Characterization of a Knock-In Mouse Line Expressing a Fusion Protein of κ Opioid Receptor Conjugated with tdTomato: 3-Dimensional Brain Imaging via CLARITY

Abstract

Activation of κ opioid receptor (KOR) produces analgesia, antipruritic effect, sedation and dysphoria. To characterize neuroanatomy of KOR at high resolutions and circumvent issues of specificity of KOR antibodies, we generated a knock-in mouse line expressing KOR fused at the C terminus with the fluorescent protein tdTomato (KtdT). The selective KOR agonist U50,488H caused anti-scratch effect and hypolocomotion, indicating intact KOR neuronal circuitries. Clearing of brains with CLARITY revealed three-dimensional (3-D) images of distribution of KOR, and any G-protein-coupled receptors, for the first time. 3-D brain images of KtdT and immunohistochemistry (IHC) on brain sections with antibodies against tdTomato show similar distribution to that of autoradiography of [³H]U69,593 binding to KOR in wild-type mice. KtdT was observed in regions involved in reward and aversion, pain modulation, and neuroendocrine regulation. KOR is present in several areas with unknown roles, including the claustrum (CLA), dorsal endopiriform nucleus, paraventricular nucleus of the thalamus (PVT), lateral habenula (LHb), and substantia nigra pars reticulata (SNr), which are discussed. Prominent KtdT-containing fibers were observed to project from caudate putamen (CP) and nucleus accumbens (ACB) to substantia innominata (SI) and SNr. Double IHC revealed co-localization of KtdT with tyrosine hydroxylase (TH) in brain regions, including CP, ACB, and ventral tegmental area (VTA). KOR was visualized at the cellular level, such as co-localization with TH and agonist-induced KOR translocation into intracellular space in some VTA neurons. These mice thus represent a powerful and heretofore unparalleled tool for neuroanatomy of KOR at both the 3-D and cellular levels.

Keywords: 3-D imaging; CLARITY; neuroanatomy; κ opioid receptor.

Figure 1.

Generation and characterization of KOR-tdTomato (KtdT/KtdT) knock-in mice and comparison with the wild KOR (K/K) mice. A, Targeting strategy. Oprk1 exons 3 and 4, tdTomato cDNA, and the floxed neomycin cassette are shown as empty, gray, and black boxes, respectively. Homologous recombination (HR) was followed by Cre recombinase treatment (Cre) in ES cells. B, Genotyping by PCR analysis using primer pair Ef/Er as shown in A and genomic DNA from mouse ears as templates. C, [³H]U69,593 binding (dpm) to receptor in mouse whole-brain membranes (1.8-mg protein) was performed with 4.5 nm [³H]U69,593 and non-specific binding was determined by naloxone (10 μm). Data are mean ± SEM (n = 4) **p < 0.01 and ***p < 0.001. D, Receptor mRNA levels were determined by real-time qRT-PCR with RNA preparations from mouse whole brains. Data are mean ± SEM (K/K, n = 4; KtdT/KtdT, n = 3); **p <0.01 and ***p < 0.001, compared with the wild type (K/K) by two-tailed Student’s t test. E, U50,488H inhibited scratching behavior induced by compound 48/80 in both KtdT/KtdT and K/K mice. Male mice were pretreated subcutaneously with saline or U50,488 (0.2–5 mg/kg) and 20 min later, compound 48/80 was injected into the nape of the neck and bouts of scratching were counted for 30 min and normalized to the saline group of each mouse strain. Each value represents mean ± SEM (n = 10). F, U50,488H inhibited novelty-induced locomotor activities in both KtdT/KtdT and K/K mice. Male mice were treated subcutaneously with saline or U50,488H (at doses of A50 values in the anti-scratching test) and put into locomotor chambers right after injections. Total locomotor activities (breaks of infrared beams) were continuously monitored using a Digiscan D. Micro System over a 90-min period. Cumulative data between 20 and 50 min postinjection are shown here. Each value represents mean ± SEM (n = 8). Data were analyzed by two-way ANOVA followed by Tukey’s post hoc test; *p < 0.05, compared with the respective saline group.

Figure 2.

A, A 3-D image of mouse brain cleared with CLARITY showing brain-wide distribution of KtdT. The 3-D whole-brain image was reconstructed from ventral and dorsal stacks acquired separately with optical settings described in Materials and Methods. Directions: anterior (A), posterior (P), ventral (V), dorsal (D), lateral (L), and medial (M). The autofluorescence in green channel was used as a reference (Ref) to whole-brain volume which is pseudo-colored in cyan. Experiments were performed on three brains with similar results. See Movies 1, 2 for video clips. B, An enlarged view of a portion of the 3-D image showing prominent KtdT fiber bundles between SI and SNr. C, Images (I–V) were presented as optical sections from 1-mm coronal blocks of mouse brains showing KtdT neurons in the prefrontal cortex (PFC), claustrum (CLA), lateral septum (LS), ventral tagmental area (VTA), paraventricular nuleus of thalamus (PVT), lateral habenula (LHb). The brain region of the images is marked by circles in the horizontal view of 3-D rendering (gray).

Figure 3.

A, Comparison of IHC staining of KtdT (top row) with autoradiography of [³H]U69,593 binding to the KOR (bottom row) in two coronal brain sections (30 μm for IHC and 20 μm for autoradiography in thickness). IHC images of KtdT were captured with a wide field fluorescence microscope. Experiments were performed on three brains each with similar results. B, Distribution of KtdT in coronal brain sections. IHC images of KtdT were captured with a wide field fluorescence microscope. Rostral-caudal coordinates in reference to bregma are indicated. The numbers in parenthesis are the coronal section numbers in the Mouse Atlas of Allen Brain Institute (https://mouse.brain-map.org/experiment/thumbnails/100048576?image_type=atlas). Experiments were performed on three brains each with similar results. For neuroanatomical sites, see list of abbreviations for brain nucleus and regions. C, Three higher resolution images with relative intensities from high to low. The images were taken with 20× objective on confocal microscope. The regions CLA, BLA, and VTA were chosen as representatives of high to low KtdT expression levels. List of abbreviations for brain nucleus and regions (per Mouse Brain Atlas of Allen Brain Institute): A11: A11 DA neurons; aco: anterior commissure; AC: anterior cingulate cortex; ACB: nucleus accumbens; alv: alveus of hippocampus; ARH: arcuate nucleus of hypothalamus; BLA: basolateral amygdala nucleus; BMA: basomedial amygdala nucleus; BST: bed nucleus of the stria terminalis; CB: cerebellum; CEA: central amygdalar nucleus; CLA: claustrum; CP: caudate putamen; CS: superior central nucleus raphe; CTX: cortex; ECT: ectorhinal cortex; ENT: entorhinal cortex; EP: endopiriform nucleus; EPd: dorsal endopiriform nucleus; DMH: dorsomedial nucleus of the hypothalamus; DR: dorsal raphe nucleus; EW: Edinger–Westphal nucleus; Hb: habenula; HIP: hippocampus; LA: lateral amygdalar nucleus; LHb: lateral habenula; LM: lateral mammillary nucleus; MEA: medial amygdalar nucleus; MM: Medial mammary nucleus; MPO: medial preoptic area; nst: nigrostriatal tract; OB: olfactory bulb; OT: olfactory tubercle; PAG: periaqueductal gray; PER: perirhinal cortex; PF: prefrontal cortex; PH: posterior hypothalamic area; PVH: paraventricular hypothalamus nucleus; PVT: paraventricular nucleus of thalamus; RE: reuniens nucleus of thalamus; RSP: retrosplenial cortex; SI: substantia innominata; SNc: substantia nigra, pars compacta; SNr: substantia nigra, pars reticulata; st: stria terminalis; STN: subthalamic nucleus; SC: superior colliculus; VTA: ventral tegmental area.

Figure 4.

A thick horizontal brain section (1 mm) of the KtdT mouse showing IHC staining of (A) KtdT (red) alone and (B) both KtdT (red) and TH (green). KtdT/KtdT mice were perfused and cleared with the CLARITY method. Cleared brains were sectioned at 1 mm with a Vibratome, and IHC was performed on floating sections for TH and KtdT. IHC image Z-stacks were captured with a confocal microscope, and the tiles were stitched in real time with NIS-Element software. The images shown are maximum intensity projection (MaxIP) of the stacks to demonstrate long-range projections. In these images, TH-containing or KtdT-containing fibers are clearly visible between SN and CP. A, B, This section is approximately −4.4 mm ventral to bregma. C, D, This section is approximately −5.6 mm ventral to bregma.

Figure 5.

A sagittal brain section (1 mm) of the KtdT mouse showing IHC staining of (A) KtdT (red) alone and (B) both KtdT (red) and TH (green). See Figure 4 legend. Note that nerve fibers are clearly visible. A, B, This section is ∼2.15 mm lateral to the midline. C, D, This section is ∼1.35 mm lateral to the midline.

Figure 6.

Anterograde tracing shows projection of KtdT-containing neurons from CP and ACB to SI and SNr. A, ISH of KOR mRNA: wild-type C57BL/6N mice were used. ISH was performed with RNAscope on coronal brain sections at regions containing CP and ACB. Experiments were performed on three brains with similar results. Some KOR mRNA (red) is present in D1 (green) or D2 (white) DA receptor-expressing neurons. B–D, Anterograde tracing from CP and ACB. Adult KtdT/KtdT mice were injected with scAAV2-GFP tracer (0.2 μl) into the CP and ACB. Three to four weeks after injection, mice were perfused, cleared with CLARITY, processed for KtdT IHC on 1-mm sagittal sections and confocal microscopy as described in Figure 4 legend. B, Red: KtdT. C, Green: scAAV2-GFP. D, Red + green + blue (DAPI). Yellow color indicates overlap of KtdT and scAAV2-GFP. The results indicate projections from CP and ACB to SI and SNr. The experiment was performed three times with similar results.

Figure 7.

Double IHC staining of KtdT (red) and TH (green) in the VTA. IHC staining for KtdT (red) and TH (green) was performed on coronal sections 30 μm) containing the VTA. A, The three figures show macro views of the VTA region with a 4× objective on a wide-field fluoresce microscope. B and C, The two figures show KtdT (red) and TH (green) staining acquired with a NikonA1R confocal microscope and a 60× objective, each of which is a MaxIP of three focal planes from a Z-stack. Co-localization of both in some neurons are indicated by arrows. The Z-stacks of confocal images are shown as a video in Movie 3.

Figure 8.

U50,488 induced KOR translocation into cells in the VTA. KtdT/KtdT mice were injected subcutaneously with the agonist U50,488 at 5 mg/kg or vehicle (n = 4 each); 30 min later, mice were anesthetized and perfused and coronal sections containing the VTA were processed for IHC for KtdT (red) and ribosomal protein S6 used to define cytosol space (green). A, Macro view of the region containing VTA and PAG. The image was acquired with a 10× objective on confocal microscope. The dotted circles define the area where KtdT neurons were selected for analysis. B, Vehicle group: KtdT in VTA neurons (marked by arrows) are clearly observed as red sharp outlines (most likely in cell membranes). The image is a single focal plane of a Z-stack acquired with a 60× objective on confocal microscope. C, U50,488 group: the image was obtained similarly as in B. Note that U50,488-induced KtdT translocation in KtdT neurons are evidenced by simultaneously exhibiting broken and dotted KtdT surface outlines and punctate staining in cytosol space as contrasted by S6 staining. D, KtdT translocation was quantified with the method described in Extended Data Figure 8-1 and shown as graph. Each value represents mean ± SEM (n = 4). Data were analyzed with Student’s t test; ****p < 0.0001.