To produce NPY-Cre mice, B6.FVB(Cg)-Tg(NPY-cre)RH26Gsat/Mmucd (037423-UCD, NPY-Cre) sperm were purchased from Mutant Mouse Resource & Research Centers (Auckland, CA, USA). We entrusted the generation of NPY-Cre mice from the sperm to the RIKEN BioResource Research Center (Ibaraki, Japan). The mice were kept under a 12-h light/dark cycle (lights on at 8 a.m.). Standard food pellets and water were provided ad libitum. All animal experiments, such as production, maintenance protocols, and behavioral studies, were reviewed and approved by the Animal Care and Use Committee of the Kyoto Prefectural University of Medicine (M2019-155&157, M2020-162&165).

pAAV-EF1a-DIO-mCherry (Addgene plasmid# 50462; http://n2t.net/addgene:50462; RRID: Addgene_50462) was a gift from Bryan Roth. pAAV-phSyn1(S)-FLEX-tdTomato-T2A-SypEGFP-WPRE was a gift from Hongkui Zeng (Addgene plasmid# 51509; http://n2t.net/addgene:51509; RRID: Addgene_51509) (Oh et al., 2014). pAAV-EF1a-FLEX-TVA-mCherry was a gift from Naoshige Uchida (Addgene plasmid# 38044; http://n2t.net/addgene:38044; RRID: Addgene_38044) (Watabe-Uchida et al., 2012). pAAV-FLEX-SADcvsG was generated (Mori and Morimoto, 2014). For recombinant AAV (rAAV) production, the vectors were cotransfected into HEK293 cells with pAAV-DJ (Cell Biolabs, San Diego, CA, USA) and pHelper (Takara Bio Inc., Shiga, Japan) using AAVpro Helper-Free system (Takara Bio Inc.). rAAV particles were extracted and purified using the AAVpro Purification Kit (Takara Bio Inc.), according to the instructions of the manufacturer. rAAV titers were determined by quantitative PCR using the AAVpro Titration Kit (Takara Bio Inc.), and aliquoted rAAVs were stored at −80°C.

We produced the HEP-Flury (HEP) strain RABV using plasmid encoding glycoprotein-deficient HEP-Flury expressing GFP (pcDNA-HEPdG-GFP), helper plasmids encoding SADcvsG, which is the cytoplasmic domain of SAD glycoprotein replaced by that of CVSG (pCAGGS-SADcvsG), helper plasmids encoding HEP nucleoprotein (pcDNA-HEPN), phosphoprotein (pcDNA-HEPP), and polymerase (pcDNA-HEPL) as described in our previous study (Mori and Morimoto, 2014). About 1.0 × 10⁴ BHK-T7/9 cells, expressing T7 RNA polymerase, were seeded into each well of a 100-mm dish in the medium containing 2% fetal bovine serum (FBS). Cells were transfected with plasmid encoding pcDNA-HEP-ΔG-GFP (2 μg), pcDNA-HEPN (1 μg), pcDNA-HEPP (0.5 μg), pcDNA-HEPL (0.25 μg), and pCAGGS-SADcvsG (0.5 μg) using TransIT-LT1 (Mirus Bio LLC, Madison, WI, USA), as described by the manufacturer. The plates were incubated for 2 days at 32°C in an atmosphere containing 5% CO₂, and then, the supernatant was collected as HEPdG-GFP. To produce EnvA-enveloped HEPdG-GFP, HEPdG-GFP was added to wells of 80% confluent BHK-T7/9 cells on 100-mm plates and then incubated for 2 days at 32°C in an atmosphere containing 5% CO₂. Thereafter, the supernatant was removed, and the cells were washed with PBS (5×) to eliminate the remaining HEPdG-GFP and incubated as before. The next day, the HEPdG-GFP-infected cells were transfected with pCAGGS-EnvA and incubated for 4 days. The supernatant containing EnvA-HEPdG-GFP was concentrated by ultracentrifugation, and aliquots of EnvA-HEPdG-GFP were stored at −80°C.

Mice were anesthetized with a midazolam/medetomidine/butorphanol cocktail (4, 0.3, and 5 mg/kg, respectively) intraperitoneally and placed on a stereotaxic apparatus (Narishige, Tokyo, Japan). AAVs were unilaterally microinjected (0.2–1 μl/site) into the NAc (AP, +1.2 mm from the bregma; ML, +0.8 mm from the midline; and DV, 4.3 mm below the skull surface) using a 30-gauge Hamilton syringe needle (Hamilton, Reno, NV, USA) at a rate of 0.1 μl/min. The needle was kept in place for 5–10 min after each injection before the needle was slowly removed. Some mice that received a second brain injection were housed individually, while the others were returned to their group home cage for recovery. Viral injections were verified by fluorescent image and location of cannula tip, and data from mice that were not correctly injected virus and CTb into the target region were excluded.

AAV(dj)-FLEX-mCherry (0.8 × 10⁷ viral genome particles/ml) or AAV(dj)-FLEX-tdTomato-T2A-SypEGFP (2.7 × 10¹¹ viral genome particles/ml) of 0.5 μl was injected into the NAc (n = 3 each). mCherry, tdTomato, and synaptophysin-EGFP (sypEGFP) expressions were allowed to develop for more than 2 weeks before the mice were deeply anesthetized and perfused.

To label NAc NPY neurons, 0.5 μl of AAV(dj)-FLEX-mCherry was injected into the NAc. Two weeks later, 0.5 μl of CTb (10 μg/μl; C-7771, Sigma-Aldrich, Steinheim, Germany), a retrograde tracer, was injected into the LH (AP, −1.2 mm from the bregma; ML, +0.9 mm from the midline; and DV, 4.3 mm below the skull surface) at a rate of 0.1 μl/min. CTb was allowed to undergo retrograde transport from the LH to the NAc for 6 days before the mice were deeply anesthetized and perfused (n = 5).

AAV(dj)-FLEX-TVA-mCherry (1.5 × 10¹¹ viral genome particles/ml) of 0.5 μl was injected into the NAc. Two weeks later, 1 μl of EnvA-HEPdG-GFP was injected into the LH, and the virus was allowed to undergo retrograde transport from the LH to the NAc for 1 week before the mice were deeply anesthetized and perfused (n = 4).

To target global input to NPY neurons in the NAc, we performed monosynaptic retrograde tracing (Watabe-Uchida et al., 2012; Krashes et al., 2014). A mixture containing 0.5 μl of AAV(dj)-FLEX-TVA-mCherry and AAV(dj)-FLEX-SADcvsG (2.7 × 10¹¹ viral genome particles/ml) was injected into the NAc. Then, 2 weeks later, 1 μl of EnvA-HEPdG-GFP was injected into the same region, and the virus was allowed to undergo retrograde transport for 1 week before the mice were deeply anesthetized and perfused (n = 6).

Mice were anesthetized with pentobarbital (Somnopentyl; Kyoritsu Seiyaku, Tokyo, Japan) intraperitoneally and perfused with physiological saline followed by 4% paraformaldehyde in 0.05 M phosphate buffer. The brain was immediately removed, postfixed with the same fixative overnight at 4°C, and then kept in 30% sucrose in 0.05 M phosphate buffer at 4°C. Serial coronal sections (40 μm) were obtained using a cryostat (CM 3050 S; Leica, Wetzlar, Germany).

For visualization and anterograde tracing of NPY neurons in the NAc, we performed mCherry IHC. Every fourth section was sequentially incubated with 0.3% H₂O₂ and 0.3% Triton X-100 in PBS for 30 min, and then in PBS containing 2% normal goat serum (NGS) and 0.1% Triton X-100 for 1 h at room temperature (RT). Sections were then incubated with primary rabbit antiserum against mCherry (1:2,000; ab167453, Abcam, Cambridge, UK) for 72 h at 4°C. Immunoreactive neurons were visualized with a streptavidin-biotin kit (Nichirei, Tokyo, Japan), followed by 3,3′-diaminobenzidine (DAB), as described in our previous studies (Takanami et al., 2010; Yamada and Kawata, 2014).

To confirm AAV(dj)-FLEX-tdTomato-T2A-SypEGFP infection, the brain sections were processed for GFP immunofluorescence. Every fourth section was incubated with 2% NGS in PBS for 1 h and then incubated with primary rabbit antiserum against GFP (1:2,000; ab6556, Abcam) for 72 h at 4°C. After washing with PBS, the sections were incubated with Alexa Fluor 488-labeled donkey anti-rabbit IgG (1:1,000; A21206, Thermo Fisher Scientific, MA, USA).

For mCherry and CTb double IHC, every fourth section from the NAc to LH was incubated with 2% normal rabbit serum (NRS) in PBS for 1 h and then incubated with primary goat antiserum against CTb (1:1,000; #703, List Biological Laboratories Inc., CA, USA) and rat antiserum against mCherry (1:2,000; M11217, Thermo Fisher Scientific) for 72 h at 4°C. After washing with PBS, the sections were incubated with Alexa Fluor 488-labeled donkey anti-goat IgG (1:1,000; ab150129, Abcam) and Alexa Fluor 555-labeled donkey anti-rat IgG (1:1,000; ab150154, Abcam).

For mCherry and GFP double IHC, every fourth section was incubated with 2% NGS in PBS for 1 h and then with primary rabbit antiserum against GFP (1:2,000) and rat antiserum against mCherry (1:2,000) for 72 h at 4°C. After washing with PBS, the sections were incubated with Alexa Fluor 488-labeled donkey anti-rabbit IgG (1:1,000) and Alexa Fluor 555-labeled donkey anti-rat IgG (1:1,000).

Fluorescence images of high magnification were obtained on an LSM510META confocal laser-scanning microscope (Carl Zeiss, Jena, Germany). Fluorescence images of low magnification were observed under a fluorescence microscope (CKX53; Olympus, Tokyo, Japan), and images were captured using a CCD camera (DP 74; Olympus). Fluorescence intensities were manipulated to make clear the injection sites and location of neural nuclei in images of low magnification. The sections reacted with DAB were observed under a light microscope (BX50; Olympus), and images were captured using a CCD camera (DP 21; Olympus). To evaluate the numbers of labeling cells and the ratio, we determined the region of interest (ROI) in the medial part of the NAc. The size of ROI is defined as 0.45 mm × 0.45 mm, and this is the size of the photo frame in our confocal laser-scanning microscope using a 20 × objective lens. This frame is able to cover the medial part of NAc where is the target region of AAV. mCherry-positive, GFP-positive, mCherry and CTb double-positive, and mCherry and GFP double-positive cells were counted in these images. All values are expressed as mean ± SEM.

To visualize NPY neurons in the NAc, we injected AAV(dj)- FLEX-mCherry into the NAc in NPY-Cre mice, and mCherry was visualized by IHC. Previously, we confirmed that our AAV injection into the NAc of NPY-Cre mice caused specific expression of mCherry in accumbal NPY neurons (Yamada et al., 2020). We observed that a 0.2-μl volume of AAV injection at 1.2-mm anterior from the bregma spread through the NAc, with a longitudinal distance from 1.34 mm to 0.74 mm anterior to the bregma in the brain atlas (Franklin and Paxinos, 2007). NPY neurons in the medial part of the NAc were labeled by the injection without distinction of the core and shell of the nucleus (Figure 1). Other than NAc, we found a few mCherry-positive cells in the lateral septum (LS) and nucleus of the vertical limb of the diagonal band (VDB).

We analyzed mCherry projections originating from the NAc in whole-brain sections. We found many dot-like mCherry-expressing fibers in the anterior (Figures 2A,B) and posterior (Figures 2C,D) parts of the LH in coronal sections. In sagittal sections, we also found some mCherry-expressing fibers in the LH (Figures 2E,F). In addition, we found a few mCherry-expressing fibers around the LH, such as the VP and extended amygdala. mCherry-positive cell bodies and fibers were dominant in the NAc and LH, respectively, demonstrating that NPY neurons in the NAc project to the LH.

To confirm the finding of NPY neural connections between the NAc and the LH, we injected other Cre-dependent AAV into the NAc in NPY-Cre mice (Figure 3A). This AAV encoded the synaptic protein synaptophysin fused with EGFP (sypEGFP; AAV-FLEX-tdTomato-sypEGFP) (Oh et al., 2014). Many double-labeled cell bodies for tdTomato and sypEGFP were detected in the NAc around the injection site (Figures 3B,D–F). In these mice, we also observed EGFP-positive presynaptic puncta in the LH (Figures 3C,G–I).

To further clarify the neuroanatomical connections of NPY neurons between the NAc and LH, we performed retrograde tracing using CTb (n = 5). NPY is a neurotransmitter that is released from axon terminals and is usually transported to the pre-synapse after synthesis and modification of the precursor peptide within the endoplasmic reticulum and Golgi apparatus (von Hörsten et al., 2004). To visualize NPY neuron cell bodies, we injected Cre-dependent AAV encoding mCherry into the NAc in NPY-Cre mice 2 weeks before CTb injection into the LH (Figure 4A). Injection of CTb into the LH (Figure 4B) caused the ipsilateral distribution of CTb-positive neurons in the NAc (Figures 4C,E,H). We also found several CTb and mCherry double-positive neurons in the NAc (Figures 4G,J). The numbers of mCherry-positive and CTb and mCherry double-positive neurons in the ROI (0.45 × 0.45 mm) in the NAc were 142.0 ± 27.9 and 10.2 ±2.52, respectively. The percentage of CTb-positive neurons among total mCherry-positive neurons was 7.19 ± 1.16% (Table 1).

We investigated whether other retrograde tracing methods would also reveal projections of NAc NPY neurons to the LH (n = 4). To this end, we employed Cre-dependent AAV encoding avian sarcoma leukosis virus receptor TVA (AAV-FLEX-TVA-mCherry) and avian sarcoma leukosis virus envelop protein (EnvA)-enveloped RABVdG (EnvA-HEPdG-GFP) (Sun et al., 2019). First, AAV-FLEX-TVA-mCherry was injected into the NAc in NPY-Cre mice to express TVA in NPY neural cell bodies and terminals, and then, EnvA-HEPdG-GFP was injected into the LH (Figures 5A,B). If there are neural terminals containing TVA from accumbal NPY neurons, EnvA-HEPdG-GFP would infect via TVA and get retrogradely transported to the NAc. We found that a considerable number of GFP-positive neurons were distributed ipsilaterally through the rostral to caudal regions of the NAc, with caudal dominance (Figure 5C). We also observed that most GFP-positive neurons in the NAc overlapped with mCherry-positive neurons (Figures 5D–F). The numbers of mCherry-positive and GFP and mCherry double-positive neurons in the ROI (0.45 × 0.45 mm) in the NAc were 99.5 ± 12.1 and 10.3 ±3.5, respectively. The percentage of GFP-positive neurons among total mCherry-positive neurons was 9.55 ± 2.15% (Table 2). When AAV-FLEX-TVA-mCherry or EnvA-HEPdG-GFP was injected into the NAc in wild-type mice, there was no mCherry- or GFP-positive neurons in the NAc (data not shown) indicating viral specificity for Cre or TVA expressing cells, respectively.

Finally, we performed global mapping of neurons projecting directly to the NAc NPY neurons by monosynaptic retrograde tracing with rRABV (for coronal and sagittal sections, n = 3 each). To this end, we injected AAV(dj)-FLEX-TVA-mCherry and AAV(dj)-FLEX-SADcvsG into the NAc of NPY-Cre mice. TVA and SADcvsG are required for EnvA-HEPdG-GFP infection and subsequent retrograde spread, respectively. After 14 days, EnvA-HEPdG-GFP was injected into the same area, and the brain was analyzed 7 days later (Figure 6A). Successful infection and expression of TVA and EnvA-HEPdG-GFP in the NAc were confirmed by mCherry and GFP immunostaining, respectively (Figures 6B,C). The first-infected cells were identified by coexpression of TVA-mCherry and GFP. In our experiment, neurons expressing both mCherry and GFP were detected only in the NAc (Figures 6D–F). At the injection site, there were some GFP-positive/mCherry-negative neurons, indicating that they were interneurons projecting to the NPY neurons in the same area (Figures 6D–F). Furthermore, we evaluated input neurons that were labeled retrogradely from NPY neurons in the NAc. We found that the vast majority were located in two regions, the midline thalamic nuclei (Figure 6G) and posterior regions of the basomedial amygdala (pBMA; Figure 6H). We also observed a smaller number of GFP-positive neurons in other areas, such as the parafascicular thalamic nucleus, mPFC, entorhinal cortex, insular cortex, prelimbic cortex, posterior region of the basolateral amygdala, lateral preoptic area, and VP.