. Author manuscript; available in PMC: 2026 Sep 25.

Published in final edited form as: Alcohol. 2025 Sep 25;129:106–114. doi: 10.1016/j.alcohol.2025.09.006

Sex differences in the impact of trauma on alcohol self-administration

Nabiha Mahmood ¹, Marian L Logrip ^1,^2,³

PMCID: PMC12634047 NIHMSID: NIHMS2115123 PMID: 41015128

Abstract

Challenging encounters can lead to escalated alcohol consumption, as evidenced by individuals grappling with posttraumatic stress disorder (PTSD). However, the precise influence of prior stress experiences, as opposed to acute stressors, on alcohol intake remains incompletely understood. This study sought to simulate the enduring repercussions of trauma using a stress paradigm involving foot shocks presented in an odor-enriched environment. The presence of a scent throughout contextual fear conditioning is more likely to cause stress effects to generalize across various environments, and this paradigm previously reduced working memory performance, one hallmark of PTSD. Male and female Wistar rats were exposed to the stress or control condition (no foot shock), then trained to perform lever presses to obtain alcohol reinforcement. The findings revealed intriguing disparities between the sexes in past stress effects on the acquisition of alcohol self-administration. Specifically, female rats exhibited divergent patterns of alcohol acquisition across days, with the control group showing a swifter acquisition compared to their previously stressed counterparts. This pattern was not observed in males, nor did either sex show differences in relapse-like self-administration after a 5-week abstinence period. Unexpectedly, presentation of the stress session odor cue did not alter alcohol self-administration behavior. Together, these data support heightened sensitivity of females to a trauma-like stressor, although in rats this decreased self-administration, contrary to human data.

Keywords: sex differences, operant self-administration, foot shock, stress history

INTRODUCTION

Alcohol use disorder (AUD) is a significant public health challenge, impacting about 10% of the population worldwide, depending on region and source of statistics [1–3]. Alcohol drinking may be used as a form of self-medication, resulting in high rates of comorbidity between AUD and stress-related psychopathology like posttraumatic stress disorder (PTSD) [4, 5]. According to the National Center for PTSD, up to 75% of people who have experienced trauma report drinking problems, regardless of whether they develop PTSD [6]. Comorbid diagnosis of anxiety or PTSD and AUD results in poorer treatment outcomes [7, 8], including for women [5]. Research suggests stress provides a greater challenge for women, as lifetime exposure to stress results in greater diagnosis of psychopathology in women than in men, including for anxiety disorders, PTSD, and AUD [9]. Additionally, studies indicate that traumatic experiences may more frequently precede the onset of AUD in women [10] and experience heightened craving and relapse compared to males [11, 12]. Together, these data indicate that biological sex significantly modulates how stress remodels the brain and thus influences behavior patterns, including for alcohol consumption. However, this area of research remains under-explored [5, 13, 14].

Preclinical research can provide insight into the mechanisms by which stress history influences alcohol intake. In rodent models, past stress exposure has been shown to modulate self-administration patterns in a sex-dependent manner, although results vary based on experimental parameters [13, 14]. Relevant to the current study, the impact of past stress using a 3-day, light-cued foot-shock exposure paradigm to examine lasting effects on alcohol self-administration over several months did not impact acquisition but rather reacquisition of alcohol self-administration after an extinction period in a sexually divergent pattern [15]. Specifically, enhanced reacquisition was observed in some male rats [16], but lower alcohol consumption in female rats, relative to same-sex stress-naïve controls [15]. However, this multi-day stress paradigm may not accurately model the experience of a single traumatic event that often triggers the development of PTSD. One modified paradigm for adult foot shock-based trauma employing a two-exposure approach differentially increased alcohol drinking and preference by sex based on the familiarity or novelty of the second shock environment [17]. Predator odor has been employed to generate a trauma history, with drinking effects of past stress modulated by susceptibility to the stress with [18] or without [19] sex-specific impacts of the trauma. Together, these studies indicate that a more discrete, trauma-like exposure may impact not only relapse-like, but also acquisition, of alcohol self-administration.

Recently, a single-session rat foot-shock exposure paradigm was developed that generated lasting working memory deficits [20]. This model differs from previous shock models used in alcohol research by presence of an odorant providing greater contextual complexity throughout the shock session. Previous studies in mice have shown that contextual scents may facilitate generalization of the stress experience [21], as generalization of fear reactions to both predictive and nonpredictive but similar cues is observed in PTSD patients [22]. Generation of working memory deficits in both sexes, a common feature of PTSD, supports examination of this odor-enriched foot-shock trauma model as a putative behavioral approach for examining comorbid PTSD and AUD in rats. The current study examines the impact of this prior trauma on acquisition of alcohol self-administration beginning 1 week after the foot-shock session, as well as relapse-like self-administration after a 5-week deprivation period in male and female Wistar rats. By addressing the interplay of trauma, sex, and alcohol consumption, this study aims to bridge existing gaps in understanding of the differential sensitivity of men and women to stress-induced alcohol misuse.

MATERIALS AND METHODS

Subjects

Adult male and female Wistar rats (Charles River, Kingston, NY, USA), 8–9 weeks old at arrival, were used in the study. A total of 32 rats, consisting of 16 females and 16 males, were used in the study. The rats were housed in same-sex pairs in a temperature- and humidity-controlled vivarium with a 12-hour reversed light/dark cycle (dark period from 7:00 am to 7:00 pm) and ad libitum access to food and water. Rats were well handled beginning one week before the onset of experimentation. Experiments were performed entirely by 2 female experimenters. All experimental procedures conformed to the National Institutes of Health Guide for the Care and Use of Research Animals and were approved by the Indiana University Indianapolis Institutional Animal Care and Use Committee.

Equipment

Operant self-administration sessions were conducted in standard behavioral testing chambers manufactured by Med Associates (Fairfax, VT, USA). These chambers are designed with plexiglass front and rear walls, stainless steel side walls, and stainless-steel rod floors. The chambers feature a recessed stainless steel liquid reinforcer receptacle located between two retractable levers, situated 2.1 cm above the floor. The chambers are housed within light- and sound-attenuating cubicles that are equipped with fans to minimize external distractions and regulate temperature. Foot-shock stress sessions were carried out in the same chambers, but the receptacle was replaced with a flat wall panel and the levers remained retracted. All experimental sessions were controlled and recorded using Med-PC IV software (Med Associates).

Home Cage Pre-Exposure

The experimental timeline, shown in Figure 1A, began with a 48-hour 2-bottle choice phase in which rats had access to water and 10% (v/v) alcohol in their home cage. On the following five days, rats were given 1-hour home-cage access to two-bottle choice, except for the day they were acquiring the lever-press response (Lever-discrimination training, below), as described below. For home-cage drinking sessions, rats were moved into a cage separated by a plexiglass divider along the long access, allowing for individual bottle access for the rats without necessitating solitary housing. When not engaging in 2-bottle choice drinking, rats were return to their normal pair-housed setting without the divider. This procedure established the time of day of alcohol access, as 1-hour home cage access was time-matched to future operant sessions and circumvented the risk of stress-related neophobia impacting alcohol self-administration [23].

Trauma Exposure

The rats were subjected to a 20-min period of foot-shock stress exposure, receiving 20 0.8-mA, 1-sec shocks on a variable interval schedule with an average intertrial interval of 40 seconds (VI40) beginning after a 5-min acclimation to the chamber. This foot-shock paradigm has previously been shown to generate sleep disturbances [24] and working memory deficits [20], two hallmarks of PTSD, and thus is referred to as a trauma exposure. To generate an enriched contextual environment during foot shock, sessions were performed in the presence of coriander powder, placed under the bedding and in a cup within the sound-attenuating cubicle so that physical interaction with the powder was minimal to none. This session was at a distinct time of day (late in the dark cycle) and rats were placed in different individual chambers than those to which they were assigned for operant studies. Group assignments were balanced such that no significant differences were present for 1-h home-cage alcohol intake, active lever presses during the 12-h water session, and inactive lever presses during the 12-h water session.

Operant Self-Administration

Lever-discrimination training:

Rats were trained to press the active, or reinforced, lever using a single, 12-h water-reinforced session, run during a complete dark cycle, as described in Logrip and Gainey [15]. Briefly, each depression of the active (reinforced) lever triggered delivery of 100 μl water into the reward port. During reinforcer delivery, presses on the active lever were counted but did not result in reward delivery. A second, inactive, unreinforced lever was present throughout the session; pressing this lever yielded no consequences. Sessions ended after 12 h or upon receipt of 300 reinforcers (30 mL). Food pellets were present in the box throughout the session to avoid food deprivation and to encourage learning of the lever-press task due to food intake-related thirst. This single training session was performed prior to trauma exposure to avoid potential stress-related new learning deficits [25].

Alcohol self-administration training:

One week after trauma exposure, a time point at which sleep disturbances have been observed with the same stress paradigm [24], operant self-administration training began. Rats first underwent five fixed-ratio 1 (FR1) operant sessions, run every other day, with each active lever press delivering 0.1 ml of 10% (v/v) alcohol. Subsequently, the rats participated in 12 sessions of FR3 alcohol self-administration every other day to assess the impact of past stress on subsequent alcohol-seeking behavior. Across all sessions, lever presses performed during reinforcer delivery were counted as time-out lever presses but did not contribute to earning the next reinforcer. These presses are included with the reinforced lever presses in the total lever-press data presented here. Sessions were 1 h in duration, run between hours 2–4 of the dark cycle.

Relapse-like reacquisition of alcohol self-administration:

The rats were given 5 weeks of abstinence in their home cages, with no operant sessions provided. This was followed by five sessions of FR3 alcohol self-administration to evaluate relapse-like behavior. Data were evaluated relative to baseline performance, averaged across the final 4 sessions before abstinence.

Trauma odor cue effects:

A final 2 sessions were run to determine whether the presence of coriander odor, a contextual cue present only during the foot shock trauma exposure, would alter alcohol self-administration since it had been present during the foot-shock trauma exposure. Because of the risk of carryover effects in the event that coriander altered alcohol self-administration, the order of presentation (no cue vs. cue) was not counterbalanced, but rather, the coriander was present during the last session for all rats.

Data Analysis

Behavioral criteria for exclusion:

We excluded 2 females (1 per group) that did not acquire operant self-administration by the conclusion of the acquisition period. Lack of acquisition was defined as having no alcohol-reinforced session in which active lever presses exceeded 10; on most days for the excluded rats, lever presses were zero (Excluded control female: average 2.67 presses per session, maximum single-day performance, 7; excluded stress female: average 0.24 presses per session, maximum single-day performance, 1 press). One control female became ill during the reacquisition phase and could no longer be run. A second female from the stress history group died prior to the coriander testing days. These differences in numbers are indicated in the appropriate figure legends for clarity.

Data transformations:

Data are expressed as mean +/− standard error of the mean (S.E.M.). Weight-normalized alcohol intake is calculated based on converting the volume of alcohol reinforcer earned within each session to g alcohol and dividing by the rat’s body weight in kg. For examination of changes in self-administration after abstinence, data are expressed as fold change in weight-normalized alcohol intake, relative to each rat’s average intake over the final 4 sessions prior to abstinence (baseline). For individual data point comparisons between baseline and relapse, 5-day averages for each rat were used. To examine the possibility that low- and high-drinking subpopulations exist within the current cohort and display different behavioral profiles in the relapse-like period, male data were subdivided into low and high drinkers based on the final 4 days of FR3 self-administration prior to abstinence, according to Logrip & Zorrilla [16] using a threshold of 0.5 g/kg. Because this classification left all but two females in the “high drinkers” category, a median split was performed on female data, resulting in a threshold of 1.11 g/kg. For coriander exposure, data are expressed as ratios to performance (active lever presses) or intake (g/kg) during the session immediately preceding the single coriander exposure session.

Statistical analyses:

Analyses of behavior across days evaluated the between-subject factors sex and group and the within-subject factor session using three-way analysis of variance (ANOVA), with two-way ANOVAs conducted separately by sex based on a priori hypotheses of potential sex differences that could be obscured by the complexity of observing three-way interactions. Pre-stress behavioral measures and coriander exposure effects were analyzed via 2-way ANOVA with the between-subjects factors sex and group. Greenhouse-Geisser corrections were applied to eliminate violations of sphericity in ANOVAs, ensuring the robustness of the statistical analyses. Significant ANOVA effects, defined as yielding p values less than 0.05, were followed by post hoc Tukey’s tests. Because some rats did not consume all the alcohol earned during the FR1-reinforced sessions, these sessions were excluded from analysis and presentation of alcohol intake data.

RESULTS

Alcohol self-administration after traumatic stress: Lever presses

To test the effects of a trauma-like stressor on acquisition of and relapse to alcohol operant self-administration, rats were exposed to a single 20-min foot-shock session 1 week before the onset of alcohol self-administration. Group assignments were balanced according to home cage alcohol intake and behavior at the active and inactive levers during the 12-h water-reinforced session (Table 1). As shown in Figure 1, behaviors on the alcohol-reinforced active lever developed differently across sessions in male (Figure 1B) and female (Figure 1C) rats. Analysis of active lever presses by 3-way RM-ANOVA revealed significant interactions between sex, group and session (F_16,416 = 1.69, p = 0.047) and between group and session (F_16,416 = 1.96, p = 0.015), as well as a main effect of session (F_6.79,176.6 = 16.34, p < 0.0001). Follow-up analyses within sex by 2-way RM-ANOVA yielded differential effects in males and females. Males showed a significant effect of session (F_5.27,73.8 = 7.18, p < 0.0001) without effect of group or interaction between the factors (F’s < 0.78, p’s > 0.54; Figure 1B). Conversely, females displayed a significant group by session interaction (F_16,192 = 2.75, p = 0.0005) in addition to a main effect of session (F_5.39,64.7 = 9.94, p < 0.0001; Figure 1C). Post hoc investigation by Tukey’s test showed escalation in active lever presses among female controls, with significant increases from session 1 observed for sessions 13 (p = 0.040), 15 (p = 0.0055), and 16 (p = 0.0004), whereas females in the stress group showed no significant differences between session 1 and any subsequent sessions. Additionally, female control and stress group rats differed on session 15. Following the 5-week abstinence period, rats returned to operant self-administration for 5 sessions (Figure 1D). Rats with a history of trauma exposure produced less behavior at the active lever than control rats, regardless of sex, yielding a main effect of group in 3-way ANOVA analysis (F_1,156 = 5.68, p = 0.018) without any other main effects or interactions (F’s < 2.27, p’s > 0.13). Sex-specific 2-way ANOVAs only found effects of session (male: F_{3.9, 55.3} = 2.67, p = 0.043; female: F_3.2,
38.6 = 3.65, p = 0.019), without group effects or interactions (F’s < 2.32, p’s > 0.08).

Table 1.

Equivalence of parameters across stress exposure groups

	Male Control	Male Stress	Female Control	Female Stress	Statistics

48-h alcohol intake (g/kg)	7.06 ± 2.08	6.75 ± 1.57	5.66 ± 1.26	6.56 ± 1.18	Group: F_1,26 = 0.03, p = 0.86
					Sex: F_1,26 = 0.23, p = 0.63
					Int: F_1,26 = 0.14, p = 0.71
Final 1-h alcohol intake (g/kg)	0.68 ± 0.13	0.75 ± 0.19	0.93 ± 0.14	1.18 ± 0.31	Group: F_1,26 = 0.57, p = 0.46
					Sex: F_1,26 = 2.60, p = 0.12
					Int: F_1,26 = 0.20, p = 0.66
Active Lever Presses, Water Session	315.4 ± 21.2	311.4 ± 24.0	258.7 ± 43.2	217.1 ± 50.8	Group: F_1,26 = 0.37, p = 0.55
					Sex: F_1,26 = 4.08, p = 0.054
					Int: F_1,26 = 0.25, p = 0.62
Inactive Lever Presses, Water Session	87.9 ± 16.8	70.3 ± 14.6	74.9 ± 13.7	59.6 ± 13.2	Group: F_1,26 = 1.17, p = 0.29
					Sex: F_1,26 = 0.61, p = 0.44
					Int: F_1,26 = 0.006, p = 0.94

Open in a new tab

Data are presented as mean +/− S.E.M. n = 8 per group for males, n = 7 per group for females. Females who did not acquire operant self-administration were eliminated from these data to demonstrate that this elimination did not unbalance parameters that might affect learning of operant alcohol self-administration. Statistics show results of 2-way ANOVA with the factors Group (Control, Stress) and Sex (Male, Female). No main effects of Group or Group x Sex interactions were found.

Increases in lever presses across acquisition sessions were specific to the active lever, as presses on the inactive lever were substantially lower across all sessions (Figure 1E,F) and decreased over time. Analysis of inactive lever behavior by 3-way RM-ANOVA yielded a main effect of session (F_{6.93, 180.1} = 9.49, p < 0.0001) and interactions between session and group (F_16,416 = 1.72, p = 0.040), sex and group (F_16,416 = 4.89, p = 0.036), and session by sex by group (F_{16, 416} = 2.88, p = 0.0002). Follow-up within-sex 2-way RM-ANOVAs produced a main effect of session in males (F_5.38,75.3 = 6.55, p = 0.0001; Figure 1E) and a trend for group (F_{1, 14} = 3.98, p = 0.066), without a significant session by group interaction. Post hoc Tukey’s tests indicated significant reductions in inactive lever presses, relative to session 1, for sessions 10 (p = 0.0498), 16 (p = 0.038) and 17 (p = 0.024) when collapsed by group. Females again showed divergence between the groups, with a main effect of session (F_5.38,64.5 = 4.027, p = 0.0025) as well as a group by session interaction (F_{16, 192} = 2.74, p = 0.0006; Figure 1F). Post hoc comparison by Tukey’s test showed group differences for sessions 4 (p = 0.049), 5 (p = 0.041), and 6 (p = 0.024), with stress history females pressing significantly more on the inactive lever than control females. Following the 5-week abstinence period, rats no longer displayed any differences in inactive lever-directed behavior (Figure 1G), with no significant main effects or interactions in the 3-way ANOVA or sex-specific 2-way ANOVAs (F’s < 1.36, p’s > 0.25).

Effects of trauma-like stress and abstinence on alcohol intake

Similarly to active lever presses, alcohol intake during FR3 sessions showed different patterns for stress effects by sex, as illustrated in Figure 2. Analysis by 3-way ANOVA produced significant main effects of session (F_8.44,211.1 = 5.40, p < 0.0001) and sex (F_1,25 = 16.30, p = 0.0005), as well as significant session by sex (F_16,400 = 1.97, p = 0.014), session by group (F_16,400 = 2.25, p = 0.0038), and session by sex by and group interactions (F_16,400 = 2.31, p = 0.0030). Comparisons within sex by 2-way RM-ANOVA revealed that whereas both sexes displayed significant main effects of session (male: F_{6.13, 85.8} = 2.69, p = 0.019, Figure 2A; female: F_6.36,70.0 = 3.54, p = 0.0035, Figure 2B), notably, females showed a statistically significant interaction between session and group (F_16,176 = 2.52, p = 0.0017). Post hoc group comparisons by Tukey’s test yielded a significant difference between female groups for session 15 (p = 0.045), indicating that the female groups largely differ in their intake patterns across all sessions rather than on a session-by-session basis.

Figure 2. — Rats acquired alcohol self-administration with one lever reinforced by delivery of 0.1 mL 10% (v/v) alcohol and the other lever unreinforced. Following acquisition, rats were exposed to a forced abstinence period of 5 weeks, after which alcohol self-administration sessions resumed for 5 sessions (relapse sessions). Male (A,C) and female (B,D) data are displayed separately for ease of visualization. (A,B) Panels show weight-normalized alcohol intake across all fixed-ratio 3-reinforced sessions. Data are calculated as (g alcohol consumed based on reinforcers earned) / (body weight in kg). (C,D) Panels display how alcohol intake changed after the forced-abstinence period, relative to intake pre-abstinence, with fold change in intake calculated as (weight-normalized intake during the relapse session) / (average weight-normalized intake across final 4 sessions prior to abstinence). Data are presented as mean +/− S.E.M. averaged by group for each session. (E,F) Panels show individual subjects’ data for comparative observation of pre- and post-abstinence average performance across a 5-day period. * p < 0.05, Control vs. Stress. Male, n = 8/group; Female Control, n = 6; Female Stress, n = 7.

To more closely examine the impact of a trauma-like stressor on relapse-like alcohol intake, two additional measures were employed. First, the change in alcohol self-administration was assessed following a 5-week abstinence period. Analysis of the change in alcohol intake, relative to the pre-abstinence baseline, by 3-way RM-ANOVA yielded no significant main effects or interactions (F’s < 2.56, p’s > 0.09). Within-sex 2-way RM-ANOVAs similarly yielded no significant effects in males (F’s < 1.52, p’s > 0.23; Figure 2C) or females (F’s < 2.94, p’s > 0.07; Figure 2D). Direct visual examination of individual subjects’ average alcohol intake before (Baseline) and after (Relapse) the abstinence period for each subject demonstrated that some individuals increased their alcohol self-administration after abstinence while many decreased or did not change their alcohol consumption (Figure 2E,F).

Since data show some variability among the rats, further analyses of subgroups divided according to pre-abstinence intake levels were performed, as in Logrip & Zorrilla [16]. As shown in Figure 3, Low-Drinkers differed by group when data were analyzed by 3-way ANOVA (F_1,9 = 7.11, p = 0.026) without any other main effects or interactions (F’s < 2.48, p’s > 0.23). Further sex-specific analyses of Low Drinkers failed to demonstrate any significant main effects or interactions for males (F’s < 2.82, p’s > 0.07, Figure 3A) or females (F’s < 0.90, p’s > 0.46, Figure 3B). Unlike Low Drinkers, rats categorized as High Drinkers displayed different levels of self-administration by sex, session, and their interaction (Sex: F_1,12 = 24.85, p = 0.0003; Session: F_3.3,40.0 = 10.61, p < 0.0001; Sex x Session: F_3.3,40.0 = 3.25, p = 0.028). Sex-specific 2-way ANOVAs reinforced the main effects of session (males: F_3.5,24.3 = 3.14, p = 0.038, Figure 3C; females: F_1.9,9.3 = 9.64, p = 0.0060, Figure 3D). It should be noted that this group subdivision yielded small numbers among all groups, particularly low-drinking control females (n = 2), so statistical analyses within Low and High Drinkers should be interpreted with caution.

Figure 3. — Rats were subdivided into Low Drinkers and High Drinkers according to their average alcohol intake across 4 sessions late in acquisition (Baseline). Panels show average alcohol intake at baseline and for each relapse session for low-drinking male (A) and female (B) rats and for high-drinking male (C) and female (D) rats. Male and female acquisition data are displayed separately for ease of visualization. Data are presented as mean +/− S.E.M. averaged by group for each session. Number of subjects in each group is displayed on the figure labels.

Coriander trauma reminder did not alter alcohol self-administration

To examine whether an olfactory reminder of the trauma exposure would change operant self-administration, rats were exposed to the coriander scent used as a contextual component during foot-shock exposure during the final operant self-administration session. Analysis of fold change in active lever presses on the coriander-exposure day, relative to the previous day, by 2-way ANOVA revealed no significant main effects of sex or group, nor any interactions between the factors (F’s < 0.67, p’s > 0.42; Figure 4A). Further examination by subject of active lever presses on the pre-exposure and exposure days show mostly similar patterns of pressing in these sessions for male (Figure 4B) and female (Figure 4C) rats, although some individuals display different patterns. Analysis of alcohol intake on coriander-exposure day, relative to the day before, also indicated no significant differences between stress and control groups in either males or females (F’s < 0.23, p’s > 0.64; Figure 4D), with a bit more spread among individual response patterns among males (Figure 4E) and females (Figure 4F), particularly for control rats who previously experienced the coriander odor in the absence of any adverse consequences. These results suggest that the presence of this trauma reminder during an alcohol self-administration session does not significantly alter alcohol self-administration behavior in a group-wide fashion but may have differential impact on individual subjects.

Figure 4. — On the final day of alcohol self-administration, rats were exposed to the presence of coriander in the operant environment as a reminder of the stress exposure session. Presses on the alcohol-reinforced lever during the coriander session, expressed as a ratio to the previous day’s performance (Fold Change), are displayed as a histogram overlaid with individual rats’ data points (A). Additionally, lever-press data are shown for all male (B) and female (C) rats with connecting lines emphasizing each individual pattern. Weight-normalized alcohol intake, calculated based on reinforcers earned, is also displayed as a ratio to intake on the day prior (D),with intake data shown for all males (E) and females (F), demonstrating differences in individual behavioral patterns over the sessions. Histograms present as mean +/− S.E.M. averaged by group, with individual subjects’ data points overlaid. Male, n = 8/group; Female, n = 7/group.

DISCUSSION

The present study examines the complex interplay between a history of trauma-like stress and alcohol self-administration in rats, providing insight into sex differences in the effects of past trauma on alcohol self-administration. The data show that past trauma affects alcohol-seeking behavior in a sex-dependent manner, with females demonstrating delayed acquisition of alcohol self-administration after trauma, while males did not show such a divergence. These differences were resolved following a 5-week period of abstinence, and no new effects of group or sex emerged in that relapse-like period. This aligns with previous literature suggesting that females may process stress differently [15, 26], although it contrasts with the increased psychopathology, including related to alcohol, observed in women more than men relative to past stress [9].

In contrast to the repeated stress paradigm previously employed [15], which utilized light-cued foot-shock stress across multiple sessions, the present study implemented a single-session unpredictable foot-shock trauma model. This stressor differed on several factors. First, the sessions lacked a discrete light cue preceding each shock but added a contextual coriander odor cue, a change intended to increase generalizability of the stress [21]. Second, the previous model utilized more presentations of a milder shock, delivered across 3 days, whereas the current model involved a single day of 20 stronger-intensity shocks. Finally, alcohol self-administration commenced a week after the stress session in the current model, to allow for incubation of the stress effects, whereas the previous model employed a 36-hour delay. This delay was selected as a time point when sleep disturbances have been shown using the same stressor [24] and because past foot-shock models have only demonstrated changes during relapse-like self-administration, which occurs more than 2 months after stress exposure [15, 16, 27]. These features of the current model suggest a more discrete trauma-like stress exposure may be able to model PTSD-like effects on alcohol self-administration.

The current data also investigate trauma’s impact on relapse-like behavior after a period of forced abstinence, as compared to the previously modeled reacquisition following extinction training. While both models produced sexually divergent effects, the pattern of effects differed substantially. Specifically, the multi-day cued foot-shock paradigm produced enhanced reacquisition of operant responding after extinction in males [16] and attenuated responding in females [15], whereas the current data demonstrate that trauma-exposed females exhibit delayed acquisition of alcohol self-administration, with group differences in females normalized after the abstinence period. Additional analyses subdivided by baseline levels of alcohol intake in the operant paradigm highlighted alcohol self-administration among low drinkers in the stress group did not change after the abstinence period, whereas high-drinking females consumed more than high-drinking males, independent of stress group. This divergence in behavioral effects of the stressors between the studies may reflect differences in stress chronicity and procedural complexity, with the impact of the current trauma model materializing earlier in training than the previously used multi-day stress paradigm, whose effects only emerged after an extinction period. The abstinence period was designed to match the length of extinction training, which takes 5 weeks to complete if run on alternate days. Thus, time since last alcohol self-administration session does not differ between the models. However, two main differences do exist between these studies that might impact the outcomes. First, in the previous study [15], rats were individually housed from the start of home-cage drinking through the end of the study, whereas here, solitary housing was avoided. As isolation activates the hypothalamic-adrenal-pituitary axis in females but not males [28, 29], the findings during the relapse-like period in females only suggest a possible role for compounding stress that did not exist in the current study design. Second, extinction training is a stressful procedure due to regular exposure to the operant context in the absence of the desired reinforcer [30], a feature lacking in an abstinence period. Together, these features suggest that extinction training and isolation likely elevated stress levels in all rats by the relapse-like period, which may have shaped behavior in the previous study differently than the current study’s 5-week abstinence period. It should be noted that despite the temporal differences of behavioral effects, both foot-shock models yield suppression of alcohol self-administration in females, contrary to the potentiation of alcohol pathology observed in women [9, 12]. Nonetheless, the potentiated behavioral effect of the current single-day stress model aligns with other data suggesting this model may reproduce a subset of PTSD symptoms.

Unlike the dampening of alcohol-reinforced lever presses among previously trauma-exposed females, in the early stages of training the same group showed increased inactive lever presses. This contrasts with previously observed lower inactive lever pressing in early training among stress history males, as compared to stress-naïve controls [16]. This difference in behavior was interpreted as increased focus on the alcohol-paired lever in the male study, which would parallel consideration of the current female data as displaying poorer lever discrimination in those previously exposed to stress during early acquisition. Yet this is not the only possible explanation. Despite efforts to familiarize rats with behavioral procedures and alcohol’s sensory properties and effects prior to trauma exposure, behavior in the stress group may be impacted by neophobia. While neophobia is commonly viewed as a reduction in motor activity or avoidance of novelty [31], rats with more sensation-seeking traits, termed high responders or low avoiders, employ active coping strategies [32, 33]. Active coping manifesting as escape behavior rather than freezing is more common in females in fear conditioning paradigms [34], making increased pressing on the nonreinforced lever a plausible result of neophobia. Alternatively, alcohol drinking in the home cage may not generalize to familiarity when presented in the operant box as the role of context has been suggested to be important for habituating rodent neophobia [35]. A third possibility lies in increased perseverative responding, which has been observed in a rat model of PTSD [36]. The delayed decrease in inactive lever presses until after the sixth alcohol-reinforced operant session supports the possibility that females in the stress group might struggle to alter that nonproductive behavior. As the current data cannot directly address these possibilities, further exploration of the underlying motivation for elevated inactive lever presses among stress group females is warranted.

The stress model utilized herein was developed to investigate its potential to reproduce the working memory deficits common among individuals with PTSD [37]. Using an odor span task requiring digging in the correctly scented pot to obtain a reward, both male and female rats who had experienced trauma displayed working memory deficits [20] at a time point that would align with the latter half of FR3 training. Importantly, presentation of the coriander odor previously employed as a contextual odor cue during the foot-shock session in one of the pots further impaired working memory on the test day. These results suggest that the contextual odor cued foot-shock session can reproduce some of the symptoms of PTSD. Unexpectedly, while this stressor generated behavioral deficits in females, the behavioral impact was contrary to the hypothesis of increased alcohol self-administration. Because of the operant nature of the task, delayed acquisition could be a sign of stress-related neophobia [23]. Previous foot-shock models suggested that the water self-administration session to acquire active lever preference and home-cage experience with alcohol predating the trauma session sufficiently reduce task-related neophobia challenges. It is possible that the trauma-like stress model presents a sufficiently stronger stressor to exacerbate neophobia risks. Alternatively, the lack of consistent effects of stress in both sexes in past studies [13], even when utilizing similar stressors, suggests that various parameters can impact stress effects on alcohol-related behaviors and that the result is not consistently an increase in alcohol intake. Regardless, the observed delay in alcohol acquisition in stressed females but not males points to underlying mechanisms triggered by the trauma that differ between the sexes and require further study.

Surprisingly, presence of the coriander odor during a final operant self-administration session did not impact alcohol-related behavior. The coriander powder was included during the trauma exposure session as an olfactory contextual cue to serve as a potential trauma-associated reminder during alcohol self-administration. However, re-exposure to the coriander odor after the abstinence period did not elicit significant behavioral changes in either active lever pressing or alcohol intake, although examination of individual subjects’ data suggests possible divergence in effect of coriander presence, particularly among control rats. Nonetheless, the lack of overall effect may be attributed to discrepancies in contextual encoding between the initial trauma session and the later operant sessions. Specifically, although the coriander cue was consistently present during foot-shock exposure, the trauma session lacked operant components such as the levers and alcohol reinforcement. In contrast, the coriander cue was reintroduced in the operant environment where alcohol and lever-pressing behaviors were present – elements not associated with the original foot-shock context. While these contextual differences may seem minor in comparison to the shifts used to demonstrate contextual dependence of reinstatement [38], they nonetheless suggest that the presence of levers and a reward port may be sufficient contextual cues to demonstrate safety, even in the presence of the coriander odor. The inability of the coriander to manipulate alcohol self-administration in the presence of the described contextual mismatches might indicate that the trauma-associated cue did not sufficiently generalize to provoke stress-related changes in alcohol self-administration; however, confirmation of the presence or lack of cue generalization is beyond the scope of the current data set. Nonetheless, although this enriched environment foot-shock model has been shown to induce persistent working memory deficits, the current findings highlight the complexity of contextual cue perception and generalization in rodents, particularly in the presence of novel behavioral contingencies such as alcohol reinforcement.

Summary

Trauma-like stress exposure significantly impacted the acquisition of alcohol self-administration in female but not male rats but had no effect on relapse-like self-administration after a period of forced abstinence. In the context of previous work examining sex differences in alcohol self-administration after stress, these data suggest the need to investigate differential neuroadaptations in the aftermath of stressors and their long-term impact on alcohol use. Importantly, the findings add to data supporting a more sex-specific approach when investigating the impact of trauma on various behaviors, including AUD. These findings underscore the importance of understanding how stress influences alcohol-seeking behavior and the role sex plays in these dynamics.

Highlights.

Trauma-like stress blunts acquisition of alcohol self-administration in females not males
Relapse-like self-administration was not impacted in either sex
Trauma-associated odors do not alter established alcohol self-administration

Acknowledgments

This research was supported by Indiana University Indianapolis and by P60 AA007611 (MLL). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the National Institute of Alcohol Abuse and Alcoholism. The authors have no conflicts of interest to disclose.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

1.Ross HE, DSM-III-R alcohol abuse and dependence and psychiatric comorbidity in Ontario: results from the Mental Health Supplement to the Ontario Health Survey. Drug Alcohol Depend, 1995. 39(2): p. 111–28. [DOI] [PubMed] [Google Scholar]
2.Grant BF, et al. , Co-occurrence of 12-month alcohol and drug use disorders and personality disorders in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch Gen Psychiatry, 2004. 61(4): p. 361–8. [DOI] [PubMed] [Google Scholar]
3.World Health Organization, Global status report on alcohol and health 2018. 2018, Geneva, Switzerland: World Health Organization. [Google Scholar]
4.McLean CP, et al. , Gender differences in anxiety disorders: prevalence, course of illness, comorbidity and burden of illness. J Psychiatr Res, 2011. 45(8): p. 1027–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Saraiya TC, et al. , Sex and Gender Differences in Co-Occurring Alcohol Use Disorder and PTSD. Curr Addict Rep, 2023. 10(4): p. 617–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.U.S. Department of Veterans Affairs. PTSD: National Center for PTSD. 2023. [cited 2023; Available from: https://www.ptsd.va.gov/.
7.Taylor M, Petrakis I, and Ralevski E, Treatment of alcohol use disorder and co-occurring PTSD. Am J Drug Alcohol Abuse, 2017. 43(4): p. 391–401. [DOI] [PubMed] [Google Scholar]
8.Brown PJ and Wolfe J, Substance abuse and post-traumatic stress disorder comorbidity. Drug Alcohol Depend, 1994. 35(1): p. 51–9. [DOI] [PubMed] [Google Scholar]
9.Armstrong JL, et al. , Gender moderates the relationship between stressful life events and psychopathology: Findings from a national study. J Psychiatr Res, 2018. 107: p. 34–41. [DOI] [PubMed] [Google Scholar]
10.Sonne SC, et al. , Gender differences in individuals with comorbid alcohol dependence and post-traumatic stress disorder. Am J Addict, 2003. 12(5): p. 412–23. [PubMed] [Google Scholar]
11.Lehavot K, et al. , Gender differences in relationships among PTSD severity, drinking motives, and alcohol use in a comorbid alcohol dependence and PTSD sample. Psychol Addict Behav, 2014. 28(1): p. 42–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Heffner JL, Blom TJ, and Anthenelli RM, Gender differences in trauma history and symptoms as predictors of relapse to alcohol and drug use. Am J Addict, 2011. 20(4): p. 307–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Logrip ML, et al. , Sexual dimorphism in the neural impact of stress and alcohol. Alcohol, 2018. 72: p. 49–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Finn DA, Clark CD, and Ryabinin AE, Traumatic stress-enhanced alcohol drinking: Sex differences and animal model perspectives. Curr Addict Rep, 2024. 11(2): p. 327–341. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Logrip ML and Gainey SC, Sex differences in the long-term effects of past stress on alcohol self-administration, glucocorticoid sensitivity and phosphodiesterase 10A expression. Neuropharmacology, 2020. 164: p. 107857. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Logrip ML and Zorrilla EP, Stress history increases alcohol intake in relapse: relation to phosphodiesterase 10A. Addict Biol, 2012. 17(5): p. 920–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Kirson D, et al. , Sex and context differences in the effects of trauma on comorbid alcohol use and post-traumatic stress phenotypes in actively drinking rats. J Neurosci Res, 2021. 99(12): p. 3354–3372. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Alavi M, et al. , Sensitivity and Resilience to Predator Stress-Enhanced Ethanol Drinking Is Associated With Sex-Dependent Differences in Stress-Regulating Systems. Front Behav Neurosci, 2022. 16: p. 834880. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Weera MM, et al. , Traumatic Stress-Induced Increases in Anxiety-like Behavior and Alcohol Self-Administration Are Mediated by Central Amygdala CRF1 Neurons That Project to the Lateral Hypothalamus. J Neurosci, 2023. 43(50): p. 8690–8699. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.McGonigle CE, Lapish CC, and Logrip ML, Male and female impairments in odor span are observed in a rat model of PTSD. Learn Mem, 2023. 30(1): p. 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Huckleberry KA, Ferguson LB, and Drew MR, Behavioral mechanisms of context fear generalization in mice. Learn Mem, 2016. 23(12): p. 703–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Jovanovic T, et al. , Posttraumatic stress disorder may be associated with impaired fear inhibition: relation to symptom severity. Psychiatry Res, 2009. 167(1–2): p. 151–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Job RF and Barnes BW, Stress and consumption: inescapable shock, neophobia, and quinine finickiness in rats. Behav Neurosci, 1995. 109(1): p. 106–16. [DOI] [PubMed] [Google Scholar]
24.Wellman LL, et al. , The basolateral amygdala determines the effects of fear memory on sleep in an animal model of PTSD. Exp Brain Res, 2014. 232(5): p. 1555–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Schoner J, et al. , Post-traumatic stress disorder and beyond: an overview of rodent stress models. J Cell Mol Med, 2017. 21(10): p. 2248–2256. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Finn DA, et al. , Binge Ethanol Drinking Produces Sexually Divergent and Distinct Changes in Nucleus Accumbens Signaling Cascades and Pathways in Adult C57BL/6J Mice. Front Genet, 2018. 9: p. 325. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Moench KM and Logrip ML, Housing Condition Differentially Impacts Escalation of Alcohol Intake, Relapse-Like Drinking, Anxiety-Like Behavior, and Stress History Effects by Sex. Alcohol Clin Exp Res, 2021. 45(2): p. 480–489. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Westenbroek C, Den Boer JA, and Ter Horst GJ, Gender-specific effects of social housing on chronic stress-induced limbic Fos expression. Neuroscience, 2003. 121(1): p. 189–99. [DOI] [PubMed] [Google Scholar]
29.Brown KJ and Grunberg NE, Effects of housing on male and female rats: crowding stresses male but calm females. Physiol Behav, 1995. 58(6): p. 1085–9. [DOI] [PubMed] [Google Scholar]
30.Davis H, et al. , Pituitary-adrenal activity under different appetitive extinction procedures. Physiol Behav, 1976. 17(4): p. 687–90. [DOI] [PubMed] [Google Scholar]
31.Hughes RN, Neotic preferences in laboratory rodents: issues, assessment and substrates. Neurosci Biobehav Rev, 2007. 31(3): p. 441–64. [DOI] [PubMed] [Google Scholar]
32.Steimer T and Driscoll P, Divergent stress responses and coping styles in psychogenetically selected Roman high-(RHA) and low-(RLA) avoidance rats: behavioural, neuroendocrine and developmental aspects. Stress, 2003. 6(2): p. 87–100. [DOI] [PubMed] [Google Scholar]
33.Blanchard MM, Mendelsohn D, and Stamp JA, The HR/LR model: Further evidence as an animal model of sensation seeking. Neurosci Biobehav Rev, 2009. 33(7): p. 1145–54. [DOI] [PubMed] [Google Scholar]
34.Gruene TM, et al. , Sexually divergent expression of active and passive conditioned fear responses in rats. Elife, 2015. 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Vicente L and De la Casa LG, Context Properties Modulate Flavor Neophobia Habituation. Psicothema, 2021. 33(4): p. 617–622. [DOI] [PubMed] [Google Scholar]
36.George SA, et al. , Alterations in cognitive flexibility in a rat model of post-traumatic stress disorder. Behav Brain Res, 2015. 286: p. 256–64. [DOI] [PubMed] [Google Scholar]
37.Girotti M, Bulin SE, and Carreno FR, Effects of chronic stress on cognitive function - From neurobiology to intervention. Neurobiol Stress, 2024. 33: p. 100670. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Janak PH and Chaudhri N, The Potent Effect of Environmental Context on Relapse to Alcohol-Seeking After Extinction. Open Addict J, 2010. 3: p. 76–87. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Sex differences in the impact of trauma on alcohol self-administration

Nabiha Mahmood

Marian L Logrip

Abstract

INTRODUCTION