mTORC1 pathway is involved in the kappa opioid receptor activation-induced increase in excessive alcohol drinking in mice

Abstract

KOP-r agonist U50,488H produces strong aversion and anxiety/depression-like behaviors that enhance alcohol intake and promote alcohol seeking and relapse-like drinking in rodents. Mammalian target of rapamycin complex 1 (mTORC1) pathway in mouse striatum is highly involved in excessive alcohol intake and seeking, and in the U50,488H-induced conditioned place aversion. Therefore, we hypothesized that KOP-r activation increases alcohol consumption through the mTORC1 activation. This study focuses on: (1) how chronic excessive alcohol drinking (4-day drinking-in-the-dark paradigm followed by 3-week chronic intermittent access drinking paradigm [two-bottle choice, 24-h access every other day]) affected nuclear transcript levels of the mTORC1 pathway genes in mouse nucleus accumbens shell (NAcs), using transcriptome-wide RNA sequencing analysis; and (2) whether selective mTORC1 inhibitor rapamycin could alter excessive alcohol drinking and prevent U50,488H-promoted alcohol intake. Thirteen nuclear transcripts of mTORC1 pathway genes showed significant up-regulation in the NAcs, with two genes down-regulated, after excessive alcohol drinking, suggesting the mTORC1 pathway was profoundly disrupted. Single administration of rapamycin decreased alcohol drinking in a dose-dependent manner. U50,488H increased alcohol drinking, and pretreatment with rapamycin, at a dose lower than effective doses, blocked the U50,488H-promoted alcohol intake in a dose-dependent manner, indicating a mTORC1-mediated mechanism. Our results provide supportive and direct evidence relevant to the transcriptional profiling of the critical mTORC1 genes in mouse NAc shell: with functional and pharmacological effects of rapamycin, altered nuclear transcripts in the mTORC1 signaling pathway after excessive alcohol drinking may contribute to increased alcohol intake triggered by KOP-r activation.

INTRODUCTION

Endogenous opioid systems are profoundly affected after excessive alcohol consumption. Specifically, the kappa opioid receptor (KOP-r)/dynorphin system has been found to be activated after chronic alcohol exposure, which is involved in the negative reinforcing aspects of alcohol addiction in rodents [Koob and Kreek 2007; Anderson and Becker 2017]. KOP-r agonists or antagonists alter alcohol drinking and alcohol-related behaviors in rodent models [Lindholm et al 2007; Rose et al 2016; Zhou and Kreek 2019a]. Alcohol exposure increases dopamine release in the nucleus accumbens (NAc), leading to rewarding effects, and KOP-r agonists oppose these effects by decreasing the dopamine release [Spanagel et al 1990; Lindholm et al 2007], suggesting an interaction between KOP-r activation and dopamine in alcohol-induced reward. In both humans and rodents, neurobiological studies have provided strong supportive results, demonstrating that among multiple actions of alcohol in the CNS, KOP-r activities are profoundly affected by chronic alcohol exposure [Logrip et al 2008; Sperling et al 2010; Rácz et al 2013; Kissler et al 2014; de Laat et al 2019]. Together, these findings have suggested that the KOP-r systems play an important role in the developments of alcohol excessive consumption and addiction.

Rapamycin, a mammalian target of rapamycin complex 1 (mTORC1) inhibitor, has immunosuppressant functions in humans and has been recently examined in rodent alcohol studies [Ron and Berger 2018]. For many years, rapamycin has been found to decrease alcohol-induced place preference, excessive alcohol drinking and alcohol seeking behaviors in several rodent models [Neasta et al 2010; Ron and Berger 2018]. Pharmacologically, intra-NAc inhibition of the mTORC1 by rapamycin decreases alcohol drinking [Cozzoli et al 2016; Laguesse et al 2017a]. Using an unbiased RNA sequencing (RNA-seq) approach, it has been further characterized that the mTORC1 (functioning as a nutrient/energy/redox sensor and controlling protein synthesis) in mouse NAc is highly involved in alcohol intake, seeking and reward [Laguesse et al 2017b]. As altered mRNA changes in the nuclear compartment are more direct and sensitive to the gene transcriptional activity than those in the cytoplasmic compartment [Mitchell et al 2012; Grindberg et al 2013; Zhou and Lapingo 2014], the present study analysed gene transcripts using nuclear RNAs, to determine transcriptional alterations in mouse NAc shell (NAcs). The changes in molecular profiling of the mTORC1 and other signaling pathways in the NAcs further provided a tool to generate an assumption for the mTORC1 pathway that could lead to the KOP-r-related drinking behavioral response.

A recent study has found that U50,488H activated the mTORC1 pathway in mouse striatum, and the rapamycin-induced inhibition of the mTORC1 pathway blocked U50,488H-induced aversion in a conditioned place aversion model, without altering other behaviors in mice [Liu et al 2019]. Most KOP-r agonists like U50,488H and U69,593 cause strong sedation, aversion and dysphoria in both humans and rodents [Walsh et al 2001; Bruchas and Chavkin 2010; Chavkin and Koob 2016; Zhou and Kreek 2018]. KOP-r agonists have been found to enhance alcohol drinking in alcohol-dependent mice [Rose et al 2016], induce alcohol-seeking behavior in rats [Funk et al 2019] and promote alcohol relapse-like drinking in rats [Holter et al 2000]. Pharmacological blockade of KOP-r decreases alcohol seeking or drinking in rodents [Walker and Koob 2008; Kissler et al 2014; Domi et al 2018], as well as anxiety/depression-like behaviors [Anderson and Becker 2017].

Based on these findings, we proposed a novel hypothesis that KOP-r agonists increase alcohol consumption through the mTORC1 activation, contributing to negative reinforcing mechanisms. Our 1^st objective was to perform high-throughput RNA-Seq of total nuclear RNAs for comprehensive molecular profiling of the mouse NAcs to identify potential gene expression changes in the mTORC1 pathway after chronic excessive alcohol drinking. In the present study, mice accessed to voluntary alcohol drinking in a 4-day drinking-in-the-dark paradigm followed by a 3-week chronic intermittent access drinking paradigm (two-bottle choice, 24-h access every other day), and rapidly developed excessive alcohol intake (~20 g/kg/day), which constitute an appropriate animal model for studying excessive alcohol consumption [Hwa et al 2011; Zhou and Kreek 2018]. In parallel, we evaluated the functional effect of pharmacological blockade of mTORC1 with rapamycin on alcohol drinking. In our 2^nd objective, we investigated whether inhibition of the mTORC1 pathway by rapamycin could prevent U50,488H-promoted alcohol intake, and found that pretreatment with rapamycin blunted the KOP-r agonist-induced alcohol drinking. Our results provide supportive and direct evidence relevant to the transcriptional profiling of the critical mTORC1 genes in mouse NAcs: altered gene transcripts in the mTORC1 pathway after excessive alcohol drinking may contribute to alcohol drinking triggered by KOP-r activation.

METHODS AND MATERIALS

1. ANIMALS

Male adult C57BL/6J (B6) mice (8 weeks of age) were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) and housed in a temperature-controlled room (21 °C). Upon arrival, mice were placed on a 12-hour reverse light-dark cycle (lights off at 7:00 am) and acclimated for a week prior to testing. Mice were housed individually in ventilated cages with steel lids and filter tops and given ad libitum access to food and water. Animal care and experimental procedures were conducted according to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources Commission on Life Sciences 1996) and were approved by the Institutional Animal Care and Use Committee of the Rockefeller University.

2. PROCEDURES

2.1. The drinking-in-the-dark (DID) procedure followed by chronic (3 weeks) intermittent access (IA) drinking procedure (Table S1).

In this one-bottle protocol with alcohol DID drinking model [Rhodes et al 2005], mice had access to alcohol drinking after the beginning of the dark period with 4 hours (limited time/day) in their home cages every day for 4 days with food available. The basic paradigm with our modifications was as follows [Zhou et al 2019b]: At the time when the mice started individual housing (1 week before the experiments), the water tube was replaced with sipper tube bottles to acclimate the mice to the sipper tube. Beginning at 10:00 am (3 hours after lights off) the water tube was replaced with an alcohol tube fitted with a stainless-steel straight sipper tube (containing a ball bearing at the end for preventing alcohol leakage) and sealed with a rubber stopper. The alcohol tube was filled with fresh alcohol solution, kept for 4 hours, and then replaced with a water tube. In all the experiments, 15% alcohol solution (v ⁄ v) was prepared by mixing alcohol with tap water to reach 15% alcohol concentration. Body weight was recorded every day, and alcohol intake value (i.e., g ⁄ kg) was recorded after 4 hours of alcohol access every day.

After the 4-day DID in the first week, the mice then accessed to alcohol in the home cage for 3 weeks with food and water available in a two-bottle free choice paradigm, with alcohol drinking every other day [Hwa et al 2011; Zhou et al 2019b]. Briefly, this intermediate access (IA) procedures were identical to the DID described above with the following exceptions: Starting at 3 hours after lights off, both the 15% alcohol solution and water tubes were provided on home cages. The right/left position of the tubes was randomly changed every other day to avoid the possible side preference. After 4, 8 and 24 hours of alcohol access, both water and alcohol intake values were recorded, and these data were used to calculate alcohol intake (i.e., g ⁄ kg) and relative preference ratio for alcohol (i.e., alcohol intake ⁄ total fluid intake). Access to alcohol following the 3-week IA procedure led to high alcohol intake in the mice [Zhou et al 2019b].

In the experiment of transcriptome profiling (Table S1A), the mice were assigned randomly to the separate groups: the alcohol group (n=12), and the water control group (n=12). The water mice were run in parallel under identical procedures (e.g., one-bottle water for 4 hours during the DID and two-bottle water every other day during 3-week IA), but without alcohol available.

2.2. RNA Extraction.

Mice in both water and alcohol groups were sacrificed by decapitation with brief CO₂ exposure 24 hours after the last IA session; the nucleus accumbens shell (NAcs) was immediately dissected out from the brain and frozen on dry ice (Table S1A). The snap-frozen NAcs tissues pooled from two mice were fractionated into cytoplasmic and nuclear phases using the double-detergent lysis buffer and tuberculin syringes with 22-gauge needles (0.40 mm id) as homogenizers [Zhou et al 2019b]. Tissues (12 samples in total: 6 samples for water control group and 6 samples for alcohol group) were then lysed by the addition of 0.5 ml/sample 0.3 M sucrose lysis buffer and layered over a 0.80-ml cushion of 0.4 M sucrose lysis buffer, and centrifuged at 300 × g for 15 min at 4 °C. The cytoplasmic supernatant fraction was transferred to a fresh tube, and treated with proteinase K (Boehringer Mannheim, Indianapolis, IN) for 1 h at 45 °C. The remaining 0.4-M sucrose cushion was removed, and the nuclear pellet was washed with 0.5 ml 0.4 M sucrose and centrifuged again. After the supernatant was removed, the nuclear pellet was treated with RNase-free DNase-1 (Worthington, Biochemicals, Freehold, NJ) for 5 min at 37 °C, followed by a 30 min proteinase K treatment at 45 °C. Finally, the nuclear RNA was added with Qiazol (Qiagen, Valencia, CA), and the total nuclear RNA was isolated using the miRNeasy kit (Qiagen), and the quality and quantity of nuclear RNA from each sample was determined using an Agilent 2100 Bioanalyzer. All the samples have RNA integrity number above 8.0. This method permits efficient lysis of the brain tissues as evidenced by absence of cytoplasmic tRNA in the nuclear extract fraction, with minimal rupture of nuclei as indicated by absence of DNA in cytoplasmic extract fraction, even with very vigorous homogenization [Zhou and Lapingo 2014].

2.3. RNA-seq library preparation and sequencing.

Both the RNA-seq library preparation and sequencing of nuclear RNA samples isolated from mouse NAcs was performed by the Genomic Resource Center at the Rockefeller University: (1) The RNA-seq libraries were prepared using Illumina’s TruSeq® Stranded Total RNA Library Prep Kit with Ribo-Zero following manufacturer protocol with unique barcodes and pooled at equal molar ratios. Briefly, starting with 100 ng of nuclear RNA extract, the total RNA was fragmented by incubating at 94°C for 8 minutes with divalent cations. The cleaved RNA fragments were copied into first strand cDNA using random primers and reverse transcriptase, followed by second strand cDNA synthesis using DNA polymerase I and RNaseH. During the subsequent addition of the indexing adapters, the double stranded cDNA fragments then had the addition of a single ‘A’ nucleotide to prevent self-ligation. PCR was then used to enrich only those DNA fragments that had adapter molecules on both ends to amplify the amount of DNA in the library. Libraries were validated using Agilent Tape Station High Sensitivity DNA kits and normalized; and (2) Libraries were multiplexed, 12 samples per lane and sequenced. Illumina NextSeq 500 sequencer using high output V2 reagents and NextSeq Control Software v1.4 generates 75 bp paired end reads, following manufacture protocol. As each lane of NextSeq provides 400 million reads, the 12 samples on one NextSeq lane were set up with 30 million+ reads per sample.

2.4. RNA-seq data quality assessment and differential transcript analysis.

The fastq files were generated by configuring BclToFastq.pl from CASAVA v1.8.2 with the following parameters: --ignore-missing-stats, --ignore-missing-bcl, --ignore-missing-control, --positions-format, clocs, --fastq-cluster-count 350000000. Then they were examined using FASTQC [Andrews 2010]. The reads were aligned to the mouse reference genome (version mm10) using STAR v2.3 aligner with default parameters [Dobin et al 2013]. The alignment results were evaluated through qualimap v2.2 [https://academic.oup.com/bioinformatics/article/32/2/292/1744356] to ensure that all the samples had a consistent alignment rate, and no obvious 5’ or 3’ bias. Aligned reads were summarized through feature Counts with the gene model from Ensembl (Mus_musculus.GRCm38.75.gtf) at gene level: specifically, the uniquely mapped reads (NH ‘tag’ in bam file) that overlapped with an exon (feature) by at least 1bp were counted and then the counts of all exons annotated to an Ensembl gene (meta features) were summed into a single number [Liao et al 2014]. Only the protein coding genes were used for the analysis of this study. Principal Component Analysis (PCA) was then applied to the normalized count of all the samples from the NAcs to detect potential outliers. As shown in Figure S1, one sample in water group did not cluster with the rest of the samples under the same condition. After careful review of the procedures (dissection, extraction, etc.), we noted that the individual variability was not contributed by any obvious technical issues and could not exclude that this level of variation was not biological. Therefore, we analyzed all the samples (water control and alcohol groups; n=6 for each group) so that we could present our data in an unbiased way.

The genes selected for analysis were mTORC1 pathway genes, glutamate receptor genes, G protein-coupled receptor kinase genes, regulator of G protein signaling genes, mitogen-activated protein kinase genes, glycogen synthase kinase genes, dopamine receptor genes and opioid genes. The rationale for such selections came from publications to provide gene expression or protein [Laguesse et al 2017b; Ron and Berger 2018; Liu et al 2019; Zhou et al 2019b]. Advanced bioinformatics tool Ingenuity Pathway Analysis [IPA] was also used to obtain the potential interaction of alcohol with the mTORC1 pathway. We used DESeq2 (https://doi.org/10.1186/s13059-014-0550-8) [Love et al 2014], a method for differential analysis of RNA-seq data to estimate fold-change and significance testing. Specifically, DESeq2 estimate gene-wise log fold of changes (LFCs) between conditions from the standard Generalized Linear Model (GLM) fits to obtain maximum-likelihood estimates, and then fit a zero-centered normal distribution to the observed distribution of Maximum-like hood Estimate over all genes. This distribution is used as a prior on LFC in a second round of GLM fits, and the maximum posteriory estimate is kept as final estimates of LFC. For significance testing, DESeq2 uses a Wald test: the shrunken estimate of LFC is divided by its standard error, resulting in a z-statistic, which is compared to a standard normal distribution. The Wald test P values from the subset of genes that pass an independent filtering step are adjusted for multiple testing using the procedure of Benjamini and Hochberg [Benjamini et al 1995].

2.5. Alcohol drinking with single administration of rapamycin, U50,488H or rapamycin+U50,488H.

The first objective of these experiments was to determine dose-dependent effects of rapamycin on alcohol drinking. For each dose of rapamycin (0, 2, 4 or 8 mg/kg), separate groups of mice were used. After chronic (4-week) excessive alcohol drinking (Table S1B), the mice were assigned randomly to the vehicle-treated and drug-treated groups, with similar alcohol intake one day before the test day. The range of rapamycin doses was based on the published study [Neasta et al 2010]. On the test day, alcohol was presented 10 min after a single injection of rapamycin or vehicle, and then alcohol and water intake values were recorded.

The second objective was to determine dose-related effects of U50,488H on alcohol drinking. Like the above experiment, for each dose of U50,488H (0, 0.75 or 1.5 mg/kg), separate groups of mice were used. The range of U50,488H doses was based on the published study [Rose et al 2016]. On the test day, a single injection of U50,488H or vehicle was administered 4 hours after the start of the drinking session. The time point for U50,488H administration was chosen due to a slow onset of rapamycin’s effect at the 4–8-hour (but not 0–4-hour) interval, consistent with early studies [e.g., Neasta et al 2010]. In comparison with the excessive alcohol drinking model (1-week DID + 3-week IA), we tested the effective dose of U50,488H (1.5 mg/kg) after 4 weeks of DID. A single injection of U50,488H or vehicle was administered at the beginning of the last DID drinking session and recorded alcohol intake after 4 hours.

Finally, we tested whether rapamycin pretreatment could block the U50,488H-induced increase in alcohol drinking, to confirm that the U50,488H effects were mediated via the mTORC1 pathway. The IA mice were pretreated with rapamycin (1, 2 or 4 mg/kg) in vehicle (i.p.) 10 min before the drinking session, followed by one U50,488H (1.5 mg/kg) or vehicle injection 4 hours after the start of the drinking session.

2.6. Sucrose (caloric reinforcer) and saccharin (non-caloric reinforcer) drinking with single administration of rapamycin.

With 8 mg/kg rapamycin (the effective dose tested for reducing alcohol intake), we further tested the specificity of the action of rapamycin on alcohol drinking using sucrose or saccharin drinking behavior in mice after chronic excessive alcohol drinking. In the following experiments, after 4-week alcohol exposure (identical to those in the above experiment), the alcohol tube was changed to 4% sucrose for 3 sessions with stable intakes. The mice in the vehicle- or rapamycin-treated groups had similar sucrose intake 24 hours before the test day. On the test day (over 6 days of sucrose or saccharin drinking), 4% sucrose and water intake values were recorded after 4, 8 and 24 hours of sucrose access. In parallel separate experiments, 0.1% saccharin drinking was tested with an identical procedure.

2.7. Materials.

Ethanol solution was prepared from 190-proof absolute ethyl alcohol (Pharmco-AAPER, Brookfield, CT, USA) and dissolved in tap water. (±)U50,488H was obtained from the NIDA Division of Drug Supply and Analytical Services and dissolved in saline for systemic i.p. administration. Rapamycin was purchased from LC Laboratories (Woburn, MA), and dissolved in 10% DMSO for systemic i.p. administration.

2.8. Data analysis.

Power analyses were performed to determine the number of mice (about 6–8 mice per group) required to provide statistically significant RNA-seq and behavioral results, based on the levels of differences reported before (see details in Supplementary Information, Data analysis section) [Zhou and Kreek 2019a, 2019b]. DESeq2 was applied to the normalized counts to estimate the fold change between the samples from mice that had chronic alcohol drinking and those from water controls, using negative binomial distribution [Love et al 2014]. An adjusted p-values of less than 0.05 (FDR<0.05) was used to select genes that have a significant expression change. To explore possible relationships between individual gene transcript level and vulnerability to alcohol drinking, the last 24-h alcohol intake and gene transcript were examined through regression.

In the experiments with single rapamycin or U50,488H, the group differences in alcohol (sucrose or saccharin) intake, water intake, total fluid and preference ratio were analyzed using 2-way ANOVA with repeated measures for treatment (vehicle vs drug) and for time interval (0–4, 4–8 vs. 8–24). For dose-response analysis on rapamycin, U50,488H or rapamycin+ U50,488H, the group differences for alcohol intake and preference ratios were analyzed using 1-way ANOVA for treatments with different doses. In the experiments with single U50,488H after rapamycin pretreatment, the group differences for alcohol intake and preference ratios were analyzed using 2-way ANOVA for pretreatment (vehicle vs rapamycin) and for treatment (vehicle vs U50,488H). All the ANOVAs were followed by Newman-Keuls post-hoc tests. The accepted level of significance was p<0.05 (Statistica version 5.5, StatSoft Inc, Tulsa, OK).

RESULTS

1. Effects on nuclear transcript levels in the NAcs after chronic (4-week) excessive alcohol drinking.

• 1.1.Alcohol intake and preference during chronic alcohol drinking in DID and IA models. After 4 days of DID, there was significantly more alcohol intake on day 4 than that on day 1 [Student’s t-tests, F(1,22)=5.31, p<0.05] (Table S2A). Then after exposed to the IA drinking for 3 weeks, the mice showed alcohol intake averagin ~20 g/kg/day, with high preference ratio (>0.8). For alcohol intake (Table S2B), 2-way ANOVA revealed a significant effect of session [F(1,22)=7.8, p<0.01]. Post hoc analysis showed that there were significantly more alcohol intakes at 0–4 h and 8–24 h intervals in the 10^th session than those in the 1^st session [p<0.05 for both]. For alcohol preference, 2-way ANOVA revealed a significant effect of session [F(1,22)=6.9, p<0.01]. Post hoc analysis showed that there was a significantly more preference at 0–4 h interval in the 10^th session than that in the 1^st session [p<0.05].
• 1.2.mTORC1 pathway, mitogen-activated protein (MAP) kinase pathway and related genes. RNA-seq analysis revealed significant changes in response to excessive alcohol (Table 1): thirteen genes showed significant up-regulation in the nuclear transcript levels, with two genes down-regulated. For many other genes, there was no significant effect of excessive alcohol (Table S3).
• 1.3.Discs large MAGUK scaffold protein (dlg), glutamate receptors and Homers genes. As shown in Table 2, three Dlg, five glutamate receptor and homer 2 genes showed significant up-regulation in the nuclear transcript levels. For other genes, there was no significant effect of excessive alcohol (Table S4).
• 1.4.G protein-coupled receptor kinase (Grk), regulator of G protein signaling (Rgs) and arrestin beta (arrb) genes. As shown in Table 3, Grk5 and Rgs17 genes showed significant up-regulation and down-regulation in the nuclear transcript levels, respectively, with no change of others (Table S5).
• 1.5.Opioid genes. As shown in Table 4, proopiomelanocortin (Pomc) gene showed significant up-regulation in the nuclear transcript levels only.
• 1.6.mTORC2 pathway and related genes. As shown in Table 5, serum/glucocorticoid regulated kinase 1 (Sgk1) gene showed significant up-regulation in the nuclear transcript levels only.
• 1.7.Nuclear transcripts of tyrosine 3-monooxygenase /tryptophan 5-monooxygenase activation protein genes (Table S6), dopamine receptor genes and melanocortin receptor genes (Table S7). For these genes, there was no significant effect of excessive alcohol.

Table 1.

Nuclear transcript levels of mTORC1 pathway, mitogen-activated protein (MAP) kinase pathway and related genes in the nucleus accumbens shell were altered after chronic (4-week) excessive alcohol drinking (FDR < 0.05). FC, fold change; FDR, false discovery rate; average DESeq2 normalized count (NC) for water and alcohol -exposed mice (n=6 samples for each group).

Gene	Fold	p-Value	FDR	Water NC	Alcohol NC	Name
Crmp2	1.61	2.91E-04	0.043	9404	15,097	Collapsin response mediator protein 2
Crmp5	1.25	7.83E-04	0.021	274	344	Collapsin response mediator protein 5
Eif4g1	1.16	8.53E-04	0.022	3166	3679	Eukaryotic translation initiation factor 4, gamma 1
Eif4g3	1.23	1.58E-03	0.031	1757	2167	Eukaryotic translation initiation factor 4, gamma 3
Eif4ebp1	1.63	3.04E-04	0.044	12	19	Eukaryotic translation initiation factor 4E binding protein 1
Fam102a	1.21	4.10E-04	0.015	811	985	Family with sequence similarity 102, member A
Fam19a4	0.71	6.19E-04	0.019	27	18	Family with sequence similarity 19, member 4
Fam19a5	1.21	2.79E-03	0.042	759	920	Family with sequence similarity 19, member 5
Prosapip1	1.25	2.47E-08	2.6E-05	2852	3592	ProSAP/shank-interacting protein 1
Pdk1	1.16	1.80E-03	0.033	584	680	Pyruvate dehydrogenase kinase, isoenzyme 1
Rasgrp2	1.54	6.70E-04	0.031	1061	1737	RAS guanyl releasing protein 2
Stk11	1.22	3.35E-03	0.046	425	518	Serine/threonine kinase 11
Tnrc6b	1.23	7.18E-06	0.001	2607	3217	Trinucleotide repeat containing 6b
Tnrc6c	1.29	2.58E-07	0.001	1338	1725	Trinucleotide repeat containing 6c
Tsc22d3	0.81	2.08E-03	0.036	1050	815	TSC22 domain family, member 3
Mapk8ip2	1.22	1.10E-04	0.008	2154	2633	Mitogen-activated protein kinase 8 interacting protein 2
Mapkbp1	1.14	1.20E-03	0.027	952	1081	Mitogen-activated protein kinase binding protein 1

Table 2.

Nuclear transcript levels of discs large MAGUK scaffold protein (Dlg), glutamate receptor and Homer genes in the nucleus accumbens shell were altered after chronic (4-week) excessive alcohol drinking (FDR < 0.05). FC, fold change; FDR, false discovery rate; average DESeq2 normalized count (NC) for water and alcohol -exposed mice (n=6 samples for each group).

Gene	Fold	p-Value	FDR	Water NC	Alcohol NC	Name
Dlg3	1.20	2.29E-04	0.011	2384	2901	Discs large MAGUK scaffold protein 3
Dlg4	1.36	5.50E-06	0.001	2437	3534	Discs large MAGUK scaffold protein 4
Dlg5	1.21	2.46E-03	0.040	372	469	Discs large MAGUK scaffold protein 5
Gria1	1.17	7.37E-05	0.007	8539	9979	Glutamate receptor, ionotropic, AMPA1 (alpha 1)
Grid1	1.16	5.04E-04	0.017	1310	1519	Glutamate receptor, ionotropic, delta 1
Grid2ip	1.28	2.06E-03	0.020	7	19	Glutamate receptor, ionotropic, delta 2 interacting protein 1
Grin2b	1.24	1.88E-03	0.034	13740	17818	Glutamate receptor, ionotropic, NMDA2B (epsilon 2)
Grin2c	1.38	4.06E-06	0.001	316	455	Glutamate receptor, ionotropic, NMDA2C (epsilon 3)
Homer2	1.35	2.62E-04	0.012	589	901	Homer scaffolding protein 2

Table 3.

Nuclear transcript levels of G protein-coupled receptor kinase (Grk)/regulator of G protein signaling (Rgs) genes in the nucleus accumbens shell were altered after chronic (4-week) excessive alcohol drinking (FDR < 0.05). FC, fold change; FDR, false discovery rate; average DESeq2 normalized count (NC) for water and alcohol -exposed mice (n=6 samples for each group).

Gene	Fold	p-Value	FDR	Water NC	Alcohol NC	Name
Grk5	1.27	1.31E-04	8.9E-03	264	344	G protein-coupled receptor kinase 5
Grk4	1.08	0.319	0.546	47	51	G protein-coupled receptor kinase 4
Grk6	1.09	0.060	0.215	584	639	G protein-coupled receptor kinase 6
Rgs17	0.87	5.35E-05	0.006	2773	2419	Regulator of G-protein signaling 17

Table 4.

Nuclear transcript levels of opioid genes in the nucleus accumbens shell were altered after chronic (4-week) excessive alcohol drinking (FDR < 0.05). FC, fold change; FDR, false discovery rate; average DESeq2 normalized count (NC) for water and alcohol -exposed mice (n=6 samples for each group).

Gene	Fold	p-Value	FDR	Water NC	Alcohol NC	Name
Pomc	1.32	9.10E-04	0.023	75	175	Proopiomelanocortin
Oprm1	0.92	0.284	0.508	404	373	Mu opioid receptor
Pdyn	0.93	0.351	0.574	2388	2192	Prodynorphin
Oprk1	0.95	0.485	0.688	757	720	Kappa opioid receptor
Penk	1.08	0.330	0.555	9684	10756	Proenkephalin
Oprd1	1.15	0.077	0.245	259	319	Delta opioid receptor
Pnoc	1.07	0.401	0.616	43	52	Pronociceptin
Oprl1	0.98	0.877	0.942	765	753	Nociceptin receptor

Table 5.

Nuclear transcript levels of mTORC2 pathway and related genes in the nucleus accumbens shell were altered after chronic (4-week) excessive alcohol drinking (FDR < 0.05). FC, fold change; FDR, false discovery rate; average DESeq2 normalized count (NC) for water and alcohol -exposed mice (n=6 samples for each group).

Gene	Fold	p-Value	FDR	Water NC	Alcohol NC	Name
Sgk1	1.36	3.75E-08	3.11E-05	1216	1727	serum/glucocorticoid regulated kinase 1
Acta1	1.24	0.012	0.089	35	48	actin alpha 1
Acta2	1.03	0.703	0.703	22	24	actin alpha 2
Actb	1.09	0.058	0.212	18222	19925	actin beta
Actbl2	1.01	0.755	0.755	0.64	0.80	actin beta like 2
Actc1	1.00	0.788	0.788	0.140	0.147	actin alpha cardiac muscle 1
Actg1	1.09	0.046	0.187	980	1084	actin gamma 1
Actg2	1.00	0.892	0.893	0.572	0.354	actin gamma 2
Actl6a	1.00	0.990	0.996	227	227	actin-like protein 6A
Actl6b	1.05	0.391	0.608	562	595	actin-like protein 6B
Actl7b	0.99	0.864	0.864	0.925	0.596	actin-like protein 7A
Actl10	1.03	0.238	0.238	0.120	0.639	actin-like protein 10
Actr1a	1.00	0.912	0.960	2418	2426	actin related protein 1A
Actr1b	1.08	0.133	0.335	1537	1677	actin related protein 1B
Actr2	1.10	0.092	0.274	4607	5129	actin related protein 2
Actr3	1.13	0.014	0.100	1875	2141	actin related protein 3
Actr3b	1.02	0.687	0.831	924	943	actin related protein 3B
Actr5	0.99	0.793	0.896	162	164	actin related protein 5
Actr6	1.07	0.146	0.351	310	336	actin related protein 6
Actr8	1.03	0.493	0.696	778	797	actin related protein 8
Actr10	1.03	0.341	0.565	1480	1534	actin related protein 10
Actrt3	1.04	0.674	0.674	10	13	actin related protein T3
Akt1	1.10	0.040	0.403	1165	1278	serine/threonine-protein kinase, RAC-alpha
Akt1s1	1.09	0.101	0.301	231	251	proline-rich AKT1 substrate 1
Akt2	1.22	0.006	0.059	730	890	serine/threonine-protein kinase, RAC-beta
Akt3	1.20	0.055	0.168	3807	4555	serine/threonine-protein kinase, RAC-gamma
Aktip	1.01	0.919	0.942	1690	1703	AKT-interacting protein
Fyn	1.04	0.381	0.599	989	1032	proto-oncogene tyrosine-protein kinase
Gsk3a	1.17	0.005	0.059	663	795	glycogen synthase kinase 3 alpha
Gsk3b	1.12	0.084	0.261	3903	4486	glycogen synthase kinase 3 beta
Gskip	1.06	0.140	0.345	473	504	glycogen synthase kinase 3 beta interacting protein
Prkca	1.17	0.035	0.161	2450	2996	protein kinase C alpha
Prkcb	1.24	0.005	0.058	7773	10352	protein kinase C beta
Prkcd	1.20	0.025	0.134	213	416	protein kinase C delta
Prkcdbp	0.96	0.528	0.722	134	124	protein kinase C delta binding protein
Prkce	1.20	0.014	0.096	3809	4783	protein kinase C epsilon
Prkcg	1.16	0.067	0.227	1159	1428	protein kinase C gamma
Prkch	1.04	0.641	0.803	836	885	protein kinase C eta
Prkci	1.15	0.038	0.168	852	1007	protein kinase C iota
Prkcq	1.15	0.023	0.129	164	202	protein kinase C theta
Prkcz	1.10	0.073	0.238	2418	2685	protein kinase C zeta
Rac1	1.10	0.023	0.130	2413	2684	ras-related C3 botulinum toxin substrate 1
Rac2	1.18	0.064	0.645	16	21	ras-related C3 botulinum toxin substrate 2
Rac3	1.02	0.735	0.860	83	86	ras-related C3 botulinum toxin substrate 3
Rictor	1.03	0.491	0.694	2374	2440	rapamycin-insensitive companion of mTOR
Sgk2	0.84	0.048	0.480	27	18	serum/glucocorticoid regulated kinase 2
Sgk3	0.98	0.752	0.871	654	642	serum/glucocorticoid regulated kinase 3
Tiam1	1.12	0.079	0.250	2298	2616	T-lymphoma invasion and metastasis-inducing protein 1
Tiam2	1.11	0.108	0.296	1499	1701	T-lymphoma invasion and metastasis-inducing protein 2

2. Correlations between the individual alcohol intake and nuclear transcript levels in the NAcs.

To further analyze the relationship between the last 24-hour alcohol intake in session 10 and subsequent individual changes in the nuclear transcript levels, all the altered genes listed in Tables 1–5 were examined to determine whether there was any correlation of the altered nuclear transcript levels in relationship with the individual amount of alcohol intake. Of note, the intake values were the average of two mice as each of the nuclear RNA samples was pooled from the two mice. This analysis revealed significant two positive (Crmp2, Homer2) and two negative (Tsc22d3, Grin2c) correlations (Figure S2).

3. Single administration of rapamycin decreased alcohol intake and preference in a dose-dependent manner after chronic drinking.

After two low doses (2 and 4 mg/kg) of rapamycin, there was no effect on alcohol intake, water intake or alcohol preference ratio [Figure S3]. At 8 mg/kg, rapamycin significantly decreased alcohol intake [2-way ANOVA, F(1,13)=5.71, p<0.05] at the 4–8-hour interval [post-hoc test p<0.05] (Figure 1A). There was a compensatory increase in water intake [F(1,13)=6.12, p<0.05] (Figure 1B), resulting in unaltered total fluid intake (Table S8). There was also a significantly reduction of preference ratio [F(1,13)=5.77, p<0.05] at the 4–8-hour interval [p<0.05] (Figure 1C). Of note, the present study observed a relatively slow onset of the effect after 4 hours of rapamycin administration on alcohol drinking behavior. Figure 2 presents a dose response of single rapamycin administration (2–8 mg/kg) in alcohol intake and preference at the 4–8-hour interval. For alcohol intake (Figure 2A), there was a main effect of rapamycin [1-way ANOVA, F(3,25)=11.1, p<0.01], and post hoc analysis revealed that (1) compared with the control group, the rapamycin-treated mice had less intake at 8 mg/kg [p<0.05]; and (2) the decrease at 8 mg/kg was greater than that at 2 mg/kg [p<0.05]. For preference ratio (Figure 2B), there was a main effect of rapamycin [1-way ANOVA, F(3,25)=8.2, p<0.01], and post hoc analysis revealed that compared with the control group, the rapamycin-treated mice had less preference at 8 mg/kg [p<0.05].

Figure 1.

Effects of single administration of rapamycin (8 mg/kg) on alcohol intake (A), water intake (B), and preference ratio (C) in mice after 4-week chronic excessive drinking. (1) Control groups: mice (n=7) received one vehicle injection (i.p.) before the drinking test; and (2) Rapamycin groups: mice (n=8) received one rapamycin injection (8 mg/kg, i.p.) 10 min before the drinking session. Alcohol (15%) and water intake values were recorded after 4, 8 and 24 hours of alcohol access. *p<0.05 vs. control at the same time point.

Figure 2.

Dose responses of single administration of rapamycin (0, 2, 4 or 8 mg/kg) on reducing alcohol intake (A) and alcohol preference ratio (B) in mice after 4-week chronic excessive drinking. Data were collected at the 4–8-hour interval on the baseline and testing day and are expressed as a percentage of baseline alcohol intake to account for the differences in baseline that contribute to variation between experiments. *p<0.05 vs. control (0 mg/kg); +p<0.05 between 2 mg/kg and 8 mg/kg treatment groups (n=6–8).

4. No effects of single rapamycin (8 mg/kg) on sucrose or saccharin intake or preference in mice.

Rapamycin had no significant effect of 8 mg/kg rapamycin on either intake or preference of 4% sucrose at the 4–8-hour interval in mice following excessive alcohol drinking (n = 6) (control mice: 7.0±0.61 g/kg; rapamycin mice: 7.1±0.88 g/kg) or preference ratio (control: 0.91±0.06; rapamycin: 0.90±0.07). Similarly, there was no effect of rapamycin on 0.1% saccharin intake or preference (n = 6/per group) (control: 0.13±0.03 g/kg; rapamycin: 0.11±0.04 g/kg) or preference ratio (control: 0.90±0.03; rapamycin: 0.92±0.03). At either the 0–4 or 8–24-hour interval, there were no significant effects of rapamycin on either 4% sucrose or 0.1% saccharin drinking (data not shown).

5. Single administration of U50,488H increased alcohol consumption after chronic drinking.

At 0.75 and 1.5 mg/kg, U50,488H increased alcohol intake at 4–8-hour interval in a dose-related manner (Figure 3A): 1-way ANOVA showed a significant effect on alcohol intake [F (2, 17) = 5.74, p<0.05], and Post hoc analysis showed that the mice after 1.5 mg/kg U50,488H had significantly more alcohol intake than the vehicle control ones [p<0.05]. For preference ratio, there was no significant effect of U50,488H (Figure 3B). At the 8–24-hour interval, there were no significant effects of U50,488H on alcohol intake (Data not shown).

Figure 3.

Dose responses of single administration of U50,488H (0, 0.75 or 1.5 mg/kg) on increasing alcohol intake in mice after 4-week chronic excessive drinking. Data were collected at the 4–8-hour interval on the baseline and testing day and are expressed as a percentage of baseline alcohol intake to account for the differences in baseline that contribute to variation between experiments. *p<0.05 vs. control (0 mg/kg) (n=6–8).

To assess the effect of U50,488H on the short access “binge” alcohol drinking, the mice received U50,488H at 1.5 mg/kg or saline (n = 6/per group) after 4 weeks of DID. There was no effect on alcohol intake after 4 hours (control mice: 5.1±0.39 g; U50,488H mice: 4.6±0.69 g).

6. Rapamycin pretreatment blocked the effect of U50,488H on increasing alcohol drinking.

In a pilot study, rapamycin at 1 mg/kg had no effect on U50,488H-increased alcohol intake [data not shown]. As shown in Figure 4, pretreatment with rapamycin at 2 mg/kg showed the effect of U50,488H on alcohol intake: 2-way ANOVA showed significant effects of U50,488H [F(1,24)=6.07, p<0.05] and rapamycin pretreatment [F(1,24)=6.18, p<0.05], with a marginally significant interaction between the rapamycin pretreatment and U50,488H [F(1,24)=3.72, p=0.07]. U50,488H at 1.5 mg/kg significantly increased alcohol intake [p<0.05], and pretreatment with rapamycin at 2 mg/kg blunted the effect of U50,488H.

Figure 4.

Pretreatment with mTORC1 inhibitor rapamycin (2 mg/kg) blocks the effect of U50,488H (1.5 mg/kg) on increasing 15% alcohol drinking in mice (n=6–8). On the test day, alcohol and water intake values were recorded at the 4–8-hour interval. *p<0.05 vs. vehicle control without pretreatment.

Figure 5 presents a dose response of single administration of U50,488H (0 or 1.5 mg/kg) alone or pretreated with rapamycin (1–4 mg/kg) in alcohol intake and preference at the 4–8-hour interval. For alcohol intake (Figure 5A), there was a main effect of U50,488H [1-way ANOVA, F(7,44)=8.1, p<0.01], and post hoc analysis revealed that (1) compared with the control group, the U50,488H-treated mice had more intake [p<0.05]; and (2) pretreatment with rapamycin at 2 or 4 mg/kg blunted the effect of U50,488H [p<0.05 for both]. For preference ratio (Figure 5B), there was a main effect of U50,488H [1-way ANOVA, F(7,44)=5.2, p<0.05], and post hoc analysis revealed that the mice pretreated with rapamycin at 4 mg/kg before U50,488H had significantly less alcohol preference than the ones without rapamycin pretreatment before U50,488H [p<0.05].

Figure 5.

Dose responses of single administration of U50,488H (U50, 0 or 1.5 mg/kg) alone or pretreated with rapamycin (Rap, 0, 1, 2, or 4 mg/kg) on increasing 15% alcohol intake (A) and alcohol preference (B) in mice after 4-week chronic excessive drinking. Data were collected at the 4–8-hour interval on the baseline and testing day and are expressed as a percentage of baseline alcohol intake to account for the differences in baseline that contribute to variation between experiments. *p<0.05 vs. control (U50 and Rap at 0mg/kg, respectively), and # p<0.05 between the treatment groups (n=6–8).

DISCUSSION

In mice, the mTORC1 pathway activation is associated with the negative reinforcement aspects of alcohol addiction, especially after excessive alcohol [Neasta et al 2010, Ron and Berger 2018]. In the present study, we found significant increases and decreases in many gene transcripts in the mTORC1 pathway in the NAcs of mice in response to chronic excessive alcohol drinking (Table 1). In the upstream of the mTORC1 pathway, we observed a significant increase in Stk11 transcript and a decrease in Tsc22d3 transcript, which could affect the mTORC1 activity [Liu et al 2019]. The mTORC1 substrates included proteins of Crmp2, Crmp5, Eif4g1, Eif4g3, Eif4ebp1 and Prosapip1 genes, which are key regulators of the translation initiation [Ma and Blenis, 2009, Ron and Berger 2018]. We then examined whether excessive alcohol drinking was enough to stimulate transcriptional activity of functional protein genes that are regulated by mTORC1, including dlg, glutamate receptors, and Homers, and found that many of these gene transcripts were increased (Table 2). Therefore, our results suggest that excessive alcohol profoundly altered many component genes of the mTORC1 signaling pathway and its related functional proteins in the NAcs, consistent with previous studies at protein phosphorylation levels by other groups [Neasta et al 2010; Ron and Berger 2018; Starski et al 2019]. In addition, we functionally examined the effect of pharmacological blockade of mTORC1 pathway and found that the selective mTORC1 inhibitor rapamycin dose-dependently reduced excessive alcohol drinking in mice (Figure 2). Consistent with many early reports, the present study further confirmed that the effect of rapamycin on alcohol intake was not secondary to a general suppression of appetitive (anhedonic effect) or consumption behaviors, since rapamycin did not decrease sucrose (caloric reinforcer) or saccharin (non-caloric reinforcer) consumption [see review by Ron and Berger in 2018]. Our behavioral result is also consistent with previous findings that the mTORC1 pathway in the NAc plays an important functional role in alcohol taking and seeking behaviors [Neasta et al 2010; Cozzoli et al 2016; Laguesse et al 2017a].

The altered nuclear transcript levels have been found to correlate with transcriptional activity in the in vivo studies [Grindberg et al 2013; Lacar et al 2016], as nuclear RNA quantity can present transcriptional activity in the cells for protein-coding genes in response to stimuli. Furthermore, many studies have demonstrated that altered nuclear transcript levels parallel the changes of cytoplasmic mRNA levels, as the accumulation of cytoplasmic mRNAs are mainly due to the changes of nuclear transcriptional activity [Grindberg et al 2013; Zhou and Lapingo 2014]. Therefore, our nuclear RNA-Seq data represent nascent transcriptional activity of the mTORC1 genes in response to excessive alcohol and may recapitulate the transcript levels associated with altered neuronal transcriptome profile, enabling the identification of nascent activity-associated mRNAs. Finally, our results of the pharmacological analyses using mTORC1 inhibitor rapamycin provide functional validation about the mTORC1 transcriptome profiling.

Though the mTORC1 pathway was activated to a high degree after excessive alcohol in the NAcs, we observed several nuclear transcript changes in other signaling pathways: two genes of MAPK pathway (Mapk8ip2 and Mapkbp1), two of G protein-coupled receptor kinase and regulator of G-protein signaling pathway (Grk5, Rgs17) and one of mTORC2 pathway (Sgk1) (Table 1, Table 3, Table 5). KOP-r activation triggers the mTORC1 [Liu et al 2019], the MAPK [Bruchas and Chavkin 2010] and the GRK/RGS [Gross et al 2019] pathways in a region-specific manner, and all of them have been found to be directly or indirectly involved in the KOP-r triggered-aversion [Bruchas and Chavkin 2010; Gross et al 2019; Liu et al 2019] or alcohol-related changes [Logrip et al 2008; Laguesse et al 2018; Funk et al 2019]. Indeed, after chronic excessive alcohol drinking, the dynorphin/KOP-r activity is increased in several neuronal regions (including NAc, bed nucleus of the stria terminalis [BNST] and central amygdala), which could produce anxiety- or depression-like behavior [Walker and Koob 2008; D’Addario et al 2013; Kissler et al 2014; Rose et al 2016; Anderson and Becker 2017; Zhou and Kreek 2018]. It is reasonable to speculate that the enhanced KOP-r activity is contributed by the profound changes of the mTORC1 and other signaling pathways, rather than the KOP-r expression levels (Table 4). In support of this concept, previous studies have found a sensitized KOP-r function in alcohol “dependent” rats and mice, which also results in anxiety- or depression-like negative affective states, and in turn leads to excessive drinking (a negative reinforcing mechanism) [Kissler et al 2014; Rose et al 2016].

In the present study, we further determined a dose-relationship of selective KOP-r agonist U50,488H with alcohol consumption in mice after chronic excessive alcohol drinking. Specifically, we found that U50,488H at 1.5 mg/kg dose (but not 0.75 mg/kg) significantly increased alcohol intake at the 4–8-hour interval session (Figure 3). In contrast to mice in 4-week chronic “binge” DID paradigm with 4-hour short-access, which had low alcohol intake (~5 g/kg/day), the mice in 4-week chronic excessive drinking paradigm (1-week DID + 3-week IA) showed excessive drinking with ~20 g/kg/day [Rhodes et al. 2005; Hwa et al. 2011; Zhou and Kreek, 2019a]. Here we purposely compared the effects of U50,488H on excessive drinking with “binge” DID, and found that U50,488H at 1.5 mg/kg did not alter alcohol drinking after chronic DID. It is likely that the increasing effect of U50,488H on alcohol drinking is related to its excessive alcohol intake, probably with related anxiogenic and depression-like behaviors [Rose et al 2016; Anderson and Becker 2017; Zhou and Kreek 2018]. It has been demonstrated that KOP-r activation inhibits glutamate transmission in the BNST, which is involved in anxiety-like behavior [Crowley et al 2016].

Based on our transcriptional profiling of the mTORC1 genes in the NAcs, we further tested a novel hypothesis that the KOP-r agonist U50,488H enhances alcohol consumption through the mTORC1 activation. For this purpose, we investigated the potential involvement of the mTORC1 pathway in alcohol intake promoted by KOP-r activation, and found, for the first time, that the effect of U50,488H on alcohol drinking was mTORC1-mediated, as the U50,488H effect was blocked by the selective mTORC1 inhibitor rapamycin in a dose-dependent manner (Figures 4 and 5). To our knowledge, this is the first study showing that KOP-r agonists increase alcohol consumption through the mTORC1 activation as a potential reinforcing mechanism. Also, our hypothesis generated based on the RNA-seq data were validated at behavioral levels. Though our initial and preliminary evidence supported a potential role for the NAcs mTORC1 pathway signaling involved in the U50,488H-induced alcohol drinking, further experiments are necessary to fill important knowledge gaps such as the particular and specific brain regions and/or neuronal circuits of the KOP-r and mTORC1 interactions.

Using nuclear RNA-Seq, the present study confirmed the alcohol-induced increase in Pomc gene expression: the mice after chronic alcohol drinking, in comparison with the water control mice, had increased nuclear Pomc transcript levels in the NAcs (Table 4), as reported before at cytoplasmic Pomc mRNA levels in alcohol-preferring rats [Zhou et al 2013]. Activation of mu-opioid receptor [MOP-r] by beta-endorphin [encoded by Pomc] produces rewarding effects, which plays an important role in the reinforcing actions and motivational behaviors of alcohol drinking in mice. Given the fact that Prosapip1 [ProSAP/shank-interacting protein 1] also contributes to alcohol reward in the NAc shell [Laguesse et al 2017b], our finding that the increased Prosapip1 transcript (Table 1) was positively correlated with Pomc transcript increases in the NAc shell of alcohol-experienced mice (Figure S4), suggests that Prosapip1 may participate in the mechanisms (including potential epigenetic modifications) [Kokare et al 2017] that enhance Pomc gene expression and then beta-endorphin activity, which in turn increase individual vulnerability to consume more alcohol. However, it remains unclear whether the Pomc gene expression in the NAcs can be processed to beta-endorphin or not.

Finally, in line with many previous reports using RNA-seq, protein and genetic analyses in different rodent models [Eisenhardt et al 2015; McClintick et al 2015; Beckley et al 2016; McGinn et al 2016; Laguesse et al 2017b], we observed that several glutamate receptors gene transcripts (Grin2c, Grin2b, Gria1, Grid1 and Grid2ip) were increased in the NAcs after excessive alcohol drinking (Table 2), suggesting an altered glutamate receptor function in this rewarding region. As NR2C-containing receptors (encoded by Grin2c) are less sensitive to ethanol inhibitory effects than those containing NR2B receptors (encoded by Grin2b) [Lovinger and Roberto 2013], so the Grin2b and Grin2c transcript increases with different magnitudes may produce an altered sensitivity of the NMDA receptors to ethanol in the NAcs. Given the fact that GluA1 receptors (encoded by Gria1) are highly involved in alcohol reinforcing effects at an initial stage [Beckley et al 2016], during nicotine withdrawal [McGinn et al 2016] and in relapse-like drinking [Eisenhardt et al 2015], the increased Gria1 transcript may play an important role in excessive alcohol consumption.

A limitation of the present study is that, though the nuclear transcripts studied here were highly correlated with gene transcriptional profile, cytoplasmic mRNA (assayed by RT-PCR) and protein changes (assayed by Western blot) are more relevant to protein translational activity and bio-function, respectively. Though our pharmacological data with rapamycin support a potential role for the mTORC1 pathway in alcohol drinking, the experiments to determine the effects of rapamycin on gene expression and protein function could provide further understanding of the interactions between the mTORC1 pathway and alcohol drinking behaviors. These limitations warrant the need for these analysis in the future.

Conclusion:

The comprehensive characterization of transcriptional activity represents a key step in the understanding of chronic alcohol exposure in which gene expressions take place in the CNS, as many previous studies have found that excessive alcohol drinking lead to profound alterations in cytoplasmic mRNA levels of mTORC1 pathway genes in mouse NAc. In this study, we hypothesized that alcohol-induced transcriptional regulation in the nucleus is another critical step for keeping the persistent synthesis of cytoplasmic mRNAs and then peptide levels that is critically involved in subsequent alcohol drinking behavior. Therefore, we determined the effects of chronic excessive alcohol drinking on the mTORC1 and other signaling pathways at nuclear transcriptional levels in the NAc shell, and observed a profound effect on the mTORC1 and related function protein gene transcripts, which may contribute to the sensitized responses to KOP-r agonists. The result of a new set point of the mTORC1 gene transcription activities may contribute to individual vulnerability to excessive alcohol drinking after stress triggered by KOP-r activation or stress.

Highlights:

• mTORC1 genes in nucleus accumbens shell after excessive alcohol drinking

• rapamycin reduced excessive alcohol drinking

• kappa opioid receptor activation-induced alcohol intake;

• rapamycin prevented kappa promoted alcohol drinking