UQCRC1 is a Key Pathogenic Determinant and Potential Therapeutic Target for Cognitive Impairment in Alzheimer’s Disease

Abstract

Alzheimer’s disease (AD) is a chronic neurodegenerative disorder and the most prevalent cause of dementia, yet it remains incurable. Ubiquinol-cytochrome c reductase core protein 1 (UQCRC1), a pivotal subunit of mitochondrial complex III, has been implicated in Alzheimer’s disease pathogenesis, though its precise mechanistic contributions remain undefined. In this study, we systematically investigated the mechanistic involvement of UQCRC1 in AD pathogenesis. Our findings reveal significant downregulation of UQCRC1 expression in hippocampal tissues from both AD patients and APP/PS1 transgenic mice. Conditional UQCRC1 knockdown in AD model mice exacerbated cognitive deficits while transmission electron microscopy analysis demonstrated that UQCRC1 deficiency induces pathological lysosomal enlargement, lipofuscin accumulation, and increased neuronal apoptosis in hippocampal neurons. Mechanistic interrogation revealed that UQCRC1 depletion triggers lysosomal Ca²⁺ overload-mediated proteolytic dysfunction coupled with activation of neuronal apoptotic pathways. Notably, adeno-associated virus-mediated UQCRC1 overexpression effectively reversed these pathological manifestations. Molecular dissection identified AMP-activated protein kinase (AMPK) signaling as the critical mechanistic mediator of this rescue effect, as pharmacological AMPK inhibition completely abrogated the therapeutic benefits. Together, our findings delineate a novel pathogenic axis linking mitochondrial complex III dysfunction to lysosomal degradation failure through UQCRC1-mediated AMPK regulation. These results position UQCRC1 not only as a promising biomarker for AD progression but also as a mechanistically validated therapeutic target, offering new insights into mitochondrial-lysosomal crosstalk in Alzheimer’s disease pathology.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12035-025-05171-2.

Introduction

Alzheimer’s disease (AD) is the most prevalent neurodegenerative dementia, accounting for 60–80% of global cases. It presents a critical paradox in modern medicine: while its neuropathological signatures—including amyloid-beta (Aβ) plaques, neurofibrillary tangles, and progressive synaptic loss—are well-characterized, the fundamental mechanisms driving initial pathogenesis remain enigmatic [1]. The global health impact is staggering, with 5–10% of individuals over 65 affected worldwide, a prevalence projected to triple by 2050 with population aging [1]. Clinically progressing from subtle memory deficits to complete functional dependency, AD not only devastates patient autonomy but also imposes immense socioeconomic burdens, with annual global costs exceeding $1 trillion [2]. Current therapeutic strategies confront three principal challenges: (1) incomplete understanding of multifactorial pathogenesis; (2) blood–brain barrier limitations compromising drug bioavailability; and (3) discordance between molecular biomarker progression and clinical symptom emergence [3–6]. Particularly critical are knowledge gaps regarding crosstalk between genetic susceptibility and environmental factors, as well as molecular events during the decade-long preclinical phase preceding cognitive decline—deficiencies that continue to hinder the development of disease-modifying interventions.

In neural systems, mitochondria function as metabolic command centers, coordinating energy production through oxidative phosphorylation while regulating calcium homeostasis, redox balance, and apoptotic signaling [7, 8]. This dual role renders mitochondrial integrity essential for maintaining neuronal viability, particularly in energy-intensive synaptic regions. Emerging evidence positions mitochondrial dysfunction as a unifying mechanism in neurodegeneration, with early-stage impairments preceding clinical symptom onset across multiple dementia subtypes [9, 10]. Notably, disease-specific pathogenic proteins such as Aβ in Alzheimer’s disease and α-synuclein in Parkinson’s disease exhibit direct mitochondrial targeting. These aberrant proteins establish a vicious cycle by impairing mitochondrial function while simultaneously amplifying their own neurotoxic effects through mitochondrial-mediated pathways [11, 12]. This multifactorial damage architecture positions mitochondrial dysfunction as a critical interface connecting genetic predispositions, aging-related vulnerabilities, and environmental stressors. Importantly, mitochondrial failure operates both as an initiating driver in early neurodegeneration and a perpetuating mechanism that accelerates irreversible disease progression through self-reinforcing pathological loops.

The respiratory chain is composed of five enzyme complexes (complexes I–V) and two mobile electron carriers (coenzyme Q and cytochrome c). Respiratory supercomplexes (comprising complexes I, III, and IV) pump protons from the mitochondrial matrix to the intermembrane space, which establishes an electrochemical gradient that powers adenosine triphosphate (ATP) synthase activity [13, 14]. UQCRC1 functions as an essential structural subunit for complex III assembly while simultaneously stabilizing the conformational architecture of respiratory supercomplexes [15, 16]. Previous research has shown that multiple complex III subunits (including UQCRC1) are significantly downregulated in AD brains [17]. Notably, UQCRC1 has been highlighted in transcriptomic studies as a differentially expressed “hub” gene in aging and AD tissues [18]. Indeed, even a moderate reduction of UQCRC1 in otherwise normal mice leads to decreased complex III activity and ATP levels in the brain, causing learning and memory impairment [19]. Such studies raised the possibility that UQCRC1 might serve as a marker of mitochondrial dysfunction in AD, but its functional significance was unclear.

In this study, we delineate the pathogenic role of UQCRC1 in Alzheimer’s disease (AD). Mechanistically, UQCRC1 deficiency drives lysosomal dysfunction through AMPK signaling deactivation, ultimately exacerbating AD-associated cognitive deterioration. Notably, therapeutic modulation of UQCRC1 expression ameliorated cognitive impairment in AD murine models. These findings nominate UQCRC1 as a mechanistically grounded therapeutic target for AD intervention.

Methods

Animals

Eight male C57BL/6 wild-type (WT) mice and 56 APP/PS1 transgenic mice aged 24–28 weeks (body weight 20–30 g) were procured from Changzhou Cavens Experimental Animal Co., Ltd. (Jiangsu, China). All animals were maintained under specific pathogen-free (SPF) conditions in standard polycarbonate cages (dimensions: 330 × 210 × 170 mm; 5 mice/cage) at the Army Medical University Animal Research Facility. Housing parameters included the following: 12-h light/dark cycle (08:00–20:00), controlled ambient temperature (20–23 °C), humidity (50–60%), and ad libitum access to autoclaved food/water. All experimental procedures were prospectively designed and strictly adhered to institutional ethical standards approved by the Army Medical University Laboratory Animal Welfare Committee (Protocol No. AMUWEC2024528110; Approval Date: 10 January 2024).

Cell Culture and Genetic Manipulation

Primary hippocampal neurons (ScienCell Research Laboratories, CAT# M1540-57) were maintained in neurobasal medium, following standardized protocols. Gene-specific siRNA targeting UQCRC1 (sense: 5′-GCACAGACUUGACUGACUAdTdT-3′; antisense: 5′-UAGUCAGUCAAGUCUGUGCdTdT-3′; Tsingke Biotechnology Co., Ltd) was complexed with Lipofectamine 3000 transfection reagent (Thermo Fisher Scientific, CAT# L3000015) at a 1:1 (v/v) ratio. Neurons were incubated with 50 nM siRNA complexes in Opti-MEM medium for 12 h before replacing the transfection medium with complete culture medium. Recombinant adeno-associated virus AAV2/9-hSyn-UQCRC1-Flag-P2A-EGFP-WPRE-pA (Taitool Bioscience, 2.15 × 10¹³ vg/mL) was diluted in PBS/MgCl₂ (1 mM) and applied to neuronal cultures at an MOI of 1 × 10⁵ viral genomes per cell for 24 h under standard culture conditions (37 °C, 5% CO₂). All cellular experiments were performed in triplicate with data obtained from three independent biological replicates.

Public Database

Methodologically, we conducted a systematic search in the Gene Expression Omnibus (GEO) database to identify whole-genome transcriptome sequencing datasets from neurodegenerative disease studies focusing on cognitive impairment phenotypes. Through cross-dataset analysis of the collected profiles, we identified commonly downregulated genes and performed subsequent prioritization analysis to assess the biological significance of UQCRC1 in disease pathogenesis. To investigate cell-type specific expression patterns, we extended our analysis to single-cell RNA sequencing datasets, systematically quantifying UQCRC1 expression across distinct neuronal subtypes. Finally, we performed a meta-analysis of multiple Alzheimer’s disease (AD) transcriptomic datasets encompassing different brain regions, enabling comprehensive evaluation of regional UQCRC1 expression alterations in AD pathophysiology.

Western Blotting

Tissue lysates were prepared using a commercial protein extraction buffer (Sangon Biotech, CAT# C510003) formulated at a ratio of 1000 µL lysis buffer:10 µL PMSF:5 µL phosphatase inhibitor:1 µL protease inhibitor. Then, the hippocampus or cell samples were homogenized in ice-cold protein extraction buffer. Following the manufacturer’s instructions, the protein concentrations were determined using a bicinchoninic acid (BCA) protein assay kit (Epizyme, CAT# ZJ103). Twenty micrograms of protein was loaded onto 4–20% gradient gels (ACE Biotechnology, CAT# ET15420LGel). Proteins were then transferred to a PVDF membrane using the Bio-Rad transfer system. Membranes were treated with a fast-blocking buffer (MedChemExpress, CAT# HY-K1027) for 10 min to prevent non-specific binding. Primary antibodies were applied and incubated overnight at 4 °C. The antibodies used included rabbit polyclonal anti-UQCRC1 (diluted 1:1000; Proteintech, CAT# 21705–1-AP), rabbit polyclonal anti-cathepsin D (diluted 1:1000; Proteintech, CAT# 21327–1-AP), rabbit monoclonal anti-procathepsin D (diluted 1:1000; abcam, CAT# ab134169), rabbit polyclonal anti-LC3B (diluted 1:1000, Abcam, CAT# ab48394), rabbit polyclonal anti-AMPKα (diluted 1:1000, Cell Signaling Technology, CAT# 2532), rabbit monoclonal anti-phospho-AMPKα (Thr172) (diluted 1:1000, Cell Signaling Technology, CAT# 2535), and rabbit monoclonal anti-GAPDH (diluted 1:3000; Proteintech, CAT# 10,494–1-AP). The HRP-conjugated goat anti-rabbit IgG secondary antibody (ZSGB-BIO, CAT# ZB-2301) was diluted at a ratio of 1:3000 and incubated at room temperature for 1 h. Finally, the resulting images were analyzed quantitatively using densitometry via the ImageJ software (available at https://imagej.nih.gov/ij/, version 1.54).

Measurement of Reactive Oxygen Species Levels

Reactive oxygen species (ROS) levels were quantitatively assessed using the Organic Reactive Oxygen Species Detection Kit (Bestbio, CAT# BB-470515) following the manufacturer’s protocol. Fluorescence intensity was measured at 510 nm excitation and 610 nm emission wavelengths using a fluorescence microplate reader. Additionally, 50 µl of the supernatant was extracted for the quantification of protein concentration using the BCA assay. The level of reactive oxygen species within the tissue was assessed by calculating the ratio of fluorescence intensity to protein concentration.

ATP Assay

ATP levels were measured according to the manual of a commercial assay kit (Servicebio, CAT # G4309-48 T). Briefly, samples were lysed in extraction buffer, boiled at 100 °C for 2 min, cooled to room temperature, and subsequently centrifuged at 10,000 × g for 10 min at 4 °C. The ATP detection working solution was freshly prepared according to the manufacturer’s protocol. Subsequently, 20 µL of the resultant supernatant was added to 100 µL of the ATP detection working solution in opaque 96-well plates. After thorough mixing, luminescence intensity was quantified using a microplate reader.

Behavioral Testing

Each behavioral test began at 10 a.m. Two hours before the start of every trial, mice were allowed to become acquainted with the testing room. Different behavioral assessments were spaced 1 day apart. Following every test, the equipment was carefully cleaned with 75% alcohol before the next one. Examiners left the room during testing to reduce outside influence. EthoVision XT 11.5 (Noldus Inc., Netherlands) was used for data capture and analysis.

Novel Object Recognition Test (NOR)

Hippocampal-dependent short-term memory was evaluated through a standardized Novel Object Recognition paradigm [20]. Testing occurred in a 40 × 40 × 40 cm acrylic arena (Fig. 3A) across three sequential phases: habituation (10-min free exploration), familiarization (24-h post-habituation; two identical objects, F), and testing (2 h later; replacement of one familiar with a novel object, N). Each trial involved 10-min exploration periods with exploration times recorded. Recognition index (RI) was calculated as RI = tN/(tF + tN); discrimination index (DI) was calculated as DI = (tN − tF)/(tF + tN).

Fig. 3.

UQCRC1 knockdown exacerbates cognitive deficits in AD model mice. A–E Behavioral outcomes of the novel object recognition (NOR) test. A Schematic diagram of the NOR experimental paradigm. B, C No significant differences were observed in total movement distance (B) or total object sniffing duration (C) between experimental groups. D, E Notably, AAV-shUQCRC1 mice demonstrated significant impairments in both recognition rate (D) and novel object discrimination index (E) compared to control mice. F–I Spatial learning and memory assessment using the Barnes maze. F Illustration of the Barnes maze apparatus. G Escape latency increased during Barnes maze training sessions in AAV-shUQCRC1 mice. H Escape latency was prolonged during Barnes maze probe trials in AAV-shUQCRC1 mice. I AAV-shUQCRC1 mice required significantly more attempts to locate the target quadrant. J, K Nest-building analysis. J Schematic representation of the nest-building test (NBT). K Nesting behaviors were impaired in AAV-shUQCRC1 mice, as quantified by nesting scores. Data presented as median with interquartile range (attempts to locate the target hole and nesting score) or mean ± SEM (other parameters). *P < 0.05, **P < 0.01, ***P < 0.001

Nest Building Test (NBT)

We assessed hippocampal-dependent cognition through nestlet-based testing in UQCRC1-deficient mice. Animals were individually housed with standard nesting materials (5 cm² cotton pad and wood shavings) under ad libitum feeding conditions. Nest construction was documented photographically 24 h later, with particular focus on cotton pad utilization (Fig. 3J). Nest quality was quantified using established scoring criteria [21], excluding animals with compromised nesting materials from analysis.

Barnes Maze

The Barnes maze effectively assessed spatial learning and memory in mice using a circular platform with 20 holes, one containing a detachable refuge chamber (Fig. 3F). The protocol comprised training and testing phases. During training, mice were centrally positioned and observed until they entered the target chamber or reached a 3-min cutoff. Unsuccessful mice were gently guided to the chamber for 60-s habituation. Animals underwent three sessions per day over four consecutive days, with entry latency as the primary outcome measure. The testing phase, conducted 24 h post-training, maintained identical parameters and additionally quantifying search attempts.

Transmission Electron Microscope (TEM)

Mice were initially perfused with ice-cold PBS. The hippocampus was carefully isolated and sectioned into small fragments. The tissue samples were then fixed, dehydrated, infiltrated, and embedded as required. Ultrathin sections were prepared and stained with 2% uranyl acetate. Quantitative analysis of the images was performed using ImageJ software (NIH, https://imagej.nih.gov/ij/, version 1.54).

Cathepsin D Activity Assessment

Cathepsin D (CTSD) enzymatic activity was quantified using fluorometric analysis according to the manufacturer’s specifications (Abcam, CAT# ab65302, Fig. 4G). Fluorescence intensity measurements (excitation/emission: 328/460 nm) were recorded to indicators of enzymatic activity. The relative fluorescence units (RFU) obtained were subsequently normalized to total protein concentration for each sample to calculate specific activity values.

Fig. 4.

UQCRC1 knockdown induces lysosomal dysfunction and promotes apoptosis in AD model mice. A Representative transmission electron microscopy (TEM) images comparing lysosomal (white arrows) ultrastructure in hippocampal tissues between AAV-Ctrl and AAV-shUQCRC1 groups. Scale bar, 1 µm. B TEM visualization of lipofuscin aggregates (red arrows) in hippocampal neurons. Scale bar, 1 µm. C Nuclear morphological features distinguishing normal and apoptotic cells: Normal nuclei (left panel) exhibit intact nuclear membranes with smooth contours and homogeneous chromatin distribution, while apoptotic nuclei (right panel) display characteristic chromatin condensation, nuclear shrinkage, and peripheral chromatin margination. Scale bar, 1 µm. D Quantitative assessment of lysosomal volume distribution across experimental groups. E Lysosomal size frequency histogram demonstrating decreased small lysosomes and increased enlarged lysosomes in AAV-shUQCRC1 mice. F Lipofuscin accumulation elevated in AAV-shUQCRC1 mice compared to controls. G Quantitative analysis revealed a significant increase in apoptotic cells within the hippocampal region of AAV-shUQCRC1 mice. H Western blot images of LC3B-I, LC3B-II, procathepsin D, and cathepsin D in hippocampal lysates. I Densitometric quantification of LC3B-II, procathepsin D, and cathepsin D protein expression. J Cathepsin D (CTSD) enzymatic activity assays showing functional impairment in AAV-shUQCRC1 mice. K Hippocampal tissues from AAV-shUQCRC1 mice showed increased activation of both caspase-3 and caspase-9. Data are presented as mean ± SEM. *P < 0.05, ***P < 0.001

Assessment of Caspase 3 and Caspase 9

The Caspase 3/Caspase 9 Activity Assay Kit (MULTI SCIENCES, CAT# APC03/APC09) was used to evaluate the activation of caspase 3 and caspase 9. Homogenized hippocampal tissue was centrifuged at 12,000 rpm for 15 min at 4 °C after extraction. Following the manufacturer’s instructions, we collected the supernatant and prepared the reaction mixture. After 4 h of incubation at 37 °C, the samples were tested for absorbance at 405 nm.

Drugs and Cellular Treatments

Pharmacological agents were applied under the following experimental conditions: Aβ1–42 oligomers (Eagle Biosciences, CAT# SPR-488B) at 1 µM concentration for 24-h exposure, and the AMPK inhibitor BAY 3827 (Selleck, CAT# S9833) at 0.5 µM for 25-h treatment. All compounds were prepared in dimethyl sulfoxide (DMSO; Sigma-Aldrich, CAT# D8418) with final solvent concentration maintained ≤ 0.1%, serving as vehicle control in parallel experiments.

MTT Assay for Cell Viability Assessment

Cells (1 × 10⁴ per well) in 96-well plates were exposed to Aβ42 oligomers for 24 h. MTT reagent (Sang Biotech, CAT# A600799; 0.5 mg/mL final) was added for 4-h incubation under standard culture conditions. Following medium removal, formazan crystals were dissolved in 100 µL DMSO with vortex-assisted solubilization. Cell viability was quantified by measuring absorbance at 570 nm, calculated using the formula: [(OD₅₇₀_treatment − OD₅₇₀_blank)/(OD₅₇₀_control − OD₅₇₀_blank)] × 100.

qPCR

Total RNA was isolated from murine hippocampal tissues using the Magbead RNA Extraction Kit (Cwbiotech, CAT# W3711S) followed by genomic DNA elimination with RNase-free DNase I (Thermo Fisher Scientific, CAT# EN0521). Complementary DNA synthesis and subsequent amplification were performed using the PrimeScript RT-PCR Kit (Takara Bio, CAT# RR037A). Quantitative analysis employed the ΔΔCT method normalized to GAPDH expression. The primers are listed in Supplementary Table S1.

Lysosomal Ca²⁺ Assessment

Lysosomal calcium levels were evaluated using an intracellular calcium detection assay kit (BestBio, CAT# BB-481125) according to the manufacturer’s protocol. Briefly, cells were washed three times with Hanks’ Balanced Salt Solution (HBSS) and seeded into 96-well plates at a density of 5 × 10⁴ cells per well. Subsequently, the probe solution was applied to the cells, which were then incubated in a 37 °C dry incubator for 20 min. Fluorescence intensity was measured using a microplate reader with excitation/emission wavelengths set to 340/510 nm. Readings were recorded at 5-s intervals over a 30-s period, with the average value calculated as the baseline level. To induce lysosomal Ca²⁺ release, glycyl-L-phenylalanine 2-naphthylamide (GPN) (Sigma-Aldrich, CAT# P5891) was added to achieve a final concentration of 50 µM. Fluorescence measurements were then continued at 5-s intervals for 3 min to monitor dynamic calcium flux.

DQ-BSA

Lysosomal proteolytic activity was quantified using the fluorogenic substrate DQ™ Green BSA (Thermo Fisher Scientific, CAT# D12050). Cells were seeded 24 h prior to experimentation in black-walled, clear-bottom 96-well plates at a density of 1 × 10⁴ cells/well. Following serum starvation, cells were loaded with 10 µg/ml DQ Green BSA in high-glucose DMEM (Thermo Scientific, CAT# A33822) and incubated for 1 h at 37 °C under 5% CO₂. Subsequent to two PBS washes, protease activity was initiated by replacing the medium with 100 µL Earle’s Balanced Salt Solution (EBSS, starvation buffer, Servicebio, CAT# G4214). Real-time fluorescence measurements (excitation/emission: 495/525 nm) were acquired every 5 min over a 4-h monitoring period using a microplate reader. Proteolytic flux rates were determined from the linear range during the initial 90-min EBSS treatment period.

Intracranial Injection

Following craniotomy surgery, stereotaxic intracranial injection of adeno-associated virus (AAV) vectors was performed using a glass micropipette. Three experimental groups were established based on the viral constructs administered: (1) AAV-shUQCRC1 mice received AAV2/9-hSyn-shUQCRC1-EGFP-WPRE-pA; (2) AAV-UQCRC1 mice received AAV2/9-hSyn-UQCRC1-Flag-P2A-EGFP-WPRE-pA; and (3) AAV-Ctrl mice received the control vector AAV2/9-hSyn-EGFP-3 × Flag-WPRE-pA. The injection coordinates relative to bregma were determined as follows: anteroposterior − 2.0 mm, mediolateral ± 1.2 mm, and dorsoventral − 1.9 mm. Viral suspensions were administered at a volume of 250 nL per hemisphere using an automated microinjection system, with a 10-min infusion duration to ensure optimal viral diffusion.

Statistical Analysis

The Kruskal–Wallis H test, unpaired two-tailed t-tests, Mann–Whitney U test, one-way ANOVA, and two-way ANOVA were used to assess differences between groups. Statistical significance was defined as a p-value of less than 0.05. Data analysis was performed using GraphPad Prism (GraphPad Software, Boston, USA, version 9.5) and SPSS (IBM, New York, USA, version 27.0). Statistical comparisons were represented as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Results

The Expression of UQCRC1 in Hippocampal Tissues of AD Patients and AD Model Mice Was Downregulated

By analyzing transcriptomic datasets (GSE122063, GSE36980, GSE26927) from four cognitive impairment-associated neurodegenerative disorders (AD, Huntington’s disease, vascular dementia, and multiple sclerosis), we identified 200 consistently downregulated genes across all four conditions (Fig. 1A). Protein–protein interaction (PPI) network analysis of these co-downregulated genes revealed UQCRC1 as a top-ranked hub gene (top 10 connectivity score, Fig. 1B). To resolve cellular specificity, we reanalyzed a single-cell RNA sequencing dataset (GSE67835) containing 466 cells classified into eight neural subtypes (Fig. 1C). Notably, UQCRC1 exhibited predominant neuronal expression (Fig. 1D, E). Multicohort validation using whole-genome transcriptome datasets from AD patients (GSE48350, GSE36980, GSE5281) demonstrated significant UQCRC1 downregulation in hippocampal tissues (Fig. 1F; P = 0.000995), primary visual cortex (Fig. 1F; P = 0.021843), and entorhinal cortex (Fig. 1F; P = 0.024774) compared to controls. Consistent with human findings, APP/PS1 transgenic mice showed a reduction in hippocampal UQCRC1 protein levels by immunoblotting (Fig. 1G, H; P = 0.0002).

Fig. 1.

Significant downregulation of UQCRC1 expression in hippocampal tissues of AD patients and AD model mice. A Bioinformatics analysis of multiple neurodegenerative disease datasets from the GEO database (GSE122063, GSE36980, GSE26927) identified 200 consistently downregulated genes across four disease conditions. B Protein–protein interaction (PPI) network analysis of these 200 co-downregulated genes revealed UQCRC1 (green arrow) as one of the top 10 hub genes. C–E Re-analysis of a public single-cell RNA sequencing dataset (GSE67835) with eight annotated neural cell types (C) demonstrated predominant neuronal expression of UQCRC1 (D, E). F Meta-analysis of multiple AD datasets (GSE48350, GSE36980, GSE5281) showed the most pronounced decrease in UQCRC1 expression in hippocampal tissues of AD patients. G Representative western blot images comparing UQCRC1 protein levels in wild-type (WT) and APP/PS1 mice. H Quantitative analysis confirmed significant reduction of UQCRC1 protein expression in APP/PS1 mice. Data represent mean ± SEM. *P < 0.05, ***P < 0.001

UQCRC1 Deficiency Impairs Mitochondrial Function in AD Model Mice

To investigate the pathophysiological role of UQCRC1 in Alzheimer’s disease progression, we performed stereotactic knockdown of UQCRC1 in APP/PS1 mice through bilateral hippocampal delivery of AAV-shUQCRC1 (Fig. 2A). Quantitative immunoblotting confirmed a reduction in UQCRC1 protein levels compared to controls (Fig. 2B, C; P = 0.0001). Transmission electron microscopy demonstrated that downregulation of UQCRC1 expression results in mitochondrial morphological alterations, characterized by decreased mitochondrial area (Fig. 2D, E; P = 0.0018) and reduced aspect ratio (Fig. 2D, F; P = 0.0002). Subsequent evaluation of energy metabolism parameters showed that AAV-shUQCRC1 mice exhibited significantly lower hippocampal ATP levels (Fig. 2G; P = 0.0012) accompanied by elevated reactive oxygen species (ROS) production (Fig. 2H; P = 0.0036) compared to controls. Furthermore, qPCR analysis revealed that UQCRC1 knockdown disrupted the expression pattern of other mitochondrial complex III subunits. Specifically, we observed the upregulated expression of Uqcrc2 (Fig. 2I; P = 0.0113) and Uqcr7 (Fig. 2I; P = 0.0299), while mt-Cyb (Fig. 2I; P = 0.0401), Uqcr (Fig. 2I; P = 0.0130), and Uqcr10 (Fig. 2I; P = 0.0453) showed significant downregulation.

Fig. 2.

UQCRC1 deficiency impairs mitochondrial function in Alzheimer’s disease (AD) model mice. A Schematic representation of the experimental timeline. B Western blot analysis of UQCRC1 expression in hippocampal tissues from APP/PS1 mice injected with either control AAV (AAV2/9-hSyn-EGFP-3 × Flag-WPRE-pA, AAV-Ctrl) or UQCRC1-targeting shRNA-expressing AAV (AAV9-hSyn-shUQCRC1-EGFP-WPRE-pA, AAV-shUQCRC1). C Quantitative analysis of UQCRC1 protein expression levels. D Representative transmission electron microscopy (TEM) images comparing mitochondrial ultrastructural morphology between AAV-Ctrl and AAV-shUQCRC1 groups. Scale bars, 0.5 µm. E UQCRC1 knockdown significantly reduced the area of mitochondrial area in hippocampus. F Mitochondrial aspect ratio was decreased following UQCRC1 downregulation. G AAV-shUQCRC1 mice demonstrated reduced ATP production in hippocampus. H AAV-shUQCRC1 mice showed increased ROS accumulation in hippocampus. I mRNA expression profiles of all mitochondrial complex III subunits in hippocampal tissues from both groups. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001

UQCRC1 Knockdown Exacerbates Cognitive Deficits in AD Model Mice

To evaluate the impact of UQCRC1 downregulation on cognitive decline in AD pathogenesis, we systematically analyzed cognitive performance in AD model mice following UQCRC1 knockdown. Multidimensional behavioral assessments revealed significantly exacerbated cognitive deficits in AAV-shUQCRC1 mice. Novel object recognition tests showed a decrease in recognition rate (Fig. 3D; P = 0.0053) and discrimination index (Fig. 3E; P = 0.0053). Importantly, these cognitive deficits occurred without confounding differences in locomotor activity (total movement distance: Fig. 3B; P = 0.6882) or exploratory motivation (total object sniffing duration: Fig. 3C; P = 0.8002) between experimental groups. Spatial learning deficiencies in AAV-shUQCRC1 mice were particularly evident in Barnes maze evaluations. UQCRC1 knockdown mice displayed prolonged escape latencies during both training sessions (Fig. 3G; P < 0.0001) and probe trials (Fig. 3H; P = 0.0022), requiring significantly more attempts to locate the target quadrant (Fig. 3I; P = 0.0244). Furthermore, AAV-shUQCRC1 mice exhibited impaired nest-building capacity, as quantified by reduced nesting scores (Fig. 3J; P = 0.0210). Taken together, these findings establish that hippocampal UQCRC1 deficiency exacerbates fundamental cognitive and behavioral deficits in AD model mice. Notably, the AAV-shUQCRC1 mice continued to demonstrate persistent cognitive dysfunction even 3 months after UQCRC1 knockdown (Fig. S1).

UQCRC1 Deficiency Exacerbates Lysosomal Dysfunction and Apoptosis in Both Cellular and Animal Models of Alzheimer’s Disease

To investigate the mechanism underlying UQCRC1-mediated cognitive impairment in Alzheimer’s disease (AD), we performed ultrastructural analysis of hippocampal tissues using transmission electron microscopy (TEM). Quantitative morphometric analysis revealed significant lysosomal enlargement (Fig. 4A, D, E; P = 0.0176) and increased lipofuscin accumulation (Fig. 4B, F; P = 0.0110) in AAV-shUQCRC1 mice compared to control animals. Concurrently, AAV-shUQCRC1 mice exhibited elevated apoptotic cell density (Fig. 4C, G; P = 0.0005). Mechanistic investigation demonstrated impaired autophagy flux in AAV-shUQCRC1 mice, as evidenced by increased LC3B-II accumulation (Fig. 4H, I; P = 0.0099). Given the crucial role of cathepsin D (CTSD), a lysosomal aspartic protease essential for protein degradation, in preventing lipofuscinogenesis and neurodegeneration, we analyzed its inactive precursor, procathepsin D (activated to CTSD in lysosomes for nonspecific proteolysis), alongside CTSD expression and enzymatic activity. Remarkably, UQCRC1-deficient mice showed both reduced CTSD protein levels (Fig. 4H, I; P = 0.0013) and diminished enzymatic activity (Fig. 4J; P = 0.0049), while there was no difference in the expression of procathepsin D (Fig. 4H, I; P = 0.5918), indicating compromised proteolytic activation of this lysosomal hydrolase. Apoptotic pathway activation was further confirmed by 1.8-fold elevation in cleaved caspase-3 (Fig. 4K; P = 0.0027) and 1.4-fold increase in activated caspase-9 (Fig. 4K; P = 0.0219).

For experimental validation, we examined the effects of UQCRC1 knockdown on lysosomal function in Aβ-treated cellular AD models (Fig. 5A). Following confirmation of siRNA transfection efficiency (Fig. 5C; P = 0.1755, Aβ vs. Aβ-siUQCRC1), we observed significant differences in cell viability between Aβ cells (86.8 ± 3.0%) and Aβ-siUQCRC1 cells (69.5 ± 2.2%) after 24-h culture (Fig. 5B; P = 0.0236). Notably, UQCRC1-silenced Aβ-treated cells demonstrated enhanced calcium efflux from lysosomes to cytoplasm compared to control Aβ-treated cells (Fig. 5D–F; P = 0.0072, Aβ vs. Aβ-siUQCRC1), suggesting exacerbated calcium dyshomeostasis following UQCRC1 suppression. To evaluate lysosomal proteolytic activity, we employed DQ Green BSA assay in Aβ-treated cells with or without UQCRC1 knockdown. This fluorometric approach detects protease-dependent dequenching of BODIPY fluorescence upon lysosomal protein hydrolysis (Fig. 5G). Aβ-siUQCRC1 cells exhibited significantly reduced DQ Green BSA hydrolysis rates compared to Aβ-treated controls (Fig. 5H; P = 0.0182, Aβ vs. Aβ-siUQCRC1), indicating impaired lysosomal protease activity. Consistent with in vivo findings, UQCRC1 knockdown in vitro resulted in enhanced activation of caspase-3 (Fig. 5I; P = 0.0003, Aβ vs. Aβ-siUQCRC1) and caspase-9 (Fig. 5I; P = 0.0126, Aβ vs. Aβ-siUQCRC1), confirming apoptosis pathway activation.

Fig. 5.

UQCRC1 downregulation induces lysosomal dysfunction and intra-lysosomal Ca²⁺ accumulation in Aβ-treated cells. A Schematic illustration of experimental design. B Cell viability assessment after different treatments. C Quantitative RT-PCR analysis of UQCRC1 mRNA expression levels. D Assessment of lysosomal Ca²⁺ levels. Cytosolic Ca²⁺ baseline was recorded for 30 s by monitoring the fluorescence intensity of the cytoplasmic Ca²⁺ indicator Fluo-4. Cells were subsequently treated with Gly-Phe β-naphthylamide (GPN), which was activated within lysosomes following cellular uptake, leading to lysosomal membrane permeabilization and subsequent release of luminal contents into the cytoplasm. The consequent elevation in cytoplasmic Ca²⁺ levels was quantified via measuring changes in the fluorescence intensity of Fluo-4, thereby enabling the evaluation of intra-lysosomal Ca²⁺ stores. E Time-course measurement of Fluo-4 fluorescence intensity reflecting cytosolic Ca²⁺ dynamics. F Quantification of peak Fluo-4 fluorescence intensity values. G Assessment of lysosomal proteolytic activity using DQ-BSA. The self-quenched substrate is endocytically taken up and delivered to lysosomes. Proteolytic cleavage by lysosomal enzymes releases fluorescent monomers, enabling quantification of proteolytic activity. H Quantitative analysis of lysosomal proteolytic activity. I Aβ exposure markedly elevated caspase-3 and caspase-9 activation in murine hippocampal neurons, with UQCRC1 knockdown exacerbating this effect. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

UQCRC1 Overexpression Attenuates Lysosomal Impairment in Aβ-Treated Hippocampal Cells Through AMPK Activation

The AMP-activated protein kinase (AMPK), a pivotal metabolic sensor coordinating cellular responses to energy stress and oxidative challenges [22], has been implicated in lysosomal dysfunction across experimental systems [23–25]. We hypothesized that UQCRC1 downregulation induces lysosomal dysfunction through AMPK pathway modulation. Western blot analysis revealed that UQCRC1 knockdown significantly suppressed AMPK activation (phosphorylation at Thr172) (Fig. 6A, C; P < 0.0001) in AD model cells without altering total AMPK expression (Fig. 6A, B; P = 0.0582). Since AMPK signaling is usually found to be increased under conditions of energy stress, the decreased AMPK activity observed in Aβ-siUQCRC1 hippocampal cells seems counterintuitive. The negative regulation of AMPK signaling remains poorly characterized in the current scientific literature. To our knowledge, only the tumor suppressor folliculin (FLCN) has been identified as a direct negative regulator of AMPK activity. FLCN forms a functional heterotrimeric complex with folliculin-interacting proteins 1 and 2 (FNIP1/2), which has been shown to suppress AMPK signaling. Notably, a previous study demonstrated transcriptional upregulation of FLCN, FNIP1, and FNIP2 in UQCRC1 knockdown cells [31]. In the present study, we confirmed the genetic upregulation in FLCN (Fig. 6D; P = 0.0044), FNIP1 (Fig. 6D; P = 0.0212), and FNIP2 (Fig. 6D; P = 0.0172).

Fig. 6.

UQCRC1 downregulation suppressed AMPK activity in Aβ-treated cells. A Representative immunoblots of phosphorylated AMPK (p-AMPK) and total AMPK in hippocampal neurons under treatment conditions: Aβ exposure alone vs. Aβ exposure combined with UQCRC1 knockdown (Aβ + siUQCRC1). B Densitometric quantification of pAMPK and AMPK protein expression levels. C pAMPK/AMPK ratio analysis showing impaired AMPK activation after UQCRC1 downregulation. D mRNA expression levels of FLCN, FNIP1, and FNIP2 are upregulated upon UQCRC1 knockdown. E Schematic timeline of experimental procedures. F qPCR analysis validating Aβ-induced UQCRC1 downregulation and subsequent restoration through UQCRC1 overexpression. G Immunoblot of p-AMPK/AMPK in hippocampal cells across experimental groups. H Quantitative comparison of pAMPK/AMPK ratios between treatment groups. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Next, to establish mechanistic causality between downregulation of UQCRC1 and AMPK inhibition, we generated UQCRC1-overexpressing hippocampal neurons using AAV vectors (Fig. 6E) and quantified AMPK-dependent rescue effects. In Aβ-treated cells, UQCRC1 overexpression reversed pathological manifestations by restoring pAMPK levels to 72% of control values compared to 33% in Aβ-only cells (Fig. 6G, H; P = 0.0010, Aβ vs. Aβ-UQCRC1), while co-treatment with the AMPK inhibitor BAY-3827 abolished this therapeutic effect (Fig. 6G, H; P < 0.0001, Aβ-UQCRC1 vs. Aβ-UQCRC1-BAY). Concurrently, UQCRC1 overexpression normalized Aβ-induced LC3B-II accumulation, an effect similarly blocked by AMPK inhibition (Fig. 7A, B; P < 0.0001, Aβ vs. Aβ-UQCRC1; P = 0.0051, Aβ-UQCRC1 vs. Aβ-UQCRC1-BAY). Comprehensive lysosomal functional analysis demonstrated that UQCRC1 overexpression ameliorated Aβ-induced pathologies including lysosomal calcium accumulation (Fig. 7C, D; P = 0.0014, Aβ vs. Aβ-UQCRC1), compromised CTSD activity (Fig. 7E, P = 0.0012, Aβ vs. Aβ-UQCRC1), and impaired proteolytic capacity (Fig. 7F; P = 0.0100, Aβ vs. Aβ-UQCRC1). The therapeutic effects extended to caspase 3/9 activation (Fig. 7G; caspase 3: P = 0.0011, caspase 9: P = 0.0356, Aβ vs. Aβ-UQCRC1), with all benefits being AMPK-dependent as evidenced by complete abrogation upon pharmacological AMPK inhibition. These findings collectively demonstrate that UQCRC1 deficiency drives lysosomal dysfunction in AD models through AMPK inactivation.

Fig. 7.

Therapeutic targeting of UQCRC1 alleviates lysosomal dysfunction through AMPK activity restoration in Aβ-treated cells. A Immunoblot of LC3B-I/II in hippocampal neurons across experimental groups. B Quantification of LC3B-II protein expression levels. C Time-course Fluo-4 fluorescence intensity profiles reflecting cytosolic Ca²⁺ dynamics. D Quantification of peak Fluo-4 fluorescence intensities. E CTSD enzyme activity assay. F Quantitative assessment of lysosomal proteolytic activity. G Aβ exposure markedly elevated caspase-3 and caspase-9 activation in murine hippocampal neurons. UQCRC1 overexpression effectively restored these activation levels to physiological ranges, whereas co-treatment with AMP inhibitor completely abolished this restorative effect. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

UQCRC1 Overexpression Attenuates Cognitive Deficits in Alzheimer’s Disease Model Mice

Building upon our in vitro investigations, we subsequently explored UQCRC1’s capacity to rescue cognitive deficits in Alzheimer’s disease (AD) model mice in vivo. Initial validation confirmed that UQCRC1 downregulation exacerbated AMPK signaling dysregulation in AD model mice, with UQCRC1 deficiency reducing hippocampal AMPK activation (Fig. 8A, C; P < 0.0001) without altering total AMPK expression (Fig. 8A, B; P = 0.8891). Bilateral hippocampal delivery of AAV-UQCRC1 vectors (Fig. 8D) achieved robust transgene expression (Fig. 8E, F; P = 0.0005), enabling mechanistic interrogation. Western blot analysis revealed that UQCRC1 overexpression restored physiological AMPK activation (Fig. 8G, H; P < 0.0001, AAV-ctrl vs. AAV-UQCRC1). Meanwhile, the autophagic flow was also improved, as manifested by LC3B-II reduction (Fig. 6G, I; P < 0.0001, AAV-ctrl vs. AAV-UQCRC1). Behavioral phenotyping across three complementary paradigms demonstrated cognitive rescue of UQCRC1 treatment compared to AAV-Ctrl mice: AAV-UQCRC1 mice exhibited enhanced nest-building proficiency (Fig. 9A, C; P < 0.001, AAV-ctrl vs. AAV-UQCRC1). Novel object recognition tests showed a discrimination index (Fig. 9F; P = 0.0004, AAV-ctrl vs. AAV-UQCRC1) and an increase in recognition rate (Fig. 9G; P = 0.0004, AAV-ctrl vs. AAV-UQCRC1), without confounding differences in exploratory motivation (total object sniffing duration: Fig. 9D; P = 0.9129, AAV-ctrl vs. AAV-UQCRC1) between experimental groups or locomotor activity (total movement distance: Fig. 9E; P = 0.9454, AAV-ctrl vs. AAV-UQCRC1). Meanwhile, AAV-UQCRC1 accelerated spatial learning in Barnes maze tests—evidenced by shorter escape latencies during training (Fig. 9H; P = 0.0021) and probe trial latency (Fig. 9I; P = 0.0027, AAV-ctrl vs. AAV-UQCRC1), and a non-significant 31% fewer attempts to locate the target escape hole during testing phase (Fig. 9J; P = 0.0948, AAV-ctrl vs. AAV-UQCRC1). These conserved therapeutic effects across molecular, cellular, and behavioral domains establish UQCRC1’s dual clinical relevance as both a mechanistically informed therapeutic target and a potential prognostic biomarker for Alzheimer’s-associated cognitive decline, with AMPK pathway modulation serving as the central mechanistic axis.

Fig. 8.

Therapeutic targeting of the UQCRC1 restored AMPK activity in AD model mice. A Representative immunoblots of pAMPK and total AMPK in hippocampal tissues from APP/PS1 mice receiving control adeno-associated virus (AAV-Ctrl) or AAV encoding short hairpin RNA targeting UQCRC1 (AAV-shUQCRC1). B Densitometric quantification of pAMPK and AMPK protein levels. C pAMPK/AMPK ratio analysis showing AMPK activity suppression following UQCRC1 knockdown in vivo. D Schematic diagram of intrahippocampal injection sites. E Fluorescence micrograph showing EGFP expression in the hippocampus, confirming viral transduction efficiency. Scale bar, 10 µm. F qPCR analysis of UQCRC1 mRNA levels, demonstrating successful hippocampal overexpression via AAV delivery. G Immunoblot verification of pAMPK, AMPK, LC3B-I, and LC3B-II expression. H Quantitative comparison of pAMPK/AMPK ratios across experimental groups. I Quantitative comparison of LC3B-II protein levels across experimental groups. Data are presented as median with interquartile range (attempts) or mean ± SEM (other parameters). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 9.

Therapeutic targeting of the UQCRC1 ameliorates cognitive deficits in AD model mice. A Representative images from the nesting behavior assessment. B Heatmap visualization of exploration patterns during the novel object recognition test (N, novel object; F, familiar object). C Nesting scores improved significantly after UQCRC1 overexpression in APP/PS1 mice. D, E Total sniffing duration (D) and total movement distance (E) during the NOR test exhibited no significant differences among experimental groups. F, G AAV-UQCRC1 mice exhibited significantly enhanced novel object discrimination rates (F) and recognition rate (G) compared to AAV-Ctrl mice. H Navigation parameters during Barnes maze training phase showing reduced escape latency in AAV-UQCRC1 mice compared to AAV-Ctrl mice. I Improved cognitive performance demonstrated by decreased escape latency in the testing phase for the AAV-UQCRC1 group compared to AAV-Ctrl mice. J AAV-UQCRC1 mice required fewer attempts to locate the target escape hole compared to AAV-Ctrl mice. Data are presented as median with interquartile range (attempts) or mean ± SEM (other parameters). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Discussion

The present study identifies UQCRC1 as a critical regulator bridging mitochondrial dysfunction to lysosomal pathology in Alzheimer’s disease (AD) via modulation of AMPK signaling. Our findings not only expand the current understanding of metabolic dysregulation in AD but also propose novel diagnostic and therapeutic strategies targeting the UQCRC1-AMPK axis (Fig. 10).

Fig. 10.

Proposed mechanism linking UQCRC1 deficiency to cognitive impairment. UQCRC1 deficiency upregulates FLCN (folliculin), which suppresses AMPK activity, thereby impairing lysosomal hydrolytic capacity. This subsequently leads to hippocampal neuronal apoptosis and exacerbates cognitive deficits in mice. NADH, nicotinamide adenine dinucleotide (reduced form); NAD⁺, nicotinamide adenine dinucleotide (oxidized form); FADH₂, flavin adenine dinucleotide (reduced form); FAD, flavin adenine dinucleotide; Q, coenzyme Q; U, UQCRC1; CC, cytochrome C; and I–IV, mitochondrial respiratory chain complexes I–IV

Our study reveals that UQCRC1 expression is significantly reduced in both Alzheimer’s disease (AD) patient brains and the APP/PS1 mouse model. Functional analyses demonstrate that partial UQCRC1 deficiency accelerates cognitive decline. Importantly, AAV-mediated restoration of UQCRC1 expression effectively mitigates this pathological alteration. This finding highlights the critical role of UQCRC1 in maintaining mitochondrial respiratory chain integrity and neuronal homeostasis, suggesting that its depletion lowers the threshold for neurodegeneration and cognitive dysfunction in AD. This conclusion aligns with existing evidence showing coordinated downregulation of multiple complex III subunits (including UQCRC1) in AD brains [17]. Transcriptomic profiling studies further identify UQCRC1 as a central hub gene exhibiting altered expression patterns in both aging and AD tissues [18]. Notably, even subcritical reductions of UQCRC1 in wild-type mice result in measurable decreases in complex III activity and cerebral ATP levels, accompanied by significant learning and memory deficits [19]. While these correlative observations suggested UQCRC1 as a potential biomarker of mitochondrial dysfunction in AD, its mechanistic involvement remained undefined prior to this investigation. Key mechanistic insights from our current work demonstrate that UQCRC1 deficiency-induced lysosomal dysfunction constitutes a pivotal pathway linking mitochondrial impairment to AD pathogenesis.

Mitochondrial dysfunction has long been recognized as a hallmark of AD, contributing to oxidative stress and bioenergetic failure [26], and parallel evidence implicates impaired lysosomal-autophagic clearance in AD neurodegeneration [27, 28]. However, direct integration of these two pathological features has remained elusive. Our study bridges this gap by demonstrating that knockdown of UQCRC1—a mitochondrial protein downregulated in AD—can directly precipitate lysosomal failure and cell death. We found that downregulation of UQCRC1 led to excessive Ca²⁺ sequestration within lysosomes (“lysosomal Ca²⁺ overload”), coincident with reduced catabolic efficiency. Regulated lysosomal calcium dynamics are crucial for normal autophagy and lysosome function: the controlled release of Ca²⁺ from lysosomes through channels like TRPML1 (MCOLN1) triggers downstream events needed for autophagosome-lysosome fusion and activation of degradative enzymes [29, 30]. In our UQCRC1-deficient neurons, this Ca²⁺ release appears to be impaired, leading to an abnormal build-up of luminal Ca²⁺ and consequent lysosomal dysfunction. This observation is in excellent agreement with recent work demonstrating that mitochondrial respiratory chain defects can interfere with lysosomal Ca²⁺ handling. For example, Raimundo and colleagues showed that respiratory chain impairment deactivates AMPK signaling (via upregulation of the AMPK inhibitor folliculin), which in turn reduces TRPML1 channel activity and causes lysosomal Ca²⁺ accumulation. The resulting failure of Ca²⁺ efflux from lysosomes was linked to a loss of lysosomal hydrolytic capacity, establishing a mechanistic link between mitochondrial malfunction and impaired lysosomal catabolism [31]. Our findings strongly support this model in a neuronal context: UQCRC1 loss (a complex III defect) likely initiates a similar cascade, culminating in lysosomal calcium overload and proteolytic failure. Importantly, the accumulation of undigested substrates in dysfunctional lysosomes (as evidenced by enlarged lysosomal structures, accumulation of lipofuscin, and elevated LC3B-II levels) can activate stress pathways that lead to neuron damage.

A pivotal discovery of this study is that UQCRC1 links mitochondrial dysfunction with lysosomal dysfunction through the AMP-activated protein kinase (AMPK) signaling. AMPK is a highly conserved protein kinase and a major responder to mitochondrial stress, which is very sensitive to changes in the ratio of AMP to ATP. AMPK is a master metabolic regulator that becomes activated under low-energy conditions to restore energy balance, partly by enhancing autophagy [32, 33]. We initially expected UQCRC1 deficiency (and the resulting drop in ATP) to activate AMPK. However, our results indicated a more complex picture of impaired AMPK signaling in UQCRC1-deficient neurons. Consistent with our observations, a recent study has shown that chronic UQCRC1 defects can deactivate AMPK signaling despite energy stress [31]. AMPK signaling normally promotes lysosomal health by regulating lysosomal biogenesis and function, and it is required for the activity of lysosomal Ca²⁺ channels such as TRPML1 (MCOLN1) [34]. This matches our observation of lysosomal Ca²⁺ overload in UQCRC1-deficient models. Beyond autophagy, deficits in AMPK signaling have been linked to exacerbated tau pathology, as seen in diabetic models where impaired AMPK coincides with increased tau hyperphosphorylation, and appropriate AMPK activation can inhibit neurotoxic processes [35]. Thus, it is a fulcrum between metabolic state and neurodegenerative changes. Our findings reinforce that maintaining normal AMPK activity is neuroprotective. Encouragingly, by rescuing UQCRC1, we restored AMPK function, which may explain the broad protective effect (spanning from lysosomal function recovery to cognitive improvement) observed in treated mice. These findings position AMPK modulation as a promising therapeutic strategy to counteract mitochondrial dysfunction-associated neurodegeneration. In fact, pharmacological interventions targeting AMPK activation or downstream effectors such as lysosomal regulators have been found to mitigate Alzheimer’s disease (AD) pathology. For example, metformin (an AMPK activator used in diabetes) and resveratrol (a SIRT1 activator that secondarily activates AMPK) have shown promising effects in preclinical AD models [36, 37]. These preclinical demonstrations corroborate our conclusion regarding the therapeutic potential of AMPK activation in neurodegenerative contexts.

Limitations

Despite the encouraging findings, our study has several limitations that warrant consideration: (1) The exclusive use of APP/PS1 introduces translational limitations, as these models primarily replicate Aβ pathology without fully recapitulating critical AD hallmarks such as neurofibrillary tangles or sporadic AD’s multifactorial etiology (e.g., APOE4 interactions, neuroinflammation). Future work integrating tauopathy strains, APOE-knock-in systems, and human cell models will better address cross-pathway interactions and translational relevance to AD heterogeneity. (2) This study exclusively used male mice to minimize confounding hormonal fluctuations and to avoid sex-specific differences in genetic risk factors (e.g., the APOE ε4 allele). However, this approach restricts the generalizability of our findings. Therefore, future studies should incorporate both sexes to fully elucidate the role of UQCRC1 in AD pathogenesis and to assess the potential sex-specific efficacy of therapeutic interventions targeting mitochondrial-lysosomal pathways. (3) Our study focused primarily on neurons due to their high UQCRC1 expression (Fig. 1D, E), their mitochondrial vulnerability [38, 39], and their critical role in neurodegeneration [40]. This neuron-centric approach excludes direct assessment of other brain cell types. Currently, it is known that UQCRC1 is undetectable in microglia [19]. However, the role of UQCRC1 in other brain cell types, such as astrocytes or oligodendrocytes, and its involvement in the progression of AD requires further research. Future studies integrating single-cell omics and cross-species models will clarify UQCRC1’s cell-type-specific contributions and optimize pan-cellular therapeutic strategies.

Conclusion

In summary, our study reveals that UQCRC1, a mitochondrial complex III subunit, plays a pivotal role in linking energy metabolism to proteostatic balance in the brain. UQCRC1 deficiency exacerbates the pathological cascade of AD by crippling AMPK signaling and lysosomal clearance, thereby accelerating cognitive decline and neuronal damage. Conversely, restoring UQCRC1 revives metabolic and lysosomal functions, leading to preserved neuronal integrity and improved cognition. These findings not only improve our understanding of AD pathogenesis but also identify UQCRC1 as a promising dual-purpose target: a marker of disease state and a lever for therapeutic intervention.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AAV

Adeno-associated virus

AD

Alzheimer’s disease

AMP

Adenosine monophosphate

AMPK

AMP-activated protein kinase

AMUWEC

Army Medical University Laboratory Animal Welfare Committee

ANOVA

Analysis of variance

APP

Amyloid precursor protein

APP/PS1

Amyloid precursor protein/presenilin 1

ATP

Adenosine triphosphate

Aβ

Amyloid-beta

Aβ1–42

Amyloid-beta 1–42

BSA

Bovine serum albumin

caspase-3/9

Cysteine-aspartic proteases 3/9

Ctrl

Control

CTSD

Cathepsin D

DMEM

Dulbecco’s Modified Eagle Medium

DMSO

Dimethyl sulfoxide

DQ-BSA

Dye-Quenched™ bovine serum albumin

EBSS

Earle’s balanced salt solution

EGFP

Enhanced green fluorescent protein

Fluo-4 AM

Fluo-4 acetoxymethyl ester

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

GEO

Gene Expression Omnibus

GPN

Glycyl-L-phenylalanine 2-naphthylamide

HBSS

Hanks’ balanced salt solution

HRP

Horseradish peroxidase

IgG

Immunoglobulin G

iPSC

Induced pluripotent stem cell

LC3B-II

Microtubule-associated protein 1A/1B-light chain 3-II

MCOLN1

Mucolipin 1

MOI

Multiplicity of infection

MTT

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

NBT

Nest building test

NIH

National Institutes of Health

NOR

Novel object recognition test

pA

Polyadenylation signal

pAMPK

Phosphorylated AMP-activated protein kinase

PBS

Phosphate-buffered saline

PPI

Protein-protein interaction

PS1

Presenilin 1

PVDF

Polyvinylidene fluoride

qPCR

Quantitative polymerase chain reaction

RFU

Relative fluorescence units

RNA

Ribonucleic acid

ROS

Reactive oxygen species

SEM

Standard error of the mean

shRNA

Short hairpin RNA

shUQCRC1

Short hairpin RNA targeting UQCRC1

siRNA

Small interfering RNA

SIRT1

Sirtuin 1

SPF

Specific pathogen-free

TEM

Transmission electron microscope

TRPML1

Transient receptor potential cation channel subfamily M member 1

UQCRC1

Ubiquinol-cytochrome c reductase core protein 1

vg/mL

Viral genomes per milliliter

WPRE

Woodchuck hepatitis virus posttranscriptional regulatory element

WT

Wild-type

Author Contributions

Jing Zhang, Fuhai Bai, Hong Li conceived the overall project and experimental design. Jing Zhang, Zuoxi Wu, Zonghong Long performed the experiments. Jing Zhang, Ceng Feng analyzed the data. Jing Zhang and Zonghong Long wrote the manuscript. Fuhai Bai provided technical supports for the experiment. Fuhai Bai, Hong Li gave guidance to the experiment and revised the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by the General Program of National Natural Science Foundation of China (No. 82171265), National Natural Science Foundation of China (No. 82101273), and Second Affiliated Hospital of Army Medical University Incubation Program for Young Doctoral Talents (No. 2023YQB007).

Data Availability

The original datasets generated and analyzed during this study are available from the corresponding author upon reasonable request. Publicly available bioinformatics datasets (GSE122063, GSE36980, GSE26927, GSE67835, GSE48350, and GSE5281) supporting this research were obtained from the Gene Expression Omnibus (GEO) repository, accessible at https://www.ncbi.nlm.nih.gov/geo/.

Declarations

Ethics Approval and Consent to Participate

All animal experimental procedures were prospectively designed and strictly adhered to institutional ethical standards approved by the Army Medical University Laboratory Animal Welfare Committee (Protocol No. AMUWEC2024528110; Approval Date: 10 January 2024). All efforts were made to minimize suffering and the number of animals used. The individuals involved in the GEO database have obtained ethical approval. Our study is based on open data, there are no ethical issues and other conflicts of interest.

Consent for Publication

The article is original, has not already been published in a journal, and is not currently under consideration by another journal.

Competing Interests

The authors declare no competing interests.

Clinical Trial Number

Not applicable.