To investigate the role of TAAR1 in motor regulation and related neural circuits, an adeno-associated virus (AAV) carrying TAAR1-specific shRNA was delivered bilaterally into the striatum and substantia nigra of Sprague-Dawley (SD) rats via stereotaxic injection to establish a TAAR1 knockdown (TAAR1-KD) rat model (Figure 1A). Western blotting and immunohistochemistry (IHC) analyses revealed that TAAR1 protein expression in both the striatum and substantia nigra was significantly reduced in the TAAR1-AAV group compared to the Control (CN) and negative control (NC-AAV) groups (P < 0.001; Figures 1B,C), indicating successful model establishment.

To assess whether the knockdown of TAAR1 in substantia nigra and striatum induces schizophrenia-like behavior, spontaneous locomotor activity was measured in TAAR1-KD rats using the open field test (OFT). The results showed that spontaneous locomotor activity was significantly increased in the TAAR1-KD group compared to the NC-AAV group (Wolinsky et al., 2006) (P < 0.01; Figure 2), suggesting that TAAR1 may be involved in modulating motor behavior within the nigrostriatal circuit. This finding provides a behavioral basis for further investigation into the role of TAAR1 in schizophrenia-related motor phenotypes and underlying neural mechanisms.

TAAR1 is a G protein-coupled receptor known to couple to Gs and Gq proteins (Xu et al., 2023), whose activation is implicated in modulating protein kinase A (PKA) and protein kinase B (AKT)/Glycogen Synthase Kinase-3β (GSK-3β)pathway (Xie and Miller, 2007; Espinoza et al., 2015).

To further investigate the in vivo signaling mechanisms of TAAR1, key signaling molecules were systematically examined in the TAAR1-KD rat model. Immunohistochemistry (IHC) analysis revealed that the phosphorylation levels of protein kinase A (PKA) (Zhang et al., 2018), cAMP Response Element-Binding Protein (CREB) (Wang et al., 2018), and Dopamine- and cAMP-regulated neuronal phosphoprotein 32 kDa (DARPP32) (which can be phosphorylated by PKA at Thr34) (Matsuyama et al., 2002) were significantly reduced in the striatal tissue of TAAR1-KD rats compared to NC-AAV group (Figure 3A), indicating that TAAR1 deficiency impairs its canonical cAMP/PKA signaling pathway and downstream regulatory network. In the substantia nigra, a similar reduction in the phosphorylation levels of PKA, CREB, and DARPP-32 was observed (Figure 3B), which was consistent with the changes observed in the striatum.

Since IHC results revealed a significant downregulation of the PKA pathway in the striatum and substantia nigra following TAAR1 knockdown, brain tissues from the striatum and substantia nigra were selected for Western blot validation. The results demonstrated that, compared NC-AAV group, the phosphorylation levels of PKA, CREB, and DARPP-32 were all significantly reduced in the striatal tissues of TAAR1-KD rats (Figure 4).

Notably, the total and phosphorylated protein levels of AKT and GSK-3β showed no significant differences between TAAR1-AAV and NC-AAV groups (see Supplementary Figure S1), suggesting that TAAR1 knockdown did not significantly alter the activity of the AKT/GSK-3β pathway in the striatum. Furthermore, the expression of other relevant signaling molecules, including ERK1/2 (Shi et al., 2019), MEK1/2 (Michael et al., 2019), and BDNF (Zheng et al., 2023), was examined, and no significant alterations were observed.

We designed and synthesised Selutaront, characterising and validating it via proton nuclear magnetic resonance (¹HNMR), high-performance liquid chromatography (HPLC), and liquid chromatography-mass spectrometry (LC-MS) (see Supplementary Figures S2–S4).

The Selutaront exhibited potent TAAR1 agonist activity in stably transfected human TAAR1-expressing cells, with an EC₅₀ of 2.33 μM and an E_max of 83% (Figure 5). This confirms that Selutaront is a potent TAAR1 agonist. To assess target specificity, agonist/antagonist activity of Selutaront was evaluated against schizophrenia-relevant receptors (namely 5-HT_1A, D_2L, and D_2S). Functional assays revealed no significant agonist or antagonist activity at concentrations up to 100 μM (EC₅₀/IC₅₀ > 100 μM) for these receptors (Table 1). These findings demonstrate that Selutaront is a highly selective TAAR1 agonist.

Molecular docking simulations were conducted using the human TAAR1 crystal structure (PDB ID: 8JLO) in Discovery Studio to elucidate the structural mechanism underlying Selutaront-TAAR1 interactions. The simulations revealed stable binding of Selutaront within the TAAR1 orthosteric pocket through multiple interactions (Figures 6A,B): The protonated amino group formed an attractive charge with Asp103, which determines primary binding orientation. Its thiophene moiety engaged in dual π-sulfur and T-shaped π-π interactions with Phe267. A conventional hydrogen bond occurred with Ser107 via the thiophene sulfur atom. A carbon-hydrogen bond formed between the methyl group and Ile290, with additional π-alkyl interactions between the thiophene ring and Ile104. To validate computational predictions, TAAR1 point mutants were generated for functional analysis (Figure 6C). Mutations of critical residues (D103A/F267A/S107A) nearly abolished Selutaront-mediated TAAR1 activation, confirming their essential role in ligand recognition. Mutations of fine-tuning residues (I104A, I290A) reduced cAMP accumulation, indicating their involvement in binding energy optimization and conformational modulation. These findings provide precise structure-based targeting strategies for developing high-selectivity TAAR1 agonists.

To evaluate the antipsychotic effects of Selutaront, behavioral tests were conducted in a C57BL/6J mouse model. Thirty minutes after a single oral administration of Selutaront (0.01, 0.03, or 0.1 mg/kg), schizophrenia-like behaviors were induced via an intraperitoneal injection of the NMDA receptor antagonist MK-801 (0.8 mg/kg) (Kruk-Slomka et al., 2017; Fan et al., 2018). The results demonstrated that Selutaront (0.03, 0.1 mg/kg) did not affect the animals’ spontaneous activity and exhibited a dose-dependent suppression of MK-801-induced schizophreniform behavioural abnormalities (Figures 7A–C), confirming its efficacy in alleviating core positive symptoms of schizophrenia. To investigate the mechanism of action, a TAAR1 partial knockdown model was subsequently established by stereotaxically injecting a TAAR1- AAV virus into the striatum and substantia nigra of SD rats. The open field test showed that TAAR1 knockdown resulted in increased locomotor activity in rats (Figure 2). However, following oral administration of Selutaront (5 mg/kg), the expected suppression of activity observed in the control group was absent in the knockdown group (Figure 7D). This finding indicates that the antipsychotic-like behavioral effects of Selutaront are dependent on an intact TAAR1 signaling pathway, providing key mechanistic evidence for its potential as a targeted therapy for schizophrenia.

The above findings suggested that TAAR1 exerted regulatory effects on the PKA/DARPP32 and PKA/CREB pathways. To further investigate whether this regulation is mediated by TAAR1 agonism, we treated MK-801-induced hyperactive mice with the TAAR1 agonist Selutaront. As shown in Figure 8, compared to the Control group, administration of MK-801 (0.8 mg/kg) alone significantly reduced the phosphorylation levels of p-PKA (Thr197) and p-CREB (Ser133) (P < 0.01). In contrast, pretreatment with Selutaront (10 mg/kg) significantly reversed the MK-801-induced decrease in p-PKA and p-CREB levels (P < 0.01 vs. the MK-801 group), restoring their phosphorylation to levels that showed no statistically significant difference from the Control group (P > 0.05). These results confirm that by functionally antagonizing the effects of MK-801, Selutaront molecularly rescues and restores the suppressed TAAR1 downstream cAMP-PKA-CREB signaling pathway, thereby providing direct mechanistic evidence for its ameliorative effects on schizophrenia-like behaviors.

The above findings thoroughly analyze the downstream pathways mediated by TAAR1 and the therapeutic effects of its agonist (Selutaront) on schizophrenia. To evaluate the drug potential of Selutaront, we analyzed its pharmacokinetic behavior in vivo and in vitro (Table 2). Pharmacokinetic studies in rats revealed favorable oral absorption and systemic exposure characteristics for Selutaront. Following oral administration (5 mg/kg), the following parameters were observed: time to maximum concentration (T_max) = 0.139 h, the terminal elimination half-life (T_1/2) = 3.46 h, area under the concentration time curve (AUC) = 4,582 h·ng/mL, and absolute oral bioavailability (F) = 48.1% (Figure 9).

Selutaront exhibited favorable membrane permeability and metabolic stability (Tables 3, 4). It demonstrated high permeability (P_app A→B = 36.0 × 10⁻⁶ cm/s) and low efflux risk (efflux ratio [ER] = 0.788, significantly below the threshold of 2.0), indicating minimal risk of efflux transporter-mediated intestinal efflux. Selutaront displayed substantially greater metabolic stability in human liver microsomes compared to rodent liver microsomes: 103% remaining at 60 min, half-life (T_1/2) >186 min, and intrinsic clearance (CL_int) <6.75 mL/min/kg (low clearance). The human hepatic intrinsic clearance was less than one-sixth of the murine value (human: <6.75 vs. mouse: 43.6 mL/min/kg), with the half-life exceeding 1.5-fold that in mice (human: >186 vs. mouse: 126 min). These properties collectively indicate favorable oral absorption potential in humans, likely resulting in low hepatic first-pass effect, extended in vivo half-life, and high oral bioavailability, providing crucial pharmacokinetic advantages for clinical development.

Selutaront was provided by WuXi AppTec (Shanghai, China). (+)-MK-801 hydrogen maleate was purchased from Sigma-Aldrich (St. Louis, MO, United States). Fetal Bovine Serum (FBS, Value-Added) and Trypsin-EDTA (0.25%) solution were obtained from Gibco™, Thermo Fisher Scientific (Waltham, MA, United States). The HitHunter® cAMP Assay for Small Molecules kit was sourced from Eurofins Discovery (Fremont, CA, United States), and the FLIPR® Calcium 6 Assay Kit was from Molecular Devices (San Jose, CA, United States). Cell culture media, including Ham’s F-12K (Kaighn’s) Medium and MEM (Minimum Essential Medium), were supplied by Invitrogen (Camarillo, CA, United States). Antibiotics (Penicillin, Streptomycin, and Hygromycin B) were purchased from Beyotime Biotechnology (Shanghai, China).

The TAAR1-targeting shRNA adeno-associated virus (AAV9-TAAR1-shRNA; target sequence: 5′-GCG​CCA​CAA​AGC​AAG​GAA​ACA-3′) and the control virus (AAV9-U6-scramble-shRNA-EGFP) were provided by GeneChem (Shanghai, China) at a titer of 1.50 × 10¹³ vector genomes (vg)/mL.

Male Sprague-Dawley rats (6–8 weeks; 190–210 g) and male C57BL/6J mice (6–8 weeks; 18–22 g) were procured for the purposes of this study by Vital River Company (Beijing, China). All animals resided under specific pathogen-free (SPF) conditions in a climate-controlled environment with a 12-h light/dark cycle (8:00–20:00), constant temperature of 22 °C ± 2 °C, and 40%–70% humidity. Standard rodent diet and water were provided ad libitum. All animals underwent a 1-week habituation period before experimentation. All behavioral assessments occurred between 08:00 and 14:00 h.

All animal experiments described in this manuscript were conducted in accordance with relevant guidelines and regulations. The study was jointly approved by the Laboratory Animal Care and Use Committee of Yantai University and the Laboratory Animal Care and Use Committee of Shandong Luye Pharmaceutical Group.

Adult male Sprague-Dawley rats were anesthetized via intraperitoneal (i.p.) injection of sodium pentobarbital (40 mg/kg; prepared in sterile 0.9% saline) and subsequently immobilized in a Stoelting stereotaxic apparatus (Model 51700D, Stoelting Co., Wood Dale, IL, United States). Bilateral injections targeting striatum and substantia nigra were performed using coordinates relative to bregma (Paxinos and Watson, 6th ed.) (Paxinos and Watson, 2006): Striatum: Antero-posterior (AP): +0.43 mm, Medio-Lateral (ML): ±2.7 mm, Dorso-Ventral (DV): −5.4 mm; Substantia Nigra: Antero-Posterior (AP): −5.0 mm, Medio-Lateral (ML): ±2.1 mm, Dorso-Ventral (DV): −8.5 mm. Viruses (2 μL/site) were infused at 0.3 μL/min via a Hamilton microsyringe. The needles were left in place for a period of 5 minutes following the infusion before being withdrawn. Surgical incisions were sutured, and penicillin G (50,000 IU/kg, diluted in saline, i.p.) was administered as infection prophylaxis. Animals were singly housed for 3 weeks to permit maximal viral expression, with daily monitoring for surgical recovery and health status.

Prior to testing, animals were habituated in the behavioral testing room for ≥2 h. The rats were individually positioned within an open-field arena (50 × 50 × 40 cm; length × width × height) connected to the TopScan® behavioral monitoring system (Clever Sys Inc., Reston, VA, United States). Testing was conducted under low illumination (≤50 lux) with minimal ambient noise (<50 dB). Total distance traveled (mm) over 30 min was recorded and analyzed using the TopScan system. After each experimental group, feces were removed, and the arena was wiped with 75% alcohol to eliminate potential confounding odors.

Striatal and substantia nigra tissues were isolated from Sprague-Dawley (SD) rats transfected with TAAR1-AAV (n = 6). Tissues were flash-frozen in liquid nitrogen and homogenized in RIPA lysis buffer containing protease and phosphatase inhibitors using sonication. Following protein quantification by BCA assay, 50 μg total protein was separated by 4%–20% gradient SDS-PAGE and transferred to PVDF membranes using wet transfer methodology. All membranes were blocked with a TBST solution containing 5% non-fat dry milk. Subsequently, the membranes were incubated with the corresponding primary antibodies overnight at 4 °C. After washing, they were incubated with the HRP-labeled secondary antibodies (1:5000) for 1 h at room temperature. For the target proteins that needed to be detected through “membrane stripping and re-incubation”, after the first chemiluminescence imaging was completed, the membranes were treated with a mild membrane stripping buffer for 10 min to thoroughly elute the bound primary and secondary antibodies. After stripping, the membranes were thoroughly washed with TBST and the blocking step was repeated. Then, they were sequentially incubated with the primary antibodies against the next target proteins. After each stripping and re-incubation, the membranes were developed using the BeyoECL Plus chemiluminescence substrate (Beyotime, Shanghai, China; Cat #P0018S) and imaged using the SageCapture Pro 5.0 system.

The signal intensities of all protein bands were quantitatively analyzed using ImageJ software. TAAR1 expression levels were normalized to β-actin (target protein band intensity/internal reference band intensity). The expression levels of the target phosphorylated proteins were normalized to their respective total proteins and verified using β-actin as an internal reference.

The antibodies used were as follows:

TAAR1 (rabbit, 1:1000, Immunoway, Product # PA5-115999)

The SD rats were rendered anaesthetic with sodium pentobarbital (40 mg/kg, intraperitoneal) and subsequently perfused transcardially with Phosphate Buffered Solution (pH 7.4). Brain tissues were post-fixed in 4% paraformaldehyde for a period of 48 h, following which they underwent a series of dehydration steps using a graded ethanol solution. The tissues were then embedded in paraffin. Serial sections (4 μm) were produced and exposed to citrate buffer in a microwave oven to retrieve antigens. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide, followed by blocking of non-specific binding sites using 5% normal goat serum. Sections were subsequently incubated overnight at 4 °C in a humidified chamber with primary antibody, then with species-matched horseradish peroxidase (HRP)-conjugated polymer secondary antibody for 30 min at room temperature. Following DAB chromogenic development, sections were mounted in neutral balsam. From the target brain region sections of each rat, select three representative sections with adequate spacing. Image analysis of tissue sections was conducted using ImageJ software (v2.14.0); National Institutes of Health, Bethesda, MD, United States (Varghese et al., 2014), with target protein expression levels quantitatively assessed through the H-score system (Wen et al., 2024). The final H-score value for each animal was obtained by averaging the scores from its three sections.

H-score was calculated using the formula: H ‐ score = ∑ 1 × % weak intensity + 2 × % moderate intensity + 3 × % strong intensity

Antibodies employed in this study:

TAAR1: Rabbit, 1:200 (Thermo Fisher Scientific, Waltham, MA, United States; Cat # PA5-115999)

The assay utilized cAMP Hunter™ CHO-K1 AGTRL1 Gi cells stably engineered to express TAAR1. The maintenance of cells was undertaken in Ham’s F-12K medium containing 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. The cells were distributed into 384-well plates at a volume of 20 μL per well, following which they were subjected to a preincubation process at a temperature of 37 °C. Following the addition of test compounds at graded concentrations and a 10-min incubation, cAMP levels were quantified. Tyramine (100 μM) served as the maximal agonist control (MAX). Agonist activity was calculated as: a ctivity % = meanRLU test   sample ‐ meanRLU   vehicle   control meanRLU MAX   control ‐ meanRLU   vehicle   control × 100 %

The performance of the assays was undertaken utilising Flp-In™-CHO cells that exhibited stable expression of the D_2L or D_2S receptor. Procedures for cell culture, seeding, compound treatment, and detection were conducted as described for the TAAR1 assay. Haloperidol (10 µM) served as the maximum inhibitory control (MAX).

Inhibitory activity was calculated using the formula: inhibition % = meanRLU test   sample ‐ meanRLU   vehicle   control meanRLU MAX   control ‐ meanRLU   vehicle   control × 100 %

The assays utilized 5-HT_1A Gα15-NFAT-bla CHO-K1 cells. These cells were cultured in MEM containing 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, and 600 μg/mL hygromycin B. Cells were exposed to serial dilutions of test compounds for a period of 2 hours at a temperature of 37 °C. Serotonin (5-HT, 10 µM) served as the maximal agonist control (MAX). Calcium flux was measured utilising the FLIPR® Calcium 6 Assay Kits, with agonist activity calculated as previously described for the TAAR1 assay.

Protein-ligand interactions were analyzed using the LibDock module in Discovery Studio 2019 (Dassault Systèmes BIOVIA, San Diego, CA, United States). TAAR1 crystal structure (PDB ID: 8JLO) was preprocessed through hydrogen atom addition, charge assignment (CHARMm force field), and energy minimization (Smart Minimizer algorithm with Generalized Born implicit solvent model). The optimized receptor structure was subsequently utilized for docking. The active site was defined as the co-crystallized ligand’s binding pocket. All remaining parameters employed LibDock’s default settings.

HEK293-cAMP-biosensor-22F-H-1B2 cells in logarithmic growth phase were transfected with TAAR1 or mutant plasmids (2 μg per transfection) using jetPRIME^® reagent. Following 36-h incubation, cells were seeded into 96-well plates and pretreated with equilibration medium containing 2% GloSensor™ cAMP Reagent for 1 h. Subsequently, serially diluted Selutaront (initial concentration 100 μM, 3.16-fold dilutions, 10 concentrations), Forskolin (10 μM), and ZH8651 (100 μM) were administered, with triplicate wells per treatment group. Kinetic luminescence readings were acquired every minute for 30 min using a microplate luminometer to quantify compound-induced cAMP dynamics. Data were normalized to forskolin (100% cAMP response) and vehicle (0% response) controls.

Selutaront and (+)-MK-801 hydrogen maleate were dissolved in physiological saline (0.9% NaCl) and administered at 0.1 mL per 10 g body weight.

Forty male C57BL/6J mice were allocated to four groups for the experimental phase and habituated to the behavioral testing room for 2 h prior to experimentation. Treatments were physiological saline or Selutaront (0.01, 0.03, or 0.1 mg/kg; i.g.). Baseline locomotor activity was assessed during a 30-min open field test in an arena (20 cm × 20 cm × 40 cm, length × width × height, GeneandI, Beijing, China). Following baseline assessment, all animals received MK-801 (0.8 mg/kg, i.p.), and total movement distance was quantified over a subsequent 30-min observation period.

To investigate the molecular mechanism by which Selutaront antagonizes MK801-induced schizophrenia-like behaviors, the effects on the phosphorylation levels of key signaling pathway proteins were examined.

Thirty male C57BL/6J mice (6–8 weeks; 18–22 g) were randomly assigned to three groups (n = 10/group): (1) Control group: administered saline (i.g.), followed 30 min later by an intraperitoneal (i.p.) injection of saline; (2) MK-801 group: administered saline (i.g.), followed 30 min later by an i.p. injection of MK-801 (0.8 mg/kg); (3) Selutaront +MK-801 group: administered Selutaront (10 mg/kg, dissolved in saline, i.g.), followed 30 min later by an i.p. injection of MK-801 (0.8 mg/kg). Thirty minutes after the final injection (MK-801 or saline), all mice were deeply anesthetized and euthanized by decapitation. The target brain regions were rapidly dissected, flash-frozen in liquid nitrogen, and stored at −80 °C for subsequent analysis. The detailed experimental procedures for Western blotting, including tissue lysis, protein concentration determination (BCA method), electrophoresis, membrane transfer, antibody incubation, and signal detection, were consistent with the methods described in the “Western blotting” section of this article. The primary antibodies used included: rabbit anti-p-PKA (Thr197) (1:1000), rabbit anti-total PKA (1:1000), rabbit anti-p-CREB (Ser133) (1:1000), rabbit anti-total CREB (1:1000), and mouse anti-β-actin (1:5000).

Following single-dose administration of Selutaront by oral gavage (1 mg/kg) or intravenous injection (5 mg/kg) to Sprague-Dawley rats, plasma concentrations were determined. Blood samples were collected via the retro-orbital venous plexus at the following time points post-administration: 0.083, 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h. The plasma concentrations of Selutaront were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS) with an Agilent 1100 series HPLC interfaced to a TSQ Quantum Access triple quadrupole mass spectrometer (Thermo Fisher Scientific, San Jose, CA, United States). The processing of drug-concentration data was undertaken using a non-compartmental model with WinNonlin™ 6.3 (Pharsight, Mountain View, CA, United States).

Pharmacokinetic (PK) parameters, including elimination half-life (T_1/2), maximum plasma concentration (C_max), area under the curve (AUC), and mean residence time (MRT), were calculated.

Caco-2 cells were cultivated in MEM that had been enriched with 20% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. For the purpose of transport studies, cells were seeded at 1 × 10⁵ cells/well on 96-well Transwell^® polycarbonate membranes and maintained for a period of 24 days to establish fully differentiated monolayers, with medium changes performed every 48 h. Test compounds were applied apically with or without verapamil (2 μM). After a 120-min incubation, samples from apical and basolateral compartments were collected. Urea content was quantified by LC-MS/MS (Agilent 1290/6470 system). Apparent permeability coefficients (Papp, cm/s) and efflux ratios (ER = Papp (B→A)/Papp (A→B)) were calculated to assess permeability and transporter involvement.

The metabolic stability of Selutaront was assessed by incubation with liver microsomes (mouse, rat, and human) and an NADPH regeneration system at 37 °C for 5, 10, 20, 30, or 60 min. The termination of reactions was achieved through the use of ice-cold acetonitrile, which was employed as the internal standard for tolbutamide. Testosterone (a CYP3A4 substrate) and dextromethorphan (a CYP2D6 substrate) were included as positive controls. The remaining concentrations of the compounds were determined using LC-MS/MS. Chromatogram processing, including retention time analysis, data acquisition, and integration for test compounds and positive controls, was performed using Analyst software (AB Sciex, Framingham, MA, United States).

All statistical analyses were performed using Prism 9 (GraphPad Software, La Jolla, CA, United States). Data are presented as mean ± standard error of the mean (SEM). Intergroup differences were assessed by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Statistical significance was defined as P < 0.05.