Impact of Sex: Determination of Alcohol Neuroadaptation and Reinforcement

Kristine M. Wiren and Joel G. Hashimoto

Chronic alcohol (EtOH) abuse is an intractable problem. One defining aspect of drug dependence is the emergence and elaboration of a constellation of symptoms after the abrupt cessation of drug use, collectively termed withdrawal. The EtOH withdrawal syndrome reflects a state of neuronal hyperexcitability resulting from a variety of cellular adaptive responses to long‐term alcohol consumption, including changes in gene expression. The identification of gene “targets” that are altered by chronic EtOH administration will provide information about such neuroadaptive changes that may reflect or underlie physical dependence and relapse potential.

The role of gender (sex) in the transcriptional response following chronic EtOH exposure is poorly understood. The vast majority of studies looking at the transcriptional response to EtOH exposure have primarily focused on males with little or no mention of how the response may be similar or different in females. Interestingly, there is ample evidence in the literature to suggest that the responses of males and females to EtOH are different, and recently there has been increased focus on the implications of these findings (Hommer et al., 2001; Mann et al., 1992, 2005). Males and females demonstrate marked behavioral and physiological differences in response to alcohol. For example, women develop greater organ damage (EtOH‐induced cirrhosis, cardiomyopathy, and peripheral neuropathy) after fewer years of heavy drinking than do alcoholic men (Ammendola et al., 2000; Fernandez‐Sola and Nicolas‐Arfelis, 2002; Hommer, 2003; Loft et al., 1987; Urbano‐Marquez et al., 1995). However, women's sensitivity to EtOH‐induced brain damage remains controversial and understudied.

As the prefrontal cortex is part of the brain circuitry involved in EtOH withdrawal (physical dependence), is important in aspects of cognitive function, and finally is a site of damage and neurodegeneration in human alcoholics, we have compared the effects of withdrawal following chronic EtOH exposure on prefrontal cortex gene expression between male and female mice. We hypothesized that gender would influence the neuroadaptive responses to chronic EtOH exposure and withdrawal.

The selected lines used in this study were developed from heterogeneous stock (Hs/Ibg, 8‐way cross of inbred mouse lines) that exhibit enhanced or diminished withdrawal severity after chronic EtOH exposure (Crabbe et al., 1983, 1985). The Withdrawal Seizure‐Prone (WSP) and Withdrawal Seizure‐Resistant (WSR) selected lines were developed independently in replicate resulting in the WSP‐1, WSP‐2 and WSR‐1, WSR‐2 lines. Chronic continuous EtOH exposure was achieved with a vapor chamber inhalation procedure, which allowed for high blood EtOH concentration (BEC, ∼2.2–2.5 mg/ml) and temporally controlled withdrawal. The mice were 49 to 95 days old at the start of the experiments and were naïve to any drug exposure. Mice were separated into 2 treatment groups, air/pyrazole, the air control group, and EtOH/pyrazole. Pyrazole, an alcohol dehydrogenase inhibitor, was used to stabilize BECs. Following 72 hours of EtOH exposure, mice were removed from the chambers and allowed to proceed through 8 hours of withdrawal. Brains were harvested following euthanasia, and the prefrontal cortex was isolated by dissection. Using DNase‐treated total RNA isolated from the prefrontal cortex of 2 animals, a complex probe was produced by direct reverse transcription in the presence of [α‐³³P‐dATP]. Following overnight hybridization, GF400 (Invitrogen, Carlsbad, CA) membranes were washed and exposed to phosphoimager screens. Clone intensities were analyzed using a phosphoimager and imported into Pathways software. Raw expression values were then exported, normalized, and analyzed using Vector Xpression 3.1 (Invitrogen). Briefly, each array was normalized using global z scores following Log₂ transformation multiplied by 2 and added to 8 so that each array had a mean of 8, a variance of 4, and a standard deviation of 2. Differential regulation was determined by comparing groups in the following manner. Male WSR EtOH regulation was determined by comparing 4 WSR arrays from 8 male air‐exposed WSR mice to 4 WSR arrays from 8 male EtOH‐exposed WSR mice. Replicate lines were combined to identify withdrawal severity phenotype‐relevant changes, reduce the contribution of randomly fixed genes, and to increase statistical power. For cluster analysis, genes that were significant in one or more comparison were converted to z ratios.

Ethanol‐regulated genes from each comparison (Male WSR, Male WSP, Female WSR, Female WSP) were analyzed using Gene Ontology Tree Machine (GOTM; Zhang et al., 2004), Database for Annotation, Visualization, and Integrated Discovery (DAVID; Dennis et al., 2003), and Expression Analysis Systematic Explorer (EASE; Hosack et al., 2003). Unsupervised hierarchical clustering analysis (average linkage clustering) was performed using the Euclidean distance metric, calculated in TMEV software (tigir.org) using z ratios.

A subset of expression differences identified in the array analysis were confirmed by real‐time RT‐PCR (qRT‐PCR). Relative expression was determined using the comparative ΔΔC_t method (Winer et al., 1999), after normalizing expression with fluorescence to RiboGreen as described previously (Hashimoto et al., 2004). Real‐time qRT‐PCR efficiency was determined for each primer set using a 5‐fold dilution series of total RNA and did not differ significantly from 100%. Individual reaction kinetics were also analyzed to ensure that each qRT‐PCR did not differ significantly from 100%. Following PCR, the specificity of the PCR reaction was confirmed with melt curve analysis to ensure that only the expected PCR product was amplified per reaction.

We identified 295 EtOH‐regulated genes in one or more comparisons following 8 hours of withdrawal after chronic EtOH exposure. Several of these genes have been previously identified as influenced by EtOH exposure or in a variety of models of neurodegenerative diseases, including adenylate cyclase 7 and natural cytotoxicity triggering receptor 1 (Niemann‐Pick type C) Ncr1. Niemann‐Pick type C disease is an inherited lipid storage disorder with major neurological involvement that is characterized by progressive neurodegeneration.

Gene expression patterns were also analyzed by unsupervised hierarchical clustering. Interestingly, with replicates collapsed, overall gene expression differences from the arrays grouped clearly according to sex (male vs female) and not the withdrawal severity phenotype (WSP vs WSR). Bootstrap analysis using 100 iterations confirmed the overall tree topography at 100% support. These results provide evidence for a fundamental difference in the gene expression response to EtOH withdrawal between males and females, irrespective of the divergent behavioral responses during EtOH withdrawal exhibited by the WSP and WSR mouse lines.

Thus, the strongest influence on the gene expression response to chronic EtOH withdrawal was gender, not the selected phenotype. Gene ontology overrepresentation analysis in male WSP and WSR mice showed robust regulation of antiapoptotic and anti‐inflammatory genes and pathways influencing ubiquitination/protein catabolism, protein folding, secretion, and lipid metabolism. Analysis in female WSP and WSR mice showed significant regulation of pathways influencing DNA binding, transcription/translation, RNA editing, apoptosis/cell death, stimulation of gene expression in glia, and inflammation.

Confirmation of expression patterns identified by microarray analysis is a necessity. In males, we have confirmed regulation of Smad3 (both WSP and WSR), fractalkine (Cx3cl1) in WSR, and Hbxip in WSP by qRT‐PCR. Smad3 is known to bind androgen receptor (AR) and prevent signaling through AR, which has been shown to be important in excitotoxic damage (Rahman et al., 2004). Also, Smad3 inhibits Class II major histocompatibility complex expression, which would lead to an anti‐inflammatory response in the central nervous system (Dong et al., 2001). Down‐regulation of fractalkine in WSR males is interesting as fractalkine is a chemokine that aids in the infiltration of microglia as part of microglial activation following injury (Inoue et al., 2005; Nanki et al., 2004). Up‐regulation of Hbxip in WSP males is interesting as Hbxip forms a complex with survivin and binds pro‐caspase‐9, preventing the recruitment of caspase‐9 to Apaf1 and resulting in a suppression of apoptosis (Marusawa et al., 2003). In females, we have confirmed up‐regulation of Smac/DIABLO in WSP and Btf (Bclaf1) in WSR by qRT‐PCR. Both Smac/DIABLO and Btf participate in the apoptotic pathway and their up‐regulation would be consistent with a proapopototic state (Kasof et al., 1999). Although the same broad category is influenced in a sex‐specific fashion, the same genes are not necessarily regulated in the same fashion in both WSP and WSR.

Combined, these results indicate that chronic EtOH exposure/withdrawal differentially alters gene expression in the prefrontal cortex in mice, as has been previously described for human alcoholic individuals. As the prefrontal cortex is involved in executive function, disruption of the normal inhibitory functions (i.e., inhibition of unnecessary or unwanted behavior) may contribute to excessive drinking and the self‐sustaining nature of alcoholism. Notably, the present findings indicate that EtOH withdrawal influences gene expression in the prefrontal cortex with differences that reflect gender more than the selected EtOH withdrawal phenotype. Furthermore, the signaling pathways that are differentially affected suggest that females may have increased vulnerability for EtOH‐induced brain damage.

Paul E. Alele and Leslie L. Devaud

Being female, rather than male, is an important context that needs to be recognized because an increasing body of evidence supports the existence of significant biological differences between men and women that impact health and disease, including alcoholism (Ashley et al., 1977; Matsumoto, 2000; Wizeman and Pardue, 2001). The underlying mechanisms for differences in risks for harm from various diseases between men and women likely involve hormonal influences, owing to the differing hormonal milieu between males and females. Animal models provide insights into human behaviors and health risks because these models allow us to study neurobiological and basic physiological contributions removed from societal and other environmental influences.

Sex differences in levels of gonadal steroid hormones (notably estrogens) occur early in development and continue throughout adulthood. As mentioned in the Introduction to this symposium, gonadal steroid organizational influences are prominent during development, resulting in structural and functional differences between the male and female brain. Conversely, activational influences involve actions of gonadal steroids present throughout youth and adulthood. At this point, it is largely unclear what role organizational and/or activational hormonal influences have on many of the health risks in men and women, including harm from alcohol.

Alcohol abuse and dependence are prevalent health concerns in our society (Alcohol and Health, 2000). Epidemiological studies have shown that women appear to have a greater initial sensitivity to the effects of EtOH than men and generally have a lower prevalence of alcohol abuse and dependence compared with men (Alcohol and Health, 2000). However, women who abuse alcohol or become dependent on it tend to undergo pathophysiological consequences of excessive consumption more quickly and at smaller doses than men (Ashley et al., 1977; Schenker, 1997), including a greater impact on brain morphology and function (Mann et al., 1992, 2005; Pfefferbaum et al., 2001; Schweinsburg et al., 2003; Wang et al., 2003). An important question raised by these findings is to what extent do gender differences in alcohol abuse and dependence involve inherent neurobiological underpinnings. We hypothesize that innate sex differences in brain structure and function influence responses to chronic EtOH exposure and withdrawal.

Our lab research is currently focused on exploring sex differences in responses of EtOH withdrawal. There are many negative symptoms of EtOH withdrawal, including agitation, anxiety, insomnia, dysphoria, and tremors. In severe withdrawal, tremors can lead to seizures and these seizures may be life‐threatening. The severity and extent of withdrawal symptoms are believed to make a significant contribution to risk for relapse in recovering alcoholic individuals. Moreover, a recent study found that male alcoholic patients displayed more EtOH withdrawal symptoms than female alcoholic patients, even when matched for history of EtOH consumption (Deshmukh et al., 2003). These findings have led to our exploration of neurobiological contributions to sex differences in EtOH dependence and withdrawal.

We used an animal model of dependence and withdrawal, young adult Sprague‐Dawley rats. We now routinely use 3 sex conditions: males, intact females, and ovariectomized (OVX) females. Rats were made to become dependent on EtOH by administration of 6% EtOH in a nutritionally complete liquid diet for 14 to 16 days. Paired control animals were provided the same liquid diet, with dextrose substituted isocalorically for EtOH to provide equivalent caloric intake. To study EtOH withdrawal, the liquid diet was replaced with lab chow and animals were tested at various time points of withdrawal. Estrus cyclicity was monitored daily in the intact female rats by histological examination of vaginal smears. Test days were scheduled for when females are in estrus or diestrus I. In general, female rats tend to largely synchronize cycles within the time frame of the experiment and often show a prolonged estrus stage.

We used both behavioral and neurochemical assays to assess sex differences in the effects of EtOH withdrawal. These included behavioral tests of seizure risk in which we administered a convulsant drug, bicuculline or pentylenetetrazol, by the intravenous route (Devaud and Chadda, 2001; Devaud et al., 2003, 1995). Neurochemical assays included radioligand‐binding analysis, γ‐aminobutyric acid (GABA_A) receptor–mediated chloride flux measurements, and Western blotting focused on identification of changes in subunit levels of GABA_A receptors and the N‐methyl‐d‐aspartate (NMDA) subtype of glutamate receptors. Our experiments have targeted GABAergic and glutamatergic systems as extensive evidence shows a critical role of these 2 systems in mediation of responses to EtOH. Ethanol consumption tilts the balance toward central nervous system depression through enhancement of GABAergic neurotransmission and inhibition of glutamatergic neurotransmission (see Crews et al., 1996; Grobin et al., 1998, for review).

To date, we have found consistent sex and brain‐region selective differences in the expression of GABA_A and NMDA receptor subunit protein levels occurring with EtOH dependence (Devaud and Alele, 2004; Devaud and Morrow, 1999). Of special interest was the correlation between increased levels of GABA_A receptor α4 subunit protein in the cerebral cortex and the hippocampus with behavioral expression of EtOH dependence and withdrawal. The α4 subunit protein remained elevated in male, but not female, rats through 3 days of EtOH withdrawal. A number of investigators have provided evidence for associations between elevations in α4 subunit–containing GABA_A receptors and alterations in activation of this receptor, further implicating α4‐containing GABA_A receptors as having a role in increased anxiety and seizure risk (e.g., see Cagetti et al., 2003; Smith et al., 1998; Smith and Gong, 2004; Sundstrom‐Poromaa et al., 2002). There were also sex‐ and region‐selective alterations in subunit levels for NMDA receptors occurring with EtOH dependence, with the functional significance of these sex‐selective responses being less clear. Moreover, there were chronic EtOH‐induced changes in levels of various other proteins involved in GABAergic and glutamatergic transmission that were also sex‐ and region‐selective, providing further support of a neurobiological bases for sex differences in chronic EtOH‐induced alterations in neurotransmission (Alele and Devaud, 2005a; Devaud, 2001).

In earlier sets of behavioral studies, ovariectomy did not alter the consistently observed increased seizure susceptibility of EtOH withdrawal, with OVX females having patterns similar to those of intact females, for both control and EtOH‐withdrawn groups (Devaud et al., 2000). We also found that ovariectomy had minimal effects on chronic EtOH‐induced alterations in GABA_A and NMDA receptor subunit levels or in GABA and glutamate transporter levels or function (Devaud, 2001). This supports our hypothesis that innate differences in brain structure and function provide a neurobiological basis for observed sex differences in response to EtOH.

We continued to explore sex differences in EtOH withdrawal behaviors by measuring seizure risk at early (24 hours) or later (3 day) time points of withdrawal. Male and female rats showed a reliable reduction in seizure threshold (increased seizure susceptibility) at 24 hours of withdrawal compared with control susceptibilities. In contrast, there was a consistent and robust sex difference in seizure susceptibility at 3 days of withdrawal, with female rats appearing largely recovered but male rats continuing to show significantly increased seizure susceptibility. More recently, we observed significant sex differences in seizure activity following a bolus injection of pentylenetetrazol (Alele and Devaud, 2005b). There were important sex differences between control sex groups (males, intact females, and OVX females) as well as at 24 hours and 3 days of EtOH withdrawal. Consistent with earlier reports, we observed greater sensitivity to the anticonvulsant effects of the neuroactive steroid, pregnanolone, by female rats during early withdrawal (Devaud et al., 1995; Devaud and Morrow, 1999). Furthermore, intact female, but not male, rats displayed tolerance to the anticonvulsant effect of pregnanolone at 3 days of EtOH withdrawal. Male and OVX female rats remained sensitized at this time. These data suggest that activational influences of gonadal steroids also exert effects on responses to EtOH.

In summary, brain adaptations that occur during recovery from EtOH withdrawal are multifaceted. Our findings suggest that the precise sequelae of events vary by sex and arise from both organizational and activational influences. These findings provide additional support for an important role of GABAergic adaptations as part of the EtOH withdrawal process. It is clear that recovery from EtOH withdrawal involves more than a simple “resetting” back to basal conditions, highlighting the context‐dependent nature of responses to EtOH. The more we look, the more we find important sex differences in risk for and responses to a number of disorders, including alcoholism and alcohol withdrawal. Continued exploration of sex‐selective responses to pharmacological management of alcohol withdrawal should help identify new approaches for improved treatment according to gender and hormonal status.

Kimber L. Price and Lawrence D. Middaugh

The use of laboratory animals in experimental preclinical research may provide information on the impact of sexually divergent biological influences on both the physiological and rewarding effects of alcohol. Rodents have been used extensively in preclinical alcohol studies and have recently been examined to model the sex differences in alcohol use seen in humans. However, in addition to the sex influence, physiological responses to alcohol in rodents vary according to species and/or strain in ways that may or not correlate with human sex differences. For example, hepatic alcohol dehydrogenase (ADH) levels differ across 7 strains of mice, with varying degrees and direction of sexual dimorphism (Rao et al., 1997) and unlike what has been reported in humans (Baraona et al., 2001), female C57BL/6 (B6) mice exhibited higher ADH activity than their male counterparts (Rao et al., 1997). As has been shown for humans (Mumenthaler et al., 1999), female rats reportedly have higher blood EtOH levels (BELs) and faster elimination rates than males (Erickson, 1984); BEL differences reported for male and female mice are contradictory (Crippens et al., 1999; Hutchins et al., 1981; Middaugh et al., 1992; Seitz et al., 1992).

Behavioral responses to alcohol reward also vary according to sex, and the results fluctuate between species and experimental conditions as well. Human studies have shown that males typically drink more alcohol than females, but preclinical research has more frequently shown the opposite in rodents, with females drinking more than males. Since the use of rodents as subjects in preclinical alcohol studies is likely to continue, attention should be paid to the conditions that may influence the degree to which an animal model may mimic what is seen in humans. Experimental procedural variables likely contribute to whether (and in which direction) sex differences in EtOH‐related behavior are observed, and should therefore be considered when animal models are used. For instance, the influence of sex on EtOH consumption may be impacted by the following: (1) EtOH dose or concentration, (2) test duration, (3) circadian factors, (4) behavioral economy, and (5) motivational variables. Two common measures of EtOH reward have been used in our laboratory to examine the potential influence of sex on EtOH reward in C57BL/6 (B6) mice, free‐access EtOH consumption and oral EtOH operant reinforcement.

Kathleen A. Grant

Fundamental questions in the study of alcoholism and alcohol abuse revolve around the biological predisposition to drink excessive amounts of alcohol and the physiological adaptations that contribute to repeated intoxications. These questions are difficult to answer in humans for logistical, ethical, and financial reasons, as well as subject bias in self‐reporting alcohol consumption. Alternatives to these problems exist in a variety of animal models with predictive validity toward one correlate of alcoholism and utilize either invertebrates, mice, rats, gerbils, guinea pigs, pigeons, dogs, pigs, or monkeys. Nonhuman primates afford the opportunity to assess the relevance of findings from species such as rodents because nonhuman primate physiology and behavior more closely resemble human physiological responses to alcohol abuse and alcoholism. Macaque monkeys have many characteristics that make them useful for EtOH research. They have greater than 95% gene homology with humans, similar comparative anatomy and physiology, similar lengths and patterns of gonadotropin and sex steroid secretion, and similar absorption and metabolism of EtOH (see Grant and Bennett, 2003). In addition, macaques use complex social relationships and have the capacity for complex cognitive function. Perhaps most significant to alcohol research, macaques have the propensity to self‐administer large quantities of EtOH orally.

Investigations of oral EtOH self‐administration in nonhuman primates have revealed important parallels with human alcohol use and abuse. For example, between 15 and 20% of vervet and rhesus monkeys drink EtOH in amounts significantly greater than the amounts in their respective cohorts, with intakes exceeding 3 g/kg/d in monkeys identified as EtOH‐preferring (Ervin et al., 1990; Higley et al., 1996). Not surprisingly, BEC is correlated with EtOH intake, and the heaviest drinking monkeys attain BECs greater than 100 mg/dL (Vivian et al., 1999). These findings were obtained with monkeys in a variety of environmental contexts, suggesting that proportions of nonhuman primates are susceptible to biological factors, including genetics, which may result in heavy EtOH consumption. In addition, the proportion of monkeys that self‐administer excessive amounts of EtOH actually exceeds the 3% to 10% lifetime prevalence of alcoholism in humans (O'Brien, 1996), an observation that suggests that nonhuman primates may be particularly valuable in research on the susceptibility to alcoholism.

We have developed a method that exposes a population of male and female cynomolgus monkeys (Macaca fascicularis) to EtOH while allowing for extreme control over intake measures. A population of 9 adult male and 9 adult female monkeys were induced to self‐administer EtOH (4% w/v) as described in Vivian et al. The monkeys were induced to consume water and then EtOH (4% w/v) under a fixed‐time (FT) schedule of pellet delivery (i.e., schedule‐induced polydipsia; Falk, 1961) in 16‐hour sessions. The amount of EtOH induced each day increased in a stepwise fashion over 30‐day epochs. Specifically, monkeys were induced to drink EtOH 0.5 g/kg/d (2–3 drinks) for 30 consecutive days, 1 g/kg/day (4–5 drinks) for 30 consecutive days, and finally 1.5 g/kg day (6–7 drinks) for 30 consecutive days. The purpose of this regimen of EtOH exposure was to avoid conditioned taste aversions by limiting the amount of EtOH the monkeys could drink upon their initial exposures and to induce all animals to drink the same amount of alcohol in a similar amount of time. By inducing the consumption of EtOH over a short period of time (1–2 hours), the schedule also optimized the learned association between EtOH and the experience of intoxication.

After completion of the induction phase, the FT schedule of pellet delivery was discontinued. Concurrent EtOH and water became available continuously during daily 16‐hour sessions for 3 months. The daily sessions were then extended to 22 hours for 3 additional months. The monkeys were then placed in 12 months of abstinence and then allowed access to EtOH, water, and food for 22 h/d for 18 months. Thus, following the induction of EtOH self‐administration, the monkeys had a total of 24 months in which they could choose to drink unlimited amounts of EtOH or water each day. During the course of the experiments, sex and individual differences in EtOH intake became apparent. The males as a group consistently drank more EtOH than females, resulting in mean daily intakes of 2.6±0.3 g/kg/d (10–12 drinks) and 1.7±0.3 g/kg/d (approximately 7–8 drinks), respectively. Using the daily EtOH intakes, EtOH consumption was normally distributed in males. The female distribution of daily EtOH intake was positively skewed, with a greater likelihood of sessions with lower EtOH consumption (<1.0 g/kg/d; approximately 4 drinks).

Blood samples were taken 8 hours after the beginning of the session every fifth day during the 16‐ to 22‐hour sessions (over 500 samples were obtained and analyzed). Although EtOH intakes varied across monkeys, BEC was linearly and positively correlated with EtOH intake [sexes combined (r(505)=0.77, p<0.01)]. Similarly, in all but 1 monkey (monkey 5496), individual correlations of BEC and g/kg EtOH intake were significant (r² range 0.4–0.91; p<0.05). The BECs attained in the males were higher than in the female monkeys. It should be noted that BECs in the male population were frequently above 100 mg/mL and often in the 200 to 400 mg/dL range (Vivian et al., 2001). These data are, to our knowledge, the highest BEC levels consistently obtained in an oral EtOH self‐administration procedure with monkeys and highlight the probability that physiological and neurobiological adaptations will occur in response to these EtOH intakes.

Important to the documentation of sex differences in the self‐administration of EtOH and subsequent BEC measurements is the metabolic rate of EtOH by the cynomolgus monkey. We conducted an extensive study of both sex and menstrual cycle phase on the metabolism of EtOH in cynomolgus monkeys (Green et al., 1999). In this study, we measured peak EtOH concentrations (60 minutes postgavage) and EtOH elimination rates (over a 5‐hour period) in EtOH‐experienced male (n=4) and female (n=4) monkeys gavaged with 1.0 g/kg EtOH (15% w/v). There were no sex differences in peak BEC levels 60 minutes following gavage (male 86±2 mg/dL; female 82±5 mg/dL). These are similar levels as reported for social drinkers that range from 90 to 128 mg/dL after drinking 0.85 to 1.0 g/kg EtOH. The lack of a gender difference in peak BEC probably reflects the relatively lean body weights of our female monkeys (10%–15%) compared with women subjects, where body fat can range from 5% to 50%. As men are generally leaner than women and EtOH is more soluble in water than fat, EtOH would be expected to be a higher concentration in the blood (water compartment) in a woman when compared with a man.

In contrast to peak BEC, we did find that female monkeys have a faster EtOH elimination rate (34±2 mg/dL/h) than male monkeys (23±1 mg/dL/h). There are reports of women having faster EtOH elimination rates, particularly when doses are given that produce BEC of 80 to 90 mg/dL (reviewed in Green et al., 1999). Ethanol elimination rates in the male monkeys match very well the elimination rates in male alcoholic subjects (Majchrowicz and Mendelson, 1970), but in general the rates were higher in our monkeys than in nonalcoholic subjects. The ranges of elimination rates in human subject studies range from 8 to 36 mg/dL/h (Jones, 1993). This wide range indicates that EtOH elimination rates are highly variable in humans. Even when studied in the same individuals under the same experimental conditions, the correlation between test–retest elimination rates are poor (r=0.42–0.55) (Thomasson, 1995; Wilson and Erwin, 1983). Overall, the EtOH elimination rates found with cynomolgus monkeys are certainly within the range reported for human subjects.

One aspect of EtOH self‐administration that became apparent in the course of these experiments is the effect on the menstrual cycle. Menstrual cycle quality was assessed in female cynomolgus monkeys with serum progesterone levels taken every Monday, Wednesday, and Friday and by counting the length of the menstrual cycle based on daily swabs of the vaginal area. We determined luteal phase serum progesterone concentrations 3 times per week as an indicator of ovulation and cycle phase. Peak luteal phase progesterone concentrations less than 2.0 ng/mL indicated an anovulatory cycle, whereas peak luteal phase progesterone concentrations that fell between 2.0 and 4.0 ng/mL might be considered indicative of luteal phase deficiency, and progesterone concentrations above 5 ng/mL are indicative of ovulation (Wilks et al., 1979).

Ethanol self‐administration altered menstrual cycle quality, as reflected in a dose‐related suppression in luteal phase progesterone concentrations. Indeed, impairment was apparent after drinking 1.0 g/kg/d EtOH and anovulation, indicated by severe suppression of progesterone concentrations, was apparent with intakes over 1.5 g/kg EtOH. Progesterone concentrations and EtOH intakes were analyzed statistically by calculating the progesterone area under the curve (AUC) during a menstrual cycle and the corresponding EtOH intake (in g/kg) AUC for the corresponding day of the menstrual cycle. In the 8 monkeys over the 2‐year time frame we had corresponding data from 143 cycles (a maximum of 25 cycles in any single subject). We found that alcohol intake was highly associated with progesterone levels (AUC) with a p value of <0.0001 and also with an increase in cycle length (p<0.0003). We also investigated whether luteal phase progesterone sensitivity to EtOH during the induction phase was significantly related to menstrual cycle disruption during self‐administration following the induction phase. We found that while sensitivity on its own was not a significant predictor of decreased progesterone levels, there was a significant interaction between average alcohol intake and monkey sensitivity in addition to the significant main effect of average alcohol intake. Finally, the data analyses found that progesterone levels recovered after 9 months (3 months induction, 6 months open access) of EtOH self‐administration, when the monkeys were in 12 months of forced abstinence (p<0.02).

These disruptions in ovarian progesterone concentrations in heavier‐drinking females are consistent with disruptions of reproductive function observed under short‐term studies in rhesus monkeys and women (LaPaglia et al., 1997; Littleton, 1995; Mendelson and Mello, 1988). Furthermore, EtOH‐induced decreases in progesterone and estrogen would be predicted to have a negative impact on heart and cognitive function in addition to the direct effects of EtOH. Our model allows a unique characterization of these sex‐based effects of chronic EtOH self‐administration.