Brain-wide mapping reveals temporal and sexually dimorphic opioid actions

While the molecular and cellular effects of opioids have been extensively studied, the precise mechanisms by which these drugs target specific brain regions over Ɵme remain unclear. Similarly, despite well-documented sex differences in opioid responses, the anatomical basis for this sexual dimorphism is not well characterized. To address these ques Ɵons, we developed an automated, scalable, and unbiased approach for whole- brain anatomical mapping of the neuronal ac Ɵvity marker c-Fos in response to acute morphine exposure. Using ribbon scanning confocal microscopy, we imaged whole cleared brains from male and female wild-type mice at 1 hour and 4 hours post-morphine administraƟon. Our whole-brain analysis of c-Fos expression revealed disƟnct paƩ erns of morphine-induced regional brain acƟvaƟon across Ɵme and sex. Notably, we observed a greater number of structures with significant acƟvity differences at 4 hours compared to 1 ho ur. In male mice, significant changes were primarily localized to regions within the dopamine system, whereas in female mice, they were concentrated in corƟcal regions. By combining high-throughput imaging with whole-brain expression analysis, parƟcularly in the context of opioid acƟons, our approach provides a more comprehensive understanding of how drugs of abuse affect the brain. IntroducƟ on Opioids are some of the most used and abused substances today, playing a central role in the treatment of pain. However, these drugs can also induce tolerance and dependence, contribu Ɵng significantly to their high potenƟal for abuse 1. As a result, opioid abuse prevalence has skyrocketed in the United States and globally 2–4. Current pharmacological treatments for opioid use disorder (OUD) are effecƟve for miƟgaƟng cravings, although ~90% of pa Ɵents relapse within several months of treatment cessaƟon5. Thus, to idenƟfy novel treatments for OUD and intervenƟons capable of slowing the opioid epidemic, we must elucidate the neurobiology of opioid addicƟon and its relaƟonships to factors that contribute to craving and relapse vulnerability IniƟal, acute exposure to opioids may lead to rapid adapƟve changes in the brain that reflect the early mechanisms of opioid dependence and risk of subsequent addicƟon. Indeed, treaƟng humans or rodents with opioid antagonists, such as naltrexone, elicits signs of withdrawal from opioids6. Changes in neuronal ac Ɵvity across various populaƟons of neuron s occur following acute administraƟon of opioids, likely underlying the adap Ɵve changes in the brain that contribute to tolerance and withdrawal7–10. Revealing where these changes in ac Ɵvity occur and their rela Ɵonship to drug -related behaviors is criƟcal for understanding the fundamental acƟons of drugs of abuse, including opioids , and for providing poten Ɵal neurobiological targets for therapeuƟc interven Ɵon. Consequently, the neuroscience field has been grappling .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 2/41 with how to effecƟvely invesƟgate acƟvity across mul Ɵple scales within the same brain, from the single cell to ensembles to enƟre brain regions, and how drugs modulate these changes in acƟvity. While acknowledging the biological importance of neuronal heterogeneity to addic Ɵon and the need for scalability, most current approaches are sƟll limited by their ability to resolve disƟnct changes in acƟvity in response to drugs of abuse at the single-cell level within larger populaƟons across the whole brain. For example, immunolabeling of neuronal acƟvity markers across various brain regions has been valuable for revealing new neurobiological substrates of complex drug-related behaviors7,9,11,12. Yet, because of various constraints of throughput and resolu Ɵon, earlier work has typically focused on invesƟgaƟng discrete regions in areas previously implicated in addicƟon. As a result, these prior limita Ɵons have created a bo Ʃ leneck for the development of a more integrated understanding of the biological underpinnings of addicƟon at a whole brain level. The ability to widely survey the whole brain at sufficient resoluƟon permits us to move beyond single brain regions. Rather, we can commence examining current opioid acƟons across mulƟple regions – a key next step in the genera Ɵon of a holis Ɵc, more integrated understanding of drug acƟons. We have developed a workflow to address these challenges which offer the capability to readily move across scales throughout whole intact brains in three dimensions (3D). Rather than being limited to a single brain region or smaller subsets of neurons, newer whole brain imaging and analysis methods provide the opportunity to probe drug- induced effects in an unbiased, discovery-based manner . Herein we used Ɵssue clearing 13 and high-speed ribbon scanning confocal microscopy 14 (RSCM) to collect high- resoluƟon imagery of whole brains and applied a novel automated high- performance compu Ɵng workflow that detects and maps cells within the brain. The development of this integrated workflow allowed us to digitally reconstruct en Ɵre adult mouse brains at sub-cellular resoluƟon and therefore to comprehensively map populaƟons of acƟvated neurons across the enƟre brain in dozens of animals. This automated approach to data analysis allows us to tackle experimental ques Ɵons like Ɵme-, sex-, and brain region-dependent differences when examining opioid acƟons. We u Ɵlized our workflow to comprehensively characterize the global anatomical expression paƩ erns of the immediate early gene (IEG) c-Fos, a classic neuronal acƟvity marker, in response to the administraƟon of the prototypical opioid, morphine. Our data demonstrated broad pa Ʃ erns of increased ac Ɵvity within the brains of morphine-exposed wildtype mice . The pa Ʃ erns of morphine-induced ac Ɵvity demonstrated disƟnct differences in neuronal ac Ɵvity across Ɵme of morphine exposure as well as across male and female animals. Specifically, we find that the mesolimbic dopamine system is ac Ɵvated more strongly in male mice, whereas cor Ɵcal regions are ac Ɵvated more strongly in female mice. In summary, our imaging and analysis approaches open the avenue for answering long- standing ques Ɵons in opioid acƟons, including resolving cellular heterogeneity within larger populaƟons and mapping their changes in response to drug exposure. Following peer review, a t the Ɵme of publicaƟon, and in line with FAIR 15,16 principles, we will make our tools openly available by deposiƟng a comprehensive set of code, microscopy data and examples into public repositories so that the community can reproduce our work or apply our methods to novel quesƟons.

Development of a whole brain imaging and analysis workflow. To inves Ɵgate the impact of different dura Ɵons of opioid exposure on the acƟviƟes of cell subpopula Ɵons throughout the whole adult mouse brain, we administered morphine to male and female wild-type mice versus a vehicle control (Fig. 1a). 1h or 4h following exposure, brains were collected and subsequently fluorescently stained for c-Fos, followed by opƟcal clearing. We then imaged the cleared and c-Fos-labeled brains via RSCM. AŌ er imaging, data from each brain was processed through an automated analysis pipeline (Fig. 1b) which consisted of parallel steps for down-scaling the data for alignment to the Allen Mouse Brain Common Coordinate Framework version 3 (CCFv3 17) and detecƟon of c-Fos-posiƟve cells throughout the brain. We used a combinaƟon of convenƟonal and deep-learning spot detecƟon approaches followed by clustering duplicate detecƟons and neural network-based discriminaƟon of true cells and arƟfacts. All aspects of the computa Ɵonal workflow were completed semi-autonomously on local high-performance compuƟng resources managed by the Simple Linux U Ɵlity for Resource Management (SLURM 18). Following RSCM, data was automaƟcally queued for s Ɵtching and assembly into 3D datasets. A Ō er manual verifica Ɵon of each 3D .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 3/41 dataset, spot detec Ɵon, alignment and classifica Ɵon were queued as a single job, outpuƫ ng intermediate processing products like the alignment deforma Ɵon field and finally the spot tables which included CCFv3 mapped structures. Where possible, we reused already available open-source toolkits, developing novel means to knit the soŌ ware together and produce an automated workflow requiring minimal intervenƟon. Following an iniƟal pass, t he results of each alignment and cell classifica Ɵon were manually reviewed and when appropriate, reprocessed with modified parameters. Although we used SLURM to manage this process, the workflow can be run on a single mid- to high-end workstaƟon w ith a NVIDIA GPU making it accessible to most research labs. The p erformance of automated spot detec Ɵon and classificaƟon was measured against a ground truth data sample collected for each brain. Acceptance criterion was f1-score >80%. For the brains having a lower f1-score, the base classificaƟon model was fine -tuned to reach the 80% threshold. For each brain, the coordinates of c-Fos-posiƟve cells were transformed into CCFv3 space and exported as CSV files that would be the basis for all subsequent data mining. During analysis, we took an unbiased approach by inspecƟng morphine’s effects on every brain structure at every depth level in the CCFv3 structure tree. This included calculaƟng the total numbers and densi Ɵes of c -Fos- posiƟve cells for each respec Ɵve brain structure . We also calculated mean cell densi Ɵes for each structure, for “all Figure 1. Experimental design. (a) 10-week-old male and female mice were injected with morphine or saline. 1h or 4h post injecƟon, their brains were harvested, stained for c-Fos and rendered opƟ cally transparent. Cleared brains were imaged using RSCM then reconstructed in 3D. The 3D images were run through the quanƟtaƟve data analysis pipeline (b). The 3D brain reconstrucƟons were first downsampled to ~10 µm resolu Ɵon. Images were subsequently registered to the CCFv3. In parallel, c -Fos-posiƟve cells were detected and each locaƟon recorded in the 3D dataset (world space). Each spot was then cla ssified as cell / non-cell using a neural network. Coordinates for each cell were transformed to the CCFv3 space using the deforma Ɵon field obtained from the image registraƟon procedure and associated with a CCFv3 brain region (spot mapping). The resul Ɵng tables of spot centroids and corresponding brain regions were aggregated by brain region and across treatment groups for staƟsƟcal analysis. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 4/41 morphine” and “all saline” groups separately, and then calculated the morphine/saline ra Ɵo along with corresponding p-values. We hereaŌ er refer to the density raƟo simply as density fold-change. Morphine induces broad increases in c-Fos acƟ vity across the whole brain To determine whether morphine exposure induced measurable changes in c-Fos expression across the mouse brain, we ploƩ ed coronal heatmaps of the average cellular density of c-Fos- posiƟve cells from saline- (Fig. 2a) and morphine- exposed (Fig. 2b) animals at 1mm projecƟons. There were clearly observable increases in c-Fos- posiƟve cell density across the brains of morphine-exposed animals with parƟcularly apparent increases in the cor Ɵcal layers. QuanƟtaƟve analysis confirmed that more c-Fos- posiƟve cells were present in brains of morphine-exposed mice compared to brains of saline-treated controls. This increase in morphine-induced acƟvity was observed across most brain regions (Fig. 2c) and remained consistent overall when examining individual experimental groups (Fig. 3). Out of 840 structures in the CCFv3, there were 713 structures with an increase in c-Fos-posiƟve cells following morphine exposure. In contrast, 125 structures showed a decrease in c-Fos-posiƟve cells, and 2 structures showed no change following either treatment. Of the structures with decreased ac Ɵvity, the highest number s were in the Isocortex (52 structures), Thalamus (17 structures) and Midbrain (12 structures). In the Thalamus, ~25% of substructures showed reduced acƟvity. At a coarser scale, we examined large brain regions that collecƟvely consƟtute 90% of the brain volume. Our data showed that each of these structures on average had more c-Fos-posiƟve cells in the brain from morphine-exposed animals compared to saline controls (Table 1, Fig. S1). Morphine-treated brains demonstrated the greatest fold- change in Pons and Medulla, and the lowest fold-change in the Thalamus and Cerebellum. Sta ƟsƟcally significant increases at this depth of the structure tree were only observed in the Pons, Cor Ɵcal Subplate and Hippocampal FormaƟon. Overall, the highest density of morphine- acƟvated c -Fos puncta was observed in the Cor Ɵcal subplate and Isocortex. Most of the large structures contained subregions with both decreased and increased c- Fos signal. For example, the Isocortex showed mostly increased signal in layers 4-6, and decreased signal in corƟcal layers 1-3 (Fig. 2d). However, layers 1-3 displayed increased c-Fos signal in regions associated with auditory and visual processing. Of the smaller structures within the CCFv3, 46 exhibited significant differences a Ō er morphine administraƟon (p0.25 mm3 volume (to reduce error imposed from registraƟon arƟfacts), and 3) possessed ≥25 c-Fos spots detected on average in brains of both morphine and saline groups (Fig. 4a, Table S1). Among these structures approximately half (24) were within the Isocortex including the Anterior cingulate area (ACA), Auditory areas (AUD), Visual areas (VIS), Somatosensory areas (SS), Prelimbic area (PL), Temporal associa Ɵon area (TEa), Visceral area (VISC), Ectorhinal area (ECT), Infralimbic area (ILA), and the Agranular Insular Area (AI). The remaining structures were found within the Hippocampal FormaƟon, Midbrain including the Ventral Tegmental Area (VTA), CorƟcal Subplate, Pons and Cerebellum. In addiƟon, there were 31 sufficiently large structures (>0.25 mm3, >25 cells) that approached significance (0.05 <= p < 0.1, Table S2). These included the Nucleus Accumbens (ACB, p=0.07), STRUCTURE DENSITY MORPHINE [CELLS/MM3] DENSITY SALINE [CELLS/MM3] DENSITY FOLD CHANGE P-VALUE PONS 121.47 32.28 3.76 0.0297 * MEDULLA 48.93 17.46 2.80 0.1524 CORTICAL SUBPLATE 463.94 212.38 2.18 0.0301 * HYPOTHALAMUS 331.58 163.64 2.03 0.1069 HIPPOCAMPAL FORMATION 393.18 215.53 1.82 0.0376 * STRIATUM 119.69 71.3 1.68 0.2260 PALLIDUM 121.84 74.6 1.63 0.2921 ISOCORTEX 417.89 256.76 1.63 0.2879 MIDBRAIN 180.87 119.48 1.51 0.1869 OLFACTORY AREAS 329.66 222.46 1.48 0.2126 CEREBELLUM 49.68 39.4 1.26 0.6441 THALAMUS 63.82 52.47 1.22 0.6373 Table 1. c-Fos cell densiƟes on major brain structures. *P<0.05 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 5/41 Figure 2. Global c-Fos expression in mice exposed to morphine. 1mm thick coronal heatmaps* of c-Fos posiƟve neurons for all a) saline and b) morphine treated animals. Areas containing the brightest signal are labeled according to the CCFv3 structure. Morphine/saline c-Fos density fold-change is displayed for all grey maƩ er brain structures c) within the CCFv3 and grouped by larger brain region as indicated by color. Cor Ɵ cal areas d) are grouped by layer with color represen Ɵng log2(fold-change). Blue indicates depressed acƟ vity in a region due to morphine exposure and red indicates enhanced acƟvity. *Heatmaps in panels a and b were calculated by aggregaƟng all c-FOS posiƟve neurons, dividing by the total number of animals and then smoothing with a gaussian kernel. All images are displayed on the same scale (CCFv3). .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 6/41 periaqueductal grey (PAG 19, p=0.07) and the midbrain Raphe nuclei (RAmb, p=0.06) which have been previously shown20 to respond following morphine exposure21. Due to the hierarchical nature of the CCFv3, a child structure having a staƟsƟcally significant difference would oŌ en cause its parent structures to have a staƟsƟcally significant difference. For simplicity, we excluded such parent structures from the above lists (see Table S8 for a full list of structures and their significance). We observed that many small structures only contained c-Fos-posiƟve puncta in either the morphine exposed animals or the saline controls (density=0). We herea Ō er refer to these structures as having a binary effect (Table S3, Fig. S3a). When calcula Ɵng fold-change, the binary structures were excluded due to the undefined nature of dividing by 0. Most of these structures fell below our size threshold but were intriguing because in most of the structures, c-Fos acƟvity was observed exclusively in response to morphine treatment. Among all morphine and Figure 3. c-Fos-posiƟ ve cell densiƟ es. A heat map displaying the density of c-Fos expressing cells (cells/mm3) organized according to experimental groups and arranged according to prominent brain structures. Structures were selected based on their depth in the CCFv3 structure tree (leaves, deepest structures) and volume > 0.25 mm 3. Experimental groups were the smallest sets of animals that differed by all 3 variables (treatment, Ɵ me-point, sex). There were 8 experimental groups: morphine-1h-male (M-1h- m, N=5), morphine-1h-female (M-1h-f, N=3), morphine-4h-male (M-4h-m, N=4), morphine-4h-female (M-4h-f, N=3), saline-1h- male (S-1h-m, N=3), saline-1h-female (S-1h-f, N=2), saline-4h-male (S-4h-m, N=3), saline-4h-female (S-4h-f, N=4) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 7/41 all saline brains, morphine brains exhibited binary effect in a total of 39/41 structures: in the Medulla, Isocortex, Midbrain, Pons, Hypothalamus and Thalamus (ordered by decreasing number of “binary” structures). In contrast, 2 structures with binary effect appeared exclusively in the saline condi Ɵon: RPA (Nucleus Raphe Pallidus), SCO (Subcommissural organ). Acute exposure to morphine induces rapid increases in c- Fos acƟ vity in regions implicated in addicƟ on. We then determined the impacts of different duraƟons of morphine exposure on brain region ac ƟvaƟon by comparing the respecƟve morphine and saline groups at 1h and 4h post- administraƟon (Fig. 5a). c-Fos posiƟve cells increased compared to saline controls in both 1h and 4h groups. However, at 4h, the difference between morphine and saline brains was more pronounced with the average fold change for the whole brain increasing from 1.2 at 1h to 2.3 at 4h. Similarly, the number of brain structures displaying sta ƟsƟcally significant differences from saline control increased from 12 structures at 1h (Fig. 4b, Table S4) to 34 structures (Fig. 4c, Table S5) at 4h. Non-overlapping binary effects were observed for both groups (Table S3, Fig. S3a). When we compared differences between 1h and 4h morphine groups, c- Fos density varia Ɵons suggest ed dynamic changes in brain ac Ɵvity ( Fig. S4a). Overall, 29 structures displayed staƟsƟcally significant increases between 1h and 4h post administraƟon (Fig. 5b). Whereas the Midbrain Re Ɵcular Nucleus (MRN) was the only structure to show a staƟsƟcally significant decrease over this period at 2.49 fold (Fig. 5c). Many of the significant structures at 4h belonged to the cortex-basal ganglia loop22,23, reward/moƟvaƟon20,24 and memory 25,26 circuits including the Nucleus Accumbens (ACB), Prelimbic area (PL), Agranular Insular Cortex (AI) 25, Orbitofrontal Cortex Figure 4. Structures with significant differences in c-Fos expression following morphine exposure. Displayed are brain regions that show significant differences between saline control and morphine exposed animals. Structures menƟoned in the literature as implicated in opioid addicƟon are indicated along the outer circle whereas all significant structures are represented in the middle. Panel a) represents 46 structures (p<0.05) between all morphine and all saline animals from table S1. Time points and sex differences are compared in panels b-c (tables S4,S5) and d-e (tables S6, S7 ), respecƟvely. Color represents logarithm of density fold-change. * - p < 0.05, ** - p <0.01, -ˆ binary effect. Structures with binary effects are shown in black. Dorsal Raphe nucleus falls below the size threshold, however it is included because it displays a sta ƟsƟcally significant difference in all subg roups (see Table S8). Other Raphe nuclei are shown since in many cases they had binary effects (RPO, RPA, RO). .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 8/41 Figure 5. Time- and sex-dependent effects of morphine on region- specific brain ac Ɵ vity. (a) The distribuƟon of morphine over saline density fold-change by brain region at 1h (purple) and 4h (green) post-exposure as displayed in a volcano plot. (b) Brain regions where increased ac Ɵvity (pright) according to greatest fold-change. (d) The distribuƟon of morphine over saline density fold-change by brain region for male (blue) and female (orange) animals is displayed in a volcano plot. ( e) Brain regions where increased ac Ɵ vity (p<0.05) was observed in male (blue) and female ( orange) are highlighted in a cartoon model of the brain. Density fold-change for each region is displayed where acƟvity is greater in (f) male and (g) female animals. Bar plots are sorted (leŌ ->right) according to greatest difference between male and female animals. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 9/41 (ORB), Primary and Secondary motor areas (MOp, MOs), Anterior Cingulate Area (ACA) and Subiculum (SUB). Male and female mice display dis Ɵ nct pa Ʃ erns of c -Fos expression in response to morphine We then examined whether the response to morphine differed based on sex. When we compared morphine and saline controls (Fig. 5d), acƟvity was overall greater in male animals which displayed a 1.63 fold increase compared to female animals that displayed a 1.55 fold increase. The number of structures with staƟsƟcally significant responses to morphine were similar between male and female groups with 32 structures responding in males (Fig. 4d, Table S7) and 31 structures responding in females (Fig. 4e, Table S6). Among those structures, 9 regions were seen in both sexes: Dorsal auditory area, layer 5 (AUDd5), Primary auditory area, layer 5 (AUDp5), AUDv5, Field CA3 in the Hippocampus, Entorhinal area, lateral part, layer 5 (ENTl5), Parabrachial nucleus (PB), Supplemental somatosensory area, layer 5 (SSs5), Visceral area, layer 5 (VISC5), Postrhinal area, layer 5 (VISpor5). The remaining structures that differed between sexes displayed disƟnct differences in the anatomical distribu Ɵon (Fig. 4d-e, S4b) Notably, more structures displayed binary effects in male mice with 25 structures in contrast to female mice with only 1 binary structure (Table S3). When we directly compared fold-change in morphine exposed mice between male and female groups, we idenƟfied 31 brain structures with significant differences between the sexes. Male mice displayed 14 structures with a greater fold-change when compared to females (Fig. 5f) whereas female mice demonstrated a greater fold change in the remaining 17 structures (Fig. 5g). Among the structures with more pronounced ac Ɵvity in m ale mice were members of the reward/moƟvaƟon circuit including the Nucleus Accumbens (ACB), Pallidum (PAL) and Lateral Hypothalamus (LHA) 31; Limbic system and the extended Amygdala (LA, BLAa, BLAp, BMAp, sAMY , PA, ECT)32; and the Memory circuits including Anterior Cingulate Area (ACA) and the Subiculum (SUB) 30,31. In contrast, female mice showed more pronounced acƟvity in structures associated with sensory inputs, associa Ɵon and memory, including visual areas (prefix: VIS), auditory areas (prefix: AUD), Inferior Colliculus (IC), Somatosensory areas (SS), Visceral areas (VISC), Entorhinal areas (ENT), Perirhinal areas (PERI), and Temporal AssociaƟon areas (TEa).

In our study, we present a fast, high-throughput approach for mapping drug-evoked neuronal acƟvity in the brains of mice acutely exposed to morphine. Our approach combined high speed microscopic imaging techniques, high-performance compu Ɵng workflows and validated open-source soŌ ware tools to idenƟfy and map cells within the whole mouse brain. We collected high- resoluƟon whole brain data across sexes and Ɵmes a Ō er morphine exposure. The datasets will be made publicly available following peer review and at the Ɵme of publicaƟon, along with all derived data, the code and computaƟonal environments used for analysis. Our workflow revealed Ɵme-, region-, and sex-dependent differences in c-Fos acƟvity throughout discrete regions of the whole brain. Indeed, our results indicate dynamic subregion-specific changes in cell acƟvity that became more pronounced 4h following drug exposure. This includes changes in the reward circuit20 and more broadly, cortex-basal ganglia loop that is involved in reinforcement learning and choice of behavioral programs33. Moreover, male and female animals displayed differences in their morphine responses with males displaying increas ed ac Ɵvity overall compared to females. This included higher ac ƟvaƟon in brain regions that belong to the mesolimbic dopamine system34, broader reward circuits20, and the extended amygdala 35 implicated in addicƟon including Nucleus Accumbens, Basolateral Amygdala and Cingulate Cortex. While the mesolimbic dopamine system is ac Ɵvated more extensively in male mice, the associaƟon and sensory related cor Ɵcal regions are acƟvated more extensively in female mice. Male mice also displayed greater numbers of regions with binary effects where acƟvity was only detected following exposure to morphine but not to saline. CollecƟvely, our work opens the door to a robust, unbiased approach to iden Ɵfy novel drug-induced temporal and spaƟal changes in whole brain acƟvity at cellular resoluƟon. We found region-specific neuronal ac ƟvaƟon in response to acute morphine treatment across 46 brain structures. Although our data confirmed ac Ɵvity in previously reported brain regions 36, we also found unexpected morphine effects in regions not tradi Ɵonally studied in the context of opioid ac Ɵons including Raphe nuclei (dorsal, midbrain, ponƟne), Hippocampo-amygdalar transiƟon area , and basolateral amygdala (BLA). Notably, our data are consistent with previous work showing that most of the structures also express mu opioid receptors (MOR) according to a brain atlas of MOR expression 37. InteresƟngly, we iden Ɵfied strong morphine -induced .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 10/41 acƟvaƟon of c-Fos expression in the BLA despite rela Ɵvely weak MOR. We posit that this discrepancy may be due to an indirect circuit-level effect where upstream MOR acƟvaƟon may drive the acƟvity of BLA cells downstream of the original morphine sƟmulaƟon. Consistent with this, opioids modulate the pain circuitry both directly and indirectly via acƟons on serotonergic neurons that extend from the PAG through the rostral ventromedial medulla enroute to the spinal cord 38. These poten Ɵal MOR - serotonin links are also evident in the prominent morphine- induced acƟvaƟon of the DR and related Raphe nuclei, structures classically associated with the serotonin system, which is in line with previous research38–42. We idenƟfied disƟnct sex differences in the response to morphine in male versus female mice, consistent with growing evidence of sex differences associated with opioid acƟons both clinically and in preclinical models 43,44. In preclinical models, females demonstrate greater opioid self-administraƟon compared to males 45–48. Likewise, clinically, females are ~4-fold more likely to inject heroin compared to males49,50. InteresƟngly, we found that males exhibited more extensive morphine-induced cell acƟvaƟon in the Striatum, Pallidum, Amygdala and Hippocampus while female mice showed significantly more c-Fos expression primarily in the Isocortex. This raises the possibility that these sex differences are due to sexually dimorphic pa Ʃ erns of MOR expression. Consistent with this, females possess a higher density of MOR-expressing neurons in several cor Ɵcal regions including ACA and SS versus males51. It is also possible that these sex differences are at least parƟally mediated by estrogen, given estrogen’s emerging role as a modulator of the brain opioid system49,52. Future work will be required to elucidate the mechanisms underlying the interac Ɵons between opioid acƟons, sex, and brain regions. We have also found temporal differences in cell acƟvaƟon at 1h versus 4h Ɵme points. At 1h following drug administraƟon, there were fewer anatomic structures with significant differences between morphine and saline condiƟons. Behavioral studies have demonstrated onset of morphine-induced hyperlocomoƟon within 20-min of drug administraƟon53, suggesƟng that 1h of exposure is sufficient for alteraƟons in cell ac Ɵvity within select brain regions. However, by 4h post-administraƟon, the data are less variable, with clearer cell ac Ɵvity within structures tradiƟonally related to the ac Ɵons of drugs of abuse including the Nucleus Accumbens, prelimbic area, infralimbic area, orbital area, motor area, agranular insular area and anterior cingulate area. Our findings are therefore in agreement with previous studies showing that c-Fos expression peaks in both morphine and saline brains at 30- 60 minutes post-injecƟon followed by a return to baseline in saline but not morphine condiƟons54. We note that our use of c- Fos labeling to iden Ɵfy morphine-induced neuronal ac Ɵvity may capture the strongest sƟmuli but sƟll miss other groups of cells with less robust pa Ʃ erns of ac ƟvaƟon55. This can some Ɵmes be exacerbated by the inherent properƟes of the brain Ɵssue and imperfecƟons in clearing which lead to light absorpƟon and sca Ʃ ering, ulƟmately reduc ing signal-to-noise deep within the Ɵssue. To increase our chances of detecƟng low signal-to-noise events, our workflow leveraged two

of cell detec Ɵon combining convenƟonal threshold-based detecƟon and neural networks that are threshold independent. While c- Fos detec Ɵon acknowledges ac Ɵvity-driven expression, this approach does not permit measurement of the magnitude of cell acƟvaƟon. Thus, we cannot use our approach to quanƟtaƟvely ascertain whether already -acƟvated cells exhibit change of expression in response to different duraƟons of morphine exposure. Our analysis did not make conclusions about structures in the CCFv3 below a 0.25mm 3 volume threshold which included 270 structures within the grey maƩ er. This was a deliberate choice designed to reduce any inaccuracies in our analysis that might have been introduced due to errors in the alignment to the CCFv3 template. Very small brain regions require exceedingly accurate alignment. Inaccuracies in the mul Ɵmodal alignment of large volumetric datasets are to be expected. Modality-specific atlas reference images56 exist for other imaging modaliƟes and might improve accuracy, but currently there is no confocal-specific template for the CCFv3. We expect that small inconsistencies in alignments would average out with greater numbers of animals per group. Thus, we expect that future studies will focus on imaging a greater number of animals which will enable us to obtain both increased accuracy and staƟsƟcal power.

Recent t echnological advances in Ɵssue clearing and whole brain imaging have enabled us to digitally reconstruct enƟre mouse brains at cellular resolu Ɵon. By merging high-speed RSCM 14 and automated high- performance compuƟng workflows designed to detect and map individual cells to loca Ɵons within the brain, we can comprehensively map drug- acƟvated neurons across dozens of animals in an unbiased manner at single-cell resoluƟon. The resulƟng data demonstrate broad paƩ erns .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 11/41 of increased acƟvity within the brains of morphine-exposed animals in a sex- and region-specific manner. Importantly, a single experiment allowed us to confirm the role of structures described in a diverse literature focused on specific brain regions and simultaneously iden Ɵfy novel structures which vary with Ɵme and sex . Overall, our fast whole brain imaging combined with massively parallel analysis opens avenues for exploring a large parameter space including ac Ɵons of different drugs, different Ɵme courses of drug ac Ɵon, as well as sex- and region-specific effects. These studies therefore offer a more comprehensive understanding of the acƟons of opioids and offer targets for the development of novel therapeu Ɵc approaches.

Animals Adult male and female mice (C57BL/6J, Jackson Laboratory, stock # 000664), (Total N=27; 15 males, 12 females, 8–10-weeks-old). All animals were housed in a 12/12 light cycle (7 am lights on, 7 pm lights off). Water and rodent chow were provided ad libitum throughout the experiment. Animals were habituated to the laboratory environment for 1 week prior to experimental tes Ɵng. All procedures were approved and performed in compliance with the Ins ƟtuƟonal Animal Care and Use Commi Ʃ ee at Boston University and the University of PiƩ sburgh. Drugs All drugs were obtained from Sigma-Aldrich (St. Louis, MO) unless noted otherwise. Morphine sulfate was dissolved in a soluƟon of filtered (0.22µm) 0.9% saline. Drug Treatment and Brain CollecƟ on Mice were randomly assigned to 1h or 4h treatment groups. Animals were injected i.p. with either morphine (10mg/kg) or saline (10mg/kg, w/v) . Morphine soluƟon or saline were injected intraperitoneally (i.p.) in awake animals that were lightly restrained. Mice were immediately placed individually in a novel environment consisƟng of a plexiglass chamber with animal bedding. Animals were allowed to explore this environment for 1h before undergoing brain collecƟon (1h group) or returned to their home cage for 3- hours before brain collec Ɵon at the 4h Ɵme point ( 4h group). Animals underwent transcardiac perfusion with ice- cold PBS un Ɵl blood had been cleared from the mouse. This was followed by 4% paraformaldehyde (PFA) perfusion for 10 minutes. Fixed brains were dissected and then transferred to 4% PFA at 4ºC for 24 hours before transferring to ice-cold PBS. Tissue Clearing and Staining Brains were prepared using a mixture of CUBIC 57, 3DISCO58, and SWITCH59 clearing and staining protocols. All steps took place at room temperature. The samples were pre-cleared using 50% CUBIC R1 for 1-2 weeks, changing the solu Ɵon every 24 -48 hours. A Ō er pre-clearing, CUBIC R1 was rinsed out and samples were stained with a rabbit monoclonal anƟ-phospho-c-Fos primary anƟbody (catalog# 8677S, Cell Signaling Technology, Danvers, MA) according to the SWITCH protocol. Samples were equilibrated in SWITCH-Off solu Ɵon (5 mM SDS + 0.04% NaAz) for 2 4h, followed by addi Ɵon of the anƟ-c-Fos primary an Ɵbody (Rb) to the samples (1:250 in SWITCH-Off soluƟon). Samples incubated in primary an Ɵbody for 1 week, then washed for 24-48 hours in SWITCH- On soluƟon (1X PBS + 0.04% NaAz + TritonX). This process was repeated with Cy5- conjugated goat and d onkey an Ɵ-rabbit secondary anƟbodies (1: 2000, catalog#s 111-605-003 and 711-175- 152, Jackson Immunoresearch, West Grove, PA) in SWITCH- Off. Samples were post-fixed in 4% paraformaldehyde for 1h. A Ō er staining, samples were dehydrated in 2 -hour cycles of increasing concentra Ɵons of tetrahydrofuran (%THF: 30, 50, 70, 90 (x2), 100 (x2)). Following dehydraƟon, samples were cleared and mounted in dibenzyl ether to be imaged. Imaging Brains were imaged via ribbon scanning confocal microscopy (RSCM, CALIBER I.D. RS-G4, Andover, MA, USA) as described previously 14. Briefly, up to 8 brains were immersed in dibenzyl ether and arrayed in a single imaging chamber. The chamber was sealed with a glass coverslip and then filled with glycerol. We used a Nikon 20x, 1.0NA, 8.2mm WD, glycerol immersion (CF190) objec Ɵve specifically designed for cleared Ɵssue imaging. Data were acquired using a 488 (autofluorescence filter Semrock FF01-520/44 filter) and 647 (Cy5 c-Fos signal - filter Semrock FF01-670/30) filter with full sequencing enabled. Four samples were imaged at 0 .337µm x 0 .337µm x 4.57µm (X,Y ,Z) voxel resoluƟon, and twenty-three samples were imaged with 0 .361µm x 0 .361µm x 5 .33µm (X,Y ,Z) voxel resoluƟon. The data were s Ɵtched and assembled into 2D images each represenƟng a single channel and z- plane. Shot noise was then removed from the images using a convoluƟonal neural network based on noise -2-noise60 and trained on RSCM data. Image planes were then assembled into a 3D volume using the freely available Imaris File Converter (Bitplane), resulƟng in a single Imaris file composed of a mul Ɵscale chunked data structure that .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 12/41 represented the enƟre whole brain dataset. All subsequent downstream analysis was completed from these files. Compute We produced an automated c-Fos mapping pipeline which included the detecƟon of c-Fos-posiƟve cells and the alignment of each brain to the Allen Mouse Brain Common Coordinate Framework v317 (CCFv3). Image processing was performed using the Center for Biologic Imaging’s high performance compuƟng environment. The environment is schedulable via SLURM18 and consists of a 18 node compute parƟƟ on having 24 cores and 96GB RAM and a 5 node GPU parƟƟ on each with 72 cores, 1.5TB RAM and 8x NVIDIA P40. All nodes mount two BeeGFS high performance file systems consisƟng of 7PB of HDD and 160TB of SSD. RSCM data was acquired directly as uns Ɵtched ribbons to the BeeGFS SSD storage system prior to being s Ɵtched and assembled on the compute nodes as 2D images. Data was then copied to the HDD storage system where they were queued for denoising on the GPU nodes. Following denoising, all images were assembled into Imaris (.ims) files which were the basis for all remaining image analysis. The final size of each full brain dataset was on average ∼1.5 terabytes, comprising 40TB across 27 brains. During image generaƟon and processing, these data were transformed 3 Ɵmes transiently genera Ɵng approximately 200TB. The final denoised, ims files were retained and are publicly available at the Brain Image Library as discussed below. Data mining and analysis was developed and tested on custom built workstaƟons consisƟng of 32 -core, 64-thread AMD Ryzen Threadripper PRO, 256GB RAM, and NVIDIA A6000 GPU. c-Fos puncta detecƟ on Data contained fluorescent c-Fos- posiƟve puncta of different sizes and intensiƟes. We found that the larger and brighter spots were well detected with cellfinder 61 python tool, whereas small and dim spots were detectable with the deepblink parƟcle model 62 based on UNet architecture 63. Both algorithms were applied to each imaged brain. Together, these algorithms were purposely biased towards over detecƟon to ensure that all cells were idenƟfied. Thus, to avoid duplicate detecƟons, the resulƟng spots were run through DBSCAN clustering algorithm64 which clustered the duplicated detecƟons together and replaced these clusters by a single point. In addi Ɵon, detected points contained arƟfacts such as edges, noise peaks and autofluorescent vasculature. To eliminate these ar Ɵfacts, each detected spot was fed to a deep learning binary classifier based on ResNet-5065. The model was trained on >40,000 cell and non-cell examples across mulƟple brains from this study. To confirm precision, recall, and f166 score of the detecƟon and classificaƟon process, ground truth data was annotated manually in a small patch of size of 25 x 250 x 250 pixels from each brain. For brains where the f1 score was <80%, addiƟonal improvement of classifica Ɵon was needed. For this, the model was fine-tuned on a relaƟvely small number (50-200 examples of cells and non-cells) of manually annotated data from each of these brains, using fastai 67. Finally, the data was inspected visually, and for a few brains where classificaƟon sƟll did not reach a f1 score of ≥80%, minor manual removal of residual ar Ɵfacts was done. OpƟonally, removal of background detecƟons can be done to reduce the Ɵme spent on clustering duplicated detecƟons and cell/arƟfact classifica Ɵon. We created background-foreground masks for each brain using accelerated pixel and object classifica Ɵon plugin for napari68 that employs random forest machine learning algorithm69. These masks will be provided alongside the raw data. RegistraƟ on to the CCFv3 Each brain was registered to the 10µm CCFv3 following downsampling of the full resolu Ɵon to 10 µm isotropic resoluƟon. Brainreg 70 a fully automated 3D registra Ɵon python package was used for all registra Ɵons. The registraƟon results were inspected visually . If the registraƟon quality was not sa Ɵsfactory, the raw data was pre-processed by filtering in the frequency domain, contrast stretching, and background subtrac Ɵon unƟl a visually good alignment was obtained. The resul Ɵng deformaƟon field was used to transform each of the c -Fos points into CCFv3 space. Every c-Fos- posiƟve cell was associated with an Allen Mouse Brain Atlas parcella Ɵon where it was located. CriƟcally, Ɵssue clearing results in some physical deformaƟon of the mouse brain. Although rela Ɵvely rare, the stresses placed on delicate brain Ɵssue can someƟmes cause tearing or result in Ɵssue loss. Though nonlinear alignment algorithms are used in brainreg as an aƩ empt to correct for changes in Ɵssue shape, they are never enƟrely accurate. Thus, to reduce errors resul Ɵng from minor inaccuracies in alignment, we eliminated the smallest brain structures by focusing our downstream analysis on structures that were at least 0.25 mm 3. At this scale, we expect that inaccuracies in alignment will be compensated for by averaging data from mulƟple brains. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 13/41 Data analysis Our workflow produced a CSV table, which included the locaƟons of c -Fos-posiƟve cells mapped to the CCFv3 and labeled with the associated parcella Ɵon. These secondary data tables will be made public alongside the raw data as menƟoned below. For analysis, tables for all of the brains were combined and used for data mining. Each row in a corresponding table contained the coordinates for a spot in the raw brain image (world space), coordinates for a spot the CCFv3, and the spot’s corresponding CCFv3 parcellaƟon (i.e., brain structure). CSV files for all brains were loaded into an interacƟve Jupyter71 notebook for analysis. For each brain, the points were aggregated by brain structure, and the density of c-Fos- posiƟve cells for each structure was calculated as the number of spots in the structure over the volume of the structure. Densi Ɵes were the main metric used to make comparisons of structures across brains from different animals and different experimental condiƟons. VisualizaƟ ons Raw data was stored in the Imaris file format, and Imaris viewer was used for ini Ɵal visual inspec Ɵon. Volumetric brain cartoons were made using brainrender72. Bar and pie charts were done using Plotly 73 or Seaborn 74. Heatmaps in the CCFv3 were made in napari75. StaƟ sƟ cal analysis To calculate p-values between different experimental condiƟons in bulk, in an automa Ɵc way, for all brain structures, we used student’s 2-sample t-test, available through scipy76. Numbers of animals for each group ranged from 2-5. For the analysis, we pooled staƟsƟcally independent groups together, therefore the n was always ≥5. Unless otherwise indicated, data is represented as mean ± standard error of the mean. Data Availability At the Ɵme of publica Ɵon, a ll full resolu Ɵon imaging data described herein will be deposited in the Brain Image Library (BIL) and be made available along with the derived analysis data under an associated DOI. Code Availability All code used to generate the results described in this manuscript will be deposited in github and/or made available alongside the imaging data deposited at BIL. Jupyter notebooks will be provided which reproduce the figures and tables.

We are grateful for the discussions and technical assistance provided by Dr. Sam Golden. This study was supported by the Na Ɵonal Ins Ɵtutes of Health (R01DK124219 to Z.F.; R01ES034037 to Z.F., R36DA057972 to J.K.; T32GM133353 to J.K., R01DA061243 to R.W.L. and Z.F., R21DA052419 to R.W.L., Z.F., and A.M.W., R24MH114793 to A.M.W.), the Department of Defense (PR210207 to Z.F.), the Chan Zuckerberg Ini ƟaƟve DAF (2023-329680 to K.M.K) and the Lundbeck Founda Ɵon (R276-2018-792 and R359-2020-2301 to U.G.). Author contribuƟ ons statement I.V., M.C.S., S.P ., J.K., P .N.J., A.M.W. conducted the experiments. I.V., M.C.S, E.A., K.H., T.L., K.M.K., J.J.S., Z.F., M.D.L., A.M.W. analyzed the results. I.V., M.C.S., Z.F., A.M.W wrote the manuscript. R.W.L., Z.F., A.M.W. conceived the experiment(s). U.G., R.W.L., Z.F., A.M.W. supervised the work, and contributed funding. All authors reviewed the manuscript. Disclosure Z.F. is funded by an invesƟgator-iniƟated award from UPMC. The other authors do not report any conflicts of interest. AbbreviaƟ ons BIL - Brain Image Library CCFv3 – Allen Brain Atlas Common Coordinate Framework Version 3 CSV - Character Separated Values FAIR - Findable, Accessible, Interoperable, Reusable GPU - Graphical Processing Unit HDD - Hard Disk Drive IEG - Immediate Early Gene i.p - intraperitoneal MOR - Mu Opioid Receptor OUD - opioid use disorder PB - petabyte RAM - Random Access Memory RSCM – Ribbon Scanning Confocal Microscopy SLURM - Simple Linux UƟ lity for Resource Management SSD - Solid State Drive TB - Terabyte Brain Regions found within the CCFv3 AAA - Anterior Amygdalar Area ACA - Anterior Cingulate Area ACAd1 - Anterior Cingulate Area, dorsal part, layer 1 ACAd2/3 - Anterior Cingulate Area, dorsal part, layers 2/3 ACAd5 - Anterior Cingulate Area, dorsal part, layer 5 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 14/41 ACAd6a - Anterior Cingulate Area, dorsal part, layer 6a ACAv1 - Anterior Cingulate Area, ventral part, layer 1 ACAv2/3 - Anterior Cingulate Area, ventral part, layers 2/3 ACAv5 - Anterior Cingulate Area, ventral part, layer 5 ACAv6a - Anterior Cingulate Area, ventral part, layer 6a ACB - Nucleus Accumbens AId - Agranular insular area, dorsal part AId1 - Agranular Insular Area, dorsal part, layer 1 AId2/3 - Agranular Insular Area, dorsal part, layers 2/3 AId5 - Agranular Insular Area, dorsal part, layer 5 AId6a - Agranular Insular Area, dorsal part, layer 6a AI - Agranular Insular Area AIp1 - Agranular Insular Area, posterior part, layer 1 AIp2/3 - Agranular Insular Area, posterior part, layers 2/3 AIp5 - Agranular Insular Area, posterior part, layer 5 AIp6a - Agranular Insular Area, posterior part, layer 6a AIv2/3 - Agranular Insular Area, ventral part, layers 2/3 AIv5 - Agranular Insular Area, ventral part, layer 5 AHN - Anterior Hypothalamic Nucleus AOBmi - Accessory Olfactory Bulb, mitral layer AON - Anterior Olfactory Nucleus APN - Anterior Pretectal Nucleus APr - Area Prostriata ARH - Arcuate Hypothalamic Nucleus AM - Anteromedial Nucleus AV - Anteroventral Nucleus BLA - Basolateral amygdalar nucleus BLAa - Basolateral Amygdalar Nucleus, anterior part BLAp - Basolateral Amygdalar Nucleus, posterior part BLAv - Basolateral Amygdalar Nucleus, ventral part BMA - Basomedial amygdalar nucleus BMAa - Basomedial Amygdalar Nucleus, anterior part BMAp - Basomedial Amygdalar Nucleus, posterior part BST, BNST - Bed nuclei of the stria terminalis CA1 - Cornu Ammonis area 1 CA2 - Cornu Ammonis area 2 CA3 - Cornu Ammonis area 3 COA - CorƟ cal amygdalar area COAa - CorƟ cal Amygdalar Area, anterior part COApl - CorƟ cal Amygdalar Area, posterolateral part COApm - CorƟcal Amygdalar Area, posteromedial part CP - Caudoputamen CTXsp - CorƟcal Subplate DG-mo - Dentate Gyrus, molecular layer DG-po - Dentate Gyrus, polymorphic layer DG-sg - Dentate Gyrus, granule cell layer DMH - Dorsomedial Hypothalamic Nucleus DMX - Dorsal motor nucleus of the vagus nerve DP - Dorsal Peduncular Area DCO - Dorsal Cochlear Nucleus DR - Dorsal Raphe Nucleus ECT - Ectorhinal Area ECT1 - Ectorhinal Area, layer 1 ECT2/3 - Ectorhinal Area, layers 2/3 ECT5 - Ectorhinal Area, layer 5 ECT6a - Ectorhinal Area, layer 6a EP - Endopiriform nucleus EPd - Endopiriform Nucleus, dorsal part EPv - Endopiriform Nucleus, ventral part ENT - Entorhinal Area ENTl1 - Lateral Entorhinal Area, layer 1 ENTl2 - Lateral Entorhinal Area, layer 2 ENTl3 - Lateral Entorhinal Area, layer 3 ENTl5 - Lateral Entorhinal Area, layer 5 ENTl5 - Lateral Entorhinal Area, layer 5 ENTl6a - Lateral Entorhinal Area, layer 6a ENTm1 - Medial Entorhinal Area, layer 1 ENTm2 - Medial Entorhinal Area, layer 2 ENTm3 - Medial Entorhinal Area, layer 3 ENTm5 - Medial Entorhinal Area, layer 5 ENTm6 - Medial Entorhinal Area, layer 6 FRP - Frontal pole, cerebral cortex FRP1 - Frontal Pole, layer 1 FRP5 - Frontal Pole, layer 5 FS - Fundus of Striatum GU - Gustatory areas GU2/3 - Gustatory Area, layers 2/3 GU5 - Gustatory Area, layer 5 GU6a - Gustatory Area, layer 6a GPe - Globus Pallidus, external segment GPi - Globus Pallidus, internal segment GRN - Gracile Nucleus HPF - Hippocampal FormaƟon HATA - Hippocampus-Amygdala TransiƟon Area HY – Hypothalamus ILA - Infralimbic Area IC - Inferior Colliculus ICc - Inferior Colliculus, central nucleus ICd - Inferior Colliculus, dorsal cortex ICe - Inferior Colliculus, external cortex IO - Inferior Olive IPN - Interpeduncular Nucleus IRN - Intermediate ReƟ cular Nucleus LA - Lateral Amygdalar Nucleus LAV - Lateral VesƟbular Nucleus LD - Lateral Dorsal Nucleus LHA - Lateral Hypothalamic Area LGd-co - Lateral Geniculate Nucleus, dorsal, core region LGv - Lateral Geniculate Nucleus, ventral part LP - Lateral Posterior Nucleus LS - Lateral septal nucleus LSc - Lateral Septum, caudal part LSr - Lateral Septum, rostral part LSv - Lateral Septum, ventral part .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 15/41 LPO - Lateral preopƟc area LRNm - Lateral ReƟcular Nucleus, medial part MA - Medial Amygdala MEA - Medial Amygdalar Nucleus MB - Midbrain MGm - Medial Geniculate Nucleus, magnocellular division MGv - Medial Geniculate Nucleus, ventral division MM - Medial Mammillary Nucleus MO - Somatomotor areas MOB - Main Olfactory Bulb MOp - Primary Motor Area MOp - Primary motor area MOp1 - Primary Motor Area, layer 1 MOp2/3 - Primary Motor Area, layers 2/3 MOp5 - Primary Motor Area, layer 5 MOp6a - Primary Motor Area, layer 6a MOs - Secondary Motor Area MOs1 - Secondary Motor Area, layer 1 MOs2/3 - Secondary Motor Area, layers 2/3 MOs5 - Secondary Motor Area, layer 5 MOs6a - Secondary Motor Area, layer 6a MARN - Magnocellular ReƟcular Nucleus MDRN - Medullary ReƟcular Nucleus MS - Medial Septum MRN - Midbrain ReƟcular Nucleus MY - Medulla NAc - Nucleus Accumbens NDB - Nucleus of the Diagonal Band NPC - Nucleus of the Posterior Commissure NLOT - Nucleus of the lateral olfactory tract NLL - Nucleus of the lateral lemniscus NTS - Nucleus of the Solitary Tract ORB - Orbital Area ORBl - Orbital area, lateral part ORBl1 - Orbital Area, lateral part, layer 1 ORBl2/3 - Orbital Area, lateral part, layers 2/3 ORBl5 - Orbital Area, lateral part, layer 5 ORBl6a - Orbital Area, lateral part, layer 6a ORBm - Orbital area, medial part ORBm1 - Orbital Area, medial part, layer 1 ORBm2/3 - Orbital Area, medial part, layers 2/3 ORBm5 - Orbital Area, medial part, layer 5 ORBvl - Orbital area, ventrolateral part ORBvl1 - Orbital Area, ventrolateral part, layer 1 ORBvl2/3 - Orbital Area, ventrolateral part, layers 2/3 ORBvl5 - Orbital Area, ventrolateral part, layer 5 OT - Olfactory Tubercle PA - Posterior Amygdalar Nucleus PAA - Piriform-Amygdalar Area PAL - Pallidum PAG - Periaqueductal Gray PB - Parabrachial Nucleus p5 - PonƟne ReƟcular FormaƟon, p5 Subregion PCG - PonƟne Central Gray PG - PonƟne Gray PIR - Piriform area POR - PonƟne ReƟcular FormaƟon, Oral Part PRNc - PonƟne ReƟcular Nucleus, Caudal Part PRNr - PonƟne ReƟ cular Nucleus, Rostral Part PS - Parastrial nucleus PSV - PonƟne Superior VesƟbular Nucleus PVT - Paraventricular nucleus of the thalamus RAmb - Nucleus ambiguus, rostral part RPA - Raphe Pallidus RE - Nucleus of reuniens RPO - Raphe PonƟs Nucleus RO - Raphe Obscurus Nucleus RSP - Retrosplenial area RSPagl - Retrosplenial area, lateral agranular part RSPagl1 - Retrosplenial Area, agranular, lateral part, layer 1 RSPagl2/3 - Retrosplenial Area, agranular, lateral part, layers 2/3 RSPagl5 - Retrosplenial Area, agranular, lateral part, layer 5 RSPagl6a - Retrosplenial Area, agranular, lateral part, layer 6a RSPd - Retrosplenial area, dorsal part RSPd1 - Retrosplenial Area, dorsal part, layer 1 RSPd2/3 - Retrosplenial Area, dorsal part, layers 2/3 RSPd5 - Retrosplenial Area, dorsal part, layer 5 RSPd6a - Retrosplenial Area, dorsal part, layer 6a RSPv - Retrosplenial area, ventral part RSPv1 - Retrosplenial Area, ventral part, layer 1 RSPv2/3 - Retrosplenial Area, ventral part, layers 2/3 RSPv5 - Retrosplenial Area, ventral part, layer 5 RSPv6a - Retrosplenial Area, ventral part, layer 6a RN - Red Nucleus SCO - Subcommissural Organ SC - Superior colliculus SCdg - Superior Colliculus, deep gray layer SCdw - Superior Colliculus, deep white layer SCig - Superior Colliculus, intermediate gray layer SCiw - Superior Colliculus, intermediate white layer SCop - Superior Colliculus, opƟc layer SCsg - Superior Colliculus, superficial gray layer SCzo - Superior Colliculus, zonal layer SF - Septofimbrial Nucleus SOCl - Superior Olivary Complex, Lateral Part SPIV - Spinal Trigeminal Nucleus, ventral part SPVC - Spinal Trigeminal Nucleus, caudal part SPVI - Spinal Trigeminal Nucleus, intermediate part SPVO - Spinal Trigeminal Nucleus, oral part SS - Somatosensory Area SSP-bfd1 - Primary Somatosensory Area, Barrel Field, layer 1 SSP-bfd2/3 - Primary Somatosensory Area, Barrel Field, layers 2/3 SSP-bfd4 - Primary Somatosensory Area, Barrel Field, layer 4 SSP-bfd5 - Primary Somatosensory Area, Barrel Field, layer 5 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 16/41 SSP-bfd6a - Primary Somatosensory Area, Barrel Field, layer 6a SSp-bfd2/3 - Primary Somatosensory Area, Barrel Field, layers 2/3 SSp-ll - Primary Somatosensory Area, Lower Limb SSp-ll1 - Primary Somatosensory Area, Lower Limb, layer 1 SSp-ll2/3 - Primary Somatosensory Area, Lower Limb, layers 2/3 SSp-ll4 - Primary Somatosensory Area, Lower Limb, layer 4 SSp-ll5 - Primary Somatosensory Area, Lower Limb, layer 5 SSp-ll6a - Primary Somatosensory Area, Lower Limb, layer 6a SSp-m - Primary Somatosensory Area, Mouth SSp-m1 - Primary Somatosensory Area, Mouth, layer 1 SSp-m2/3 - Primary Somatosensory Area, Mouth, layers 2/3 SSp-m4 - Primary Somatosensory Area, Mouth, layer 4 SSp-m5 - Primary Somatosensory Area, Mouth, layer 5 SSp-m6a - Primary Somatosensory Area, Mouth, layer 6a SSp-n - Primary Somatosensory Area, Nose SSp-n1 - Primary Somatosensory Area, Nose, layer 1 SSp-n2/3 - Primary Somatosensory Area, Nose, layers 2/3 SSp-n4 - Primary Somatosensory Area, Nose, layer 4 SSp-n5 - Primary Somatosensory Area, Nose, layer 5 SSp-n6a - Primary Somatosensory Area, Nose, layer 6a SSp-tr - Primary Somatosensory Area, Trunk SSp-tr2/3 - Primary Somatosensory Area, Trunk, layers 2/3 SSp-tr5 - Primary Somatosensory Area, Trunk, layer 5 SSp-ul - Primary Somatosensory Area, Upper Limb SSp-ul1 - Primary Somatosensory Area, Upper Limb, layer 1 SSp-ul2/3 - Primary Somatosensory Area, Upper Limb, layers 2/3 SSp-ul4 - Primary Somatosensory Area, Upper Limb, layer 4 SSp-ul5 - Primary Somatosensory Area, Upper Limb, layer 5 SSp-ul6a - Primary Somatosensory Area, Upper Limb, layer 6a SSp-un - Primary Somatosensory Area, Unassigned Region SSp-un2/3 - Primary Somatosensory Area, Unassigned Region, layers 2/3 SSp-un5 - Primary Somatosensory Area, Unassigned Region, layer 5 SSp-un6a - Primary Somatosensory Area, Unassigned Region, layer 6a SSs1 - Secondary Somatosensory Area, layer 1 SSs2/3 - Secondary Somatosensory Area, layers 2/3 SSs4 - Secondary Somatosensory Area, layer 4 SSs5 - Secondary Somatosensory Area, layer 5 SSs6a - Secondary Somatosensory Area, layer 6a SUT - Supratrigeminal Nucleus SUM - Supramammillary Nucleus TEa - Temporal AssociaƟon Area TEa1 - Temporal AssociaƟon Area, layer 1 TEa2/3 - Temporal AssociaƟon Area, layers 2/3 TEa4 - Temporal AssociaƟon Area, layer 4 TEa5 - Temporal AssociaƟon Area, layer 5 TEa6a - Temporal AssociaƟon Area, layer 6a TR - Taenia Tecta, rostral part TRN - Tegmental ReƟcular Nucleus TRS - Triangular Septal Nucleus TU - Tuberal Nucleus UVU - Uvula V - VesƟbular Nucleus VISC - Visceral Area VISC1 - Visceral Area, layer 1 VISC2/3 - Visceral Area, layers 2/3 VISC4 - Visceral Area, layer 4 VISC5 - Visceral Area, layer 5 VISC6a - Visceral Area, layer 6a VIS - Visual Area VISa - Anterior area VISa2/3 - Visual AssociaƟon Area, layers 2/3 VISa5 - Visual AssociaƟon Area, layer 5 VISal - Anterolateral Visual Area VISam - Anteromedial Visual Area VISl - Lateral visual area VISl2/3 - Lateral Visual Area, layers 2/3 VISl5 - Lateral Visual Area, layer 5 VISli - Laterointermediate Visual Area VISp - Primary visual area VISp1 - Primary Visual Area, layer 1 VISp2/3 - Primary Visual Area, layers 2/3 VISp4 - Primary Visual Area, layer 4 VISp5 - Primary Visual Area, layer 5 VISp6a - Primary Visual Area, layer 6a VISpl - Posterolateral Visual Area VISpm - Posteromedial Visual Area VISpm2/3 - Posteromedial Visual Area, layers 2/3 VISpm5 - Posteromedial Visual Area, layer 5 VISpor - Postrhinal visual area VISpor2/3 - Postrhinal (Visual) Area, layers 2/3 VISpor5 - Postrhinal (Visual) Area, layer 5 VTA - Ventral Tegmental Area VM - Ventromedial Nucleus VPL - Ventral Posterolateral Nucleus VPM - Ventral Posteromedial Nucleus VAL - Ventral Anterior-Lateral Nucleus XII - Hypoglossal Nucleus ZI - Zona Incerta

1. Koob, G. F. Neurobiology of Opioid AddicƟon: Opponent Process, HyperkaƟ feia, and NegaƟve Reinforcement. Biol Psychiatry 87, 44–53 (2020). 2. Jones, A. A., Segel, J. E., Skogseth, E. M., Apsley, H. B. & Santos-Lozada, A. R. Drug overdose deaths among women 1999-2021 in the United States: Differences by race, ethnicity, and age. Womens Health (Lond) 20, 17455057241307088 (2024). 3. Tiwari, K., Rahimian, M. A., Roberts, M. S., Kumar, P . & Buchanich, J. M. Measuring network dynamics of opioid .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 17/41 overdose deaths in the United States. Sci Rep 14, 29563 (2024). 4. Jalal, H. et al. Changing dynamics of the drug overdose epidemic in the United States from 1979 through 2016. Science 361, eaau1184 (2018). 5. Gossop, M., Green, L., Phillips, G. & Bradley, B. Lapse, relapse and survival among opiate addicts aŌ er treatment. A prospecƟve follow-up study. Br J Psychiatry 154, 348–353 (1989). 6. Harris, A. C. & Gewirtz, J. C. Acute opioid dependence: characterizing the early adaptaƟons underlying drug withdrawal. Psychopharmacology (Berl) 178, 353–366 (2005). 7. Cruz, F. C., Javier Rubio, F. & Hope, B. T. Using c-fos to study neuronal ensembles in corƟcostriatal circuitry of addicƟon. Brain Res 1628, 157–173 (2015). 8. Cruz, F. C. et al. Role of nucleus accumbens shell neuronal ensembles in context-induced reinstatement of cocaine- seeking. J Neurosci 34, 7437–7446 (2014). 9. Whitaker, L. R. & Hope, B. T. Chasing the addicted engram: idenƟfying funcƟonal alteraƟons in Fos-expressing neuronal ensembles that mediate drug-related learned behavior. Learn Mem 25, 455–460 (2018). 10. Salery, M., Godino, A. & Nestler, E. J. Drug- acƟvated cells: From immediate early genes to neuronal ensembles in addicƟon. Adv Pharmacol 90, 173–216 (2021). 11. Koya, E., MargeƩ s-Smith, G. & Hope, B. T. Daun02 InacƟvaƟon of Behaviorally AcƟvated Fos-Expressing Neuronal Ensembles. Curr Protoc Neurosci 76, 8.36.1- 8.36.17 (2016). 12. Warren, B. L., Suto, N. & Hope, B. T. MechanisƟc ResoluƟon Required to Mediate Operant Learned Behaviors: Insights from Neuronal Ensemble-Specific InacƟvaƟon. Front Neural Circuits 11, 28 (2017). 13. MunƟfering, M. et al. Clearing for Deep Tissue Imaging. Curr Protoc Cytom 86, e38 (2018). 14. Watson, A. M. et al. Ribbon scanning confocal for high- speed high-resoluƟon volume imaging of brain. PLoS One 12, e0180486 (2017). 15. FAIR Principles. GO FAIR hƩ ps://www.go-fair .org/fair- principles/. 16. Wilkinson, M. D. et al. The FAIR Guiding Principles for scienƟfic data management and stewardship. Sci Data 3, 160018 (2016). 17. Wang, Q. et al. The Allen Mouse Brain Common Coordinate Framework: A 3D Reference Atlas. Cell 181, 936-953.e20 (2020). 18. Yoo, A. B., JeƩ e, M. A. & Grondona, M. SLURM: Simple Linux UƟlity for Resource Management. in Job Scheduling Strategies for Parallel Processing (eds. Feitelson, D., Rudolph, L. & Schwiegelshohn, U.) vol. 2862 44–60 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2003). 19. Jacquet, Y . F. & Lajtha, A. The periaqueductal gray: site of morphine analgesia and tolerance as shown by 2-way cross tolerance between systemic and intracerebral injecƟons. Brain Res 103, 501–513 (1976). 20. Ikemoto, S. Brain reward circuitry beyond the mesolimbic dopamine system: a neurobiological theory. Neurosci Biobehav Rev 35, 129–150 (2010). 21. Welsch, L. et al. Mu Opioid Receptor-Expressing Neurons in the Dorsal Raphe Nucleus Are Involved in Reward Processing and AffecƟve Behaviors. Biol Psychiatry 94, 842– 851 (2023). 22. Miyachi, S. [CorƟ co-basal ganglia circuits--parallel closed loops and convergent/divergent connecƟons]. Brain Nerve 61, 351–359 (2009). 23. Haber, S. N. & Calzavara, R. The corƟco-basal ganglia integraƟve network: the role of the thalamus. Brain Res Bull 78, 69–74 (2009). 24. Haber, S. N. & Knutson, B. The reward circuit: linking primate anatomy and human imaging. Neuropsychopharmacology 35, 4–26 (2010). 25. Kibaly, C., Xu, C., Cahill, C. M., Evans, C. J. & Law, P .-Y . Non- nocicepƟve roles of opioids in the CNS: opioids’ effects on neurogenesis, learning, memory and affect. Nat Rev Neurosci 20, 5–18 (2019). 26. Fredriksson, I. et al. Role of ventral subiculum neuronal ensembles in incubaƟon of oxycodone craving aŌ er electric barrier-induced voluntary absƟnence. Sci Adv 9, eadd8687 (2023). 27. Ceceli, A. O., Bradberry, C. W . & Goldstein, R. Z. The neurobiology of drug addicƟon: cross-species insights into the dysfuncƟon and recovery of the prefrontal cortex. Neuropsychopharmacology 47, 276–291 (2022). 28. McGregor, M. S. & LaLumiere, R. T. SƟll a ‘hidden island’? The rodent insular cortex in drug seeking, reward, and risk. Neurosci Biobehav Rev 153, 105334 (2023). 29. McGregor, M. S., Cosme, C. V. & LaLumiere, R. T. Insular cortex subregions have disƟnct roles in cued heroin seeking aŌ er exƟncƟon learning and prolonged withdrawal in rats. Neuropsychopharmacol. 49, 1540–1549 (2024). 30. McKendrick, G., McDeviƩ , D. S., Shafeek, P ., CoƩ rill, A. & Graziane, N. M. Anterior cingulate cortex and its projecƟons to the ventral tegmental area regulate opioid withdrawal, the formaƟon of opioid context associaƟons and context- induced drug seeking. Front Neurosci 16, 972658 (2022). 31. Gf, K. & Nd, V. Neurocircuitry of addicƟon. Neuropsychopharmacology : official publicaƟon of the American College of Neuropsychopharmacology 35, (2010). 32. Sheng, H. et al. Nucleus accumbens circuit disinhibits lateral hypothalamus glutamatergic neurons contribuƟng to morphine withdrawal memory in male mice. Nat Commun 14, 71 (2023). 33. Baladron, J. & Hamker, F. H. Habit learning in hierarchical cortex-basal ganglia loops. Eur J Neurosci 52, 4613–4638 (2020). 34. Salamone, J. D. & Correa, M. The mysterious moƟvaƟonal funcƟons of mesolimbic dopamine. Neuron 76, 470–485 (2012). .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 18/41 35. Kaplan, G. B. & Thompson, B. L. NeuroplasƟcity of the extended amygdala in opioid withdrawal and prolonged opioid absƟnence. Front Pharmacol 14, 1253736 (2023). 36. Ammon-Treiber, S. & Höllt, V. Morphine-induced changes of gene expression in the brain. Addict Biol 10, 81–89 (2005). 37. E, E. et al. A mu-delta opioid receptor brain atlas reveals neuronal co-occurrence in subcorƟcal networks. Brain structure & funcƟon 220, (2015). 38. FaƩ , M. P . et al. Morphine-responsive neurons that regulate mechanical anƟnocicepƟon. Science 385, eado6593 (2024). 39. Tao, R. & Auerbach, S. B. Involvement of the dorsal raphe but not median raphe nucleus in morphine-induced increases in serotonin release in the rat forebrain. Neuroscience 68, 553–561 (1995). 40. Tao, R. & Auerbach, S. B. GABAergic and glutamatergic afferents in the dorsal raphe nucleus mediate morphine- induced increases in serotonin efflux in the rat central nervous system. J Pharmacol Exp Ther 303, 704–710 (2002). 41. Proudfit, H. K. Effects of raphe magnus and raphe pallidus lesions on morphine-induced analgesia and spinal cord monoamines. Pharmacol Biochem Behav 13, 705–714 (1980). 42. Kalyuzhny, A. E., Arvidsson, U., Wu, W. & Wessendorf, M. W. mu-Opioid and delta-opioid receptors are expressed in brainstem anƟnocicepƟve circuits: studies using immunocytochemistry and retrograde tract-tracing. J Neurosci 16, 6490–6503 (1996). 43. Huhn, A. S., Berry, M. S. & Dunn, K. E. SystemaƟc review of sex-based differences in opioid-based effects. Int Rev Psychiatry 30, 107–116 (2018). 44. Sharp, J. L., Pearson, T. & Smith, M. A. Sex differences in opioid receptor mediated effects: Role of androgens. Neurosci Biobehav Rev 134, 104522 (2022). 45. Wj, L. & Me, C. Sex differences in the acquisiƟon of intravenously self-administered cocaine and heroin in rats. Psychopharmacology 144, (1999). 46. Carroll, M. E., Morgan, A. D., Lynch, W. J., Campbell, U. C. & Dess, N. K. Intravenous cocaine and heroin self- administraƟon in rats selecƟvely bred for differenƟal saccharin intake: phenotype and sex differences. Psychopharmacology (Berl) 161, 304–313 (2002). 47. Cicero, T. J., Aylward, S. C. & Meyer, E. R. Gender differences in the intravenous self-administraƟon of mu opiate agonists. Pharmacol Biochem Behav 74, 541–549 (2003). 48. Klein, L. C., Popke, E. J. & Grunberg, N. E. Sex differences in effects of predictable and unpredictable footshock on fentanyl self-administraƟon in rats. Exp Clin Psychopharmacol 5, 99–106 (1997). 49. Lee, C. W.-S. & Ho, I.-K. Sex differences in opioid analgesia and addicƟon: interacƟons among opioid receptors and estrogen receptors. Mol Pain 9, 45 (2013). 50. Wu, L.-T. & Howard, M. O. Is inhalant use a risk factor for heroin and injecƟon drug use among adolescents in the United States? Addict Behav 32, 265–281 (2007). 51. M, Z. et al. DisƟnct and sex-specific expression of mu opioid receptors in anterior cingulate and somatosensory S1 corƟ cal areas. Pain 164, (2023). 52. Sb, E. & Ma, S. Estradiol and Mu opioid-mediated reward: The role of estrogen receptors in opioid use. AddicƟon neuroscience 9, (2023). 53. Hodgson, S. R., Hofford, R. S., Roberts, K. W ., Wellman, P . J. & Eitan, S. Socially induced morphine pseudosensiƟzaƟon in adolescent mice. Behav Pharmacol 21, 112–120 (2010). 54. ZióŁkowska, B., Gieryk, A., Solecki, W . & PrzewŁocki, R. Temporal and anatomic paƩ erns of immediate-early gene expression in the forebrain of C57BL/6 and DBA/2 mice aŌ er morphine administraƟon. Neuroscience 284, 107–124 (2015). 55. McReynolds, J. R., ChrisƟanson, J. P ., Blacktop, J. M. & Mantsch, J. R. What does the Fos say? Using Fos-based approaches to understand the contribuƟon of stress to substance use disorders. Neurobiol Stress 9, 271–285 (2018). 56. Tyson, A. L. & Margrie, T. W. Mesoscale microscopy and image analysis tools for understanding the brain. Prog Biophys Mol Biol 168, 81–93 (2022). 57. Susaki, E. A. et al. Advanced CUBIC protocols for whole- brain and whole-body clearing and imaging. Nat Protoc 10, 1709–1727 (2015). 58. Ertürk, A. et al. Three-dimensional imaging of solvent- cleared organs using 3DISCO. Nat. Protocols 7, 1983–1995 (2012). 59. Murray, E. et al. Simple, Scalable Proteomic Imaging for High-Dimensional Profiling of Intact Systems. Cell 163, 1500–1514 (2015). 60. LehƟnen, J. et al. Noise2Noise: Learning Image RestoraƟon without Clean Data. Preprint at hƩ ps://doi.org/10.48550/ARXIV.1803.04189 (2018). 61. Tyson, A. L. et al. A deep learning algorithm for 3D cell detecƟon in whole mouse brain image datasets. PLOS ComputaƟonal Biology 17, e1009074 (2021). 62. Eichenberger, B. T., Z han, Y ., Rempfler, M., Giorgeƫ , L. & Chao, J. A. deepBlink: threshold-independent detecƟon and localizaƟon of diffracƟon-limited spots. Nucleic Acids Res 49, 7292–7297 (2021). 63. Ronneberger, O., Fischer, P . & Brox, T. U-Net: ConvoluƟonal Networks for Biomedical Image SegmentaƟon. Preprint at hƩ ps://doi.org/10.48550/ARXIV.1505.04597 (2015). 64. Ester, M., Kriegel, H.-P ., Sander, J. & Xu, X. A Density-Based Algorithm for Discovering Clusters in Large SpaƟal Databases with Noise. 226–331 (1996). 65. He, K., Zhang, X., Ren, S. & Sun, J. Deep Residual Learning for Image RecogniƟon. Preprint at hƩ ps://doi.org/10.48550/ARXIV.1512.03385 (2015). 66. GouƩ e, C. & Gaussier, E. A ProbabilisƟc InterpretaƟon of Precision, Recall and F-Score, with ImplicaƟon for EvaluaƟon. in Advances in InformaƟon Retrieval (eds. Losada, D. E. & Fernández-Luna, J. M.) vol. 3408 345–359 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2005). .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 19/41 67. Howard, J. & Gugger, S. fastai: A Layered API for Deep Learning. (2020) doi:10.48550/ARXIV.2002.04688. 68. Haase, R. haesleinhuepf/apoc. (2025). 69. Breiman, L. Random Forests. Machine Learning 45, 5–32 (2001). 70. Tyson, A. L. et al. Accurate determinaƟon of marker locaƟon within whole-brain microscopy images. Sci Rep 12, 867 (2022). 71. Perez, F. & Granger, B. E. IPython: A System for InteracƟve ScienƟfic CompuƟng. Comput. Sci. Eng. 9, 21–29 (2007). 72. Claudi, F. et al. Visualizing anatomically registered data with brainrender. Elife 10, e65751 (2021). 73. Inc, P . T. CollaboraƟve data science. hƩ ps://plot.ly (2015). 74. Waskom, M. et al. mwaskom/seaborn: v0.8.1 (September 2017). Zenodo hƩ ps://doi.org/10.5281/ZENODO.883859 (2017). 75. Sofroniew, N. et al. napari: a mulƟ-dimensional image viewer for Python. Zenodo hƩ ps://doi.org/10.5281/ZENODO.3555620 (2025). 76. Virtanen, P . et al. SciPy 1.0: Fundamental Algorithms for ScienƟfic CompuƟng in Python. Nature Methods 17, 261– 272 (2020). .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint Supplemental Materials Figure S1. Morphine effect across large brain structures. A cartoon representaƟon of the mouse brain with a heat map which displays the average density fold-change across structures summarized in Table 1. Average morphine cell densiƟ es are calculated across all 15 morphine treated brains in the study. Average saline densiƟ es are calculated across all 12 saline treated brains. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 21/41 Acronym Region Density Ra Ɵo (fold-change) p-value AUDd5 Isocortex 6.05 0.0002 ACAv6a Isocortex 5.87 0.0133 SSs5 Isocortex 5.68 0.0008 ACAv5 Isocortex 4.68 0.0112 VISC5 Isocortex 4.62 0.0039 PPN MB 4.6 0.033 PB P 4.53 0.0025 P-sat P 4.4 0.0352 EPd CTXsp 4.21 0.0018 PL6a Isocortex 4.07 0.0034 VISl5 Isocortex 4.03 0.0032 VISp5 Isocortex 3.99 0.0074 ACAv2/3 Isocortex 3.94 0.027 SSs6a Isocortex 3.88 0.0269 SSp-bfd5 Isocortex 3.75 0.0057 ACAd6a Isocortex 3.69 0.0208 AUDv5 Isocortex 3.68 0.0004 VISC6a Isocortex 3.65 0.022 HATA HPF 3.65 0.0374 AIp5 Isocortex 3.63 0.0024 ACAd5 Isocortex 3.61 0.025 ENTl5 HPF 3.57 0.0003 VISpor5 Isocortex 3.56 0.0008 ENTm6 HPF 3.49 0.001 ENTm5 HPF 3.42 0.0014 VTA MB 3.38 0.0393 PL5 Isocortex 3.23 0.0054 AUDp5 Isocortex 3.21 0.0022 ENTl6a HPF 3.15 0.0032 ECT6a Isocortex 3.11 0.0043 ENTm3 HPF 3.10 0.0068 CA2 HPF 3.01 0.0237 TEa5 Isocortex 2.9 0.0009 CA3 HPF 2.79 0.0028 TEa6a Isocortex 2.63 0.0408 CENT3 CB 2.58 0.0223 ECT5 Isocortex 2.52 0.0089 ICc MB 2.52 0.0092 CENT2 CB 2.49 0.0116 EPv CTXsp 2.4 0.0396 BLAp CTXsp 2.34 0.0282 ILA5 Isocortex 2.24 0.025 MRN MB 2.08 0.0352 CA1 HPF 1.91 0.0131 ProS HPF 1.77 0.029 DG-mo HPF 1.6 0.0419 Table S1. Allen Mouse Brain Atlas structures that exhibited a significant morphine effect. Table summarizing gray maƩ er structures with significant morphine-induced changes in c-Fos expression (p0.25 mm 3 volume, to reduce effect of registra Ɵon imperfec Ɵons) and had at least 25 spots detected on average in both morphine and saline brains (to ensure reliable c-Fos puncta idenƟficaƟon). 46 structures saƟsfied these condiƟons. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 22/41 Acronym Region Density RaƟo ( fold- change) p-value CLA CTXsp 9.32 0.071 AIp6a Isocortex 5.82 0.066 AId6a Isocortex 4.90 0.051 GU5 Isocortex 4.24 0.057 RSPagl6a Isocortex 4.14 0.066 MV MY 3.99 0.097 MOp5 Isocortex 3.93 0.087 MOs6a Isocortex 3.72 0.089 LPO HY 3.70 0.053 MY-sen MY 3.51 0.068 P-mot P 3.44 0.071 RAmb MB 3.37 0.06 SSp-m5 Isocortex 3.36 0.085 VISp6a Isocortex 3.32 0.057 ORBl6a Isocortex 3.15 0.091 SSp-bfd6a Isocortex 2.87 0.06 MPN HY 2.85 0.093 MPO HY 2.77 0.065 AHN HY 2.69 0.06 AAA STR 2.57 0.073 PL2/3 Isocortex 2.5 0.082 AIv5 Isocortex 2.45 0.055 BLAv CTXsp 2.42 0.065 SI PAL 2.41 0.077 SCiw MB 2.39 0.058 ACB STR 2.34 0.072 VISli Isocortex 2.31 0.081 ACAv1 Isocortex 2.29 0.061 AUDp Isocortex 2.11 0.098 LHA HY 2.08 0.055 PAG MB 1.89 0.067 Table S2. Allen Mouse Brain Atlas structures that trended towards significant morphine-induced effects. All CCFv3 structures that had morphine effects approaching significance (0.05 <= p 0.25 mm3 volume, to reduce effect of registraƟon imperfecƟons and had at least 25 spots detected on average in both morphine and saline brains. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 23/41 Comparison Morphine only Saline only All morphine vs All saline Medulla: AMBd, AMBv, LIN, LRNp, PAS, VI, y, RO, AP , GR, ECU, Pa5 Pons: PC5, SG, V, SLC Hypothalamus: ME, SFO, ASO Thalamus: RH Midbrain: DT, III, LT, MA3, Su3, PN, IPA, IPDM Isocortex: ACAd6b, AId6b, GU6b, ORBm6b, ORBvl6b, PL6b, SSp-ll6b, SSp-n6b, SSp-tr6b, SSp-ul6b, SSpun6b Medulla: RPA Midbrain: SCO Morphine 1h vs saline 1h Medulla: AMB, AMBd, AMBv, IO, LIN, LRN, LRNm, LRNp, PGRNl, PPY , XII, x, MY-sat, RM, VCO, GR, ECU, Pa5 Pons: PC5, SUT, V, LC, NI, RPO, SLC, SLD, KF Hypothalamus: MMd, ASO Thalamus: IAM, IntG, PCN, PR, SMT, VPLpc Midbrain: LT, MA3, MT, Su3, PN, MEV, IPDM, IPRL Isocortex: AIv6b, ORBl6b, PL6b, SSp-ll6b, SSp-tr6b, SSp-ul6b, SSpun6b Medulla: PAS Hypothalamus: OV, VMPO Striatum: BA Morphine 4h vs saline 4h Medulla: PAS, NR, VI, RO, AP , GR, ECU Pons: P5, PC5, SG, V, SLC Hypothalamus: ME, TMd, OV, SCH, SFO, VLPO, VMPO, ARH, ASO, PVa Midbrain: DT, III, LT, MA3, Su3, PN, CLI, IPA, IPDM, IPI Pallidum: BAC Striatum: BA Isocortex: ACAd6b, AId6b, AUDd6b, GU6b, ILA6b, MOp6b, MOs6b, ORBm6b, ORBvl6b, PL6b, VISa6b, VISrl6b, RSPd6b, RSPv6b, SSp-ll6b, SSp-m6b, SSp-n6b, SSp- tr6b, SSpul6b, SSp-un6b, VISpl6b Medulla: LIN, RPA Thalamus: IGL, LGd-ip, LGd-sh Midbrain: SCO Isocortex: VISam6b, VISpm6b .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 24/41 Morphine male vs saline male Medulla: AMB, AMBd, AMBv, LIN, NR, VI, XII, x, RO, AP , GR, ECU, Pa5 Pons: PC5, V, RPO, SLC, SLD, KF Hypothalamus: RCH, ME, TMd, AVP , AVPV, MEPO, OV, SCH, SFO, VLPO, VMPO, ASO, PVa Thalamus: IntG, PR, VPLpc, VPMpc Midbrain: LT, MA3, Su3, PN, MEV, CLI, IPA, IPDM Pallidum: BAC Striatum: BA Isocortex: ACAd6b, ACAv6b, AId6b, AIp6b, AIv6b, AUDd6b, ORBl6b, ORBm6b, PL6b, SSp-ll6b, SSp-tr6b, SSp-ul6b, SSp-un6b Olfactory areas: NLOT1 Medulla: RPA Midbrain: SCO Morphine female vs saline female Medulla: ICB, PAS, PRP , AP , GR, ECU Pons: P5, SG, V, SLC Hypothalamus: ME, SFO Thalamus: RH Midbrain: III, LT, INC, Su3, IPA, IPDL, IPDM, IPRL Isocortex: GU6b, ILA6b, MOs6b, ORBvl6b, VISa1, VISrl1, VISrl6b, SSp-ll6b, SSp-m6b, SSp-n6b, SSp- tr1, SSp-tr6b, SSp-un6b, VISal6b Medulla: AMB, AMBv, PPY , RPA Pons: PC5 Thalamus: LGd-ip Midbrain: PBG, SCO Cerebellum: LING Striatum: SH, BA Isocortex: VISam6b Comparison 1h only 4h only Morphine 1h vs morphine 4h Medulla: AMBd, ICB, LIN, y, Pa5 Thalamus: IGL, IntG, LGd-ip, LGdsh Isocortex: AIv6b, VISam6b, VISpm6b Medulla: PAS, NR, VI, RO, AP Hypothalamus: OV, SFO, VMPO Thalamus: RH Striatum: BA Isocortex: ORBm6b Comparison male only female only Morphine male vs morphine female Medulla: AMB, AMBd, AMBv, LRNp, NR, PPY , VI, y, RO, Pa5 Pons: PC5 Thalamus: IGL, IntG, LGd-ip, LGdsh Midbrain: DT, PBG Cerebellum: LING Striatum: SH, BA Isocortex: AIv6b, ORBm6b, RSPv6b, VISam6b, VISpm6b Thalamus: RH Table S3. Binary Effects: Regions of c-Fos acƟ vity in the brain which are only present in one experimental group. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 25/41 Acronym Density RaƟo p-value PB 11.11 0.0444 EPd 5.27 0.0235 ENTl5 4.37 0.008 ENTm3 3.94 0.0409 ENTm5 3.81 0.0347 VISpor5 3.51 0.0413 AUDd5 3.48 0.0197 SSs5 3.42 0.0365 VISp5 3.40 0.031 ENTm6 3.14 0.0374 MRN 3.08 0.0462 CA3 2.61 0.0281 Table S4. Structures with significant differences in c- Fos+ cell densiƟes (morphine vs saline) at 1h (sorted by decreasing density fold- change) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 26/41 Acronym Density RaƟo p-value PL6a 15.16 0.0052 ACAv5 13.52 0.0328 SSs5 10.87 0.0185 VISl5 7.99 0.0369 VISC5 7.85 0.0033 SSs6a 7.41 0.0373 AUDp5 7.29 0.0002 GU5 6.24 0.039 AUDv5 5.97 0.0004 DMH 5.87 0.0402 ACAd 5.66 0.0449 PL5 5.42 0.0063 ACB 5.06 0.0356 VISC6a 4.74 0.0444 AUDd 4.70 0.0343 AIp5 4.19 0.0334 VISp5 3.77 0.0122 ENTm6 3.60 0.0151 TEa5 3.48 0.0018 TEa6a 3.43 0.0025 VISpor5 3.43 0.0089 VISp6a 3.41 0.0099 PB 3.37 0.0298 ENTl6a 3.33 0.0246 CENT2 3.26 0.0053 ECT6a 3.16 0.0075 VISli 2.97 0.0368 EPd 2.92 0.0308 ECT5 2.90 0.0149 ENTm5 2.89 0.0253 ENTl5 2.64 0.0194 AUDpo 2.64 0.0494 ILA5 2.58 0.0397 ICc 2.23 0.0306 Table S5. Structures with significant differences in c-Fos+ cell densiƟ es (morphine vs saline) at 4h (sorted by decreasing fold-change) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 27/41 Acronym Density RaƟo p-value ACB 15.6 0.0098 ACAv5 10.29 0.0486 LA 9.47 0.0247 PAL 8.49 0.035 EPd 7.83 0.0089 BLAp 7.72 0.0176 AUDd5 7 0.0064 PL6a 6.91 0.0295 ACAd6a 6.86 0.045 VISl5 5.95 0.0273 ECT6a 5.82 0.0141 PL5 5.77 0.0363 SSs5 5.52 0.0054 BMAp 5.4 0.0255 ENTl5 5.15 0.0121 LHA 5.14 0.0314 PB 5.04 0.0303 TEa6a 5 0.05 ENTm5 4.86 0.0153 ENTm6 4.83 0.0098 AIp5 4.72 0.027 PA 4.54 0.0156 VISC5 4.36 0.0412 ENTl6a 4.25 0.0256 VISpor5 3.48 0.0491 AUDp5 3.14 0.0385 AUDv5 3.14 0.0322 CENT2 2.96 0.0413 CA3 2.53 0.0341 ProS 2.32 0.0162 CA1 2.15 0.0399 SUB 1.91 0.0431 Table S6. Male-specific morphine effects. Structures with significant differences in c-Fos+ cell densiƟes (morphine/saline) in male mice (sorted by decreasing fold-change) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 28/41 Acronym Density RaƟo p-value TEa4 6.87 0.0003 AUDp2/3 6.53 0.0027 ENTm1 5.36 0.0301 RSPagl5 4.84 0.0069 SSs5 4.67 0.0232 VISa 4.62 0.0038 AUDd5 4.57 0.0075 AUDv5 4.41 0.0010 VISC5 4.39 0.0206 TEa2/3 4.32 0.0125 SSp-bfd5 4.16 0.0349 SSp-tr 3.73 0.0413 CA2 3.71 0.0271 PB 3.65 0.0431 VISpor5 3.6 0.0003 PERI2/3 3.53 0.0399 ICc 3.52 0.0187 ECT5 3.47 0.0077 ECT2/3 3.38 0.036 ENTm3 3.31 0.0376 TEa5 3.26 0.0009 CA3 3.19 0.0454 VISrl 3.13 0.0368 AUDp5 3.1 0.0281 VISp5 2.86 0.0359 AIv5 2.8 0.0124 ICe 2.76 0.0332 ENTl5 2.7 0.0127 ENTl3 2.57 0.022 RSPagl 2.47 0.0411 RSPd5 2.43 0.038 AUDpo 2.31 0.0382 Table S7. Female-specific morphine effects. Structures with significant differences in c-Fos+ cell densiƟes (morphine/saline) in female mice (sorted by decreasing fold-change) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 29/41 acronym all male female 1h 4h 1h male 1h female 4h male 4h female FRP1 * FRP2/3 FRP5 FRP6a * FRP6b bin bin bin bin bin bin bin bin bin MOp1 MOp2/3 MOp5 MOp6a MOp6b bin MOs1 ** MOs2/3 MOs5 MOs6a * MOs6b bin SSp-n1 SSp-n2/3 * SSp-n4 SSp-n5 * SSp-n6a * SSp-n6b bin bin bin SSp-bfd1 * SSp-bfd2/3 * SSp-bfd4 SSp-bfd5 ** * * SSp-bfd6a **** SSp-bfd6b * SSp-ll1 SSp-ll2/3 SSp-ll4 * SSp-ll5 * * * SSp-ll6a SSp-ll6b bin bin SSp-m1 SSp-m2/3 SSp-m4 SSp-m5 SSp-m6a SSp-m6b SSp-ul1 ** SSp-ul2/3 SSp-ul4 SSp-ul5 SSp-ul6a SSp-ul6b SSp-tr1 bin SSp-tr2/3 SSp-tr4 SSp-tr5 * * SSp-tr6a * SSp-tr6b SSp-un1 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 30/41 SSp-un2/3 SSp-un4 SSp-un5 * * SSp-un6a SSp-un6b bin SSs1 SSs2/3 SSs4 SSs5 *** ** * * * * * SSs6a * * ** SSs6b GU1 GU2/3 GU4 GU5 * * GU6a * GU6b bin VISC1 VISC2/3 * VISC4 * VISC5 ** * * ** * VISC6a * * * VISC6b AUDd1 AUDd2/3 ** * AUDd4 * AUDd5 *** ** ** * ** * * AUDd6a * AUDd6b AUDp1 * AUDp2/3 ** AUDp4 ** * * AUDp5 ** * * *** * ** AUDp6a * AUDp6b AUDpo1 AUDpo2/3 * ** AUDpo4 * AUDpo5 ** * *** ** AUDpo6a AUDpo6b ** * * ** AUDv1 AUDv2/3 * AUDv4 ** * * * AUDv5 *** * *** *** ** AUDv6a * ** AUDv6b VISal1 VISal2/3 VISal4 * VISal5 * * * VISal6a * VISal6b * bin VISam1 bin VISam2/3 * * .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 31/41 VISam4 ** bin VISam5 VISam6a VISam6b * VISl1 VISl2/3 VISl4 * VISl5 ** * * ** VISl6a * * VISl6b VISp1 VISp2/3 VISp4 VISp5 ** * * * * VISp6a ** * VISp6b VISpl1 VISpl2/3 VISpl4 * * VISpl5 ** * * VISpl6a ** ** *** VISpl6b * * VISpm1 bin VISpm2/3 * VISpm4 * VISpm5 VISpm6a VISpm6b VISli1 * VISli2/3 VISli4 VISli5 *** * * * * * ** VISli6a * VISli6b VISpor1 VISpor2/3 VISpor4 * * * VISpor5 *** * *** * ** ** VISpor6a ** * * * * VISpor6b * * * * * ACAd1 ACAd2/3 * ACAd5 * * ACAd6a * * * * ACAd6b bin bin bin ACAv1 * ACAv2/3 * ACAv5 * * * ** ACAv6a * * * ** ACAv6b * * PL1 PL2/3 PL5 ** * ** * PL6a ** * ** ** PL6b bin .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 32/41 ILA1 ** ILA2/3 ILA5 * * ILA6a * * ILA6b bin ORBl1 * ORBl2/3 * ORBl5 ORBl6a ORBl6b bin ORBm1 ORBm2/3 ORBm5 ORBm6a * * * * ORBm6b bin bin bin ORBvl1 ORBvl2/3 ** ORBvl5 * ORBvl6a * * ORBvl6b bin bin bin AId1 AId2/3 AId5 * AId6a ** AId6b AIp1 AIp2/3 AIp5 ** * * AIp6a * AIp6b AIv1 AIv2/3 AIv5 * * AIv6a * * * ** AIv6b bin bin bin bin bin bin bin bin bin RSPagl1 RSPagl2/3 RSPagl5 ** * * RSPagl6a * RSPagl6b RSPd1 RSPd2/3 RSPd4 bin bin bin bin bin bin bin bin bin RSPd5 * RSPd6a * RSPd6b bin RSPv1 RSPv2/3 RSPv5 RSPv6a * RSPv6b bin VISa1 bin VISa2/3 *** ** * VISa4 ** * VISa5 * .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 33/41 VISa6a * VISa6b * VISrl1 ** VISrl2/3 VISrl4 * VISrl5 * * VISrl6a * VISrl6b TEa1 TEa2/3 * * TEa4 *** * ** TEa5 *** *** ** ** TEa6a * * ** **** TEa6b * * ** PERI1 PERI2/3 * PERI5 ** ** ** PERI6a ** * * * PERI6b * * * ECT1 * * ECT2/3 * ECT5 ** ** * ECT6a ** * ** * ECT6b ** * * * MOB AOBgl AOBgr AOBmi * AON TTd TTv DP PIR NLOT1 bin NLOT2 * bin NLOT3 bin COAa bin COApl COApm PAA * TR CA1 * * * CA2 * * CA3 ** * * * DG-mo * * DG-po DG-sg FC IG ENTl1 ENTl2 ENTl3 * ENTl5 *** * * ** * * ENTl6a ** * * ENTm1 * * .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 34/41 ENTm2 ENTm3 ** * * ENTm5 ** * * * ENTm6 ** ** * * * PAR POST PRE SUB * * * ProS * * *** HATA * APr CLA EPd ** ** * * ** EPv * LA * BLAa BLAp * * * BLAv bin BMAa BMAp * PA * * CP ACB ** * FS OT LSc LSr LSv SF SH bin AAA * BA bin bin CEAc CEAl CEAm IA MEA * GPe GPi SI MA MS * bin NDB TRS BST BAC bin bin bin VAL VM VPL **** VPLpc bin bin bin VPM VPMpc PoT * SPFm bin SPFp .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 35/41 SPA bin PP ** * ** MGd MGv ** MGm * LGd-sh LGd-co LGd-ip LP PO POL * SGN Eth bin AV AMd AMv AD IAM bin IAD LD IMD bin MD SMT bin PR bin PVT PT RE bin Xi bin RH bin bin bin CM bin PCN CL PF PIL RT IGL IntG bin bin bin bin LGv SubG bin bin bin bin bin MH LH SO ASO * bin PVH bin PVa bin bin bin PVi ARH ADP AVP bin AVPV bin DMH * * * MEPO MPO OV bin PD bin .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 36/41 PS bin PVp PVpo SBPV bin SCH bin bin bin bin bin SFO bin bin bin bin bin VMPO bin bin bin VLPO bin AHN LM * MMme MMl MMm MMp bin bin bin bin bin bin bin bin bin MMd SUM TMd TMv bin ** MPN PMd PMv PVHd VMH PH LHA * LPO PST PSTN PeF * RCH bin STN TU FF ME bin bin bin SCop SCsg SCzo ICc ** * * * ICd ICe * NB SAG PBG MEV * bin SCO bin SNr VTA * PN RR ** * * bin MRN * * SCdg SCdw SCiw * SCig PRC * .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 37/41 INC bin ND Su3 bin APN MPT NOT NPC ** OP PPT RPF * CUN * RN III bin MA3 bin EW bin IV bin bin bin bin bin bin bin bin bin Pa4 bin bin bin bin bin VTN bin bin bin bin AT bin bin bin bin bin bin LT bin bin bin bin bin bin bin bin bin DT MT * SNc PPN * IF bin IPR IPC bin IPA bin IPL IPI bin IPDM bin bin bin bin bin bin bin bin bin IPDL IPRL bin RL CLI bin bin DR ** * * * * NLL PSV KF bin POR bin SOCm bin SOCl bin B bin DTN PDTg bin PCG * PG PRNc SG bin SUT ** * bin ** bin bin TRN V bin P5 * * bin Acs5 bin bin bin bin bin PC5 bin bin bin .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 38/41 I5 bin bin bin bin bin bin bin bin bin CS LC * LDT * ** * NI bin PRNr RPO bin bin bin SLC * bin bin bin bin bin SLD ** * bin bin AP bin bin bin bin bin DCO bin VCO bin CU bin GR bin bin bin ECU bin NTB bin NTS bin bin SPVC bin SPVI bin SPVO bin Pa5 bin bin bin bin bin VI bin bin bin bin bin VII bin ACVII bin bin bin bin bin AMBd bin bin bin bin bin AMBv bin bin bin bin bin bin DMX bin bin GRN ICB bin bin IO bin bin bin IRN bin ISN bin bin bin bin bin bin LIN bin bin LRNm bin LRNp bin MARN bin MDRNd bin MDRNv bin bin PARN bin PAS bin bin PGRNd bin bin PGRNl bin NR bin bin bin bin bin PRP bin PPY bin LAV bin MV bin SPIV bin SUV bin x bin XII bin bin y bin bin bin bin bin RM bin RPA bin bin bin RO bin bin bin bin .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 39/41 LING bin bin CENT2 * * ** * CENT3 * DEC FOTU bin PYR bin UVU NOD bin SIM ANcr1 ANcr2 PRM COPY PFL FL bin FN bin IP bin DN VeCB bin * Table S8. Significant effects of morphine on all leaf structures within the CCFv3 tree containing grey maƩ er. In each column, significance was calculated between respecƟve morphine and saline subgroups. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 40/41 Figure S3. Binary Effects. Cartoon visualiza Ɵon of brain regions with c -Fos acƟvity in the brain which are only present in one experimental group. a) red = morphine only, blue = saline only; b) leŌ : red = male only, blue = female only; middle, right: red = morphine only, blue = saline only; c) leŌ : red = 1h only, blue = 4h only; middle, right: red = morphine only, blue = saline only. .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint 41/41 Figure S4. Temporally and sexually dimorphic acƟ vaƟ on paƩ erns aŌ er morphine exposure: Maximum intensity projecƟons of c-Fos puncta locaƟons in the CCFv3 for morphine brains normalized (divided) by corresponding saline controls (smoothed with Gaussian filter, kernel size=10) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 24, 2025. ; https://doi.org/10.1101/2025.02.19.638902doi: bioRxiv preprint