PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease Authors: Kuan-Yin Tseng, Tung-Tai Kuo, Pi-Kai Chang, Jian-Liang Chou, Eagle Yi-Kung Huang, Nigel H Greig, Ho-Il Choi, Lars Olson, Barry J Hoffer, Yuan-Hao Chen Published: 2026 ABSTRACT BackgroundNeuropsychiatric symptoms such as anxiety and depression substantially impair quality of life in Parkinson's disease (PD), yet the underlying neural circuits remain poorly defined. The MitoPark (MP) mouse, a dopaminergic mitochondrial dysfunction model, recapitulates both motor and non-motor features of PD.ObjectiveTo determine whether the sustained-release GLP-1 receptor agonist PT320 (exenatide) alleviates anxiety- and depression-like behaviors in MP mice and to identify the involved neural substrates.MethodsThe temporal progression of anxiety- and depression-like behaviors was characterized in MP mice. PT320 was administered biweekly starting at either 5 weeks (early treatment) or 15 weeks (late treatment), with longitudinal evaluation until 20 weeks. Behavioral outcomes were correlated with molecular, transcriptomic, and neurochemical analyses in the nucleus accumbens (NAc), including Western blotting, bulk RNA sequencing, fast-scan cyclic voltammetry, and tyrosine hydroxylase immunostaining.ResultsEarly PT320 treatment effectively prevented the emergence of anxiety- and depression-like phenotypes in MP mice. Behavioral improvement was associated with restoration of BDNF signaling and activation of the Akt–CREB pathway in the NAc. Transcriptomic analysis revealed increased expression of Akt3, CREB, BDNF, and TrkB, along with modulation of genes related to mitochondrial homeostasis. Late PT320 treatment partially ameliorated neuropsychiatric deficits, coinciding with enhanced phasic dopamine release and recovery of tyrosine hydroxylase expression in the NAc.ConclusionsThese findings identify the NAc as a critical regulator of affective disturbances in this PD model. By restoring neurotrophic signaling and dopaminergic function, PT320 represents a promising therapeutic strategy for PD-related anxiety and depression. Introduction Parkinson's disease (PD) is not limited to its characteristic motor deficits. 1 A wide range of non-motor symptoms (NMSs) appear throughout the disease course and impact most PD patients. 2 These symptoms involve several functional domains, including mood and emotional regulation, cognitive processing, sensory perception, sleep architecture, autonomic function, and pain.1,3 Although these manifestations are common, they are often overlooked in clinical settings and represent major contributors to disability, reduced quality of life, and increased healthcare demands.3–6 Some NMSs, such as hyposmia, constipation, REM sleep behavioral disorder (RBD), and depressive symptoms, can emerge years before the onset of classical motor signs and are increasingly regarded as indicators of the prodromal phase of PD.7–10 Among the many NMSs, affective disturbances are especially debilitating. Depression in particular can impair daily functioning to a greater extent than motor symptoms. 11 In addition, individuals with PD have a substantially higher risk of developing cognitive decline compared with healthy peers. 12 Current findings suggest that dysregulation of monoaminergic neurotransmission, 13 persistent neuroinflammation, 14 and abnormalities in the hypothalamic pituitary adrenal axis 15 may contribute to these affective and cognitive difficulties. Given the significant burden caused by NMSs, improving their detection and treatment has become an essential goal in PD management. Early identification and targeted therapeutic strategies address a major unmet clinical need and offer opportunities to enhance long-term outcomes. These efforts can also support a more comprehensive and patient-centered approach to PD care. A variety of animal models have been developed to investigate depressive symptoms associated with PD. Neurotoxin-based models such as MPTP in mice or non-human primates can display depression-like phenotypes; however, the acute nature of the toxin exposure limits their ability to reproduce the gradual neurodegenerative process and the broad spectrum of non-motor manifestations observed in PD.16–18 The 6-hydroxydopamine (6-OHDA) rat model also shows reduced exploratory behavior and increased immobility in behavioral despair paradigms, yet the unilateral lesion design does not fully represent the widespread dopaminergic degeneration characteristic of PD.19,20 In contrast, the MitoPark (MP) mouse model offers a progressive phenotype that more closely resembles the natural evolution of affective disturbances in PD. MP mice carry a conditional deletion of mitochondrial transcription factor A (Tfam) in midbrain dopamine neurons, leading to mitochondrial dysfunction and progressive degeneration of both nigrostriatal and mesolimbic dopaminergic circuits. 21 Behavioral studies have shown that MP mice develop reduced locomotor activity, cognitive impairment, and clear depression- and anxiety-like behaviors as the disease advances.22–24 These affective abnormalities are partially reversed by the antidepressant desipramine and are accompanied by alterations in CREB phosphorylation, BDNF expression, and catecholamine-related transcriptional pathways, supporting the construct validity of this model for studying PD-related depression. 22 Importantly, this temporal pattern closely mirrors the clinical course of PD, in which non-motor symptoms commonly coincide with motor deterioration. 22 The nucleus accumbens (NAc) is increasingly recognized as a key region contributing to depressive symptoms in PD.25,26 As an essential component of the mesolimbic dopaminergic system, the NAc integrates prefrontal cortical inputs with midbrain dopamine signals to regulate motivation, reward processing, and emotional responses. Clinical neuroimaging studies have reported reduced NAc volume and diminished functional connectivity in PD, which correlate with depression, anxiety, apathy, and cognitive impairment. 27 In preclinical work, disrupted dopaminergic receptor signaling, glutamatergic dysfunction and synaptic plasticity deficits within the NAc have all been implicated in depression-related behaviors.25–27 Conversely, deep brain stimulation of the NAc has been shown to alleviate treatment-resistant depression by enhancing BDNF expression. 26 Glucagon-like peptide-1 (GLP-1) receptor agonists, like Exenatide, widely used in the treatment of diabetes and weight loss, and have recently gained attention as potential neuroprotective agents in PD. 28 Our previous work demonstrated that extended-release Exenatide (PT320: a once biweekly subcutaneous formulation) slows dopaminergic degeneration in MP mice, preserves mitochondrial function, and delays the onset of motor deficits. 29 GLP-1 receptor activation also modulates central pathways involved in inflammation, oxidative stress, and neural plasticity, 30 as well as has been associated with improvements in stress-related behavioral abnormalities. 31 Furthermore, prior studies indicate that GLP-1 receptor agonist treatment mitigates depression-like behaviors, potentially via enhancement of neurotrophic signaling and increased resilience against cellular stress.32,33 Based on the above rationale, we sought to determine whether progressive dopaminergic denervation, together with reduced neurotrophic signaling in the NAc, contributes to the development of depression- and anxiety-like behaviors in MP mice. Based on the above foundation, the effects of PT320 on depressive or anxious symptoms and its potential underlying mechanisms were further investigated. Materials and methods Animals & treatment Animals and Treatment Groups: Male MitoPark (MP) mice (a C57BL/6J background strain with dopaminergic neuron-specific Tfam deletion) and littermate wild-type (WT) controls were used. The breeding strategy employed to generate MP mice and rational for using only males at this point have been detailed in previous publications,34–36 as male MP mice display a PD phenotype earlier and more severely across a broad array of measures, over a 5–24 week time frame, than female MP mice consequent to the protective effects of estrogen. 24 Hence, only male mice were used in the current study. The MP mice utilized in this series of experiments were heterozygous for DAT-cre expression (DAT/DATcre), and homozygous for the loxP-flanked Tfam gene (TfamloxP/TfamloxP). Breeding pairs used for the generation of MP mice were sourced from a colony maintained at the NIDA (NIH) Intramural Program. Subsequently, these MP mice were sent to the NDMC in Taiwan, where comprehensive genotyping procedures were undertaken. Mice were housed under standard conditions with ad libitum access to food and water. We implemented two PT320 treatment paradigms. In the early intervention group, MP mice received PT320 beginning at 5 weeks of age (prior to the onset of PD-like symptoms) and continuing up to 20 weeks. In the late intervention group, MP mice were started on PT320 at 15 weeks (when depression-like behaviors are already present) and treated until 20 weeks. PT320 is a sustained-release GLP-1 receptor agonist based on Exenatide; it was prepared freshly and administered via subcutaneous injection at a dose calibrated to provide continuous exposure. The PT320 dose and frequency were based on our prior studies to provide a clinically translatable dose29,37 in line with FDA guidelines. Specifically, clinical grade PT320 was administered subcutaneously once every 2 weeks to provide an Exenatide dose of 0.6 mg/kg. As exenatide accounts for only 2% of the drug available within the PT320 formulation, the remaining 98% being polymers to control the Exenatide release rate from PT320; the total administered PT320 dose was 30 mg/kg biweekly. MP mice in the control group received vehicle (PBS) injections according to the same dosing schedule as the PT320-treated group. WT mice (with or without PT320) were also monitored to confirm that PT320 does not alter normal behavior (see Supplementary Figure 1). Those male mice were housed at 25˚C with a 12/12 light/dark cycle and continuous water and food supply in the animal facility at National Defense Medical University (NDMU) with full AAALAC accreditation. All animal experiments in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of the National Defense Medical University under approval number IACUC-23-027. For comparison, additional groups of 20-week-old MP mice received either L-DOPA (co-administered with benserazide) to evaluate dopaminergic restoration of behavior, or the selective serotonin reuptake inhibitor citalopram (20 mg/kg) to examine serotonergic modulation of neuropsychiatric symptoms (Figure 1A). These treatments served as reference controls to benchmark the therapeutic effects of PT320 against established dopaminergic and serotonergic interventions. The number of animals used for each experimental group and assay is summarized in Table 1 and Supplementary Table 1. Progressive development of depression- and anxiety-like behaviors in MP mice. (A) Experimental protocol for age-dependent behavioral assessments. (B) In the open-field test, MP mice spent significantly less time exploring the center zone starting at 12 weeks of age, indicating increased anxiety-like behavior. (C) In the marble burying test (MBT), MP mice displayed an increased number of buried marbles beginning at 14 weeks of age, further supporting an age-dependent increase in anxiety-like behavior. (D) MP mice exhibited significantly increased immobility time in the forced swim test (FST) beginning at 8 weeks of age compared with wild-type (WT) controls. (E) In the tail suspension test (TST), MP mice showed a significant increase in immobility time starting at 10 weeks of age, indicating progressive depression-like behavior. (F) Citalopram treatment increased center-zone exploration, indicating reduced anxiety-like behavior in MP mice. (G) The elevated number of buried marbles observed in MP mice was significantly reduced following citalopram administration. (H) Citalopram treatment reduced immobility time in the FST, indicating an antidepressant-like effect. (I) Improved performance was also observed in the TST following citalopram treatment. (J) 5-HTP treatment increased center-zone exploration, indicating attenuation of anxiety-like behavior. (K) 5-HTP also reduced the number of buried marbles in the MBT. (L) Administration of 5-HTP significantly reduced immobility time in the FST in MP mice. (M) Horizontal locomotor activity was not significantly altered by 5-HTP treatment. (N) Vertical activity likewise showed no significant difference between 5-HTP–treated and untreated MP mice. Data are presented as mean ± SEM with individual values shown. Statistical analyses were performed using one-way or two-way analysis of variance (ANOVA), followed by Bonferroni post hoc tests for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001 vs. WT; #p < 0.05, ###p < 0.001 vs. MP. Table 1.Summary of sample sizes (N values) for all panels presented in the main figures. For each figure and corresponding panel, the number of animals included in the analysis is indicated.FigurePanelGroup (N)Assay Figure 1. Depression- and anxiety-like behaviors in MP mice Figure 1 B (4W–20W)WT (8), MP (8)Center zone activity Figure 1 C (4W–20W)WT (4), MP (4)Marble burying test Figure 1 D (4W–20W)WT (8), MP (8)Forced swim test Figure 1 E (4W–20W)WT (8), MP (8)Tail suspension test Figure 1 F (20W)WT (8), MP (8), MP+Citalopram (10)Center zone activity Figure 1 G (20W)WT (10), MP (14), MP+Citalopram (9)Marble burying test Figure 1 H (20W)WT (8), MP (8), MP+Citalopram (15)Forced swim test Figure 1 I (20W)WT (8), MP (7), MP+Citalopram (8)Tail suspension test Figure 1 J (20W)WT (8), MP (8), MP+5-HTP (3)Center zone activity Figure 1 K (20W)WT (4), MP (3), MP+5-HTP (3)Marble burying test Figure 1 L (20W)WT (8), MP (8), MP+5-HTP (3)Forced swim test Figure 1 M (20W)WT (8), MP (8), MP+5-HTP (3)Horizontal activity Figure 1 N (20W)WT (8), MP (8), MP+5-HTP (3)Vertical activity Figure 2. Effects of early PT320 treatment on behavioral phenotypes Figure 2 B (4W–20W)WT (8), MP (8), MP+PT320 (9)Forced swim test Figure 2 C (4W–20W)WT (8), MP (8), MP+PT320 (9)Tail suspension test Figure 2 D (8W)WT (4), MP (4), MP+PT320 (4)Marble burying test Figure 2 D (14W)WT (4), MP (4), MP+PT320 (4)Marble burying test Figure 2 D (20W)WT (4), MP (4), MP+PT320 (4)Marble burying test Figure 2 E (8W)WT (8), MP (8), MP+PT320 (9)Center zone activity Figure 2 E (14W)WT (8), MP (8), MP+PT320 (4)Center zone activity Figure 2 E (20W)WT (8), MP (8), MP+PT320 (8)Center zone activity Figure 3. Effects of early PT320 treatment on p-Akt/p-CREB/BDNF expression Figure 3 B (5W)WT (7), MP (7)WB (p-Akt/Akt ratio, NAc) Figure 3 B (8W)WT (6), MP (6), MP+PT320 (6)WB (p-Akt/Akt ratio, NAc) Figure 3 C (5W)WT (7), MP (7)WB (p-CREB/CREB ratio, NAc) Figure 3 C (8W)WT (6), MP (6), MP+PT320 (6)WB (p-CREB/CREB ratio, NAc) Figure 3 D (5W)WT (7), MP (7)WB (BDNF, NAc) Figure 3 D (8W)WT (8), MP (8), MP+PT320 (8)WB (BDNF, NAc) Figure 3 F (5W)WT (6), MP (6)WB (p-Akt/Akt ratio, NAc) Figure 3 F (8W)WT (6), MP (6), MP+PT320 (6)WB (p-Akt/Akt ratio, NAc) Figure 3 G (5W)WT (7), MP (7)WB (p-CREB/CREB ratio, NAc) Figure 3 G (8W)WT (6), MP (6), MP+PT320 (6)WB (p-CREB/CREB ratio, NAc) Figure 3 H (5W)WT (7), MP 7)WB (BDNF, NAc) Figure 3 H (8W)WT (6), MP (6), MP+PT320 (6)WB (BDNF, NAc) Figure 5. Late PT320 effects on depression-like behaviors in 20W MP mice Figure 5 B (20W)WT (8), MP (8), MP+PT320 (7)Horizontal activity Figure 5 C (20W)WT (8), MP (8), MP+PT320 (7)Vertical activity Figure 5 E (20W)WT (8), MP (8), MP+PT320 (7)Center zone activity Figure 5 F (20W)WT (10), MP (14), MP+PT320 (7)Marble burying test Figure 5 G (20W)WT (8), MP (8), MP+PT320 (7)Forced swim test Figure 5 H (20W)WT (8), MP (7), MP+PT320 (7)Tail suspension test Figure 6. Late PT320 effects on dopamine release in the NAc of 20W MP mice Figure 6 B (20W)WT (5), MP (5), MP+PT320 (5)FSCV (Tonic, NAc) Figure 6 B (20W)WT (5), MP (5), MP+PT320 (5)FSCV (Phasic, NAc) Figure 7. Late PT320 effects on TH and p-CREB expression in the NAc of 20W MP mice Figure 7 B (20W)WT (3), MP (3), MP+PT320 (3)IHC (TH, NAc) Figure 7 D (20W)WT (4), MP (4), MP+PT320 (4)IHC (p-CREB, NAc) Figure 7 F (20W)WT (17), MP (17), MP+PT320 (17)WB (TH, NAc) Figure 7 G (20W)WT (4), MP (4), MP+PT320 (4)WB (p-Akt/Akt ratio, NAc) Figure 7 H (20W)WT (5), MP (4), MP+PT320 (5)WB (p-CREB/CREB ratio, NAc) Figure 7 I (20W)WT (7), MP (6), MP+PT320 (7)WB (BDNF, NAc) Summary of sample sizes (N values) for all panels presented in the main figures. For each figure and corresponding panel, the number of animals included in the analysis is indicated. Behavioral tests Mice underwent a series of tests to evaluate locomotor activity, anxiety-like behavior and depression-like behavior. Locomotor activity After allowing the mice to habituate in a low-noise environment for 1 h, we used activity boxes (45×45 cm) to evaluate their horizontal and vertical movements. The protocol involved the use of a grid of infrared beams to monitor mouse movements for 1 h, quantifying both the horizontal distance traveled and the frequency of rearing behaviors. The open-field test was performed in a clear acrylic arena to evaluate horizontal locomotor activity (total distance traveled) and vertical activity (rearing duration) as indicators of spontaneous movement. The percentage of time spent within the central zone relative to the periphery was analyzed as an index of anxiety-like behavior. Marble burying test This was performed by placing 20 glass marbles evenly on the bedding of a testing cage for a 30-min duration. The total number of marbles buried (out of 20) was quantified as a measure of anxiety-like behavior, where increased marble burying reflected heightened anxiety tendencies. Tail suspension test and forced swim test The tail suspension and forced swim tests, commonly employed to evaluate behavioral “despair” and coping responses under inescapable stress conditions, have been extensively characterized in prior studies. 38 In brief, during the tail suspension tests (TST), each mouse was individually suspended by the tail at a height of 30 cm using adhesive tape attached to a horizontal ring stand bar. Each 6-min test session was video-recorded and analyzed to quantify escape-oriented movements and bouts of immobility. The total duration of immobility for each mouse was recorded as an index of depression-like behavior. For the forced swim test (FST), mice were individually placed in a transparent glass cylinder (24×16 cm) containing 15 cm of water maintained at 25 °C. Behavior was recorded from the side of the cylinder for 6 min. Following an initial 2-min habituation period, the total duration of immobility was measured during the subsequent 4-min test phase. A mouse was considered immobile when it remained floating passively, making only minimal movements to keep its head above water, for at least 2.5 s (≥ 65% passive behavior). In both assays, immobility time (total seconds spent immobile) served as the primary behavioral endpoint, where increased immobility was interpreted as reflecting greater depression-like behavior. All behavioral testing was performed during the light phase, and mice were acclimated to the testing room prior to experimentation. Tests such as the FST and TST were conducted on separate days to minimize carryover effects. Where applicable, multiple age points (8, 14, and 20 weeks) were assessed to monitor the progression of behavioral deficits and the effects of PT320 treatment over time. Fast-scan cyclic voltammetry (FSCV) for measurement of DA dynamics in striatum/NAc brain slice preparations A detailed description of brain slice preparation is provided in our previous publications.39,40 In brief, following decapitation, the brain was rapidly removed and placed in an ice-cold, oxygenated cutting solution. Coronal sections (280 μm) containing the striatum and nucleus accumbens (NAc) were obtained using a vibrating microtome (VT100, Leica). Slices were then transferred to oxygenated artificial cerebrospinal fluid and equilibrated at 30°C for 30 min before being used for FSCV measurements. FSCV procedures followed our established protocols.41,42 Briefly, 280-μm coronal slices from 22-week-old mice were maintained in a 0.5-ml chamber at 31–33°C with continuous aCSF perfusion. A 7-μm carbon fiber electrode (Goodfellow Corp.) was positioned 100 μm deep in the dorsal striatum between bipolar stimulating electrodes (FHC Inc.). The electrode was backfilled with 4 M potassium acetate and 150 mM KCl. Dopamine release was measured using a triangular waveform scanned between −0.4 and 1.0 V at 400 V/s, applied every 100 ms, with electrical stimulation delivered at 25 Hz. Signals were acquired using National Instruments hardware and LabView-based analysis software (TarHeel CV). For NAc recordings, acute slices from 22 week WT, MP, and MP+PT320 mice were prepared, and monoamine release was evoked using the same carbon fiber electrode approach. Evoked dopamine signals were quantified as indicators of terminal neurotransmitter function. Only recordings matching the characteristic voltammogram for dopamine were included in the analysis. 43 RNA sequence data analysis Total RNA was extracted from brain of 8-week-old mice using the TRIzol reagent (Invitrogen, USA). RNA integrity and concentration were measured by a Qubit™ RNA High Sensitivity (HS) (Thermo Fisher Scientific, USA) and Agilent 4150 Tapestation (Agilent Technologies, USA). High-quality RNA samples (RIN > 7) were used for library construction by Illumina Stranded mRNA Prep (Illumina, USA). The libraries were sequenced on Illumina NovaSeq 6000 to generate 150-bp paired-end reads. Sequencing reads were analyzed by CLC (Qiagen, USA). After trimming reads to remove adapters and low-quality bases, the clean reads were aligned to the mouse reference genome (GRCm39). Gene expression levels and differential expression analysis was analyzed by DESeq2 R package. Genes with a fold change > 2 and a false discovery rate (FDR) adjusted p-value <0.05 were identified as differentially expressed genes (DEGs). To analyze the biological functions pathway, Gene Ontology (GO) enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and Gene Set Enrichment Analysis (GSEA) were conducted by the clusterProfiler and msigdbr package in R. Principal Component Analysis (PCA) was performed to visualize the segregation of experimental groups based on their global gene expression profiles. Western blot analysis In the early intervention group, MP mice received PT320 beginning at 5 weeks of age. Subsets of animals were sacrificed at 5 and 8 weeks of age for early-stage molecular analyses, while the remaining animals continued treatment until 20 weeks of age. In the late intervention group, MP mice were started on PT320 at 15 weeks of age and treated continuously until 20 weeks of age, at which point the animals were sacrificed. Following sacrifice, mice were decapitated, and the dorsal striatum, hippocampus, and NAc were rapidly dissected according to a standard mouse brain atlas for subsequent biochemical analyses. Tissues underwent homogenization in RIPA buffer (TAAR-ZBZ5, Biotools Co., Ltd, Taiwan) enriched with a protease inhibitor cocktail (ab20111, Abcam, Cambridgeshire, United Kingdom). Protein concentration was determined using the BCA Protein Assay Kit II (K813, Biovision, Lausen, Switzerland). These proteins were then electrophoresed on a 12% SDS-polyacrylamide gel and electrotransferred onto a polyvinyl difluoride membrane. Following a 45-min incubation in a blocking buffer (2% bovine serum albumin and Tris-buffered saline with Tween 20), the membranes were subjected to overnight incubation with primary antibodies: Phospho-AKT(Ser473) (1:1000, Mouse, 66444-1-Ig, Proteintech), AKT (1:1000, rabbit, 10176-2-AP, Proteintech), BDNF (1:1000, rabbit, 25699-1-AP, Proteintech), Phospho-CREB (Ser133) (1:1000, rabbit, MA5-11192, Thermo Fisher Scientific), CREB (1:1000, Mouse, MA1-083, Thermo Fisher Scientific), Anti-Dopamine Receptor D3/DRD3 (1:1000, rabbit, ab155098, abcam), and β-actin polyclonal antibody (1:5000, rabbit, GTX109639), GeneTex). Following three washing steps, the membranes were incubated for 1 h with an HRP-conjugated goat anti-rabbit IgG antibody (1:20,000; ab6721, Abcam). Signal detection was performed using an enhanced chemiluminescence (ECL) kit (Amersham Life Sciences, Piscataway, NJ, USA), and images were captured with a UVP ChemStudio Plus system (Analytik Jena AG, Germany). Protein signals were normalized to GAPDH to control for loading variability, and band intensities were quantified relative to their respective controls using NIH ImageJ software. Tissue collection and immunostaining of section For collection of the whole brain for immunostaining mice were anesthetized with a high dose of sodium pentobarbital (90 mg/kg intraperitoneally [i.p.], Orion Pharma, Finland) and perfused transcardially with PBS and then 4% PFA. For paraffin sections, brains were put into 4% PFA for a few days (with changes to fresh 4% PFA regularly), then embedded in paraffin. Sections were cut on a microtome into 5 mm thickness and stored at 4°C for further processing. For staining of paraffin sections, after deparaffinization, sections underwent antigen retrieval by heating them in citraconic anhydride (0.05% citraconic anhydride in Milli-Q water [pH 7.4]) without boiling. Slides were then cooled for 30 min at RT. They were washed with TBS, and then sections were encircled with a PAP pen, washed with TBS-T, and blocked with 1.5% normal horse serum (Vector) in TBS-T for approximately 20 min. Sections were labeled with primary antibodies including TH (1:500, rabbit, ab137869, Abcam) and p-CREB at 4°C overnight, followed by incubation with fluorescent dye-conjugated secondary antibodies for 2 h at 37°C. Finally, the slices were mounted with a FluoroshieldTM with nuclear dye 4′,6-diamidino-2-phenylindole (DAPI) (GTX30920; Genetex) for fluorescence detection. Sections were evaluated by blinded observers and controls consisted of omission of the primary antibodies. Statistical analysis Behavioral data were analyzed with one-way or two-way ANOVA where appropriate, with genotype/treatment and age as factors. Significant ANOVAs were followed by Bonferroni post hoc tests for multiple comparisons. Key planned comparisons included MP vs. WT, MP+PT320 vs. MP, and drug-treated vs. untreated within genotypes. For quality control, data points exceeding two-fold the standard deviation (±2 SD) from the group mean were considered outliers and excluded from further analysis. 44 Significance was reported at p < 0.05, p < 0.01, *p < 0.001 for comparisons to WT, and # for comparisons between treated and untreated MP groups (## for p < 0.01, ### for p < 0.001). Data are presented as mean ± SEM. Sample sizes for all analyses are detailed in Table 1 and Supplementary Table 1. Results 5-HTP or citalopram ameliorated depression- and anxiety-like behaviors in MP mice A previous study identified anxiety-like and depression-like changes present starting from 14 weeks in MP mice, 22 concurrent with the onset of motor deficits in this model. However, in our laboratory, the forced swim test (FST) revealed depressive-like behavior as indicated by increased immobility time in MP mice at 8 weeks when compared to age-matched littermate control mice (Figure 1B). During the tail suspension test (TST), MP mice exhibited a significant increase in immobility beginning at 10 weeks of age, while immobility in control mice remained relatively constant over time (Figure 1C). We also performed a marble burying test to evaluate anxiety/compulsive-like behavior in MP mice. The number of buried marbles increased significantly in MP mice beginning at 14 weeks of age (Figure 1D). Likewise, from 14 weeks onward, MP mice spent significantly less time in the center zone of the open field, as compared with age-matched controls, suggesting a progressive increase in anxiety-like behavior (Figure 1E). Taken together, our findings indicate that depression-like behavior emerges as early as 8 weeks of age in MP mice, preceding the onset of motor deficits or anxiety-like behaviors in this model. To further support that these findings were due to affective changes and not motor dysfunction, a subset of 20-week-old MP mice received two weeks of treatment with 5-HTP or citalopram, an antidepressant that elevates brain serotonin levels. Behavioral tests, including the TST, FST, marble-burying, and open-field assays, were conducted at 22 weeks (Figure 1A). Behavioral analyses revealed that citalopram treatment exerted both anxiolytic- and antidepressant-like effects in MP mice (Figure 1F–I). Specifically, citalopram increased the time spent in the center zone and reduced marble-burying behavior, indicating reduced anxiety-like behavior (Figure 1F–G), while also decreasing immobility time in the FST and TST, consistent with an antidepressant-like effect (Figure 1H–I). Similarly, 5-HTP treatment alleviated both anxiety- and depression-like behaviors in 22-week-old MP mice, as evidenced by increased center-zone exploration, fewer buried marbles, and reduced immobility in the FST (Figure 1J–L). Importantly, locomotor activity remained unchanged following either treatment (Figure 1M–N), indicating that reduced immobility in these behavioral assays was not secondary to alterations in motor function. Although selective serotonin reuptake inhibitors (SSRIs) may provide therapeutic benefit for neuropsychiatric symptoms in PD, their clinical utility can be limited by adverse effects, including the risk of serotonin syndrome or potential exacerbation of akinetic symptoms (Supplementary Figure 1). PT320 suppressed the development of depression- and anxiety-like behaviors in MP mice Considering the limitations of serotonergic therapies, recent research has focused on alternative pharmacological strategies capable of simultaneously improving PD motor deficits and associated behavioral disturbances. Based on a previously established protocol, 29 PT320 was administered starting at 5 weeks of age (early treatment with PT320) and the time line of biweekly PT320 treatments is shown in Figure 2A. Our prior pharmacokinetic studies in mice demonstrated that the selected PT320 dose of 30 mg/kg (0.6 mg/kg Exenatide) produces clinically comparable GLP-1 exposure levels, 37 confirming its translational relevance. Before assessing whether early PT320 administration could ameliorate neuropsychiatric disturbances in MP mice, we evaluated its effects in WT mice and found that PT320 did not influence locomotor activity or anxiety- and depression-like behaviors from 8 to 20 weeks of age (Supplementary Figure 2). In MP mice, horizontal and vertical locomotor activities progressively declined beginning at 10 and 14 weeks of age, respectively (Supplementary Figure 3). Early PT320 treatment improved horizontal activity at 14 weeks but had no effect on vertical activity, suggesting a delay in spontaneous motor decline in MP mice. Assessment of depression-like behavior showed that PT320-treated MP mice exhibited significantly shorter immobility durations in the FST from 8 to 20 weeks (Figure 2B) and in the TST (Figure 2C) from 10 to 20 weeks, demonstrating a sustained antidepressant-like effect compared with untreated MP mice. In the marble-burying test, MP mice buried significantly more marbles than WT mice beginning at 14 weeks; in contrast, PT320 treatment restored marble-burying behavior to near WT levels, with this effect persisting through 20 weeks (Figure 2D). Similarly, MP mice exhibited strong avoidance of the center zone in the open-field test compared with WT mice beginning at 14 weeks, whereas PT320-treated MP mice spent significantly more time in the center zone, and this anxiolytic effect persisted through 20 weeks (Figure 2E). Collectively, these resuls indicated that early treatment of PT320 not only postponed the progression of spontaneous motor impairments, but also conferred protection against the emergence of neuropsychiatric disturbances in MP mice, highlighting its dual therapeutic potential in PD. Early PT320 treatment ameliorated anxiety- and depression-like phenotypes in MP mice. (A) Experimental protocol for early PT320 treatment and longitudinal behavioral assessments. (B) In the FST, untreated MP mice exhibited a significant increase in immobility time beginning at 8 weeks of age, whereas the PT320 early-treatment group showed this increase at 12 weeks, indicating a delayed onset of depressive-like behavior. PT320 treatment also significantly reduced immobility duration compared with age-matched untreated MP mice starting at 8 weeks. (C) In the TST, early PT320 treatment delayed the onset of increased immobility from 10 weeks in MP mice to 12 weeks in the treatment group, and significantly reduced immobility time relative to untreated MP mice. (D) In the MBT, PT320 treatment significantly decreased the number of buried marbles at 14 and 20 weeks, indicating attenuation of anxiety-like behaviors in MP mice. (E) In the open-field test, MP mice spent significantly less time exploring the center zone at 14 and 20 weeks compared with WT controls, whereas PT320 treatment improved this deficit, suggesting a rescue of anxiety-related exploratory behavior. Data are presented as mean ± SEM with individual data points shown. Statistical significance was determined using one-way or two-way ANOVA followed by Bonferroni post hoc tests. * p < 0.05, ** p < 0.01, *** p < 0.001 compare with WT; # p < 0.05, ## p < 0.01, ### p < 0.001 MP vs. MP +PT320(5W). Early PT320 treatment restored the downregulated BDNF signaling pathway in NAc of MP mice To further elucidate the mechanisms underlying the antidepressant- and anxiolytic-like effects of PT320, brain tissues were collected from 8-week-old MP mice. Given that alterations in CREB phosphorylation and BDNF expression are linked to depressive phenotypes in MP mice, 45 we performed Western blot analyses to quantify BDNF protein levels in the prefrontal cortex, hippocampus, dorsal striatum (Figure 3A), and nucleus accumbens (NAc) (Figure 3E) of WT and MP mice, with or without PT320 treatment. In both the prefrontal cortex (Supplementary Figure 4A) and hippocampus (Supplementary Figure 4B), BDNF levels did not differ between MP and WT mice at 8 weeks of age. A significant reduction in BDNF protein level was observed in the dorsal striatum and NAc of MP mice (Figure 3D and H)). Notably, PT320 treatment reversed the downregulation of BDNF and concomitantly enhanced CREB and AKT phosphorylation in dorsal striatum (Figure 3A–D) and NAc (Figure 3E–H), which are closely associated with dopaminergic innervation underlying motor and affective behaviors, respectively. Taken together, this set of experiments indicates that the affective improvements induced by PT320 are mediated by a drug-dependent potentiation of the AKT–CREB–BDNF axis in the NAc of MP mice. Early PT320 treatment counteracted the reduction in p-Akt/p-CREB/BDNF protein expression levels in the dorsal striatum and nucleus accumbens (NAc) of MP mice. (A) Representative immunoblot images showing Akt, CREB, and BDNF protein expressions in the dorsal striatum at 5 and 8 weeks. (B) MP mice showed significantly lower p-Akt/Akt ratios at both 5 and 8 weeks relative to WT controls, which were restored by early PT320 treatment. (C) MP mice showed a significant decrease in p-CREB/CREB ratios at 8 weeks relative to WT controls; this reduction was observed in the PT320-treated group. (D) PT320 treatment partially restored the reduction in BDNF protein levels observed in MP mice at 8 weeks. (E) Representative immunoblot images showing Akt, CREB, and BDNF protein expression in the nucleus accumbens at 5 and 8 weeks. (F) PT320 treatment did not rescue the significantly reduced p-Akt/Akt ratio observed in MP mice at 8 weeks. (G) PT320 treatment counteracted the reduction in p-CREB/CREB ratios observed in MP mice at 8 weeks of age. (H) PT320 administration reinstated BDNF protein expression levels in MP mice at 8 weeks of age. Data are presented as mean ± SEM with individual data points shown. Statistical significance was determined using one-way ANOVA followed by Bonferroni post hoc tests. **p < 0.01, ***p < 0.001 vs. WT; #p < 0.05, ###p < 0.001 for MP vs. MP + PT320 (5W). Early administration of PT320 normalized the altered transcriptional profiles of BDNF–Trkβ, dopamine receptor 3 and mitochondrial fusion/fission–related genes within the NAc of MP mice To identify the gene regulation underlying the therapeutic effects of PT320, RNA sequencing was conducted on NAc tissues from 8-week-old WT, MP, and PT320-treated MP mice (Figure 4). Principal Component Analysis (PCA) revealed distinct clustering of three groups (Figure 4A). The gene expression profile of PT320-treated MP mice was similar with WT, and it suggested that a global restoration of gene expressions after PT320 treatment. This restorative effect was corroborated by heatmap, which displayed a reversal pattern between MP and PT320 (Figure 4B). Early PT320 treatment ameliorated transcriptional dysregulation involving BDNF–Trkβ, DRD3, and mitochondrial fusion/fission pathways in the NAc of MP mice. (A) Principal component analysis (PCA) revealed distinct clustering among WT, MP, and PT320-treated MP mice, indicating marked transcriptional divergence between groups. (B) Heatmap showing the top 50 differentially expressed genes (DEGs) distinguishing MP mice from PT320-treated MP mice across the three groups. (C) Gene Ontology (GO) enrichment analysis of upregulated genes identified in MP and PT320-treated MP mice. (D) GO enrichment analysis of downregulated genes. (E) Heatmap illustrating expression patterns of neurotrophic signaling, dopaminergic system, and mitochondrial dynamics–related genes in the NAc across the three groups. (F) Ingenuity Pathway Analysis (IPA) was employed to evaluate the associations between DEGs related to neurotrophic signaling and mitochondrial dynamics and their corresponding disease pathways. Gene Ontology (GO) enrichment analysis underscored the relevance of biological processes modulated by PT320. Genes which were upregulated in MP mice and suppressed by PT320 were associated with cellular stress responses (Figure 4C), implying that PT320 rescues mitochondrial stress and aberrant protein synthesis. In contrast, downregulated genes that were restored by PT320 were significantly enriched in neuronal plasticity and connectivity related pathways (Figure 4D). To further support the neuroprotective effects of PT320, we performed Gene Set Enrichment Analysis (GSEA) using the Hallmark gene set collection (Supplementary Figure 5). We compared the pathological state by comparing MP mice with WT controls (Supplementary Figure 5A). This analysis revealed that dysregulation in MP mice was associated with metabolic compensation and oxidative stress, including Oxidative Phosphorylation, Reactive Oxygen Species (ROS) Pathway, Glycolysis, and Fatty Acid Metabolism. This suggests that the loss of Tfam in MP mice drives dopaminergic neurons into a state of metabolic hyperactivity and oxidative burden. In addition, pathways related to genotoxic stress and apoptosis, such as DNA Repair, P53 Pathway, and Unfolded Protein Response (UPR), were significantly enriched, reflecting severe cellular damage and the activation of cell death programs. Treatment with PT320 resulted in a reversal of these cellular disruptive pathways (Supplementary Figure 5B). The pathways that were aberrantly upregulated in MP mice, specifically those governing metabolic stress (Oxidative Phosphorylation, Glycolysis) and cellular damage (ROS Pathway, DNA Repair, P53 Pathway) were significantly downregulated following PT320 administration (NES < 0). Furthermore, PT320 treatment suppressed mTORC1 Signaling, which is often hyperactivated in neurodegeneration and inhibits autophagy, suggesting a restoration of autophagic potential. We next focused on specific signaling cascades involved in mood regulation and mitochondrial dynamics (Figure 4E and Supplementary Figure 6). Transcriptomic data confirmed that the Akt3–Creb1–Bdnf–Ntrk2 (TrkB) axis was significantly downregulated in the MP mice compared to WT controls (Supplementary Figure 6C–F; Supplementary Table 2). PT320 treatment restored the expression of these neurotrophic genes. Furthermore, MP mice exhibited a reduction in dopamine D3 receptor (DRD3) expression, a critical receptor for motivation and affect. Especially, PT320 treatment significantly normalized DRD3 expression levels (Supplementary Figure 6G; Supplementary Table 3). With respect to mitochondrial homeostasis, MP mice exhibited an imbalance favoring fission by the downregulation of fusion genes (Opa1, Mfn1/2) and upregulation of fission genes (Fis1, Dnm11). However, PT320 treatment partially reversed this condition, specifically by upregulating Opa1 and suppressing Fis1 (Figure 4E and Supplementary Figure 6H–N; Supplementary Table 4). Ingenuity Pathway Analysis revealed that differentially expressed genes across the three groups, particularly those involved in BDNF neurotrophic signaling and mitochondrial organization, were significantly linked to mitochondrial dysfunction, depressive disorders, and Parkinson's disease (Figure 4F). Collectively, these transcriptomic findings support that the antidepressant and anxiolytic benefits of early PT320 intervention are mediated through restoration of BDNF–TrkB neurotrophic signaling, preservation of Drd3-mediated dopaminergic transmission, and stabilization of mitochondrial homeostasis. Late administration of PT320 mitigated the depressive-like behavioral phenotypes observed in MP mice Next, we examined the therapeutic potential of PT320 when administered after the onset of depressive and anxiety-like behavioral symptoms. PT320 administration was initiated at 15 weeks of age in MP mice, representing the “late-stage” treatment paradigm (Figure 5A). Locomotor activity, measured as the total horizontal distance traveled and rearling frequency during a 10-min session, was significantly reduced in 20-week-old MP mice relative to WT controls. Even after three biweekly administrations of PT320, no significant improvement in horizental activity was observed (Figure 5B). Likewise, vertical rearing activity remained markedly reduced in MP mice, with no appreciable difference between PT320-treated and vehicle groups (Figure 5C). In the open-field test, PT320-treated MP mice exhibited a tendency to spend more time in the central zone compared with untreated MP mice (Figure 5D). In the marble-burying test, late administration of PT320 normalized the number of buried marbles in MP mice to levels comparable to WT controls, indicating that PT320 tratment mitigated the progression of certain anxiety-like behaviors in MP mice (Figure 5E). Late treatment of PT320 alleviated depression-like behaviors of 20-week-old MP mice. (A) experimental protocol for late PT320 treatment (15–20 weeks). (B) Late PT320 treatment did not improve the reduced horizontal locomotor activity observed in MP mice. (C) The decreased vertical activity in 20-week-old MP mice was not rescued by late PT320 treatment. (D) Late PT320 treatment did not ameliorate the reduced center exploration time in the open-field test. (E) Anxiety-like behavior assessed by the MBT was significantly improved by late PT320 treatment in MP mice. (F) Late PT320 treatment reduced immobility time in the FST, indicating an improvement in depressive-like behavior. (G) Late PT320 treatment improved depressive-like behavior, as evidenced by decreased immobility duration in the TST. Data are presented as mean ± SEM with individual data points shown. One-way analysis of variance (ANOVA) followed by a Bonferroni post hoc test for multiple comparisons; *** p < 0.001 compare with WT; ### p < 0.001 MP vs. MP +PT320(15W). In the FST, PT320-treated MP mice exhibited an approximately 40% reduction in immobility duration compared with untreated MP mice (Figure 5F), consistent with the results of the TST (Figure 5G), indicating that PT320 exerts a strong antidepressant effect even when administered after the onset of depressive-related symptoms. Late PT320 treatment partially improved dopamine transmission dynamics in the NAc of MP mice To determine whether the antidepressant effects observed following late PT320 administration were mediated by changes in dopaminergic transmission dynamics, evoked dopamine release in the NAc was quantified at 20 weeks of age using FSCV. In agreement with our earlier observations in the dorsal striatum, tonic or phasic electrical stimulation of NAc slices from WT mice elicited marked dopamine release. In contrast, 20-week-old MP mice displayed severely attenuated dopamine signals, reflecting pronounced dopaminergic terminal degeneration or dysfunction. With tonic stimulation, dopamine release in the NAc of PT320-treated MP mice remained comparable to that of untreated MP mice. In contrast phasic dopamine release was significantly elevated in the PT320-treated group, indicating partial restoration of dynamic dopaminergic function by PT320 treatment (Figure 6). Collectively, these findings indicate that late-stage administration of PT320 augments the functional dopaminergic output of residual NAc projections. Also, this finding aligns with the behavioral data, suggesting that improved dopaminergic input within the NAc likely underlies the reductions in depression-like behaviors observed following PT320 treatment. Late PT320 treatment improved dopamine release in the nucleus accumbens of 20 W MP mice. (A) Representative voltammetric traces and cyclic voltammograms recorded at a stimulation intensity of 7V. (B) Dopamine release in the nucleus accumbens was significantly higher in the MP + PT320 group compared with the MP group during phasic stimulation (10 pulses, 25 Hz), whereas no significant differences were observed under tonic stimulation (1 pulse, 25 Hz). Data are presented as mean ± SEM. Statistical significance was determined using one-way analysis of variance (ANOVA) followed by Bonferroni post hoc tests for multiple comparisons. ***p < 0.001 vs. WT; #p < 0.05 for MP vs. MP + PT320 (15W). Late PT320 administration enhanced CREB–BDNF signaling cascades and mitigated the reduction of tyrosine hydroxylase (TH) expression in the NAc of MP mice Previous studies demonstrated that early intervention with PT320 confers neuroprotection by mitigating TH loss in the striatum and substantia nigra. 46 Here, we sought to investigate whether PT320 treatment, initiated after the onset of loss of dopaminergic input or behavioral symptoms, could still slow or prevent further loss of dopaminergic input. As illustrated in Figure 7A and B, the NAc of 20-week-old MP mice showed a pronounced reduction in TH expression, which was restored by late PT320 treatment, indicating that GLP-1 receptor activation could preserve dopaminergic input at advanced disease stages. In accordance with the reduction of TH expression, MP mice showed a marked decline in phosphorylated CREB–positive cells within the NAc compared with WT controls. Late PT320 administration restored CREB phosphorylation to near-normal levels (Figure 7C–D), indicating reactivation of neurotrophic signaling. In the quantitative immunoblotting analysis (Figure 7E–I), no significant differences in p-AKT/AKT levels were observed among the three groups (Figure 7G). However, the levels of TH (Figure 7F), p-CREB/CREB (Figure 7H), and BDNF (Figure 7I) were significantly decreased in the NAc of MP mice compared with WT controls, and these reductions were restored by late PT320 treatment. Overall, these findings suggest that PT320, a GLP-1 receptor agonist, upregulates the decreased CREB–BDNF signaling cascade and mitigates loss of dopaminergic input in the NAc during the late stage of PD, thereby attenuating NAc-associated neuropsychiatric symptoms as the disease progresses in this MP mouse model. Late PT320 treatment rescued TH expression and counteracted the reduction in p-CREB in the NAc of 20-week-old MP mice. (A) Representative immunofluorescence images showing TH expression in the NAc. (B) TH fiber intensity was significantly reduced in MP mice compared with WT mice, and late PT320 treatment ameliorated this reduction. (C) Representative immunofluorescence images showing CREB expression in the NAc. (D) Late PT320 treatment improved the reduced CREB expression observed in MP mice. (E) Representative immunoblot images showing Akt, CREB, BDNF, and TH protein expression in NAc brain tissue. (F) Late PT320 treatment alleviated the decrease in TH protein levels in MP mice. (G) The p-Akt/Akt ratio did not differ significantly among the three groups. (H) MP mice exhibited significantly reduced p-CREB/CREB ratios compared with WT controls, and PT320 treatment mitigated this reduction. (I) Late PT320 treatment improved the decreased BDNF protein expression observed in MP mice relative to WT controls. Data are presented as mean ± SEM with individual data points shown. One-way analysis of variance (ANOVA) followed by a Bonferroni post hoc test for multiple comparisons; * p < 0.05, ** p < 0.01, *** p < 0.001 compare with WT; # p < 0.05, ### p < 0.001 MP vs. MP +PT320(15W). Discussion In this study, we investigated the relationship between dopaminergic degeneration, neurotrophic signaling, and affective behavioral alterations in MP mice. We found that dopaminergic deficits were associated with impaired CREB–BDNF signaling in the NAc, concomitant with the emergence of depression- and anxiety-like behaviors. Using both early and late PT320 intervention paradigms, our results further demonstrate that modulation of this pathway is sensitive to treatment timing and disease stage, highlighting the importance of therapeutic window in targeting neuropsychiatric symptoms associated with PD (Supplementary Table 5). Depression and anxiety are among the earliest non-motor symptoms in patients with PD, and clinical studies have shown that the mesolimbic system, particularly in the NAc, plays a key role in the affective disturbances associated with PD.47–50 Although extensive evidence suggests that suppression of dopaminergic transmission in the mesolimbic system contributes to mood disturbances, these conclusions have largely been derived from models using neurotoxins or environmental stress paradigms.51,52 Such approaches establish correlation but not causality, leaving the questions whether dopaminergic circuits alone are responsible, and which downstream signaling pathways mediate the neurobiological basis of depression and anxiety. To clarify these mechanisms, we employed MP mice, a genetic model of PD in which mitochondrial function is disrupted by selective deletion of the nuclear genome–encoded mitochondrial transcription factor A (Tfam) gene specifically in dopaminergic neurons rather than in other neuronal systems. This model faithfully recapitulates several features of human PD, including progressive dopaminergic degeneration, gradual onset of motor deficits, and L-DOPA–responsive symptoms.53–56 Monica et al. provided additional characterization of non-motor behavioral changes in MP mice but found that most non-motor deficits overlapped temporally with motor dysfunction, owing to the rapid onset of motor symptoms once the mice reached 12 weeks of age. 22 Considering that gender can significantly affect physical parameters such as body weight and locomotor activity, 24 this study focused on chronologically characterizing depression- and anxiety-like phenotypes in male MP mice throughout disease progression. Notably, our prior studies showed gender differences in disease progression of MP mice, 24 with males demonstrating a more rapid decline. With this knowledge, males were evaluated in the current study thereby providing a more severely progressive PD model to allow dissection of non-motor behavioral actions from motor ones. According to our previous results, male MP mice exhibited stable motor performance in the RotaRod test up to 16 weeks of age, after which a significant decline in motor ability was observed.56,57 To capture the temporal evolution of affective symptoms independent of motor deficits, we performed biweekly behavioral assessments for depressive- and anxiety-like phenotypes from 4 to 20 weeks of age. In contrast to previous studies, 22 male MP mice in our experiments exhibited prolonged immobility in the FST and TST starting at 8 and 10 weeks of age, respectively, earlier than motor deficits detected in the rotarod and locomotor tests. This indicates that motor dysfunction is not the cause of immobility observed in these assays. This observed time course of depressive phenotypes aligns with clinical observation that affective disturbances precede the onset of motor impairment in patients with PD.52,58 In addition, anxiety-like phenotypes emerged at 14 weeks in male MP mice, as evidenced by increased marble burying and reduced center-zone exploration, following the occurrence of depressive-like symptoms. Although several clinical studies suggest that anxiety emerges before depression in the prodromal stage of PD,50,59 the discrepancy with our findings may stem from compensatory responses that help maintain performance during early disease stages. Increased catecholamine release or recruitment of auxiliary neural circuits during specific behavioral tasks may transiently offset anxiety-like phenotypes.60–62 Furthermore, our data showed that Citalopram, a SERT inhibitor, and 5-HTP, a serotonin precursor, not only attenuated behavioral depression but also partially improved anxiety-like symptoms. In line with our results, GLP-1 receptor agonists have been reported to modulate dorsal raphe serotonergic neuron activity 63 and enhance central serotonin signaling, effects that have been linked to antidepressant- and anxiolytic-like behaviors in preclinical models, suggesting that serotonergic circuits may also contribute to the observed affective improvements. 64 These findings are consistent with clinical observations that serotonergic agents are commonly prescribed to manage depression or anxiety in patients with PD. 65 Accordingly, our findings support the view that male MP mice constitute a valuable model for reproducing the neuropsychiatric disturbances, such as early depressive and anxiety-like phenotypes, that characterize the prodromal phase of PD. Given that BDNF levels within the mesolimbic system are closely associated with motivation, reward processing, and affective regulation,66–68 we sought to examine BDNF alterations in this circuitry and evaluate their relationship to the neuropsychiatric phenotypes observed in MP mice. Interestingly, we found that BDNF levels are preferentially downregulated in the NAc and dorsal striatum, but not in the hippocampus or prefrontal cortex, of 8-week-old MP mice coinciding with the initial appearance of depressive-like behavior. This region-specific downregulation aligns with prior studies demonstrating that diminished BDNF expression or deletion of TrkB within dopaminergic circuits contributes to the development of depressive or anxious phenotypes.69,70 While early PT320 administration did not modify TH expression or dopamine release in the NAc of 8-week-old MP mice (Supplementary Figure 7), the long-lasting GLP-1 analogue, potentiating BDNF signaling, could restore the diminished p-AKT, p-CREB, and BDNF levels in this region, thereby slowing the onset of neuropsychiatric disturbances in this transgenic mouse model of PD. These results imply that PT320 does not initially act by preserving dopamine synthesis in presynaptic terminals; instead, it may exert early therapeutic effects by promoting CREB-mediated BDNF upregulation, thereby enhancing synaptic plasticity and reinforcing neurotrophic support within the vulnerable microenvironment. Our described actions of PT320 on BDNF are in line with prior studies illustrating GLP-1 receptor agonist mitigation of decreased BDNF levels in the substantia nigra and striatum of mice challenged with MPTP.71,72 GLP-1 analogue-mediated upregulation of brain BDNF levels have likewise been reported in other preclinical animal models too, including type 2 diabetes and Alzheimer's disease.73,74 Several preclinical investigations reported that GLP-1 analogues decrease immobility time in the FST and improve behavioral outcomes across additional paradigms in models of depression and obesity. However, it remains possible that some of the observed behavioral effects may not be attributable solely to PT320 itself, but rather to secondary factors such as reduced caloric intake. Caloric restriction has been shown to influence both latency to immobility and total immobility time in mice and is also associated with alterations in brain BDNF levels. Therefore, the potential contribution of caloric restriction should be carefully considered when interpreting the behavioral effects attributed to PT320.75–78 In human clinical studies, however, some studies report potential antidepressant benefits, 79 others, suggest a possible elevated risk of depression and anxiety, 80 potentially implying context-driven effects. Notably, in a phase 2 clinical trial of Exenatide in early to moderate PD, a post-hoc analysis of data reported that weekly Exenatide provided greater improvements in individual domains that assessed mood/depression across all observer-rated outcome measures when evaluated after 48 weeks treatment, including the mood/apathy domains within the non-motor symptom scale. 81 However, a recent phase 3 randomized controlled trial of exenatide in PD did not demonstrate disease-modifying effects on motor progression. 82 This negative finding does not exclude potential effects on non-motor or affective symptom domains, which were not primary outcomes of that study. In this context, the lack of disease-modifying efficacy in the phase 3 trial underscores the importance of improved target engagement or the identification of responsive subgroups when evaluating GLP-1 receptor agonists in PD. Our RNA-sequencing results demonstrated a selective reduction in the transcriptional abundance of the dopamine D3 receptor (D3R), but not D1R or D2R, in the NAc of 8-week-old MP mice. In contrast, PT320 treatment reversed this reduction, indicating that modulation of D3 signaling may contribute to alleviating subsequent depressive-like behavior. These observations align with earlier reports showing that the D3R, which is highly enriched in mesolimbic dopaminergic circuits, plays a critical role in affective regulation. In adult mice, selective knockdown of D3R in the NAc is sufficient to produce depressive-like behaviors, whereas reinstating D3R expression in the same region markedly alleviates depression-like symptoms. 83 Recently, accumulated evidence has indicated that D3Rs are expressed on both pre- and postsynaptic elements of dopaminergic circuits, where they modulate dopaminergic transmission by regulating dopamine synthesis and release. 84 In addition, some studies have reported that D3Rs are expressed in microglial populations within the NAc. This microglial D3R expression has been implicated in neuroinflammatory signaling cascades that contribute to the onset of depression-like behaviors. 83 In this regard, the potential crosstalk between D3R expressed neurons and microglia may alter the functional state of each cell type within dopaminergic circuitry, providing a possible mechanistic basis for the regulation of affective behaviors. 83 A few investigations have demonstrated that activation of the BDNF/TrkB pathway via AKT signaling is crucial for regulating D3R function and thus represents a promising therapeutic mechanism for anxiety and depression in PD rat models.67,83,85 This is consistent with our findings that PT320 not only reactivates BDNF signaling but also increases D3R transcript abundance in the NAc, thereby suppressing the development of depression- and anxiety-like behaviors in MP mice. Compared with other neuroprotective strategies, PT320 demonstrated a notable therapeutic advantage: even when delivered after the emergence of neuropsychiatric deficits and advanced dopaminergic terminal loss; PT320 robustly restored CREB–BDNF pathway activation in the NAc. These results underscore that late PT320 intervention still confers substantial BDNF-driven neurotrophic support capable of reversing dopaminergic dysfunction and producing antidepressant and anxiolytic effects. In parallel, mitochondrial regulation represents another plausible contributor to affect-related changes in the NAc. Accumbal MFN2 has been causally linked to anxiety- and depression-like behaviors through mitochondrial and neuronal mechanisms, supporting the plausibility that altered mitochondrial dynamics may participate in affective phenotypes. 86 In the current study, we therefore interpret D3R- and MFN2-related signals as hypothesis-generating molecular correlates that may contribute to PT320-associated behavioral improvement. Importantly, formal testing of these candidates will require region-specific gain- and loss-of-function approaches in the NAc, together with direct comparison to dorsal striatal mechanisms, which we now highlight as a key future direction. Accumulating evidence indicates that the NAc plays a central role in the regulation of affective and motivational behaviors, and that intracellular signaling pathways involving AKT, CREB, and BDNF within this region are closely associated with depression- and anxiety-like phenotypes. Previous studies have demonstrated that alterations in CREB activity and BDNF signaling in the NAc accompany stress susceptibility and mood-related behavioral outcomes, although causal relationships generally require region- or pathway-specific manipulations to be established.87,88 In line with this framework, dysregulation of AKT–CREB–BDNF–related signaling within the mesolimbic system has been reported in preclinical models of affective disorders, and normalization of these pathways often coincides with behavioral improvement following pharmacological or circuit-level interventions.68,89 Importantly, such molecular changes are typically interpreted as correlates of behavioral state rather than definitive mediators unless supported by targeted manipulations. In the context of PD, emerging evidence suggests that mesolimbic dysfunction, including alterations in NAc neurotrophic signaling, may contribute to non-motor affective symptoms such as depression and anxiety.90,91 Although most PD studies have focused on nigrostriatal degeneration, changes in BDNF/TrkB- and CREB-related signaling within limbic striatal regions have been increasingly implicated in affective symptomatology. 67 Consistent with this literature, our findings demonstrate that PT320 treatment is associated with normalization of AKT–CREB–BDNF–related measures in the NAc, coinciding with improvements in depression- and anxiety-like behaviors. However, given that parallel molecular changes were also observed in the dorsal striatum, the present data does not allow definitive attribution of behavioral effects to NAc-specific mechanisms. Future studies employing region- or pathway-specific approaches will be required to determine the relative contributions of NAc versus dorsal striatal signaling, and to clarify whether modulation of the VTA–NAc pathway is necessary for the observed affective improvements. Given that PT320 treatment was found to modulate the expression of optic atrophy 1 (OPA1), mitofusin-2 (MFN2), and mitochondrial fission 1 (Fis1)—key regulators of mitochondrial fusion and fission dynamics that collectively preserve mitochondrial architecture, turnover, and functional integrity, 29 we propose that the regional selectivity of PT320 may relate to the unique metabolic profile of mesolimbic dopaminergic neurons projecting to the NAc. These neurons may exhibit distinct metabolic demands and mitochondrial adaptive capacity, rendering them particularly responsive to interventions that stabilize mitochondrial homeostasis. In addition, the NAc is highly enriched in trophic signaling pathways, including the BDNF–TrkB–CREB cascade, 92 which was significantly modulated by PT320 in our study. The convergence of mitochondrial regulation and region-specific trophic signaling may therefore contribute to the preferential effects observed in this mesolimbic circuit. 86 While our findings strengthen the understanding of how a long-acting GLP-1 receptor agonist provides enduring neurotrophic protection against dopaminergic degeneration in PD-associated depression and anxiety, several limitations warrant consideration. First, while MitoPark model offers strong translational value, particularly in recapitulating progressive dopaminergic degeneration, motor decline, and emerging non-motor symptoms, it does not prominently reproduce α-synuclein aggregation pathology, nor does it consistently exhibit the robust neuroinflammatory signatures often described in idiopathic PD. Accordingly, future studies will incorporate complementary synucleinopathy-based models and inflammatory paradigms to further strengthen the translational framework and better define the therapeutic potential of PT320 in PD patients. To minimize the effect caused by estrogen-associated issues and simplify the experimental variables, the evaluation of exclusively male mice was undertaken in this study. This would limit the generalizability of our findings to the broader clinical population. Hence, to achieve a more comprehensive understanding of PD-associated mood disturbances and to enhance the translational value of therapeutic approaches, it will be important for subsequent studies to include both male and female subjects. Moreover, this study specifically focused on the role of NAc dopamine input in depression or anxiety pathophysiology and did not investigate other neural circuits in affective neurobiology. Lastly, our study did not include structural analyses to determine whether PT320-mediated BDNF upregulation influences dendritic morphology in NAc medium spiny neurons. These neurons receive dense mesolimbic dopaminergic input and exhibit structural remodeling in mood disorders. In future research, we will refine the above studies to further investigate the contribution of PT320's neurotrophic effects to neuronal structure and function in a cell type–specific manner. Supplemental Material sj-pptx-1-pkn-10.1177_1877718X261477621 - Supplemental material for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease Supplemental material, sj-pptx-1-pkn-10.1177_1877718X261477621 for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease by Kuan-Yin Tseng, Tung-Tai Kuo, Pi-Kai Chang, Jian-Liang Chou, Eagle Yi-Kung Huang, Nigel H Greig, Ho-Il Choi, Lars Olson, Barry J Hoffer and Yuan-Hao Chen in Journal of Parkinson's Disease sj-pptx-2-pkn-10.1177_1877718X261477621 - Supplemental material for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease Supplemental material, sj-pptx-2-pkn-10.1177_1877718X261477621 for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease by Kuan-Yin Tseng, Tung-Tai Kuo, Pi-Kai Chang, Jian-Liang Chou, Eagle Yi-Kung Huang, Nigel H Greig, Ho-Il Choi, Lars Olson, Barry J Hoffer and Yuan-Hao Chen in Journal of Parkinson's Disease sj-docx-3-pkn-10.1177_1877718X261477621 - Supplemental material for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease Supplemental material, sj-docx-3-pkn-10.1177_1877718X261477621 for PT320, a GLP-1 receptor agonist, mitigates nucleus accumbens–associated depression- and anxiety-like behaviors in MitoPark mice of Parkinson's disease by Kuan-Yin Tseng, Tung-Tai Kuo, Pi-Kai Chang, Jian-Liang Chou, Eagle Yi-Kung Huang, Nigel H Greig, Ho-Il Choi, Lars Olson, Barry J Hoffer and Yuan-Hao Chen in Journal of Parkinson's Disease