Mu Opioid Receptor-Positive Neurons in the Dorsal Raphe Nucleus Are Impaired by Morphine Abstinence

Abstract

Background: Chronic opioid exposure leads to hedonic deficits and enhanced vulnerability to addiction, which are observed and even strengthen after a period of abstinence, but the underlying circuit mechanisms are poorly understood. In this study, using both molecular and behavioral approaches, we tested the hypothesis that neurons expressing mu opioid receptors (MORs) in the dorsal raphe nucleus (DRN) are involved in addiction vulnerability associated with morphine abstinence.

Methods: MOR-Cre mice were exposed to chronic morphine and then went through spontaneous withdrawal for 4 weeks, a well-established mouse model of morphine abstinence. We studied DRN-MOR neurons of abstinent mice using 1) viral translating ribosome affinity for transcriptome profiling, 2) fiber photometry to measure neuronal activity, and 3) an opto-intracranial self-stimulation paradigm applied to DRN-MOR neurons to assess responses related to addiction vulnerability including persistence to respond, motivation to obtain the stimulation, self-stimulation despite punishment, and cue-induced reinstatement.

Results: DRN-MOR neurons of abstinent animals showed a downregulation of genes involved in ion conductance and MOR-mediated signaling, as well as altered responding to acute morphine. Opto-intracranial self-stimulation data showed that abstinent animals executed more impulsive-like and persistent responses during acquisition and scored higher on addiction-like criteria.

Conclusions: Our data suggest that protracted abstinence to chronic morphine leads to reduced MOR function in DRN-MOR neurons and abnormal self-stimulation of these neurons. We propose that DRN-MOR neurons have partially lost their reward-facilitating properties, which in turn may lead to increased propensity to perform addiction-related behaviors.

Keywords: Addiction-related behaviors; Chronic morphine; Mu opioid signaling; Opto-intracranial self-stimulation; vTRAP-seq.

Figure 1.. Gene transcription is modified in abstinent mice.

(A-D) Viral translating ribosome affinity purification allows to sequence the transcriptome of DRN-MOR neurons. (A) Schematic of experimental timeline: MOR-Cre mice were injected in the DRN to express L10a-mCherry protein in DRN-MOR neurons, and were subsequently injected chronically either with saline CTL) or morphine (ABS). Four weeks after the last morphine exposure the DRN was micro-dissected. A fraction of total tissue mRNAs was conserved (input samples) and we immuno-precipitated mRNAs bounds to mCherry-L10a (IP samples) (n=6/6). (B) Principal component analysis: Factor 1 explains 41.7% of the variance of the and mostly corresponds to the fraction samples (IP vs input). (C) Quantification of Oprm1 transcripts: IP samples were enriched for Oprm1 transcripts compare to input samples for CTL and ABS. (D) Quantification of mCherry transcripts: IP samples were enriched for mCherry transcripts compare to input samples for CTL and ABS. (E-G) Gene expression is modified in DRN-MOR neurons from abstinent mice. (E) Differentially expressed genes after morphine abstinence (CTL vs ABS) are mainly distinct genes in input vs IP samples. (F) Comparison of the fold change of differentially expressed genes in CTL vs ABS: the fold change of differentially expressed genes is higher in IP versus input samples. (G) Comparison of the fold change of differentially expressed genes in IP vs input for all genes or genes differentially expressed in IP CTL vs IP ABS: the fold change of differentially expressed genes were higher when looking only at genes affected by morphine abstinence (IP CTL vs ABS differentially expressed genes) than at all genes. Data are represented as mean ± SEM. *: p<0.05; **: p<0.01; ****: p<0.0001.

Figure 2.. DRN-MOR neuron function is altered in abstinent mice.

(A-C) DRN-MOR neurons of abstinent mice show decreased gene expression related neuronal excitability and opioid signaling. (A) Over representation-analysis: the gene ontology (GO) terms underrepresented in DRN-MOR neurons of ABS animals included: Molecular function (passive transmembrane transporter activity, channel activity, cation channel activity, ion channel activity, substrate specific channel activity), Cellular component (cilium, ciliary part, axoneme part) and Biological processes (metanephros development, kidney morphogenesis). (B) Gene Set Enrichment Analysis (GSEA) with gene set REACTOME_OPIOID_SIGNALLING (33). (E) GSEA performed on IP ABS vs CTL differentially expressed genes indicated transcriptional depletion of opioid signaling pathway in DRN-MOR neurons of ABS animals. (C) Enrichment profile for differentially expressed genes between IP CTL vs ABS. Data are represented as mean ± SEM. *: p<0.05; ****: p<0.0001. (D). DRN-MOR neurons from ABS mice respond differently to morphine. Left: schematic of viral injection of the GCaMP expressing Cre-dependent virus and optic fiber (OF) implantation in the DRN of MOR-Cre mice. Middle: time courses of average GCaMP6m z-scores following morphine administration (10 mg/kg, i.p) in CTL and ABS animals showing a reduction of DRN-MOR neurons calcium activity with a maximum effect 4min after injection. Right: the average of z-score of DRN-MOR neurons fluorescence levels is reduced from 6 to 12 min after morphine injection in ABS animals compared to the CTL animals. n=5 for CTL and n=6 for ABS. Welch t-test: t_(6.45)=2.64; p<0.05.

Figure 3.. Control and abstinent mice learn to self-stimulate DRN-MOR neurons.

(A) Schematic of the experimental timeline: MOR-cre mice were injected in the DRN to express mCherry-ChR2 proteins and implanted with an optic fiber to activate DRN-MOR neurons with laser stimulation. Three weeks after surgeries, mice were injected either with saline (CTL, n=17) or with escalating doses of morphine (ABS, n=17). Four weeks after the last injection, mice were trained to nosepoke for the laser stimulation according to a fixed ratio of 1 (FR1; 4 days), FR3 (9 days) and FR5 (15 days) (acquisition). Mice were also evaluated for different addiction-related behaviors (persistence to respond, motivation, compulsion and reinstatement after extinction training). (B) Left. Total number of nosepokes performed during each 1 hr self-stimulation session during acquisition (28 days). Right. Average nosepokes performed during each reinforcement schedule: CTL and ABS discriminated the active from the inactive and increased the number of active nosepoke according to the schedule. (C) Left. Total number of earned laser stimulation bouts during each self-stimulation session during acquisition. Left. Average laser stimulations earned during each reinforcement schedule: CTL and ABS mice earned similar number laser stimulation across schedules. Data are represented as mean ± SEM.**** / ˚˚˚˚: p<0.0001.

Figure 4.. Abstinent mice perform more impulsivity-like and persistent responses when evaluated for several addiction-related behaviors.

(A) Left. Total number of impulsivity-like responses during each 1 hr self-stimulation session during acquisition (28 days. Right. Average impulsivity-like nose pokes per session for each reinforcement schedule: ABS mice performed in average more impulsivity-like nose pokes during acquisition. (B) Left. Persistence to respond FR3: ABS mice performed more persistent nose pokes than CTL mice. Right. Persistence to respond FR5: ABS did not perform more persistent responses than CTL mice. (C) Motivation: ABS mice did not perform more active nose pokes during the progressive ratio session (D) Compulsivity-like: ABS mice did not earn more laser stimulation during the foot-shock session. (E) Extinction and reinstatement: cue presentation reinstated active nose pokes after extinction in CTL and ABS mice. Data are represented as mean ± SEM. #: p<0.05; **: p<0.01; **** / ˚˚˚˚: p<0.001.

Figure 5.. Abstinent mice score higher for addiction-related criteria in the oICSS behavior.

(A-C) Comparison of addiction-related behaviors in mice presenting 0, 1, 2 or 3 criteria of addiction-related behaviors. The higher the criteria of addiction-related behavior the higher the animal scored for (A) the persistence to respond (B) the motivation (C) the compulsivity-like responses. (D) % of mice in each criterion of addiction-related behaviors for CTL and ABS mice. (E) ABS mice have higher criteria of addiction-related behaviors than CTL mice. (F) Principal component analysis: Factor 1 explains 45.3% of the total variance between individuals and takes into account four behavioral measures (compulsivity, motivation, persistence to respond and reinstatement). (G-H) Representation of individual values in Factor 1 and 2 according to (G) the group-criteria of addiction-related behaviors: Factor 1 ranks the individuals according to their criteria of addiction-related behavior. (H) the treatment: Factor 1 ranks the individuals according to their treatment (morphine or saline). Data are represented as mean ± SEM. *: p<0.05; **: p<0.01; ***: p<0.001; ****: p<0.0001.