Abstract
Opioids initiate dynamic maladaptation in brain reward and affect circuits that occur throughout chronic exposure and withdrawal that persist beyond cessation. Protracted abstinence is characterized by negative affective behaviors such as heightened anxiety, irritability, dysphoria, and anhedonia, which pose a significant risk factor for relapse. While the ventral tegmental area (VTA) and μ-opioid receptors (MORs) are critical for opioid reinforcement, the specific contributions of VTAMOR neurons in mediating protracted abstinence-induced negative affect is not fully understood. In our study, we elucidate the role of VTAMOR neurons in mediating negative affect and altered brain-wide neuronal activities following forced opioid exposure and abstinence in male and female mice. Utilizing a chronic oral morphine administration model, we observe increased social deficit, anxiety-related, and despair-like behaviors during protracted forced abstinence. VTAMOR neurons show heightened neuronal FOS activation at the onset of withdrawal and connect to an array of brain regions that mediate reward and affective processes. Viral re-expression of MORs selectively within the VTA of MOR knock-out mice demonstrates that the disrupted social interaction observed during protracted abstinence is facilitated by this neural population, without affecting other protracted abstinence behaviors. Lastly, VTAMORs contribute to heightened neuronal FOS activation in the anterior cingulate cortex (ACC) in response to an acute morphine challenge, suggesting their unique role in modulating ACC-specific neuronal activity. These findings identify VTAMOR neurons as critical modulators of low sociability during protracted abstinence and highlight their potential as a mechanistic target to alleviate negative affective behaviors associated with opioid abstinence.
Significance Statement
The compelling urge for relief from negative affective states during long-term opioid abstinence presents a crucial challenge for maintaining abstinence. The ventral tegmental area (VTA) and its μ-opioid receptor-expressing (VTAMOR) neurons represent a critical target of opioidergic action that underlie dependence and abstinence. Chronic activation of VTAMOR neurons during opioid exposure induces maladaptations within these neurons and their structurally connected circuitries, which alter reward processing and contribute to negative affect. Using an oral morphine drinking paradigm to induce dependence, we demonstrate that withdrawal engages VTAMOR neurons and identify this neuronal population as key mediators of opioid abstinence-induced social deficits. These findings hold promise to inform development of targeted therapies aimed at alleviating negative affective states associated with protracted opioid abstinence.
Introduction
Protracted opioid abstinence—the weeks-to-month–long period after drug cessation—is marked by negative affective behaviors such as heightened anxiety, irritability, dysphoria, and anhedonia, which continue long after acute physical withdrawal symptoms have subsided. The compelling urge to relieve this negative state poses a significant challenge for sustained abstinence, with relapse rates among individuals with opioid use disorder exceeding 90% (Weiss, 2011; Nwaefuna, 2023). Preclinical models of protracted abstinence using primarily investigator-administered opioids have extensively assessed the enduring effects of negative affect during spontaneous withdrawal. These effects, including increased anxiety- and despair-related behaviors, anhedonia, and social deficits, persist long after the last opioid administration (Goeldner et al., 2011; Bravo et al., 2020; Welsch et al., 2020; Becker et al., 2021; Pomrenze et al., 2022; Ozdemir et al., 2023). These observations underscore the critical need to investigate the underlying neuroadaptations resulting from repeated opioid exposure and abstinence that facilitate the development of these maladaptive behaviors (Koob and Volkow, 2010, 2016).
Central to the brain's reward circuitry, the ventral tegmental area (VTA) is a critical site for opioid action and reinforcement and is enriched with μ-opioid receptor (MOR)-containing neurons. MORs are essential for opioid-induced analgesia, nociception, and dependence (Matthes et al., 1996), and their activation, specifically in the VTA, is necessary for opioid-induced reward (Phillips and LePiane, 1980; Britt and Wise, 1983; Olmstead and Franklin, 1997; Fields and Margolis, 2015) and the reinstatement of drug-seeking behavior (Stewart, 1984). Beyond their well-established role in opioidergic activity and reward processing, MORs have been proposed to play a fundamental role in mediating social behaviors, with previous studies demonstrating MORs are critical for social interaction (Becker et al., 2014; Pellissier et al., 2018; Toddes et al., 2021).
Although various regions such as the nucleus accumbens (Pomrenze et al., 2022; Fox et al., 2023), central amygdala (CeA; Jiang et al., 2021), bed nucleus of the stria terminalis (BNST; Delfs et al., 2000), habenula (Valentinova et al., 2019; Bailly et al., 2023), and dorsal raphe nucleus (DRN; Lutz et al., 2014; Pomrenze et al., 2022; Welsch et al., 2023) have been studied for their role in negative affect circuitry, the specific contributions of VTAMOR neurons to behavioral effects during protracted abstinence remain elusive. Characterized as a central hub within the mesocorticolimbic circuitry, the VTA's extensive afferent and efferent projection targets (Bouarab et al., 2019) position the region to influence a wide range of behaviors related not only to reward but also to stress and aversion. These unique structural connections, alongside the interplay of opioid action on VTAMOR neurons during morphine administration, suggest the fundamental role of the VTA in the emergence of maladaptive behaviors associated with repeated opioid use and withdrawal.
Here, we aimed to elucidate the contribution of VTAMOR neurons in the development of morphine dependence and negative affect during protracted abstinence. We utilized a chronic oral morphine drinking paradigm followed by a protracted forced abstinence period and conducted a targeted examination of VTAMOR neurons using MOR-specific mouse lines and Cre-dependent viral tools. Leveraging the genetic neural activity marker, FOS, we show that VTAMOR neurons across the anterior–posterior axis are activated during dependence and naloxone-precipitated withdrawal, indicating a neural adaptation within this cell type that may be involved in the long-lasting behavioral effects observed during protracted abstinence. Additionally, using viral antero- and retrograde tracing to map the input–output connectivity architecture of VTAMOR neurons, we discovered a diverse array of brain-wide connections with regions implicated in stress and negative affect, including the prefrontal cortex and locus ceruleus (LC). Notably, we demonstrated that VTAMOR neurons mediate reduced sociability during protracted abstinence through viral MOR re-expression within the VTA of MOR KO mice. Furthermore, these neurons contribute to increased FOS expression in the anterior cingulate cortex (ACC) following an acute morphine challenge injection, an effect that is diminished in mice with a prior history of morphine exposure. In total, our results highlight the critical role of VTAMOR neurons in facilitating low sociability during protracted opioid abstinence and propose this neuronal population as a potential target for interventions aimed at alleviating the distressing symptoms associated with protracted abstinence.
Materials and Methods
Animals
Male and female mice aged 8–20 weeks old were used from the following genetic lines: C57BL/6J mice were purchased from Taconic Biosciences and bred in-house at the University of Pennsylvania. A MOR-specific T2A-cleaved Cre-recombinase (Mengaziol et al., 2022) mouse line that will be referred to as Oprm1MOR-T2A-Cre (C57BL/6NTac-Oprm1em1(cre)Jabl/Mmnc; stock #070963-UNC) was generated from Dr. Julie Blendy's laboratory and bred in-house; briefly, the line was generated by inserting a T2A cleavable peptide sequence and the Cre coding sequence into the MOR 3′-UTR, as previously described (Mengaziol et al., 2022). An Oprm1Cre:GFP knock-in/knock-out mouse line that will be referred to as MOR KO (B6.Cg-Oprm1tm1.1(cre/GFP)Rpa/J; stock #035574) was generated from Dr. Richard Palmiter's laboratory and bred in-house; briefly, the MOR KO line was generated by inserting a cassette encoding Cre:GFP 5′ of the initiation codon in Oprm1's first coding exon, as previously described (Liu et al., 2021); and mice homozygous for Cre:GFP lack MOR expression and therefore do not respond to endogenous or exogenous ligands. For both the Oprm1MOR-T2A-Cre and MOR KO lines, only mice homozygous for Cre were used for behavioral studies. Littermate controls of the MOR KO line, Oprm1+/+, which lack the Cre allele, were used to validate expression using quantitative PCR (qPCR). Mice were maintained on a 12 h reverse light/dark cycle (lights on at 6:00 P.M.) and provided with food and water ad libitum. All mice in the drinking experiments were singly housed to measure individual daily volume consumption. All conducted experiments were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.
Stereotaxic surgeries and viral injections
Mice were anesthetized with isoflurane (2.5–3.0% for induction, 1.5–2.0% for maintenance) and head-fixed in a stereotaxic frame. For brain-wide MOR+ neuron labeling, AAV.PHP.eB-mMORp-eYFP (titer, 8.60 × 1011 GC/ml) was administered intravenously through the retro-orbital sinus using a 30 G insulin syringe, as previously described (Salimando et al., 2023). Other viruses in the study were injected intracranially into the VTA (AP, −3.2 mm; ML, ±0.55 mm; DV, −4.75 mm) using a 10 µl NanoFil Hamilton syringe (World Precision Instruments) with a 33 G beveled needle. For the rabies monosynaptic labeling, 150 nl of a 1:1 mixture of AAV5-CAG-FLEX-TVA-mCherry and AAV8-CAG-FLEX-RABV-G (titer, 1.00 × 1012 GC/ml, Beier Lab) was unilaterally injected into the left VTA of Oprm1MOR-T2A-Cre mice. Two weeks postinjection, 300 nl of RABVΔG-GFP (titer, 8.00 × 108 GC/ml, Beier Lab) was administered into the same injection site, and mice were killed 5 d later. To map axonal density outputs of VTAMOR neurons, we unilaterally injected 15 nl of AAV5-CAG-FLEX-tdTomato (titer, 2.10 × 1013 GC/ml; Addgene; catalog #28306-AAV5) into the left VTA of Oprm1MOR-T2A-Cre mice. For re-expression of VTAMORs, MOR KO mice received bilateral VTA injections of 300 nl per side of AAVDJ-hSyn1-FLEX-mCh-T2A-FLAG-hMOR-WPRE (AAV-hMOR; titer, 1.30 × 1012 GC/ml; Banghart Lab) or control AAV5-hSyn1-DIO-EGFP (AAV-EGFP; titer, 1.30 × 1012 GC/ml; Addgene; catalog #50457-AAV5). Following surgery, mice received meloxicam (5 mg/kg, s.c.) and recovered for 1–2 weeks before behavioral assessments.
Morphine treatment
Morphine sulfate (NIDA Drug Supply) was dissolved in water at an initial concentration of 0.3 mg/ml for the first 2 d. The concentration of the morphine solution escalated to 0.5 mg/ml for the following 3 d and was subsequently maintained at 0.75 mg/ml for the remainder of the total 13 d exposure paradigm, as previously established and described (Belknap, 1990). We employed a forced drinking paradigm as the morphine solution was presented in 15 ml tubes within the home cage as the sole drinking source for continuous 24 h access. Opioid-naive mice were administered water. Both the volumes of the morphine solution in the drinking tubes and the mice's body weights were recorded daily at 12:00 P.M.
Behavior
Mice were acclimated for 1 h in the behavior room prior to all behavior assessments.
Somatic withdrawal signs
For precipitated withdrawal, mice were injected with naloxone (1 mg/kg, s.c.) on Day 13 of the morphine exposure paradigm and subsequently placed in a clear observation box to monitor physical withdrawal signs. For spontaneous withdrawal, somatic signs were observed on Day 14, 24 h after cessation of morphine exposure. Somatic withdrawal signs were conducted for 20 min under indirect lighting (15 lux), during which the frequency of episodes of diarrhea, resting tremor, paw tremors, jumping, gnawing on limbs, genital licking and grooming, head shakes, body shakes, backing and defensive treading, and scratching were video-recorded and quantified, as previously described (Eacret et al., 2023) using an ethological keyboard in the BORIS (Friard and Gamba, 2016) software. The total frequency of each individual withdrawal behavior culminated to the global withdrawal score.
Social interaction
Sex-, weight-, and age-matched unfamiliar mice that were drug-naive were introduced into an open-field arena (40 × 40 cm) under indirect lighting (15 lux) for 10 min. Social interactions, including nose contacts (nose-to-nose, nose-to-body, or nose-to-anogenital region), grooming, and following episodes, were scored using an ethological keyboard in the BORIS software (Friard and Gamba, 2016), following established protocols (Goeldner et al., 2011; Lutz et al., 2014; Becker et al., 2021). The total duration of these episodes was summed as the social interaction duration.
Elevated zero maze
Mice were placed in an elevated zero maze (50 cm diameter, 5 cm track width, and 50 cm from the floor) under indirect lighting (15 lux) for 5 min. The total duration spent in the open and closed arms were recorded and analyzed using the Noldus EthoVision XT software.
Tail suspension test
Mice were suspended by the tail under indirect lighting (15 lux) for 6 min using a nonirritating adhesive tape. Total immobility duration was assessed, as previously described (Can et al., 2011).
Hot plate test
Mice were placed on a hot plate set to 50°C for 1 min under indirect lighting (15 lux). The duration of hindpaw and forepaw licking behaviors was recorded and analyzed using an ethological keyboard in BORIS (Friard and Gamba, 2016), and the total duration was recorded as the paw lick duration.
Immunohistochemistry
Ninety minutes following an injection of saline, naloxone (1 mg/kg, s.c.; Fig. 2A), or an acute morphine challenge (20 mg/kg, s.c.; Figs. 5A, 6A), mice were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and perfused with ice-cold 0.01 M PBS followed by 4% paraformaldehyde (PFA). Brains were left in 4% PFA at 4°C overnight and subsequently transferred to 30% sucrose in PBS. Brains were then sectioned frozen at −20°C on a cryostat (CryoStar NX50, Thermo Fisher Scientific), and 30-µm-thick coronal sections were collected and stored in 0.01 M PBS.
Sections were washed three times for 10 min in 0.01 M PBS and then incubated for 2 h in a blocking buffer solution consisting of 5% normal donkey serum and 0.1% Triton X-100 in 0.01 M PBS. Sections were incubated overnight in room temperature with either rabbit anti-c-Fos (1:500, Cell Signaling Technology, catalog #2250S), rabbit anti-dsRed (1:1,000, Takara, catalog #632496), chicken anti-GFP (1:500, Abcam, catalog #AB13970), or mouse anti-tyrosine hydroxylase (1:1,000, Sigma-Aldrich, catalog #MAB318) in the blocking buffer solution. Sections were then washed three times for 10 min in 0.01 M PBS the following day and incubated with Alexa Fluor 647 donkey anti-rabbit (1:500, Invitrogen Thermo Fisher A31573), Alexa Fluor 594 donkey anti-rabbit (1:500, Invitrogen Thermo Fisher, catalog #A21207), Alexa Fluor 488 donkey anti-chicken (1:500, Jackson ImmunoResearch Laboratories, catalog #703-545-155), or Alexa Fluor 647 donkey anti-mouse (1:500, Invitrogen Thermo Fisher, catalog #A31571) in the blocking buffer solution for 2 h at room temperature. Sections were then washed three times for 10 min in 0.01 M PBS, incubated in DAPI (1:10,000, Sigma-Aldrich, catalog #D9542), mounted, and then coverslipped prior to imaging on a Keyence BZ-X810 fluorescence microscope at 4× or 20× objectives.
Fluorescent in situ hybridization
Following the 13 d forced morphine drinking exposure, C57BL/6J mice were injected with naloxone (1 mg/kg, s.c.; Fig. 3A) on Day 13 to precipitate withdrawal. Twenty minutes postinjection, mice were anesthetized with isoflurane gas in oxygen. Brains were collected, embedded in TissueTek O.C.T (Thermo Fisher Scientific) and stored at −80°C. The 16-um-thick coronal VTA sections at −3.28 mm from the bregma were sectioned on a cryostat and mounted on SUPERFROST PLUS slides (Thermo Fisher Scientific). RNAscope fluorescent in situ hybridization was performed using the RNAscope Multiplex Fluorescent v2 Assay (Advanced Cell Diagnostics, catalog #323100) following the manufacturer's instructions for the fresh-frozen tissue. Briefly, slices were postfixed in prechilled 10% neutral buffered formalin at 4°C and then dehydrated through graded ethanol washes. Sections were stored overnight in 100% ethanol at −20°C. The next day, sections were air-dried and incubated with the provided hydrogen peroxide solution, then the provided Protease IV reagent, with washes in distilled water between steps. Sections were then immediately incubated in a mixture of cDNA hybridization probes. We used probes specific to the Mm-Oprm1-C1 (Advanced Cell Diagnostics, catalog #315841), Mm-Fos-C4 (Advanced Cell Diagnostics, catalog #316921-C4), and Mm-Slc17a6-E1-E3-C3 (i.e., Vglut2; ACDBio #456751-C3) or Mm-Slc32a1-C3 (i.e., Vgat; ACDBio #319191-C3), mixed according to the manufacturer's instruction to a ratio of 1:1:50 for C3:C4:C1. After washing in the provided wash buffer, three sequential amplifications were performed using the provided AMP1, AMP2, and AMP3 reagents, with washes in the provided wash buffer between steps. Sections were then fluorescently stained using the provided TSA Plus HRP solution, followed by incubation with the selected fluorescent dye and finally the provided HRP blocker solution, with washes in wash buffer between each step. This process was repeated for each channel. We used the fluorescent dyes TSA Vivid 520 (ACDBio #323271), TSA Vivid 570 (ACDBio #323272), and TSA Vivid 650 (ACDBio #323273), all diluted to 1:3,000 in the provided TSA buffer. Sections were counterstained using the provided DAPI reagent, coverslipped, and imaged on a Keyence BZ-X800 fluorescent microscope at 40× objective. Cell quantification was completed semimanually using the FISH analysis module in the HALO-AI software (Indica Labs). Only puncta associated with DAPI-labeled nuclei were included in analysis.
Quantification
Naloxone-induced FOS expression in VTAMOR neurons
FOS- and YFP-labeled MOR neurons were quantified across the anterior to posterior axis of the VTA. Ten coronal VTA sections per mouse were assigned as anterior (−2.80, −2.92, and −3.08 mm from the bregma), central (−3.16, −3.28, and −3.4 mm from the bregma), or posterior (−3.52, −3.64, −3.8, and −3.88 mm from the bregma) VTA sections, as based on the Franklin and Paxinos mouse brain atlas, third edition (Franklin and Paxinos, 2007). The percentage of FOS+ MOR+ neurons was calculated by dividing the total number of colabeled FOS+ MOR+ neurons by the total number of MOR+ neurons in the VTA for each coronal section.
Acute morphine-induced FOS expression
FOS expression was analyzed in 22 regions implicated in opioid abstinence (Welsch et al., 2020) and identified as key VTA afferent and efferent projection targets (Beier et al., 2015). These regions included the orbital cortex (ORB), ACC, infralimbic cortex, insular cortex, nucleus accumbens core, nucleus accumbens shell, caudate putamen (CP), globus pallidus (GP), BNST, ventral pallidum (VP), CeA, basolateral amygdala (BLA), hippocampus (HIP), paraventricular nucleus of the thalamus, lateral habenula (LHb), medial habenula (MHb), lateral hypothalamus (LH), VTA, periaqueductal gray (PAG), DRN, parabrachial nucleus (PBN), and the LC. FOS expression was quantified and analyzed using a semimanual detection method previously developed in our lab (Xie et al., 2024).
Rabies monosynaptic VTAMOR input labeling
Every sixth 30-µm-thick coronal section throughout the entire brain was included in analysis. GFP-positive VTAMOR input cells from the left hemisphere were analyzed using Semi-Automated Workflow for Brain Slice Histology Alignment, Registration, and Cell Quantification (Lauridsen et al., 2022) and quantified, as previously described (Schwarz et al., 2015; Xie et al., 2024). Counts from each region were adjusted by a factor of 6 to account for the sampling rate, as previously reported (Beier et al., 2015). The fraction of GFP+ counts per region by the total sum of all rabies-labeled inputs throughout the entire brain was assessed for these regions chosen for analysis, which included major VTA inputs outside of the excluded region near the injection site, as previously described (Beier et al., 2015).
Axon density quantification of VTAMOR outputs
Four 30-µm-thick coronal sections per region were selected across the region's anterior to posterior segment. An inclusive set of regions was selected based on their involvement in mediating opioid dependence and abstinence (Welsch et al., 2020; Ozdemir et al., 2023), as well as their structural connectivity to the VTA (Beier et al., 2015). Coronal histology images were processed using AxoDen, a custom-written algorithm developed in the Corder Lab (https://corderlab.com) for the automated quantification of axonal density in defined regions (https://github.com/raqueladaia/AxoDen; web application link: https://axoden.streamlit.app/; Sandoval Ortega et al., 2024). Briefly, this algorithm converts images to gray scale and later employs dynamic thresholding to accurately binarize the image and segregate signal from background fluorescence. This method facilitates the quantification of the density of axon projections in selected brain regions.
qPCR for mouse Oprm1 and human OPRM1
Tissue samples from the VTA were collected as 1 mm punches from Oprm1+/+ mice (littermate controls of the MOR KO line), MOR KO mice, and MOR KO mice with VTAMOR re-expression through bilateral injections of AAVDJ-hSyn1-FLEX-mCh-T2A-FLAG-hMOR-WPRE (AAV-hMOR). Both the mouse Oprm1 and human OPRM1 transcripts were quantified. Total tissue RNA was extracted with RNAzol (Sigma-Aldrich, R4533) according to manufacturer's protocol, and cDNA was synthesized from 1 µg RNA (Applied Biosystems, catalog #4374966). cDNA was diluted 1:10 and assessed for mRNA transcript levels by qPCR with SYBR Green Mix (Applied Biosystems, catalog #A25741) on a QuantStudio7 Flex Real-Time PCR System (Thermo Fisher Scientific). Oligonucleotide primer sequences for target and reference genes are as follows: mouse_Oprm1 (forward, CTGCAAGAGTTGCATGGACAG; reverse, TCAGATGACATTCACCTGCCAA); human_OPRM1 (forward, ACTGATCGACTTGTCCCACTTAGATGGC; reverse, ACTGACTGACTGACCATGGGTCGGACAGGT); and mouse_L30 (forward, ATGGTGGCCGCAAAGAAGACGAA; reverse, CCTCAAAGCTGGACAGTTGTTGGCA). Target gene expression was normalized to an internal control (mL30 mRNA) and sample control (Oprm1+/+) and analyzed by the 2−ΔΔCT method (Livak and Schmittgen, 2001). Each sample reaction was performed in triplicate.
Experimental design and statistical analysis
Behavioral assessments
Behavioral experiments were conducted to assess the effects of chronic morphine exposure on somatic withdrawal signs alongside negative affective behaviors. Mice underwent a 13 d morphine drinking paradigm followed by behavioral experiments including somatic withdrawal signs, social interaction, elevated zero maze, tail suspension test, and hot plate test. For C57BL/6J mice, 37 mice (n = 23 males; n = 14 females) were divided into two groups: opioid-naive (n = 16) and morphine-treated (n = 21). Behavioral data were analyzed between opioid-naive and morphine-treated groups using unpaired t tests (Fig. 1D–H, 2B). For MOR KO mice, 48 mice (n = 28 males; n = 20 females) were divided into four groups (n = 10–13/group): AAV-EGFP virus + opioid-naive, AAV-EGFP virus + morphine-treated, AAV-hMOR virus + opioid-naive, and AAV-hMOR virus + morphine-treated (Fig. 6F–J). A two-way ANOVA and Bonferroni's multiple-comparison test was performed to analyze the factors of viral injection (AAV-hMOR vs AAV-EGFP) and treatment condition (opioid-naive vs morphine-treated).
Neuronal activation studies
Neuronal activation was assessed through FOS expression in male and female C57BL/6J and MOR KO mice at the culmination of behavioral assessments. To assess precipitated withdrawal-induced FOS expression in VTAMOR neurons, mice were divided into two groups (n = 3–5/group): opioid-naive + naloxone injection and morphine-treated + naloxone injection (Fig. 2A). To assess potential subregional differences in neuronal activation within the VTA, we analyzed 10 sections per mouse spanning the anterior to posterior VTA axis from −2.80 to −3.88 mm from the bregma. An unpaired t test was used to assess neuronal activation in the opioid-naive versus morphine-treated groups (Fig. 2D,G), or two-way ANOVA and Bonferroni's multiple comparison was performed to evaluate neuronal activation across treatment groups (opioid-naive vs morphine-treated) and VTA subregions (anterior, central, and posterior; Fig. 2E,H) or specific bregma levels (Fig. 2F,I). To assess the effects of a morphine challenge injection during protracted morphine abstinence on neuronal activation, we analyzed FOS expression in 22 regions (Fig. 5B) after a morphine challenge during 4 week protracted abstinence across four groups (n = 4–6/group): opioid-naive + saline injection, morphine-treated + saline injection, opioid-naive + morphine challenge injection, and morphine-treated + morphine challenge injection. Two-way ANOVA and Bonferroni's multiple comparisons were performed to evaluate factors of treatment (opioid-naive vs morphine-treated) and injection (saline vs morphine challenge) on neuronal activation (Fig. 5B–F). In MOR KO mice, neuronal activation in the ACC and LC was analyzed following a morphine challenge injection during the 4 week protracted abstinence across eight groups (n = 5–6/group): AAV-EGFP + opioid-naive + saline injection, AAV-EGFP + morphine-treated + saline injection, AAV-hMOR + opioid-naive + saline injection, AAV-hMOR + morphine-treated + saline injection, AAV-EGFP + opioid-naive + morphine challenge injection, AAV-EGFP + morphine-treated + morphine challenge injection, AAV-hMOR + opioid-naive + morphine challenge injection, and AAV-hMOR + morphine-treated + morphine challenge injection (Fig. 7B,D). We performed a three-way ANOVA and Bonferroni's multiple comparisons to assess viral groups (AAV-hMOR vs AAV-EGFP), treatment conditions (opioid-naive vs morphine-treated), and injection (saline vs morphine challenge) on neuronal activation.
Structural input–output connectivity studies
For the VTAMOR structural connectivity mapping, 6 Oprm1MOR-T2A-Cre mice (n = 4 males; n = 2 females) were used for the rabies virus monosynaptic labeling (n = 3) or axonal output mapping (n = 3). Although the use of n = 3 per group in our baseline connectivity mapping of VTAMOR neurons is a lower sample size, this sample size is within range of established literature using similar tracing techniques (Beier et al., 2015, 2017), and our objective was to comprehensively map out the native structural connectivity of VTAMOR neurons in an opioid-naive state.
All data are presented as mean ± the standard error of the mean (SEM) for each group and analyzed using an unpaired t test or one-way, two-way, or three-way ANOVA and Bonferroni's multiple comparison. Statistical analyses were performed using Prism 10 (GraphPad). Statistical outliers were identified using Grubbs’ test with an α = 0.05 significance level. One statistically significant outlier was excluded from the MHb FOS expression analysis in Figure 5. One statistically significant outlier from the social interaction test and one statistically significant outlier from the elevated zero maze test were excluded from behavioral assessment in Figure 6.
Results
Protracted morphine abstinence leads to the emergence of negative affective behaviors
To establish a model of morphine dependence and negative affective behaviors during protracted abstinence, we employed a chronic, forced oral morphine drinking paradigm with escalating concentrations over 13 d, followed by 4 weeks of protracted abstinence, and a behavioral battery (social interaction test, elevated zero maze, tail suspension test, and hot plate test; Fig. 1A). Morphine-treated mice show an increased volume consumption of morphine relative to Day 1 throughout the paradigm (Fig. 1B). This increase corresponds to the experimental escalation of morphine concentration, which results in a progressive increase in morphine dose intake over the 13 d exposure period (Fig. 1C). Twenty-four hours following the cessation of morphine administration, mice undergoing spontaneous withdrawal demonstrate physical dependence, as shown through an increased global withdrawal score relative to opioid-naive controls (Fig. 1D; unpaired t test; t = 8.014; df = 27; p < 0.0001). This result is driven by increased frequencies of paw tremors, head shakes, body shakes, and genital licking and grooming bouts (Extended Data Fig. 1-1B–E).
Protracted morphine abstinence leads to the emergence of negative affective behaviors. A, Experimental schematic of the 13 d chronic forced morphine drinking exposure with escalating concentrations, assessment of somatic withdrawal signs 24 h following cessation of morphine administration, a 4 week protracted abstinence period, and behavioral battery assessment (social interaction, elevated zero maze, tail suspension test, and hot plate test). B, Daily consumed volumes in the opioid-naive and morphine-treated groups across the 13 d drinking paradigm. C, Daily morphine sulfate dose administration across the 13 d drinking exposure period. D, At 24 h following cessation of morphine administration, morphine-treated mice undergoing spontaneous abstinence exhibit an increased global withdrawal score (n = 13 opioid-naive/16 morphine-treated, unpaired t test; t = 8.014; df = 27; p < 0.0001). The individual somatic signs contributing to the global withdrawal score are detailed in Extended Data Figure 1-1. At 4 weeks of protracted morphine abstinence, mice display (E) low sociability as assessed through the social interaction test (unpaired t test; t = 2.974; df = 27; p = 0.0061), (F) anxiety-like behavior as assessed through the elevated zero maze assay (unpaired t test; t = 2.357; df = 27; p = 0.0259), (G) increased immobility as evaluated through the tail suspension test (unpaired t test; t = 3.937; df = 26; p = 0.0006), and (H) unaltered nociceptive hypersensitivity as assessed through the hot plate test (unpaired t test; t = 0.6185; df = 27; p = 0.5414). Data are represented as mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.
Figure 1-1
Individual somatic withdrawal signs in morphine-dependent mice undergoing 24-hour spontaneous abstinence. (A) Morphine-dependent mice undergoing spontaneous withdrawal demonstrate increased frequencies of (B) paw tremors (unpaired t test, t = 7.187, df = 27, p < 0.0001), (C) head shakes (unpaired t test, t = 2.443, df = 27, p = 0.0214), and (d) body shakes (unpaired t test, t = 7.342, df = 27, p < 0.0001), while the frequencies of (A) resting tremors, (E) genital licking and grooming, (F) gnawing on limb, and (G) backing and defensive treading bouts (H) diarrhea, (I) jumps, and (J) digging bouts remain unchanged relative to opioid naïve. Data are represented as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Download Figure 1-1, TIF file.
Following 4 weeks of forced protracted morphine abstinence, morphine-treated mice demonstrate social deficit-like behavior in the social interaction test, as demonstrated through a decreased social interaction duration (Fig. 1E; unpaired t test; t = 2.974; df = 27; p = 0.0061). Morphine-treated mice also exhibited increased anxiety-like behavior in the elevated zero maze, as shown through decreased time spent in the open arms (Fig. 1F; unpaired t test, t = 2.357, df = 27, p = 0.0259), as well as increased despair-related behavior through an increased duration of immobility time in the tail suspension test (Fig. 1G; unpaired t test; t = 3.937; df = 26; p = 0.0006). Opioid-naive and morphine-treated mice did not show differences in the total paw lick duration in the hot plate test (Fig. 1H; unpaired t test; t = 0.6185; df = 27; p = 0.5414), indicating a lack of persistent alterations in nociceptive hypersensitivity during protracted morphine abstinence.
VTAMOR neurons show increased neural activation during withdrawal in morphine-dependent mice
Having observed the emergence of negative affective behaviors, including reduced social interactions, during protracted morphine abstinence, we next sought to determine if VTAMOR neurons are engaged during dependence. Given the prominent role of the VTA in opioid withdrawal and social reward and dysfunction (Kaufling and Aston-Jones, 2015; Bariselli et al., 2016; Hung et al., 2017; Langlois and Nugent, 2017; Porcelli et al., 2019; Solié et al., 2022), we aimed to assess the withdrawal-related activation of VTAMOR neurons, as these opioid receptor-expressing cell types would be directly engaged throughout morphine exposure and thus subject to intracellular adaptations leading to dependence. Identifying the anatomical location of precipitated withdrawal-active VTAMOR neurons is key to understanding their role in the early stages of dependence and how they may contribute to the enduring effects observed during protracted abstinence. Given the low reported expression of FOS during spontaneous withdrawal (removal of morphine drinking access), we used the MOR antagonist naloxone to precipitate withdrawal and provide a salient stimulus for induction of the immediate early gene, c-fos. This precipitated approach allowed us to localize the activation of specific neural populations in the VTA during the critical window during the transition from acute withdrawal to protracted morphine abstinence.
To determine if MOR-expressing neurons exhibit differential neuronal activation during morphine withdrawal, we employed a targeted adeno-associated viral (AAV) approach to label MOR-containing neurons, which allowed us to label FOS in these neurons. A retro-orbital injection of AAV.PHP.eB-mMORp-eYFP (a synthetic MOR promoter driving enhanced yellow fluorescent protein; Salimando et al., 2023) was administered prior to initiating the 13 d chronic morphine exposure paradigm that culminated in naloxone-precipitated withdrawal and immediate assessment of somatic withdrawal signs (Fig. 2A,C). Mice exposed to the chronic morphine drinking paradigm with escalating concentrations show a heightened global withdrawal score following naloxone-precipitated (1 mg/kg, s.c.) withdrawal, compared with opioid-naive control mice (Fig. 2B; unpaired t test; t = 14.14; df = 6, p < 0.0001), which confirmed the presence of physical dependence to morphine. The increased global withdrawal score was primarily driven increased frequencies of diarrhea and jumps (Extended Data Fig. 2-1H–I).
VTAMOR neurons show increased neural activation during withdrawal in morphine-dependent mice. A, Experimental schematic of a retro-orbital injection of AAV.PHP.eB-mMORp-eYFP, followed by 13 d chronic morphine drinking with escalating concentrations, naloxone-precipitated withdrawal, assessment of somatic withdrawal signs, and tissue collection 90 min postnaloxone injection. B, Morphine-treated mice undergoing withdrawal demonstrate an increased global withdrawal score relative to opioid-naive mice (n = 3 opioid-naive/5 morphine-treated; unpaired t test; t = 14.14; df = 6; p < 0.0001). The individual somatic signs contributing to the global withdrawal score are detailed in Extended Data Figure 2-1A–J. C, Representative images of coronal sections showing mMORp-eYFP and FOS staining [left, coronal section of VTA (dashed white box highlights VTA); scale bar, 1 mm] in the anterior, central, and posterior VTA in opioid-naive and morphine-treated conditions (right; scale bar , 200 µm). Representative images of tyrosine hydroxylase expression across the anterior, central, and posterior VTA are shown in Extended Data Figure 2-1K. D, Quantification of precipitated withdrawal-induced FOS in the VTA shows increased FOS expression in morphine-dependent mice (unpaired t test; t = 2.849; df = 6; p = 0.0292), an effect observed in the (E) central and posterior VTA [two-way ANOVA + Bonferroni's multiple comparison; subregion F(2,18) = 10.96; p = 0.0008; treatment F(1,18) = 20.27; p = 0.0003; interaction (subregion × treatment) F(2,18) = 1.834; p = 0.1884]. F, FOS quantifications across varying bregma levels of the VTA show increased FOS expression at −3.80 mm from the bregma in morphine-treated mice relative to opioid-naive counterparts [two-way ANOVA + Bonferroni's multiple comparison, bregma F(9,59) = 2.476; p = 0.0180; treatment F(1,59) = 31.47; p < 0.0001; interaction (bregma × treatment) F(9,59) = 0.5615; p = 0.8228]. G, Morphine-dependent mice exhibit an increased percentage of FOS+ VTAMOR neurons (unpaired t test; t = 7.661; df = 6; p = 0.0003), an effect observed in the (H) anterior, central, and posterior VTA [two-way ANOVA + Bonferroni's multiple comparison; subregion F(2,18) = 14.71; p = 0.0002; treatment F(1,18) = 108.6; p < 0.0001; interaction (subregion × treatment) F(2,18) = 3.100; p = 0.0697]. I, The percentage FOS+ VTAMOR neurons across varying bregma levels shows an increased percentage at −3.28, −3.40, −3.52, −3.80, and −3.88 mm from the bregma in morphine-dependent mice [two-way ANOVA + Bonferroni's multiple comparison; bregma F(9,59) = 2.654; p = 0.0118; treatment F(1,59) = 68.84; p < 0.0001; interaction (region × treatment) F(9,59) = 1.092; p = 0.3828]. Data are represented as mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.
Figure 2-1
Individual somatic withdrawal signs in morphine-dependent mice undergoing precipitated withdrawal and neuronal activation of MOR neurons in the anterior, central, and posterior VTA. Morphine-dependent mice undergoing precipitated withdrawal demonstrate increased frequencies of (H) diarrhea (unpaired t test, t = 3.509, df = 6, p = 0.0127) and (I) jumps (unpaired t test, t = 5.046, df = 6, p = 0.0023), while the frequencies of (A) resting tremors, (B) paw tremors, (C) head shakes, (D) body shakes, (E) genital licking and grooming, (F) gnawing on limb, and (G) backing and defensive treading bouts remained unchanged relative to opioid naïve mice. (K) Representative images of coronal sections showing tyrosine hydroxylase (TH) staining with mMORp-eYFP and FOS staining in the anterior, central, and posterior VTA (scale, 200 µm). Data are represented as mean ± SEM, *p < 0.05, **p < 0.01. Scale, 200 µm. Download Figure 2-1, TIF file.
We first evaluated whether the VTA shows differential withdrawal-induced neuronal activation (Fig. 2D–F). We observed an increase in naloxone-precipitated FOS expression across the anterior–posterior axis of the VTA in morphine-treated mice relative to opioid-naive mice (Fig. 2D; unpaired t test; t = 2.849; df = 6; p = 0.0292). Interestingly, a significant elevation in withdrawal-induced FOS was observed in the central and posterior VTA relative to the anterior subregion [Fig. 2E; two-way ANOVA + Bonferroni's multiple comparison, subregion F(2,18) = 10.96; p = 0.0008; treatment F(1,18) = 20.27; p = 0.0003; interaction (subregion × treatment) F(2,18) = 1.834; p = 0.1884]. Indeed, a higher percentage of withdrawal FOS+ MOR+ neurons was observed throughout the VTA (Fig. 2G; unpaired t test; t = 7.661; df = 6; p = 0.0003) across the anterior, central, and posterior VTA in morphine-treated mice [Fig. 2H; two-way ANOVA + Bonferroni's multiple comparison, subregion F(2,18) = 14.71; p = 0.0002; treatment F(1,18) = 108.6; p < 0.0001; interaction (subregion × treatment) F(2,18) = 3.100; p = 0.0697]. This finding highlights the altered activation of VTAMOR neurons following the development of dependence and at the initiation of withdrawal, suggesting that this neural population undergoes adaptive processes that may contribute to the negative affective behavioral effects observed during protracted abstinence.
Neurochemical identification of withdrawal-active VTAMOR neurons
Given prior evidence that MORs in the VTA are expressed in both GABAergic and glutamatergic neurons (McGovern et al., 2023), we next investigated the neurochemical identity of VTAMOR neurons activated during withdrawal using fluorescent in situ hybridization. C57BL/6J mice were exposed to the 13 d chronic morphine exposure drinking paradigm, which culminated in naloxone-precipitated withdrawal prior to brain collection (Fig. 3A). Thin coronal VTA sections from these mice were assessed for the expression of transcripts encoding Oprm1 (MOR-expressing neurons), Vglut2 (glutamate-releasing neurons), Vgat (GABA-releasing neurons), and Fos (withdrawal-active neurons). Within the VTA, an average of 26.2% of neurons exhibiting withdrawal-induced activation expressed Vglut2 and Oprm1 (Fig. 3C), and an average of 31.3% of neurons exhibiting withdrawal-induced activation expressed Vgat and Oprm1 (Fig. 3E). In the Vglut2+ population, we observed a trend suggesting that a greater percentage of withdrawal-activated neurons are Vglut2+/Oprm1− or Vglut2+/Oprm1+ relative to Vglut2-/Oprm1+ (Fig. 3C; RM one-way ANOVA, cell type F(1.1,3.3) = 0.6817; p = 0.4778). Within the Vgat+ population, we observed an increased percentage of withdrawal-active cells that are Vgat+/Oprm1− or Vgat+/Oprm1+ relative to Vgat−/Oprm1+ neurons (Fig. 3E; RM one-way ANOVA, cell type F(1.2,4.9) = 20.02; p = 0.0058). These findings suggest that a greater proportion of GABAergic MOR neurons may be activated during withdrawal compared with MOR-expressing neurons without Vgat+ coexpression. Together, these data suggest that VTAMOR neurons are activated during withdrawal and include populations that express markers of GABAergic and glutamatergic neurotransmission.
Neurochemical identification of withdrawal-active VTAMOR neurons. A, Experimental schematic illustrating the 13 d chronic morphine drinking with escalating concentrations, naloxone-precipitated withdrawal, and tissue collection 20 min postnaloxone injection. B, Representative VTA coronal section exhibiting RNA transcripts for Oprm1, Fos, and Vglut2. C, Percentages of withdrawal-activated neurons among Vglut2+ and Oprm1+ populations (n = 4 morphine-treated; RM one-way ANOVA; cell type F(1.1,3.3) = 0.6817; p = 0.4778). D, Representative VTA coronal section showing RNA transcripts for Oprm1, Fos, and Vgat. E, Percentages of withdrawal-activated neurons among Vgat+ and Oprm1+ populations (n = 5 morphine-treated, RM one-way ANOVA, cell type F(1.2,4.9) = 20.02; p = 0.0058). Data are represented as mean ± SEM.
Structural input–output mapping of VTAMOR neurons
To further investigate the circuitry associated with the activation of VTAMOR neurons during abstinence, we mapped the structural connectivity of this VTA cell type. Identifying the direct input and output structural connections to and from VTAMOR neurons can establish the distribution architecture of abstinence-related neurotransmissions during early and protracted morphine abstinence. We first used Oprm1MOR-T2A-Cre mice and Cre-dependent AAV vectors for retrograde and anterograde brain-wide tracing. While the structural circuit architecture of dopaminergic and GABAergic neurons in the VTA is known (Sesack and Grace, 2010; Russo and Nestler, 2013; Beier et al., 2015), the specific afferent and efferent projections of VTAMOR neurons have not been directly explored. Using a modified rabies virus for monosynaptic input labeling in Oprm1MOR-T2A-Cre mice (Fig. 4A), we injected Cre-dependent helper viruses and the rabies virus in the VTA (Fig. 4B). We found that VTAMOR neurons receive a diverse range of monosynaptic inputs (Fig. 4C,D; Extended Data Fig. 4-1B; Table 1) from regions critical to reward and aversion processing, such as the striatum (NAcC, NAcSh, DLS, DMS, VLS, VMS), pallidum (VP, GP, BNST), amygdala (CeA), thalamic nuclei (MD, LHb, MHb), hypothalamus (LH, ZI, POA), midbrain structures (PAG, DRN), and the pons (PBN).
Structural input–output mapping of VTAMOR neurons. A, Experimental schematic illustrating rabies-mediated monosynaptic labeling of VTAMOR inputs in Oprm1MOR-T2A-Cre mice through a unilateral injection of AAVs expressing Cre-dependent TVA-mCherry (AAV5-CAG-FLEX-TVA-mCherry) and rabies glycoprotein (AAV8-CAG-FLEX-RABV-G) helper viruses in the VTA, followed by a subsequent injection of the EnvA-pseudotyped, G-deleted, GFP-expressing rabies virus (RABVΔG-GFP) into the same VTA injection site. B, Representative image of the unilateral VTA injection site in Oprm1MOR-T2A-Cre mice (scale bar, 200 µm). C, Coronal images of rabies-infected GFP–positive VTAMOR inputs (scale bar, 1,000 µm). Magnified images of regions with denser GFP-positive expression are provided in Extended Data Figure 4-1B. D, Quantification of VTAMOR inputs presented as the fraction of GFP-positive counts per region by the total sum of all rabies-labeled inputs (n = 3 Oprm1MOR-T2A-Cre mice). E, Experimental schematic of anterograde axon tracing through a unilateral viral injection of AAV5-FLEX-tdTomato in the VTA of Oprm1MOR-T2A-Cre mice. F, Representative image of the unilateral VTA injection site of the Cre-dependent anterograde tracer (scale bar, 200 µm). G, Coronal images of mCherry-positive VTAMOR axon density projections (scale bar, 1,000 µm). Magnified images of regions with denser mCherry-positive expression are provided in Extended Data Figure 4-1D. H, Quantification of VTAMOR axon density outputs (n = 3 Oprm1MOR-T2A-Cre mice) . Data are represented as mean ± SEM.
Figure 4-1
Rabies-mediated monosynaptic inputs to VTAMOR neurons and anterograde tracing of VTAMOR axonal outputs. (A) Schematic of helper AAVs and rabies virus unilaterally injected in the VTA of Oprm1MOR-T2A-Cre mice to map (B) monosynaptic inputs of VTAMOR neurons. (C) Anterograde virus unilaterally injected into the VTA of Oprm1MOR-T2A-Cre mice to label (D) axonal outputs of VTAMOR neurons. Scale, 1000 µm. Download Figure 4-1, TIF file.
Abbreviations of regions
We also mapped the axonal projections of VTAMOR outputs through anterograde tracing also in Oprm1MOR-T2A-Cre mice (Fig. 4E) through an injection of a Cre-dependent anterograde tracer into the VTA (Fig. 4F). Axonal density analyses show VTAMOR neurons project densely to the striatum (NAcC, NAcSh, DLS, DMS, VLS, VMS), pallidum (VP, GP, BNST), the amygdala (CeA), thalamic nuclei (AV, AM, AD, CM, MD, VL, VM, LHb), hypothalamus (ZI, LH, POA), HIP (DG, CA1-3), midbrain (PAG, DRN), and pons (LC; Fig. 4G–H; Extended Data Fig. 4-1D; Table 1). Together, these extensive and diverse structural connections of VTAMOR neurons with key regions involved in processing reward and aversion (Russo and Nestler, 2013) suggest how these neurons may facilitate the development of negative affect throughout dependence and abstinence, a process likely involving disrupted reward processing (Koob and Volkow, 2016).
Morphine abstinence leads to decreased neuronal activation following an acute morphine challenge in protracted abstinence in the ACC and LC
From the comprehensive, brain-wide structural connectivity map of VTAMOR neurons, we next aimed to understand how these anatomically connected regions, as well as additional structures implicated in abstinence, respond to an acute morphine challenge injection in mice undergoing protracted morphine abstinence. Specifically, we sought to determine whether a prior history of chronic morphine exposure, compared with opioid-naive mice, would induce enduring alterations in patterns of neural activation in response to a dose of morphine previously shown to produce rewarding effects (Do Couto et al., 2003, 2005; Brynildsen et al., 2020b; Ahmadian Salami et al., 2022). To this end, within the same behavioral cohort from Figure 1, we examined FOS expression across 22 regions that are structurally connected to VTAMOR neurons or involved in abstinence processes following a morphine challenge administered 90 min prior to tissue collection (Fig. 5A; Extended Data Table 5-1).
Protracted morphine abstinence leads to decreased neuronal activation following acute morphine challenge in the ACC and LC. A, Experimental schematic of the 13 d chronic forced morphine drinking exposure with escalating concentrations, assessment of somatic withdrawal signs 24 h following cessation of morphine administration, a 4 week protracted abstinence period, behavioral battery assessment, and tissue collection 90 min following a morphine challenge injection. B, Neuronal activation as represented by FOS density in opioid-naive and morphine-treated mice following a saline or morphine challenge injection (20 mg/kg, s.c.) across 22 regions connected to VTAMOR neurons or associated with withdrawal. Chronic morphine exposure results in decreased neuronal activation following an acute morphine challenge in protracted abstinence in the (C, D) ACC (two-way ANOVA + Bonferroni's multiple comparison; treatment F(1,16) = 4.312; p = 0.0543; injection F(1,16) = 190.3; p < 0.0001; interaction F(1,16) = 6.456; p = 0.0218; scale bar, 200 µm) and (E, F) LC (two-way ANOVA + Bonferroni's multiple comparison; treatment F(1,15) = 5.847; p = 0.0288; injection F(1,15) = 14.48; p = 0.0017; interaction F(1,15) = 9.757; p = 0.0070; scale bar, 200 µm). Two-way ANOVA statistics across all 22 assessed regions are provided in Extended Data Table 5-1. Data are represented as mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.
Table 5-1
Two-Way ANOVA results for neuronal activation in response to drinking and injection in Figure 5. Download Table 5-1, DOCX file.
Opioid-naive mice displayed increased FOS expression in numerous regions following an acute morphine injection relative to saline-injected mice, including the ORB, ACC, IL, IC, NAcC, NAcSh, GP, CeA, BLA, MHb, LH, VTA, and LC (Fig. 5B). This widespread activation is consistent with our lab's previous findings that acute morphine elicits robust FOS expression in opioid-naive mice (Brynildsen et al., 2020a). Our finding that an acute morphine injection increases FOS in the VTA of opioid-naive animals (Fig. 5B) may be largely attributed to the activation of dopaminergic neurons within this region (Georges et al., 2006; Corre et al., 2018). The VTA represents a key component of the mesocorticolimbic dopamine circuitry, and prior studies have found that morphine increases the impulse activity and burst firing of VTA dopamine neurons (Georges et al., 2006) and that opioids such as heroin activate dopaminergic neurons in the medial VTA (Corre et al., 2018). In mice with a prior history of chronic morphine exposure, an acute morphine injection induced heightened FOS density relative to saline-injected animals, albeit in a more focused set of regions including the ORB, ACC, IL, IC, CP, GP, MHb, and LH.
Interestingly, in mice with a prior history of chronic morphine exposure, a morphine challenge injection during protracted abstinence induced blunted FOS expression relative to opioid-naive mice in the ACC (Fig. 5C,D; two-way ANOVA + Bonferroni's multiple comparison; treatment F(1,16) = 4.312; p = 0.0543; injection F(1,16) = 190.3; p < 0.0001; interaction F(1,16) = 6.456; p = 0.0218) and in the LC (Fig. 5E,F; two-way ANOVA + Bonferroni's multiple comparison; treatment F(1,15) = 5.847; p = 0.0288; injection F(1,15) = 14.48; p = 0.0017; interaction F(1,15) = 9.757; p = 0.0070). This decreased neuronal activation in response to a dose of morphine previously shown to produce rewarding effects suggests persistent neuroadaptive changes resulting from chronic morphine exposure and subsequent 4 week abstinence. These changes may contribute to dysregulated neural activity underlying maladaptive negative affect behaviors observed during abstinence. These findings suggest that the ACC and LC may be uniquely susceptible to chronic morphine and protracted abstinence-induced plasticity, potentially contributing to the persistence of negative affective states during abstinence.
VTAMORs mediate low sociability during protracted morphine abstinence
Our investigations have highlighted VTAMOR neurons' increased activation following the development of dependence and at the onset of morphine withdrawal. Furthermore, our mapping of the baseline structural connectivity of VTAMOR neurons to diverse projections targets implicated in reward, stress, and aversion indicates that this population could contribute to the negative affect behaviors characteristic of protracted abstinence. Building on these findings, we next sought to directly assess the sufficiency of VTAMORs in mediating protracted abstinence-induced negative affect behaviors.
To re-express MORs specifically in the VTA, we performed bilateral injections of an AAV expressing a Cre-dependent human MOR (AAV-hMOR; AAVDJ-hSyn1-FLEX-mCh-T2A-FLAG-hMOR-WPRE) or a control virus (AAV-EGFP; AAV5-hSyn-FLEX-EGFP) into the VTA of Cre-expressing MOR KO mice (Fig. 6A,B). To validate our viral VTAMOR re-expression in the MOR KO line, we used qPCR and confirmed reduced expression of mouse Oprm1 in this line (Fig. 6C; unpaired t test; t = 8.949; df = 10; p < 0.0001). Further qPCR analysis confirmed the viral-mediated re-expression of human OPRM1 in the VTA of MOR KO mice (Fig. 6D; unpaired t test, t = 8.551; df = 5; p = 0.0004), shown through increased human OPRM1 expression compared with baseline levels in MOR KO mice.
VTAMORs mediate protracted abstinence-induced low sociability. A, Experimental schematic illustrating VTAMOR re-expression in MOR KO mice through bilateral VTA injections of AAVDJ-hSyn1-FLEX-mCh-T2A-FLAG-hMOR-WPRE, followed by the protracted morphine withdrawal paradigm and culminating in a morphine challenge injection (20 mg/kg, s.c.). Total consumed volumes across all groups are provided in Extended Data Figure 6-1A. B, Representative images of AAV-hSyn-EGFP viral control or AAVDJ-hSyn1-hMOR in the VTA injection site (scale bar, 200 µm). C, Validation of MOR knock-out in the MOR KO mice demonstrated by decreased mouse Oprm1 mRNA expression relative to the line's littermate controls (n = 8 Oprm1+/+ mice/4 MOR KO mice; unpaired t test; t = 8.949; df = 10; p < 0.0001). D, Validation of VTAMOR re-expression through bilateral injections of AAVDJ-hSyn1-FLEX-mCh-T2A-FLAG-hMOR-WPRE in MOR KO mice through increased expression of human OPRM1 mRNA relative to MOR KOs (n = 4 MOR KO mice/3 MOR KO mice with VTAMOR bilateral re-expression; unpaired t test; t = 8.551; df = 5; p = 0.0004). Expression levels in panel D are adjusted by a factor of 105. The large expression values in the AAV-hMOR group result from the absence of endogenous human OPRM1 expression in the control MOR KO line and the specificity of the viral re-expression in the experimental group. E, Daily morphine sulfate intake across 13 d forced drinking exposure paradigm. F, At 24 h following cessation of morphine exposure, neither the AAV-EGFP–injected control or AAV-hMOR–injected treatment group shows alterations in physical somatic signs as assessed through global withdrawal scores (n = 13 AAV-EGFP–injected + opioid-naive mice/12 AAV-hMOR–injected + opioid-naive mice/10 AAV-EGFP–injected + morphine-treated mice/11 AAV-hMOR–injected + morphine-treated mice; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.01213; p = 0.9128; treatment F(1,42) = 0.3826; p = 0.5395; interaction F(1,42) = 0.3809; p = 0.4502). The individual somatic signs contributing to the global withdrawal score are detailed in Extended Data Figure 6-1B–K. G, Morphine-treated MOR KO mice with VTAMOR re-expression show decreased sociability relative to their opioid-naive counterparts during the 4 week protracted abstinence period, while the morphine-treated AAV-EGFP–injected viral controls remain unaffected during protracted morphine (two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.8630; p = 0.3582; treatment F(1,42) = 9.539; p = 0.0036; interaction F(1,42) = 5.849; p = 0.0200). VTAMOR re-expression in MOR KO mice does not alter protracted morphine withdrawal behaviors related to (H) anxiety as assessed through the elevated zero maze (two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.9462; p = 0.3363; treatment F(1,42) = 0.1989; p = 0.6579; interaction F(1,42) = 0.6083; p = 0.4398), (I) despair as assessed through tail suspension (two-way ANOVA + Bonferroni's multiple comparison; virus F(1,40) = 0.03410; p = 0.8544; treatment F(1,40) = 0.3161; p = 0.5771; interaction F(1,40) = 1.753; p = 0.1930), and (J) nociceptive hypersensitivity as assessed through the hot plate test (two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.1684; p = 0.6836; treatment F(1,42) = 0.01332; p = 0.9087; interaction F(1,42) = 0.7436; p = 0.3934). Social interaction scores of MOR KO mice compared with the line's littermate controls are presented in Extended Data Fig. 6-1L. Data are represented as mean ± SEM; *p < 0.05; ***p < 0.001; ****p < 0.0001.
Figure 6-1
Total consumed drinking volume, individual somatic withdrawal signs 24-hour following cessation of morphine exposure, and social interaction in opioid naive Oprm1+/+ mice compared to the MOR KO line. (A) Total consumed volume in AAV-EGFP and AAV-hMOR injected groups exposed to water or chronic morphine show no differences in volume consumed each day compared to day 1 across all groups (two-way ANOVA + Bonferroni’s multiple comparison, day F5.051,238.7 = 13.75, p < 0.0001; group F3,48 = 4.044, p = 0.0121; interaction (day x group) F36,567 = 0.6448, p = 0.9474). Frequencies of (B) resting tremors, (C) paw tremors, (D) head shakes, (E) body shakes, (F) genital licking and grooming, (G) gnawing on limb, (H) backing and defensive treading bouts, (I) jumps, (J) jumps, and (K) digging bouts across viral groups remain unchanged across treatment conditions. (L) Social interaction duration is comparable between Oprm1+/+ mice (littermate controls of the MOR KO line) relative to MOR KO mice (injected with AAV-EGFP control virus 8 weeks prior to the social interaction test). Although these two cohorts were not conducted concurrently, all experiments were conducted under consistent conditions, including the same experimenter, facility, equipment, and handling protocols. This comparison aims to demonstrate no inherent social interaction deficits in the MOR KO line used in our study. Data are represented as mean ± SEM. Download Figure 6-1, TIF file.
Mice were exposed to the 13 day forced morphine drinking paradigm with escalating concentrations (Fig. 6A,E). Morphine-treated mice in both the AAV-hMOR- and AAV-EGFP–injected groups did not show an increased consumption of morphine solution throughout the exposure paradigm relative to Day 1 (Extended Data Fig. 6A). This result contrasts with the morphine-treated wild–type C57BL/6J mice, which have functional MORs globally and exhibit a significant increase in morphine solution consumption relative to Day 1 (Fig. 1B). We assessed somatic withdrawal signs 24 h following the cessation of morphine administration and found that neither the AAV-EGFP- nor the AAV-hMOR–injected groups showed differences in global withdrawal score between water- and morphine-treated animals (Fig. 6F; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.01213; p = 0.9128; treatment F(1,42) = 0.3826; p = 0.5395; interaction F(1,42) = 0.5809; p = 0.4502). While the VTAMOR population is a critical pharmacological target of morphine, this result suggests that MORs in the VTA are not sufficient to induce spontaneous morphine withdrawal-induced physical dependence signs.
At 4 weeks of protracted morphine abstinence, morphine-treated mice with VTA-specific re-expression of AAV-hMOR demonstrated decreased sociability relative to their opioid-naive counterparts, while the AAV-EGFP–injected group, which effectively lack functional MORs, remain unaffected compared with opioid-naive mice (Fig. 6G; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.8630; p = 0.3582; treatment F(1,42) = 9.539; p = 0.0036; interaction F(1,42) = 5.849; p = 0.0200). This outcome highlights the sufficiency of VTAMORs in mediating low sociability during protracted morphine abstinence. Moreover, in the opioid-naive groups, VTA-specific re-expression of AAV-hMOR induces a trending increase in social interaction compared with the AAV-EGFP–injected group. Additional behavioral assessments including the elevated zero maze (Fig. 6H; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.9462; p = 0.3363; treatment F(1,42) = 0.1989; p = 0.6579; interaction F(1,42) = 0.6083; p = 0.4398), tail suspension test (Fig. 6I; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,40) = 0.03410; p = 0.8544; treatment F(1,40) = 0.3161; p = 0.5771; interaction F(1,40) = 1.753; p = 0.1930), and hot plate test (Fig. 6J; two-way ANOVA + Bonferroni's multiple comparison; virus F(1,42) = 0.1684; p = 0.6836; treatment F(1,42) = 0.01332; p = 0.9087; interaction F(1,42) = 0.7436; p = 0.3934) did not reveal differences between morphine-treated and opioid-naive mice within the AAV-EGFP- and AAV-hMOR–injected groups. These findings suggest that the effects of chronic morphine exposure on VTAMORs may primarily influence sociability.
VTAMORs contribute to acute morphine-induced neural activation in the ACC
Given our previous findings of blunted FOS expression in the ACC and LC in morphine-treated C57 mice during protracted abstinence (Fig. 5C–F), we sought to assess whether re-expression of VTAMORs in MOR KO mice may modulate FOS expression in these regions following a morphine challenge.
Interestingly, in opioid-naive MOR KO mice with VTAMOR re-expression, FOS expression in the ACC is increased following an acute morphine challenge relative to the saline controls (Fig. 7A,B; three-way ANOVA + Bonferroni's multiple comparison; treatment F(1,38) = 0.4619; p = 0.5008; virus F(1,38) = 2.828; p = 0.1008; injection F(1,38) = 12.37; p = 0.0011; interaction; injection × virus × drinking F(1,38) = 1.595; p = 0.2142). This effect contrasts with results in the opioid-naive group injected with AAV-EGFP, where FOS expression remains unaltered following the morphine challenge injection relative to the saline controls. This finding highlights the role of VTAMORs in modulating the ACC's neural activation response to a dose of morphine shown to produce rewarding effects. This effect suggests that activation of VTAMORs induced by a morphine challenge contributes to enhanced neural responsiveness in the ACC.
VTAMORs contribute to acute morphine-induced FOS increase in the ACC. FOS density following a saline or morphine challenge injection (20 mg/kg, s.c.) in the (A, B) ACC (three-way ANOVA + Bonferroni's multiple comparison; treatment F(1,38) = 0.4619; p = 0.5008; virus F(1,38) = 2.828; p = 0.1008; injection F(1,38) = 12.37; p = 0.0011; interaction (injection × virus × drinking F(1,38) = 1.595; p = 0.2142; scale bar, 200 µm) and (C, D) LC [three-way ANOVA + Bonferroni's multiple comparison; treatment F(1,38) = 11.08; p = 0.0020; virus F(1,38) = 0.1604; p = 0.6911; injection F(1,38) = 12.70; p = 0.0010; interaction (injection × virus × drinking) F(1,38) = 9.740; p = 0.0034; scale bar, 200 µm]. In the ACC, a morphine challenge injection leads to increased (A, B) FOS in AAV-hMOR-injected opioid–naive mice. In the LC, a morphine challenge injection leads to increased (C, D) FOS in AAV-EGFP–injected opioid-naive mice. E, Schematic representation of VTAMOR neurons' structural connectivity as explored across our VTAMOR input–output mapping experiments. Arrow thickness represents estimated connection density. F, Schematic summary of morphine bound to a VTAMOR neuron and its subsequent effects on low sociability during protracted abstinence. Data are represented as mean ± SEM; *p < 0.05.
Additionally, we observed altered neuronal activation patterns in the LC in morphine-treated MOR KO mice injected with the AAV-EGFP virus. Following an acute morphine challenge injection, these MOR KO mice exhibited increased FOS expression in the LC (Fig. 7C,D; three-way ANOVA + Bonferroni's multiple comparison; treatment F(1,38) = 11.08; p = 0.0020; virus F(1,38) = 0.1604; p = 0.6911; injection F(1,38) = 12.70; p = 0.0010; interaction; injection × virus × drinking F(1,38) = 9.740; p = 0.0034). Given this effect was not observed in morphine-treated C57BL/6J mice following a morphine challenge injection (Fig. 5E,F), this result may suggest compensatory changes in response to morphine that may be mediated by alterations in noradrenergic signaling, as the LC is the major source of norepinephrine (Berridge and Waterhouse, 2003). The differential response in morphine-treated mice that is not apparent in the opioid-naive mice may also suggest unique opioid-induced plasticity in this area following chronic morphine exposure and protracted abstinence.
Discussion
In this study, we aimed to elucidate the role of MORs in the VTA in facilitating neuroadaptive changes and negative affective behaviors during protracted morphine abstinence. We first modeled the emergence of negative affective behaviors through increased social deficit-, anxiety-, and despair-related behaviors during protracted abstinence and identified VTAMOR neurons' involvement in the onset of withdrawal across the VTA's anterior to posterior axis. We identify withdrawal-active VTAMOR neurons to show markers of GABA and glutamate release through expression of Vgat and Vglut2, respectively. These VTAMOR neurons are structurally connected to an array of cortical, subcortical, and hindbrain regions central to motivation, abstinence, affect, and reward, suggesting that abstinence-related maladaptations in VTAMOR neurons might affect and impact reward processes in upstream and downstream circuits. Indeed, chronic morphine exposure diminishes the ability of an acute, rewarding dose of morphine to engage these regions during the protracted abstinence period, in particular the ACC and LC. We also demonstrate that VTAMORs are sufficient to facilitate low sociability during protracted abstinence and contribute to acute morphine-induced neural activation in the ACC. Thus, our findings offer a neuronal-type, region-specific target implicated in a maladaptive phenotype of protracted opioid abstinence—low sociability and diminished opioid-reward responsiveness (Chou et al., 2011).
Consistent with prior literature, the observed negative affective behaviors following 4 weeks of protracted morphine abstinence parallel findings from various opioid administration paradigms. For instance, protracted abstinence 3–6 weeks following investigator-administered opioid injections also induces low sociability (Goeldner et al., 2011; Bravo et al., 2020; Becker et al., 2021; Pomrenze et al., 2022), despair-like behaviors (Goeldner et al., 2011; Becker et al., 2021), and anxiety-related behaviors (Bravo et al., 2020; Becker et al., 2021), which reinforces the robust nature of these maladaptive behaviors as key features of protracted opioid abstinence's pathology. Our adoption of an oral morphine drinking paradigm in the home cage minimizes potential restraint stress during frequent injections (Stuart and Robinson, 2015) and mirrors the pharmacokinetic properties of oral opioid consumption in humans that enable a naturalistic framework of morphine dose escalation and maintenance.
Furthermore, VTAMOR neurons do not appear to mediate somatic withdrawal signs during protracted morphine abstinence. Our finding that both morphine-treated AAV-hMOR and AAV-EGFP–injected groups exhibited no differences in physical withdrawal signs relative to the opioid-naive mice suggests that VTAMORs do not mediate physical signs of morphine withdrawal. This finding aligns with established literature that MORs in other regions, including the LC (Maldonado et al., 1992; Maldonado and Koob, 1993) and CeA (Chaudun et al., 2024), are critical in mediating the physical signs of opioid withdrawal. Additionally, studies have shown that knockdown of MORs in the VTA in floxed Oprm1fl/fl mice, which possess MORs flanked by loxP sites, does not alter the physical somatic signs characteristic of fentanyl withdrawal (Chaudun et al., 2024). This finding further supports our observation of no differences in physical withdrawal signs following morphine cessation.
Our findings reveal the nuanced contribution of VTAMOR neurons in social interaction during protracted abstinence. This result extends beyond the established influence of dopaminergic transmission (Solié et al., 2022) and oxytocin signaling (Hung et al., 2017) in social reward in the VTA. Notably, in opioid-naive mice, VTA-specific re-expression of AAV-hMOR induces a trending increase in social interaction compared with the AAV-EGFP viral controls, further supporting the role of VTAMORs in facilitating sociability. Dysregulation of MOR activity is implicated in a variety of neuropsychiatric disorders, including depression, anxiety, and substance use disorders, where social behavior is often adversely affected (Fone and Porkess, 2008; Nummenmaa et al., 2020; Christie, 2021). The reduced sociability previously observed in other models of MOR KO mice underscores MORs' integral function in modulating social behavior (Moles et al., 2004; Becker et al., 2021; Toddes et al., 2021), a role that may be disrupted during repeated opioid exposure and protracted abstinence. The involvement of MOR-expressing neurons in negative affect has been investigated within regions such as the DRN (Lutz et al., 2014; Bailly et al., 2023) and habenula (Bailly et al., 2023), as the deletion of MORs in the DRN prevents the emergence of social deficits during chronic heroin abstinence (Lutz et al., 2014). Moreover, activation of MOR neurons in the habenula mediates projection-specific aversion, as optogenetic stimulation of habenula MOR neurons projecting to the DRN increase anxiety-like behavior, while optogenetic stimulation of these neurons projecting to the interpeduncular nucleus produces despair-related behavior and avoidance (Bailly et al., 2023). Despite the established role of the VTA as a key target for opioids, the field's understanding of the behavioral implications of VTAMOR neurons during dependence and protracted abstinence remains incomplete. Our work addresses this gap by demonstrating this neuronal population's role in facilitating social deficit-related behavior during abstinence. Interestingly, other behavioral measures such as anxiety-, despair-, and nociceptive-related behaviors did not show differences between morphine-treated and opioid-naive mice upon VTAMOR viral re-expression, suggesting that MORs in the VTA may be more specific to social behaviors during protracted abstinence.
In this study, we examined the neurochemical identity of VTAMOR neurons implicated in naloxone-induced withdrawal using fluorescent in situ hybridization. Future studies could build on these findings by examining the dopaminergic VTA population by labeling for tyrosine hydroxylase, a well-established marker for dopaminergic neurons. Moreover, investigating the neurochemical identity of withdrawal-active VTAMOR neurons along the anterior-to-posterior axis of the VTA could uncover potential topographical differences in cell-type contributions during withdrawal. Notably, prior work from Dr. David Root's lab has demonstrated subregional variability in the neurochemical identity VTAMOR neurons, showing that glutamatergic VTAMOR neurons are primarily distributed in the anterior VTA, while GABAergic VTAMOR neurons are concentrated in the posterior VTA (McGovern et al., 2023). Finally, conducting simultaneous fluorescent in situ hybridization for Vglut2 and Vgat would allow for direct statistical comparisons between glutamatergic and GABAergic neural populations to understand their relative contributions to withdrawal-active VTAMOR neurons. This expanded approach would enhance the current findings and deepen our understanding of the VTA cell types that express MOR during withdrawal and abstinence.
While we have identified the role of VTAMORs in facilitating low sociability during protracted morphine abstinence, an important consideration to acknowledge is the use of certain strains of MOR KO mice, which have been previously reported to exhibit low sociability (Moles et al., 2004; Becker et al., 2014; Toddes et al., 2021; Derieux et al., 2022). However, it is worth noting that the MOR KO line used in these previous studies (Matthes et al., 1996) was generated by Dr. Brigitte Kieffer using different technologies and knock-in strategies (129S2/SvPas × C57BL/6-derived P1 ES cells inserted into Exon 2), compared with the MOR KO line used here, which was created by Dr. Richard Palmiter (129S6 × C57BL/7-derived F1-derived G4 ES cells through insertion of the Cre:GFP construct into Exon 1). To our knowledge, the MOR KO line generated by Dr. Richard Palmiter's lab, used in this study, has not been reported to exhibit such social deficits. Consistently, we did not observe apparent differences in social interaction between this MOR KO line compared with the line's littermate controls (Extended Data Fig. 6-1L). Differences in genetic background or environmental factors, such as vivarium conditions or experimental procedures, could contribute to behavioral discrepancies and variability in phenotypic outcomes (Crabbe et al., 1999). Moreover, our primary objective was to investigate the role of VTA-specific MORs in mediating social behaviors following chronic opioid exposure and abstinence. The use of Cre-dependent viral tools enabled the selective re-expression of MORs within the VTA of MOR KO mice to dissect the specific contributions of this neural population to abstinence-related negative affective behaviors. Although VTAMOR re-expression was associated with decreased sociability in morphine-treated mice during protracted abstinence relative to opioid-naive mice, with no effect observed in viral control mice, a potential limitation is the absence of a statistically significant difference between the AAV-EGFP and AAV-hMOR groups within the morphine-treated condition, despite a trending decrease in sociability in the AAV-hMOR group. Future studies to further elucidate the contribution of VTAMORs in facilitating abstinence-related social behaviors include utilizing conditional knock-out models and knocking down MORs within the VTA of floxed Oprm1fl/fl mice. Additionally, expanding the behavioral repertoire of social behaviors through other assessments, such as the three-chamber social test or social conditioned place preference, could provide an enhanced understanding of the role of VTAMORs in diverse social contexts.
We also found that mice with a prior history of morphine exposure exhibit decreased neuronal activation in the ACC following an acute morphine injection. The ACC has been implicated in encoding and multiplexing various aspects of reward processing (Hayden and Platt, 2010; Cai and Padoa-Schioppa, 2012; Monosov, 2017), including social interaction (Rudebeck et al., 2008). In our study, we observe blunted morphine challenge-induced ACC FOS in mice undergoing protracted morphine abstinence. The diminished neural responsiveness in the ACC to a dose of morphine previously shown to produce rewarding effects may reflect a dysregulation in processing rewarding stimuli including social reward, an effect that may inform our observation of social deficits during protracted abstinence. Specifically, our results shed light on the role of VTAMOR neurons in modulating the ACC's response to acute morphine (Fig. 7A,B). Despite the relatively minimal direct structural connections between VTAMOR neurons and the ACC, as observed in our rabies input mapping (Fig. 4D) and axonal projection data (Fig. 4H), increased ACC FOS following a morphine challenge in MOR KO mice with restored VTAMOR expression suggests that indirect or neuromodulatory pathways may mediate this effect. Opioid-induced MOR activation induces downstream effects, notably the inhibition of local GABAergic interneurons within the VTA (Johnson and North, 1992). Given the characterized projection of dopaminergic VTA neurons to the ACC (Beier et al., 2015; Breton et al., 2019), it is plausible that this altered dopaminergic signaling following acute morphine may contribute to the heightened FOS density in the ACC. This aligns with prior evidence that VTA dopaminergic projections to the ACC are critical for the acquisition and maintenance of morphine-conditioned place preference (Narita et al., 2010). Moreover, the variety of neuronal types within the VTA, such as glutamate-transmitting neurons (Yamaguchi et al., 2007; Nair-Roberts et al.,2008; Root et al., 2020), may also contribute to morphine-induced FOS expression. However, the projection targets of VTA glutamatergic neurons remain to be elucidated. Future studies leveraging projection-specific techniques, such as retrograde viral tracing combined with chemogenetic or optogenetic approaches, could provide valuable insights into the functional role of the VTAMOR to ACC pathway in neural activation and its contribution to protracted abstinence-related negative affective behaviors.
In conclusion, our study highlights the role of VTAMORs in mediating low sociability during protracted abstinence. We demonstrate that chronic morphine exposure leads to increased negative affective behaviors during abstinence, which are linked to heightened neuronal activity in VTAMOR neurons and their structurally connected regions implicated in reward and aversion processing. By highlighting the region- and neuronal-specific mechanisms underlying neuroadaptive alterations from chronic opioid exposure and abstinence, this study advances our understanding of neural substrates involved in dependence and abstinence. Future research may expand investigations of VTAMOR neurons' role in protracted abstinence-induced low sociability, as this neuronal population may represent a potential target for interventions aimed at alleviating the debilitating signs and symptoms associated with protracted abstinence.
Footnotes
This research was funded by National Institutes of Health R01-DA-054374, R01-DA-056599, DP2-GM1-40923, and T32-DA028874. We thank the University Laboratory Animal Resources (ULAR) group at the University of Pennsylvania's Translational Research Laboratory for the assistance with rodent husbandry and veterinary support. We thank Jacqueline Wu and Malaika Mahmood for their technical assistance. We also thank Matthew Banghart and Karl Deisseroth for their virus contributions.
The authors declare no competing financial interests.
- Correspondence should be addressed to Julie A. Blendy at blendy{at}pennmedicine.upenn.edu or Gregory Corder at gcorder{at}upenn.edu.
