The effects of moderate prenatal alcohol exposure on performance in hippocampal-sensitive spatial memory and anxiety tasks by adult male and female rat offspring

Abstract

Moderate prenatal alcohol exposure (mPAE) results in structural alterations to the hippocampus. Previous studies have reported impairments in hippocampal-sensitive tasks, but have not compared performance between male and female animals. In the present study, performance in hippocampal-sensitive spatial memory and anxiety behavior tests were compared across adult male and female saccharin (SACC) control mPAE Long-Evans rat offspring. Two tests of spatial memory were conducted that were aimed at assessing memory for recently acquired spatial information: A delayed spatial alternation task using an M-shaped maze and a delayed match-to-place task in the Morris water task. In both tasks, rats in SACC and mPAE groups showed similar learning and retention of a spatial location even after a 2-h interval between encoding and retention. A separate group of adult male and female SACC and mPAE rat offspring were tested for anxiety-like behaviors in the elevated plus-maze paradigm. In this test, both male and female mPAE rats exhibited a significantly greater amount of time and a greater number of head dips in the open arms, while locomotion and open arm entries did not differ between groups. The results suggest that mPAE produces a reduction in anxiety-like behaviors in both male and female rats in the elevated plus-maze.

Introduction

Alcohol exposure during the gestation period can lead to Fetal Alcohol Spectrum Disorders (FASD), which describe a range of physical malformations in offspring, including neurobiological alterations, cognitive and learning problems, delays in language development, and anxiety disorders (Hoyme et al., 2016; Mattson et al., 2019; O’Connor & Paley, 2009; Popova et al., 2023). Although individuals with FASD often require long-term care including workplace and social support, there is increasing evidence that the vast majority are on the less severe end of the spectrum with milder forms of intellectual impairments as well as the absence of physical dysmorphology. Because these milder forms of FASD are difficult to diagnose, establishing behavioral assessments for early detection will help optimize interventions and health outcomes (Popova et al., 2023; Stevens et al., 2020).

Studies in humans diagnosed with an FASD indicate that the hippocampus is particularly vulnerable (Lebel et al., 2011), and studies using animal models of FASD report that alcohol exposure even at moderate doses can damage the hippocampus (Harvey et al., 2019; Marquardt & Brigman, 2016; Valenzuela et al., 2012). Specifically, moderate prenatal alcohol exposure (mPAE; blood ethanol concentration (BEC): ~30–100 mg/dL) produces significant reductions in mouse and rat hippocampal long-term potentiation (Brady et al., 2013; Fontaine et al., 2019; Patten, Brocardo, et al., 2013; Patten, Sickmann, et al., 2013; Sutherland et al., 1997; Varaschin et al., 1997; 2014; reviewed in Goncalves-Garcia & Hamilton, 2024), disrupts NMDA subunit expression in the mouse dentate gyrus (Brady et al., 2013), and disrupts enrichment induced neurogenesis in the mouse dentate gyrus (Choi et al., 2005). In addition, mPAE alters the expression of interneurons in the hippocampus (Madden et al., 2020) and related cortical regions (Kenton et al., 2020; Licheri et al., 2023). Recent work reports that mPAE can also alter rat hippocampal place cell activity and theta modulation of hippocampal neurons (Harvey et al., 2020), which are thought to play a pivotal role in the mapping of the spatial extent of the environment (O’Keefe & Nadel, 1978) and contribute more broadly to learning and memory (Buzsaki & Moser, 2013). Importantly, disruptions to place cell and hippocampal synaptic plasticity have been associated with spatial learning and memory deficits (Cacucci et al., 2008; Lester et al., 2017), and hippocampal interneuron function is thought to be critical for the expression of hippocampal rhythmic activity involved in spatial memory (Buzsaki, 2005; 2015).

Consistent with reports of hippocampal dysfunction, deficits in spatial behaviors have frequently been reported after mPAE (Harvey et al., 2019; Marquardt & Brigman, 2016), particularly on tests assessing the acquisition and retention of recent spatial experiences. For instance, Sutherland et al. (2000) demonstrated that male mPAE rats expressed impairments in a variant of the Morris water task in which animals were required to rapidly learn and retain spatial information across a delay interval (also see Savage et al., 2002). Other studies have shown that retention of learned spatial locations is impaired in rats, but these deficits have been reported only in adult male offspring (Savage et al., 2002, 2010). Thus, while many studies report impairments in adult male rodent models of mPAE (Brady et al., 2012; Sanchez et al., 2019; Savage et al., 2010), comparisons of performance of both males and females in spatial navigation tests has been limited. Recently, Kenton et al., 2020 demonstrated that mPAE mice exhibit sex-specific deficits in a non-match visual discrimination task in a touch screen apparatus. Here, female mice exhibited less accurate performance in selecting the correct visual stimulus, especially after a 15sec delay interval. Whether similar sex-specific differences in rat models of mPAE can be observed in spatial navigation tasks remains to be investigated.

In addition to spatial memory, the hippocampus plays a critical role in regulating anxiety-like behavior, particularly the ventral region of the hippocampus (for detailed reviews see Canteras et al., 2010; Fanselow & Dong, 2010). Indeed, previous work indicates that hippocampal place cells, which are significantly altered after mPAE (Harvey et al., 2020), can also be modulated by fear inducing and angiogenic stimuli in the environment (Malagon-Vina et al., 2023; Moita et al., 2004). Damage to the ventral hippocampus is known to reduce the expression of anxiety in tasks such as the elevated plus-maze (EPM), where rats are exposed to a four-armed maze with two arms open and two arms closed (Degroot & Treit, 2004; Kjelstrup et al., 2002). Thus, rats with ventral hippocampal damage spend more time in the open arms compared to controls, indicating a reduction in anxiety-like behavior. Although several experiments investigating affective behaviors after developmental alcohol exposure have reported mixed findings (reviewed in Marquardt & Brigman, 2016), one study reported a trend toward reduced anxiety-like behaviors in the EPM by adult female mPAE rats (Staples et al., 2013). Whether a similar pattern might be observed in male rats after mPAE has not been investigated. However, recent studies have demonstrated that low dose PAE can alter EPM behaviors in both male and female rats (Cullen et al., 2013), and a single exposure of alcohol on gestational day 12 can result in heightened anxiety-like behaviors in adolescent male rats (Rouzer et al., 2017).

The purpose of the present study was to investigate whether mPAE produces sex-dependent deficits in hippocampal-sensitive spatial memory and anxiety tasks. Because a previous study reported impairments in mPAE rats using tasks designed to assess retention of a recent experience (e.g., Kenton et al., 2020), we evaluated the performance of rats in a delayed spatial alternation task in an M-maze environment (sometimes referred to as a Wtrack), which requires rats to navigate between two spatial locations after a delay (Kim & Frank, 2009; Roy et al., 2022). Performance in the M-maze and in similar spatial alternation designs, depends on intact hippocampal and limbic neuronal activity and circuitry (e.g., Jadhav et al., 2012; Kim & Frank, 2009; Roy et al., 2022; Zhang et al., 2021). The M-maze task has also been extensively used to investigate the relationship between hippocampal place cell and hippocampal rhythmic activity during spatial learning and memory (e.g., Fernández-Ruiz et al., 2019; Frank et al., 2000; Jadhav et al., 2012; Singer et al., 2010, 2013; Zhang et al., 2021). Thus, delayed spatial alternation in the M-maze is well suited to investigate the impact of mPAE on hippocampal-sensitive spatial information processing.

We also tested rats in a delayed match-to-place task in a Morris water maze, which requires rats to retain the location of a novel platform after a variable delay, and is known to be dependent on the hippocampus and interconnected brain regions (Hales et al., 2021; Harker & Whishaw, 2002; Steele & Morris, 1999). While previous work has shown that male mPAE rats are impaired in similar non-match and match-to-place tasks (Brady et al., 2012; Savage et al., 2002; Sutherland et al., 2000), comparisons of performance between male and female mPAE rats has not been conducted. In addition to retention of recent spatial experience, a separate cohort of control and mPAE rats was evaluated for anxiety-like behavior in an EPM. Although a previous study from our group demonstrated that female mPAE rats exhibited altered behaviors on the EPM (Staples et al., 2013), similar studies have not been conducted in male rats using the same mPAE model. Here, we report that while mPAE does not produce impairments in retention of recent spatial experience in the M-maze and water task, alterations to anxiety-like behaviors were observed in both male and female mPAE rats.

Methods

Breeding colony and ethanol consumption procedures

Female adult rat breeders (Envigo Corporation, Indianapolis, IN) were exposed to a voluntary ethanol drinking paradigm using previously described procedures (Fig. 1) (Davies et al., 2023; Osterlund Oltmanns et al., 2022; Schaeffer et al., 2024-a). Briefly, Long-Evans rats were single-housed at 22 °C and maintained on a reverse 12h dark/12h light schedule and were provided PMI Picolab 5L0D laboratory rodent diet (LabDiet Incorporated, St Louis, MS) and tap water ad libitum. Pre-pregnancy drinking levels in female rats were evaluated while gradually acclimating them to drinking 5% EtOH in 0.066% saccharin in tap water for 4h each day from 10:00–14:00 h for at least two weeks, and the mean daily ethanol consumption was determined for each female rat. Following two weeks, females who drank at levels less than one standard deviation below the entire group mean were removed from the study the remaining breeders were assigned to either a saccharin control or 5% ethanol drinking group. The female breeder rats were then housed with male breeders until pregnancy was achieved. Ethanol consumption did not occur during breeding. Beginning on day one of gestation, dams were given access to saccharin water containing either 0% (v/v) or 5% (v/v) ethanol for 4 h each day (10:00–14:00). The volume of saccharin water provided to the control group was matched to the mean volume of ethanol saccharin water consumed by the ethanol group. The volume of ethanol consumed was recorded daily for each rat dam. In the current study, the mean consumption was 2.24 g/kg. A recent study has shown that this level of ethanol consumption, coupled with the 5L0D diet, produces a mean peak maternal serum ethanol concentration of 46.0 + 3.2 mg/dL (Davies et al., 2023). At birth, daily ethanol consumption was discontinued, and the litters were weighed and culled to 10 pups. To minimize potential litter effects, only one or two female or male rats were used from each litter.

Fig. 1.

Experimental timeline for moderate prenatal alcohol exposure procedures and behavioral testing. In adulthood, male and female rats were assigned to one of the spatial memory and anxiety tasks. GD = gestational day; PND = postnatal day

Subjects

Subjects included 52 female and 52 male Long-Evans rats from each of the two prenatal treatment groups obtained from the University of New Mexico Health Sciences Animal Resource Facility (see breeding protocol above). Following weaning, all rats were pair-housed in standard plastic cages on a reverse r12h dark/12h light schedule at a room temperature of 22 °C with food (5L0D) and water provided ad libitum. The University of New Mexico central campus and Health Sciences Center Institutional Animal Care and Use Committee approved all procedures for the current study.

M-maze apparatus

To investigate delayed non-match spatial alternation behavior, we used an M-maze configuration (Fig. 2B). Behavioral testing was conducted within a custom M-maze configuration, comprised of three vertical arms measuring 140 cm in length, and two horizontal tracks measuring 110 cm in width, positioned at the ends of the arms. Together the apparatus formed a figure eight modifiable by removable and adjustable dividers. Each arm and track had a width of 10.2 cm, and cemented walls measuring 4.4 cm in height. The entire apparatus was elevated 13.9 cm above a blue-gray floor, providing a suitable point for observation. The maze was located in a room containing visible objects such as a white noise generator, computers, computer benches, a sink, a cabinet, and shelves. An overhead camera connected to a Mac computer (QuickTime Player) was used to create digital recordings of each test session.

Fig. 2.

Timeline for habituation, training, and probe tests in the M-maze delayed spatial alternation task (A). Schematic of the M-maze showing the sample, delay, and choice phases of the task (B)

Delayed spatial alternation training and probe test

Saccharin (SACC) (n = 20; 10 male and 10 female) and mPAE rats (n = 20; 10 male and 10 female) were tested on a delayed spatial alternation task in the M-maze (Fig. 2). Rats were tested in small cohorts of ~2 rats per sex/group but with all animals receiving training and probe testing between PND 92–137. Prior to testing, rats were placed on a food restricted diet of 85% of their ad libitum weight and given access to water ad libitum. Food restriction continued throughout pre-training and training in the task. Pretraining included habituation trials in which rats were placed in the maze daily for 10 min or until all rewards were collected, for four days, during which food rewards (Fruit Loops) were distributed along the maze surface (Fig. 2A). During training on the task, food rewards were allocated at the end of each arm of the apparatus and the rats were trained to alternate between the outer arms of the maze. During the Sample phase of the task, rats navigate from the center arm toward either the left or right outbound arm, where they collect a reward (Fig. 2B). Only one of the arms was open during the Sample phase. The open arm was randomly selected but the number of left and right arms choices were counterbalanced during the Sample phase. After consuming the reward, rats returned to the center stem reference point, where they collected another treat and were held for a 15 s delay. This segment of the task was called the Delay phase (Fig. 2B). The Choice phase followed the delay phase in which rats were released from the central stem and were then required to navigate to the opposite arm for reward (Fig. 2B). During the Choice phase, both outer arms were open. Thus, rats were required to recall what direction they came from initially (e.g., left arm) during the Sample phase and produce the correct trajectory for reward (e.g., proceed to the right arm) during the Choice phase. Failing to navigate to the opposite arm during the Choice phase of the task was considered an error and resulted in the animal not receiving a reward. Rats were trained until they achieved 90% accuracy in 10 trials for two consecutive days (i.e., 9 correct trials per day). Once rats reached a criterion of 90% accuracy for two consecutive days, probe tests were administered on two separate (consecutive) days. During each probe test, rats were given ten trials in which the delay interval was either 15 or 30 s in duration. Delay durations were evenly counterbalanced during each probe session (5 trials/day of each delay interval). Rats that failed to reach criteria after 18 days of training were excluded from probe testing.

Morris water maze

As in our previous studies (e.g., Berkowitz et al., 2018; Pentkowski et al., 2018), a high walled circular pool was used for the delayed match-to-place task (150 cm diameter, 48 cm high). A plastic escape platform covered with a metal grate (16 cm × 16 cm, 25 cm high) was placed within the pool. The pool was filled with water (20–22 °C) until the level reached ~2.5 cm above the top of the platform. Non-toxic white paint was used to make the water opaque. Environmental cues (posters and furniture) were maintained in fixed locations throughout the duration of the experiment. An overhead camera recorded swim behavior on a Mac computer (QuickTime Player) for subsequent analysis.

Delayed match-to-place task

A separate group of SACC (n = 16; counterbalanced for sex) and mPAE (n = 16 counterbalanced for sex) rats were tested between PND 197e208 on the delayed match-to-place variant of the water task (Fig. 4A). The delayed match-to-place task was conducted similar to our previous work (Pentkowski et al., 2018; also see Hales et al., 2021; Harker & Whishaw, 2002; Steele & Morris, 1999). Briefly, rats were given 2 trials per day for 9 consecutive days, but with the hidden platform moved to a new location each day. On trial 1, the rat was required to search and locate the new location of the hidden platform (i.e., the encoding trial). On trial 2, the rat was tested for memory for the new platform location (i.e., the retention trial). The starting position for a given subject remained the same for both trials. Daily platform positions varied with respect to quadrant location and distance from the pool wall. Start positions were selected pseudo randomly from 4 locations located along the perimeter of the pool. On each trial, rats were placed by hand into the water facing the wall of the pool at 1 of the 4 starting positions. If the rat failed to reach the platform within 90 s, it was directed to the platform location with assistance from the experimenter. Once on the platform, rats remained there for 30 s. After the first (encoding) trial, rats were placed in a holding cage for a delay interval of either 15 s, 20 min, or 120 min, before beginning the second (retention) trial. Each delay interval was used three times across the 9 testing days. Rats were not habituated to the pool or environment prior to testing in the delayed to match-to-place task.

Fig. 4.

Effects of mPAE or SACC exposure and sex on retention of spatial learning in the delayed match-to-place task. Illustration showing the encoding (trial 1), delay, and retention (trial 2) phases of the water task (A). Representative swim path is plotted for encoding and retention trials. The Bar plots represent the latency (B) and path length difference (C) for each of the three delay intervals.

Elevated plus-maze

The same EPM design has been used in our previous work (Pentkowski et al., 2018, 2022). Briefly, the EPM apparatus consisted of 4 Plexiglas arms arranged in a plus-shaped configuration elevated 75 cm above the floor. Each arm was 10-cm wide and 50-cm long, and each arm was joined at the center by a 10-cm square platform. The 2 “open” arms contained no walls, whereas the 2 opposite “closed” arms contained 40-cm tall opaque sides. A white noise generator was used during EPM testing to standardize background noise.

Elevated plus-maze task

A separate group of adult SACC (n = 16; counterbalanced for sex) and mPAE rats (n = 16; counterbalanced for sex) were tested between PND 161–166 on the EPM using previously published protocols (Pentkowski et al., 2018, 2022). Briefly, each individual rat was initially placed in the center arm of the apparatus facing 1 of the 2 closed arms. Tests were 5min in duration and were conducted under red light. The EPM was thoroughly cleaned between each trial using a 10% ethanol solution. Importantly, rats were not habituated to the EPM or testing room before the experiment. All test trials were recorded from a camera positioned over the center of the EPM apparatus with a second camera providing a “side view” facing the open arms.

Behavioral analysis

Delayed spatial alternation task

Behaviors in the delayed spatial alternation task were classified into three categories: correct, incorrect, or non-responsive errors. A trial was deemed correct if the animal successfully navigated down the alternate arm during the choice phase, deviating from the initial (sample) trajectory. Conversely, a trial was classified as incorrect if, during the choice phase, the rat continued along the arm they explored during the sample phase. When the rat failed to navigate toward an outbound arm during the choice phase, it was categorized as a non-responsive error.

Delayed match-to-place task

Video records of swim paths were collected by a camera mounted directly above the swimming pool. Tracked x-y- coordinates of the animal’s back were acquired for each video frame (10 frames/sec) using DeepLabCut software (Mathis et al., 2018). From the swim path recordings, latency and path length were measured for each trial using Matlab scripts (The MathWorks, Natick, MA). Task performance is typically characterized by long paths on trial 1 (when the platform location is unknown), and shorter and direct paths on trial 2. Thus, a difference score (i.e., savings score; Steele & Morris, 1999) was calculated from latency and path length measures by taking the difference in measures between trial 1 and trial 2. Difference scores greater than 0 indicate the use of shorter swim paths during trial 2.

Elevated plus-maze

As in our previous studies (Pentkowski et al., 2018, 2022), behavioral measures in the EPM included the duration of time spent in the open versus closed arms; head-dips which were quantified as an extension of the rats head over the edge of an open arm; the number of open arm entries defined as all four paws moving from a closed-to an open-arm; and total distance traveled measured as total movement in meters during the test period. Location measures in the EPM were analyzed using ANY-maze (Stoelting Co, Wood Dale, IL); rat movements were automatically tracked using the midline as the frame of reference. Head dips were scored by a single trained observer blind to group assignment.

Statistical analysis

Independent sample t-tests were used to compare mPAE paradigm outcome measures. Manne–Whitney U tests were conducted on non-normally distributed data. ANOVAs and linear mixed model approaches were used to investigate treatment group and sex main effects and interactions. For two-way ANOVAs, data were log transformed in cases of violations of homogeneity of variance (a significant Levene’s test). For repeated measures ANOVAs, a Greenhouse-Geisser correction was used in cases of violations of the assumption of sphericity. Partial eta squared (${\text{η}}^2\text{p}$) values were reported as a measure of effect size. For probe tests in the delayed spatial alternation task, a linear mixed model was implemented which allowed us to account for the imbalance in sample size (not all animals reached acquisition training criteria). Nested random effects of subjects within cohort, and subjects within litter were accounted for in the model to address potential sources of variation. All ANOVAs were conducted with JAMOVI (version 2.3). Linear mixed model analyses were conducted using R.

Results

Paradigm outcome measures

A summary of the mPAE paradigm outcome measures for the rat dams providing offspring for the studies reported here are shown in Table 1. In this study, the rat dams in the 5% ethanol group consumed a mean of 2.24 ± 0.06 g ethanol/kg body weight on average throughout gestation. In a recent study using a separate set of rat dams, a similar level of consumption produced a mean serum ethanol concentration of 46.0 ± 3.0 mg/dL when sampled 2 h after the introduction of drinking tubes (Davies et al., 2023). This mPAE paradigm had no significant effects on maternal weight gain during pregnancy (t = 0035, p = 0.97) nor mean pup weight per litter (t = −0.835, p = 0.41), nor on litter size, data which was not normally distributed (Manne–Whitney U = 314, p = 0.51).

Table 1.

Effects of daily 4-h consumption of 5% EtOH on female rat dams and their offspring.

	Saccharin Control	5% Ethanol Group
Daily 4 h 5% ethanol consumption	NA	2.24 ± 0.06a (27)
Maternal Weight Gain during pregnancy	111 ± 6b (26)	111 ± 7 (27)
Litter Size	11.5 ± 0.3c (26)	10.9 ± 0.5 (27)
Pup birth weight	7.78 ± 0.19d (26)	8.00 ± 0.17 (27)

NA-not applicable.

(n)-Group sample size.

^aMean ± S.E.M. grams ethanol consumed/kg body weight/day.

^bMean ± S.E.M. grams increase in body weight from GD 1 through GD 21.

^cMean ± S.E.M. number of live births/litter.

^dMean ± S.E.M. grams pup birth weight.

Delayed spatial alternation task

Acquisition

Fig. 3A plots the number of days to criteria for each group in the M-maze delayed spatial alternation task. The ANOVA indicted a significant difference in days to criteria based on the sex variable (F (1,36) = 6.92, p = 0.01, ${\text{η}}_{\text{p}}^2=0.161$). This finding indicates that sex has an effect on the time it took to meet the criteria, with females (mean ± SEM; SACC: 10.2 ± 1.4 days; mPAE: 9.8 ± 1.6 days) requiring more testing days to reach criteria compared to male rats (SACC: 5.7 ± 1.2 days; mPAE: 6.7 ± 1.5 days). Although 3 rats within the mPAE treatment group, comprising two females and one male, did not meet criterion over the course of an 18-day training period, the ANOVA did not indicate significant differences for the type of treatment administered on days to criteria (F (1,36) = 0.043, p = 0.84, ${\text{η}}_{\text{p}}^2=0.001$). Thus, the results indicate that the type of treatment administered did not appear to have any significant impact on the time it took animals to meet the criteria.

Fig. 3.

Effects of mPAE or SACC exposure and sex on acquisition and probe testing behaviors in the delayed spatial alternation task. Bar graphs represent the number of days to criterion during acquisition (A), and percent correct as a function of delay interval during probe tests (B) (Mean ± SEM). Open circles represent data from each rat within each group. Pound signs (##) represents a significant sex difference (p < 0.01).

Probe Test

After reaching criteria, rats were given two consecutive daily probe tests with 15 s or 30 s delay intervals randomly presented across trials (Fig. 3B). A linear mixed model was implemented to examine both the main effect and interaction of probe trials, sex, and treatment. Nested random effects of subjects within cohort, and subjects within litter were accounted for in the model to address potential sources of variation. Given the limited number of offspring sampled per litter, the influence of the litter factor on the statistical outcomes was negligible. Instead, the random effect of cohort was accounted for in the model to address variations across multiple cohorts trained at different periods. The analysis indicated that the prenatal treatment administered to rats had no significant effect on performance accuracy (β₁ = 0.16, p = 0.12), indicating that mPAE animals performed, on average, 0.16 units higher than the animals in the control group. In addition, the effect of sex (β₂ = 0.17, p = 0.11) was not statistically significant, suggesting that sex did not influence performance accuracy during probe testing. Additionally, delay intervals (β₃ = 0, p = 0.52) were also not significant, suggesting that delay intervals do not impact animals’ performance in the retention delayed testing in the task. At a 15-s delay, the mean score was 4.28 (SE = 0.07, 95% CI [4.15, 4.42]), and at a 30-s delay, the mean score was 4.23 (SE = 0.07, 95% CI [4.09, 4.36]). Although a slight decrease in mean score was observed with a longer delay, these differences were not statistically significant.

The two-way interactions between treatment and delay (β = 0.06, SE = 0.18, t(DF) = 0.359, p = 0.72), and between delay and sex (β = −0.05, SE = 0.18, t(DF) = −0.283, p = 0.78) were also found to be non-significant. Moreover, the analysis indicated a non-significant three-way interaction among treatment, sex, and delay (β = −0.28, SE = 0.36, t(DF) = −0.773, p = 0.44). This suggests that the relationship between the outcome variable and the predictors is not significantly influenced by the simultaneous presence and interaction of prenatal treatment, sex, and delay. Overall, the delay did not have a significant effect on the animals’ performance during the probe test.

Delayed match-to-place task

A separate group of SACC and mPAE rats were tested on the delayed match-to-place variant of the water task (Fig. 4A). Fig. 4B shows difference scores for latency and path length for rats in SACC and mPAE groups. Difference scores, which were generally greater than 0 for each delay interval, suggested that rats in mPAE and SACC group displayed evidence of 1-trial learning for daily repositioned platform locations. ANOVAs indicated a significant effect of delay for latency difference scores (F(2,54) = 3.6, p = 0.03, η²_p = 0.12), but did not detect group effects or interactions for either difference score (ps > 0.10). Thus, male and female mPAE rats were unimpaired in tests of recent spatial learning in the delayed match-to-place variant of the water task.

Elevated plus-maze

A separate group of SACC and mPAE rats were tested for anxiety-like behavior on the EPM. Two rats (1 female from the SACC group and 1 male from the mPAE group) fell from the maze during testing and were removed from subsequent analyses. Fig. 5 displays measures of anxiety-like behaviors and locomotor activity for groups during EPM performance. On average, mPAE rats spent more time in the open arms and made more head dips in the open arms than SACC controls. This observation was supported by significant differences between mPAE and SACC animals for time in the open arms (F(1,26) = 4.70, p = 0.04, ${\text{η}}_{\text{p}}^2=0.15$), and the number of head dips, which was log transformed for statistical purposes due to a violation of homogeneity of variance (F(1,26) = 10.8, p = 0.003, ${\text{η}}_{\text{p}}^2=0.29$). Although there were no significant sex or interaction effects for these measures (ps > 0.13), the ANOVA trended toward a significant sex difference for head dips (F(1,26) = 3.63, p = 0.07, ${\text{η}}_{\text{p}}^2=0.13$). There were no significant treatment differences on measures of open arm entries or total distance travelled (ps > 0.45), however, the ANOVA detected a significant sex difference in total distance traveled with male rats walking less during the test (F(1,26) = 5.15, p = 0.03, ${\text{η}}_{\text{p}}^2=0.17$).

Fig. 5.

Effects of mPAE or SACC exposure and sex on anxiety-like behavior in the EPM. Bar plots show the percent of time in the open arms (A), head dips (B), total distance traveled (C), and the number of open arm entries (D). Regardless of sex, mPAE reduced anxiety-like behavior compared to SACC controls, including an increase in percent open-arm time (3A) and the number of head dips (3B). For the number of head dips, the raw data is shown but for statistical purposes was log transformed due to a violation of homogeneity of variance. There were no effects of mPAE on total distance travelled (3C) or the number of open-arm entries (3D). Females exhibited an increase in locomotor activity compared to males (3C). Asterisks (*p < 0.05; ^**p < 0.01) represents a significant difference between mPAE and SACC and pound sign (#p < 0.05) represents a significant sex effect.

Discussion

In three experiments, we investigated the potential role of sex in mediating the impact of mPAE on hippocampal-sensitive spatial memory and anxiety tasks in adult Long-Evans rats. The results of the study support two general conclusions. First, the findings from the delayed spatial alternation in the M-maze and the delayed match-to-place task in the Morris water task demonstrate that retention or recall of recent spatial experiences is unaffected in both male and female mPAE rats. In the M-maze alternation task, rats from both groups required a similar number of training sessions to achieve performance criteria (Fig. 3A), however, the study found that female rats required a significantly greater number of training days to reach criteria in the M-maze regardless of group. After reaching criteria in the M-maze, all rats were administered probe tests in which the delay between sample and choice was doubled in time. In sum, the probe test did not reveal significant treatment or sex differences (Fig. 3B). In the delayed match-to-place water task, our analysis demonstrated that both sexes of mPAE and control groups displayed evidence of 1-trial memory retention with latency and path length measures decreasing from trial 1 to trial 2 (i.e., higher latency and path length difference measures; see Fig. 4B–C). There was some evidence that longer delay intervals were more challenging in that recall latencies and path lengths were longer especially after the 2 h delay. Nonetheless, the findings from these navigation tasks support the conclusion that retention of recent spatial experience does not appear impaired after mPAE and does not significantly vary with sex.

Although the results from the present spatial navigation studies are consistent with previous reports that rats from both sexes can learn a hidden platform location in the Morris water task after mPAE (also see Savage et al., 2002, 2010; Sutherland et al., 2000), the absence of retention deficits stands in contrast with studies reporting an impact of mPAE on adult male rats performance in spatial tasks. Deficits in male mPAE rats have been reported in tasks requiring discrimination of objects conditionally associated with places (Sanchez et al., 2019), discriminating between spatial locations in a radial maze (Brady et al., 2012), and discriminating between stimuli among an array of visual-spatial stimuli (Kenton et al., 2020). It is possible that the discrepancy between the present study and previous studies relate to task design. For instance, Brady et al. (2012) reported deficits only when spatial locations (sample and choice arms) had considerable feature overlap. In the present study, we used an M-maze which has previously been shown to elicit substantial hippocampal place cell activity along the maze arms and memory-related hippocampal rhythmic activity (i.e., sharp wave-ripples) in the delay area (see Fernández-Ruiz et al., 2019; Frank et al., 2000). However, the arms are located on opposite sides of the maze and did not carry featural (spatial cue) overlap (as in Brady et al., 2012). In other words, the two outbound segments were located on opposite sides of the test room and were likely associated with distinct distal (extra-maze) landmarks.

An additional consideration is that retention deficits may be observed with longer delay intervals, especially in the delayed spatial alternation task where the delay was limited to a maximum of 30 s during probe tests. The initial design of the studies was based on experiments showing deficits in a non-match visual discrimination task where female mPAE mice were observed to make a greater number of errors especially after 15-sec delay intervals (Kenton et al., 2020). Further, studies report that damage to afferent input from the entorhinal cortex or hippocampus can produce impairments in a delayed match-to-place water task with similar retention intervals used in the present study (Hales et al., 2021; Harker & Whishaw, 2002; Steele & Morris, 1999). Support for the idea that greater retention intervals are needed to expose deficits after mPAE comes from studies by Savage et al. (2002; 2010) which are also supported by work from Cullen et al. (2014). First, Savage et al. trained rats in a fixed hidden platform variant of the Morris water task in a single session composed of 12 trials followed by a retention test 4 days later in which they were given 12 additional test trials. As expected, rats in both the mPAE and control groups expressed similar place learning during initial training. However, when returned for retention testing, rats showed greater escape latencies on the first few trials. Thus, retention of spatial experience across the 4-day interval was significantly impaired in the mPAE rats. In the study by Cullen et al. the authors reported intact retention of a learned water maze location after 24 h by rats exposed to a low dose of alcohol prenatally. Collectively, these previous studies suggest that retention deficits are more likely to be observed after delays greater than 24 h.

In the M-maze task, we found that female rats in both control and mPAE groups required a greater number of training days to reach criteria (Fig. 3A). The present findings are consistent with a recent meta-analysis reporting a small but appreciable trend toward an advantage in performance by male rats in spatial tasks (Jonasson, 2005). For instance, Seymoure et al. (1996) reported a male advantage by rats trained in a radial arm task which consisted of reinforcing navigation to a set of baited maze arms that were spatially fixed related to distal environmental cues. On average, male rats made fewer errors (entering the incorrect arm or revisiting a reinforced arm) compared to female rats. Sex differences in spatial behavior suggest the involvement of sex hormones which are known to influence the structural plasticity of the hippocampus (Duarte-Guterman et al., 2015; Hamson et al., 2016; Sheppard et al., 2019). Sex differences in spatial behavior may also be attributed to variation in the types of learning and navigation strategies deployed by male and female rodents (Devan et al., 2016; Kanit et al., 2000; Saucier et al., 2002). In the Morris water task, Devan et al. (2016) reported that female rats spent more time searching along the periphery (thigmotaxis), while male rats exhibited a more directed spatial search toward the location of the platform. It has also been argued that variations in task procedures such as the sex of the experimenter can influence performance (in the present study, the M-maze was conducted by a female experimenter) (e.g., Faraji et al., 2022; Georgiou et al., 2022). It has been previously noted that the strain or species of rodent used in the study presents an important variable for the study of sex differences in spatial behavior including the study of the neural correlates of learning and memory (Hok et al., 2016; Jonasson, 2005; Mou et al., 2018). Future studies could investigate these variables with respect to sex differences in M-maze performance.

A second conclusion stemming from the present results is that mPAE decreases anxiety-like behavior in adult male and female offspring. Specifically, regardless of sex, mPAE offspring showed an increase in the duration of time spent in the open arms (Fig. 5A) and an increase in the number of head dips (Fig. 5B), results that are similar to the effects produced by standard anxiolytic drugs (for a detailed review see: Carobrez & Bertoglio, 2005). There were no differences between SACC and mPAE groups in the number of open arm entries (Fig. 5C) or total distance traveled (Fig. 5D), indicating that there were no differences in general activity or the opportunity to explore the open arms. Interestingly, using the same mPAE model that was used in the present study, Staples et al. (2013) reported a trend toward reduced anxiety-like behavior in adult female offspring during EPM testing. Although the latter effects were only statistical trends, mPAE rats showed a 47% increase in time spent in the open arms compared to SACC controls. Collectively, these data indicate that mPAE throughout gestation decreases anxiety-like behavior in adult male and female rats.

The literature examining the impact of PAE on measures of anxiety has produced mixed outcomes, which likely involve critical experimental differences that impact the development of anxiety in adulthood including the route of alcohol delivery, and the dose, pattern and timing of alcohol exposure. This limitation, coupled with different experimental procedures for exposing pregnant dams to alcohol (e.g., dose, timing and route of administration) and the role of the time of testing (adolescence vs. adulthood) has led to an unclear understanding of the consequences of PAE on anxiety-like behavior in adult offspring (Marquardt & Brigman, 2016). Indeed, differential effects on anxiety measures vary depending on alcohol dose (moderate vs. heavy) exposure (1st, 2nd or 3rd trimester) and anxiety model (EPM vs. open field test – OFT; for a detailed review see Marquardt & Brigman, 2016). For example, heavy exposure (~350 mg/dL BAC) via oral gavage during portions of the 1st and 2nd trimesters either fail to alter or increase, respectively, anxiety-like behavior in adult male rats during testing in the OFT or EPM (Dursun et al., 2006). Similarly, following heavy PAE exposure (~150–160 mg/dL BAC) via liquid diet during the 1st and 2nd trimesters, adult male and female offspring either show no changes in anxiety-like behavior in the OFT (Hellemans et al., 2010) or EPM (Hellemans et al., 2008), or show decreased anxiety-like behavior in the EPM (Gabriel et al., 2006; Lam et al., 2018). Studies using mPAE models incorporating a liquid diet or voluntary drinking approach indicate that offspring either show increased (EPM, 1st and 2nd trimesters, BAC ~30 mg/dL, adult male and female rats; Cullen et al., 2013), decreased (zero maze, 2nd and 3rd trimesters, BAC ~20 mg/dL, adult male rats; Cagiano et al., 2002) or null effects (EPM, 2nd and 3rd trimesters, BAC ~15 mg/dL, adult male rats; Barbaccia et al., 2007) on anxiety-like behavior. Collectively, this literature indicates that the directional effects of PAE on anxiety is dependent on the model used, suggesting that the effects of PAE in human offspring likely depend on alcohol dose, timing and duration of exposure. Variability across studies may also reflect differences in task features, suggesting that future study designs might benefit by testing animals in a “battery” of assessments for anxiety-like behavior (e.g., Rouzer et al., 2017).

The present results indicating that mPAE offspring show reduced anxiety-like behavior seems to conflict with outcomes in humans exposed to gestational alcohol. Indeed, children and adults with FASD have increased rates of various psychopathologies, including anxiety (Famy et al., 1998; Pei et al., 2011; Steinhausen & Spohr, 1998). Although increases in open arm time and the number of head dips in PAE offspring represent two classic indices of reduced anxiety-like behavior in the EPM (Carobrez & Bertoglio, 2005), two alternative interpretations should be considered. First, rather than reduced anxiety per se, these behaviors could represent inappropriate defensive behaviors in response to potential threat (i.e., increased risk for predation in open spaces). As suggested by Raineki et al. (2016), this inappropriate behavior might reflect a pathological increase in novelty seeking or an inability of mPAE offspring to effectively use environmental cues to inhibit inappropriate behavior. Alternatively, the present results indicative of reduced anxiety may be dependent on specific features related to timing or dosage, with other models using lower doses or with exposure restricted to specific gestational days showing enhancement of anxiety (e.g., Cullen et al., 2013; Rouzer et al., 2017). Future research probing the neural circuity underlying these potential dose and timing-dependent effects are needed in order to draw firm conclusions regarding the directional effects of mPAE on anxiety.

The neurobiological mechanisms contributing to anxiety-like behavior likely involves several interconnected neural circuits (Canteras et al., 2010; Fanselow & Dong, 2010; reviewed in Pentkowski et al., 2021), but the hippocampal formation is proposed to have a prominent role and is known to undergo significant alterations following mPAE. Studies have revealed functional differentiation in the functions of the dorsal vs. ventral hippocampus (Risold & Swanson, 1996; Moser & Moser, 1998), with studies suggesting that the dorsal hippocampus has a more prominent role in spatial memory and the ventral hippocampus has greater involvement in emotional behaviors including modulation of anxiety-related behaviors. The ventral hippocampus is comprised of a heterogenous cell populations that are thought to make an important contribution to anxiety-like behaviors, and are also affected by PAE. Notably, glutamate antagonists (Hackl & Carobrez, 2007; Marrocco et al., 2012) and GABA_A receptor agonists (McEown & Treit, 2010; Zhang et al., 2014) infused into the ventral hippocampus disrupts anxiety-like behavior in the EPM. Additionally, previous work has demonstrated excitatory and inhibitory system alterations after PAE. For instance, studies have reported that PAE leads to GABA_A receptor hyperexcitability (Olney, 2002) and alters the expression of parvalbumin positive cells in dorsal but not ventral CA1 (Reid et al., 2021). Using the same model as the present study, Madden et al. (2020) reported reductions in parvalbumin positive interneuron cell number in dorsal hippocampus of both male and female adult rats, but expression in ventral hippocampus was not investigated. Recent work has also shown that CRF (corticotropin-releasing factor) receptors, which are found in the hippocampal formation and throughout the limbic system and are known to have a significant influence on the expression anxiety-like behaviors (Pentkowski et al., 2009), can be altered after PAE (Rouzer & Diaz, 2022). Indeed, a recent study reported functional deficits in the expression of CRF type-1 receptors in the central amygdala in adult male rats following a single exposure of alcohol on gestational day 12 (Rouzer & Diaz, 2022). By extension, future work could be directed at determining whether the mPAE model used in the present study impacts similar CRF-associated mechanisms and excitatory-inhibitory signaling within ventral hippocampal circuitry.

In conclusion, the present results suggest that mPAE does not have a considerable effect on delayed non-match spatial alternation behaviors in the M-maze or in the delayed match-to-place task in the Morris water task. However, mPAE led to reduced anxiety-like behaviors in the EPM, effects that occurred in both male and female mPAE rats. Future studies should begin examining the neurobiological bases of anxiety-like behaviors in mPAE offspring. Specifically, studies should evaluate whether the mPAE model used here impacts excitatory-inhibitory systems, and CRF-associated mechanisms involved in the expression of anxiety phenotypes in mPAE offspring.