香京julia种子在线播放

Alcohol use disorder (AUD) represents one of the most prevalent and costly neuropsychiatric disorders globally and domestically, costing the United States economy an estimated $249 billion due to losses in workplace productivity, and health care and criminal justice expenses (Sacks et al., 2015). Withdrawal from alcohol (both acute and protracted withdrawal) produces a host of negative emotional conditions, such as depression and anxiety (Hershon, 1977; Becker, 2014; Smith et al., 2019). These emotional states can increase the risk for relapse and further hamper an individual’s ability to abstain from alcohol. Importantly, the literature suggests that risk of relapse may be different across the sexes, with women more likely to relapse following negative affect-related situations, and men more likely to relapse when in the presence of other alcohol consumers (White et al., 2015; Peltier et al., 2019). Complementary research suggests that alleviating depressive symptoms following abstinence from alcohol may improve treatment outcomes for women (Annis et al., 1998; Holzhauer and Gamble, 2017). Taken together, these data point to a complex relationship between AUD, depression, and treatment outcome, likely moderated by sex.

Previous research has shown that forced abstinence (FA) from alcohol produces depressive-like behavior and a variety of neurobiological changes in rodent models. For instance, Vranjkovic et al. (2018) demonstrated that 24-h immediately following 6 weeks of alcohol drinking, female C57BL/6J mice showed decreased time spent in the open arm of the elevated plus maze (EPM), and increased latency to feed in the novelty suppression of feeding test (NSFT), two commonly used models of anxiety-like behavior. In addition, this model produced increased immobility in the forced swim test (FST) during protracted withdrawal (21 days of abstinence), though it should be noted males were not investigated in these studies (Holleran et al., 2016; Vranjkovic et al., 2018). Similarly, Valdez and Harshberger (2012) demonstrated that male Wistar rats exposed to chronic alcohol show increased immobility in the FST, and that this effect is further enhanced during protracted withdrawal.

Alcohol use disorder and major depressive disorder (MDD) are highly comorbid disorders with overlapping etiology. Novel fast acting antidepressants are able to reduce both binge drinking (Crowley et al., 2019a) and depressive-like behavior following alcohol exposure (Holleran et al., 2016; Vranjkovic et al., 2018) further highlighting the potential overlapping neural circuitry involved in AUD and MDD. Importantly, both the clinical and preclinical depression literature has continuously pointed to a novel subpopulation of gamma aminobutyric acidergic (GABAergic) neurons as a protective, resiliency-conferring population. Somatostatin (SST) neurons are a group of GABAergic neurons that express and release the neuropeptide SST in the cortex, hippocampus, thalamus and amygdala (Fuchs et al., 2017; Abbas et al., 2018; Ahrens et al., 2018). Recent reports have implicated these neurons in a host of neuropsychiatric disorders in addition to MDD, such as bipolar disorder, anxiety disorder, and schizophrenia (Fee et al., 2017). Global genetic upregulation of SST neuronal function reduces depression- and anxiety- like behaviors in male and female mice (Fuchs et al., 2017), and deficits in SST expression are seen in the amygdala of postmortem samples of MDD patients (Douillard-Guilloux et al., 2017). Ketamine reverses the effects of chronic stress on GABAergic transmission in PFC pyramidal neurons by restoring the excitation/inhibition balance in SST neurons (Ghosal et al., 2020). Despite the clear correlation between SST neuronal markers and their function in MDD, both in the human and preclinical animal literature, thus far SST neurons have been poorly investigated in the context of AUD and depression-like phenotypes seen during withdrawal from alcohol.

The current study had two aims: first, to replicate previous behavioral models of forced abstinence from alcohol with both male and female mice. The second aim was to bridge the resiliency-like effect of SST neuronal function seen in the animal depression literature with the alcohol literature, in order to establish whether SST neurons within the prefrontal cortex (PFC), bed nucleus of the stria terminalis (BNST), basolateral amygdala (BLA) and lateral central amygdala (CEA), brain regions known for their role in MDD (Fee et al., 2017) and chronic alcohol exposure (Pleil et al., 2015a) are altered following forced abstinence from alcohol. These findings may shed new insights into the overlapping neural etiology of these highly comorbid disorders, particularly the SST neurons, and provide new directions to the discovery of more effective treatments.

Male and female C57BL/6J mice underwent either an alcohol ramp two-bottle choice procedure and forced abstinence (N = 11 males and 10 females) or control condition (N = 12 males and 10 females) (Figure 1A). On average, over the 6 weeks of alcohol exposure, male mice consumed significantly less alcohol per body weight as compared to female mice [F_time(5, 95) = 6.314, p < 0.001, F_sex(1, 19) = 7.060, p = 0.015, F_{sex × time}(5, 95) = 0.685, p = 0.635] (Figure 1B). Alcohol preference over water did not differ between male and female mice [F_time(5, 95) = 5.839, p < 0.001, F_sex(1, 19) = 0.041, p = 0.84, F_{sex × time}(5, 95) = 0.535, p = 0.749], and both sexes displayed a strong preference for alcohol over water [male: one-sample t(10) = 5.391, p < 0.001, female: one-sample t(9) = 7.102, p < 0.001] (Figure 1C).

The current study aimed to replicate previously published findings of abstinence-induced depression in both male and female rodents, and to understand the role SST neurons throughout the brain may play in this phenotype. Both male and female C57BL/6J mice showed an increased depressive-like behavioral profile following forced abstinence from alcohol, without aberrations in anxiety-like behaviors. Though both sexes showed a decrease in sucrose preference (a behavioral phenotype classically interpreted as anhedonia), only male mice showed an increased time spent immobile in the FST. Additionally, we observed plasticity in cortical and amygdalar areas of FA-exposed female mice in response to forced swim stress, including alterations in neuronal activation and intrinsic excitability of the GABAergic SST neurons in the PFC and vBNST. These findings highlight the efficacy of this alcohol exposure procedure to model protracted, forced abstinence-induced affective disturbances in both sexes, as well as shed insight into the circuit and cell type-specific adaptations.

Our results corroborate previous preclinical animal studies on sex differences in voluntary alcohol drinking (Almeida et al., 1998; Priddy et al., 2017; Crowley et al., 2019a; Peltier et al., 2019) in which female mice consistently consumed more alcohol than male mice across 6 weeks of access, despite both sexes showing similar alcohol preference over water (Figures 1B,C). In humans, while women historically tend to consume less alcohol than men (Rehm et al., 2010), this gap is rapidly closing with substantial increases in the prevalence of alcohol use and binge drinking among women and not men (White et al., 2015; Grucza et al., 2018). Women are also more susceptible to development of alcohol-associated neuropsychiatric disorders, including MDD, anxiety, post-traumatic stress disorder (PTSD) and stress-mediated relapse in alcohol use (Grant et al., 2017; Peltier et al., 2019). A lack of observable differences in anxiety-like behaviors in the EPM and OFT between control and FA-exposed mice in our study could indicate a transition from the heightened anxious state associated with acute withdrawal and early abstinence (Lee et al., 2015; Pleil et al., 2015a) to a more depressive-like state during protracted abstinence (Driessen, 2001; Stevenson et al., 2009; Heilig et al., 2010).

Previous studies using the same two-bottle choice paradigm demonstrated that female mice develop a heightened depressive-like behavioral phenotype during protracted abstinence from alcohol, as indicated by longer time spent immobile in the FST and longer latency to feed in the novelty-suppressed feeding test (NSFT), that can be alleviated by physical activity (Pang et al., 2013) or administration of the novel fast-acting antidepressant ketamine (Holleran et al., 2016; Vranjkovic et al., 2018). Interestingly, here we observed a more robust depressive-like behavioral phenotype in the FST in FA-exposed male mice, and not FA-exposed female mice. One hypothesis for these differences is that females transition to the abstinence-induced depressive-like state at a more rapid rate than males, such that heightened depressive-like behaviors in females may be observable at an earlier timepoint during abstinence. Alcohol-dependent men and women progress through courses of mood states at different rates during alcohol withdrawal (Bokström et al., 1991). Future experiments should assess the same depressive-like behavioral assays across multiple timepoints during protracted withdrawal. Another possibility is that male and female mice cope with acute stress and manifest depressive-like behaviors differently. Multiple studies in the literature (Kokras et al., 2015; Colom-Lapetina et al., 2017, 2019; Molendijk and de Kloet, 2019) have cautioned against over-interpretation of immobility duration in the FST, which classically has been interpreted as ‘behavioral despair’ or diminished motivation to escape a stressful environment (Porsolt et al., 1977). Recent work from Colom-Lapetina et al. (2017, 2019) suggests that female rats employ more active strategies to cope with the FST, including climbing and headshaking, while males employ the more classic strategy of immobility. Similar sex differences in coping strategy have also been found in conditioned fear response (Gruene et al., 2015a). Furthermore, we observed a sex difference in the frequency of open arm entries in the EPM (Figure 2B), where more entries into the open arm is classically interpreted as lower level of anxiety. Overall, our results suggest that the behavioral adaptations in response to acute swim stress following forced abstinence may be more nuanced in female mice, and future studies should employ a more fine-tuned behavioral analysis to fully elucidate the affective perturbations associated with alcohol exposure and abstinence.

The corticolimbic circuit comprises of highly interconnected regions, including the medial prefrontal cortex (mPFC) and the amygdala and its subregions, that are critically involved in the regulation of emotional behaviors, stress response and reward seeking (Koob, 2009; Koob and Volkow, 2016). The mPFC, including the prelimbic (PL) and infralimbic (IL) subregions, receives glutamatergic inputs from the BLA that can promote anxiety-like behaviors (Felix-Ortiz et al., 2016) and are modulated by chronic stress (Lowery-Gionta et al., 2018; Marcus et al., 2019). Acute withdrawal from alcohol exposure has been found to disrupt synaptic transmission and intrinsic excitability of pyramidal neurons of the mPFC at different stages of alcohol dependence; particularly, the dependence-inducing chronic intermittent ethanol (CIE) model enhances excitatory drive and intrinsic excitability in pyramidal neurons (Pleil et al., 2015b; Cannady et al., 2018), whereas the drinking-in-the-dark (DID) model of pre-dependence binge-like drinking impaired excitatory transmission (Crowley et al., 2019b). Recent evidence further demonstrates an interaction between sex and alcohol exposure on the PFC neurophysiology. Chronic gavage administration of high-dose ethanol enhances excitability of Martinotti neurons (presumably SST neurons) in male PL but decreases excitability of female Martinotti neurons via a reduction in hyperpolarization-induced cation channel (HCN) current (Hughes et al., 2020). Interestingly, this model of alcohol exposure disrupts inhibitory transmission in layer 5/6 pyramidal neurons in both sexes (Hughes et al., 2019), likely via an uncharacterized microcircuit mechanism. Excitatory transmission in PL layer 2/3 pyramidal neurons is decreased following four cycles of DID via distinct cellular mechanisms between the sexes (Crowley et al., 2019a), in which DID targeted cell-surface expression of NMDA and AMPA receptors alterations occur in female PL only. Sex also modulates the effects of stress on the cortical circuitry. Male rats with high level of freezing in fear conditioning have higher spine density in IL neurons that project to the BLA than male rats with low level of freezing (Gruene et al., 2015b), but the effect is not observable in females. Reciprocal projections from the BLA to the PL and IL are more severely affected by early life adversity in females than in males (Honeycutt et al., 2020). Chronic unpredictable stress induces sex-specific alterations in the transcriptomic profiles of SST neurons in the PFC (Girgenti et al., 2019). Here our FST-induced cFos expression data revealed that forced abstinence from alcohol dampened neuronal activation in the female PFC (Figure 3B). Hypoactivation of the mPFC in response to stress and subsequent dysregulation of amygdala activity have indeed been observed in humans with MDD and alcoholism (Johnstone et al., 2007; Covington et al., 2010; Seo et al., 2013), as well as rodents in chronic social defeat (Covington et al., 2010) and chronic unpredictable stress (Lam et al., 2018). Withdrawal from CIE exposure similarly reduces cFos expression in the PL and IL of male mice (Smith et al., 2019). Our model, which combines both stress (potential hypoactivation of PFC) and alcohol (potential hyperactivation) may result in more nuanced alterations in this region.

The downregulation of neuronal activation in the PFC may have resulted from augmented inhibitory tone from the GABAergic SST neurons, as we identified an increase in intrinsic excitability of these neurons (Figures 3H, 6). In the cortex, SST neurons are often postulated to be the gatekeeper of thalamo- and cortico-cortical excitatory inputs to pyramidal neurons, via their soma- and dendrite-targeting synapses (Urban-Ciecko and Barth, 2016). These neurons recently gained substantial interest for their role in a host of neuropsychiatric disorders, including MDD, bipolar disorder, and schizophrenia (Fee et al., 2017; Pantazopoulos et al., 2017; Abbas et al., 2018). Global disinhibition of SST neurons via deletion of GABA_A receptors expressing the Υ₂ subunit produces an antidepressant-like behavioral and neural profile (Fuchs et al., 2017). Acute, chemogenetically induced inhibition of SST neurons in the mPFC is anxiogenic (Soumier and Sibille, 2014), pointing to an overall resiliency-conferring role of SST neurons in the cortex. In our study, it is unclear whether the hyperexcitability of SST neurons following forced abstinence from alcohol directly contributes to the depressive-like states, or is in fact a compensating mechanism to counteract the alcohol-induced adaptations. The latter interpretation, that SST hyperexcitability may be in response to alcohol-induced adaptations, is further supported by the opposing results seen in the vBNST. Future studies should further examine whether SST neurons in the PFC has a causal role in alcohol-related pathological behavioral adaptations.

The extended amygdala, including the BNST and the CeA, is critically involved in regulation of alcohol consumption and withdrawal-associated negative affect (Koob, 2009; Koob and Volkow, 2016; Vranjkovic et al., 2017; Torruella-Suárez et al., 2020). As in the mPFC, withdrawal from both the CIE and DID models modifies synaptic transmission and excitability in the extended amygdala, with heavy emphasis on hyperexcitability in the vBNST and enhanced inhibition of the CeA (Roberto et al., 2010; Lee et al., 2015; Pleil et al., 2015a; Smith et al., 2019). In accordance with these results, our FST-induced cFos expression data demonstrated elevated neuronal activation in the vBNST of FA-exposed female mice (Figure 4D). SST neurons in the vBNST were also highly activated in response to FST in FA-exposed female, whereas this activation was very minimal in all other groups (Figure 4E). Contrastingly, FA dampens intrinsic excitability in SST neurons of female mice, with fewer numbers of action potentials in response to depolarizing step currents (Figure 4J), which could reflect a compensating mechanism to counteract the heightened engagement of vBNST SST neurons following FST. These data suggest that forced abstinence from alcohol may remodel the engagement of the stress-responsive networks in the vBNST, including SST neurons (Figure 6).

The circuit organization and functionality of these neuronal populations in the extended amygdala are not fully characterized. There is evidence suggesting that SST neurons in the lateral CeA directly synapse onto SST neurons in the oval nucleus of the BNST (Ahrens et al., 2018). Enhancing excitatory drive on the CeA SST neurons paradoxically disinhibits SST neurons in the oval nucleus in a kappa opioid receptor dependent manner to promote anxiety (Ahrens et al., 2018). As we did not observe any changes in anxiety-like behaviors, or any SST neuron-specific plasticity in the lateral CeA and the dorsal BNST, it is possible that the neuroadaptations in these anxiety-promoting populations may be transient and occur during acute withdrawal, subsiding before transition to a depressive-like state. The otherwise robust plasticity in the vBNST SST neurons may implicate a functional distinction between the dorsal and ventral subregions of the BNST. In addition to a dense population of corticotropin-releasing factor (CRF) neurons that are anxiogenic and alcohol drinking-promoting (Pleil et al., 2015b), the vBNST includes an anxiolytic population of GABAergic neurons that project to the ventral tegmental area and the lateral hypothalamus (Marcinkiewcz et al., 2016; Pati et al., 2019). Withdrawal from CIE exposure augments excitability of CRF neurons and increases inhibition onto the anxiolytic midbrain-projecting neurons (Pati et al., 2019). Forced abstinence-induced decrease in intrinsic excitability of SST neurons in the vBNST may result in disinhibition of the CRF neurons to exacerbate negative affect and heighten the risk of stress-induced relapse. Nevertheless, there is no current evidence of direct synaptic input from SST neurons to CRF neurons or midbrain-projecting neurons in the vBNST. Further complicating the matter is the fact that the BNST organization and functionality in females have not been examined to the same degree as in males. In adult men, somatostatin and vasoactive intestinal peptide (VIP) immunostaining reveal larger BNST volume than in adult women (Chung et al., 2002). In rats, female BNST SST neurons do not express estrogen receptors and androgen receptors that are prevalent in male SST neurons (Herbison and Theodosis, 1993; Fernández-Guasti et al., 2000), suggesting that the FA-induced behavioral and neuronal adaptations in females might not be modulated by gonadal hormones. Studies examining the sex-specific properties of the BNST in regulation of emotions and reward seeking, and sex differences in AUD-related neuroadaptations in the BNST, are lacking. Future studies should examine the input and output pathways of these SST populations, as well as their functional role in regulation of alcohol drinking and emotional behaviors with an eye toward sex-specific properties.

Notably, we observed a consistent sex effect in total cFos expression the PFC and the dorsal and ventral subregions of the BNST, in which male mice showed markedly higher level of FST-induced cFos expression (Figures 3B, 4B,D). Given the fact that the majority of the previously cited studies on alcohol-induced neuroplasticity were performed in males, it is interesting to see that FA-exposed male mice seem to be relatively resilient against protracted forced abstinence-induced alterations. One possibility is that, since male mice consumed less alcohol than female mice in the two-bottle choice paradigm when normalized to body weight, the effect of forced abstinence from alcohol in male mice may be too subtle to be detected by the methods employed here. Additionally, as discussed with the sex differences in coping strategies in response to forced swim stress, passive coping in male mice may engage a different circuit and cell-type specificity that could be overlooked. These observations further highlight the importance of sex as a biological variate in neurobiological studies in order to have a better understanding of the pathophysiology underlying neuropsychiatric disorders.

We did not observe any changes in SST protein expression in any cortical or amygdalar regions. We have previously reported that SST neurons in the PL cortex release SST peptide tonically and phasically in a frequency-dependent manner (Dao et al., 2019), hence altered excitability of these neurons in the PFC and the BNST may affect their capacity for neuropeptide release. The SST peptide acts on a family of G protein-coupled receptors, that are mainly inhibitory through Gi/o-dependent signaling (Barnett, 2003). However, SST receptor signaling in these regions has not been characterized. Future studies may merit from examination of the modulatory action of SST peptide, in healthy states and alcohol abuse-associated conditions.

The current study identified a host of sex-dependent behavioral adaptations and neuroplasticity in corticolimbic regions that may underlie forced abstinence-induced affective perturbations. SST neurons in the PFC and vBNST emerge as a strong neural candidate that undergo robust plasticity following alcohol exposure and protracted abstinence. These results shed new insight into the neurobiological bases of highly comorbid neuropsychiatric diseases, including MDD and AUDs, and may aid in the development of new and effective treatments.

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 the Pennsylvania State University Institutional Animal Care and Use Committee.

ND and NC conceived and designed the experiments, analyzed and interpreted the experiments, and wrote the manuscript with feedback from all authors. ND, MS, SM, JM, VS, and DB conducted the experiments.

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.