Remifentanil self-administration promotes circuit- and sex-specific adaptations within the prefrontal-accumbens pathways

Glutamate neurotransmission, plasticity, sex differences, D1 MSNs, D2 MSNs, 10 Infralimbic-NAc shell, Prelimbic-NAc core 11 12 Number of words (manuscript main body): 8151 13 Number of figures: 6 (main) and 1 (supplemental) figures 14 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint

15 The nucleus accumbens (NAc) and its excitatory input from the medial prefrontal cortex (mPFC) 16 form a critical circuit underlying drug -induced plasticity associated with addiction -related 17 behaviors. However, baseline differences in excitatory signaling across NAc subcircuits and sex-18 specific neuroadaptations following opioid self -administration remain poorly understood. Here, 19 we examined synaptic signaling in mPFC –NAc pathways in drug -naïve mice and after 20 abstinence from remifentanil self-administration. Under drug-naïve conditions, AMPA receptor–21 mediated glutamatergic signaling was generally elevated in D2 medium spiny neurons (MSNs) 22 of both the NAc core and shell across sexes, while females exhibited greater excitatory signaling 23 in D1 MSNs of the NAc core compared with males. Pathway -specific analyses revealed that 24 prelimbic cortex (PL) inputs to NAc core D2 MSNs displayed enhanced calcium -permeable 25 AMPA receptor (CP-AMPAR) signaling and increased presynaptic release relative to D1 MSNs. 26 Following abstinence from remifentanil self -administration, miniature excitatory postsynaptic 27 current analyses showed increased excitatory drive at D1 MSNs and decreased drive at D2 28 MSNs, largely restricted to the NAc core. At PL –Core D1 MSN synapses, remifentanil reduced 29 AMPA/NMDA ratios, consistent with increased CP-AMPAR incorporation in males and females, 30 while increasing presynaptic signaling exclusively in males. In contrast, PL-Core D2 MSN 31 synapses showed a reduction in presynaptic signaling across sex, while ostensibly weakening 32 postsynaptic signaling selectively in males through reductions in CP -AMPAR signaling. At 33 infralimbic cortex (IL) –shell inputs , a reduction in AMPAR rectification indices at D1 MSN 34 synapses was produced by remifentanil , while release probability was decreased at D2 MSN 35 synapses in males only . Together, these findings reveal sex - and pathway -specific synaptic 36 adaptations within mPFC–NAc circuits that may be obscured by global measures of excitatory 37 transmission and identify baseline circuit differences that may shape opioid-induced plasticity. 38 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 1 INTRODUCTION 39 Opioid use disorder (OUD) develops from deficits in cognitive control that escalate opioid 40 intake, strengthen drug-seeking behaviors, and increase an individual’s propensity to relapse. 41 Long-lasting molecular and physiological neuroadaptations in key nodes of the reward pathway 42 form the molecular and cellular basis of these maladaptive behaviors (Koob et al., 2004; Koob 43 & Volkow, 2016) . Clinical findings indicate that dysfunction of the prefrontal cortex ( PFC) 44 emerges across substance use disorders (Koob & Volkow, 2010) . The rodent medial PFC 45 (mPFC), a region homologous to the primate PFC, plays a key role in governing numerous 46 aspects of drug reward- and drug seeking behaviors (Everitt & Robbins, 2005; Koob & Volkow, 47 2010). The mPFC receives and sends input to multiple limbic and midbrain targets implicated in 48 opioid relapse . Importantly, e xtensive anatomical and causal evidence demonstrates that 49 divergent mPFC-NAc projections exert robust, bidirectional control over opioid seeking 50 (LaLumiere & Kalivas, 2008; Rogers et al., 2008; Van den Oever et al., 2010; Shen et al., 2011; 51 Bossert et al., 2012; Marchant et al., 2016; Hearing, 2019). Given the role of the mPFC and its 52 interconnected nature, understanding how this region and its subcircuits are altered by opioids 53 will provide a mechanistically precise and translationally relevant framework for understanding 54 cortical control of opioid seeking. 55 The mPFC-NAc pathway can be divided into two primary projections consisting of the 56 prelimbic cortex projection to the core region of the NAc (PL -Core) and infralimbic cortical 57 projections to the shell subdivision (IL-Shell) (Sesack et al., 1989; Berendse, 1992; Brog et al., 58 1993; Heidbreder & Groenewegen, 2003; Heimer, 2003; Vertes, 2004; Hoover & Vertes, 2007). 59 Numerous lines of research to date have demonstrated the necessity of the PL to NAc core (PL-60 Core) pathway in many different types of reinstatement to drug-seeking (McFarland et al., 2003; 61 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint LaLumiere & Kalivas, 2008; Rogers et al., 2008). Alternatively, while the IL-Shell pathway has 62 been noted to inhibit psychostimulant seeking , evidence supports a potential role in promoting 63 heroin seeking (Rogers et al., 2008; Van den Oever et al., 2010; Bossert et al., 2012; Peters et 64 al., 2013; Hearing et al., 2016; Marchant et al., 2016). . Recent findings have identified opioid-65 induced adaptations in mPFC output neurons (Anderson et al., 2021), including those projecting 66 to the NAc (Kokane et al., 2023) that are causally linked to impairments in cognition and opioid-67 seeking, respectively. However, what remains unknown is how these adaptations translate to 68 plasticity downstream in the NAc to change behavior. 69 Like subregions of the medial PFC, the NAc core and shell show divergent responses to 70 non-contingent morphine with dissociable and overlapping roles in opioid reward and seeking 71 behavior (Zahm & Brog, 1992; Pontieri et al., 1995; Hutcheson et al., 2001; Pecina & Berridge, 72 2005; LaLumiere & Kalivas, 2008; Rogers et al., 2008; Bossert et al., 2012). GABAergic medium 73 spiny neurons (MSNs) are the principle output neurons in the NAc and are canonically divided 74 into subpopulations based on the expression of the dopamine receptor type 1 (D1 MSN) or D2 75 MSNs (Kalivas, 2009; Gerfen & Surmeier, 2011; Scofield et al., 2016). These two cell types show 76 divergent roles in reinforcement (Kravitz et al., 2012) and respond differently to drug exposure 77 (Kalivas, 2009; Lobo et al., 2010; Gerfen & Surmeier, 2011; Calipari et al., 2016; Scofield et al., 78 2016; van Zessen et al., 2021). Past work has examined plasticity in D1- and D2-MSNs produced 79 by opioid exposure (Glass et al., 2008; Graziane et al., 2016; Hearing et al., 2016; Zhu et al., 80 2016; Lefevre et al., 2023); however, these studies have been largely input agnostic, (except 81 Zhu et al., 2016 ). Further, it remains unclear whether 1) these adaptations map onto plasticity 82 associated with more translational models of volitional drug taking, 2) plasticity is unique within 83 mPFC-NAc circuits, and 3) adaptations are similar in females, as studies to date have been 84 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint selectively in males. To address these gaps, the present study investigated sex-specific effects 85 of self-administering remifentanil, the μOR-specific and highly potent synthetic opioid , on 86 enduring synaptic properties of D1 and D2 MSNs in the NAc core and shell that receive afferents 87 from the PL and IL, respectively. 88 89 2 MATERIALS AND METHODS 90 2.1 Animals 91 Adult male (N = 100; n = 51 – Remifentanil; n = 49 – Yoked Saline) and female (N = 48; n = 24 92 – Remifentanil; n = 24 – Yoked Saline) mice (PN45-60) were bred and raised in the animal facility 93 at Marquette University. Animals were housed in a temperature- and humidity-controlled room 94 maintained on a 12 -h reverse light/dark cycle. BAC transgenic mice double homozygous for 95 Drd1a−tdTomato and Drd2−eGFP mice (graciously provided by Dr. Rob Malenka , Stanford 96 University) (Pascoli et al., 2011; Nelson et al., 2012) and subsequently bred with C57BL/J6 mice 97 (Jackson Laboratories) to generate mice heterozygous for both transgenes. In some instances, 98 mice only expressing tdTomato in D1 MSNs (Jackson Laboratories) were used as tdTomato 99 signaling. For these mice, D1 MSNs were identified as td Tomato(+) and D2 MSNs were 100 identified as tdTomato(-), an approach previously used (Hearing et al., 2016; Anderson et al., 101 2019). All animals were group-housed and provided with food and water ad libitum until the start 102 of surgical procedures. All procedures were approved by Marquette University’s IACUC and 103 conform to the guidelines on humane animal care. 104 105 2.2 Intravenous catheter surgery. 106 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Intravenous (i.v.) catheter surgeries were performed under general isoflurane (1-3%; Covetrus) 107 anesthesia as previously described (Anderson et al., 2021). Briefly, mice were implanted with a 108 standard mouse catheter in the right jugular vein (Access Technologies; 2/3Fr. x 6cm silicone, 109 Cat No. AT -MJVC-2914A) and connected to a back mount (PlasticsOne; 22GA, Part No. 110 8I31300BM01). Following catheter surgery, mice were single-housed and allowed to recover for 111 at least 5 days prior to beginning self-administration. Before and after self-administration 112 sessions, catheters were immediately flushed with 0.05 mL (i.v.) of heparinized (20 IU/m L; 113 Hospira, Inc.) bacteriostatic 0.9% saline (Hospira, Inc.) containing gentamicin sulphate 114 (Sparhawk Laboratories, Inc.; 0.25 mg/mL) and containing enrofloxacin (Norbrook Laboratories; 115 4.994 mg/mL) respectively. Intravenous application of 0.05 mL of Brevital Sodium (5mg/mL) was 116 periodically used to assess catheter patency. 117 118 2.3 Intracranial virus injection of ChR2-AAV. 119 Intracranial viral surgeries were either performed the week prior to IV catheter surgeries or within 120 48 hours after the final day of remifentanil self-administration. pAAV9-CaMKIIa-hChR2(H134R)-121 EYFP was a gift from Karl Deisseroth (Addgene plasmid #26969; http://n2t.net/addgene:26969; 122 RRID:Addgene_26969) (Lee et al., 2010). Virus was bilaterally injected (0.5 µl; 0.1 µl /min) into 123 the PL (Males: AP: +1.8, ML: ± 0.4, DV: -2.3 mm; Females: AP: +1.75, ML: ± 0.4, DV: -2.3 mm) 124 or IL (Males: AP: +1.8, ML: +/- 0.40, DV: -3.2mm; Females: AP: +1.75, ML: +/- 0.40, DV: -3.2mm) 125 using standard stereotaxic apparatus attached with a 10 µL Hamilton syringe (Hamilton). Post 126 injections, the syringe was left in place for 5 min to allow for diffusion of the virus into the tissue. 127 128 2.4 Self-administration. 129 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint We used a ‘paired’ remifentanil self-administration paradigm that has been previously published 130 (Anderson et al., 2021). Briefly, all animals were initially food restricted to 85-90% of their original 131 weight and habituated overnight in their home cage to 25% liquid Ensure® diluted with water. 132 On the following day, mice began training in standard operant boxes installed with a liquid dipper 133 system (Med Associates, Inc .). Mice were trained to press the active lever which resulted in a 134 20 s presentation of Ensure® dipper and a simultaneous infusion of either saline (0.025 mL) or 135 remifentanil HCl (0.025mL of 5µg/kg/infusion; Ultiva®; Mylan Institutional, LLC; purchased from 136 Froedtert Hospital Pharmacy, Milwaukee, WI), as previously published (Anderson et al., 2021). 137 Training was conducted using an increasing fixed-ratio (FR) schedule. On the first day of training, 138 each active lever press resulted in a 20 s presentation of the dipper containing Ensure® and 139 saline/remifentanil infusion (FR1) for a maximum of 25 dipper/infusions. The following day, mice 140 that obtained the maximum number of reinforcer s underwent FR2 with a total of 50 141 dipper/infusion pairings. On the final day of training, 3 active lever presses were required (FR3) 142 for each dipper/infusion presentation with a maximum of 100 reward presentations. Animals 143 were required to reach the maximum number of reinforcers for each day (FR1 = 25, FR2 = 50, 144 FR3 = 100) before proceeding with the next FR training. Following completion of training, mice 145 underwent 1 -3 days of forced abstinence followed by one self-administration session of the 146 remifentanil or saline alone before food was returned ad libitum. 147 Male and female m ice that successfully acquired training progressed to 148 remifentanil/saline self-administration. Self-administration was conducted for an average of 5 149 days/week with each session lasting for 2 -3 hours per day . During self-administration, each 150 active lever press resulted in an i.v. infusion of remifentanil or saline (FR1) and was paired with 151 the presentation of visual cue (light) . Between infusions , a 20 -second time out period was 152 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint present during which levers were extended but not reinforced . A minimum of 10 remifentanil 153 infusions and discrimination between the active and inactive lever was used as self-154 administration criterion. Following self-administration, mice underwent 2-3 weeks of forced 155 abstinence in their home cage prior to electrophysiology recordings. 156 157 2.5 Slice electrophysiology. 158 Two to three weeks after the last self-administration session, mice were mildly anesthetized with 159 isoflurane (Covetrus), decapitated, the brain removed and put in oxygenated (95% O2/5% CO2) 160 ice-cold sucrose artificial cerebrospinal fluid ( sucrose ACSF; containing in mM: Sucrose 228, 161 KCl 2.5, NaH2PO4 1.0, MgSO 4 7.0, CaCl 2 0.5, NaHCO 3 26, and glucose 11) solution. Acute 162 coronal slices (300 µm) containing the NAc core and shell were prepared with a vibratome (Leica 163 VT1000S) and transferred to an incubation chamber filled with oxygenated ACSF (containing in 164 mM: NaCl 119, KCl 2.5, NaH2PO4 1.0, MgCl2 1.3, CaCl 2 2.5, NaHCO3 26.2 and glucose 11). 165 Slices were incubated at room temperature for 45 mins before being placed in the recording 166 chamber. Slices were continuously superfused with oxygenated ACSF maintained at 29 -32°C 167 using a single inline solution heater (Warner Instruments) throughout recordings. MSNs were 168 identified by their morphology and capacitance (>50 pF). Anatomical organization was used to 169 distinguish between MSNs in NA core vs NA shell. Fluorescent tags were used to identify MSNs 170 receiving direct input from PL or IL. When using Drd1a-tdTomato mice, D1 MSNs were identified 171 as td Tomato(+) and putative D2 MSNs as td Tomato(-). When using Drd1a-tdTomato/Drd2-172 eGFP double transgenics, D1 MSNs were identified as tdTomato(+) and D2 MSNs identified as 173 eGFP(+). Cells were checked for overlap in fluorescence in these mice to ensure recordings 174 were from cells selectively expressing one fluorescent tag. Additionally, e lectrophysiological 175 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint recordings were selectively conducted in MSNs located in dorsomedial NAc core and dorsal 176 aspect of rostral NAc shell. 177 For miniature excitatory postsynaptic currents (mEPSCs), 0.7 mM lidocaine was added 178 to the recording solution. mEPSCs were recorded at a holding potential (Vh) of -72 mV with 179 borosilicate glass pipettes (2.5-4.5 MΩ; Sutter Instruments) filled with Cs-based internal solution 180 (containing in mM: CsMeSO4 120, CsCl 15, TEA-Cl 10, NaCl 8, HEPES 10, EGTA 5, spermine 181 0.1, QX-314 5, ATP-Mg 4, and GTP-Na 0.3). Measurement, data collection, and analysis were 182 performed as described (M. C. Hearing et al., 2016; Kourrich et al., 2007; Madayag et al., 2019). Series 183 resistance was monitored continuously during all recordings, and a change beyond 20% resulted 184 in exclusion of the cell from data analyses. Data were filtered at 2 kHz and digitized at 5 kHz via 185 a Sutter Integrated Patch Amplifier and Igor Pro software (Wavemetrics). At the beginning of 186 each sweep, a depolarizing step (4 mV, 100 ms) was generated to monitor series (10 –40 MΩ) 187 and input resistance (>400 MΩ). Holding potentials were corrected for liquid junction potential 188 (∼8 mV). 189 AMPAR/NMDAR (A/N) ratios were computed from optically-evoked EPSCs (oEPSCs) at 190 +40 mV with and without 50 µM D-aminophosphoonovaleric acid (D -APV; selective NMDAR 191 antagonist). EPSCs were obtained at 0.1 Hz, with pulses of 473 -nm wavelength full -field light 192 (submerged 40x objective, Olympus) using a SOLA SE II354 Light Engine (Lumencor). Pulse 193 duration (0.3-2-ms) and light intensity were adjusted to obtain EPSC amplitudes of ~150-500 pA 194 across cells. For assessment of AMPAR subunit composition, a current-voltage (I-V) curve was 195 plotted, with AMPAR-mediated EPSCs measured at -72, +8, and +48 mV following correction of 196 holding potential for liquid junction potential. AMPAR rectification indices were calculated by 197 dividing the amplitude of AMPAR current at +48, +8 and -72 mV by the amplitude of AMPAR 198 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint current at -72 mV. For paired-pulse ratio experiments, EPSCs were measured at 50, 100, and 199 200 ms interstimulus intervals (ISI). The paired-pulse ratio at individual ISI was calculated by 200 dividing the amplitude of EPSC2:EPSC1. 201 202 2.6 Statistical Analysis. 203 Statistical analyses employed mixed-, multivariate- or univariate-ANOVAs and were conducted 204 using IBM SPSS (v30, New York NY, USA). Two separate mixed-ANOVAs were used to assess 205 sex-specific differences in lever pressing behavior of mice that received saline or remifentanil. 206 We also used unpaired t -test employing Welch’s correction to assess sex differences in 207 cumulative infusions. Electrophysiology data from yoked-saline control groups was compared to 208 assess baseline cell type- and sex-specific differences in synaptic properties of D1 vs D2 MSNs. 209 Separate ANOVAs were conducted for analyzing baseline differences in synaptic properties of 210 PL-Core and IL-Shell MSNs. To evaluate the influence of remifentanil treatment and sex on 211 synaptic properties, we conducted separate ANOVAs on D1 vs D2 PL-Core MSNs. Similarly, we 212 conducted separate ANOVAs on D1 vs D2 IL-Shell. Multivariate-ANOVA was used to assess 213 group differences in mEPSC amplitude and frequency. Group differences in A/N ratios and 214 AMPA rectification index were assessed using separate univariate -ANOVAs. A mixed-ANOVA 215 was used to assess group differences in paired-pulse ratio at different ISIs. We also analyzed 216 group differences in paired-pulse ratio at each ISI using separate univariate -ANOVAs. All 217 ANOVAs employed Bonferroni correction for pairwise comparisons. Whenever tests of sphericity 218 were violated, Greenhouse -Geisser corrected degrees of freedom are reported. GraphPad 219 Prism 10.4.2 was used to graph the data. 220 221 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 3 RESULTS 222 To determine how contingent administration of a clinically relevant opioid, remifentanil, 223 impacts mPFC-NAc circuit function, male and female mice underwent 10 –14 days of saline or 224 remifentanil self-administration followed by 14–22 days of forced abstinence (Fig. 1A). During 225 maintenance of remifentanil self-administration (days 5-14), both males and females showed 226 significantly greater pressing of the active lever compared to the inactive lever across all self-227 administration days (Lever x Day: F(4.14, 302.32) = 5.044, p < 0.001), with no differences in active 228 or inactive presses across sex (Sex x Lever x Day interaction: F(4.14, 302.32) = 1.244, p = 0.292) 229 (Fig. 1B). We also compared the number of remifentanil infusions between males and females 230 during the maintenance phase of remifentanil self-administration. Number of infusions changed 231 significantly across maintenance of remifentanil self-administration but did not differ between 232 males and females (Infusions: F(4.71, 343.67) = 4.745, p < 0.001; Infusions x Sex : F(4.71, 343.67) = 233 1.626, p = 0.157) (Fig. 1C). Examination of c umulative infusions showed a marginall y greater 234 number of infusions in females compared to males (t(45.89) = 2.010, p = 0.050) (Fig. 1D). 235 236 3.1 Sex- and cell-type specific baseline properties of NAc core and shell MSN synaptic 237 transmission 238 We assessed sex- and cell-type specific baseline differences in the synaptic transmission of D1 239 and D2 NAc core and shell MSNs, agonistic of input, via amplitude and frequency of mEPSCs. 240 In the NAc core, a significant sex and cell-type interaction was observed for mEPSC amplitude 241 (Cell-type x Sex: F(1,39) = 4.785, p = 0.035; Fig. 2A, B). In males, D2 MSNs exhibited greater 242 mEPSC amplitude compared to D1 MSNs (p = 0.003), while the amplitude of mEPSCs in D1 243 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint MSNs were greater in females compared to males (p = 0.010). Examination of mEPSC 244 frequency showed a significant effect of cell-type (F(1, 39) = 9.964; p = 0.003) but not sex (F(1, 39) 245 = 4.006; p = 0.052) nor an interaction between cell-type and sex (F(1,39) = 1.546; p = 0.221). 246 Specifically, the mEPSC frequency of D2 MSNs was significantly greater than D1 MSNs (p = 247 0.003) in the NAc core (Fig. 2C). In the NAc shell, no significant differences in mEPSC amplitude 248 were observed across cell-type or sex (Cell-type: F(1,37) = 0.113; p = 0.738; Sex: F(1,37) = 1.437; 249 p = 0.238; Cell-type x Sex: F(1,37) = 0.043; p = 0.837; Fig. 2D, E). Conversely, mEPSC frequency 250 of NAc shell MSNs showed a significant effect of cell -type (F(1,37) = 5.673; p = 0.022) with no 251 differences observed across sex (F(1,37) = 0.608; p = 0.441) or an interaction (F(1,37) = 1.469; p = 252 0.233). Specifically, the mEPSC frequency of D2 MSNs was significantly greater than D1 MSNs 253 (p = 0.022) in the NAc shell (Fig. 2F). Taken together, these data indicate that in males, excitatory 254 transmission in the NAc core is greater at D2 versus D1 MSNs and that compared to males, 255 females exhibit higher transmission at D1 MSNs. 256 257 3.2 Sex- and cell -type specific baseline properties of PL -Core and IL -Shell MSN 258 synaptic transmission 259 We next examined whether cell-type or sex -specific differences in baseline pre- and 260 postsynaptic signaling were present at isolated PL-Core and IL-Shell synapses using oEPSCs. 261 Examination of A/N ratio in PL-Core circuits showed a significant effect of cell -type (F(1, 26) = 262 31.431, p < 0.001) but not sex (F(1, 26) = 1.890, p = 0.181), or an interaction between cell-type 263 and sex (F(1, 26) = 0.097, p = 0.758). Specifically, A/N ratios were greater at D1 compared to D2 264 PL-Core synapses (p < 0.001; Fig. 3A, B). To test whether these differences reflected 265 distinctions in AMPAR subunit composition, we evaluated current -voltage (I-V) relationships of 266 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint oEPSCs. A significant effect of cell-type (F(1, 26) = 10.284, p = 0.004) but no effect of sex (F(1, 26) 267 = 1.809, p = 0.190) or interaction between cell-type and sex (F(1, 26) = 2.050, p = 0.164) was 268 observed, with AMPAR rectification indices lower in D2 versus D1 PL-Core synapses (p = 0.004; 269 Fig. 3C-E). These differences in the AMPAR rectification indices indicate that lower A/N ratios 270 observed at D2 PL-Core synapses likely reflect an increased presence of calcium-permeable 271 AMPARs (CP-AMPARs) at baseline . Examination of presynaptic release probability using 272 paired-pulse ratio s showed that paired-pulse ratio s significantly changed at increasing ISIs 273 (F(1.25, 42.36) = 24.755, p < 0.001). But there was no significant effect of cell -type (F(1.25, 42.36) = 274 1.539, p = 0.222), sex (F(1.25, 42.36) = 0.497, p = 0.525) or an interaction between cell-type and sex 275 (F(1.25, 42.36) = 0.318, p = 0.625). However, because shorter ISI durations are more often shaped 276 by residual presynaptic Ca2+ while longer ISIs may reflect recover kinetics at presynaptic sites 277 (Dittman et al., 2000), we conducted subsequent analyses in paired-pulse ratio at 50, 100 and 278 200 ms ISI separately. At 50 ms ISI, we found a significant effect of cell-type (F(1, 34) = 11.297, p 279 = 0.002) and sex (F(1, 34) = 6.980, p = 0.012) but did not see a significant interaction between cell-280 type and sex (F(1, 34) = 0.381, p = 0.541). At 100 ms ISI, we found a significant effect of cell-type 281 (F(1, 34) = 4.224, p = 0.048) and sex ( F(1, 34) = 6.892, p = 0.013) but did not see a significant 282 interaction between cell -type and sex ( F(1, 34) = 1.501, p = 0.229). At 200 ms ISI, we found a 283 significant effect of sex (F(1, 34) = 8.164, p = 0.007), but did not see an effect of cell-type (F(1, 34) = 284 2.711, p = 0.109) nor a significant interaction between cell -type and sex ( F(1, 34) = 0.893, p = 285 0.351; Fig. 3F, G). Taken together, these results indicate that the decrease in presynaptic release 286 probability at D1 vs D2 PL-Core synapses is likely driven by males while not differing in females. 287 Examination of IL-Shell inputs showed distinctly different patterns of baseline signaling. 288 While not significant, a cell-type (F(1, 27) = 3.507, p = 0.072) and sex-specific (F(1, 27) = 3.346, p = 289 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 0.078) trend towards increased A/N ratio was observed at D1 vs D2 IL-Shell synapses (p = 290 0.072) likely driven by increased A/N ratio at D1 IL-Shell synapses of females vs males ( p = 291 0.058; Fig. 3H, I). AMPAR rectification indices at D1 and D2 IL-Shell synapses did not show 292 significant differences across cell-type (F(1, 25) = 1.007, p = 0.325), sex (F(1, 25) = 1.737, p = 0.199), 293 nor was there an interaction between cell-type and sex (F(1, 25) = 1.007, p = 0.325; Fig. 3J-L). In 294 contrast, assessment of release probability via paired-pulse ratio s showed a significant 295 interaction of cell -type and sex across all ISIs (F(2, 6 6) = 7.861, p < 0.001). Post hoc pairwise 296 comparisons revealed that within sex, paired-pulse ratios at male D1 IL-Shell synapses were 297 higher at the 50 ms ISI compared to D2 IL-Shell synapses (p = 0.006), whereas ratios at female 298 D1 IL-Shell synapses were marginally higher at the 200 ms ISI compared D2 IL-Shell synapses 299 (p = 0.059; Fig. 3M, N). 300 301 3.3 Cell-type specific effects of remifentanil on D1 and D2 MSN signaling in NAc core 302 and shell 303 Previous work has shown that non -contingent opioid exposure promotes cell-type and region-304 specific adaptations in the NAc (Hearing et al., 2016; Madayag et al., 2019); however, it remains 305 unclear whether similar forms of pathway agnostic plasticity are observed after volitional opioid 306 taking and whether this plasticity is isolated to select pathways within mPFC-NAc circuits. 307 Initial examination of agnostic synaptic transmission (assessed by measuring mEPSCs) 308 showed a significant interaction of treatment (saline, remifentanil) and sex (male, female) in the 309 amplitude of mEPSCs of D1 NAc core MSNs (F(1, 56) = 5.485, p = 0.023). Within sex, remifentanil 310 males (p < 0.001) but not females (p = 0.573) showed significantly greater mEPSC amplitudes 311 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint compared to saline counterparts. Additionally, within treatment groups, mean amplitude of 312 mEPSCs of D1 NAc core MSNs was significantly greater in saline females compared to saline 313 males (p = 0.003), however no difference was observed in remifentanil treated mice (p = 0.914; 314 Fig. 4A, B). Assessment of mEPSC frequency in D1 NAc core MSNs showed a significant effect 315 of treatment (F(1, 56) = 5.780, p = 0.020) in that remifentanil increased mEPSC frequency. 316 However, no significant effect of sex ( F(1, 56) = 1.604, p = 0.211) or a significant interaction 317 between treatment and sex (F(1, 56) = 0.143, p = 0.706; Fig. 4C) was observed. In D2 NAc core 318 MSNs, a main effect of treatment was observed for both mEPSC amplitude (F(1, 32) = 6.392, p = 319 0.017) and frequency (F(1, 32) = 16.886; p < 0.001), with remifentanil reducing both, compared to 320 saline controls. No significant effect of sex (amplitude: F(1, 32) = 0.095, p = 0.760); frequency: (F(1, 321 32) = 2.57, p = 0.119) or a significant interaction between treatment and sex (amplitude: F(1, 32) = 322 1.231; p = 0.276; frequency: F(1, 32) = 2.742, p = 0.108) was observed (Fig. 4D-F). Taken together, 323 these data indicate that remifentanil upregulates excitatory signaling at D1 NAc core MSNs 324 primarily in males, whereas there is a general reduction at D2 NAc core MSNs in both sexes. 325 To assess pathway agnostic plasticity in D1 and D2 NAc shell MSNs, we measured 326 remifentanil-induced alterations in the amplitude and frequency of mEPSCs. Remifentanil 327 significantly increased the amplitude of mEPSCs of D1 NAc shell MSNs ( F(1, 36) = 4.368, p = 328 0.044). However, there was no significant effect of sex (F(1, 36) = 0.305, p = 0.584) or a significant 329 interaction between treatment and sex (F(1, 36) = 2.005, p = 0.165; Fig4. G, H). The frequency of 330 mEPSCs of D1 NAc shell MSNs showed a significant interaction between treatment and sex 331 (F(1, 36) = 8.598, p = 0.006). Specifically, mEPSC frequency increased significantly in D1 NAc 332 shell MSNs of males but not females (p < 0.001) after remifentanil treatment vs saline (Fig. 4I). 333 In D2 NAc shell MSNs, there was n either a significant effect of remifentanil treatment on the 334 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint amplitude of mEPSCs (F(1, 40) = 0.141, p = 0.709) across sex (F(1, 40) = 1.049, p = 0.312) nor was 335 there a significant interaction between treatment and sex (F(1, 40) = 0.271, p = 0.606; Fig. 4J, K). 336 However, remifentanil treatment significantly decreased the frequency of D2 NAc shell MSNs 337 (F(1, 40) = 12.865, p < 0.001) in both males and females ( F(1, 40) = 5.304, p = 0.027) but no 338 significant interaction between treatment and sex (F(1, 40) = 0.003, p = 0.958; Fig. 4L) was 339 observed. 340 341 3.4 Impact of remifentanil on PL-Core D1 and D2 MSN pre- and postsynaptic signaling 342 Past work has examined cell-type specific plasticity following non -contingent opioid exposure 343 (Graziane et al., 2016; Hearing et al., 2016) or assessed plasticity within select circuits including 344 inputs from the IL to D1 MSNs (Hearing et al., 2016) and paraventricular nucleus of the thalamus 345 (Zhu et al. , 2016) input to D1 and D2 MSN s. However, no studies to date have compare d 346 plasticity associated with volitional opioid taking or directly compared plasticity within sub-circuits 347 and cell -types. Given the roles of PL and IL subregions in opioid -related behavior, we next 348 examined whether inputs from these cortical regions to NAc MSNs differs following remifentanil 349 and abstinence. 350 Examination of A/N ratios at D1 PL -Core synapses showed a significant effect of 351 treatment, with remifentanil reducing A/N ratios in both male s and females compared to their 352 saline counterparts (F(1, 32) = 41.864, p < 0.001). No significant effect of sex (F(1, 32) = 0.623, p = 353 0.436) or interaction between sex and treatment (F(1, 32) = 0.137, p = 0.714) was observed (Fig. 354 5A, B). Assessment of AMPAR rectification showed a significant reduction in indices in 355 remifentanil male and female mice compared to saline controls ( F(1, 30) = 33.057, p < 0.001). 356 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint However, we did not observe a significant effect of sex (F(1, 30) = 0.741, p = 0.396) or a significant 357 interaction between sex and treatment ( F(1, 30) = 0.568, p = 0.457; Fig. 5C-E). Examination of 358 release probability showed a marginally significant impact of remifentanil treatment on paired-359 pulse ratio at D1 PL-Core synapses (ISI x Treatment: F(1.31, 52.28) = 3.485, p = 0.057). However, 360 we did not observe any significant effect of sex (ISI x Sex: F(1.31, 52.28) = 0.731, p = 0.484) or an 361 interaction between sex and treatment on paired-pulse ratio (ISI x Treatment x Sex: F(1.31, 52.28) 362 = 0.925, p = 0.366). We also assessed changes in paired-pulse ratio of D1 PL-Core synapses 363 at 50, 100 and 200 ms ISI separately. At 50 ms ISI, we found a significant effect of remifentanil 364 treatment (F(1, 40) = 11.790, p = 0.001) on paired-pulse ratio of D1 PL-Core synapses such that 365 remifentanil treatment decreased paired-pulse ratio vs saline treatment . We also found a 366 significant effect of sex (F(1, 40) = 7.073, p = 0.011) with paired-pulse ratio at D1 PL-Core synapses 367 of males being greater than females . However, the interaction between treatment and sex (F(1, 368 40) = 1.011, p = 0.321) was not significant. At 100 ms ISI, we found a significant effect of sex (F(1, 369 40) = 5.858, p = 0.020) and marginally significant interaction between treatment and sex (F(1, 40) = 370 3.346, p = 0.075), but no significant effect of remifentanil treatment (F(1, 40) = 2.118, p = 0.153). 371 Specifically, remifentanil treatment significantly decreased paired-pulse ratio at D1 PL -Core 372 synapses of males vs saline ( p = 0.024) but not in females (p = 0.795). Within saline treated 373 mice, however, paired-pulse ratio at D1 PL-Core synapses of males was significantly greater 374 than that of saline treated females ( p = 0.006). At 200 ms ISI, we found a significant effect of 375 treatment (F(1, 40) = 6.531, p = 0. 015), with the paired-pulse ratio at D1 PL -NAc synapses 376 decreasing significantly after remifentanil treatment vs saline (p = 0.015). We also observed a 377 significant effect of sex (F(1, 40) = 9.826, p = 0.003) with paired-pulse ratio at D1 PL-Core synapses 378 being greater for males vs females. However, we did not see a significant interaction between 379 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint treatment and sex (F(1, 40) = 2.694, p = 0.109; Fig. 5F, G). These data indicate that remifentanil 380 increased postsynaptic strength at D1 PL-Core synapses across sex through upregulation of 381 CP-AMPA receptors, whereas increases in presynaptic strength are specific to males, due in 382 part to higher release under drug naïve conditions in females. 383 In contrast to D1 MSNs, a significant effect of remifentanil treatment (F(1, 26) = 5.687, p = 384 0.025) and sex (F(1, 26) = 7.245, p = 0.012) was observed with A/N ratios at D2 PL-Core synapses. 385 While not significant, a trend towards an interaction between sex and treatment was observed 386 (F(1, 26) = 3.604, p = 0.069). Assessment of pairwise comparisons demonstrated that a significant 387 effect of remifentanil treatment was driven by an increase in A/N ratios at D2 PL-Core synapses 388 of remifentanil treated males compared to saline controls (p = 0.006; Fig. 5H, I). Examination of 389 AMPAR rectification indices of D2 PL-Core synapses identified an interaction between treatment 390 and sex (F(1, 27) = 6.821, p = 0.015), with indices at D2 PL-Core synapses of males significantly 391 greater in remifentanil treated mice vs saline controls (p = 0.033). No significant difference s 392 were observed at D2 PL-Core synapses between male vs females within saline ( p = 0.089) or 393 remifentanil groups ( p = 0.061), nor was and difference observed across treatment in females 394 (p = 0.150; Fig. 5J-L). Examination of release probability at D2 PL-Core synapses did not 395 demonstrate an overall significant effect of remifentanil treatment on paired-pulse ratio across 396 ISIs (F(1.25, 41.20) = 0.001, p = 0.986) nor a significant effect of sex (F(1.25, 41.20) = 0.621, p = 0.469). 397 There was also no significant interaction between treatment and sex across ISIs ( F(1.25, 41.20) = 398 0.098, p = 0. 811). However, examinations of changes in paired-pulse ratio of D 2 PL-Core 399 synapses at 50, 100 and 200 ms ISI separately demonstrated significant group differences . At 400 50 ms ISI, we found a significant effect of remifentanil treatment (F(1, 33) = 11.902, p = 0.002) and 401 a marginally significant effect of sex ( F(1, 33) = 3.634, p = 0.065) on paired-pulse ratio of D2 PL-402 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Core MSNs. Specifically, paired-pulse ratio was significantly increased at D2 PL-Core synapses 403 after remifentanil treatment vs saline, and it was significantly greater in males vs females (p = 404 0.065). However, the interaction between treatment and sex (F(1, 33) = 0.559, p = 0.460) was not 405 significant. At 100 ms ISI, we found a significant effect of treatment (F(1, 33) = 17.602, p < 0.001) 406 in that paired-pulse ratio was increased at D2 PL-Core synapses after remifentanil treatment vs 407 saline. However, we did not observe a significant effect of sex (F(1, 33) = 2.476, p = 0.125) or a 408 significant interaction between treatment and sex (F(1, 33) = 0.313, p = 0.579) at these synapses. 409 At 200 ms ISI, we found a significant effect of treatment (F(1, 33) = 15.755, p < 0.001) with paired-410 pulse ratio at D2 PL-Core synapses increasing significantly after remifentanil treatment vs saline. 411 We also observed a significant effect of sex (F(1, 33) = 5.059, p = 0.031) with paired-pulse ratio at 412 D2 PL-Core synapses being greater for males vs females. However, we did not see a significant 413 interaction between treatment and sex ( F(1, 33) = 0.525, p = 0.474; Fig. 5M, N). Together these 414 data indicated that in males but not females, remifentanil alters the composition of AMPAR 415 subunits through downregulation of CP-AMPA receptors, while presynaptic strength is 416 increased. 417 418 3.5 Impact of remifentanil on IL-Shell D1 and D2 MSN pre- and postsynaptic signaling 419 Previous work has shown that non-contingent morphine exposure promotes pathway 420 non-specific adaptations in NAc shell MSNs and that adaptations at D1 IL-Shell synapses play 421 a role in opioid reward behavior (Hearing et al., 2016). Therefore, we next examined whether 422 remifentanil produced overlapping or divergent forms of plasticity at IL-Shell synapses. At D1 IL-423 Shell synapses, mean A/N ratios were marginally increased by remifentanil treatment (F(1, 31) = 424 3.925, p = 0.056) across sex. There was also no significant effect of sex (F(1, 31) = 2.499, p = 425 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 0.124), but we found a marginally significant interaction between sex x treatment (F(1, 31) = 3.966, 426 p = 0.055). Assessment of pairwise comparisons demonstrated that while A/N ratio did not differ 427 between males vs females (p = 0.760) after remifentanil treatment, it was decreased at D1 IL-428 Shell synapses of females vs saline control ( p = 0.017; Fig. 6A, B). Assessment of AMPAR 429 rectification index showed a significant treatment effect, with indices lower in remifentanil treated 430 mice compared to saline controls (F(1, 30) = 5.584, p = 0.025). However, there was no significant 431 effect of sex (F(1, 23) = 0.234, p = 0.632) or an interaction effect between sex and treatment (F(1, 432 30) = 0.857, p = 0. 362; Fig. 6C-E). Examination of paired-pulse ratio s showed that release 433 probability remained unchanged by r emifentanil treatment across ISIs (ISI x Treatment: F(1.54, 434 64.70) = 0.066, p = 0.894), with no impact of sex (ISI x Sex: F(1.54, 64.70) = 0.211, p = 0.752) nor a 435 significant interaction between treatment and sex ( ISI x Treatment x Sex: F(1.54, 64.70) = 0.395, p 436 = 0.621) at D1 IL-Shell synapses. However, subsequent comparisons within 50, 100 and 200 437 ms ISIs separately demonstrated significant alterations . Specifically, we found a significant 438 reduction in paired-pulse ratio at D1 IL-Shell synapses of remifentanil versus saline controls at 439 50 ms ISI (F(1, 42) = 4.582, p = 0.038), with trends towards a reduction at 100 ms (Treatment: F(1, 440 42) = 3.145, p = 0.083) and 200 ms (F(1, 42) = 3.860, p = 0.056). However, no significant effect of 441 sex (50 ms: F(1, 42) = 0.665, p = 0.419; 100 ms: F(1, 42) = 0.308, p = 0.582; 200 ms: F(1, 42) = 0.165, 442 p = 0.687) or a significant interaction between treatment and sex ( 50 ms: F(1, 42) = 0.299, p = 443 0.587; 100 ms: F(1, 42) = 0.018, p = 0.893; 200 ms: F(1, 42) = 0.320, p = 0.575) was observed at 444 any ISI at these synapses (Fig. 6F, G). 445 At D2 IL-Shell synapses, there was no significant effect of remifentanil treatment (F(1, 25) 446 = 0.406, p = 0.530), sex (F(1, 25) = 1.239, p = 0.276) or interaction between treatment and sex 447 (F(1, 25) = 0.005, p = 0.947) on A/N ratios (Fig. 6H, I). In contrast to D1 IL -Shell synapses, no 448 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint impact of treatment (F(1, 23) = 0.007, p = 0.935) or sex (F(1, 23) = 1.436, p = 0.243), or an interaction 449 of treatment and sex (F(1, 23) = 0.428, p = 0.519) was observed for AMPAR rectification indices 450 at D2 IL-Shell synapses (Fig. 6J-L). Assessment of paired-pulse ratios at D2 IL-Shell synapses 451 showed a significant interaction between treatment and sex across ISIs (F(2, 64) = 15.286, p < 452 0.001). Assessment of pairwise comparisons showed that at D2 IL -Shell synapses in males, 453 remifentanil treatment increased paired-pulse ratio at the 50 ms ISI compared to saline (p = 454 0.004). Examination within individual ISIs showed that at 50 ms ISI, a significant interaction 455 between treatment and sex ( F(1, 32) = 5.922, p = 0.021) was observed , with males exhibiting 456 increased paired-pulse ratio following remifentanil treatment vs females. At 100 ms ISI, we found 457 a significant effect of sex ( F(1, 32) = 4.209, p = 0.048) but no significant effect of remifentanil 458 treatment (F(1, 32) = 0.429, p = 0.517) or an interaction between treatment and sex (F(1, 32) = 1.367, 459 p = 0.251). At 200 ms ISI, we also found a significant effect of sex (F(1, 32) = 7.249, p = 0.011) 460 with paired-pulse ratio at D2 IL-Shell synapses of males being significantly greater than females 461 (p = 0.011). However, we did not see a significant effect of treatment (F(1, 32) = 0.070, p = 0.793) 462 or a significant interaction between treatment and sex ( F(1, 32) = 0.486, p = 0.491; Fig. 6M, N). 463 Together, these data suggest that decreased presynaptic strength at D2 MSNs is likely due to 464 decreased presynaptic release probability at D2 IL-Shell synapses in males. 465 466 4 DISCUSSION 467 Previous research, including our own, has characterized opioid -induced 468 neuroadaptations in NAc glutamate signaling at the level of either specific cell types or 469 subregions (Graziane et al. , 2016; Hearing et al. , 2016; Madayag et al. , 2019) . However, 470 examination of sex-, cell type-, and input-specific baseline differences in glutamate signaling as 471 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint well as whether these variables influence opioid -induced plasticity has yet to be explored. 472 Furthermore, this past work has been confined to non -contingent models of opioid 473 administration. Our study reveals a complex pattern of baseline glutamatergic signaling and 474 synaptic plasticity arising during remifentanil abstinence that is specific to sex, NAc subregion 475 (core vs. shell), and neuronal cell type (D1 vs. D2 MSNs). 476 4.1 Differences in baseline MSN glutamate transmission 477 We first examined baseline differences in synaptic properties of D1 and D2 MSNs in the 478 NAc core and shell. In the core, D2 MSNs exhibited significantly greater mEPSC amplitude and 479 frequency compared to D1 counterparts in males but not females . While increase in amplitude 480 are likely attributed to enhanced postsynaptic signaling, elevations in frequency may be a result 481 of increased receptor (or synapse) number or increased presynaptic release probability 482 (Kerchner & Nicoll, 2008; Han & Stevens, 2009; Graziane et al., 2016; Hearing et al., 2016). 483 While p revious work has demonstrated higher excitatory drive at D2 MSNs in dorsal striatal 484 regions (Kreitzer & Malenka, 2007), this is the first known evidence that this phenotype is also 485 present in ventral striatal regions . Unlike previous studies in which no baseline differences in 486 glutamate signaling were observed at ventral hippocampus and BLA inputs to the NAc (MacAskill 487 et al., 2014), the present study observed cell-type and projection-specific baseline differences in 488 synaptic signaling. We observed lower A/N ratios and AMPAR rectification indices at D2 PL-Core 489 synapses compared to D1 synapses which likely reflect higher levels of CP-AMPAR mediated 490 signaling and greater synaptic strength. We also observed greater levels of release probability 491 at D2 PL-Core vs D1 synapses. The implications of this difference are unclear, it is possible that 492 it represents a baseline “protective” measure that when removed, may lead to maladaptive drug 493 related behavior, as reductions in synaptic strength at D2 NAc co re MSNs has been linked to 494 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint increased habit-like cocaine seeking behavior (Bock et al., 2013). Notably, we observed greater 495 levels of release probability at D2 vs D1 PL-Core synapses specifically in males. The lack of 496 differences in cell-type observed in females suggests a sex difference in baseline transmission. 497 In agreement with our findings, past work has shown that MSNs in the NAc core of females have 498 greater spine density (Forlano & Woolley, 2010; Wissman et al., 2011) and large spine diameter 499 (Forlano & Woolley, 2010) – findings that align with greater frequency of mEPSCs (Wissman et 500 al., 2011). Importantly, the present study is the first to demonstrate that these differences likely 501 exhibit cell-type and input -specificity. Therefore, it is possible that baseline diff erences in 502 synaptic properties between males and females may underlie sex differences in reward-related 503 disorders or increase vulnerability to transition to use disorders more rapidly (Becker & Hu, 2008; 504 Kniffin & Briand, 2024). 505 In contrast to the NAc core, baseline differences across cell-types were much more 506 restricted, with higher mEPSC frequency observed at D2 IL-Shell synapses of both males and 507 females. While a trend towards increased A/N ratios at IL-Shell D1 versus D2 synapses, 508 distinctions in presynaptic release were confined to males at these synapses. It is worth noting 509 that the lack of sex differences in the NAc shell agrees with previous pathway non -specific 510 findings that while MSNs in females have larger spines, no differences in MSN intrinsic 511 properties or mEPSCs were present (Willett et al., 2016). Regardless, increasing research is 512 demonstrating the influence of sex, and sex hormones on intrinsic and synaptic properties of D1 513 and D2 MSNs (Cao et al., 2018; Proano et al., 2018). Thus, future studies controlling for the 514 estrus cycle stage of female mice would help identify the role of sex hormones on the synaptic 515 properties of these neurons. 516 4.2 Cell- and Sex-Specific Neuroadaptations in the NAc Core after remifentanil 517 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint The most pronounced differences identified during remifentanil abstinence emerged in 518 the NAc core. In male mice, we observed a pathway non-specific increase in mEPSC amplitude 519 in D1 MSNs – an effect that was not observed in females. However, both sexes exhibited an 520 increase in mEPSC frequency, highlighting an upregulation of excitatory drive that is overlapping 521 and divergent across sex . Alternatively, isolation of changes at PL-Core synapses showed a 522 robust reduction of A/N ratios that aligned with a reduction in the AMPAR rectification indices in 523 remifentanil treated males and females, indicative of increased postsynaptic CP-AMPAR 524 signaling. These data coupled with a decreased paired-pulse ratio, provide convergent evidence 525 for enhanced presynaptic glutamate release and increased postsynaptic AMPA receptor 526 sensitivity that may be more pronounced at PL -Core synapses. Notably, this p otentiation of 527 excitatory signaling at D1 NAc MSNs has long been thought to reflect a hallmark of addiction -528 related neuroplasticity produced by both opioids and psychostimulants (Bock et al., 2013; Hikida 529 et al., 2016; Hearing et al., 2018; Klawonn & Malenka, 2018); however, this is the first indication 530 that this occurs following opioid self -administration. This is particularly pertinent , as pathway 531 non-specific increases at D1 MSNs of NAc core were not previously observed following non -532 contingent morphine exposure (Hearing et al. , 2016) . While unclear, it is possible that the 533 emergence of plasticity in a more motor-centric region of the NAc reflects the use of an operant- 534 and goal-directed model of drug exposure. 535 Co-occurring with augmented excitatory drive at D1 MSNs was an observed weakening 536 of glutamate signaling at D2 MSNs. This was prominent when assessing pathway non-specific 537 inputs (e.g., both mEPSC amplitude and frequency). Subsequent examination of PL -Core 538 plasticity showed more nuanced effects, however. Evidenced by an increase in the paired-pulse 539 ratio, presynaptic release was decreased in both males and females. Alternatively, examination 540 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint of postsynaptic plasticity showed effects selectively in males , with an increase in A/N ratios in 541 remifentanil treated mice. While CP-AMPARs are not traditionally thought to be the predominant 542 sub-unit under baseline conditions, rectification indices were lower and A/N ratios higher at D2 543 PL-Core synapses compared to D1 synapses in saline treated males. Thus, it is possible that 544 remifentanil promotes a subunit-composition shift by reducing expression of CP-AMPARs and 545 increasing expression of AMPARs that contain the GluA2 subunit. In support, GluA2-containing 546 AMPARs have a roughly linear current-voltage relationship that permits more outward current at 547 positive voltages, which in turn may lead to the observed increase in A/N ratios following 548 remifentanil self -administration. Given that D2 MSNs are part of the "no -go" pathway, which 549 suppresses inappropriate or maladaptive actions (Bock et al., 2013; Soares-Cunha et al., 2018; 550 Soares-Cunha et al., 2020), the simultaneous strengthening of the D1 pathway and weakening 551 of the D2 pathway may reflect a neurobiological bias towards drug-related behavior that arises 552 during abstinence and leaves individuals vulnerable to subsequent relapse. 553 554 4.2 Cell- and Sex-Specific Neuroadaptations in the NAc Shell after remifentanil 555 The shell subdivision of the NAc is highly connected with limbic and autonomic brain 556 regions and has been shown to be heavily involved in drug-associated motivation, reward 557 learning, and relapse (Heimer et al., 1997; Sesack & Grace, 2010; Bossert et al., 2012; Pascoli 558 et al., 2014; Hearing et al., 2016; Hikida et al., 2016). Previous work has shown that at protracted 559 withdrawal timepoints following repeated non-contingent morphine exposure, synaptic strength 560 is increased and decreased at NAc shell D1 and D2 MSNs, respectively (Graziane et al., 2016; 561 Hearing et al., 2016; Madayag et al., 2019). To our knowledge, our study is the first to directly 562 examine the impact of self -administered opioids on synaptic transmission in the NAc shell and 563 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint how this may differ based on biological sex. Similar to non -contingent exposure, remifentanil 564 increased mEPSC amplitude at D1 MSNs in both sexes, indicative of enhanced postsynaptic 565 AMPAR signaling. Unexpectedly, increases in mEPSC frequency were only observed in males. 566 Additionally, similar to non -contingent morphine, neuroadaptations at D2 MSNs were only 567 observed through decreases in mEPSC frequency – a phenomenon that was previously shown 568 to reflect reduced presynaptic release from pooled afferents (Hearing et al., 2016). This divergent 569 regulation of synaptic transmission in D1 vs. D2 MSNs is congruent with their antagonistic roles 570 in addiction behavior and the notion that MSN subpopulations receive differential afferent 571 innervation (Wall et al., 2013; MacAskill et al., 2014; Boxer et al., 2023). The exact mechanism 572 behind cell-type specific plasticity remains unclear; however, past work suggests that increased 573 release at D1 MSNs may reflect a reduction in presynaptic mu opioid receptor inhibition (James 574 et al., 2013). Conversely, reduction in release at D2 MSNs reflect alterations in endocannabinoid-575 mediated LTD (Grueter et al., 2010). Regardless, the overlap of these adaptations with non -576 contingent exposure as well as across sex likely highlight key adaptations associated with opioid 577 motivation and associative learning that modulate behavior. 578 Previous studies have indirectly highlighted the importance of the IL -Shell pathway in 579 relapse and reinstatement of conditioned place preference (Bossert & Daniel, 2006; Hearing et 580 al., 2016; Kruyer et al., 2019; Madayag et al., 2019; Chioma et al., 2021); however, the cell-type 581 and sex-specific nature of adaptations associated with behavior was previously unknown . As 582 seen in the PL-Core circuit, remifentanil produced a decrease in both the A/N ratio and AMPAR 583 rectification at D1 IL-Shell synapses. However, unlike the PL-Core, this effect was only observed 584 in males. Potentiation of this circuit aligns with previous findings following non -contingent 585 morphine and cocaine self -administration in male mice ; however, it highlights an intriguing 586 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint divergence in plasticity across sex. Furthermore, as no effects of remifentanil were observed in 587 the amplitude of mEPSCs in pathway agnostic D1 NAc shell MSNs, it suggests that opioid self-588 administration may specifically and more prominently impact the IL -Shell network to impact 589 behavior. Indeed, previous work has shown a selective potentiation of IL but not amygdala or 590 ventral hippocampal input at D1 NAc shell MSNs following cocaine self -administration (Pascoli 591 et al., 2014) – a phenomenon that is particularly intriguing given that shell MSNs have been 592 shown to receive greater innervation from ventral hippocampus afferents (Britt et al. , 2012) . 593 While the functional implications of these adaptations are unclear, they may underlie increased 594 negative affect and dysphoria during abstinence that motivates continued drug use (Flagel et al., 595 2007; Hikida et al., 2016; Madayag et al., 2019). 596 There are several limitations to this study that should be acknowledged. First, food-based 597 lever training and a fading procedure were used to initiate acquisition of self -administration. 598 Although this fading approach was utilized to expedite acquisition of lever pressing and reduce 599 the rate of failed acquisition, it also resulted in an atypical maintenance of lever pressing in saline 600 treated mice (see supplemental materials). We do not believe this reflects a lack of reinforcing 601 properties of remifentanil and sustained responding for Ensure, but rather, a lack of or slowed 602 extinction of responding for Ensure ®. There are a number of reasons for this : first, past work in 603 mice shows that the average time to extinguish responding for a non-drug reward is ~10-12 days 604 (Bobadilla et al., 2017). Second, previous work from our group using a similar fading approach 605 showed that remifentanil treated mice demonstrated increased motivation compared to saline in 606 a progressive ratio test (Anderson et al., 2021). Finally, it has been reported that mice will actively 607 press a lever to produce a light cue in an operant task (Olsen & Winder, 2009). We also do not 608 think that observed differences in physiological properties at baseline , as well as across 609 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint treatment reflect adaptations associated with extinction in saline mice . This is because in a 610 previous study we did not find differences in physiological properties of mPFC pyramidal neurons 611 between mice that underwent fading followed by saline self-administration compared to 612 behaviorally naïve mice (Anderson et al., 2021). Furthermore, baseline mEPSC amplitudes and 613 oEPSCs in the present study are nearly identical to our previous work where mice received non-614 contingent saline without extinction (Hearing et al., 2016). A second caveat of this study is that 615 a non -drug reward was not used for comparison with remifentanil , thus we are unable to 616 determine how selective these adaptations are for opioids. However, we have previously 617 demonstrated that two weeks of Ensure® self-administration did not promote plasticity in mPFC 618 pyramidal neurons (Anderson et al., 2021). Finally, our experiments do not address alterations 619 in gonadal hormones. As recent work has shown a role of estrous stage on cue/drug associations 620 (Johnson et al., 2019), heroin self-administration (Lacy et al., 2016), and remifentanil demand 621 (Lacy et al., 2020), a critical step in future studies will be to determine what role, if any, these 622 hormones have on plasticity. 623 624 5 CONCLUSIONS 625 Our ability to effectively treat substance use disorders is likely hindered by variability within 626 diagnosed populations. Biological sex is known to dictate drug-related behavior and outcomes, 627 as females exhibit heightened risk for substance use disorders. Data from the present study 628 collectively align with the broader research on sex differences in addiction, which often shows 629 that mechanisms underlying substance abuse may differ between males and females (Becker 630 & Chartoff, 2019; Kokane & Perrotti, 2020; Nicolas et al., 2022). While adaptations observed in 631 male and female mice cannot definitively be mapped on to alterations in humans, our findings 632 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint highlight several potentially key points of overlap produced by opioids as well as other illicit 633 substances that may represent targetable changes across sex and use disorders. 634 635 6 Conflict of Interest 636 The authors declare that the research was conducted in the absence of any commercial or 637 financial relationships that could be construed as a potential conflict of interest. 638 639 640 7 Author Contributions 641 M.C.H., A.C.M. and E.M.A. designed experiments; M.C.H., A.C.M., E.M.A., S.D., A.E., and L.F. 642 performed experiments; S.S.K and S.I.A . analyzed dat a, prepared figures and drafted 643 manuscript; S.S.K, S.I.A. and M.C.H. interpreted results of experiments, edited and revised the 644 manuscript. 645 646 8 Funding 647 This study was supported by NIH R00 DA038706-03 and R01 DA055956-01 awarded to M.C.H 648 by the National Institutes of Health’s National Institute for Drug Addiction. The content is solely 649 the responsibility of the authors and does not necessarily represent the official views of the 650 National Institutes of Health’s National Institute for Drug Addiction. 651 652 9 Acknowledgements 653 We would like acknowledge Megan Matre for her contributions in conducting behavioral 654 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint experiments include d in this st udy. We would also like to thank all members of the Hearing 655 laboratory, as well as Dr. Mark Thomas for initial technical insight and equipment support. 656 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Figures and captions 657 658 Figure 1. Experimental design and remifentanil self-administration in males and females. 659 A. From left to right: Location of i ntracranial channelrhodopsin (ChR) virus (AAV9-CaMKIIa-660 hChR2(H134R)-eYFP) injections; self-administration, and abstinence timeline ; and location of 661 whole cell patch electrophysiology recordings. Insert: Representative confocal image of the NAc 662 of Drd1a−tdTomato/Drd2−eGFP heterozygous mice expressing TdTomato (red) in D1 + MSNs 663 and eGFP (green) in D2 + MSNs. B. Summary graph of lever press ing behavior during the 664 maintenance phase (day 5-14) of remifentanil self-administration. Males (light blue circles) and 665 females (dark blue circles) had significantly greater number of active lever presses (solid circles) 666 vs inactive lever presses (bordered circles). C. Summary graph of average r emifentanil 667 infusions. Males (light blue squares) and females (dark blue squares) did not differ in daily 668 number of remifentanil infusions. D. Cumulative infusions of remifentanil in females (dark blue) 669 were marginally greater than males (light blue). Summary data are presented as Mean ± SEM. 670 ***p < .001 (Active vs Inactive); #p < .05 for sex effects. 671 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 672 Figure 2. D1 and D2 MSNs of NAc core and shell show sex -specific differences in their 673 baseline synaptic transmission 674 A. Representative traces of mEPSCs of D1 (light red) and D2 (light blue) NAc core MSNs of 675 saline treated males (top), and D1 (light orange) and D2 (light green) NAc core MSNs of saline 676 treated females (bottom). B. At baseline, mEPSC amplitude of D2 NAc core MSNs (light blue 677 bar) is higher than D1 NAc core MSNs (light red) in males. The amplitude of mEPSCs does not 678 differ between D1 (light orange) vs D2 (light green) NAc core MSNs of females. The amplitude 679 of mEPSCs of females is greater than males in D1 but not D2 NAc core MSNs. C. The frequency 680 of mEPSCs of D1 NAc core MSNs of males (light red) and females (light blue) is lower than 681 mEPSC frequency of D2 NAc core MSNs of males (light blue) and females (light green). D. 682 Representative traces of mEPSCs of D1 (light red) and D2 (light blue) NAc shell MSNs of saline 683 treated males (top), and D1 (light orange) and D2 (light green) NAc shell MSNs of saline treated 684 females (bottom). E. There were no differences in the mEPSC amplitude of D1 (light red) or D2 685 (light blue) NAc shell MSNs of males, and D1 (light orange) or D2 (light green) NAc shell MSNs 686 of females. F. The frequency of mEPSCs of D1 NAc core MSNs of males (light red) and females 687 (light blue) is lower than mEPSC frequency of D2 NAc core MSNs of males (light blue) and 688 females (light green). D1+/D2+ - D1/D2-expressing MSNs; M – males, F – females; Summary 689 data are presented as Mean ± SEM. ^, ^^p < 0.05, 0.01 for cell-type differences; #p < .05 for sex 690 effects. 691 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 692 Figure 3. Sex- and cell -type specific differences in baseline pre - and postsynaptic 693 signaling at PL-Core and IL-Shell synapses 694 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint A. Representative baseline traces of AMPAR and NMDAR mediated currents at PL -Core 695 synapses. NMDAR current (black) was measured at +40 mV and AMPAR current (colored) was 696 measured at +40 mV in the presence of D-APV. B. Summary graphs of AMPA/NMDA (A/N) ratios 697 showing that at baseline, D1 PL -Core synapses in both males (light red) and females (light 698 orange) exhibited significantly higher A/N ratios compared to D2 PL -Core synapses of males 699 (light blue) and females (light green), with no effect of sex. C. Representative baseline oEPSC 700 traces at -72, +8, and +48 mV at D1 and D2 PL -Core synapses in males and females. D. 701 Baseline I -V relationships representing AMPAR rectification curve at D1 and D2 PL -Core 702 synapses. E. Summary graph of AMPAR rectification index at +48 mV. At baseline, AMPAR 703 rectification was significantly lower at D2 PL -Core synapses of males (light blue) and females 704 (light green) vs. D1 PL -Core synapses of males (light red) and females (light orange). F. 705 Representative traces of paired oEPSCs at 50 ms ISI at baseline in D1 and D2 PL -Core 706 synapses. G. Summary graph of baseline paired -pulse ratios at 50, 100 and 200 ms ISI. At 50 707 and 100 ms ISI, paired-pulse ratio at D1 PL-Core synapses were higher than that at D2 PL-Core 708 synapses. D1 PL -Core synapses of males also had higher PPRs compared to females at all 709 ISIs. H. Representative traces of A/N ratios at IL -Shell synapses. NMDA current (black) was 710 measured at +40 mV and AMPA current (colored) was measured at +40 mV in the presence of 711 D-APV. I. Summary graphs of A/N ratios showing a trend toward increased A/N ratios was 712 observed at D1 vs. D 2 IL -Shell synapses, potentially driven by higher ratios in females . J. 713 Representative oEPSC traces at -72, +8, and +48 mV at D1 and D2 IL -Shell synapses. K. I-V 714 relationships representing AMPAR rectification curve at D1 and D2 IL -Shell synapses. L. 715 Summary graph of AMPAR rectification index at +48 mV at D1 vs D2 IL-Shell synapses showing 716 no differences in rectification. F. Representative traces of paired pulses evoked with 50 ms ISI 717 at D1 and D2 IL -Shell synapses. G. Summary graph of paired -pulse ratios at 50, 100 and 200 718 ms ISI. At 50 ms ISI, paired-pulse ratio at D1 IL-Shell synapses were higher than that at D2 IL -719 Shell synapses while at 200 ms ISI, they were marginally higher. D1+/D2+ - D1/D2-expressing 720 MSNs, M – males, F – females; Summary data are presented as Mean ± SEM. ^, ^^, ^^^ p < 721 0.05, 0.01, 0.001 for cell-type differences; #, ##p < 0.05, 0.01 for sex effects. 722 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 723 Figure 4. Divergent sex- and cell-type specific changes in synaptic transmission of NAc 724 core and shell MSNs are induced after remifentanil treatment. 725 A. Representative mEPSC traces of D1 NAc core MSNs. B. Summary graph of mEPSC 726 amplitude in D1 NAc core MSNs demonstrating that remifentanil treatment selectively increased 727 mEPSC amplitude in males (red bars) and females (orange bars) , while saline-treated females 728 (light orange bars) exhibited higher baseline amplitudes compared to saline males (light red 729 bars). C. Summary graph of mEPSC frequency in D1 NAc core MSNs showing that the 730 frequency of mEPSCs was increased after remifentanil treatment across both sexes . D. 731 Representative mEPSC traces of D2 NAc core MSNs. E, F. In D2 NAc core MSNs remifentanil 732 treatment selectively decreased amplitude (E) and frequency ( F) of mEPSCs in males (blue 733 bars) and females (green bars) compared to saline treated males (light blue) and females (light 734 green). G. Representative mEPSC traces of D1 NAc shell MSNs. H. Summary graph of mEPSC 735 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint amplitude in D1 NAc shell MSNs demonstrating that remifentanil treatment selectively increased 736 mEPSC amplitude in males and females. I. Summary graph of mEPSC frequency in D1 NAc 737 shell MSNs showing that the f requency of mEPSCs was increased after remifentanil treatment 738 in males but not females. J. Representative mEPSC traces of D2 NAc shell MSNs. K. Summary 739 graph in D2 NAc shell MSNs showing that remifentanil treatment did not affect the amplitude of 740 mEPSCs in both males and females. L. Summary graph in D2 NAc shell MSNs demonstrating 741 that remifentanil treatment decreased mEPSC frequency in males (blue bars) and females 742 (green bars) compared to saline treated males (light blue) and females (light green). Sal – Saline, 743 Rem – Remifentanil, M – males, F – females; Summary data are presented as Mean ± SEM. *, 744 **, ***p < 0.05, 0.01, 0.001 for treatment effects; ##p < 0.01 for sex effects. 745 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 746 Figure 5. Pre- and postsynaptic signaling mechanisms after remifentanil treatment differ 747 between D1 and D2 PL-Core synapses 748 A. Representative traces of AMPAR and NMDAR mediated currents at D1 PL-Core synapses in 749 saline (top) vs remifentanil treat ed (bottom) males (left) and females (right). NMDAR current 750 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint (black) was measured at +40 mV and AMPAR current (colored) was measured at +40 mV in the 751 presence of D -APV. B. Summary graph of A/N ratios shows that remifentanil significantly 752 reduced A/N ratios at D1 PL-Core synapses in both males and females compared to saline 753 controls. C. Representative oEPSC traces at -72, +8, and +48 mV at D1 PL-Core synapses in 754 saline and remifentanil treated males and females. D. I-V plots derived from oEPSC traces show 755 decreased rectification following remifentanil at D1 PL -Core synapses in both males and 756 females. E. Summary graph of AMPAR rectification index at +48 mV confirmed that remifentanil 757 significantly decreased the AMPAR rectification index at D1 PL -Core synapses across both 758 sexes. F. Representative paired oEPSC traces at 50 ms ISI at D1 PL-Core synapses in saline 759 (top) and remifentanil (bottom) treated males (left) and females (right) . G. Summary graph of 760 paired-pulse ratios at 50, 100 and 200 ms ISI in D1 PL -Core synapses demonstrated that 761 remifentanil marginally decreased the paired-pulse ratio. Specifically, paired -pulse ratio was 762 significantly decreased after remifentanil treatment at 50, and 200 ms ISI. Paired-pulse ratios 763 were significantly decreased after remifentanil treatment in males but not females at all ISIs 764 tested. H. Representative traces of AMPAR and NMDAR mediated currents at D2 PL -Core 765 synapses in saline (top) and remifentanil treat ed (bottom) males (left) and females (right) . 766 NMDAR current (black) was measured at +40 mV and AMPAR current (colored) was measured 767 at +40 mV in the presence of D -APV. I. Summary graph of A/N ratios shows that remifentanil 768 significantly increased A/N ratios at D2 PL-Core synapses in males but not females compared 769 to saline controls . J. Representative oEPSC traces at -72, +8, and +48 mV at D2 P L-Core 770 synapses in saline vs remifentanil treated males and females. K. I-V plots derived from oEPSC 771 traces show increased rectification following remifentanil treatment at D2 PL-Core synapses in 772 males but not females . L. Summary graph of AMPAR rectification index at +48 mV confirmed 773 that remifentanil significantly increased the AMPAR rectification index at D2 PL-Core synapses 774 of males. M. Representative paired oEPSC traces at a 50 ms ISI at D2 PL -Core synapses in 775 saline (top) vs. remifentanil (bottom) treated males (left) and females (right). G. Summary graph 776 of paired-pulse ratios at 50, 100 and 200 ms ISI in D2 PL -Core synapses demonstrated that 777 remifentanil significantly increased the paired-pulse ratios at all ISI tested. At 200 ms ISI paired-778 pulse ratio of males was significantly higher than females. Sal – Saline, Rem – Remifentanil, M 779 – males, F – females; Summary data are presented as Mean ± SEM. *, **, *** p < 0.05, 0.01, 780 0.001 for treatment effects; #, ##p < 0.05, 0.01 for sex effects. 781 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 782 Figure 6. Divergent sex-specific presynaptic signaling mechanisms are induced in D1 and 783 D2 IL-Shell synapses after remifentanil treatment. 784 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint A. Representative traces of AMPAR and NMDAR mediated currents at D1 IL-Shell synapses in 785 saline (top) vs remifentanil trea ted (bottom) males (left) and females (right). NMDAR current 786 (black) was measured at +40 mV and AMPAR current (colored) was measured at +40 mV in the 787 presence of D -APV. B. Summary graph of A/N ratios shows that remifentanil significantly 788 reduced A/N ratios at D1 IL-Shell synapses in females but not males compared to saline controls. 789 C. Representative oEPSC traces at -72, +8, and +48 mV at D1 IL-Shell synapses of saline vs 790 remifentanil treated males and females. D. I-V plots derived from oEPSC traces show that 791 remifentanil did not change AMPAR rectification at D1 IL -Shell synapses in both males and 792 females. E. However, the summary graph of AMPAR rectification index at +48 mV demonstrates 793 that remifentanil significantly decreased the AMPAR rectification index at D1 IL-Shell synapses 794 across both sexes . F. Representative paired o EPSC traces at a 50 ms ISI at D1 IL -Shell 795 synapses in saline (top) vs remi fentanil (bottom) treated males (left) and females (right) . G. 796 Summary graph of paired -pulse ratios at 50, 100 and 200 ms ISI in D1 IL -Shell synapses 797 demonstrated that remifentanil marginally decreased the paired-pulse ratio across sexes at all 798 ISIs tested. H. Representative traces of AMPAR and NMDAR mediated currents at D2 IL-Shell 799 synapses of saline (top) and remifentanil trea ted (bottom) males (left) and females (right) . 800 NMDAR current (black) was measured at +40 mV and AMPAR current (colored) was measured 801 at +40 mV in the presence of D-APV. I. Summary graph of A/N ratios shows that remifentanil did 802 not alter A/N ratios at D2 IL-Shell synapses. J. Representative oEPSC traces at -72, +8, and 803 +48 mV at D2 IL-Shell synapses of saline and remifentanil treat ed males and females. K. I-V 804 plots derived from oEPSC traces show no differences in rectification following remifentanil 805 treatment at D2 IL-Shell synapses. L. Summary graph of AMPAR rectification index at +48 mV 806 confirmed that remifentanil did not alter AMPAR rectification index at D2 IL-Shell synapses. M. 807 Representative paired oEPSC traces at a 50 ms ISI at D2 IL-Shell synapses in saline (top) and 808 remifentanil (bottom) treated males (left) and females (right). G. Summary graph of paired-pulse 809 ratios at 50, 100 and 200 ms ISI in D2 IL -Shell synapses demonstrated that remifentanil 810 treatment significantly increased the paired-pulse ratios at 50 ms ISI. At 100 and 200 ms ISI 811 paired-pulse ratio of males was significantly higher than females. Sal – Saline, Rem – 812 Remifentanil, M – males, F – females; Summary data are presented as Mean ± SEM. *p < 0.05 813 for treatment effects; #p < 0.05 for sex effects. 814 815 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint

816 Anderson, E. M., Gomez, D., Caccamise, A., McPhail, D. and Hearing, M. (2019) Chronic 817 unpredictable stress promotes cell -specific plasticity in prefrontal cortex D1 and D2 818 pyramidal neurons. Neurobiol Stress, 10, 100152. 10.1016/j.ynstr.2019.100152. Available 819 at: https://www.ncbi.nlm.nih.gov/pubmed/30937357. 820 821 Anderson, E. M., Engelhardt, A., Demis, S., Porath, E. and Hearing, M. C. (2021) Remifentanil 822 self-administration in mice promotes sex-specific prefrontal cortex dysfunction underlying 823 deficits in cognitive flexibility. Neuropsychopharmacology, 46(10), 1734–1745. 824 10.1038/s41386-021-01028-z. Available at: 825 https://www.ncbi.nlm.nih.gov/pubmed/34012018. 826 827 Becker, J. B. and Hu, M. (2008) Sex differences in drug abuse. Front Neuroendocrinol, 29(1), 828 36–47. 10.1016/j.yfrne.2007.07.003. Available at: 829 https://www.ncbi.nlm.nih.gov/pubmed/17904621. 830 831 Becker, J. B. and Chartoff, E. (2019) Sex differences in neural mechanisms mediating reward 832 and addiction. Neuropsychopharmacology, 44(1), 166–183. 10.1038/s41386-018-0125-833 6. Available at: https://www.ncbi.nlm.nih.gov/pubmed/29946108. 834 835 Berendse, G.-d. G., Groenewegen (1992) Topographical Organization and Relationship with 836 Ventral Striatal Compartments of Prefrontal Corticostriatal Projections in the Rat. Journal 837 of Comparative Neurology. 838 839 Bobadilla, A. C., Garcia -Keller, C., Heinsbroek, J. A., Scofield, M. D., Chareunsouk, V., 840 Monforton, C. and Kalivas, P. W. (2017) Accumbens Mechanisms for Cued Sucrose 841 Seeking. Neuropsychopharmacology, 42(12), 2377–2386. 10.1038/npp.2017.153. 842 Available at: https://www.ncbi.nlm.nih.gov/pubmed/28726801. 843 844 Bock, R., Shin, J. H., Kaplan, A. R., Dobi, A., Markey, E., Kramer, P. F., Gremel, C. M., 845 Christensen, C. H., Adrover, M. F. and Alvarez, V. A. (2013) Strengthening the accumbal 846 indirect pathway promotes resilience to compulsive cocaine use. Nat Neurosci, 16(5), 847 632–8. 10.1038/nn.3369. Available at: https://www.ncbi.nlm.nih.gov/pubmed/23542690. 848 849 Bossert, J. and Daniel, C. (2006) trans -cis Photoisomerization of the styrylpyridine Ligand in 850 [Re(CO)3(2,2'-bipyridine)(t-4-styrylpyridine)]+: role of the metal-to-ligand charge-transfer 851 excited states. Chemistry, 12(18), 4835–43. 10.1002/chem.200501082. Available at: 852 https://www.ncbi.nlm.nih.gov/pubmed/16642521. 853 854 Bossert, J. M., Stern, A. L., Theberge, F. R., Marchant, N. J., Wang, H. L., Morales, M. and 855 Shaham, Y. (2012) Role of projections from ventral medial prefrontal cortex to nucleus 856 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint accumbens shell in context-induced reinstatement of heroin seeking. J Neurosci, 32(14), 857 4982–91. 10.1523/JNEUROSCI.0005 -12.2012. Available at: 858 https://www.ncbi.nlm.nih.gov/pubmed/22492053. 859 860 Boxer, E. E., Kim, J., Dunn, B. and Aoto, J. (2023) Ventral Subiculum Inputs to Nucleus 861 Accumbens Medial Shell Preferentially Innervate D2R Medium Spiny Neurons and 862 Contain Calcium Permeable AMPARs. J Neurosci, 43(7), 1166–1177. 863 10.1523/JNEUROSCI.1907-22.2022. Available at: 864 https://www.ncbi.nlm.nih.gov/pubmed/36609456. 865 866 Britt, J. P., Benaliouad, F., McDevitt, R. A., Stuber, G. D., Wise, R. A. and Bonci, A. (2012) 867 Synaptic and behavioral profile of multiple glutamatergic inputs to the nucleus 868 accumbens. Neuron, 76(4), 790–803. 10.1016/j.neuron.2012.09.040. Available at: 869 https://www.ncbi.nlm.nih.gov/pubmed/23177963. 870 871 Brog, J. S., Salyapongse, A., Deutch, A. Y. and Zahm, D. S. (1993) The patterns of afferent 872 innervation of the core and shell in the "accumbens" part of the rat ventral striatum: 873 immunohistochemical detection of retrogradely transported fluoro -gold. J Comp Neurol, 874 338(2), 255–78. 10.1002/cne.903380209. Available at: 875 https://www.ncbi.nlm.nih.gov/pubmed/8308171. 876 877 Calipari, E. S., Bagot, R. C., Purushothaman, I., Davidson, T. J., Yorgason, J. T., Pena, C. J., 878 Walker, D. M., Pirpinias, S. T., Guise, K. G., Ramakrishnan, C., Deisseroth, K. and Nestler, 879 E. J. (2016) In vivo imaging identifies temporal signature of D1 and D2 medium spiny 880 neurons in cocaine reward. Proc Natl Acad Sci U S A, 113(10), 2726–31. 881 10.1073/pnas.1521238113. Available at: 882 https://www.ncbi.nlm.nih.gov/pubmed/26831103. 883 884 Cao, J., Dorris, D. M. and Meitzen, J. (2018) Electrophysiological properties of medium spiny 885 neurons in the nucleus accumbens core of prepubertal male and female Drd1a-tdTomato 886 line 6 BAC transgenic mice. J Neurophysiol, 120(4), 1712–1727. 10.1152/jn.00257.2018. 887 Available at: https://www.ncbi.nlm.nih.gov/pubmed/29975170. 888 889 Chioma, V. C., Kruyer, A., Bobadilla, A. C., Angelis, A., Ellison, Z., Hodebourg, R., Scofield, M. 890 D. and Kalivas, P. W. (2021) Heroin Seeking and Extinction From Seeking Activate Matrix 891 Metalloproteinases at Synapses on Distinct Subpopulations of Accumbens Cells. Biol 892 Psychiatry, 89(10), 947–958. 10.1016/j.biopsych.2020.12.004. Available at: 893 https://www.ncbi.nlm.nih.gov/pubmed/33579535. 894 895 Dittman, J. S., Kreitzer, A. C. and Regehr, W. G. (2000) Interplay between facilitation, 896 depression, and residual calcium at three presynaptic terminals. J Neurosci, 20(4), 1374–897 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint 85. 10.1523/JNEUROSCI.20 -04-01374.2000. Available at: 898 https://www.ncbi.nlm.nih.gov/pubmed/10662828. 899 900 Everitt, B. J. and Robbins, T. W. (2005) Neural systems of reinforcement for drug addiction: from 901 actions to habits to compulsion. Nat Neurosci, 8(11), 1481–9. 10.1038/nn1579. Available 902 at: https://www.ncbi.nlm.nih.gov/pubmed/16251991. 903 904 Flagel, S. B., Watson, S. J., Robinson, T. E. and Akil, H. (2007) Individual differences in the 905 propensity to approach signals vs goals promote different adaptations in the dopamine 906 system of rats. Psychopharmacology (Berl), 191(3), 599–607. 10.1007/s00213 -006-907 0535-8. Available at: https://www.ncbi.nlm.nih.gov/pubmed/16972103. 908 909 Forlano, P. M. and Woolley, C. S. (2010) Quantitative analysis of pre - and postsynaptic sex 910 differences in the nucleus accumbens. J Comp Neurol, 518(8), 1330–48. 911 10.1002/cne.22279. Available at: https://www.ncbi.nlm.nih.gov/pubmed/20151363. 912 913 Gerfen, C. R. and Surmeier, D. J. (2011) Modulation of striatal projection systems by dopamine. 914 Annu Rev Neurosci, 34, 441–66. 10.1146/annurev-neuro-061010-113641. Available at: 915 https://www.ncbi.nlm.nih.gov/pubmed/21469956. 916 917 Glass, M. J., Lane, D. A., Colago, E. E., Chan, J., Schlussman, S. D., Zhou, Y., Kreek, M. J. and 918 Pickel, V. M. (2008) Chronic administration of morphine is associated with a decrease in 919 surface AMPA GluR1 receptor subunit in dopamine D1 receptor expressing neurons in 920 the shell and non -D1 receptor expressing neurons in the core of the rat nucleus 921 accumbens. Exp Neurol, 210(2), 750–61. 10.1016/j.expneurol.2008.01.012. Available at: 922 https://www.ncbi.nlm.nih.gov/pubmed/18294632. 923 924 Graziane, N. M., Sun, S., Wright, W. J., Jang, D., Liu, Z., Huang, Y. H., Nestler, E. J., Wang, Y. 925 T., Schluter, O. M. and Dong, Y. (2016) Opposing mechanisms mediate morphine - and 926 cocaine-induced generation of silent synapses. Nat Neurosci, 19(7), 915–25. 927 10.1038/nn.4313. Available at: https://www.ncbi.nlm.nih.gov/pubmed/27239940. 928 929 Grueter, B. A., Brasnjo, G. and Malenka, R. C. (2010) Postsynaptic TRPV1 triggers cell type -930 specific long-term depression in the nucleus accumbens. Nat Neurosci, 13(12), 1519–25. 931 10.1038/nn.2685. Available at: https://www.ncbi.nlm.nih.gov/pubmed/21076424. 932 933 Han, E. B. and Stevens, C. F. (2009) Development regulates a switch between post - and 934 presynaptic strengthening in response to activity deprivation. Proc Natl Acad Sci U S A, 935 106(26), 10817–22. 10.1073/pnas.0903603106. Available at: 936 https://www.ncbi.nlm.nih.gov/pubmed/19509338. 937 938 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Hearing, M., Graziane, N., Dong, Y. and Thomas, M. J. (2018) Opioid and Psychostimulant 939 Plasticity: Targeting Overlap in Nucleus Accumbens Glutamate Signaling. Trends 940 Pharmacol Sci, 39(3), 276–294. 10.1016/j.tips.2017.12.004. Available at: 941 https://www.ncbi.nlm.nih.gov/pubmed/29338873. 942 943 Hearing, M. (2019) Prefrontal -accumbens opioid plasticity: Implications for relapse and 944 dependence. Pharmacol Res, 139, 158–165. 10.1016/j.phrs.2018.11.012. Available at: 945 https://www.ncbi.nlm.nih.gov/pubmed/30465850. 946 947 Hearing, M. C., Jedynak, J., Ebner, S. R., Ingebretson, A., Asp, A. J., Fischer, R. A., Schmidt, C., 948 Larson, E. B. and Thomas, M. J. (2016) Reversal of morphine -induced cell-type-specific 949 synaptic plasticity in the nucleus accumbens shell blocks reinstatement. Proc Natl Acad 950 Sci U S A, 113(3), 757–62. 10.1073/pnas.1519248113. Available at: 951 https://www.ncbi.nlm.nih.gov/pubmed/26739562. 952 953 Heidbreder, C. A. and Groenewegen, H. J. (2003) The medial prefrontal cortex in the rat: 954 evidence for a dorso -ventral distinction based upon functional and anatomical 955 characteristics. Neurosci Biobehav Rev, 27(6), 555–79. 956 10.1016/j.neubiorev.2003.09.003. Available at: 957 https://www.ncbi.nlm.nih.gov/pubmed/14599436. 958 959 Heimer, L., Alheid, G. F., de Olmos, J. S., Groenewegen, H. J., Haber, S. N., Harlan, R. E. and 960 Zahm, D. S. (1997) The accumbens: beyond the core-shell dichotomy. J Neuropsychiatry 961 Clin Neurosci, 9(3), 354–81. 10.1176/jnp.9.3.354. Available at: 962 https://www.ncbi.nlm.nih.gov/pubmed/9276840. 963 964 Heimer, L. (2003) The legacy of the silver methods and the new anatomy of the basal forebrain: 965 implications for neuropsychiatry and drug abuse. Scand J Psychol, 44(3), 189–201. 966 10.1111/1467-9450.00336. Available at: 967 https://www.ncbi.nlm.nih.gov/pubmed/12914582. 968 969 Hikida, T., Morita, M. and Macpherson, T. (2016) Neural mechanisms of the nucleus accumbens 970 circuit in reward and aversive learning. Neurosci Res, 108, 1–5. 971 10.1016/j.neures.2016.01.004. Available at: 972 https://www.ncbi.nlm.nih.gov/pubmed/26827817. 973 974 Hoover, W. B. and Vertes, R. P. (2007) Anatomical analysis of afferent projections to the medial 975 prefrontal cortex in the rat. Brain Struct Funct, 212(2), 149–79. 10.1007/s00429 -007-976 0150-4. Available at: https://www.ncbi.nlm.nih.gov/pubmed/17717690. 977 978 Hutcheson, D. M., Parkinson, J. A., Robbins, T. W. and Everitt, B. J. (2001) The effects of nucleus 979 accumbens core and shell lesions on intravenous heroin self -administration and the 980 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint acquisition of drug -seeking behaviour under a second -order schedule of heroin 981 reinforcement. Psychopharmacology (Berl), 153(4), 464–72. 10.1007/s002130000635. 982 Available at: https://www.ncbi.nlm.nih.gov/pubmed/11243494. 983 984 James, A. S., Chen, J. Y., Cepeda, C., Mittal, N., Jentsch, J. D., Levine, M. S., Evans, C. J. and 985 Walwyn, W. (2013) Opioid self -administration results in cell -type specific adaptations of 986 striatal medium spiny neurons. Behav Brain Res, 256, 279–83. 987 10.1016/j.bbr.2013.08.009. Available at: 988 https://www.ncbi.nlm.nih.gov/pubmed/23968589. 989 990 Johnson, A. R., Thibeault, K. C., Lopez, A. J., Peck, E. G., Sands, L. P., Sanders, C. M., Kutlu, 991 M. G. and Calipari, E. S. (2019) Cues play a critical role in estrous cycle -dependent 992 enhancement of cocaine reinforcement. Neuropsychopharmacology, 44(7), 1189–1197. 993 10.1038/s41386-019-0320-0. Available at: 994 https://www.ncbi.nlm.nih.gov/pubmed/30728447. 995 996 Kalivas, P. W. (2009) The glutamate homeostasis hypothesis of addiction. Nat Rev Neurosci, 997 10(8), 561–72. 10.1038/nrn2515. Available at: 998 https://www.ncbi.nlm.nih.gov/pubmed/19571793. 999 1000 Kerchner, G. A. and Nicoll, R. A. (2008) Silent synapses and the emergence of a postsynaptic 1001 mechanism for LTP. Nat Rev Neurosci, 9(11), 813–25. 10.1038/nrn2501. Available at: 1002 https://www.ncbi.nlm.nih.gov/pubmed/18854855. 1003 1004 Klawonn, A. M. and Malenka, R. C. (2018) Nucleus Accumbens Modulation in Reward and 1005 Aversion. Cold Spring Harb Symp Quant Biol, 83, 119–129. 1006 10.1101/sqb.2018.83.037457. Available at: 1007 https://www.ncbi.nlm.nih.gov/pubmed/30674650. 1008 1009 Kniffin, A. R. and Briand, L. A. (2024) Sex differences in glutamate transmission and plasticity in 1010 reward related regions. Front Behav Neurosci, 18, 1455478. 1011 10.3389/fnbeh.2024.1455478. Available at: 1012 https://www.ncbi.nlm.nih.gov/pubmed/39359325. 1013 1014 Kokane, S. S. and Perrotti, L. I. (2020) Sex Differences and the Role of Estradiol in Mesolimbic 1015 Reward Circuits and Vulnerability to Cocaine and Opiate Addiction. Front Behav 1016 Neurosci, 14, 74. 10.3389/fnbeh.2020.00074. Available at: 1017 https://www.ncbi.nlm.nih.gov/pubmed/32508605. 1018 1019 Kokane, S. S., Cole, R. D., Bordieanu, B., Ray, C. M., Haque, I. A., Otis, J. M. and McGinty, J. 1020 F. (2023) Increased Excitability and Synaptic Plasticity of Drd1 - and Drd2 -Expressing 1021 Prelimbic Neurons Projecting to Nucleus Accumbens after Heroin Abstinence Are 1022 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Reversed by Cue-Induced Relapse and Protein Kinase A Inhibition. J Neurosci, 43(22), 1023 4019–4032. 10.1523/JNEUROSCI.0108 -23.2023. Available at: 1024 https://www.ncbi.nlm.nih.gov/pubmed/37094933. 1025 1026 Koob, G. F., Ahmed, S. H., Boutrel, B., Chen, S. A., Kenny, P. J., Markou, A., O'Dell, L. E., 1027 Parsons, L. H. and Sanna, P. P. (2004) Neurobiological mechanisms in the transition from 1028 drug use to drug dependence. Neurosci Biobehav Rev, 27(8), 739–49. 1029 10.1016/j.neubiorev.2003.11.007. Available at: 1030 https://www.ncbi.nlm.nih.gov/pubmed/15019424. 1031 1032 Koob, G. F. and Volkow, N. D. (2010) Neurocircuitry of addiction. Neuropsychopharmacology, 1033 35(1), 217–38. 10.1038/npp.2009.110. Available at: 1034 https://www.ncbi.nlm.nih.gov/pubmed/19710631. 1035 1036 Koob, G. F. and Volkow, N. D. (2016) Neurobiology of addiction: a neurocircuitry analysis. Lancet 1037 Psychiatry, 3(8), 760–773. 10.1016/S2215 -0366(16)00104-8. Available at: 1038 https://www.ncbi.nlm.nih.gov/pubmed/27475769. 1039 1040 Kravitz, A. V., Tye, L. D. and Kreitzer, A. C. (2012) Distinct roles for direct and indirect pathway 1041 striatal neurons in reinforcement. Nat Neurosci, 15(6), 816–8. 10.1038/nn.3100. Available 1042 at: https://www.ncbi.nlm.nih.gov/pubmed/22544310. 1043 1044 Kreitzer, A. C. and Malenka, R. C. (2007) Endocannabinoid-mediated rescue of striatal LTD and 1045 motor deficits in Parkinson's disease models. Nature, 445(7128), 643–7. 1046 10.1038/nature05506. Available at: https://www.ncbi.nlm.nih.gov/pubmed/17287809. 1047 1048 Kruyer, A., Scofield, M. D., Wood, D., Reissner, K. J. and Kalivas, P. W. (2019) Heroin Cue -1049 Evoked Astrocytic Structural Plasticity at Nucleus Accumbens Synapses Inhibits Heroin 1050 Seeking. Biol Psychiatry, 86(11), 811–819. 10.1016/j.biopsych.2019.06.026. Available at: 1051 https://www.ncbi.nlm.nih.gov/pubmed/31495448. 1052 1053 Lacy, R. T., Strickland, J. C., Feinstein, M. A., Robinson, A. M. and Smith, M. A. (2016) The 1054 effects of sex, estrous cycle, and social contact on cocaine and heroin self-administration 1055 in rats. Psychopharmacology (Berl), 233(17), 3201–10. 10.1007/s00213 -016-4368-9. 1056 Available at: https://www.ncbi.nlm.nih.gov/pubmed/27370020. 1057 1058 Lacy, R. T., Austin, B. P. and Strickland, J. C. (2020) The influence of sex and estrous cyclicity 1059 on cocaine and remifentanil demand in rats. Addict Biol, 25(1), e12716. 1060 10.1111/adb.12716. Available at: https://www.ncbi.nlm.nih.gov/pubmed/30779409. 1061 1062 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint LaLumiere, R. T. and Kalivas, P. W. (2008) Glutamate release in the nucleus accumbens core is 1063 necessary for heroin seeking. J Neurosci, 28(12), 3170–7. 10.1523/JNEUROSCI.5129-1064 07.2008. Available at: https://www.ncbi.nlm.nih.gov/pubmed/18354020. 1065 1066 Lefevre, E. M., Gauthier, E. A., Bystrom, L. L., Scheunemann, J. and Rothwell, P. E. (2023) 1067 Differential Patterns of Synaptic Plasticity in the Nucleus Accumbens Caused by 1068 Continuous and Interrupted Morphine Exposure. J Neurosci, 43(2), 308–318. 1069 10.1523/JNEUROSCI.0595-22.2022. Available at: 1070 https://www.ncbi.nlm.nih.gov/pubmed/36396404. 1071 1072 Lobo, M. K., Covington, H. E., 3rd, Chaudhury, D., Friedman, A. K., Sun, H., Damez-Werno, D., 1073 Dietz, D. M., Zaman, S., Koo, J. W., Kennedy, P. J., Mouzon, E., Mogri, M., Neve, R. L., 1074 Deisseroth, K., Han, M. H. and Nestler, E. J. (2010) Cell type -specific loss of BDNF 1075 signaling mimics optogenetic control of cocaine reward. Science, 330(6002), 385–90. 1076 10.1126/science.1188472. Available at: https://www.ncbi.nlm.nih.gov/pubmed/20947769. 1077 1078 MacAskill, A. F., Cassel, J. M. and Carter, A. G. (2014) Cocaine exposure reorganizes cell type- 1079 and input-specific connectivity in the nucleus accumbens. Nat Neurosci, 17(9), 1198–207. 1080 10.1038/nn.3783. Available at: https://www.ncbi.nlm.nih.gov/pubmed/25108911. 1081 1082 Madayag, A. C., Gomez, D., Anderson, E. M., Ingebretson, A. E., Thomas, M. J. and Hearing, 1083 M. C. (2019) Cell -type and region -specific nucleus accumbens AMPAR plasticity 1084 associated with morphine reward, reinstatement, and spontaneous withdrawal. Brain 1085 Struct Funct, 224(7), 2311–2324. 10.1007/s00429 -019-01903-y. Available at: 1086 https://www.ncbi.nlm.nih.gov/pubmed/31201496. 1087 1088 Marchant, N. J., Kaganovsky, K., Shaham, Y. and Bossert, J. M. (2016) Role of corticostriatal 1089 circuits in context-induced reinstatement of drug seeking. Brain Res, 1628(Pt A), 219–32. 1090 10.1016/j.brainres.2014.09.004. Available at: 1091 https://www.ncbi.nlm.nih.gov/pubmed/25199590. 1092 1093 McFarland, K., Lapish, C. C. and Kalivas, P. W. (2003) Prefrontal glutamate release into the core 1094 of the nucleus accumbens mediates cocaine -induced reinstatement of drug -seeking 1095 behavior. J Neurosci, 23(8), 3531–7. 10.1523/JNEUROSCI.23-08-03531.2003. Available 1096 at: https://www.ncbi.nlm.nih.gov/pubmed/12716962. 1097 1098 Nelson, A. B., Hang, G. B., Grueter, B. A., Pascoli, V., Luscher, C., Malenka, R. C. and Kreitzer, 1099 A. C. (2012) A comparison of striatal -dependent behaviors in wild -type and hemizygous 1100 Drd1a and Drd2 BAC transgenic mice. J Neurosci, 32(27), 9119–23. 1101 10.1523/JNEUROSCI.0224-12.2012. Available at: 1102 https://www.ncbi.nlm.nih.gov/pubmed/22764221. 1103 1104 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Nicolas, C., Zlebnik, N. E., Farokhnia, M., Leggio, L., Ikemoto, S. and Shaham, Y. (2022) Sex 1105 Differences in Opioid and Psychostimulant Craving and Relapse: A Critical Review. 1106 Pharmacol Rev, 74(1), 119–140. 10.1124/pharmrev.121.000367. Available at: 1107 https://www.ncbi.nlm.nih.gov/pubmed/34987089. 1108 1109 Olsen, C. M. and Winder, D. G. (2009) Operant sensation seeking engages similar neural 1110 substrates to operant drug seeking in C57 mice. Neuropsychopharmacology, 34(7), 1111 1685–94. 10.1038/npp.2008.226. Available at: 1112 https://www.ncbi.nlm.nih.gov/pubmed/19145223. 1113 1114 Pascoli, V., Turiault, M. and Luscher, C. (2011) Reversal of cocaine-evoked synaptic potentiation 1115 resets drug-induced adaptive behaviour. Nature, 481(7379), 71–5. 10.1038/nature10709. 1116 Available at: https://www.ncbi.nlm.nih.gov/pubmed/22158102. 1117 1118 Pascoli, V., Terrier, J., Espallergues, J., Valjent, E., O'Connor, E. C. and Luscher, C. (2014) 1119 Contrasting forms of cocaine -evoked plasticity control components of relapse. Nature, 1120 509(7501), 459–64. 10.1038/nature13257. Available at: 1121 https://www.ncbi.nlm.nih.gov/pubmed/24848058. 1122 1123 Pecina, S. and Berridge, K. C. (2005) Hedonic hot spot in nucleus accumbens shell: where do 1124 mu-opioids cause increased hedonic impact of sweetness? J Neurosci, 25(50), 11777–1125 86. 10.1523/JNEUROSCI.2329-05.2005. Available at: 1126 https://www.ncbi.nlm.nih.gov/pubmed/16354936. 1127 1128 Peters, J., Pattij, T. and De Vries, T. J. (2013) Targeting cocaine versus heroin memories: 1129 divergent roles within ventromedial prefrontal cortex. Trends Pharmacol Sci, 34(12), 689–1130 95. 10.1016/j.tips.2013.10.004. Available at: 1131 https://www.ncbi.nlm.nih.gov/pubmed/24182624. 1132 1133 Pontieri, F. E., Tanda, G. and Di Chiara, G. (1995) Intravenous cocaine, morphine, and 1134 amphetamine preferentially increase extracellular dopamine in the "shell" as compared 1135 with the "core" of the rat nucleus accumbens. Proc Natl Acad Sci U S A, 92(26), 12304–1136 8. 10.1073/pnas.92.26.12304. Available at: 1137 https://www.ncbi.nlm.nih.gov/pubmed/8618890. 1138 1139 Proano, S. B., Morris, H. J., Kunz, L. M., Dorris, D. M. and Meitzen, J. (2018) Estrous cycle -1140 induced sex differences in medium spiny neuron excitatory synaptic transmission and 1141 intrinsic excitability in adult rat nucleus accumbens core. J Neurophysiol, 120(3), 1356–1142 1373. 10.1152/jn.00263.2018. Available at: 1143 https://www.ncbi.nlm.nih.gov/pubmed/29947588. 1144 1145 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint Rogers, J. L., Ghee, S. and See, R. E. (2008) The neural circuitry underlying reinstatement of 1146 heroin-seeking behavior in an animal model of relapse. Neuroscience, 151(2), 579–88. 1147 10.1016/j.neuroscience.2007.10.012. Available at: 1148 https://www.ncbi.nlm.nih.gov/pubmed/18061358. 1149 1150 Scofield, M. D., Heinsbroek, J. A., Gipson, C. D., Kupchik, Y. M., Spencer, S., Smith, A. C., 1151 Roberts-Wolfe, D. and Kalivas, P. W. (2016) The Nucleus Accumbens: Mechanisms of 1152 Addiction across Drug Classes Reflect the Importance of Glutamate Homeostasis. 1153 Pharmacol Rev, 68(3), 816–71. 10.1124/pr.116.012484. Available at: 1154 https://www.ncbi.nlm.nih.gov/pubmed/27363441. 1155 1156 Sesack, S. R., Deutch, A. Y., Roth, R. H. and Bunney, B. S. (1989) Topographical organization 1157 of the efferent projections of the medial prefrontal cortex in the rat: an anterograde tract -1158 tracing study with Phaseolus vulgaris leucoagglutinin. J Comp Neurol, 290(2), 213–42. 1159 10.1002/cne.902900205. Available at: https://www.ncbi.nlm.nih.gov/pubmed/2592611. 1160 1161 Sesack, S. R. and Grace, A. A. (2010) Cortico -Basal Ganglia reward network: microcircuitry. 1162 Neuropsychopharmacology, 35(1), 27–47. 10.1038/npp.2009.93. Available at: 1163 https://www.ncbi.nlm.nih.gov/pubmed/19675534. 1164 1165 Shen, H., Moussawi, K., Zhou, W., Toda, S. and Kalivas, P. W. (2011) Heroin relapse requires 1166 long-term potentiation -like plasticity mediated by NMDA2b -containing receptors. Proc 1167 Natl Acad Sci U S A, 108(48), 19407–12. 10.1073/pnas.1112052108. Available at: 1168 https://www.ncbi.nlm.nih.gov/pubmed/22084102. 1169 1170 Soares-Cunha, C., Coimbra, B., Domingues, A. V., Vasconcelos, N., Sousa, N. and Rodrigues, 1171 A. J. (2018) Nucleus Accumbens Microcircuit Underlying D2 -MSN-Driven Increase in 1172 Motivation. eNeuro, 5(2). 10.1523/ENEURO.0386 -18.2018. Available at: 1173 https://www.ncbi.nlm.nih.gov/pubmed/29780881. 1174 1175 Soares-Cunha, C., de Vasconcelos, N. A. P., Coimbra, B., Domingues, A. V., Silva, J. M., 1176 Loureiro-Campos, E., Gaspar, R., Sotiropoulos, I., Sousa, N. and Rodrigues, A. J. (2020) 1177 Nucleus accumbens medium spiny neurons subtypes signal both reward and aversion. 1178 Mol Psychiatry, 25(12), 3241–3255. 10.1038/s41380 -019-0484-3. Available at: 1179 https://www.ncbi.nlm.nih.gov/pubmed/31462765. 1180 1181 Van den Oever, M. C., Spijker, S., Smit, A. B. and De Vries, T. J. (2010) Prefrontal cortex plasticity 1182 mechanisms in drug seeking and relapse. Neurosci Biobehav Rev, 35(2), 276–84. 1183 10.1016/j.neubiorev.2009.11.016. Available at: 1184 https://www.ncbi.nlm.nih.gov/pubmed/19932711. 1185 1186 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint van Zessen, R., Li, Y., Marion-Poll, L., Hulo, N., Flakowski, J. and Luscher, C. (2021) Dynamic 1187 dichotomy of accumbal population activity underlies cocaine sensitization. Elife, 10. 1188 10.7554/eLife.66048. Available at: https://www.ncbi.nlm.nih.gov/pubmed/34608866. 1189 1190 Vertes, R. P. (2004) Differential projections of the infralimbic and prelimbic cortex in the rat. 1191 Synapse, 51(1), 32–58. 10.1002/syn.10279. Available at: 1192 https://www.ncbi.nlm.nih.gov/pubmed/14579424. 1193 1194 Wall, N. R., De La Parra, M., Callaway, E. M. and Kreitzer, A. C. (2013) Differential innervation 1195 of direct - and indirect -pathway striatal projection neurons. Neuron, 79(2), 347–60. 1196 10.1016/j.neuron.2013.05.014. Available at: 1197 https://www.ncbi.nlm.nih.gov/pubmed/23810541. 1198 1199 Willett, J. A., Will, T., Hauser, C. A., Dorris, D. M., Cao, J. and Meitzen, J. (2016) No Evidence 1200 for Sex Differences in the Electrophysiological Properties and Excitatory Synaptic Input 1201 onto Nucleus Accumbens Shell Medium Spiny Neurons. eNeuro, 3(1). 1202 10.1523/ENEURO.0147-15.2016. Available at: 1203 https://www.ncbi.nlm.nih.gov/pubmed/27022621. 1204 1205 Wissman, A. M., McCollum, A. F., Huang, G. Z., Nikrodhanond, A. A. and Woolley, C. S. (2011) 1206 Sex differences and effects of cocaine on excitatory synapses in the nucleus accumbens. 1207 Neuropharmacology, 61(1-2), 217–27. 10.1016/j.neuropharm.2011.04.002. Available at: 1208 https://www.ncbi.nlm.nih.gov/pubmed/21510962. 1209 1210 Zahm, D. S. and Brog, J. S. (1992) On the significance of subterritories in the "accumbens" part 1211 of the rat ventral striatum. Neuroscience, 50(4), 751–67. 10.1016/0306-4522(92)90202-1212 d. Available at: https://www.ncbi.nlm.nih.gov/pubmed/1448200. 1213 1214 Zhu, Y., Wienecke, C. F., Nachtrab, G. and Chen, X. (2016) A thalamic input to the nucleus 1215 accumbens mediates opiate dependence. Nature, 530(7589), 219–22. 1216 10.1038/nature16954. Available at: https://www.ncbi.nlm.nih.gov/pubmed/26840481. 1217 1218 1219 .CC-BY-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 24, 2026. ; https://doi.org/10.64898/2026.03.21.713428doi: bioRxiv preprint