyzed 30 minutes after alcohol treatment. At this timepoint, mice obtained BECs of 241.5g/dl ± 13.70g/dl.

Discussion

The primary findings from these studies are that voluntary alcohol consumption via IEA increases sensitivity to subthreshold SDS, in line with our previous findings using chronic alcohol gavage. IEA also increases NFkB activity, specifically in the AMY. When assessed with subregional specificity, NFkB activity was increased in the CEA and BLA, but not in either the NAC shell or core. In contrast, chronic alcohol gavage exposure more selectively increased NFkB activity in the CEA subregion of the AMY without affecting NFkB in other subregions of the AMY or NAC.

We analyzed BECs following both of the alcohol exposure paradigms used in this study. Following two weeks of IEA, mice obtained an average BEC of 39.43g/dl two hours into the alcohol access period. Following a gavage of 3.0g/kg alcohol, mice obtained an average BEC of 241.5g/dl. There are many clear differences between IEA and chronic alcohol exposure paradigms (i.e. route of administration, schedule of exposures, BECs obtained, different stress levels from the exposure, etc.), and as such one must take caution when directly comparing the results obtained following these exposures. However, these differences can also be viewed as a strength of this study. Despite the various differences between these paradigms, both chronic alcohol exposure (Nelson et al., 2018) and IEA increase sensitivity to the same stressor (subthreshold SDS).

Here we demonstrate that voluntary alcohol consumption via IEA impacts stress sensitivity. Similar to other studies using the IEA model of consumption (Hwa et al., 2011), we observed consistent consumption levels around 20g/kg in our IEA-exposed C57BL6/J mice. Examining the effects of voluntary consumption on stress responsivity is an important, clinically relevant approach. IEA is a highly validated model that results in clinically relevant levels of consumption (Carnicella et al., 2014). Considering that alcohol misuse associates with increased rates of MDD (Fergusson et al., 2013, Boden and Fergusson, 2011), voluntary forms of alcohol access such as IEA are ideal for examining this relationship in a preclinical setting. This finding parallels the effects of chronic alcohol exposure on stress sensitivity we have previously observed (Nelson et al., 2018) and further corroborates that voluntary alcohol consumption can impact development of depressive-like behavior by altering stress sensitivity, an underlying factor of development of depression (Bale, 2006).

The neuroimmune system has recently gained attention for its involvement in both the behavioral and molecular responses to alcohol exposure (Mayfield et al., 2013, Crews et al., 2015) and the pathophysiology of stress and depression (Hodes et al., 2015, Hodes et al., 2014, Menard et al., 2016, Menard et al., 2017, Raison et al., 2006). As such, components of the NFkB signaling cascade are intriguing candidates to target in studies examining the underlying mechanisms of comorbidity between these disorders. In our study, we found a decrease in IkB expression following 4 weeks of IEA that, in the AMY specifically, coincided with an increase in NFkB activity in the CEA, BLA, and to a lesser extent, the MEA. Our data suggest that 48 hours after stress, downregulated expression of IkB potentially results in less NFkB inhibition and more freed NFkB subunit dimers to translocate to the nucleus and act as transcriptional activators. In agreement with this, NFkB activity in reporter mice was increased at the same timepoint at which IkB downregulation occurs. While IkB expression and NFkB activity in transgenic mice did not always show parallel effects, it is important to consider that the IkB expression analysis used whole region homogenates of the NAC and AMY, whereas the immunohistochemistry allowed for the assessment of NFkB in specific subregions of the NAC and AMY, and thus a higher degree of specificity.

It is interesting to note that exposure to chronic SDS has previously been found to increase protein levels of IKK, IkB, and phosphorylated-IkB in mice sensitive to SDS 48 hours after the final defeat session (Christoffel et al., 2012). In our study, we assessed IkB gene expression 48 hours after alcohol cessation, indicating critical differences in stimulation of NFkB (stress- vs. alcohol-exposure) and level of analysis (protein vs. gene expression) that may underlie these conflicting results. It appears an increase in IkB expression occurs following acute stress, while a decrease in its expression is observed following alcohol exposure and withdrawal. We would predict that if we assessed NFkB activation following the defeat stress exposure, NFkB expression would be increased to an even greater extent in IEA-exposed mice relative to water-exposed controls, as the pathway would be primed due to the decreased expression of IkB. Our results also indicate a more prominent role of NFkB in the AMY following alcohol exposure, while its activity in the NAC, where NFkB has previously been shown to mediate susceptibility to chronic SDS (Christoffel et al., 2012), is less affected.

While we have previously shown that NFkB is activated in the NAC shell following alcohol place conditioning (Nennig et al., 2017), in contrast to what is reported here, this effect was observed 2 hours following the conditioning session, and was in response to a moderate dose of alcohol administered intraperitoneally. Additionally, place preference is an associational learning protocol that may induce NFkB activation in unique ways. Here, we assessed NFkB activation 48 hours after the last IEA session or final gavage treatment. It is also important to highlight that the gene targets and effects of NFkB are extremely diverse (Nennig and Schank, 2017). As such, NFkB could influence the expression of different genes in different brain regions or cell types, which may have variable effects on the behavioral responses to alcohol. For example, NFkB in the NAC might influence the expression of opioid receptors that modulate reward value, but NFkB in the amygdala might influence stress sensitivity through its actions on stress-related peptides systems.

While we found that IEA increases NFkB activity in multiple subregions of the AMY, including the CEA, BLA, and MEA, chronic alcohol gavage had a more selective effect, only increasing NFkB in the CEA. The CEA is the major output nucleus of the amygdala and has been heavily implicated in stress- and drug-related responsivity (Gilpin et al., 2015). The CEA, along with other regions in the extended amygdala, regulates negative-affect related to stress and alcohol use disorder (Koob, 1999, Gilpin and Roberto, 2012). This study provides novel insight on not only the involvement of NFkB in response to various alcohol exposure paradigms, but also provides a specific region to target (CEA) in future experiments aimed at inhibiting NFkB activation following IEA or chronic alcohol exposure. Interestingly, NFkB activity in the BLA was increased following IEA but not alcohol gavage. The BLA is implicated in affective disorders and addiction due to its role in stress-responses, motivation, and cognition (Sharp, 2017). A projection from the BLA to the NAC shell was recently determined to control voluntary alcohol consumption (Millan et al., 2017), supporting the role of this region in both affect and voluntary intake. As chronic alcohol gavage is not a voluntary form of consumption, this may explain why no differences in activation were observed in the BLA after this exposure.

IEA also increased NFkB activity (though not with statistical significance) in the MEA. Chronic alcohol exposure and alcohol withdrawal dysregulate important mediators of stress circuitry such as corticotropin releasing hormone (CRH)(Becker, 2012, Koob and Kreek, 2007, Sommer et al., 2008). Transcript expression of Crhr1, the gene encoding the CRH receptor 1 (CRH-R1), is upregulated in the BLA and MEA following alcohol dependence (Sommer et al., 2008). Interestingly, CRF-R1 activation can directly induce NFkB activity (Smith et al., 2006), perhaps indicating a potential mechanism of IEA’s ability to increase NFkB activity within these three subregions.

Overall, our results indicate a complex relationship between alcohol-exposure, stress responsivity, and activation of the NFkB pathway. This study determined that intermittent, voluntary alcohol consumption increases sensitivity to social stress. We observed increased NFkB activity in multiple subregions of IEA-exposed mice, but only the CEA of chronic alcohol gavage exposed mice. Future experiments will focus on intra-CEA inhibition of NFkB during exposure to IEA or chronic alcohol gavage to assess alterations in stress sensitivity when NFkB is inhibited in a regionally specific manner prior to stress exposure. Identifying molecular targets involved in the pathophysiology of both alcohol abuse and depression, such as NFkB, may be extremely beneficial for development of novel pharmacotherapeutics for patients displaying comorbid alcoholism and depression.