Attenuation of cue-induced heroin-seeking by the atypical antidepressant mirtazapine Authors: Amanda L Persons, T Celeste Napier Published: 2026 ABSTRACT AbstractBackgroundThe addiction profile of opioids reflects imbalances in downstream transmitters, including dopamine and serotonin (5-HT). The atypical antidepressant mirtazapine has multiple sites of action, including antagonism at 5-HT2A receptors and inverse agonism at 5-HT2C receptors. We previously demonstrated that mirtazapine reduces drug-seeking behavior in rats that self-administer methamphetamine. Here, we tested the hypothesis that mirtazapine attenuates cue-induced heroin-seeking behavior.MethodsMale Sprague-Dawley rats were operantly trained to nose-poke to obtain heroin (0.075 mg/kg per 1 mL intravenous infusion) on a fixed ratio-1 schedule of reinforcement for 10 days (6 h/day). One day after heroin self-administration (SA), a baseline cue reactivity test (CR, 20 min) was conducted. Thereafter, rats were subjected to an intermittent SA/CR cycle wherein rats self-administered heroin for two days, followed by a CR test the following day. This cycle was repeated until each rat was tested for CR with vehicle (1 mL/kg), mirtazapine (1 mg/kg) or mirtazapine (5 mg/kg) injected intraperitoneally 30 min prior to the test.ResultsRats readily acquired the SA task. By the tenth SA session, the active hole was selected 94% of the time, and the average cumulative heroin intake across all ten sessions was 22 ± 1.5 mg/kg. Pretreatment with vehicle or mirtazapine at 1 mg/kg did not affect CR. CR was significantly reduced by 5 mg/kg mirtazapine without significantly affecting the latency to initiate cue-directed responding.ConclusionsReductions in CR indicate a decrease in drug-seeking. Thus, these findings suggest that mirtazapine may be repurposed for relapse reduction during abstinence therapy of opioid use disorder.Keywords: Opioid, Heroin, Self-administration, Relapse, Mirtazapine, Antidepressant, Rat Introduction Opioid use disorder (OUD) and mortality associated with overdosing are a worldwide public health concern. Opioids activate mu-opioid receptors (MOR), and current government-approved pharmacotherapies for OUD target this receptor. These medications promote recovery from OUD, but relapse to drug-taking remains high. The addiction profile of opioids reflects not only the direct consequences of MOR activation but also imbalances in downstream transmitter systems, including dopamine and serotonin (5-HT). 5-HT2A and 5-HT2C receptors (R) regulate dopamine release (for review, see Alex and Pehek, 2007; Fink and Göthert, 2007). The atypical antidepressant mirtazapine has multiple sites of action, including antagonism at 5-HT2A receptors (Pawlyk et al., 2006) and inverse agonism at the 5-HT2C receptors (Chanrion et al., 2008, Labasque et al., 2010). We revealed in rats that mirtazapine reduces morphine-induced place preference (Graves et al., 2012b, Graves et al., 2012a), suggesting that mirtazapine may interfere with context-associated memories linked to repeated opioid exposure. The current proof-of-concept study sought to extend this work and our previous work in rats demonstrating mirtazapine-induced reductions in methamphetamine-seeking (Graves and Napier, 2011). Here, the treatment utility of mirtazapine was tested against a more potent opioid, heroin, using a behavioral task that indicates ‘drug-seeking’, a potential contributor to relapse. To do so, rats were trained to self-administer heroin and then tested for cue reactivity (CR) as an index of the incentive motivation that can drive seeking. We hypothesized that mirtazapine attenuates cue-induced heroin-seeking behavior. Materials and methods Animals Male Sprague–Dawley rats (n = 8) were purchased from Envigo (Indianapolis, IN) weighing 225–249 g and housed two per cage with enriched bedding, in an environmentally controlled facility under a 12 h light–dark cycle. Food and water were available ad libitum. Testing was conducted between 9AM and 5PM, during the light cycle. Rats were handled in accordance with procedures established in the Guide for the Care and Use of Laboratory Animals (National Research Council, Washington, DC), and protocols were approved by the RUSH University Institutional Animal Care and Use Committee. Test drugs Heroin hydrochloride (NIDA Drug Supply Program, Bethesda, MD) was dissolved in 0.9% saline and administered intravenously as 0.075 mg/kg per 0.1 mL infusion. Mirtazapine (NIMH Chemical Synthesis and Drug Supply Program, Bethesda, MD) was dissolved in 1 N HCl, diluted with sterile H2O, and the pH was adjusted to ~6.3–6.8 using 1 N NaOH. Mirtazapine was administered intraperitoneally (ip) as 1.0 or 5.0 mg/mL/kg. Doses of mirtazapine were selected based on prior behavioral studies testing the efficacy of mirtazapine to reduce methamphetamine-seeking (Graves and Napier, 2011), expression of morphine-induced conditioned place preference (Graves et al., 2012b, Graves et al., 2012a), and effort-based decision-making in a gambling-like task (Persons et al., 2017). Additionally, mirtazapine was tested in naïve rats for motivated motor function on a rotorod; mirtazapine did not alter latency to fall at either dose. Exploratory activity was also assessed in a subset of methamphetamine self-administering rats, and it was determined that 5.0 mg/kg mirtazapine did not affect general motor function (Graves and Napier, 2011). Surgical procedures Intravenous catheter implant procedures followed our previously published protocols (Graves and Napier, 2011, Kousik et al., 2014, Persons et al., 2021). In brief, rats were anesthetized with 2–3% isoflurane, and catheters constructed of silastic tubing (0.3 mm × 0.64 mm; Dow Corning Co., Midland, Michigan, USA) were inserted into the right jugular vein and secured with sutures. The distal end of the cannula passed subcutaneously over the midscapular region and exited through a metal guide cannula (22 gauge; Plastics One, Roanoke, Virginia, USA). Rats were allowed to recover from surgery for 10 days; catheters were flushed daily with 0.1–0.2 mL 0.9% sterile saline to maintain patency. Self-administration Operant chambers (Med-Associates, St. Albans, VT) for self-administration (SA) were equipped with two nose-poke holes, a stimulus light above each hole, a tone cue, and a house light on the opposite wall. Infusions of heroin (0.075 mg/kg per 0.1 mL sterile saline) were delivered to the jugular cannula by tubing attached to a syringe set in an infusion pump on a fixed ratio-1 (FR1) schedule of reinforcement. As illustrated in the experimental timeline (Supplemental Figure 1), training involved two sessions per day, 3 h each (total 6 h per day) in a daily long-access self-administration protocol (Venniro et al., 2017). To accommodate the long-access, a 1–2 h break between sessions was implemented, wherein rats were returned to the home cage to allow access to food and water ad libitum. The exception was on day seven; due to technical issues, rats completed only one 3 h session, and data from this day were not included in outcome analyses. Selection of the “active” hole resulted in the illumination of the stimulus light, activation of the tone, and a 6 s infusion of heroin. At the end of the infusion, the stimulus light and tone were turned off and the house light was illuminated for 9 s, resulting in a total time-out period of 15 s. During the time-out period, nose-pokes in the active hole were recorded but had no consequence. Selection of the “inactive” hole at any time during the session had no programmed consequence. Rats self-administered heroin for a total of 10 consecutive days. One rat died at the end of acquisition; all data from this rat were excluded from outcome analyses. Cue reactivity One day after the last operant session, a baseline drug-seeking CR test was conducted (n = 7). In this 20 min test, explicit, discrete cues were presented upon selection of the active hole (stimulus light and tone), but no heroin was delivered. Thereafter, rats were subjected to an intermittent SA/CR cycle (Graves and Napier, 2011) during which they self-administered heroin for two days, followed by a CR test the next day. This cycle was repeated until each rat was tested for CR, in a randomized order, with vehicle (1 mL/kg, ip), mirtazapine (1 mg/kg, ip), or mirtazapine (5 mg/kg, ip) injected 30 min before the test. All CR tests were conducted between 1 and 4PM, consistent with the afternoon self-administration sessions. Given the 2–4 h half-life of mirtazapine in rats, the minimum 3-day interval between CR tests provided > 5 half-lives for complete drug clearance, minimizing carryover effects (Rouini et al., 2014). Statistical analyses All data were analyzed using a one-way or two-way repeated measures ANOVA with post hoc Newman-Keuls, where appropriate. A Pearson correlation was used to compare number of active nose-pokes and latency to first nose-poke in the mirtazapine (5.0 mg/kg) group. Data are presented as mean+SEM or mean± SEM, and significance was set at p < 0.05. Graphpad Prism v9.5 software was used for analysis (Graphpad Software, Boston, MA). Results The rats rapidly learned the heroin SA task contingencies, e.g., discriminating between active and inactive holes in the first operant session (Fig. 1A). Selection of the active hole closely mapped onto the number of infusions by protocol day 3. Daily intake stabilized, and during the last two acquisition sessions (protocol days 9–10), the average amount self-administered was 2.8 ± 0.28 mg/kg. We did not observe systematic differences in heroin infusions between the morning and afternoon sessions within individual days during the acquisition phase (two-way repeated measures ANOVA). Heroin intake was stable during the during the intermittent SA days; there was no significant effect of repeated CR testing (F=3.4, p = 0.064; Fig. 1B). That is, daily intake during the during the SA /CR cycles did not differ from the acquisition phase. Pretreatment with 1.0 mg/kg mirtazapine had no effect on CR, but CR was significantly decreased by 5.0 mg/kg mirtazapine (F=5.8, p = 0.017; Fig. 2A). This occurred without altering latency to first nose-poke (F=1.14, p = 0.34; Fig. 2B) or selection of the inactive hole (F=2.1, p = 0.17; data not shown). Additionally, there was no correlation between the number of active nose-pokes and the latency to first nose-poke when rats were treated with 5.0 mg/kg mirtazapine (r = -0.34, p = 0.46; data not shown). Heroin self-administration. (A) Rats (n = 7) acquired task contingencies and demonstrated stable self-administration of heroin (data from day 7 were not included due to technical issues). There were no significant differences among the last three days of self-administration for nose-pokes in the active hole (Active NP), nose-pokes in the inactive hole (Inactive NP), or infusions. (B) Two days of heroin self-administration followed each cue reactivity test (for experimental timeline, see Supplemental Figure 1). Intake of heroin was not significantly different among the last two sessions of acquisition (protocol days 9–10) and the intermittent self-administration days between the three cue reactivity tests (protocol days 12–13, 15–16, 18–19). One-way repeated measures ANOVA, p > 0.05. Impact of mirtazapine on cue reactivity. (A) Rats (n = 7) were tested for cue reactivity (CR) 30 min after an intraperitoneal injection of vehicle (1 mL/kg), 1.0 mg/kg mirtazapine (MIRT), or 5.0 mg/kg MIRT in randomized order. Pretreatment with MIRT at 5.0 mg/kg decreased the number of active nose-pokes (NP) during the 20 min test session as compared to vehicle and 1.0 mg/kg MIRT; there was no effect of 1.0 mg/kg MIRT. One-way repeated measures ANOVA with post hoc Newman-Keuls; * p < 0.05. (B) MIRT (1.0 or 5.0 mg/kg) did not significantly alter the amount of time taken by the rats to begin poking in the active hole during the CR test (Latency to 1st NP. One-way repeated measures ANOVA, p > 0.05. Discussion Maladaptation of the brain 5HT system is associated with drug use and addictions (for review, see Müller and Homberg, 2015). Thus, the serotonergic features of mirtazapine make the atypical antidepressant an interesting candidate for treating substance use disorders (Graves et al., 2012b, Graves et al., 2012a). Most investigations of this possibility have focused on psychostimulants that directly impact serotonergic transmission, with promising outcomes. In rats, mirtazapine reduces methamphetamine-motivated behaviors (Herrold et al., 2009, Voigt et al., 2011, Voigt and Napier, 2012, Graves and Napier, 2011). In humans, mirtazapine may reduce symptoms of withdrawal from cocaine or amphetamine use (for review, see Li and Shoptaw, 2023), and it is undergoing a phase 3 clinical trial for outpatient treatment of methamphetamine use disorder (McKetin et al., 2024). Given that serotonin is central to brain circuits that are engaged during numerous reward-motivated behaviors (for review, see Müller and Homberg, 2015; Peters et al., 2021), mirtazapine is also showing potential utility against an array of drug and behavioral addictions. To illustrate, in rats, mirtazapine reduces nicotine-seeking (Barbosa-Méndez and Salazar-Juárez, 2018) and attenuates gambling-like reinforcement tasks that do not involve drug or food reward (Persons et al., 2017). Suggesting therapeutic efficacy for OUD, in rats, mirtazapine reduces morphine-mediated conditioned place preference (Graves et al., 2012b, Graves et al., 2012a) and heroin seeking (current study); in humans, mirtazapine shows promise in managing symptoms that are associated with withdrawal from chronic opioid use (Lalani et al., 2023). Relapse to drug-taking by the abstinent individual remains a challenge for OUD therapy. The current preclinical study focused on this challenge. As drug-seeking during abstinence can be a prelude to drug-taking, we evaluated a behavioral indicator of drug-seeking, CR, i.e., active nose-pokes to discrete cues previously associated with self-administered heroin that are presented in the absence of the drug. Notably, mirtazapine reduced CR without altering latency to first nose-poke or selection of the inactive hole, suggesting that the reduction in cue-induced seeking was not due to gross motor impairment or sedation that would prevent the animal from approaching and engaging with the operant task. These findings are consistent with our prior work showing that these doses of mirtazapine do not impair motivated motor performance on a rotorod or alter open-field locomotor activity (Graves and Napier, 2011). This suggests that mirtazapine may be useful in reducing cue-induced reinstatement of opioid-seeking and/or relapse to opioid-taking during abstinence. These findings add to accumulating evidence in several rodent models of seeking and relapse to drug-taking that show benefits with mirtazapine treatment, including those with nicotine (Barbosa-Méndez et al., 2018), methamphetamine (Graves and Napier, 2011), and cocaine (Barbosa-Méndez et al., 2018). Likewise, human studies reveal that increased attentional bias toward drug-related cues is associated with increased effective connectivity in the anterior cingulate cortex-to-hippocampus neurocircuit (Ma et al., 2018, Ma et al., 2019), and both are attenuated by mirtazapine (Ma et al., 2022). Study limitations warrant consideration. First, the sample size was modest (n = 7), reflecting the proof-of-concept design of this study. While statistical power was sufficient to detect the primary effect of mirtazapine on CR, replication with larger samples is needed. Second, only male rats were studied. While the literature on sex differences in opioid self-administration, drug-seeking, and relapse behaviors is mixed (for review, see Nicolas et al., 2022), future studies should include both sexes to evaluate potential sex-dependent effects of mirtazapine on opioid-seeking. Third, the CR model employed here and elsewhere (Graves and Napier, 2011) specifically assesses cue-induced drug-seeking, only one of multiple triggers for relapse in humans. Stress-induced and drug-primed reinstatement models represent additional, critical facets of relapse that have not been evaluated with mirtazapine in the context of OUD. Lastly, the current study used an acute, single-dose administration paradigm; the effects of chronic mirtazapine treatment on heroin-seeking remain to be determined. Several features of mirtazapine add to the attractiveness of its use in treating OUD: (1) it is safe and well-tolerated by humans, (2) the most frequent adverse event is drowsiness/sedation, which can be mitigated by administering mirtazapine before going to bed, and (3) mirtazapine is off-patent, making it an inexpensive therapy option. While additional studies are needed, the current positive outcomes in heroin self-administering male rats, and its safety profile in humans, add to growing evidence suggesting that mirtazapine has the potential to be repurposed for relapse-reduction during abstinence therapy of OUD. CRediT authorship contribution statement T. Celeste Napier: Writing – review & editing, Writing – original draft, Supervision, Methodology, Conceptualization. Amanda L. Persons: Writing – review & editing, Writing – original draft, Project administration, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization. Funding sources This work was supported by the Brain Research Foundation grant number BRFORRG-2018-1. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Footnotes Supplementary data associated with this article can be found in the online version at doi:10.1016/j.dadr.2026.100441. Supplementary material Supplementary material Supplementary material References Associated Data Supplementary Materials Supplementary material Supplementary material