Animals
All procedures were performed in accordance with guidelines set in place by National Institutes of Health and approved by the Stanford University Institutional Care and Use Committee.
Both male and female mice were used equally in all experiments and were randomized to treatment conditions. Hemizygous Dat-Cre (the Jackson Laboratory, 006660) mice were bred with C57BL/6 J (the Jackson Laboratory, 000664). Optogenetic experiments were performed on animals hemizygous for Dat–Cre or homozygous for Vgat–Cre (the Jackson Laboratory, 028862). For oligodendrocyte specific expression of mGFP, homozygous mT/mGlox/lox (the Jackson Laboratory, 007676) animals were crossed with Plp1–CreERT (the Jackson Laboratory, 005975). For conditional deletion of Myrf, hemizygous Pdgfra–CreERTM (the Jackson Laboratory, 018280) mice were bred with homozygous Myrflox/lox mice48 (the Jackson Laboratory, 010607) to generate hemizygous Pdgfra–CreERTM and heterozygous Myrflox/+, which were backcrossed to homozygous Myrflox/lox to achieve hemizygous Pdgfra–CreERTM and homozygous MyrfOPC−/− (Myrflox/lox; Pdgfra–CreERTM). Littermates that lacked Pdgfra–CreERTM were used as control animals. For conditional deletion of TrkB, hemizygous Pdgfra–CreERTM (the Jackson Laboratory, 018280) mice were bred with homozygous TrkBlox/lox (MMRC, 033048-UCD) to generate hemizygous Pdgfra–CreERTM and heterozygous TrkBlox/+, which were backcrossed to homozygous TrkBlox/lox to achieve hemizygous Pdgfra-CreERTM and homozygous TrkBOPC−/− (TrkBlox/lox; Pdgfra–CreERTM), which was referred to as OPC-TrkB cKO in a previous publication5. Littermates that lack Pdgfra–CreERTM were used as control animals. For conditional deletion of Oprk, hemizygous Pdgfra-CreERTM (the Jackson Laboratory, 018280) mice were bred with homozygous Oprk1lox/lox (the Jackson Laboratory, 030076) to generate hemizygous Pdgfra-CreERTM and heterozygous Oprk1lox/+, which were backcrossed to homozygous Oprklox/lox to achieve hemizygous Pdgfra-CreERTM and homozygous Oprk1lox/lox (Oprk1lox/lox; Pdgfra-CreERTM). Littermates that lack Pdgfra-CreERTM were used as control animals. To initiate Cre-dependent deletion of Myrf, TrkB or Oprk, animals were intraperitoneally injected with 100 mg kg−1 tamoxifen (Sigma-Aldrich) for 5 days at 7 weeks or 4 weeks of age. For mGFP expression mT/mG animals were injected with the same concentration of tamoxifen for 3 days at 4 weeks of age. For testing effects of drugs on oligodendroglial cells, wild-type C57BL/6 J (the Jackson Laboratory, 000664) mice were used. All animals were housed in a 12 h light/dark cycle with unrestricted access to food and water. Behavioural experiments were performed during the same circadian period (07:00–19:00). All mouse experiments were repeated in at least two independent experiments conducted at different times. Sample sizes were guided by power calculations.
Viral vectors
For optogenetics experiments, Dat-Cre-dependent expression of channelrhodopsin was achieved by injection of AAV-DJ-EF1a-DIO-hChR2(H134R)::eYFP (virus titre 1.3 × 1012 genome copies per millilitre), Vgat-Cre-dependent expression of halorhodopsin was achieved by injection of AAV-DJ-EF1a-DIO-NpHR3.0::eYFP (virus titre 1 × 1012 genome copies per millilitre) and eYFP control by injection of AAV-DJ-EF1a-DIO-eYFP (virus titre 1.6 × 1012 genome copies per millilitre). All viral vectors for optogenetics experiments were obtained from Stanford University Gene Vector and Virus Core. For fibre photometry experiments, GrabDA expression was achieved by injection of AAV9-hSyn-DA2m (DA4.4) (virus titre, 1.64 × 1013 genome copies per millilitre) obtained from WZ Biosciences49.
Surgical procedures
Animals were anaesthetized with 1–4% isoflurane and placed in a stereotaxic apparatus. For optogenetic stimulation experiments, 1 µl of AAV-ChR2::eYFP or AAV-eYFP viral vectors were unilaterally injected using Hamilton Neurosyringe and Stoelting stereotaxic injector over 5 min. Coordinates for viral injections and optic ferrule placements were measured from bregma. Viral vectors were injected into VTA at coordinates, anterior–posterior (AP) = −2.8 mm, mediolateral (ML) = −0.3 mm, dorsoventral (DV) = −4.4 mm and an optic ferrule was placed at AP = −2.8 mm, ML = −0.3 mm, DV = −4.00 mm. For NAc stimulations an optic ferrule was placed at AP = +1.25 mm, ML = −0.75 mm, DV = −4.00 mm.
Optogenetic stimulations
Optogenetic stimulations were performed at least 3 weeks after the viral vector delivery and 1 week after optic ferrule implantation. Freely moving animals were connected to a 473 nm diode-pumped solid-state laser system with a monofibre patch cord. Phasic DA neuron stimulation, for both neuronal cell bodies and axon terminals, was performed at 30 Hz, with eight 5 ms pulses of 473 nm light delivery every 5 s at a light power output of 15 mW (about 475 mW mm−2) from the tip of the optic fibre (200 µm core diameter, numerical aperture = 0.22, Doric Lenses)31. Tonic DA neuron stimulation was performed at 1 Hz, with 24 15 ms pulses of 473 nm light delivery every 1 min (ref. 30). Acute optogenetic stimulation session lasted for 30 min for phasic or tonic stimulations. Animals were injected intraperitoneally with 40 mg kg−1 of EdU (Invitrogen, E10187) before and after the session and perfused 3 h after the start of the stimulation. Chronic phasic optogenetic stimulations were performed for 10 min a day, for 7 days, with EdU injections for all stimulation sessions. Animals were perfused 4 weeks after the cessation of stimulations. For inhibition experiments, freely moving animals were connected to a 595 nm high-power LED system with a monofibre patch cord. NpHR inhibition of GABAergic neuron cell bodies was performed with constant 595 nm light delivery at 1 Hz, for 8 s every 10 s at light power output of 15 mW (about 475 mW mm−2) from the tip of the optic fibre.
Fibre photometry
AAV9-hSyn-DA2m (DA4.4) was injected into NAc (AP = +1.25 mm, ML = −0.75 mm, DV = −4.40 mm) and an optic fibre (400 µm core diameter, NA = 0.48, Doric Lenses) was placed just above the injection site (AP = +1.25 mm, ML = −0.75 mm, DV = −4.30 mm). Mice were administered with tamoxifen 1 day to 4 weeks before starting CPP. After allowing 3 to 4 weeks for viral expression, GrabDA signal was recorded during the pre-test while the mice explored each chamber of the three-chamber CPP apparatus for 30 min. Mice then went through morphine conditioning for 5 days. The GrabDA signal was then recorded again during post-test exploration for 30 min, in an identical manner to the pre-test.
Fibre photometry data were acquired with Synapse software controlling an RZ5P lock-in amplifier (Tucker-Davis Technologies). GrabDA was excited by frequency-modulated 465 and 405 nm LEDs (Doric Lenses). Optical signals were band-pass filtered with a fluorescence mini cube FMC4 (Doric Lenses) and signals were digitized at 6 kHz. Signal processing was performed with custom scripts in MATLAB (MathWorks). Briefly, signals were debleached by fitting with a mono-exponential decay function and the resulting fluorescence traces were Z-scored. Videos were analysed for the first entry into the morphine conditioning chamber for each animal for the pre-test and post-test. Peristimulus time histograms were constructed by taking the average of 20 s epochs of fluorescence consisting of 10 s before and 10 s after the chamber entry, which is defined as time = 0. Before averaging, each epoch was offset such that the Z-score averaged from −10 to −1 s equalled 0. Area under the curve (AUC) was defined as the integral between 0 and +10 s. The AUC difference score was calculated as the (post-test AUC − pre-test AUC) for each animal.
Behavioural analysis
Real-time place preference
Real-time place preference was performed in a two-chamber acrylic box (each chamber 45 × 20 x 25 cm3) without any more contextual cues. Each mouse was connected to a 473 nm laser system with a monofibre patch cord and gently placed into the middle section of the cage at the beginning of the test. One chamber of the cage was randomly assigned for optogenetic stimulation before the test. Optogenetic stimulation was turned on every time a mouse entered into the pre-assigned chamber and turned off if the mouse moved to the other chamber. Each session lasted for 20 min. Ethovision XT software (Noldus) was used to determine the time animals spent in each chamber after the test in an automated and condition-blinded manner.
Conditioned place preference
The experimental protocol was adapted from the previously described method35. CPP was performed in an acrylic rectangular three-chamber apparatus with two distinct chambers which are connected by a neutral corridor. The ‘grid chamber’ included a 3-mm-wide grid in the floor texture and black stripes over white walls. The ‘spots chamber’ included a floor with 10-mm-diameter holes and white walls. Both chambers were cubes (18 ×18 × 18 cm3) and connected by a 10-cm-wide corridor (10 × 18 × 18 cm3). The entrance to each grid or spots chamber could be closed off with a transparent acrylic divider. On day 1, mice were tested for baseline preference towards either chamber by placing them in the cage. After 30 min of free exploration between chambers with two different kinds of flooring, mice were returned to their home cage. Mice that show strong preference (more than 75%) to either chamber during pre-test were excluded from further analysis. On the morning of day 2 (conditioning day 1), all mice were administered with saline intraperitoneally and placed in the non-conditioning chamber with either kind of flooring for 15 min and returned to their home cage. The morning session was followed by conditioning afternoon session after 4 h. In the afternoon session, the first group of mice received saline and EdU (40 mg kg−1) injections intraperitoneally and were placed in the conditioning chamber. A second group of mice were administered with 10 mg kg−1 of morphine and EdU (40 mg kg−1) intraperitoneally and placed in the conditioning chambers for 30 min. A third group received 20 mg kg−1 of morphine and EdU (40 mg kg−1) intraperitoneally and were placed in the conditioning chambers for 30 min. A fourth group of mice received 15 mg kg−1 of cocaine and EdU (40 mg kg−1) intraperitoneally and were placed in the conditioning chambers for 20 min. For naloxone experiments, animals were injected with 5 mg kg−1 of naloxone (Tocris-0599) during conditioning sessions. These conditioning sessions were repeated on the following days 3, 4, 5 and 6 (total conditioning days 1–5). On day 7, mice were again allowed to freely move around the cage for 30 min during post-test. Food CPP was performed in a similar manner to drug CPP. Food reward group and food restricted control group were food restricted down to 85% body weight 48 h before the start of CPP and maintained at that weight for the duration of the test. Mice were habituated to the chocolate sucrose pellets in the home cage during the two nights preceding CPP. All animal groups were allowed 30 min of free exploration of the apparatus during a pre-test. Both the spots and grid chambers contained empty, identical weighing boats taped to the floor in the back corner. On the morning of day 2 (conditioning day 1), mice were placed into the non-conditioning chamber with an empty weighing boat for 30 min. No saline injections were administered. In the afternoon, mice were administered EdU (40 mg kg−1) intraperitoneally and placed in the conditioning chamber with a weighing boat full of chocolate sucrose pellets for food reward group and empty weighing boats for food restricted control group for 30 min. Food reward mice were allowed to freely consume sucrose pellets throughout the conditioning session. These conditioning sessions were repeated over the next 4 days. On day 7, mice were allowed to freely move around the apparatus during a 30 min post-test, in which both the spots and grid chambers contained fresh weighing boats with no sucrose pellets. All drug conditioning sessions were recorded for capturing locomotion data. However, the initial CPP experiments which are used for the histological data were performed during the Covid-19 pandemic lockdown conditions, so we were not able to record the conditioning sessions of most experimental groups. Afterwards we ran more CPP assays to record the locomotion data and CPP preference, which resulted in different numbers of animals across these analysis in Fig. 3. The time spent on each chamber with either type of grid and the distance travelled by each mouse were analysed using Ethovision XT software (Noldus) in an automated and condition-blinded manner.
Novel-object recognition
The experimental protocol was adapted from the previously described method50. Animals were handled daily for the week leading up to the test for 5 min each day and habituated to the experimental room. Mice were placed in the acrylic experimental cage (50 × 50 × 50 cm3) to acclimatize for 10 min on the day before testing. On the day of the testing, mice were tested for anxiety by placing into the experimental cage for 10 min and video recorded. If an animal spent less than 2 min in the centre of the box (10 cm away from the walls), it was regarded as too anxious for the test and discarded from analysis. Mice were then placed in the home cage for 5 min. For the training phase, mice were placed in the experimental chamber with two identical inanimate objects (about 5 cm in size). Each time the mouse was placed in the experimental chamber it was facing away from the objects towards the opaque walls. The mouse was allowed to explore these objects for 10 min, then was returned to its home cage. Both the experimental chamber and the objects were cleaned using 70% ethanol. For the novel object testing phase, which is performed 24 h after the training phase, the mouse was returned to the cage to explore these objects for 10 min. One of these objects was returned to the experimental cage and a novel inanimate object of similar size was placed into the cage. The objects used as novel and familiar and the position of the novel object were counterbalanced from trial to trial, animal to animal. All the objects used were initially tested to ensure there was no bias or preference for the animals. All sessions were camera recorded and analysed using Ethovision XT software (Noldus) in an automated and condition-blinded manner. Any exploratory head gesturing within 2 cm of the object, including sniffing and biting, was considered as investigation but climbing onto the object was not considered. Only animals that explored the objects for a minimum of 20 s were included in the analysis.
Sucrose preference
The experimental protocol was adapted from the previously described method51. Mice were tested for sucrose preference 4 weeks after the induction of Myrf deletion by tamoxifen administration. On the first day, animals were habituated individually in experimental cages and to drinking bottles with water. Next day, one water bottle and one sucrose solution bottle (1% wt/vol) were provided for ad libitum access to mice. Water and sucrose solution were replenished every 24 h and consumption was determined per day over 3 days. Positions of the sucrose and water bottles were swapped daily to prevent a bias towards bottle location. All measurements were corrected for spillage by subtracting the volume loss from a control bottle.
Three-chamber social interaction
Mice were tested for social preference 4 weeks after the induction of Myrf deletion by tamoxifen administration. Animals were habituated to the experimental room and the three-chamber experimental cage before test. The three-chamber acrylic experimental cage (each chamber 45 × 20 × 25 cm3) contained two metal grid pencil cups (10 cm in diameter) in each of the outer chambers. The chambers were divided with transparent acrylic walls with 15-cm-wide entrance holes. One of the chambers contained a novel same-sex juvenile in the pencil cup and the other chamber contained an inanimate object in the pencil cup. The test mouse was placed into the middle chamber and 1 min after the chamber dividers were removed and the mouse was allowed to explore each chamber freely for 20 min. The locations of the novel mouse and the inanimate object were counterbalanced between sessions. The time spent on interacting with either pencil cup was analysed using Ethovision XT software (Noldus) in an automated and condition-blinded manner.
Brain tissue processing
Mice (saline control, n = 3; 10 mg kg−1 of morphine, n = 3) were subjected to CPP test as described above. On day 7, mice were again allowed to move freely around the cage for 30 min and time spent on each chamber with either type of grid was analysed. After the test, the mice were transcardially perfused with 20 ml of perfusion buffer (110 mM NaCl, 10 mM HEPES, 25 mM glucose, 75 mM sucrose, 7.5 mM MgCl2, 2.5 mM KCl, 5 µg ml−1 of actinomycin D and 10 µM triptolide in UltraPure DNase/RNase-free distilled water). The brains were promptly collected, flash frozen in isopentane on dry ice and stored at −80 °C. Frozen brains were sectioned inside a cryostat until AP = −2.8 mm and VTA from both hemispheres were punched out using a 1.75 mm biopsy punch with about 1 mm depth. Punches were stored at −80 °C for processing.
Isolation and sorting of nuclei
Punches for each mouse were added to a Wheaton Dounce homogenizer containing 1 ml of ice-cold nuclei isolation medium (10 mM Tris pH 8.0, 250 mM sucrose, 25 mM KCl, 5 mM MgCl2, 0.1% Triton X-100, 1% RNasin Plus, 1× protease inhibitor, 0.1 mM DTT, 5 µg ml−1 of actinomycin D, 10 µM triptolide and anisomycin in UltraPure DNase/RNase-free distilled water). Tissues were dissociated by ten strokes with the loose Dounce pestle followed by ten strokes with the tight pestle. Homogenates were passed through a 40 μm cell strainer and centrifugated at 900g for 15 min to pellet the nuclei. Nuclei were resuspended in 1 ml of resuspension buffer (1× phosphate buffer saline (PBS), 1% nuclease free BSA and 0.5% RNasin Plus) and centrifuged at 900g for another 15 min to pellet the nuclei. Supernatant was removed and nuclei were then stained by adding 500 µl of resuspension buffer containing 0.1 µg ml−1 of DAPI and transferred to FACS tubes. Nuclei were then FACS sorted using a Sony MA900 sorter with a 70 µm chip. A standard gating strategy was applied to all samples. First, nuclei were gated on their size and scatter properties. Doublet discrimination gates were used to exclude aggregates of nuclei. Lastly, nuclei were gated on DAPI. Single nuclei were sorted into chilled PCR tubes containing 10 µl of resuspension buffer. The counts of nuclei were verified using a haemocytometer.
RNA-seq library preparation and sequencing
10X Chromium Next GEM Single Cell 3′ HT Kit v.3.1 was used for library preparation. Following manufacturer’s instructions, nuclei were promptly mixed with the master mix and loaded onto a 10X Chromium Next GEM Chip M and ran on a 10X Chromium X instrument aiming to recover 20,000 cells from each. We generated 10X 3′ RNA-seq libraries following the manufacturer’s protocols. After complementary DNA amplification and after library construction, quality controls were performed to ensure quality of samples. Libraries were loaded at 650 pM along with 1% PhiX control on an Illumina NextSeq2000 and paired-end sequenced (28 cycles read1, 10 cycles i7 index, 10 cycles i5 index, 90 cycles read2) at a targeted depth of 20,000 reads per nucleus.
RNA-seq data analysis
Raw read (FASTQ) files were aligned to the mouse reference genome (mm10-2020-A) using Cell Ranger (v.7.0.1). For snRNA-seq, all analysis, quantification and statistical testing was completed using R (v.4.2.2). Seurat (v.4.3.0) was used for preprocessing, dimensionality reduction, clustering and differential expression testing. Cell Ranger count matrices for each sample were merged and low-quality cells with fewer than 200 detected genes or greater than 5% mitochondrial mapping unique molecular identifiers (UMIs) and putative doublets with greater than 4,000 detected genes were removed. A total of 80,302 nuclei passed these quality control criteria (39,780 saline and 40,522 morphine). UMI count data were then normalized, such that the counts per each gene per cell were divided by the total UMIs per cell, multiplied by a scale factor of 10,000 and natural log transformed. Variance stabilizing transformation was performed and the top 2,000 variable genes were identified, followed by principal component analysis. UMAP was conducted using the first 12 principal components and graph-based clustering was used to identify clusters with a resolution parameter of 0.8. To focus on oligodendroglial lineage cells, oligodendrocytes and OPCs were then subsetted and reclustered using similar parameters. Trajectory from OPCs to mature oligodendrocytes were labelled using previously defined markers for oligodendroglial subpopulations37. All differential expression testing was performed using Wilcoxon rank-sum tests with Bonferroni corrections for several comparisons. Adjusted P < 0.05 with accompanying average log2-fold changes greater than 0.25 in magnitude were considered statistically significant. ScType and differential expression testing and were first performed to identify cell types based on canonical marker gene expression and clusters of the same cell type were merged52. Gene set enrichment analysis was performed using Cluster Profiler (v.4.6.2) to identify enrichment of gene ontology biological process terms in cluster differentially expressed genes using a hypergeometric test. Cell–cell communication analysis and inference was performed using CellChat (v.1.6.1) to calculate the aggregated cell–cell communication networks and identify signals contributing to outgoing or incoming signalling of different cell groups38.
In vitro OPC analysis
To test the direct effects of morphine and dopamine, in vitro OPC proliferation assay was used. C57BL/6 P4-5 mice pups were rapidly decapitated and brains were processed in Hibernate-A medium (Thermo Fisher Scientific, A12475-01). Resulting tissue was enzymatically disassociated in buffer containing HEPES-HBSS with DNase (Worthington Biochemical LS002007) and Liberase (Roche Applied Sciences 05401054001) at 37 °C on a rotator. Tissue mixture then was triturated with 1,000 µl tip and passed through a 100 µm cell strainer. OPCs were isolated using the CD140 (Pdgfrα) Microbead kit (MACS, Miltenyi Biotex 130-101-502) according to the manufacturer’s instructions. A total of 30,000 cells were seeded per well, on laminin-coated (Thermo Fisher Scientific, 23017015) coverslips in a 24-well plate. OPC media containing DMEM (Thermo Fisher Scientific, 11320082), glutamax (Invitrogen, 35050-061), sodium pyruvate (Invitrogen, 11360070), MEM non-essential amino acids (Thermo Fisher Scientific, 11140076), antibiotic-antimytotic (GIBCO), N21-MAX (R&D systems, AR012), trace elements B (Corning, 25-022-Cl), 5 mg ml−1 of N-acetyl cysteine (Sigma-Aldrich, A9165), 10 ng ml−1 of PDGFAA (Shenandoah Biosciences, 200-54), 10 ng ml−1 of CNTF (PeproTech, 450-13) and 1 ng ml−1 of NT-3 (PeproTech, 450-03) was used. For proliferation studies, after 3 days in proliferative media, cells were treated with various concentrations of morphine or dopamine in OPC media for 24 h. In the last 4 h of this treatment, 10 µM EdU was added to the media to label dividing cells. Afterwards, cells were fixed in 4% paraformaldehyde (PFA) for 20 min and incubated in HBSS until immunohistochemistry. All in vitro experiments were performed in triple wells (technical replicate) and independently replicated (biological replicates).
Immunohistochemistry
All mice were anaesthetized with intraperitoneal injections of 2.5% Avertin (tribromoethanol; Sigma-Aldrich, T48402) and transcardially perfused with 20 ml of 0.1 M PBS. Brains were postfixed in 4% PFA overnight at 4 °C before cryoprotection in 30% sucrose solution for 48 h. For sectioning, brains were embedded in optimum cutting temperature (Tissue-Tek) and sectioned coronally at 40 µm using a sliding microtome (Leica, HM450). For immunohistochemistry, brain sections were stained using the Click-iT EdU cell proliferation kit (Invitrogen, C10339) according to manufacturer’s protocol. Tissue sections were then stained with antibodies following an incubation in blocking solution (3% normal donkey serum, 0.3% Triton X-100 in tris-buffered saline (TBS)) at room temperature for 30 min. Goat anti-Pdgfrα (1:500; R&D Systems, AF1062), rabbit anti-Olig2 (1:500; Abcam, 7349), rabbit anti-ASPA (1:250; EMD Millipore, ABN1698), rat anti-MBP (1:200; Abcam ab7349), rabbit anti-tyrosine hydroxylase (1:500; Millipore Sigma, AB152), mouse anti-tyrosine hydroxylase (1:250; Novus Biologicals, MAB7566), chicken anti-GFP (1:1,000; Aves Labs, GFP-1020), chicken anti-mCherry (1:1,000; ab205402) or rabbit anti-cfos (1:500; Santa Cruz Biotechnology, sc-52) were diluted in 1% blocking solution (1% normal donkey serum in 0.3% Triton X-100 in TBS) and incubated overnight at 4 °C. All antibodies have been validated in the literature for use in mouse immunohistochemistry. To further validate the antibodies, we confirmed that each antibody stained in the expected cellular patterns and brain-wide distributions (for example, nuclear Olig2 staining, cell membrane Pdgfrα staining, tyrosine hydroxylase staining in midbrain DA neurons). The following day, brain sections were rinsed three times in 1× TBS and incubated in secondary antibody solution for 2 h at the room temperature. All secondary antibodies were used at 1:500 concentration including Alexa 488 anti-rabbit (Jackson ImmunoResearch; 711-545-152), Alexa 488 anti-mouse (Jackson ImmunoResearch; 715-545-150), Alexa 488 anti-chicken (Jackson ImmunoResearch; 703-545-155), Alexa 594 anti-chicken (Jackson ImmunoResearch; 703-585-155), Alexa 647 donkey anti-goat (Jackson ImmunoResearch; 705-605-147), Alexa 647 anti-rat (Jackson ImmunoResearch; 712-605-150), Alexa 647 anti-rabbit (Jackson ImmunoResearch; 711-605-152), Alexa 647 donkey anti-goat (Jackson ImmunoResearch; 705-605-147). Sections were then rinsed three times in 1× TBS and mounted with ProLong Gold (Life Technologies, P36930).
Fluorescence microscopy and quantification
All image analyses were performed by experimenters blinded to the experimental conditions or genotype. EdU stereology images were taken using a Zeiss AxioObserver upright fluorescence microscope with automated stage and tile-scanning capability (Stereo Investigator, Microbrightfield) with ×20 objective. Brain tissue that was damaged during perfusion or tissue processing was excluded from histological analysis. Imaging and cell counting were performed on every sixth 40 µm slice, throughout the extend of DA neurons in the midbrain and extend of NAc. Cells surrounding ferrule-induced tissue damage were not included. Regions with tyrosine hydroxylase labelling (or Dat-Cre»GFP positive for optogenetic studies) were marked and EdU cells in these regions were counted manually using Stereo Investigator software. Resulting numbers were reported as a function of the area (cells per mm2). Higher resolution imaging was conducted by acquiring z-stacks using a Zeiss LSM710 or LSM980 confocal microscope (Carl Zeiss). For MBP intensity analyses, z-stack images were taken of VTA using confocal microscopy with the same laser and fluorescence settings across animals and sections. Mean MBP intensity of pixels over VTA, which is labelled by Dat-Cre»GFP, was quantified using Fiji and reported as a function of the area.
Electron microscopy and quantification
Four weeks after cessation of chronic optogenetic stimulations, mice were euthanized by transcardial perfusion with Karnovsky’s fixative: 2% glutaraldehyde (EMS 16000) and 4% PFA (EMS 15700) in 0.1 M sodium cacodylate (EMS 12300), pH 7.4. For regional accuracy and consistency, brains were sectioned into 250-µm-thick slices on a vibratome (Leica VT1000S) until midbrain. Then 2-mm-diameter tissue punches were collected from VTA region under the optic fibre mark. The samples were then postfixed in 1% osmium tetroxide (EMS 19100) for 1 h at room temperature, washed three times with ultrafiltered water, then en bloc stained for 2 h at room temperature. Samples were dehydrated in graded ethanol (50%, 75% and 95%) for 15 min each at 4 °C; the samples were then allowed to equilibrate to room temperature and were rinsed in 100% ethanol twice, followed by acetonitrile for 15 min. Samples were infiltrated with EMbed-812 resin (EMS 14120) mixed 1:1 with acetonitrile for 2 h followed by 2:1 EMbed-812:acetonitrile for 2 h. The samples were placed into EMbed-812 for 2 h, then placed into TAAB capsules filled with fresh resin, which were then placed into a 65 °C oven overnight. Sections were taken between 40 and 60 nm on a Leica Ultracut S (Leica) and mounted on 100-mesh Ni grids (EMS FCF100-Ni). For immunohistochemistry, microetching was done with 10% periodic acid and eluting of osmium with 10% sodium metaperiodate for 15 min at room temperature on parafilm. Grids were rinsed with water three times, followed by 0.5 M glycine quench and then incubated in blocking solution (0.5% BSA, 0.5% ovalbumin in PBST) at room temperature for 20 min. Primary rabbit anti-GFP (1:300; MBL International) was diluted in the same blocking solution and incubated overnight at 4 °C. The next day, grids were rinsed in PBS three times and incubated in secondary antibody (1:10 10 nm gold-conjugated IgG TED Pella 15732) for 1 h at room temperature and rinsed with PBST followed by water. For each staining set, secondary-only staining was simultaneously performed to control for any non-specific binding. Grids were contrast stained for 30 s in 3.5% uranyl acetate in 50% acetone followed by staining in 0.2% lead citrate for 90 s. Samples were imaged using a JEOL JEM-1400 TEM at 120 kV and images were collected using a Gatan Orius digital camera. Adjacent sections from all samples were also tested for ChR2::YFP and YFP signal using fluorescent microscopy to confirm the viral labelling. All image analyses were performed by experimenters blinded to the experimental conditions. Secondary antibody-only controls were also performed to establish immunogold labelling background amounts. Myelinated axons with more than three independent immunogold spheres were counted as positive labelling. The g-ratios were calculated by dividing the shortest axonal diameter by the corresponding axonal-plus-sheath diameter (diameter of axon/diameter of axon plus myelin sheath). Between 296 and 446 axons were analysed per animal. Myelinated axon density was analysed by quantifying the number of myelinated axons at each size per ×10,000 electron micrograph. An average of 45 images were quantified and the total number of myelinated axons was divided by the total area quantified per animal. Group means are calculated on a per mouse basis (not per axon or per image).
Statistical analysis
The experimenter was blinded to the genotype of the animals during behavioural testing, microscopy imaging and histological analyses. All optogenetic real-time place preference experiments, CPP and CPA pre-test–post-test preference comparisons, and novel-object recognition test data that showed Gaussian distribution were assessed using two-tailed paired t-tests. In cases in which normal distributions were not assumed for CPP preference data, we performed Wilcoxon non-parametric signed-rank tests. For all histological analysis and CPP Z-scores with normal distributions, group mean differences were analysed using unpaired two-tailed t-tests. For data that did not pass the normality test, Mann–Whitney non-parametric tests were used. All statistical analyses were performed using GraphPad Prism Software.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
