Skip to main content
NCBI home page
Alcohol and Alcoholism (Oxford, Oxfordshire) logoLink to Alcohol and Alcoholism (Oxford, Oxfordshire)
. 2022 Mar 8;57(4):413–420. doi: 10.1093/alcalc/agac008

Rat Model of Late Gestational Alcohol Exposure Produces Similar Life-Long Changes in Thalamic Nucleus Reuniens Following Moderate- Versus High-Dose Insult

Zachary H Gursky 1, Anna Y Klintsova 2,✉
PMCID: PMC9270984  PMID: 35258554

Abstract

Aims

Recent studies have recognized that thalamic nucleus reuniens (Re) undergoes substantial neuron loss following alcohol exposure (AE) during the brain growth spurt (BGS). As all previous studies have utilized high-dose AE paradigms, we tested whether moderate-dose AE is capable of damaging Re to a similar degree as high-dose AE.

Methods

We used a rat model of third-trimester binge AE (relative to human pregnancy) to administer ethanol to rat pups at either a high (5.25 g/kg/day) or moderate (3.00 g/kg/day) dose during the BGS (postnatal days [PD] 4–9) via intragastric intubation. In adulthood (i.e. PD72), we quantified the volume of Re as well as the total number of neurons and non-neuronal cells in the nucleus (which were further divided into microglia versus ‘other’ non-neurons), using unbiased stereological estimation of cells identified with immunofluorescent markers (i.e. nuclear label Hoechst, neuron-specific protein NeuN, and microglia-specific protein Iba1). Data were analyzed both between-treatment and correlated with peak blood alcohol concentration (BAC).

Results and conclusions

We observed significant neuronal and non-neuronal cell loss in both the high-dose and moderate-dose AE groups (relative to both procedural control and typically-developing control groups), which mediated reductions in Re volume. Outcomes did not correlate with peak BAC, further supporting that Re is vulnerable to AE-induced neurodegeneration at lower doses than previously suspected. Given the role that Re has in coordinating prefrontal cortex and hippocampus, the current study highlights the role that thalamic damage may play in the range of behavioral alterations observed in Fetal Alcohol Spectrum Disorders.

Short Summary: We administered high- or moderate-dose alcohol to rats at a neurodevelopmental stage analogue of human third trimester. When these animals were examined in adulthood, we observed significant damage to thalamic nucleus reuniens (reductions in neuron and non-neuronal cell number, regional volume loss) at equal magnitude between doses (uncorrelated with peak blood alcohol concentration).

INTRODUCTION

Alcohol consumption during pregnancy remains a pervasive problem, despite attempts at reducing its prevalence (Popova et al., 2017). An estimated 8% of all live births are impacted by alcohol exposure during late gestation (Umer et al., 2020), in a period of rapid fetal brain development often referred to as the brain growth spurt (BGS) (Dobbing and Sands, 1979). Preclinical studies using rodent models of late gestational alcohol exposure have determined that alcohol exposure during the BGS causes widespread cell death and neuron loss (Ikonomidou et al., 2000). Among the most vulnerable brain structures at this period of development are the medial prefrontal cortex (mPFC), hippocampus (HPC), and thalamus (Ikonomidou et al., 2000)—particularly thalamic nucleus reuniens (Re) (Gursky et al., 2019). These three structures (i.e. mPFC-Re-HPC) communicate to give rise to a class of complex behaviors collectively referred to as ‘executive functions’ (EF) (Dolleman-van der Weel et al., 2019).

Individuals diagnosed with Fetal Alcohol Spectrum Disorders (FASD) often display behavioral alterations in working memory, inhibition, and set shifting (Khoury et al., 2015), three prominent domains of EF which also happen to be dependent on mPFC-Re-HPC circuitry (Cassel et al., 2013; Dolleman-van der Weel et al., 2019). It is widely recognized that AE during the BGS causes damage to HPC (Berman and Hannigan, 2000), and recent studies have increased the acceptance that mPFC is also damaged (Otero et al., 2012; Hamilton et al., 2017; Heroux et al., 2019). However, the only two studies to explicitly examine Re (i.e. the structure responsible for coordinating mPFC-HPC activity) to date have been published in the last 2 years (Gursky et al., 2019; Gursky et al., 2020). The first of these two studies demonstrated that Re is selectively damaged within the ventral midline thalamus, displaying substantial neuron loss and volume loss in adult female rats following high-dose AE during the BGS (Gursky et al., 2019). No such impairments were observed in the neighboring rhomboid nucleus of the thalamus. The second study determined that alcohol-induced apoptotic neuron loss was the mechanism underlying permanent neuronal number deficits observed in Re, again using a high-dose AE paradigm, but extending the finding to both males and females (Gursky et al., 2020). Although these studies are the first of their kind, both use a high-dose AE paradigm with peak blood alcohol concentrations (BACs) of ~330–380 mg/dl. As peak BAC is a robust predictor of AE-induced neurodegeneration (Bonthius and West, 1990; Ikonomidou et al., 2000), and individuals diagnosed with FASD display a range of behavioral and neuroanatomical alterations due to variability in dose and timing of alcohol consumption (Khoury et al., 2015; Wozniak et al., 2019), it remains to be seen whether Re is damaged throughout the lifespan following exposure to more moderate doses of AE during the BGS than those used in previous studies.

The current study reconciles the paucity of knowledge regarding the sensitivity of Re to alcohol-induced alterations during the BGS. We initially hypothesized that Re would undergo a dose-dependent reduction in damage, predicting that a moderate-dose variant of the widely accepted high-dose model of AE during the BGS (Gursky and Klintsova, 2017) capable of producing peak BACs around the toxic threshold for neuronal apoptosis (i.e. 200 mg/dl for 2 hours) (Ikonomidou et al., 2000) would produce intermediate levels of neuron and volume loss. We further examine AE effects on microglia as neuroimmune alterations have been linked to neurodegeneration and neuronal survival following developmental alcohol exposure (Ahlers et al., 2015; Topper et al., 2015) and evaluated total population of non-neuronal cells.

METHODS

Rat model of FASD

All procedures performed in these experiments were carried out in accordance with the animal use protocol approved by University of Delaware (UD) Institutional Animal Care and Use Committee (IACUC). The experiments discussed in this study use two different doses of AE (i.e. 5.25 g/kg/day using 11.90% vol/vol ethanol or 3.00 g/kg/day using 6.79% vol/vol ethanol; delivered in two doses 2 hours apart) in a well-characterized rodent model of third trimester-equivalent binge-like AE (Goodlett and Johnson, 1997; Ryan et al., 2008; Gursky and Klintsova, 2017; Gursky et al., 2019) to determine the sensitivity of the ventral midline thalamic nucleus reuniens, a critical hub for of prefrontal-hippocampal coordination (Dolleman-van der Weel et al., 2019). Besides the inclusion of a moderate AE dose, the current study’s methods are identical to the cited studies and can be seen in summary in the timeline of Fig. 1A.

Figure 1.

Figure 1

Open in a new tab

Summary of neuroanatomical findings in the current study. (A) Timeline of experimental procedures. Black drop identifies treatment groups that underwent blood collection on PD4, at 90 minutes after the final alcohol administration for the day. Highlighted colors and symbols (on right) represent each treatment group throughout the figure, with #s in parentheses indicating total sample size for the experiment. (B) Representative 100 μm × 100 μm images from each treatment group, highlighting the reductions in neuron number in both AE groups while microglial number remains unchanged. Scale bars = 25 μm. (C) Both AE groups displayed a significant reduction in the total number of cells in Re (left) relative to both the SI procedural control group and the SC typically-developing control group. This included a reduction in neurons (center) and non-neurons (right). (D) AE-induced reduction of non-neurons (likely, glia) was not caused by reductions in microglia number (left), as neither AE group significantly differed from control groups. Therefore, reductions in non-neuronal cell number were driven by the number of ‘other’ cells (i.e. Hoechst+/Iba1−) (right). (E) Both AE groups displayed a reduction in total volume of Re relative to both the SI procedural control group and the SC typically-developing control group. (F) The total effect of AE on volume of Re (in μm3) is shown in parentheses. (G) Although reductions in Re cell number mediate the effect of AE on Re volume, there is no significant direct effect of AE on Re volume (i.e. effect of AE on Re volume that is not accounted for by reduction in Re cell #). Neither the mediation nor direct effect is moderated by dose of AE (indicated by P-values >0.050 over dashed grey arrows), indicating a similar outcome across doses. In all panels of this figure, black bars represent mean within that postnatal treatment with individual data points representing one animal. Data include both male and female animals, which are not differentiated due to the lack of sex effects. n.s.P > 0.050, *P ≤ 0.05, **P ≤ 0.01.

Long Evans rats were bred in-house at UD using first time mothers placed with experienced male breeders. Both male and female Long Evans rats were examined. All regression analyses used litter as a random-effect to account for between-litter variability (Galbraith et al., 2010) as has been used in past experiments of this nature (Gursky et al., 2019; Gursky et al., 2020). A total of 74 pups from 14 litters were examined (Table 1).

Table 1.

Sample sizes and mean stereological coefficients of error (CEs) for Re cell number estimates (commonly accepted as robust and reliable if CE values are below 0.100 (Slomianka and West, 2005)) for each sex and postnatal treatment group

Postnatal treatment Sex # of litters represented Total sample size Stereological CE (mean ± SEM)
SC Female 02 03 0.057 ± 0.009
SC Male 03 03 0.047 ± 0.003
SI Female 11 11 0.049 ± 0.002
SI Male 11 11 0.048 ± 0.003
AEModerate Female 08 08 0.053 ± 0.004
AEModerate Male 09 09 0.047 ± 0.002
AEHigh Female 10 15 0.057 ± 0.004
AEHigh Male 11 14 0.052 ± 0.002
Open in a new tab

SC: ‘suckle control’ (typically-developing control group); SI: ‘sham-intubated’ (procedural control group); AEHigh: ‘high-dose alcohol-exposed’ (5.25 g/kg/day) experimental treatment group, AEModerate: ‘moderate-dose alcohol-exposed’ (3.00 g/kg/day) experimental treatment group.

Although high-dose AE around 5.25 g/kg/day (causing peak BACs between 300 mg/dl and 500 mg/dl) is commonly used in rat models of FASD (Goodlett and Johnson, 1997; Ryan et al., 2008; Gursky and Klintsova, 2017; Gursky et al., 2019), we chose to use 3.00 g/kg/day AE as a moderate dose to obtain peak BACs between 100 mg/dl and 300 mg/dl, as such a relationship has been established and validated by multiple groups in both rats and mice (Bonthius and West, 1988; Ikonomidou et al., 2000; Ahlers et al., 2015). The current study includes both a procedural control group (i.e. sham intubated, ‘SI’) and a typically-developing (undisturbed, except for daily weighing) control group (i.e. suckle control, ‘SC’) to determine whether stress of intubation has detrimental impact on the neuroanatomical measures examined.

BAC analysis

Blood samples for BAC analysis were collected at exactly 90 minutes following the final alcohol dose on PD4, the time at which BAC is highest (Kelly et al., 1987). Peak BAC is a robust predictor of alcohol-induced damage to the central nervous system, more so than total amount of alcohol administered (Bonthius and West, 1990). Immediately after blood collection from pups, samples were centrifuged at 15,000 × g at 4°C for 25 minutes and supernatant plasma was collected and stored at −20°C until BAC analysis using an Analox GL5 Analyzer (Analox Instruments, Boston, MA, USA) with manufacturer’s protocol.

Peak mean ± SEM BAC achieved on PD4 for the AEModerate treatment group was 144.7 ± 17.9 mg/dl, whereas the AEHigh group achieved 323.3 ± 18.7 mg/dl, confirming a significant reduction in BAC of the moderate-dose group relative to the 5.25 g/kg/day dose typically used with this model (linear mixed-effects model, estimate = −190.999 ± 33.078, t(33.912) = −5.774, P = 1.71 × 10−6).

Tissue fixation and immunofluorescence

On PD72, each animal was weighed, then deeply anesthetized using a veterinarian-approved dose of ketamine and xylazine: 1 ml/kg of ketamine/xylazine cocktail consisting of 87 mg/ml ketamine and 13 mg/ml xylazine was delivered via i.p. injection. Brains were perfused transcardially with 0.1 M phosphate buffered saline (PBS; pH = 7.20), followed by 4% paraformaldehyde (PFA) in 0.1 M PBS, postfixed in 4% PFA for 48 hours, then equilibrated in 30% sucrose in 4% PFA three times. The forebrain was then exhaustively sectioned in a coronal plane at 40 μm using a cryostat (CM3050S; Leica Biosystems, Wetzlar, Germany), and serial sections were collected in a cryoprotectant solution (30% sucrose in 30% ethylene glycol solution in 0.1 M PBS) and stored at −20°C.

The current experiment examined the total number of cells in Re, as visualized by nuclear DNA stain Hoechst33342, as well as immunofluorescent (IF) labeling of neuronal nuclei and microglia using primary antibodies against NeuN and Iba1, respectively (Lind et al., 2005; Fernandez-Arjona et al., 2017). Every 16th serial coronal section was systematically selected (from a random starting point) and exposed to a first blocking solution (NDS-TX-TBS blocking solution) containing 3.0% normal donkey serum (NDS; Millipore Sigma, S30) and 0.5% Triton X-100 (ThermoFisher Scientific, Product #85111) in 0.1 M tris buffer solution (TBS; pH = 7.40) for 2 hours. Tissue was then transferred to the first primary antibody, Rabbit anti-Iba1 (Wako Chemicals, 019-19741, 1:2500 dilution) in NDS-TX-TBS blocking solution at 4°C for 40 hours. Negative-control sections were incubated in NDS-TX-TBS blocking solution without primary antibody. Tissue was then washed in 0.1 M TBS and transferred into the first secondary antibody (Donkey antirabbit conjugated with Alexa Fluor568, ThermoFisher Scientific, A10042, 1:250 dilution) in NDS-TX-TBS blocking solution for 3 hours at room temperature. This process was repeated with a second blocking solution (replacing 3.0% NDS with 3.0% normal goat serum: NGS; Millipore Sigma, S26), second primary antibody (Mouse anti-NeuN, Millipore Sigma, MAB377, 1:500 dilution) in NGS-TX-TBS blocking solution, and second secondary antibody (Goat antimouse conjugated with Alexa Fluor488, Jackson ImmunoResearch, 115-545-003, 1:250 dilution) in NGS-TX-TBS blocking solution. Lastly, sections were incubated in 0.4 μg/ml Hoechst33342 in PBS for 5 minutes before coverslipping using gelvatol mounting medium (Mitala et al., 2008) and allowed to set for at least 1 week prior to imaging. Representative image stacks acquired for analysis following IF labeling can be observed in Fig. 1B.

Fluorescence microscopy and unbiased stereological estimation

All microscopy utilized a Zeiss AxioImager M2 with Apotome and Colibri 7 LED illumination (Carl Zeiss AG, Oberkochen, Germany) and imaged by a high-sensitivity monochrome camera (ORCA-Flash4.0 LT+ Digital CMOS camera, Hamamatsu Corporation, Middlesex, NJ, USA).

A rigorous method of quantification for parameters such as number and volume is unbiased stereological estimation (Napper, 2018). Every 16th serial coronal section of Re was prepared and examined. Systematic random sampling of each section utilized 150 μm × 150 μm counting frame within a 400 μm × 400 μm grid. This resulted in an analysis of ~14% of the area of Re in each section. The thickness of the optical disector was 15.0 μm and guard zones of 2.0 μm, with section thickness measured at each counting site. The observed mean Gundersen coefficients of error (Gundersen and Jensen, 1987; Slomianka and West, 2005) resulting from the estimation of cell number in Re was automatically calculated by StereoInvestigator software and can be found in Table 1.

Statistical analyses

An α = 0.050 level of significance was used for all statistical analyses. All statistical analyses were performed using the R programming language version 4.0.2 (R Core Team, 2020) in the RStudio integrated development environment (RStudio Team, 2016) with the following packages: lme4 (Bates et al., 2015), lmerTest (Kuznetsova et al., 2017), mediation (Tingley et al., 2014), readxl (Wickham and Bryan, 2018), sjPlot (Lüdecke, 2020), and tidyverse (Wickham, 2017).

To account for potential source of variation within-litter (Galbraith et al., 2010), all analyses were performed as linear mixed-effects models using the ‘lmer’ function in ‘lmerTest’ package (except for BAC correlations) with litter included as a random-intercept term (except for mediation analyses due to lack of support for linear mixed-effects models for the ‘test.modmed’ function for moderated mediations in the ‘mediation’ package). Outliers were not removed from alcohol-exposed groups prior to analysis, as potential sources of biological variability that could cause outlier values (such as BAC) were important experimental questions in this study. Outliers were also not removed from the SI or SC control groups to ensure that data were treated similarly in the control groups versus the alcohol-exposed treatment groups. The observed powers for the mixed-effects models for outcomes were as follow: Total cell number = 0.896, total neuron number = 0.987, total non-neuronal cell number = 0.861, total microglial number = 0.911, total ‘other’ cell number = 0.841. These values are all above the conventionally accepted minimal threshold for power, 80% (Dumas-Mallet et al., 2017).

The procedural SI control group was considered the baseline reference group for postnatal treatment because the causal impact of alcohol can only be considered as a difference from the procedural control group (the confound of procedural stress precludes the interpretation of differences between AE groups and typically-developing groups as ‘alcohol-induced’). Nevertheless, we observed no differences between the SI and the typically-developing groups in any neuroanatomical measures. Females were considered the baseline reference group for biological sex. All simulation-based analyses (i.e. mediation and moderated mediation analyses) were scripted and utilized a randomly generated numerical seed, to allow for reproducible analysis. The data presented in this study are available upon reasonable request from the corresponding author.

RESULTS

For succinct reporting of findings, only key findings related to postnatal treatment will be discussed in this section. Comprehensive presentation of exact statistical values can be found in Tables 2 and 3.

Table 2.

Linear mixed-model regression table for total number of cells, neurons, or non-neuronal cells in Re. Table was generated using sjPlot function ‘tab_model’ (Lüdecke, 2020)

Total cell # (Hoechst+/NeuN+or-/Iba1+or-) Total neuron # (Hoechst+/NeuN+/Iba1−) Total non-neuronal cell # (Hoechst+/NeuN−/Iba1+or-)
Predictor Estimate 95% CI P df Estimate 95% CI P df Estimate 95% CI P df
(Intercept) 145,699 131,677–159,721 <0.001 64 52,988 46,152–59,823 <0.001 47 92,711 83,885–101,537 <0.001 66
SC −4900 −35,644 to –25,844 0.756 57 −5375 −20,854 to –10,104 0.500 41 −12 −19,080 to –19,055 0.999 66
AEModerate −27,538 −48,541 to −6534 0.013 57 −10,086 −19,105 to −1068 0.033 56 −16,960 −30,562 to −3358 0.017 66
AEHigh −29,324 −47,253 to −11,395 0.002 56 −14,512 −22,199 to −6825 <0.001 55 −14,700 −26,320 to −3079 0.016 66
Male 5724 −13,476 to –24,924 0.561 55 −5009 −13,202 to –3185 0.236 55 10,733 −1750 to –23,215 0.097 66
SC *Male 2966 −38,843 to –44,774 0.890 60 9213 −9086 to –27,512 0.328 61 −5760 −32,725 to –21,205 0.677 66
AEModerate *Male 4741 −24,438 to –33,921 0.751 56 4755 −7745 to –17,255 0.459 55 −250 −19,175 to –18,675 0.979 66
AEHigh *Male −2205 −27,771 to –23,361 0.866 57 6140 −4833 to –17,113 0.278 55 −8691 −25,249 to –7866 0.307 66
Random effects
σ2 527783569.41 96125008.35 223081004.40
τ00 35230136.18Litter 37679109.90Litter 0.00Litter
ICC 0.06 0.28
N 14Litter 14Litter 14Litter
Observations 74 74 74
Marginal R2, Conditional R2 0.262, 0.308 0.173, 0.406 0.290, NA
Open in a new tab

Random effect was litter (random-intercept) and df were calculated using Satterthwaite approximation. CI: confidence interval; σ2: residual variance of outcome variable; τ00: random-intercept (i.e. between-litter) variance. ICC: intraclass correlation coefficient; N: number of random-intercept groups (i.e. litters); Marginal R2: R2 associated with fixed-effects only; Conditional R2: R2 associated with both fixed-effects and random-effects.

Table 3.

Linear mixed-model regression table for total number of microglial cells, and ‘other’ (i.e. nonmicroglial glial cells) in Re. Table was generated using sjPlot function ‘tab_model’ (Lüdecke, 2020)

Total # of microglia (Hoechst+/NeuN−/Iba1+) Total # of ‘other’ cells (Hoechst+/NeuN−/Iba1−)
Predictor Estimate 95% CI P df Estimate 95% CI P df
(Intercept) 8502 7154–9850 <0.001 64 84,209 76,188–92,230 <0.001 64
SC 432 −2583 to −3447 0.779 64 −686 −18,014 to −16,641 0.938 64
AEModerate −1515 −3416 to –385 0.118 64 −15,298 −27,659 to −2937 0.015 64
AEHigh −1481 −3101 to –140 0.073 64 −13,212 −23,772 to −2652 0.014 64
Male 1246 −484 to –2976 0.158 64 9487 −1857 to –20,830 0.101 64
SC *Male −564 −4383 to –3255 0.772 64 −4954 −29,459 to –19,550 0.692 64
AEModerate *Male −224 −2861 to –2412 0.867 64 −259 −17,457 to –16,939 0.976 64
AEHigh *Male −1080 −3393 to –1232 0.360 64 −7561 −22,608 to –7486 0.325 64
Random-effects
σ2 4286761.78 184229014.56
τ00 914775.33Litter 0.00Litter
ICC 0.18
N 14Litter 14Litter
Observations 74 74
Marginal R2, Conditional R2 0.172, 0.317 0.281, NA
Open in a new tab

Random effect was litter (random-intercept) and df were calculated using Satterthwaite approximation. CI: confidence interval; σ2: residual variance of outcome variable; τ00: random-intercept (i.e. between-litter) variance; ICC: intraclass correlation coefficient; N: number of random-intercept groups (i.e. litters); Marginal R2: R2 associated with fixed-effects only; Conditional R2: R2 associated with both fixed-effects and random-effects.

Both moderate- and high-dose AE caused neuronal and non-neuronal loss in Re at similar levels, but the number of microglia was not affected by either dose

Total cell number (i.e. Hoechst+ cells) within Re was reduced in both the AEHigh and AEModerate treatment groups relative to both control groups (AEHighP = 0.002, AEModerateP = 0.013). This was caused by reductions in both total number of neurons (i.e. Hoechst+/NeuN+/Iba1− cells) in Re (AEHighP < 0.001, AEModerateP = 0.033) as well as reductions in the total number of non-neuronal cells (i.e. Hoechst+/NeuN−/Iba1+ or – cells). Within the population of non-neuronal cells, there was no influence of postnatal treatment on microglial cell number (i.e. Hoechst+/NeuN−/Iba1+ cells; AEHighP = 073, AEModerateP = 0.118). AE-induced reductions in non-neuronal cell number are caused by a decrease in the number of non-neuronal cells that are nonmicroglial (i.e. Hoechst+/NeuN−/Iba1− cells; AEHighP = 0.014, AEModerateP = 0.15). These data are presented in Fig. 1C and D.

Reduction in Re volume caused by AE was mediated by cell loss and was not moderated by dose of AE

As identified using a moderated mediation analysis (schematically represented in Fig. 1F and G), AE (regardless of dose) reduced volume of Re (total effect estimate = −1.05 × 108, 95% confidence interval [CI]: −1.97 × 108 to −1.13 × 107, P = 0.028). This total effect of AE on volume (Fig. 1E and F) was mediated by reductions in total cell number (mediated effect estimate = −7.46 × 107, 95%CI: −1.37 × 108 to −2.57 × 107, P < 0.001) rather than a direct effect of AE on Re volume (direct effect estimate = −2.99 × 107, 95%CI: −1.17 × 108 to 5.72 × 107, P = 0.501), indicating that AE-induced reductions in Re volume can be accounted for entirely by reduction in cell number, and are not caused by other influences of AE on the composition of Re (e.g. reduction in cell body size or amount of dendritic material in excess of that due to cell loss). Examination of alcohol dose as a moderating factor for this relationship indicated that neither the mediated effect nor direct effect of AE on Re volume differed between doses (respectively, estimate = 4.29 × 104, 95% CI: −8.05 × 107 to 7.98 × 107, P = 0.995 and estimate = 5.67 × 104, 95% CI: −1.23 × 108 to 1.22 × 108, P > 0.999), suggesting that this relationship between AE-induced Re cell loss and Re volume loss is consistent across both doses used in this experiment.

Peak BAC on PD4 did not correlate with any neuroanatomical outcomes, further supporting that both moderate- and high- doses of AE during the BGS have similar effects on Re

All neuroanatomical outcomes examined were additionally analyzed through correlation with peak BAC on PD4, to account for the possibility that variability among achieved BAC within alcohol dose may have obscured potential dose-related effects. Due to a highly non-normal distribution of BACs when treatment groups were pooled (Shapiro–Wilk normality test, W = 0.920, P = 0.004), correlations were performed as Spearman’s rank correlations. Peak BAC on PD4 did not correlate with total cell number (rho = −0.025, P = 0.869), total neuron number (rho = −0.121, P = 0.425), total non-neuronal cell number (rho = 0.023, P = 0.878), total microglia number (rho = 0.080, P = 0.600), total number of ‘other’ non-neuronal cells in Re (rho = 0.038, P = 0.803), or total volume of Re (rho = 0.016, P = 0.918). These data are presented in Fig. 2, and interpretation of statistics are identical when data are analyzed using Pearson correlations (all P’s ≥ 0.486).

Figure 2.

Figure 2

Open in a new tab

Peak measured BAC on PD4 did not correlate with any neuroanatomical outcomes. All correlations were performed as Spearman’s rho due to a highly non-normal distribution of BACs (Shapiro–Wilk normality test, W = 0.920, P = 0.004) when pooled between AEModerate (orange triangles, point-up, n = 17) and AEHigh (red triangles, point-down, n = 29) treatment groups. Line of best fit (with 95% confidence interval) is displayed in black over each graph (was not significant in all correlations). Interpretation of statistics are identical when analyzed using Pearson correlations (P’s > 0.485). For all panels in this figure, individual data points represent one animal and include both male and female animals (which are not differentiated due to the lack of sex effects).

DISCUSSION

Summary of key findings

The current study examined whether moderate-dose AE causes similar or dose-dependent profiles of thalamic nucleus Re damage relative to the commonly used high-dose AE paradigm in the field (Gursky and Klintsova, 2017). We observed identical patterns of thalamic damage (i.e. volume loss, neuron loss, nonmicroglial non-neuronal cell loss) following both high- and moderate-dose AE, suggesting that both doses of AE are sufficient to induce persistent neurodegeneration in Re. Outcomes did not correlate with peak BAC on PD4, supporting the interpretation of between-groups analyses. This finding substantially extends the known window of dose-related vulnerability of Re, as previous studies examining this nucleus following developmental alcohol exposure have exclusively used high-dose AE paradigms (Gursky et al., 2019; Gursky et al., 2020), and this was the first to examine Re in animals who achieved significantly and meaningfully lower peak BACs (close to 200 mg/dl).

There were no significant differences in any of the parameters between the typically-developing control group (i.e. ‘SC’) and the procedural control group (i.e. ‘SI’), suggesting that the observed changes in the AE groups were not caused by developmental stress, and is a consequence of developmental alcohol exposure alone. There was also no apparent influence of biological sex on any examined neuroanatomical measures, which is consistent with past examination of Re (Gursky et al., 2020).

Consistency of estimates of AE-induced Re neuron loss across studies

To date, there are only two published studies that explicitly examine Re following AE during the BGS. Our finding that high-dose AE caused significant Re neuron loss is consistent with the findings of our two previous studies (Gursky et al., 2019; Gursky et al., 2020). Estimated rates of neuron loss in these studies were ~21% and 52%, respectively. These previous estimates likely vary so drastically due to the relatively small sample sizes utilized in these studies (n = 5–8 females per postnatal treatment, or n = 2–4 per sex per postnatal treatment, respectively) or duration of AE (PD4–9 versus PD7 only), since the average peak BACs observed in these studies were similar (i.e. 379 ± 38 mg/dl and 333 ± 34 mg/dl, respectively). The current study used a meaningfully larger sample size (n = 8–15 per sex per experimental postnatal treatment group) to increase statistical power and sensitivity to both dose- and sex-effects, resulting in a more accurate estimation of the magnitude of neuron loss than previous studies. As a result, the current experiment observed a 28 ± 9% (mean ± SEM) reduction in Re neuron number in the AEHigh group relative to the SI group and a 20 ± 10% (mean ± SEM) reduction in the AEModerate group (also relative to the SI group), which we believe to be replicable and generalizable estimates due to the robust sample size, low stereological coefficients of error, and statistical methods utilized.

Potential explanations for similar magnitude of Re damage observed across all peak BACs achieved in the current experiment

The reduction in Re volume that we observed in both AE postnatal treatment groups was mediated by a reduction in total cell number, replicating Gursky et al. (2020) and suggesting that the mechanism of Re volume loss is likely similar. However, it is possible that neuron loss caused by moderate- and high-dose AE may have different underlying mechanisms despite similar observed outcomes. Although a substantial number of similarities were observed between the two doses (cell loss, neuron loss, specific patterns of non-neuronal cell loss, regional volume loss, lack of BAC correlations), it is critical for future experiments asserting this hypothesis to causally manipulate likely mechanisms such as apoptotic neurodegeneration (Gursky et al., 2020) in both doses (e.g. through coadministration of an apoptotic inhibitor with EtOH) to determine whether both outcomes derive from similar processes. Longitudinal examination of Re would further support or refute this hypothesis of similar mechanisms.

Another limitation in the current study was the use of only a single cross-sectional timepoint to analyze BAC and correlate with adult neuroanatomy. The collection of multiple blood samples may prove deleterious to development of neonatal pups, and the benefit of frequent BAC analysis is a consideration that should be balanced against the potential consequences of frequent blood collection, especially at early ages of postnatal development (Bonthius et al., 1988). Although peak BACs achieved from a given dose of alcohol tend to be consistent between PD4 and PD9, it is common for studies to observe a single day with higher peak BACs such as PD4 (Kelly et al., 1987) or PD6 (Bonthius et al., 1988). Therefore, the differential impact of high- verses moderate-dose AE on distinct days of the PD4–9 exposure paradigm poses an interesting future direction but is beyond the scope of the current experiment.

Lastly, the current study was limited in the number of glial markers examined. We did identify a significant reduction in cells that were neither neuronal nor microglial, raising a number of questions regarding the likely identity of these non-neuronal cells (astrocytes versus oligodendrocytes versus other cell types such as endothelial or surveilling peripheral immune cells), and the mechanism by which their number is altered. Future examination of Re should focus on potential candidate glial subtypes, including both mature and immature oligodendrocytes as they have been recently implicated in a number of behavioral and neurological alterations in FASD (Creeley et al., 2013; Newville et al., 2017; Wozniak et al., 2019), and astrocytes which likely play a role in aberrant neuronal plasticity (Paul and Medina, 2012) and neuroimmune alterations following developmental AE (Topper et al., 2015). It is possible that changes in glial profile may contribute to AE-induced neuron loss in Re, but testing this hypothesis would require prior identification of a target population. The prospect that such non-neuronal cells may involve vasculature or surveilling immune response also has promise, as AE during the developmental time period in the current study can contribute to microhemorrhage formation (Welch et al., 2016).

Contributor Information

Zachary H Gursky, Department of Psychological & Brain Sciences, University of Delaware, Newark, DE 19716, USA.

Anna Y Klintsova, Department of Psychological & Brain Sciences, University of Delaware, Newark, DE 19716, USA.

FUNDING

This work was supported by the National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism [R01 AA027269 to A.Y.K. and R21 AA026613 to A.Y.K.] and Sigma Xi, The Scientific Research Honor Society [Grants in Aid of Research (GIAR) program award to ZHG].

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

CONFLICTS OF INTEREST STATEMENT

The authors declare no real, potential or apparent conflicts of interest with respect to commercial entities. The authors have no significant financial or other relations (e.g. directorship, consultancy or speaker fee) with companies, trade associations, unions or groups (including civic associations and public interest groups) that may gain or lose financially from the results or conclusions in this study.

References

  1. Ahlers KE, Karacay B, Fuller L, et al. (2015) Transient activation of microglia following acute alcohol exposure in developing mouse neocortex is primarily driven by BAX-dependent neurodegeneration. Glia 63:1694–713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bates D, Mächler M, Bolker B, et al. (2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67:1–48. [Google Scholar]
  3. Berman RF, Hannigan JH. (2000) Effects of prenatal alcohol exposure on the hippocampus: spatial behavior, electrophysiology, and neuroanatomy. Hippocampus 10:94–110. [DOI] [PubMed] [Google Scholar]
  4. Bonthius DJ, West JR. (1988) Blood alcohol concentration and microencephaly: a dose-response study in the neonatal rat. Teratology 37:223–31. [DOI] [PubMed] [Google Scholar]
  5. Bonthius DJ, West JR. (1990) Alcohol-induced neuronal loss in developing rats: increased brain damage with binge exposure. Alcohol Clin Exp Res 14:107–18. [DOI] [PubMed] [Google Scholar]
  6. Bonthius DJ, Goodlett CR, West JR. (1988) Blood alcohol concentration and severity of microencephaly in neonatal rats depend on the pattern of alcohol administration. Alcohol 5:209–14. [DOI] [PubMed] [Google Scholar]
  7. Cassel JC, Pereira de Vasconcelos A, Loureiro M, et al. (2013) The reuniens and rhomboid nuclei: neuroanatomy, electrophysiological characteristics and behavioral implications. Prog Neurobiol 111:34–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Creeley CE, Dikranian KT, Johnson SA, et al. (2013) Alcohol-induced apoptosis of oligodendrocytes in the fetal macaque brain. Acta Neuropathol Commun 1:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dobbing J, Sands J. (1979) Comparative aspects of the brain growth spurt. Early Hum Dev 3:79–83. [DOI] [PubMed] [Google Scholar]
  10. Dolleman-van der Weel MJ, Griffin AL, Ito HT, et al. (2019) The nucleus reuniens of the thalamus sits at the nexus of a hippocampus and medial prefrontal cortex circuit enabling memory and behavior. Learn Mem 26:191–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dumas-Mallet E, Button KS, Boraud T, et al. (2017) Low statistical power in biomedical science: a review of three human research domains. R Soc Open Sci 4:160254. https://royalsocietypublishing.org/doi/10.1098/rsos.160254 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fernandez-Arjona MDM, Grondona JM, Granados-Duran P, et al. (2017) Microglia morphological categorization in a rat model of neuroinflammation by hierarchical cluster and principal components analysis. Front Cell Neurosci 11:235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Galbraith S, Daniel JA, Vissel B. (2010) A study of clustered data and approaches to its analysis. J Neurosci 30:10601–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Goodlett CR, Johnson TB. (1997) Neonatal binge ethanol exposure using intubation: timing and dose effects on place learning. Neurotoxicol Teratol 19:435–46. [DOI] [PubMed] [Google Scholar]
  15. Gundersen HJ, Jensen EB. (1987) The efficiency of systematic sampling in stereology and its prediction. J Microsc 147:229–63. [DOI] [PubMed] [Google Scholar]
  16. Gursky ZH, Klintsova AY. (2017) Wheel running and environmental complexity as a therapeutic intervention in an animal model of FASD. J Vis Exp 30:748–752. 10.1097/wnr.0000000000001270) [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gursky ZH, Savage LM, Klintsova AY. (2019) Nucleus reuniens of the midline thalamus of a rat is specifically damaged after early postnatal alcohol exposure. Neuroreport 30:748–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gursky ZH, Spillman EC, Klintsova AY. (2020) Single-day postnatal alcohol exposure induces apoptotic cell death and causes long-term neuron loss in rodent thalamic nucleus Reuniens. Neuroscience 435:124–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hamilton GF, Hernandez IJ, Krebs CP, et al. (2017) Neonatal alcohol exposure reduces number of Parvalbumin-positive interneurons in the medial prefrontal cortex and impairs passive avoidance acquisition in mice deficits not rescued from exercise. Neuroscience 352:52–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Heroux NA, Robinson-Drummer PA, Kawan M, et al. (2019) Neonatal ethanol exposure impairs long-term context memory formation and prefrontal immediate early gene expression in adolescent rats. Behav Brain Res 359:386–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Ikonomidou C, Bittigau P, Ishimaru MJ, et al. (2000) Ethanol-induced apoptotic neurodegeneration and fetal alcohol syndrome. Science 287:1056–60. [DOI] [PubMed] [Google Scholar]
  22. Kelly SJ, Bonthius DJ, West JR. (1987) Developmental changes in alcohol pharmacokinetics in rats. Alcohol Clin Exp Res 11:281–6. [DOI] [PubMed] [Google Scholar]
  23. Khoury JE, Milligan K, Girard TA. (2015) Executive functioning in children and adolescents prenatally exposed to alcohol: a meta-analytic review. Neuropsychol Rev 25:149–70. [DOI] [PubMed] [Google Scholar]
  24. Kuznetsova A, Brockhoff PB, Christensen RHB. (2017) lmerTest package: tests in linear mixed effects models. J Stat Softw 82:1–26. https://www.jstatsoft.org/article/view/v082i13) [Google Scholar]
  25. Lind D, Franken S, Kappler J, et al. (2005) Characterization of the neuronal marker NeuN as a multiply phosphorylated antigen with discrete subcellular localization. J Neurosci Res 79:295–302. [DOI] [PubMed] [Google Scholar]
  26. Lüdecke D. (2020) sjPlot: data visualization for statistics in social science. https://cran.r-project.org/web/packages/sjPlot/index.html
  27. Mitala CM, Wang Y, Borland LM, et al. (2008) Impact of microdialysis probes on vasculature and dopamine in the rat striatum: a combined fluorescence and voltammetric study. J Neurosci Methods 174:177–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Napper RMA. (2018) Total number is important: using the Disector method in design-based stereology to understand the structure of the rodent brain. Front Neuroanat 12:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Newville J, Valenzuela CF, Li L, et al. (2017) Acute oligodendrocyte loss with persistent white matter injury in a third trimester equivalent mouse model of fetal alcohol spectrum disorder. Glia 65:1317–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Otero NK, Thomas JD, Saski CA, et al. (2012) Choline supplementation and DNA methylation in the hippocampus and prefrontal cortex of rats exposed to alcohol during development. Alcohol Clin Exp Res 36:1701–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Paul AP, Medina AE. (2012) Overexpression of serum response factor in astrocytes improves neuronal plasticity in a model of early alcohol exposure. Neuroscience 221:193–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Popova S, Lange S, Probst C, et al. (2017) Estimation of national, regional, and global prevalence of alcohol use during pregnancy and fetal alcohol syndrome: a systematic review and meta-analysis. Lancet Glob Health 5:e290–9. [DOI] [PubMed] [Google Scholar]
  33. R Core Team . (2020) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.r-project.org/ [Google Scholar]
  34. Rstudio Team . (2016) RStudio: Integrated Development Environment for R, 1.1.414 edn. Boston, MA. https://www.rstudio.com/ [Google Scholar]
  35. Ryan SH, Williams JK, Thomas JD. (2008) Choline supplementation attenuates learning deficits associated with neonatal alcohol exposure in the rat: effects of varying the timing of choline administration. Brain Res 1237:91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Slomianka L, West MJ. (2005) Estimators of the precision of stereological estimates: an example based on the CA1 pyramidal cell layer of rats. Neuroscience 136:757–67. [DOI] [PubMed] [Google Scholar]
  37. Tingley D, Yamamoto T, Hirose K, et al. (2014) Mediation: R package for causal mediation analysis. J Stat Softw 59:1–38.26917999 [Google Scholar]
  38. Topper LA, Baculis BC, Valenzuela CF. (2015) Exposure of neonatal rats to alcohol has differential effects on neuroinflammation and neuronal survival in the cerebellum and hippocampus. J Neuroinflammation 12:160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Umer A, Lilly C, Hamilton C, et al. (2020) Prevalence of alcohol use in late pregnancy. Pediatr Res 88:312–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Welch JH, Mayfield JJ, Leibowitz AL, et al. (2016) Third trimester-equivalent ethanol exposure causes micro-hemorrhages in the rat brain. Neuroscience 324:107–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wickham H. (2017) Tidyverse: easily install and load the 'Tidyverse'. https://cran.r-project.org/web/packages/tidyverse/index.html
  42. Wickham H, Bryan J. (2018) Readxl: read excel files. https://cran.r-project.org/web/packages/readxl/
  43. Wozniak JR, Riley EP, Charness ME. (2019) Clinical presentation, diagnosis, and management of fetal alcohol spectrum disorder. Lancet Neurol 18:760–70. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.

Articles from Alcohol and Alcoholism (Oxford, Oxfordshire) are provided here courtesy of Oxford University Press

RESOURCES