Prenatal alcohol exposure enhances the susceptibility to NMDA‐induced generalized tonic‐clonic seizures in developing rats

Summary

Aims

Prenatal alcohol exposure (PAE) is associated with a higher likelihood of developing generalized tonic‐clonic seizures (GTCS) in infants and children. However, experimental studies of PAE‐related seizures have yielded conflicting results. Here, we investigated the effect of acute PAE on N‐methyl‐D‐aspartate (NMDA)‐induced seizures in developing rats.

Methods

Pregnant Sprague Dawley rats were given an oral dose of either ethanol (5 g/kg body weight) or vehicle on embryonic day 18. The offspring were tested for susceptibility to NMDA‐induced seizures on postnatal day 7 (P7), 21 (P21), 35 (P35), and 42 (P42). Specifically, the prevalence and latency of NMDA‐induced continuous wild running‐like behaviors (CWR), flexion seizures (FS), wild running seizures (WRS), GTCS, and tonic seizures (TS) were recorded and analyzed.

Results

N‐methyl‐D‐aspartate‐induced seizures consisted of CWR, FS, GTCS, and TS in <P21 rats, while WRS, GTCS, and TS were observed in >P21 rats. Thus, GTCS were consistently observed during development. PAE significantly increases the prevalence of GTCS in female and male P7‐P21 rats and P7‐P35 rats, respectively, but not in older rats. PAE also increases the prevalence of TS in male, but not female P21‐P35 rats.

Conclusions

The PAE animal model of GTCS may provide a new opportunity to investigate the mechanisms that underlie neuronal hyperexcitability in developing animals prenatally‐exposed to alcohol.

1. INTRODUCTION

When a pregnant woman binge drinks alcohol, the developing fetus experiences high blood alcohol levels (BAL). To the developing child, this prenatal binge‐like alcohol exposure (PAE) can have a variety of teratogenic outcomes collectively known as fetal alcohol spectrum disorder (FASD); even a single PAE episode can produce FASD.1, 2 Seizure is a common symptom of FASD; children who are exposed in utero to alcohol are up to 20 times more likely to develop seizures compared to the general population.3, 4 GTCS are the most prevalent seizure type among patients with FASD3, 4, 5, 6; these seizures can present as early as 6 month of age—the equivalent of postnatal day 7 (P7) in rats.4, 7 Thus, alcohol exposure in utero is a likely factor that can cause both structural and functional brain abnormalities, leading to increased GTCS susceptibility. In contrast to the high prevalence of GTCS among the FASD population, studies using various PAE models have yielded conflicting results. For example, a single exposure to alcohol on postnatal day 4 (P4) or repeated exposures to alcohol (from P4 to P9) reduces the threshold for pentylenetetrazol (PTZ)‐induced seizures in juvenile rats, but not adult rats.8 Furthermore, semi‐chronic PAE (from E17 to E19) enhances the prevalence of PTZ‐induced focal seizures in P15‐P25 offspring rats.9 In contrast, chronic PAE exposure (beginning on E8 and continuing through delivery) delays the onset and decreases the incidence of PTZ‐induced seizures,10, 11 as well as the occurrence of amygdala kindling‐induced limbic (but not motor) seizures.12, 13 Chronic PAE exposure (from E8 through parturition) also failed to induce acoustically evoked GTCS susceptibility in P18‐P23 rats; nevertheless, a few of these animals presented GTCS on the retest conducted 5 days later.14 Thus, there is currently a need for PAE‐related GTCS models in developing rats that mimic the human counterpart, to clarify the consequence of PAE on the development of seizure susceptibility. Interestingly, systemic administration of N‐methyl‐D‐aspartate (NMDA) induced GTCS in developing and adult rats.15, 16 Here, we used a rat NMDA‐induced seizures model to examine the effects of a single PAE episode on the susceptibility to develop GTCS in offspring rats.

2. MATERIALS AND METHODS

2.1. Animals

Timed‐pregnant Sprague Dawley rats (dams) were purchased from Taconic Farms (Germantown, NY, USA) and housed individually in standard polycarbonate cages with standard chow and water available ad libitum. The rats were housed in a temperature‐ and humidity‐controlled room with a 12/12‐hour light/dark cycle. From the time of birth (defined as P0), the pups remained with their mother until weaning at P21. Both female and male P7‐P42 rats were used for seizure testing. All experimental procedures were approved by Georgetown University Animal Care and Use Committee and were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals.17

2.2. Alcohol administration

On E18, dams were randomly assigned to receive either a single dose of ethanol (1, 2.5, 5 g/kg body weight) or control vehicle by gavage; these doses were chosen based on a previous study.18 Ethyl alcohol (95%) was freshly prepared as 30% (vol/vol) ethanol mixed with Isomil‐milk‐based infant formula (Abbott Laboratories, Abbott Park, IL, USA). Dams in the control group received isovolumetric and isocaloric solution of Isomil without ethanol. Another group of dams received a semi‐chronic treatment, once daily from E18 to E20, of either the vehicle or ethanol (1 or 2.5 g/kg body weight).

2.3. Measurements of blood alcohol levels

In a separate group of ethanol‐treated animals, blood alcohol levels (BAL) were measured on E18 from both dams (n = 6) and their fetuses (n = 24). Two hours after administering the ethanol dose, blood samples (ranging from 0.1‐1 mL in volume) were extracted by intracardiac puncture; the two‐hour interval was chosen based on a previous study.19 The blood samples were then centrifuged for 10 minutes at 1000 × g, and the plasma was isolated and stored at −80°C. Alcohol levels were measured using an Analox model GM7 or AM1 analyzer (Analox Instruments, London, UK).

2.4. Offspring seizure testing

On P7, P21, P35, and P42, NMDA (Sigma Chemicals, St. Louis, MO, USA) was dissolved in 0.9% NaCl and injected intraperitoneally at a dose of 5, 75, 120, and 175 mg/kg body weight, respectively. The doses of NMDA were determined based on published studies and on preliminary tests; these doses were the lowest doses that were able to induce seizures in less than half of tested animals.15 Two to four P7, P21, P35, and P42 rats per litter were used. The rats were then placed in clear Plexiglas boxes and monitored for 60 minutes for the occurrence of seizures. To compensate for the weak thermoregulation of P7 rats, the body temperatures of pups were maintained at 33 ± 2°C using a heating pad and a heat therapy pump (model TP700; Stryker Medical, Portage, MI, USA). In P7‐P21 rats, NMDA induced sequential behaviors consisting of stereotypy and seizures. Stereotypy included sigmoidal tail movement (or tail twisting) and caudal biting. Seizures consisted of continuous wild running‐like behaviors (CWR), flexion seizures (FS, characterized by tonic hyperflexion of the head, neck, spine, and tail), wild running seizures (WRS), and GTCS (bouncing tonic‐clonic seizures experienced while lying on the belly or side) and tonic seizures (TS, characterized by forelimb extension) (Table 1). In P7, convulsive responses (or seizure severity) were classified as follows: stage 0, no response; stage 1, CWR; stage 2, FS; stage 3, GTCS; and stage 4, TS.15 In P21, convulsive responses were classified as follows: stage 0, no response; stage 1, WRS; stage 2, FS; stage 3, GTCS; and stage 4, TS.15 In >P21 rats, NMDA‐induced seizures consisted of WRS, GTCS, and TS 15, 16 and the seizure severity was classified as follows: stage 0, no response; stage 1, WRS (1 episode); stage 2, WRS (>2 episodes); stage 3, WRS (1 episode) and GTCS; stage 4, WRS (>2 episodes) and GTCS; stage 5, WRS (1 episode), GTCS, and TS; and stage 6, WRS (>2 episodes), GTCS, and TS. We used the acute NMDA‐induced seizure model because seizures induced by this method develop gradually over a period of approximately 3‐20 minutes (in contrast, seizures induced by electrical stimulation have an abrupt onset, precluding our ability to measure latency). Following NMDA injections, animals that did not develop seizures within 60 minutes were considered to be nonresponders. For each animal, the incidences of CWR, FS, WRS, GTCS, and TS were recorded. In some cases (animals with GTCS), NMDA administration was lethal. The time interval between NMDA injection and the onset of CWR or the first FS, WRS, GTCS, or TS episode was recorded and defined as the CWR, FS, WRS, GTCS, and TS latency, respectively. For each animal, the seizure severity score was also recorded.

Table 1.

N‐methyl‐D‐aspartate (NMDA)‐induced convulsive behaviors in developing rats

Age	P7	P21	P35	P42
NMDA dose (mg/kg)	5	75	120	175
Convulsive behaviors
Continuous wild running‐like behaviors	+	–	–	–
Flexion seizures	+	+	–	–
Wild running seizures	–	+	+	+
Generalized tonic‐clonic seizures	+	+	+	+
Tonic seizures	+	+	+	+

(+): Present; (−): absent.

2.5. Rectal temperature and blood oxygen saturation

Given the importance of fetal hypoxia and intrapartum maternal fever in the pathophysiology of neonatal seizures, we evaluated the effects of ethanol administration on body temperatures and blood oxygen saturation (SpO₂) in dams.20, 21 Both rectal temperature and blood SpO2 were measured in freely moving dams using a Physitemp TCAT‐2LV controller (Physitemp Instrument INC, Clifton NJ, USA) and a MouseOx pulse oximeter system (Starr Life Sciences Corp, Oakmont, PA, USA), respectively. Core body temperatures and blood SpO₂ were recorded 30 minutes before ethanol administration (time zero corresponds to time before administering ethanol), and at 30 and 60 minutes after ethanol administration. The effects of ethanol on rectal temperatures and SpO₂ were expressed as the difference between rectal temperature or SpO₂ measured at time zero (baseline) and at 30 and 60 minutes after ethanol administration.

2.6. Statistical analyses

Data for both rectal temperatures and blood SpO₂ were assessed using repeated measurements ANOVA (and Bonferroni correction, if necessary) with time as the repeated factor and treatment as the between‐group factor. For postpartum studies, a litter was used as random factor for data analysis. The between‐group prevalence of NMDA‐induced CWR, FS, WRS, GTCS, and TS were analyzed using Chi‐square test. The between‐group BAL and seizure latency were analyzed using one‐way analysis of variance (ANOVA) and two‐way ANOVA (followed by post hoc comparison), respectively. Before performing ANOVA, data were subjected to the Shapiro‐Wilk test for normality and Levene's test for homogeneity of variances. Seizure severity scores were compared using the Mann‐Whitney rank test. Data are presented as the percentage (%) for the incidence of CWR, FS WRS, GTCS, and TS; mean ± SEM for BAL and seizure latency; and median score ± mean absolute deviation for seizure severity. Differences were considered significant at P < .05.

3. RESULTS

3.1. Effects of various prenatal alcohol exposure doses on the occurrence of NMDA‐induced GTCS in postpartum P7 rats

In the control‐treated group, the prevalence of GTCS was 27.77% in P7 male pups (Figure 1A). Of the three single oral doses of ethanol (1.0, 2.5, or 5.0 g/kg) administered to pregnant rats at E18, only doses of 2.5 and 5 g/kg of ethanol significantly increased GTCS susceptibility to 50% (χ² = 9.27, df = 1, P = .002) and 67% (χ² = 28.95, df = 1, P = .0001) in male P7 pups compared to control‐treated P7 pups, respectively (Figure 1A). PAE nonsignificantly altered GTCS latency (ethanol‐treated group: 1 mg/kg: 30.62 ± 5.26 minutes, n = 4; 2.5 mg/kg: 26.5 ± 1.71 minutes, n = 4, 5 mg/kg: 20.0 ± 1.78 minutes, n = 6) compared to the control‐treated group (26.8 ± 1.1 minutes, n = 6). Similarly, single ethanol treatment 2.5 and 5 g/kg also significantly increased the prevalence of GTCS to 45% (χ² = 6.27, df = 1, P = .012) and 56% (χ² = 16.15, df = 1, P = .0001) in female P7 pups from 26.66% in the control‐treated group (Figure 1B). No significant change in GTCS latency was observed between the control‐treated group (27.9 ± 3.2 minutes, n = 6) and the ethanol‐treated group (1 mg/kg: 38.5 ± 5.0 minutes, n = 4; 2.5 mg/kg: 31.7 ± 5.3 minutes, n = 4, 5 mg/kg: 24.0 ± 3.0 minutes, n = 6).

Figure 1.

Effects of PAE on GTCS prevalence, rectal temperature, and blood oxygen saturation. Only 2.5 and 5 g/kg ethanol increased NMDA (5 g/kg, i.p.)‐induced GTCS susceptibility in both male (A) and female (B) P7 pups. Semi‐chronic 1 and 2.5 g/kg ethanol also increased NMDA‐induced GTCS susceptibility in both male (C) and female (D) P7 pups. A single oral dose of ethanol (5 g/kg body weight) in pregnant rats at E18 decreased core body temperature (E) but did not affect blood oxygen saturation (F). For GTCS prevalence, data are shown as percentage ± standard error of proportion, while data for rectal temperature and blood oxygen saturation are shown as the mean ± SEM. Analysis was performed on litters; the number of litters in each group is indicated within the histogram and the total number of animals is reported in parentheses. Chi‐square test was used to compare seizure incidence. Repeated measure ANOVA was used to compare change in rectal temperature and blood oxygen saturation. *P < .05, **P < .01, ***P < .001, ****P < .0001

Semi‐chronic ethanol treatment (E18‐E21) at a dose of 1 and 2.5 mg/kg increased the prevalence of GTCS in male P7 pups to 47% (χ² = 4.33, df = 1, P = .037) and 67% (χ² = 28.95, df = 1, P = .0001) compared to the control‐treated group, respectively (Figure 1C). This ethanol treatment nonsignificantly altered GTCS latency (ethanol‐treated group: 1 mg/kg: 26.50 ± 3.20 minutes, n = 4; 2.5 mg/kg: 22.9 ± 2.31 minutes) compared to the control‐treated group (26.1 ± 1.5 minutes, n = 6). Likewise, semi‐chronic ethanol treatment at the dose of 1 and 2.5 g/kg also significantly increased the prevalence of GTCS in female P7 pups to 42% (χ² = 4.34, df = 1, P = .037) and 50% (χ² = 10.22, df = 1, P = .001), respectively (Figure 1D). No significant change in GTCS latency was observed between the control‐treated group (26.9 ± 3.0 minutes, n = 6) and the ethanol‐treated group (1 mg/kg: 33.5 ± 2.5 minutes, n = 4; 2.5 mg/kg: 24.3 ± 1.9 minutes). Interestingly, the increased prevalence of GTCS was similar following a semi‐chronic ethanol treatment at a dose of 2.5 g/kg or a single dose of 5.0 g/kg (compare Figure 1A‐C and Figure 1B‐D). Thus, based on the highest prevalence of GTCS and to avoid a “kindling” effect due to repetitive episodes of binge alcohol exposure, a single dose of 5.0 g/kg ethanol was chosen for subsequent experiments.

Next, we evaluated the extent to which a single episode of PAE (at the dose of 5 g/kg, p.o.) at E18 alters core body temperatures, blood SpO₂, and BAL. Quantification shows that PAE led to a transient hypothermia in dams (test between‐subjects, F _1,11 = 24.19, P = .0004; Figure 1E), but did not alter the blood SpO₂ (Figure 1F). This hypothermia had no effect on the occurrence of NMDA‐induced seizures in P7 postpartum rat (data not shown). PAE treatment also caused a significant increase in BAL in both the dams (controls: 0.007 ± 0.002, n = 6; PAE: 0.26 ± 0.02, n = 6; F _1,46 = 59.72, P = .0001) and fetuses (controls: 0.007 ± 0.003, n = 24; PAE: 0.24 ± 0.03, n = 24, F _1,10 = 470.58, P < .001). PAE at E18 had no effect on the length of the gestation (control group: 22 ± 0.2 days, n = 44 litters; PAE: 22 ± 0.2 days, n = 44 litters), the number of pups per litter (controls: 13 ± 1 pups, n = 44 litters; PAE: 12 ± 1 pups, n = 44 litters), and the gender ratio of male/female pups in a litter (control: 1.3; PAE: 1.5).

3.2. Effects of PAE on NMDA‐induced seizures in postpartum P7 rats

On P7, NMDA‐induced seizures consisted of CWR, FS, GTCS, and TS. The prevalence of NMDA‐induced CWR in the control‐treated female and male pups was 57.5% (n = 5) and 65% (n = 5), respectively (Figure 2A). Among the female and male control‐treated pups that developed CWR, the onset of CWR was 8.15 ± 0.72 minutes (n = 5) and 9.93 ± 1.07 minutes (n = 5), correspondingly (Figure 2B). The prevalence of NMDA‐induced FS in the control‐treated female and male pups was 76% (n = 5) and 90% (n = 5), respectively (Figure 2C). Among the female and male control‐treated pups that developed seizures, the mean latency to the first FS episode was 13.8 ± 0.6 minutes (n = 5) and 13.6 ± 0.7 minutes (n = 5), correspondingly (Figure 2D). The number of FS was 6.7 ± 0.5 (n = 5) and 6.8 ± 0.6 (n = 5) in control‐treated female and male pups, respectively. The prevalence of NMDA‐induced GTCS in the control‐treated female and male P7 pups was 29% (n = 5) and 25% (n = 5), respectively (Figure 2E). Among the control‐treated P7 pups that developed seizures, the mean latency to GTCS was 27.9 ± 3.5 minutes (n = 5) and 26.8 ± 1.4 minutes (n = 5) and in female and male pups, correspondingly (Figure 2F). The prevalence of NMDA‐induced TS in the control‐treated female and male P7 pups was 18% (n = 5) and 15% (n = 5), respectively (Figure 2G). The latency to TS was 38.9 ± 5.45 minutes (n = 5) and 36.8 ± 3.54 minutes (n = 5) in female and male pups, respectively (Figure 2H). The median seizure severity score was 2.0 ± 0.16 (n = 5) and 2.0 ± 0.36 (n = 5) in control‐treated male and female P7 pups, respectively.

Figure 2.

Effects of PAE on NMDA‐induced seizures in postnatal P7 rats. NMDA (5 g/kg, i.p.)‐induced convulsive behaviors consisted of CWR, FS, GTCS, and TS in postpartum P7 pups. A single dose of PAE (5 g/kg, p.o.) at E18 did not affect the prevalence of CWR (A) and FS (C) or the onset of CWR (B) and FS (D). PAE increased the prevalence of GTCS in both females and males (E), but nonsignificantly decreased GTCS latency (F). PAE did not alter the prevalence of TS (G) and the TS latency (H). Data are shown as the percentage ± standard error of proportion of CWR, FS, GTCS, and TS, as well as mean ± SEM for seizure latency. Analysis was performed on litters; the number of litters in each group is indicated within (or above) the histogram and the total number of animals is reported in parentheses. Chi‐square test was used to compare seizure incidence, whereas two‐way ANOVA was used to compare seizure latency. ***P < .001, ****P < .0001

A single episode of ethanol exposure at E18 had no effect on the prevalence of CWR, FS, and TS in female and male P7 pups compared to control‐treated groups (Figure). PAE also did not alter the latency to develop CWR, FS, or TS in females and males compared to control‐treated groups (Figure 2B,D,H). PAE nonsignificantly increased the number of FS at P7 in males (9.8 ± 1.6, n = 5 compared to 6.8 ± 0.6, n = 5 in the control‐treated group) and females (6.7 ± 0.5, n = 4 compared to 6.8 ± 0.5, n = 5 in the control‐treated group). However, PAE significantly increased the prevalence of GTCS at P7 in females to 50% (compared to 29.41% in the control‐treated group, χ² = 8.34, df = 1, P = .004; Figure 2E) and in males to 68.42% (compared to 25% in the control‐treated group, χ² = 27.31, df = 1, P = .0001; Figure 2E); the prevalence of GTCS was significantly increased in PAE males compared to PAE females (χ² = 5.97, df = 1, P = .015). This increased prevalence of GTCS was not associated with change in the seizure latency in either female or male PAE‐treated groups (Figure 2F). Finally, PAE significantly increased the median score of seizure severity at P7 in males (3.0 ± 0.4, n = 5, vs the control group, Z = −2.2, P = .02) and in females (3.0 ± 0.44, n = 4 compared to the control group, Z = −2.06, P = .04).

3.3. Effects of PAE on NMDA‐induced seizures in postpartum P21 rats

On P21, NMDA‐induced seizures consisted of FS, WRS, GTCS, and TS. The prevalence of NMDA‐induced FS, WRS, GTCS, and TS in the control‐treated females was 75%, 33%, 33%, and 25%, respectively (Figure 3). Similarly, the prevalence of NMDA‐induced FS, WRS, GTCS, and TS in the control‐treated males was 35%, 19%, 29%, and 18%, respectively (Figure 3). Among the female control‐treated rats that developed seizures, the latency to the first FS, WRS, GTCS, and TS episode was 17.65 ± 4.49 minutes, 14.0 ± 3.05, 20.33 ± 4.23 minutes, and 23.00 ± 4.16 minutes, respectively (Figure 3); the seizure severity score was 1.0 ± 0.33 (n = 3). In male control‐treated rats, the latency to FS, WRS, GTCS, and TS episode was 18 ± 3.6 minutes, 22.3 ± 1.45, 25.33 ± 1.53 minutes, and 24 ± 1.5 minutes, respectively (Figure 3); the seizure severity score was 1.0 ± 1.0 (n = 4).

Figure 3.

Effect of PAE on NMDA‐induced seizures in postnatal P21 rats. NMDA (75 g/kg, i.p.)‐induced convulsive behaviors characterized by FS, WRS, GTCS, and TS in postpartum P21 rats. A single dose of PAE at E18 did not alter the susceptibility to develop FS (A) or the FS latency (B). PAE significantly increased the prevalence of WRS in both female and male P21 rats (C), but did not alter the WRS latency (D). Similarly, PAE significantly increased the susceptibility to develop GTCS (E), but nonsignificantly decreased the GTCS latency (F). Finally, PAE significantly increased the prevalence of TS in male but not in female P21 rats (G); PAE did not alter the onset of TS (H). Data are shown as the percentage ± standard error of proportion of FS, WRS, GTCS, and TS, as well as mean ± SEM for seizure latency. Analysis was performed on litters; the number of litters in each group is indicated within (or above) the histogram and the total number of animals is reported in parentheses. Chi‐square test was used to compare seizure incidence and two‐way ANOVA for seizure latency. *P < .05, **P < .01

Prenatal alcohol exposure at E18 had no effect on the prevalence of FS and TS at P21 in females compared to the control‐treated group (Figure 3A,G). However, PAE significantly increased the prevalence of WRS to 72% (χ² = 28.95, df = 1, P = .0001) and of GTCS to 55% (χ² = 8.95, df = 1, P = .001) compared to the control‐treated group (Figure 3C,E). Prenatal alcohol exposure did not altered latency to FS, WRS, GTCS, and TS (Figure 3). PAE also nonsignificantly altered the median score of seizure severity (3.0 ± 0.44, n = 3, compared to the control‐treated group, Z = −1.78, P = .07). In postpartum P21 males, PAE had no effect on FS prevalence or latency compared to the control‐treated group (Figure 3A,B). However, PAE significantly increased the prevalence of WRS to 31% (χ² = 4.39, df = 1, P = .05) but did not alter WRS latency compared to the control‐treated group (Figure 3C,D). PAE also increased the prevalence of GTCS to 55% (χ² = 12.83, df = 1, P = .0001) and of TS to 35% (χ² = 4.83, df = 1, P = .01), but did not alter the latency to GTCS and TS latency, compared to the control‐treated group (Figure 3E,F). Finally, PAE significantly increased the seizure severity score (to 3.0 ± 0.48, n = 5, compared to the control‐treated group, Z = −1.96, P = .05).

3.4. Effects of PAE on NMDA‐induced seizures in male postpartum P35 rats

On P35, NMDA‐induced seizures consisted of WRS that evolved into GTCS and TS; no FS were observed. In control‐treated females, the prevalence of WRS, GTCS, and TS was 42%, 42%, and 25%, respectively (Figure 4A,C,E). Among rats that developed seizures, the latency to WRS, GTCS, and TS was 14 ± 3 minutes (n = 3), 16 ± 3.35 minutes (n = 3), and 22.5 ± 2.5 minutes (n = 3), respectively (Figure 4B,D,F); the seizure severity score was 2.5 ± 1.11 (n = 3). In control‐treated males, the prevalence of WRS, GTCS, and TS was 50%, 37.5%, and 25%, respectively (Figure 4A,C,E). The latency to the first WRS, GTCS, and TS episode was 11.11 ± 0.67 minutes (n = 4), 13.44 ± 2.92 minutes (n = 4), and 18 ± 1.55 minutes (n = 4), respectively (Figure 4B,D,F). The seizure severity score was 2.5.0 ± 1.12 (n = 4).

Figure 4.

Effects of PAE on NMDA‐induced seizures in postnatal P35 rats. NMDA (120 g/kg, i.p.)‐induced convulsive behaviors characterized by WRS, GTCS, and TS in postpartum P35 rats. A single dose of PAE at E18 did not alter the prevalence of WRS (A) or the onset of WRS (B) in P35 rats. PAE, however, significantly increased the susceptibility to develop NMDA‐GTCS in male but not in female P35 rats (C) and did not alter the GTCS latency (D). Similarly, PAE significantly increased the susceptibility to develop TS in male but not in female P35 rats (E) and did not alter the onset of TS (F). Data are shown as the percentage ± standard error of proportion of WR, GTCS, and TS, as well as mean ± SEM for seizure latency. Analysis was performed on litters; the number of litters in each group is indicated within the histogram and the total number of animals is reported in parentheses. Chi‐square test was used to compare seizure incidence and two‐way ANOVA for seizure latency. *P < .05

PAE at E18 nonsignificantly altered the prevalence of WRS and TS in both female and male P35 rats (Figure 4A,E). However, PAE significantly increased the prevalence of GTCS to 53% (χ² = 5.27, df = 1, P = .02) and TS to 40% (χ² = 4.67, df = 1, P = .03) in males (but not in females) compared to the control‐treated group (Figure 4C,E). Similarly, PAE significantly increased the seizure severity score in males (to 4.5 ± 0.62, n = 4, Z = −2.04, P = .04) but not in females (3.75 ± 1.87, n = 4, Z = 0.75, P = .21). Figure 4 also shows that PAE did not alter the latency to WRS, GTCS, and TS in females and males compared to the control‐treated group.

3.5. Effects of PAE on NMDA‐induced seizures in postpartum P42 rats

On P42, NMDA‐induced seizures were similar to P35, consisting of WRS that evolved into GTCS and TS. In control‐treated female rats, the prevalence of WRS, GTCS, and TS was 56%, 49.33%, and 34.33% respectively (Figure 5A,C,E). The latency to WRS, GTCS, and TS was 11.22 ± 2.15 minutes (n = 3), 13 ± 2.89 minutes (n = 3), and 18.75 ± 2.25 minutes (n = 3), respectively (Figure 5B,D,F); the seizure severity median score was 2 ± 1.22 (n = 3). The prevalence of WRS, GTCS, and TS was 50%, 45%, and 30%, respectively, in the control‐treated male rats (Figure 5A,C,E). The latency to the first episode of WRS, GTCS, and TS was 11.5 ± 1.5 minutes (n = 5), 12.61 ± 1.69 minutes (n = 5), 13.67 ± 1.2 minutes (n = 5), respectively (Figure 5B,D,F); the seizure severity median score was 3 ± 0.36 (n = 5).

Figure 5.

Effects of PAE on NMDA‐induced seizures in postnatal P42 rats. NMDA (175 g/kg, i.p.)‐induced convulsive behaviors characterized by WRS, GTCS, and TS in postpartum P42 rats. A single dose of PAE at embryonic day 18 did not affect the susceptibility to develop NMDA‐induced WRS (A), GTCS (C), and TS (E) or the latency to WRS (B), GTCS (D), and TS (F) in postpartum P42 rats. Data are shown as the percentage ± standard error of proportion of WRS, GTCS, and TS, as well as mean ± SEM for WRS, GTCS, and TS latency. Analysis was performed on litters; the number of litters in each group is indicated within the histogram and the total number of animals is reported in parentheses. Chi‐square test was used to compare seizure incidence, whereas two‐way ANOVA was used for seizure

By P42 following PAE at E18, there was no difference in the prevalence of WRS, GTCS, and TS in both females and males compared to control‐treated groups (Figure 5A,C,E). PAE also did not alter the latency to WRS, GTCS, or TS in both females and males compared to the control‐treated groups. There was also no difference in the seizure severity in PAE female rats (2.75 ± 1.37, n = 4) and male rats (3.0 ± 21.05, n = 6) compared to the control‐treated rats.

4. DISCUSSION

Here, we report that transiently exposing developing rats to high BAL in the second‐trimester equivalent of pregnancy significantly increases the susceptibility of the postpartum rats to develop NMDA‐induced seizures. The enhanced susceptibility to NMDA‐induced seizures included: (i) increase in the prevalence of GTCS in both male and female P7 pups with males being more susceptible to exhibit GTCS; (ii) increase in the prevalence of GTCS in P21 but not older female rats; (iii) increase in the prevalence of GTCS in P21‐P35, but not in P42 male rats; (iv) increase in the seizure severity of in P7 but not in older female rats; and (v) increase in the seizure severity in P7‐P35 but not P42 male rats. The increased prevalence of GTCS and seizure severity was not associated with hyperthermia and hypoxia in rat dams.20, 21 These findings suggest that PAE increases the susceptibility to develop NMDA‐induced seizures in both female and male offspring; this effect was transient, such that increased seizure susceptibility was not observed at P35 in females or P42 in males. Thus, a single PAE exposure early in development can induce pathophysiological changes in the brain, ultimately leading to increased susceptibility to seizures in offspring rats.

Generalized tonic‐clonic seizures are among the most severe convulsive seizures that can occur in infancy and early childhood among children with FASD.3, 4 Consistent with clinical studies, we found that a single PAE episode at E18 increased the prevalence of NMDA‐induced GTCS in postpartum P7‐P35 rats, which is the developmental period of a full‐term newborn to early adolescent in humans.7 Similarly, an increase in PTZ‐induced seizure susceptibility was also reported in juvenile but not in adult rats that had been subjected to a single alcohol exposure at P4.8 Although this similarity supports the translational nature of our model, an important difference exists between GTCS in our PAE model and patients with FASD. In our PAE model, GTCS occurred in response to the pharmacological activation of NMDA receptors in the brain; in patients with FASD, these seizures develop spontaneously. Nevertheless, our PAE model is an important step toward unraveling the neuronal hyperexcitability mechanisms that lead to enhanced seizure susceptibility in patients with FASD.

N‐methyl‐D‐aspartate (NMDA)‐induced seizure activity is thought to originate in the hippocampus CA1 region, where NMDA receptors are the most abundant and these seizures generalize quickly.22, 23 The enhanced susceptibility to develop NMDA‐induced GTCS following a single PAE in developing rats may reflect a reduced seizure threshold in specific brain regions that are sensitive to alcohol exposure‐related seizures. GTCS originate in the brainstem; thus, in PAE rats, the target of NMDA‐induced seizure activity is a seizure‐sensitive brainstem area, and one site of interest is the inferior colliculus (IC). NMDA‐induced GTCS in both developing and adult rats are similar to acoustically evoked GTCS in rats during alcohol withdrawal and in the genetically epilepsy‐prone rat (GEPR).24, 25, 26 With respect to the alcohol withdrawal seizure model and the GEPR, acoustically evoked GTCS are initiated by neurons in the IC.24, 25 Evidence indicates that the IC is also capable of generating seizure activity in rats as young as 3 days old.27 Interestingly, activating NMDA receptors within the IC of normal adult rats also increases the susceptibility to acoustically evoked GTCS.26 Therefore, we posit that the IC plays a role in NMDA‐induced GTCS in developing rats following PAE. Accordingly, we found that PAE increases the firing in IC neurons of postpartum rats (N'Gouemo, personal communication).

In the United States, 10%‐45% of women reported drinking alcohol at some time point during pregnancy, of which an alarming 3%‐15% reported binge drinking.28 Evidence indicates that binge drinking is associated with the increased susceptibility to develop seizures and epilepsy.5, 29 The ethanol exposure paradigm used in this study modeled a binge drinking pattern that achieved relatively higher BAL in rat dams; these BAL are similar to those reported in previous studies following acute binge alcohol administration 30 but are on the lower side compared to BAL seen in patients.31 The rat dam's BAL in this study also matched the estimated BAL in women who later give birth to an infant with FASD.2 In the present study, elevated BAL were also found in fetuses, indicating that ethanol readily crosses the placenta barrier. As alcohol is eliminated much more slowly in the fetus than the mother,32 this finding suggests that the enhanced susceptibility to NMDA‐induced seizures is a consequence of the effects of alcohol on the developing brain.

Although apoptosis has been reported as a possible mechanism underlying the teratogenic effect of ethanol on the developing brain,33 the mechanisms of how PAE increases seizure susceptibility in offspring rats are not fully understood. Nevertheless, PAE attenuates GABAergic inhibition in the basolateral amygdala 34 and CA3 hippocampal pyramidal cells,35 resulting in neuronal hyperexcitability in offspring rats. We have also found that PAE upregulates the expression of voltage‐gated Ca²⁺ channels in IC neurons of postpartum rats (N'Gouemo, personal communication).

Binge drinking during pregnancy is particularly harmful to fetal brain development and even a single episode of binge drinking is sufficient to produce FASD in humans 2, 36 and increased seizure susceptibility in rodents.3, 4, 5, 6 Interestingly, PAE‐induced neuronal cell death is particularly high during the second‐trimester equivalent of pregnancy in rats.37 Thus, our rat PAE model is designed to model the increased GTCS susceptibility by a single binge drinking episode by a pregnant woman in the second trimester of pregnancy. This PAE episode exposes the developing fetus to both a high peak in BAL and acute withdrawal from alcohol,36 leading to enhanced seizure susceptibility in postpartum developing rats. In the chronic PAE model, however, repeated episodes of alcohol withdrawal led to the increased severity of subsequent withdrawal episodes,38 making it difficult to evaluate the impact of a single PAE episode on brain structure and function.

In conclusion, using a single episode of maternal binge‐like drinking in the second‐trimester equivalent, we found that developing rats have an increased susceptibility to NMDA‐induced GTCS. Understanding how PAE contributes to the neuronal hyperexcitability that drives these GTCS may facilitate the development of targeted therapeutic approaches for treating PAE‐related seizures.

CONFLICT OF INTEREST

The authors declare no conflict of interest. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Cho SJ, Lovinger DM, N'Gouemo P. Prenatal alcohol exposure enhances the susceptibility to NMDA‐induced generalized tonic‐clonic seizures in developing rats. CNS Neurosci Ther. 2017;23:808–817. 10.1111/cns.12756