Alcohol is the most commonly used substance of abuse worldwide, and about 2.3 billion people aged 15–16years old have used this substance at least once in their life (WHO, 2018). Harmful use of alcohol brings many social and economic impairments to individuals (UNODC, 2020). Moreover, this harmful use is considered one of the most important risk factors for public health, as it is estimated that 3 million (5.3%) global deaths are attributed to alcohol intake (WHO, 2018).

Alcohol use disorders (AUD) are among the most prevalent mental disorders (American Psychiatric Association, 2013; Rhem and Shield, 2019). It is characterized by an inappropriate pattern of alcohol use, in which individuals exhibit compulsive heavy alcohol consumption, and loss of control over alcohol intake, despite adverse consequences involving their health, family, social life, and occupational activities (American Psychiatric Association, 2013; Connor et al., 2016; Carvalho et al., 2019). Epidemiological studies have shown that AUD is more prevalent in men compared to women (WHO, 2018; Rhem and Shield, 2019). In this sense, studies have shown that 237 million men and 46 million women around the world are diagnosed with AUD (WHO, 2018).

Stress is considered an important environmental risk factor for the development of alcohol and other substance use disorders, such as cocaine, amphetamine, and opiates (Piazza and Le Moal, 1998; Sinha, 2008; Camarini et al., 2018; Mukhara et al., 2018). Stress may be related to initiation, maintenance, escalation, and relapse of drug use (Piazza and Le Moal, 1998; Sinha, 2008; Camarini et al., 2018; Mukhara et al., 2018). However, the relationship between stress exposure and alcohol intake is complex and poorly understood (Becker et al., 2011; Noori et al., 2014).

Newborn rodents depend on maternal care in order to achieve appropriate nutrition, thermoregulation, and protection against predators. In addition, it can provide long-lasting effects on the brain, and decreased reactivity to stress (Novick et al., 2018). In the MS protocol, pups are separated daily from their dam, usually from post-natal day (PND) 1 to 14, for short or long periods (Cruz et al., 2008; Novick et al., 2018). Prolonged periods of MS (180min) have been shown to be an intense stressor for the litter, and can negatively impact maternal care during behavioral, physiological, and neuroendocrine maturation (Levine, 1967).

Several studies have demonstrated that MS stress in rodents increased alcohol consumption during adulthood, in different paradigms, such as two-bottle choice and operant self-administration (Huot et al., 2001; Ploj et al., 2003; Roman and Nylander, 2005; Kawakami et al., 2007; Cruz et al., 2008; Portero-Tresserra et al., 2018). In addition, male mice submitted to long periods of MS stress were more susceptible to the effects of chronic alcohol administration on plasma corticosterone levels (Kawakami et al., 2007).

The MS stress seems to produce sex-dependent effects. Roman et al. (2005) showed that alcohol intake of stressed male ethanol-preferring Alko Alcohol (AA) rats was higher than the control group consumption. However, stressed female AA rats had a reduced alcohol intake and preference relative to controls. A study conducted by Kawakami et al. (2007) also showed that prolonged maternal separation (MS) induced a faster behavioral sensitization in females compared to male mice.

Extended AMY, which consists of the Nucleus Accumbens shell (NAcc shell), Central Amygdala (CeA), and Bed Nucleus of Stria Terminalis (BNST), is complex neural circuitry that is known especially for its role in negative affects during ethanol withdrawal (Koob and Volkow, 2010; Koob, 2013). Studies show that the extended amygdala is involved in the modulation of stress responses and AUD (for review, see Centanni et al., 2019). For instance, it has been demonstrated that an acute alcohol systemic injection increases CeA and BNST c-fos expression (Sharko et al., 2016; Saalfield and Spear, 2019). In the same sense, Leriche et al. (2008) also demonstrated that intragastric alcohol administration was able to augment c-fos expression in subregions of NAcc, AMY and BNST. Furthermore, Anstrom et al. (2009) showed that social defeat stress increased dopamine release in NAcc of defeated rats. Also, social defeat stress increased c-Fos expression in CeA and BNST of mice (Numa et al., 2019). In addition, lesions in CeA were able to alter behavioral and autonomic responses in stress condition (Roozendaal et al., 1991). Davis et al. (2010) suggested that the CeA is activated and can rapidly elicit stress-related responses immediately after a stressful event is detected. However, the BNST is activates in situations of chronic stress, fear, and anxiety.

Moreover, MS induced higher c-fos levels in BNST, BLA, CeA, and other limbic regions (Horii-Hayashi et al., 2012; Aguggia et al., 2013; Nishi et al., 2013). Interestingly, Jahng et al. (2010) showed that rats exposed to early-life MS presented a blunted accumbal Fos immunoreactivity response to restraint stress relative to controls. However, it remains unclear whether alcohol-induced neuronal activity in extended amygdala regions may differ between subjects that are exposed to stress in early-life and those that are not exposed, in a sex-dependent manner.

Taken together, these data show a complex relation between environmental experiences (including a history of early life stress), and biological factors, such as sex, on vulnerability to AUD. Although evidence points to the influence of MS stress on AUD development, the neurobiological mechanisms underlying this interaction remain unclear. Thus, we aimed to investigate the long-term effects of prolonged MS stress on alcohol-related behaviors in male and female adult mice. In addition, we evaluated the role of extended amygdala sub-nuclei in this phenomenon via the analysis of neuronal activity in those areas, in the different experimental groups. We hypothesized that MS would increase mice ethanol consumption and Fos-immunoreactive neurons (Fos-IR) in extended amygdala in a sex-dependent manner.

Given the importance of environmental experiences and biological factors as predictors of ethanol consumption, we investigated the influence of MS stress on ethanol consumption in female and male adult mice, and the participation of the extended amygdala in this interaction.

We showed that during involuntary consumption, MS increased ethanol intake in female mice, while it decreased consumption in males. In addition, regardless of sex, animals submitted to early life stress were more motivated to seek ethanol in a progressive-ratio schedule of operant self-administration procedure. Moreover, female controls, as well as male MS mice, were more resistant to decreasing their consumption relative to male control mice during the quinine adulteration protocol. For immunohistochemistry, we observed an increase of Fos-immunoreactive neurons in the BLA and CeA in MS mice, but not in NAcc, nor in BNST.

Interestingly, during the involuntary consumption paradigm, female controls and MS males presented a similar pattern of consumption, showing a decrease in consumption during the second and third weeks of the procedure. In contrast, male controls exhibited a linear consumption pattern over the weeks, while MS females increased their intake during the third week relative to the second week. Although no statistical differences were found between MS and control or male and female groups, we observed a possible effect of MS stress on ethanol consumption in a sex-dependent model, as the four experimental groups presented distinct ethanol intake patterns over the weeks.

Concerning human clinical studies, women and men differ significantly in the development of AUD (Haas and Peters, 2000). Although women initiate alcohol use later in life than men, they show progress faster from drug use to abuse, especially when using psychostimulants, such as cocaine (Haas and Peters, 2000; Castro-Zavala et al., 2020).

Studies suggest that females are more susceptible to stress-potentiated ethanol consumption, to alleviate the negative stress effects and reactivity (Armstrong et al., 2018; Peltier et al., 2019). In addition, Kawakami et al. (2007) have demonstrated that after chronic exposure to ethanol, female mice submitted to long periods of MS showed faster development of behavioral sensitization when compared to male mice. In addition, Lancaster (1998) reported that female rats submitted to short periods of MS, from PND 1 to 7, increased ethanol intake. These findings agree with our results showing that stressed female mice increased their consumption over time. On the other hand, Castro-Zavala et al. (2020) demonstrated that female rats submitted to early-life stress decreased self-administration of a psychostimulant drug in adulthood.

In contrast with many studies that reported increased ethanol intake of male rodents after exposure to different stressors (Nash and Maickel, 1985; Lynch et al., 1999; Wille-Bille et al., 2017; Newman et al., 2018), including MS stress in which male rats were submitted to 360min MS stress, from PND 1 to PND 21 (Roman et al., 2005), we observed a decrease in ethanol intake of male MS mice. This kind of stress may have induced an anhedonic effect in male mice, which could lead to reduced intake. Willner et al. (1992) suggested that chronic exposure to a variety of mild stressors may cause a decreased responsiveness to reward, when rats were exposed to a sequence of different and unpredictable stressors from 5 to 9weeks, reducing the consumption and preference for saccharin or sucrose solutions (Willner et al., 1987). In the same way, Huot et al. (2001) showed that maternally-separated rats exhibited anhedonia, as they avoided sucrose solution.

Regarding the self-administration protocol in fixed-ratio schedules, we did not observe a MS effect for male or female mice. To the best of our knowledge, there are few studies using the ethanol self-administration approach in mice. For instance, Cruz et al. (2008) demonstrated that maternally-separated male mice separated from their dams during 180min from PND 1 to PND 14, presented an increased consumption of 6 and 10% ethanol+saccharin solutions. Although these data differ from ours, it is important to highlight at least three major protocol differences between the Cruz study and ours: (i) unlike that study, our protocol included a 3-week-ethanol exposure prior to operant tasks. (ii) Cruz et al. (2008) added saccharin to ethanol solutions to augment response levels, which may have increased the hedonic value assigned to the substance, leading to a different outcome. (iii) In their study they used Swiss mice, while in ours we used C57BL/6J, which have shown a higher ethanol intake compared to other strains (Rhodes et al., 2007). Therefore, the control and MS mice could have reached the ceiling effect of consumption and this could have impaired the MS stress effects.

We did not detect a sex effect, when comparing male vs. female MS groups. In agreement with our findings, Logrip and Gainey (2020) showed that previous footshock stress did not alter the acquisition of ethanol self-administration behavior of male and female rats on FR1 and FR3 schedules. Conversely, female rats exposed to predator odor stress increased responses to ethanol self-administration in comparison to stressed male rats (Ornelas et al., 2021). Regarding undisturbed control groups, we also did not observe a sex effect. In contrast, in a FR4 schedule, male rats exhibited a higher ethanol self-administration in comparison to females (Lorrai et al., 2019). In addition, Randall et al. (2017) showed that male rats presented increased operant responses for ethanol+sucrose solution relative to female rats.

The progressive-ratio schedule has been widely used to evaluate motivation and goal-directed behavior toward rewards (Randall et al., 2017). Our data reveal that regardless of sex, MS increased the motivation to take ethanol reinforcement, considering the number of active nose pokes.

Few studies have evaluated early life stress effects on motivation for ethanol, especially in mice, and the data is contradictory in the literature. Fosnocht et al. (2019) demonstrated that prolonged social isolation during adolescence increased the motivation of male and female adult mice to seek cocaine, in a progressive-ratio schedule. However, Wistar rats submitted to early pup-litter-dam deprivation exhibited reduced motivation to obtain sucrose in the same paradigm (Rüedi-Bettschen et al., 2005). Based on our data, we suggest that MS altered the motivation of mice to seek ethanol, but not necessarily hedonic value of this substance under a self-administration paradigm, since both MS and control groups presented similar numbers of reinforcements and active nose poke responses during FR tests. Furthermore, progressive-ratio experiments assessing the influence of MS on motivation toward other drugs of abuse and natural reinforcers, such as sucrose, would contribute to a better understanding of this effect.

During the binge test, we did not observe sex differences in relation to the number of reinforcements or active responses. However, Finn et al. (2018) showed that intermittent predator odor stress exposure increased ethanol binge drinking in male mice, but did not alter drug consumption in females. On the other hand, Sneddon et al. (2018) demonstrated that female mice increased binge consumption, while males did not. It is important to highlight that in these studies, binge drinking was not evaluated through an operant behavior, but by measuring ethanol consumption in home cages. Also, mice were exposed to ethanol consumption for 7weeks in our study, to acquire the operant behavior while in both the Finn and Sneddon studies, the animals were submitted to ethanol intake for up to 3weeks.

In regard to stress effect, we also did not observe differences between MS and control groups. Nevertheless, Gondré-Lewis et al. (2016) showed that MS stress increased response to ethanol during a binge session of self-administration protocol in rats. However, there are some aspects of Gondré-Lewis protocol that are worth mentioning: the multiple consecutive binge sessions (with no previous abstinence) were constituted by two interspersed 45min-periods, and sucrose was added to the ethanol solution. In comparison, in the present study the single binge session lasted 4h and followed a 2-day abstinence period, which we hypothesized may have influenced the control groups intake.

In the quinine adulteration protocol, we observed that female control mice presented increased ethanol intake relative to male controls for 0.005 and 0.01g/L quinine concentrations. This result suggests that females are more resistant to extinguishing or decreasing ethanol-seeking behavior than male mice when presented with an aversive stimulus.

Consistent with our findings, Sneddon et al. (2020) demonstrated that in an operant self-administration paradigm, male mice reduced their response to adulterated ethanol solution in high quinine concentrations, but female mice did not exhibit changes in their response to adulterated ethanol at any concentration. Furthermore, Fulenwider et al. (2019) showed that a higher concentration of quinine was required to suppress the ethanol consumption of female mice compared to male mice. In addition, the Sneddon group (2018) showed that with the two-bottle choice of access in the “Drinking in the Dark” protocol, female mice responded more to quinine-adulterated ethanol solution than male mice, but high concentrations of quinine suppressed the consumption of both male and female mice.

In relation to MS stress, we observed that only male mice exhibited increased adulterated ethanol solution intake relative to the male control group. In this sense, male mice subjected to chronic predator-exposure stress consumed more quinine-adulterated ethanol solution than non-stressed animals, and they only exhibited a reduction in ethanol intake when quinine reached the highest concentration (Shaw et al., 2020). Nevertheless, female rats submitted to acute early life stress (footshock) tolerated higher quinine concentrations compared to their control group. Male control rats were more sensitive to quinine taste, and the consumption of the stressed male group decreased only with the highest quinine concentration (Radke et al., 2019).

Results from this trial may suggest that the male MS group are more resistant to bitter quinine taste than its control group. Also, the responses to the aversive taste of quinine seems to be related to a number of experimental variables, such as type and duration of stress, quinine concentration, and the species (rats vs. mice).

Apparently, this is the first study to evaluate extended amygdala neuronal activation of mice that have been chronically exposed to both MS and ethanol in a sex-dependent model.

As described in the immunohistochemistry procedure, male and female mice were submitted to a last operant self-administration session before brain collection. This session followed the quinine adulteration protocol, and animals were allowed to consume 20% ethanol after a period of an aversive stimulus. In this sense, this last operant session may have contributed to Fos protein expression in neurons of the NAcc core and shell of male and female mice from MS and control groups.

Our results did not show any differences in Fos-IR cells of NAcc core and shell subregions among MS and control mice groups (male or female). In the same sense, Saalfield and Spear (2019) observed that after administration of two different concentrations of ethanol, adult rats did not show any differences of Fos positive cells in NAcc core. However, Leriche et al. (2008) showed that ethanol intragastric administration increased the number of Fos-immunoreactive neurons in NAcc shell of rats. Ethanol intraperitonial injection also increased the number of Fos positive cells in this same brain region (Herring et al., 2006). Further, Ryabinin et al. (1999) observed an increased number of Fos-IR neurons in the NAcc Core, but not in the NAcc shell in mice submitted to restraint stress after ethanol consumption. It is important to highlight that in these studies, ethanol administration was not evaluated through operant behavior, such in the present study. Moreover, our study assessed Fos expression 24h after the quinine adulteration protocol, an aversive stimulus. This may also have contributed to difference in the results.

We did not observe any differences in Fos protein expression in neurons of medial- and lateral-BNST subdivisions in female and male mice of control or MS groups. However, Ryabinin et al. (1999) showed that restraint stress increased the number of Fos+ cells in the BNST of mice chronically exposed to ethanol. In fact, the stressful stimulus exposure occurred after ethanol consumption, and not during the early life of the mice.

Although, we did not find any differences in neurons expressing Fos protein between MS and control groups in the BNST, we observed increased neuronal activation in the Central and Basolateral Amygdala.

In the BLA, we observed the effect of stress, and the Fos protein expression was higher in MS groups when compared to control groups. In agreement with our results, it has been demonstrated that MS stress increases the number of Fos+ cells in the BLA of male mice (Horii-Hayashi et al., 2012; Nishi et al., 2013). However, this increased Fos expression occurred on PND 14, after the final separation, but not during adulthood.

In the Central Amygdala, we observed that Fos expression of female mice was higher than in the male group. This may be due to the increased sensitivity to ethanol inhibition that male mice CeA neurons show compared to female mice CeA neurons (Logrip et al., 2017).

We also found that the number of Fos-IR neurons in the CeA of MS groups was higher than in control groups. An important integrative role between stress- and addiction-related behaviors has been attributed to the CeA (Gilpin, 2012). In relation to this, Horii-Hayashi et al. (2012) demonstrated that MS stress exposure alters CeA activation in male mice and increases Fos expression in this brain region. On the other hand, this amygdalar nucleus is also involved in ethanol response behaviors. For instance, acute ethanol exposure can increase the number of Fos-immunoreactive neurons in the CeA (Leriche et al., 2008; Sharko et al., 2016). In addition, De Guglielmo et al. (2016) demonstrated that the inactivation of a CeA neuronal ensemble during abstinence decreased ethanol consumption of dependent and non-dependent rats.

Therefore, we hypothesize that MS stress could induce neuroadaptations in the CeA, which are related to increased neuronal activation and ethanol consumption. Nonetheless, future studies assessing the effects of activation or inhibition of CeA on ethanol consumption of animals exposed to early life stress, such as MS are needed to corroborate our hypothesis.

As future perspectives, studies aimed at investigating which CeA neuroadaptations are involved in both stress and ethanol consumption responses will help to understand this complex interaction better.

In conclusion, MS stress produces long-term alterations in the ethanol consumption of adult mice and BLA and CeA activation patterns, depending on reinforcement protocol and sex.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Comissão de Ética no Uso de Animais (CEUA/UNIFESP) n. 5360240918.

FC, TM, and NB contributed to the conception and design of the study. NB, CF, PP, TY, TR, and BT performed the experiments. NB and CL performed the statistical analysis. NB wrote the first draft of the manuscript. FC, TM, RL, CF, CL, and NB contributed to manuscript revision and read and approved the submitted version. All authors contributed to the article and approved the submitted version.

The study was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). CL is recipient of Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) research fellowship through Instituto Nacional de Ciência e Tecnologia Translacional em Medicina (INCT-TM), Brazil. TY and TR are recipient of Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) research fellowship. NB is recipient of Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) fellowship (process number 2020/04389-3). FC is recipient of a FAPESP grant number 2018/15505-5.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.