Phase-locking of event-related oscillations is decreased in both young adult humans and rats with a history of adolescent alcohol exposure

Abstract

Alcohol exposure typically begins in adolescence and frequent binge drinking has been associated with health risk behaviors including alcohol use disorders (AUD). Few studies have documented the effects of a history of adolescent binge drinking on neurophysiological consequences in young adulthood. Synchrony of phase (phase-locking, PL) of event-related oscillations (EROs) within and between different brain areas reflects communication exchange between neural networks, and is a sensitive measure of adolescent development in both rats and humans, and thus may be a good translational measure of the potential harmful effects of alcohol exposure during adolescence. In this study EROs were collected from 1041 young adults of Mexican American and American Indian ancestry (age 18–30 yrs) with and without a history of adolescent binge drinking (5 drinks for boys 4 for girls per occasion at least once per month), and in 74 young adult rats with and without a history of 5 weeks of adolescent alcohol vapor exposure. PL of theta and beta frequencies between frontal and parietal cortex were estimated using an auditory-oddball paradigm in the rats and a visual facial expression paradigm in the humans. Significantly lower phase-locking between frontal and parietal cortices in the theta frequencies was seen in both the humans and the rats with a history of adolescent alcohol exposure as compared to their controls. These findings suggest that alcohol exposure during adolescence may result in decreases in synchrony between cortical neuronal networks, suggesting a developmental delay, in young adult humans and in rats.

INTRODUCTION

Alcohol use disorders (AUD), not unlike other mental health disorders, follow a neurodevelopmental trajectory, and typically emerge during young adulthood following several years of heavy drinking that can begin in early adolescence (Johnston et al., 2009). When adolescents and young adults drink, they often report “binge drinking” (5 or more drinks for boys, 4 or more for girls per drinking occasion) (Centers for Disease & Prevention, 2012). Frequent binge drinking has been linked to a number of health risk behaviors (Miller et al., 2007) including risk for AUD (Dawson et al., 2008). There are demographic differences in the rates of binge drinking with American Indian adolescents reporting the highest rate of past-month binge drinking as compared to White and Mixed-Race adolescents in the U.S. (Chen et al., 2012).

During adolescence, as during the fetal period, there are major changes that occur in brain morphology (Giedd, 2004) that may render the brain particularly vulnerable to exposure to alcohol during this time period. These changes including: the selective removal of 40–50% of synapses (i.e. synaptic pruning) (Johnston, 1995; Lidow et al., 1991; Seeman, 1999) continued myelination of cortical regions (Giedd et al., 1996) changes in neurotransmitters and their receptor levels, as well as changes in receptor sensitivity (Lidow et al., 1991). The global cognitive/behavioral and neurobiological changes seen over adolescent development most likely involve the coordination of a large number of neuronal assemblies that are in communication within individual brain areas as well as across different brain subsystems (Singer, 1993). One phenomenon that has been forwarded as a potential mechanism that could allow for such large-scale integration and communication within the brain is a process whereby groups of neuronal ensembles begin oscillating within a specific frequency range and then enter into precise phase-locking, or synchrony, with each other over a limited period of time (Lachaux et al., 1999). Measures of neural synchrony, or phase locking within a specific set of frequencies, can be measured among neurons in a local population (‘local synchrony’) or can be indexed between neuronal ensembles in separate brain regions (‘long-range synchrony’) (see (Singer, 1999; Varela et al., 2001)).

Phase locking or synchrony of local and long-range networks can be indexed in both humans (Roach & Mathalon, 2008), and more recently in rodent models (Ehlers & Criado, 2009; Ehlers et al., 2014) using event-related oscillation (ERO) paradigms. EROs are neural oscillations within specific frequency bands that are recorded in the EEG over a limited time period following the presentation of a time-locked sensory or cognitive stimuli (Basar et al., 2000; Roach & Mathalon, 2008). EROs are thought to arise by a “phase re-ordering” of the background EEG within specific frequency bands (Basar, 1980). Several recent studies have hypothesized that theta oscillations in fronto- parietal networks may be particularly important in indexing cognitive function. For instance it has been suggested that synchrony in the EEG theta (3–9 Hz) and beta (12–20 Hz) frequencies represent an index of the coordination of relevant information during visual cognitive tasks within a specific neuronal network between frontal and parietal cortices (Babapoor-Farrokhran et al., 2017; Womelsdorf & Everling, 2015). It has also been forwarded that theta oscillations may index decision making in frontal cortex and may also mediate reciprocal communications between parietal attentional control regions and frontal executive function during attentional tasks (Rajan et al., 2018).

The development of techniques to index ERO phase-locking in humans and in animal models provides a tool by which to probe how behavior in these key neuronal circuits may be impacted by adolescent developmental alcohol exposure. Using an auditory ERO paradigm in human and rat, we have previously demonstrated that both peri-adolescent humans and rats have significantly lower phase-locking of EROs between frontal and parietal cortices (FZ-PZ) than their adult counterparts in several frequency ranges (Ehlers et al., 2014). These findings are consistent with the hypothesis that increases in ERO synchrony may be a good index of whether long range neuronal networks have matured over adolescent brain development. It also suggests that this process of adolescent brain development is conserved over species.

The development of animal models of adolescent alcohol exposure also allows for the experimental control necessary to study the effects of ethanol on brain development independent of many factors that may confound human studies such as: differences in genetic risk, psychiatric comorbidity, environmental hardships, and other current and past substance use. Studies from our laboratory (see (Ehlers et al., 2011; Ehlers et al., 2013b; Ehlers et al., 2013c; Ehlers et al., 2018a) and others (for review see (Crews et al., 2016; Spear, 2016, 2018)) in rats, have demonstrated that chronic ethanol exposure during adolescence can produce persistent changes in brain and behavior in adulthood.

The present set of translational studies used ERO data collected using this rat model, as well as data from young adult humans with a history of adolescent binge drinking. The overall aim of the study was to investigate whether ERO phase-locking in the neural network between frontal and parietal cortices in the theta and beta frequency ranges, could be impacted in young adulthood by a history of adolescent alcohol exposure. A facial recognition task that requires participants to identify happy or sad faces was used to generate EROs (Orozco & Ehlers, 1998). We posited that an emotional recognition task might be a sensitive probe of the impact of binge drinking based on our previous studies using this paradigm in binge drinkers (Ehlers et al., 2007). Our studies were also informed by the hypothesis put forth by Dahl (2004), positing that adolescence is a “window of plasticity” for emotional fluency and that if something interferes with or prevents that learning process it may delay fluent control of emotion at a later time point.

Several studies, in animal models, have demonstrated that one of the consequences of adolescent alcohol exposure may be to cause a “developmental delay” or a persistence of adolescent-typical phenotypes into adulthood (Spear & Swartzwelder, 2014). Developmental delays in electrophysiological measures have also been reported in children at high risk for developing AUDs (Hill et al., 1999). Thus, based on our previous studies, and those of others, we predicted that the increase in ERO phase-locking observed over adolescent development (Ehlers et al., 2014) would be delayed, and therefore not fully matured, in both the young adult rodents and in the humans with a history of adolescent alcohol exposure compared to their controls.

MATERIALS AND METHODS

Human studies

Participants

The participants were individuals of American Indian and Mexican American ancestry that were recruited from the local community for larger studies investigating risk factors for substance dependence in these two ethnic groups. The American Indians were recruited from eight geographically contiguous reservations with a total population of about 3,000 individuals. They were recruited using a combination of a venue-based method and a respondent-driven procedure, as described previously (see (Ehlers et al., 2004)). American Indian participants were recruited in a wide age range (18–82 yrs), however, the present set of analyses focused on only those individuals who were between 18 and 30 years of age at the time of interview. Mexican American participants were recruited using a commercial mailing list that provided the addresses of individuals with Hispanic surnames in 11 zip codes in San Diego County, all of which had a population that was over 20% Hispanic and were within 25 miles of the research site. The mailed invitation stated that potential participants must be of Mexican American ancestry, be between the ages of 18 and 30 years, be residing in the United States legally, and be able to read and write in English (see (Ehlers et al., 2018b)). Based on the aims of the larger studies, for both ethnic groups, participants were excluded if they were pregnant, nursing, or currently had a major medical disorder that precluded them traveling to the research site. Participants were asked to refrain from alcohol or any other substance use for 24 hours prior to testing, and their breathalyzer blood alcohol levels had to be 0.00 g/dl to be included in the study.

Potential participants gave written informed consent, and then responded to a screening questionnaire that was used to gather information on demographics, personal medical history, ethnicity and detailed measures of current and past substance abuse history, including a retrospective report of their adolescent binge drinking, using a time line follow back format. Each participant also completed an interview with the Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) (Bucholz et al., 1994). Family history of AUD was determined by the Family History Assessment Module and was defined as the presence of alcoholism in a first-degree family member. An adolescent history of binge drinking was defined as drinking 5 or more drinks for boys and 4 or more drinks for girls, per drinking occasion, for at least one time a month during their highest drinking period at or before the age of 19 years.

Electrophysiological recordings

For ERO recordings, in humans, seven channels of EEG data (FZ, CZ, PZ, F3, F4, F7, and F8, referenced to linked ear lobes with a forehead ground, international 10–20 system) were obtained using gold-plated electrodes with impedance held below 5 kΩ. An electrode placed left lateral infraorbitally and reference to the left earlobe was used to monitor both horizontal and vertical eye movements. EEG recording signals were amplified (high pass 0.5 Hz, 35 Hz low pass) and were transferred on-line to a PC for digitation. The present study used a facial discrimination task (Gur et al., 1992) that was adapted for use in an event-related potential (ERP) paradigm (Orozco & Ehlers, 1998) to generate the ERO data. This task has been demonstrated to be sensitive to the effects of binge drinking (Ehlers et al., 2007). The stimuli were realistic digital photographs of happy, neutral and sad faces presented on a computer screen. All faces were of EuroAmericans, the use of faces from other racial groups could potentially elicit different responses. Participants were instructed to discriminate, by pressing a counter, between happy and sad faces (36 trials each) and not to respond to neutral faces (144 trials). There were 36 unique faces (12 each of happy, neutral, and sad) presented in random order for a total of 216 trials. The number of male and female faces presented was also equally distributed among neutral, sad and happy stimuli. The percentage of the stimuli that were correctly identified within a 1000 msec window following the stimuli were calculated for each participant. Only trials from correctly identified stimuli were included in the data analyses.

PLI (phase lock index) analyses were accomplished, in part, from datasets that were previously used to generate ERP data (Criado & Ehlers, 2007a; Ehlers et al., 2007). This ERO measure was selected because previous studies had indicated that it was a sensitive measure of adolescent to adult brain development (Ehlers et al., 2014). The methods for these analyses have been described in detail elsewhere (Ehlers et al., 2015). Briefly, the ERO trials were digitized at a rate of 256 Hz. Trials containing excessive artifact were eliminated prior to averaging (<5% of the trials). Data from single trials generated by the stimuli were entered into the time frequency analyses algorithm. The S-transform (ST), a generalization of the Gabor transform was used. PLI is a measure of constancy over trials of the difference in phase angle between two channels, as a function of frequency and of time relative to the start of the stimulus for each trial. The range of PLI is from zero to 1.0, with high values at a time and frequency indicating little variation, among trials, of phase angle difference between channels of the pair, at that time and frequency. Two rectangular regions of interest (ROIs) were defined within the time-frequency analysis plane by specifying, for each ROI, a band of frequencies and a time interval relative to the stimulus onset time. The ROI frequencies selected were: theta (4–7 Hz), and beta (13–30 Hz) frequencies, based on apriori hypotheses that processing of stimuli in these frequency ranges in the fronto-parietal network would be disrupted by adolescent alcohol exposure. The ROI time intervals were: theta (400–800 ms), and beta (300–800ms). ROI time intervals were selected based on ERO energy in specific ERP component location (P3) identified in previous ERP/ERO studies (Criado & Ehlers, 2007a, b). Using mean values over trials, the maximum values were calculated for each ROI, for the pair of electrode locations (FZ-PZ).

Preclinical studies

Animal subjects

Seventy-four (74), adolescent male Wistar rats were obtained from Charles River (USA) and arrived on postnatal day (PD) 21. Adolescent rats were triple-housed in standard plastic cages in a temperature-controlled room with a 12h light/dark cycle (lights on 08:00). Food and water were available ad libitum. The work described herein adheres to the guidelines stipulated in the NIH Guide for the Care and Use of Laboratory Animals (NIH publication No. 80–23, revised 1996) and was reviewed and approved by The Scripps Research Institute’s Animal Care and Use Committee.

Ethanol vapor exposure

The ethanol vapor inhalation procedure and the chambers used in this study have been previously described (see (Ehlers et al., 2011; Ehlers et al., 2013b; Ehlers et al., 2013c; Ehlers et al., 2018a). Ethanol vapor chambers were calibrated to produce high to moderate blood ethanol concentrations between 175–225 mg/dL. Rats were randomly divided into two groups; ethanol or control group. Ethanol exposed rats were housed in sealed chambers, infused with vaporized 95% ethanol from 8 p.m. to 10 a.m. For the remaining 10 hrs of the day, ethanol vapor was not infused into the chambers. Rats were exposed to this daily vapor cycle for 5 weeks (P22–57). Blood samples were collected from the tip of the tail every 3–4 days during the 5-week exposure period to assess BECs (5-week average= 204.6 ± 4.24 mg/DL). Control animals were handled identically to ethanol rats. BECs were determined using the Analox micro-statAM1 (Analox Instr. Ltd., Lunenberg, MA). Following the 5-week exposure, animals were transferred to standard polycarbonate cages for the duration of the experiment.

Surgical procedure

Surgical and electrophysiological recording procedures performed in this study have been previously described (Ehlers et al., 2013a). Following withdrawal from vapor exposure or control conditions, rats were surgically implanted (PD 55–71) with screw electrodes in the skull overlying the frontal cortex (FZ, AP: 1.5 mm, ML: ± 3.0 mm, FR1), and parietal cortex (PZ, AP: −4.5mm, ML: ± 4.5 mm, PAR3) with a corresponding reference electrode posterior to lambda overlying the cerebellum. Surgical coordinates were obtained from the Paxinos and Watson (Paxinos & Watson, 1986) atlas. EEG electrodes were connected to a multi-pin PlasticsOne® connector and the assembly was anchored to the skull with dental acrylic and anchor screws. Rats were given at least 2 weeks of recovery before EEG recordings were made.

Electrophysiological recordings

After recovery from surgery, rats were habituated to the EEG testing chambers prior to the electrophysiological testing. For waking EROs, animals were placed in individual polycarbonate chambers within a sound and electrically shielded recording chamber. Rats were unrestrained but were spatially restricted. Rats were first presented with auditory stimuli through a small speaker centered approximately 70 cm above the rat’s head. Auditory stimuli and EROs were elicited using an oddball plus “noise” paradigm described previously (Ehlers et al., 2012). This task has also been shown previously to be sensitive to developmental changes in brain in both rats and humans over the adolescent/ young adult time period (Ehlers et al., 2014). Each auditory ERO session consisted of 312 trials that lasted approximately 10 minutes and each trial presented one of three randomly generated tones: frequent tone (84% probability, 75db, 1000Hz), infrequent tone (10% probability, 85db, 2000Hz), or noise tone (6% probability, 100db, white noise). Individual trials were 1000ms in duration (200 ms pre-stimulus + 800 ms post-stimulus) and the interval between tones varied from 750ms to 1500ms. PLI (phase lock index) analyses were accomplished from the same datasets that were used to generate ERP data in the rats. Methods for estimating PLI in rats were identical to those used in the clinical studies, for the two frequency bands (theta, beta) for the infrequent tone presented in the auditory paradigm as described above.

Statistical analyses

The data analyses were based on the study aims. The first aim was to determine if a history of binge drinking during adolescence resulted in reduced ERO phase locking in the human population. Demographic variables were compared between those participants with an adolescent history of binge drinking and those who did not report binge drinking during adolescence, using Chi Square for dichotomous variables and ANOVA for continuous variables. To determine the relationships between binge drinking and ERO phase locking over cortex (Fz vs. Pz), analyses of variance (ANOVAs) were calculated for the two frequency bands (theta, beta) for the three stimuli (happy, sad, neutral) for those participants with a history of bingeing during adolescence and for those who did not have such a history co-varying for significant demographic variables (race, age and education). ANOVA was also used to determine if the percentage of correctly identified stimuli differed as a function of binge drinking, in the human participants, while covarying for significant demographic variables. In order to determine if either family history of alcoholism or current binge drinking could also influence PL in the theta and beta frequency ranges, we also tested those variables for significance co-varying for race, age, and education. Effect sizes (ES) were calculated using Cohen’s d. The second aim was to evaluate whether reductions in phase locking of EROs were associated with a history of adolescent alcohol vapor exposure in the rats. To determine the relationships between adolescent alcohol vapor exposure and phase locking over cortex (Fz vs. Pz), analyses of variance (ANOVAs) were calculated for the two frequency bands (theta, beta) for the three auditory stimuli (frequent, infrequent, noise) comparing those rats with adolescent vapor exposure to those experiencing control (air) conditions.

RESULTS

Human studies

The sample consisted of American Indian (n=460) and Mexican American (n=581) young adult (18–30 yrs old) participants. Fifty-six percent of the sample was female, 47% of the sample was currently employed and 29% of the sample had an annual income below 20K. The mean age was 22.7 ± 0.1 yrs, and the mean level of education was 12.6 ± 0.06. Fifty-two percent (n=538) of the sample reported adolescent binge drinking (5 drinks for men 4 drinks for women per drinking occasion at least once a month during their highest drinking period prior to the age of 19 years). Among the binge drinkers the mean number of drinks per occasion was 10.8. As seen in table 1 while binge drinkers did not differ on gender or income, they were younger at the time of interview (F=8.4, df=1, p<0.004), more likely to be American Indian (Chi Square= 81.4, df=1, p<0.001), and have less education (F=49.9, df=1, p<0.001), have a family history of alcoholism (Chi Square =36, p<0.0001), and report current binge drinking (Chi Square=102,p<0.0001), compared to participants without a history of adolescent binge drinking.

Table 1.

Demographic characteristics of Mexican American (n=581) and American Indian (n=460) participants according to adolescent binge drinking history

Demographic Characteristic	Adolescent Binge (n=538)	Adolescent Non-Binge (n=503)	Overall (n=1041)
	N (%)	N (%)	N (%)
Race
American Indian	310 (29.8)	150 (14.4)	460 (44.2)
Mexican American	228 (21.9)	353 (33.9)	581 (55.8)
Gender
Male	250 (24.0)	211 (20.3)	461 (44.3)
Female	288 (27.7)	292 (28.0)	580 (55.7)
Married
Yes	50 (4.8)	65 (6.2)	115 (11.0)
No	488 (46.9)	438 (42.1)	926 (89.0)
Employed
Yes	220 (21.5)	259 (25.3)	479 (46.8)
No	305 (29.8)	240 (23.4)	545 (53.2)
Income ≥ $20,000/yr
Yes	340 (35.4)	344 (35.8)	684 (71.2)
No	152 (15.8)	125 (13.0)	277 (28.8)
Recent Binge Drinking
Yes	276 (26.5)	106 (10.2)	382 (36.7)
No	262 (25.2)	397 (38.1)	659 (63.3)
	Mean (SD)	Mean (SD)	Mean (SD)
Age (yrs)	22.4 (3.9)	23.1 (3.9)	22.7 (3.9)
Education (yrs)	12.2 (1.7)	13.0 (1.9)	12.6 (1.8)
Adolescent Binge Drinking Quantity (drinks/occasion)	10.8 (8.3)	2.4 (4.3)	7.8 (8.2)

As seen in figure 1, a history of adolescent regular binge drinking was found to be significantly associated with a reduction in ERO phase locking between frontal and parietal cortex in an ANOVA that included age, race and education as covariates. Binge drinkers had lower phase lock indices to the neutral faces in the beta (F=4.4,p=0.035; ES=0.135) and theta (F=3.9,p=0.048; ES 0.121) frequencies and to the sad faces in the beta frequencies (F=5.9,p=0.015;ES=0.164). Age was also found to be associated with changes in phase-locking. When the entire population was parceled into 3 age groups (18–21 yrs, 22–25 yrs, 26–30 yrs) a significant increase in phase locking was found, overall, in both the theta and beta frequency bands (F=5.6–31.5; df=2, 1038; p<0.004–0.0001) as a function of increasing age as seen in figure 2. A post hoc ANOVA revealed that the significant increase was only between those that were in the younger age range compared to the other two older age ranges (18–21 yrs vs 22–25 yrs (p<0.003–0.0001; ES=0.23–0.50); 18–21 yrs vs 26–30 yrs (p<0.01–0.0001;ES=0.2–0.44); 22–25 yrs vs. 26–30 yrs (all p’s = NS)). Race was also found to influence phase locking with American Indians having lower phase locking indices than Mexican Americans in both theta and beta frequency ranges (F= 3.8–36.0; df= 1,1040; p< 0.05–0.0001:ES=0.13–0.38), although it should be noted that American Indians were more likely to be binge drinkers during adolescence than the Mexican Americans. Figure 3 shows a color map of the degree of ERO phase locking over the relevant frequency spectrum (1–30 Hz) in the time region of interest (300–800 msec) between controls and adolescent binge drinkers in both American Indian and Mexican American participants. There was no significant effect of currently reported binge drinking on phase locking in either the beta or theta frequency ranges to any stimuli. Family history of alcoholism was associated with lower PLI in the beta frequency range to the sad faces (F= 5.3, p<0.02; ES=0.24) but no significant findings were observed to happy or neutral faces in the beta frequency range or to any facial stimuli in the theta frequency range.

Figure 1.

Phase locking index (PLI) between frontal and parietal cortex in American Indian and Mexican American participants combined with and without a history of binge drinking during adolescence. Adjusted means for PLIs, co-varying for age, race, and education shown for the theta (top) and beta (bottom) band frequencies in response to a facial discrimination task. Significant results were observed in response to the neutral stimuli for the theta frequencies and the theta and sad facial stimuli for the beta frequencies. * p<0.05, error bars=SEM.

Figure 2.

Phase locking index (PLI) between frontal and parietal cortex in American Indian and Mexican American participants combined, grouped by their age. PLIs were estimated for the theta (top) and beta (bottom) band frequencies in response to a facial discrimination task. Participants between the ages of 18 to 21 yrs of age were observed to have significantly lower phase locking index to all (happy, neutral, sad) facial stimuli in both frequency bands compared to older age groups. * p<0.01, error bars=SEM.

Figure 3.

Color equivalents of phase locking index (PLI) between frontal and parietal cortex in Mexican American (A) and American Indian (B) young adults with a history of adolescent binge drinking and their controls. In the first vertical panel PLI to happy, in the second panel response to neutral and in the third panel response to sad facial expressions within the time regions of interest for the theta and beta frequencies are presented. Red color indicates the highest level of phase locking.

The percentage of correctly identified stimuli (happy, sad, neutral) for each participant was calculated and compared for all demographic variables (race, gender, age group, marriage status, economic status, and education) using ANOVA. The percentage of correctly identified stimuli did not differ based on gender, age group (18–21 yrs, 22–25 yrs, 26–30 yrs), or marriage status. However, a significantly lower percentage of correctly identified stimuli were found for American Indians for the sad (mean: AI=87%, MA=91%, F=20.0, p<0.0001) neutral (mean: AI=94%, MA=96%, F=23.6, p<0.0001) and happy stimuli (mean: AI=90%, MA=92%, F=8.3, p<0.004). Significant findings were also found for economic status with those participants who reported making less than $20K a year having a lower percentage of correctly identified stimuli for the sad (mean: <$20K =85%, >$20K=91%, F=35.4, p<0.0001) neutral (mean: <$20K =94%, >$20K=96%, F=10.1, p<0.001) and happy stimuli (mean: <$20K =90%, >$20K=92%, F=4.5, p<0.03). Education was also found to significantly influence responses as those with at least a high school education (HS) had higher percentages of correctly identified stimuli for the sad (mean: <HS =86%, ≥HS =90%, F=14.6, p<0.0001) neutral (mean: <HS =94%, ≥HS =96%, F=5.9, p<0.02) and happy stimuli (mean: <HS =90%, ≥HS =92%, F=4.3, p<0.04). However, when an ANOVA was conducted for the effect of binge drinking status, that co-varied for race, education and economic status, no significant differences in the percentage of correct identifications were found for any stimuli (sad (mean: binge =89%, no binge =89%, F=0.1, p<0.7); neutral (mean: binge =95%, no binge =95%, F=0.7, p<0.39); happy (mean: binge =91%, no binge =91%, F=0.0, p<0.99)).

Animal studies

Adolescent vapor exposure was found to significantly reduce phase locking indices (PLI) in theta frequencies as compared to their non-exposed controls. Phase locking was decreased in the theta band for all auditory stimuli (Figure 4); frequent (F=9.0, p=0.004), infrequent (F=7.7, p=0.007) and noise (F=5.0, p=0.028). No significant findings were observed in PLI in the beta frequencies in rats. Figure 5 shows a color map of the degree of ERO phase locking over the relevant frequency spectrum (4–30 Hz) in the time region of interest (300–800 msec) in vapor exposed rats and their controls in response to the three auditory stimuli.

Figure 4.

Phase locking index (PLI) between frontal and parietal cortex in rats with a history of exposure to five weeks of chronic intermittent alcohol vapor during adolescence and their controls. PLIs were estimated for the theta and beta band frequencies in response to each tone: frequent, non-frequent and noise. Significant results were observed in the theta frequencies in response to all three tones. * p<0.05, error bars=SEM.

Figure 5.

Color equivalents of phase-locking index (PLI) between frontal and parietal cortex in young adult rats with a history of adolescent alcohol vapor exposure and their controls. PLI responses to each tone (frequent, infrequent, and noise) are shown in the time regions of interest in the theta and beta frequencies. Red color indicates the highest level of phase locking.

DISCUSSION

Alcohol and other substance use, especially during early adolescence, has been linked to a number of long-terms health risks (Miller et al., 2007) that include elevated risk for AUDs later in life in a number of populations including American Indians (Dawson et al., 2008; Ehlers et al., 2006). In our present study, we found that a personal history of adolescent binge drinking was associated with lowered phase locking between frontal and parietal cortices in the beta and theta frequency ranges in a population of American Indians (AI) and Mexican Americans (MA). We also found some differences in ERO phase-locking in our present population of AI and MA young adults based on age. Lower phase locking was found to occur in those participants between 18 and 21 yrs of age when compared to those participants that were between 22 and 30 yrs of age. These findings are consistent with our previous studies that demonstrated that peri-adolescent humans have significantly lower phase-locking of EROs between frontal and parietal cortices (FZ-PZ) as compared to adults (Ehlers et al., 2014). Since it has been suggested that the human brain becomes “neurobiologically adult” in the early 20s (Mills et al., 2016), it is possible that ERO phase locking may represent a neurobiological marker of brain maturation during that time.

We also investigated whether the ERO task requirements, as indexed by the percentage of correctly identified facial expressions used to generate the electrophysiological responses, differed between participants in the study. We found that race (AI), education (< high school completion) and economic status (earning<$20K/yr.) all contributed to the finding of lower percentages of correctly identified facial expressions. However, the two variables that contributed to significant differences in ERO phase locking, age and binge drinking, did not. This suggests that the neuro-cognitive consequences of adolescent binge drinking, and associated indices of brain maturation, do not include deficits in simple discriminations of these facial expressions. Although the neuro-cognitive consequences of our findings of reduced cortical phase locking in participants with a history of adolescent alcohol exposure are not fully understood, event-related oscillations over the spectral range of the EEG (1–50 Hz) have been suggested to underlie a number of different cognitive processes. Oscillations in the theta frequency ranges have been associated with signal detection, decision making, inhibitory processing, conscious awareness, executive control, recognition memory, episodic retrieval and reward processing (Basar et al., 2001), all processes that mature during adolescence. It has also been suggested that high frequency oscillations (above 30 Hz) reflect synchronization of neuronal ensembles that are interacting over short distances in response to primarily sensory processes, whereas, lower frequency oscillations are generated by synchronization of ensembles interacting at longer distances during higher cognitive processing (Kopell et al., 2000). Our studies in humans suggest that a history of adolescent binge drinking may impact synchronization in both lower (theta) and higher (beta) frequency ranges in humans but only in theta frequencies in rodents. The lack of findings of beta frequency differences between rodents and humans may be due to differences in the task used to generate EROs between the two species or may represent a true species difference. Additionally, a family history of alcoholism was found to be associated with lower phase locking in the beta frequency range to the sad faces in humans. Thus, changes in the beta frequency range may also reflect the impact of trait variables as has been seen in other electrophysiological markers (Hill et al., 1999).

Deficits in connectivity have been suggested as a potential pathological feature in a number of neurodevelopmental disorders, most prominently in schizophrenia (Basar, 2013; Ford et al., 2007; Roach & Mathalon, 2008; Sigurdsson, 2016; Spellman & Gordon, 2015; Uhlhaas & Singer, 2013), but also in depression, obsessive compulsive disorders, bipolar disorder and externalizing disorders. The present set of findings add to this literature by demonstrating that lowered phase locking is also associated with developmental alcohol exposure. Additionally, it suggests that when lowered phase locking is found in individuals with psychiatric illness that developmental exposure to alcohol should be considered as a covariate.

The development of preclinical models of alcohol associated changes in ERO synchrony allows for study of the effects of adolescent ethanol exposure on EROs in young adulthood independent of many factors that may confound human studies such as differences in genetic risk, and a host of potential cultural and environmental factors. In the present study, we found that a history of adolescent alcohol vapor exposure was associated with lower levels of phase-locking between frontal and parietal cortex in the theta frequencies as compared to control conditions. Changes in resting state connectivity, as indexed by functional MRI, between frontal cortex and limbic sites in adult rats with a history of high dose intragastric adolescent alcohol administration has also been reported (Broadwater et al., 2018). In that study, a history of adolescent alcohol exposure was found to be associated with decreased baseline resting state connectivity between prefrontal cortical subregions and between prefrontal cortical-striatal regions as compared to control conditions. Thus, both electrophysiological and functional MRI data suggest that adolescent alcohol exposure, in animal models, may lead to remodeling of brain circuits that leads to lower levels of long-range synchrony. Several studies, in animal models, have demonstrated that one of the consequences of adolescent alcohol exposure may be to cause a “developmental delay” or a persistence of adolescent-typical phenotypes into adulthood (Spear & Swartzwelder, 2014). Taken together our study suggests that the reduced phase locking seen in electrophysiological and functional MRI data may represent further measures of this developmental delay associated with adolescent alcohol exposure.

We have previously shown that alcohol vapor exposure during adolescence in rats can lead to deficits in a number of physiological and behavioral responses in adulthood. For instance, reductions in prepulse startle, increases in behavioral measures of disinhibition, increased alcohol drinking, reductions in hippocampal size and measures of neurogenesis, as well as deficits in slow wave sleep have all been documented (Ehlers et al., 2011; Ehlers et al., 2013b; Ehlers et al., 2013c; Ehlers et al., 2018a). Thus, it is likely that adolescent alcohol exposure affects multiple circuits and neurotransmitter systems that could also potentially influence ERO phase locking between cortical sites. We have demonstrated, using this rat adolescent alcohol exposure model, that chronic adolescent vapor exposure produces lasting reductions in cell counts of ChAT- immunoreactive (ChAT-IR) neurons in the Ch1–2 and Ch3–4 regions of the basal forebrain. This reduction in cell counts was significantly correlated with more “disinhibitory” behavior in the open field conflict test (Ehlers et al., 2011). However, whether these reductions in ChAT-IR are also responsible for the ERO findings observed in the present study is also not known. However, we have previously demonstrated that selective cholinergic lesions to the nucleus basalis magnocellularis resulted in reductions in phase-locking in the theta and beta frequency ranges within the frontal cortex in rats (Sanchez-Alavez et al., 2014) lending some supporting evidence for this idea.

In conclusion, these data represent the first association analysis of ERO phase locking with a history of regular binge drinking during adolescence in humans and following adolescent alcohol vapor exposure in rats. The results of this study should, however, be interpreted in the context of several limitations. First, the human findings may not generalize to other American Indians or Mexican American young adults. Comparisons in responses between the American Indian and Mexican American population are also limited by differences in recruitment strategy. Comparisons of these findings to non-American Indian or non-Mexican populations may also be limited by differences in a host of potential demographic, genetic and environmental variables. This study relied on cross-sectional and retrospective data on binge drinking and a longitudinal study of the relationship between adolescent binge drinking and the ERO phase-locking would represent a more powerful study design. While we have suggested that adolescent binge drinking may cause developmental delays only longitudinal studies can demonstrate whether this hypothesis is supported. The effects of binge drinking were found to produce small effects sizes, this is likely due to the fact that a host of genetic and environmental variables can potentially impact ERO phase locking. Chronic ethanol vapor exposure during adolescence, in the rat, was also found to produce a deficit in phase locking between frontal and parietal cortical areas. While preclinical studies allow the experimental control to investigate the effects of alcohol on adolescent brain development, they do not shed light on the complex motives and risk factors for drinking in adolescence. Additionally, only a passive auditory task was used to generate EROs in the rodent studies whereas in the human studies a complex visual task was used thus further limiting direct comparisons. Despite these limitations, this report represents an important step in an ongoing investigation to understand the developmental determinants associated with substance use disorders in high-risk and understudied ethnic groups as well as to identify mechanisms and potential therapeutic targets in preclinical models.