Beyond molecular chaperoning: AHA1 reprograms autophagy flux through direct ATP5A1 interaction in ischemic neuronal injury

Abstract

Background

Mitochondrial dysfunction and excessive reactive oxygen species (ROS) generation play a pivotal role in ischemic neuronal injury. The Activator of 90kDa heat shock protein ATPase homolog 1 (AHSA1/AHA1) has been implicated in regulating ATP synthesis and energy metabolism. Yet, its role in neurological functional impairment and mitophagy under pathological conditions remains unclear.

Methods

We utilized in vivo middle cerebral artery occlusion/reperfusion (MCAO/R) mouse models and in vitro oxygen-glucose deprivation/reperfusion (OGD/R) neuronal cell models. The study integrated bioinformatics, molecular biology techniques, histological analyses, behavioral tests, and genetic knockdown (siRNA) to elucidate the underlying mechanisms.

Results

Our findings demonstrate that I/R stress induces the transcription factor STAT3 to upregulate AHA1 expression. AHA1 then translocates to the mitochondria and directly interacts with the ATP synthase subunit ATP5A1. This interaction disrupts the cellular ATP/AMP ratio and increases ROS production, leading to mitochondrial damage. The resulting energy stress triggers the aberrant activation of the AMPK/mTOR/ULK1 signaling pathway, culminating in an excessive and detrimental flux of PINK1/Parkin-mediated mitophagy. Critically, silencing of AHA1 reversed these effects, suppressing pathological mitophagy, reducing infarct volume, and improving neurological outcomes.

Conclusion

This study reveals a novel, non-canonical function for AHA1 as a pathological driver in ischemic stroke. By directly interacting with ATP5A1, AHA1 links transcriptional stress responses to mitochondrial bioenergetic failure and excessive autophagy. Targeting the AHA1-ATP5A1 axis represents a promising therapeutic strategy to inhibit maladaptive mitophagy and protect against neurological outcomes.

Graphical abstract

Highlights

• •AHA1 is upregulated in ischemic stroke; its in vivo inhibition is neuroprotective.
• •AHA1 translocates to mitochondria and directly interacts with the subunit ATP5A1.
• •The AHA1-ATP5A1 axis drives pathological mitophagy via the AMPK/mTOR pathway.
• •Reveals a non-canonical, pro-autophagic function for AHA1 beyond Hsp90 co-chaperoning.

1. Introduction

The key pathological mechanism of neurological dysfunction after ischemic stroke is cerebral ischemia-reperfusion (I/R) injury [1]. When the blood flow to the brain tissue is blocked, the supply of oxygen and glucose sharply decreases [2], which in turn hinders the mitochondrial oxidative phosphorylation process, leading to a significant decline in ATP synthesis, resulting in the collapse of energy metabolism, the accumulation of lactic acid, and the occurrence of intracellular acidosis [3]. Importantly, ischemia itself already causes mitochondrial dysfunction and excessive production of ROS, which contributes to the initiation of cellular damage. Although reperfusion can restore the supply of blood oxygen, the simultaneous generation of ROS burst, calcium overload, and inflammatory cascade reactions will further aggravate the injury [4], causing the so-called “second hit”. Among them, mitochondrial dysfunction is at the center of the pathological response in I/R injury because mitochondria are not only the main source of energy supply for neurons but also an important signaling platform for regulating cellular damage and protective mechanisms. It can lead to the imbalance of superoxide dismutase, excessive production of ROS, and then trigger lipid peroxidation and DNA damage [5,6]; simultaneously, a cascade of reactions, including impaired cellular energy metabolism and subsequent mitochondrial dysfunction, collectively exacerbates neurological damage and ultimately leads to functional deficits [7,8]. At present, strategies that promote reperfusion, such as intravenous thrombolysis and mechanical thrombectomy, are guideline-recommended treatment options for ischemic stroke. Although reperfusion inevitably induces oxidative stress and mitochondrial dysfunction, reducing oxidative stress and excessive mitochondrial autophagy while restoring blood flow promptly and effectively remains an urgent issue to be addressed.

During ischemia reperfusion, mitochondria are susceptible to a decrease in membrane potential, blockage of the electron transport chain, and impaired ATP synthesis, thereby inducing an intracellular energy crisis and excessive ROS generation. As one of the cellular stress adaptation mechanisms, mitophagy can selectively eliminate damaged mitochondria and maintain the functional stability of the mitochondrial population [9]. However, increasing evidence suggests that under I/R conditions, mitophagy may be over-activated, which accelerates mitochondrial loss, exacerbates energy depletion, and ultimately promotes neuronal death [10]. In particular, the PINK1/Parkin-mediated mitophagy pathway can play a dual role: at appropriate levels, it protects neurons by removing dysfunctional mitochondria; whereas excessive or uncontrolled activation accelerates mitochondrial depletion and aggravates injury [11].

The co-chaperone AHA1 (activator of Hsp90 ATPase 1) is a co-chaperone of heat shock protein 90 (Hsp90), which is traditionally considered to promote the folding and functional maintenance of its client proteins by enhancing the ATPase activity of Hsp90. AHA1 (activator of Hsp90 ATPase homolog 1), as a key co-factor of Hsp90, regulates protein folding and stress response and maintains cellular homeostasis by enhancing its ATPase activity [12,13]. In the nervous system, dysregulated expression of AHA1, particularly its upregulation, is closely associated with protein misfolding-related diseases such as Alzheimer's disease [14]. Studies have shown that AHA1 can relieve the proteotoxic stress in neurodegenerative diseases by promoting the correct conformation of Hsp90-dependent client proteins (such as tau protein) [14,15]. However, while AHA1 is conventionally recognized for its Hsp90 co-chaperone activity, its specific expression pattern during cerebral ischemia-reperfusion (I/R) and whether it plays a role in the mitochondrial autophagy pathway independent of Hsp90 remain completely unknown.

Intriguingly, the initial experiments within this present study revealed that AHA1 was significantly upregulated following both in vitro OGD/R and in vivo MCAO/R. This upregulation was accompanied by molecular signatures of enhanced mitophagy, including the level of PINK1 in total cellular protein was significantly increased, and LC3B-II/I ratio, and decreased p62. Further proteomic analysis identified a novel interaction between AHA1 and the mitochondrial ATP synthase F1 α subunit (ATP5A1), which correlated with altered cellular ATP levels and AMPK signaling. This suggested that AHA1 might regulate mitophagy in I/R injury through a bioenergetic mechanism, rather than solely through its canonical Hsp90 pathway.

Based on these findings, we hypothesized that AHA1, upon its upregulation, interacts with ATP5A1 to impair ATP synthesis, thereby triggering the AMPK/mTOR/ULK1 signaling axis to promote excessive, PINK1/Parkin-dependent mitophagy and exacerbate ischemic brain injury. To test this hypothesis, this study combines in vitro and in vivo models to systematically investigate the expression, function, and molecular mechanisms of AHA1 in I/R injury. Our work not only uncovers a novel pathogenic role for AHA1 but also provides a strong rationale for targeting this pathway as a therapeutic intervention for acute stroke.

2. Materials and methods

2.1. Ethical statement

All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Jinzhou Medical University(250140-5) and conducted in compliance with the NIH Guide for the Care and Use of Laboratory Animals. Male C57BL/6 mice (8–10 weeks old, 20–25 g) were sourced from the Animal Center of Liaoning Changsheng Biotechnology. Efforts were made to minimize animal suffering.

For the human study, blood samples were procured from the Department of Neurosurgery at the First Affiliated Hospital of Jinzhou Medical University. The study included patients diagnosed with acute ischemic stroke who received thrombolytic therapy. Blood samples were collected within 24 h of the recanalization treatment. Inclusion criteria were: (1) age 18–80 years; (2) first-ever ischemic stroke confirmed by imaging; (3) pre-stroke modified Rankin Scale (mRS) score of 0–1. Exclusion criteria included hemorrhagic stroke, severe systemic diseases, and pregnancy. Healthy, age and sex-matched volunteers with no history of cerebrovascular disease served as the control group. All procedures were approved by the hospital's Ethics Committee (KYLL2025349), and written informed consent was obtained from all participants or their legal guardians, in adherence to the Declaration of Helsinki.

2.2. Animal experiment

Transient focal cerebral ischemia was induced in mice via middle cerebral artery occlusion and subsequent reperfusion (MCAO/R). Anesthesia was administered using 2 % isoflurane, and a silicone-coated 6-0 nylon filament was inserted into the internal carotid artery to occlude the middle cerebral artery for 90 min. Reperfusion for 24 h was achieved by withdrawing the filament after the 90-min occlusion period. Mice that underwent sham surgery were subjected to the same procedure but without arterial blockage. AHA1 siRNA or a scrambled control siRNA (GenePharma, Shanghai, China) was administered into the lateral ventricle (coordinates: AP −0.5 mm, ML +1.0 mm, DV −2.5 mm) for in vivo gene silencing, using a stereotaxic apparatus 24h prior to MCAO surgery. A total of 5 μL of siRNA solution (2 μg/μL) was injected at a rate of 0.3 μL/min.

2.3. Neurological and behavioral assessment

Neurological function was blindly assessed at 6, 12, and 24 h after reperfusion, as previously described [16]. The scoring criteria were defined as follows: 0, no neurological deficit; 1, failure to extend the right forelimb fully; 2, spontaneous circling to the right; 3, falling to the right; and 4, absence of spontaneous movement with loss of consciousness. All evaluations were performed independently by three blinded observers.

Locomotor activity was further assessed using the open-field test. Mice were placed individually in a 50 × 50 × 40 cm arena and allowed to explore freely for 10 min. A video tracking system (ANY-maze, Stoelting Co., USA) was used to record and quantify total distance traveled and average movement speed.

2.4. Evaluation of infarct volume

Volume of brain infarction was assessed through staining with 2,3,5-triphenyltetrazolium chloride (TTC). Brain sections (2 mm thick) were exposed to 2 %TTC at 37 °C for 20 min, then fixed in 4 % paraformaldehyde and photographed. Infarct size was determined as a fraction of the entire hemisphere's volume. Hematoxylin-eosin (HE) and Nissl staining were performed on paraffin-embedded brain sections for morphological evaluation.

2.5. Cell culture and OGD/R model establishment

The human neuroblastoma cell line SH-SY5Y (purchased from ATCC, USA) was used for all in vitro experiments. This cell line was chosen because of its neuronal-like characteristics and its widespread use as a reliable model for studying neuronal injury and neurodegenerative diseases. Cells were cultured in a 1:1 mixture of DMEM and F12 medium, supplemented with 10 % fetal bovine serum. For experiments, the cells were used in their undifferentiated, proliferative state. To establish the OGD/R model, cells were incubated in low-glucose DMEM (1 g/L) without fetal bovine serum at 37 °C under 1 % O₂ and 5 % CO₂ for 2 h, followed by incubation in high-glucose DMEM (4.5 g/L) with 10 % fetal bovine serum at 37 °C under 21 % O₂ and 5 % CO₂ for 24 h.

2.6. Plasmid and siRNA transfection

Cells were transfected with siRNA targeting AHA1, STAT3, or ATP5A1, or with negative control (NC) siRNA, using Lipofectamine 300 (Invitrogen, USA) following the manufacturer's protocol. The overexpression plasmid for ATP5A1 (OE-ATP5A1) and its corresponding control vector were similarly transfected.After 6 h, the medium was replaced, and cells were used for subsequent experiments 24–48 h post-transfection.

2.7. Western blot

Brain tissues or cells were lysed using RIPA buffer supplemented with inhibitors for proteases and phosphatases. Protein concentration was assessed with a BCA assay kit. Equal protein quantities were then separated via SDS-PAGE and transferred onto PVDF membranes. Subsequently, membranes were blocked using 5 % non-fat milk and exposed overnight at 4 °C to specific primary antibodies. AHA1, ATP5A1, STAT3, p-AMPK, AMPK, LC3B, p62, PINK1, Parkin, p-mTOR, mTOR, p-ULK1, ULK1, Tubulin, Lamin B1, TOM20, and β-actin used at 1:1000 dilution. After incubation with HRP-conjugated secondary antibodies, protein bands were visualized using an ECL kit and quantified using ImageJ.

2.8. Cellular fractionation

Mitochondrial and cytoplasmic fractions were separated using a commercial mitochondria isolation kit (Thermo, USA) according to the manufacturer's instructions. Approximately 5 × 10⁷ cells were used for each fractionation procedure to ensure adequate protein yield from each compartment.

2.9. Co-immunoprecipitation (Co-IP) and mass spectrometry

Protein interactions were examined using Co-IP. For each Co-IP reaction, SH-SY5Y cells were grown to 90 % confluency in a 10 cm dish and lysed, yielding approximately 1–2 mg of total protein. Protein extracts were incubated with anti-AHA1 antibody or control IgG overnight, followed by protein A/G agarose beads (Thermo, USA). Immunoprecipitates were analyzed by Western blotting for ATP5A1. For identification of AHA1-binding partners, immunoprecipitated samples from three independent, scaled-up Co-IP experiments were pooled and subjected to LC-MS/MS analysis.

2.10. Bioinformatic analysis

Multiple online databases, including JASPAR, hTFtarget, Chip Atlas, GTRD, ENCODE, and KnockTF, were used to predict the potential transcription factors of AHA1. JASPAR was employed to analyze the binding motifs and sites of STAT3 on the AHA1 promoter. Spearman correlation analysis was conducted to analyze the expression correlation between STAT3 and AHA1 mRNA in the hippocampus of MCAO mice.

2.11. GEO transcriptome dataset

The public dataset GSE16561 for cerebral infarction was downloaded from the GEO database (https://www.ncbi.nlm.nih.gov/geo) using the R language package GEOquery (version 2.70.0). GSE16561 includes samples from 39 patients with ischemic stroke (IS) and 24 control samples. After batch correction of the samples, the limma package was used for differential analysis.

2.12. GO and KEGG pathway enrichment analysis

GO and KEGG are popular tools for large-scale functional enrichment investigations, encompassing biological processes (BP), cellular components (CC), and molecular functions (MF). The KEGG repository, a widely utilized knowledge resource, systematically gathers diverse data including genomic details, biological pathways, illnesses, and medications. Utilizing the R software package clusterProfiler (version 4.10.), we conducted GO and KEGG pathway enrichment analysis on the differentially expressed genes of IIMR. The selection criteria included a p-value<0.05 and FDR (q-value)<0.25, with p-value adjustment performed using the Benjamini-Hochberg method.

2.13. Gene Set Enrichment Analysis (GSEA)

Gene Set Enrichment Analysis (GSEA) evaluates the distribution of genes within predefined gene sets across a phenotype-associated ranked gene list to determine their contribution to the phenotype. In this study, genes in the combined GEO dataset were initially ranked by logFC. GSEA was performed using the clusterProfiler R package (v4.10), with 100 permutations and gene set sizes constrained to 10–500 genes. Gene sets were obtained from the Molecular Signatures Database (MSigDB), specifically the c2 collection “Cp.All.V2024.1.Hs.Symbols.” Analysis encompassed all standard pathways in GMT format, with significance thresholds of p < 0.05 and FDR <0.25. P-values were adjusted using the Benjamini-Hochberg method.

2.14. Immunofluorescence and confocal microscopy

Cells on coverslips were fixed using 4 % PFA, permeabilized with 0.2 % Triton X-100, and blocked with 5 % BSA. Subsequently, cells were treated with primary antibodies targeting AHA1 and LC3B, or stained using Mito-tracker Red (Beyotime). Following incubation with secondary antibodies conjugated to Alexa Fluor (Invitrogen), nuclei were stained with DAPI. Imaging was conducted with a Zeiss LSM 880 laser scanning confocal microscope in Germany. Co-localization was quantified through line scan analysis.

2.15. Immunohistochemistry analysis

Paraffin-embedded sections were prepared using a standard protocol, deparaffinized, rehydrated, and subjected to antigen retrieval by autoclaving to inactivate endogenous peroxidases. The primary antibody was applied and incubated overnight at 4 °C. On the next day, the sections were incubated with the secondary antibody at 37 °C for 30 min, followed by DAB color development and nuclear counterstaining with hematoxylin.

2.16. Measurement of ATP, ROS, and mitochondrial membrane potential (MMP)

ATP levels in cells were quantified with a firefly luciferase-based ATP Assay Kit. ROS levels inside cells were determined with the DCFH-DA probe. The assessment of mitochondrial membrane potential was conducted using a JC-1 Assay Kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) All steps were carried out following the provided guidelines, and a multi-mode microplate reader was utilized to measure fluorescence or luminescence.

2.17. Transmission electron microscopy (TEM)

Fixed cells with 2.5 % glutaraldehyde, post-fixed using 1 % osmium tetroxide, dehydrated, and embedded in epoxy resin. Cut ultrathin sections (70 nm), stained with uranyl acetate and lead citrate, and observed under a transmission electron microscope (Hitachi H-765, Japan).

2.18. Quantitative real-time PCR (qRT-PCR) and ELISA

Total RNA was extracted with TRIzol (Invitrogen) and reverse-transcribed into cDNA. Quantitative real-time PCR (qRT-PCR) was performed using SYBR Green Master Mix, and relative expression levels were calculated by the 2⁻ΔΔCt method with GAPDH as the internal control. For ELISA, antibody-coated plates were blocked with BSA, incubated with human serum, and then with the secondary antibody. Signal development was achieved using a chromogenic substrate, and absorbance was measured immediately with a microplate reader (see Table 1).

Table 1.

qRT-PCR primer.

Gene name	Sequence(5’-›3′)
STAT3(Human)
Forward primer	CCCGTACCTGAAGACCAAGT
Reverse primer	ACACTCCGAGGTCAGATCCA
ATP5A1(Human)
Forward primer	ACTGGGCGTGTCTTAAGTATTG
Reverse primer	ACCAAGGGCATCAACTACAC
AHSA1(Human)
Forward primer	GCGGCCGCTTCTAGTAGTTT
Reverse primer	CATCTCTCTCCGTCCAGTGC
Stat3(Mouse)
Forward primer	GCCGCCGTAGTGACAGAGAA
Reverse primer	GGCAGCAACATCCCCAGAGT
Atp5a1(Mouse)
Forward primer	GCCCTCGGTAATGCTATTGA
Reverse primer	GCAATCGATGTTTTCCCAGT
Ahsa1(Mouse)
Forward primer	AACTCGAGCGAGTCCTCTCCTAGGACTA
Reverse primer	TTGCGGCCGCACACAAAGCTTAAATATG
GAPDH(Human)
Forward primer	CGGAGTCAACGGATTTGGTCGTAT
Reverse primer	AGCCTTCTCCATGGTGGTGAAGAC
Gapdh(Mouse)
Forward primer	CGACTTCAACAGCAACTCCCACTCTTCC
Reverse primer	TGGGTGGTCCAGGGTTTCTTACTCCTT

2.19. Statistical analysis

Statistical analysis involved SPSS 21. for data processing and GraphPad Prism 8. for graph generation. Results were expressed as mean ± SD. Group differences were assessed with a two-tailed Student's t-test, while multiple group comparisons utilized one-way ANOVA followed by Tukey's post hoc analysis. Statistical significance was set at p < 0.05.

3. Results

3.1. Bioinformatic analysis of clinical stroke datasets reveals dysregulation of mitochondrial homeostasis and autophagy

To comprehensively characterize transcriptomic alterations in ischemic stroke, we analyzed GEO dataset GSE16561, comprising 39 patients and 24 healthy controls (Fig. 1A). Differential expression profiling revealed a robust gene signature distinguishing stroke patients from controls (Fig. 1B). Functional enrichment analysis indicated that upregulated genes were primarily associated with autophagy, including ‘regulation of autophagy’ and ‘macroautophagy,’ alongside critical mitochondrial processes such as ‘mitochondrial transport’ and ‘mitochondrial gene expression’ (Fig. 1C). Consistently, Gene Set Enrichment Analysis confirmed coordinated activation of core mitochondrial pathways, including ‘mitochondrial ATP synthesis coupled electron transport,’ ‘mitochondrial membrane organization,’ and ‘mitochondrial transport’ (Fig. 1D). Collectively, these unbiased analyses delineate a dual transcriptomic response in ischemic stroke—marked enhancement of autophagy coupled with pervasive mitochondrial dysregulation—providing a mechanistic framework to guide subsequent experimental investigations.

Fig. 1.

Transcriptomic profiling of a clinical stroke dataset reveals coordinated upregulation of autophagy and mitochondrial pathways.

(A) Normalized gene expression across all samples, showing consistent data distribution. (B) Heatmap of differentially expressed genes (DEGs) distinguishing control and stroke groups. (C) GO enrichment of DEGs, highlighting autophagy-related processes (top) and mitochondrial-related processes (bottom). (D) GSEA of key mitochondrial pathways, showing significant positive enrichment for ‘mitochondrial ATP synthesis coupled electron transport,’ ‘mitochondrial membrane organization,’ ‘mitochondrial gene expression,’ and ‘mitochondrial transport’ in stroke samples, indicating coordinated upregulation.

3.2. AHA1 is upregulated in clinical stroke and its in vivo silencing confers potent neuroprotection

To establish clinical relevance, we analyzed samples from acute ischemic stroke patients and found that AHA1 was significantly elevated in both serum (Fig. 2A) and peripheral blood mononuclear cells (Fig. 2B) compared with healthy controls, consistent with our preclinical models. To evaluate therapeutic potential, AHA1 was silenced in a MCAO/R mouse model via intracerebroventricular siRNA administration. AHA1 knockdown markedly reduced cerebral infarct volume (Fig. 2C), preserved neuronal architecture (Fig. 2D–G), and increased neuronal survival. Functionally, silenced mice exhibited restored ATP levels (Fig. 2F), improved neurological scores (Fig. 2E), and enhanced motor and exploratory behaviors in the open field test (Fig. 2H–J). Mechanistically, ischemia-reperfusion upregulated both AHA1 and ATP5A1 and activated mitophagy, as indicated by Parkin accumulation (Fig. 3A–C); AHA1 silencing attenuated this activation, directly linking its inhibition to neuroprotection. Collectively, these findings demonstrate that AHA1 is upregulated in clinical stroke and that targeted inhibition in vivo preserves neuronal viability, restores energy metabolism, and drives functional recovery.

Fig. 2.

Clinical and in vivo validation of AHA1 in ischemic stroke.

(A,B) AHA1 is elevated in serum (ELISA, n = 6) and PBMCs (qRT-PCR, n = 3) from stroke patients versus controls. (C,D) AHA1 silencing in MCAO/R mice reduces infarct volume (TTC, n = 5) and preserves cortical neurons (H&E and Nissl). (E–G) Knockdown restores neurological scores (n = 10), ATP levels (n = 6), and neuronal survival (n = 5). (H–J) Open field test shows improved locomotion and exploratory behavior (n = 5). Data are mean ± SD; ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

Fig. 3.

AHA1 knockdown attenuates ischemia/reperfusion-induced mitophagy in vivo.

(A) Immunohistochemistry showing cortical expression of AHA1 and ATP5A1 after I/R and AHA1 silencing. (B) Quantified IOD of AHA1 and ATP5A1 (n = 5). (C) Western blot and densitometry of AHA1, PINK1, Parkin, LC3B-II/I, and p62 in ischemic brain tissue (n = 3). Data are mean ± SD; ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

3.3. OGD/R induces the translocation of AHA1 to mitochondria and its Co-localization with autophagosomes

To define AHA1's subcellular dynamics during ischemic stress, SH-SY5Y cells were subjected to OGD/R followed by fractionation and confocal imaging. Under basal conditions, AHA1 localized mainly to the cytoplasm, whereas OGD/R induced its pronounced accumulation in mitochondria, as confirmed by Western blot of isolated fractions (Fig. 4A). Confocal microscopy revealed a shift from diffuse cytoplasmic AHA1 to punctate signals that co-localized extensively with mitochondria and LC3B under OGD/R (Fig. 4B). Confocal analysis verified strong spatial overlap between AHA1, mitochondrial, and autophagosomal markers (Fig. 4C). These findings demonstrate that ischemic stress drives AHA1 translocation to mitochondria, where it associates with autophagosomes, implicating AHA1 in mitophagy regulation.

Fig. 4.

Illustrates the translocation of AHA1 to mitochondria and its co-localization with LC3B during oxygen-glucose deprivation/reperfusion (OGD/R) stress.

(A) AHA1's subcellular distribution in SH-SY5Y was assessed via Western blot analysis post-cellular fractionation. (B, C) Confocal immunofluorescence images and quantification display AHA1's localization. Scale bar, 20 μm.

3.4. AHA1 promotes the mitophagic flux in OGD/R-treated neurons

To determine whether AHA1 actively modulates mitophagic flux rather than merely influencing marker abundance, we employed Bafilomycin A1 (Baf A1) to block lysosomal degradation. Confocal microscopy showed that OGD/R increased LC3-II and TOM20 co-localization, further augmented by Baf A1, indicating active mitophagic flux. AHA1 silencing markedly diminished LC3-II/TOM20 overlap, and even with Baf A1, mitophagosome accumulation remained low, signifying impaired autophagosome formation (Fig. 5A–C). Consistently, Western blot analysis revealed that Baf A1-induced accumulation of LC3B-II and p62 under OGD/R was substantially blunted by AHA1 knockdown (Fig. 5B). These findings identify AHA1 as a positive regulator of mitophagy initiation, essential for sustaining mitophagic flux under ischemic stress.

Fig. 5.

AHA1 is a positive regulator of the mitophagic flux under OGD/R stress.

(A) Representative confocal immunofluorescence images (top) showing co-localization of mitochondria (TOM20, red) and autophagosomes (LC3-II, green). Nuclei were counterstained with DAPI (blue). Scale bar, 10 μm. (B) Western blot analysis (left) and quantification (right) of LC3BII/I and p62 protein levels (n = 3). (C) Quantification of the co-localization using Pearson's correlation coefficient (n = 3 independent experiments).Data are presented as mean ± SD. ∗∗p < 0.01.

3.5. Silencing of AHA1 preserves mitochondrial integrity and restrains aberrant mitophagy in OGD/R-treated neurons

To delineate AHA1's functional role in ischemic injury, SH-SY5Y cells were transfected with AHA1 siRNA before OGD/R. JC-1 staining demonstrated that OGD/R induced marked mitochondrial depolarization (ΔΨm), which was effectively prevented by AHA1 silencing, indicating preserved mitochondrial integrity (Fig. 6A). Likewise, AHA1 knockdown markedly reduced OGD/R-induced ROS accumulation, as shown by fluorescence imaging and flow cytometry (Fig. 6B and C). Western blot verified efficient AHA1 suppression and revealed inhibition of mitophagy signaling, evidenced by reduced PINK1 and Parkin accumulation, a lower LC3B-II/I ratio, and preserved p62 levels (Fig. 6D and E). Together, these findings identify AHA1 as a key driver of mitochondrial dysfunction and excessive mitophagy during OGD/R, whose inhibition restores mitochondrial homeostasis and mitigates oxidative injury.

Fig. 6.

AHA1 silencing preserves mitochondrial integrity and restrains mitophagy under OGD/R stress.

(A) JC-1 staining. (B–C) Intracellular ROS levels assessed by DCFH-DA fluorescence imaging and flow cytometry.(D–E) Western blot and quantitative analysis of AHA1, PINK1, Parkin, LC3B, and p62 expression in SH-SY5Y cells after OGD/R.Data are mean ± SD; ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

3.6. AHA1 directly interacts with ATP5A1 to drive PINK1/Parkin-mediated mitophagy

To identify AHA1's mitochondrial binding partner, co-immunoprecipitation in SH-SY5Y cells revealed a strong interaction between endogenous AHA1 and ATP5A1, the α-subunit of mitochondrial ATP synthase (Fig. 7A). AHA1 silencing markedly reduced ATP5A1 protein abundance under OGD/R, indicating that AHA1 regulates ATP5A1 during ischemic stress (Fig. 7B). Functionally, disruption of the AHA1-ATP5A1 axis attenuated PINK1/Parkin-mediated mitophagy: ATP5A1 knockdown decreased PINK1 and Parkin expression, reduced LC3B-II accumulation, and caused p62 retention (Fig. 7C–F). These findings demonstrate that AHA1 directly binds and regulates ATP5A1 to promote PINK1/Parkin-dependent mitophagy, establishing the AHA1-ATP5A1 complex as a key upstream regulator linking mitochondrial proteostasis to mitophagy under ischemic stress.

Fig. 7.

AHA1 directly interacts with ATP5A1 and drives PINK1/Parkin-mediated mitophagy in OGD/R-treated neurons. (A) Co-immunoprecipitation (Co-IP) assays were performed on lysates from SH-SY5Y cells. (B) Western blot analysis and quantification(n = 3). (C, D) Western blot analysis and quantification of key mitophagy-related proteins(n = 3). (E) Representative immunofluorescence images. (F) Western blot analysis and quantification of mitophagy markers in cells (n = 3). Data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

3.7. ATP5A1 acts downstream of AHA1 to mediate mitophagy and mitochondrial damage

To validate ATP5A1 as a downstream mediator of AHA1-driven mitophagy, functional rescue experiments were performed. AHA1 silencing in SH-SY5Y cells suppressed OGD/R-induced mitophagy and reduced oxidative stress. Critically, ATP5A1 overexpression in AHA1-deficient cells reversed mitophagic flux, LC3B-II and p62 levels (Fig. 8A), and reinstated ROS accumulation (Fig. 8B and C). Ultrastructural analysis by transmission electron microscopy confirmed these findings: AHA1 knockdown preserved mitochondrial integrity with elongated cristae, whereas ATP5A1 reintroduction reverted mitochondria to a damaged, swollen phenotype with fragmented cristae (Fig. 8D). These results establish ATP5A1 as a key downstream effector of AHA1, mediating pathological mitophagy and mitochondrial injury under ischemic stress.

Fig. 8.

ATP5A1 acts downstream of AHA1 to mediate mitochondrial damage and the pathological mitophagy flux. (A) Western blot analysis (left) and quantification (right) of ATP5A1, LC3BII/I, and p62 protein levels(n = 3). (B, C) Representative fluorescence microscopy images and quantification of intracellular ROS levels (stained with DCFH-DA, green) (n = 6). (D) Representative transmission electron microscopy (TEM) images showing mitochondrial ultrastructure. All quantitative data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

3.8. The detrimental effects of AHA1 are mediated by aberrant AMPK activation

We next examined AMPK/mTOR signaling in AHA1-silenced SH-SY5Y cells. AHA1 knockdown induced pronounced AMPK hyper-phosphorylation that to activate downstream pro-autophagic signaling: ULK1 phosphorylation was suppressed and mTOR phosphorylation increased, indicating a dysfunctional energy-sensing state (Fig. 9A). To test causality, AHA1-deficient cells were treated with the AMPK inhibitor Compound C during OGD/R. Pharmacological inhibition recapitulated the protective effects of AHA1 silencing, suppressing mitophagy, reducing PINK1/Parkin expression, lowering the LC3B-II/I ratio, and restoring p62 levels (Fig. 9B), while rescuing mitochondrial membrane potential (Fig. 9C) and cellular ATP content (Fig. 9D). Compound C also normalized signaling, reversing AMPK hyper-phosphorylation and restoring ULK1 and mTOR phosphorylation to control levels (Fig. 9E). These results indicate that AHA1 exacerbates ischemic neuronal injury via aberrant AMPK hyper-activation, which uncouples canonical AMPK/ULK1/mTOR signaling to drive excessive mitophagy and mitochondrial dysfunction, and that pharmacological inhibition of this pathway restores mitochondrial homeostasis.

Fig. 9.

AHA1 drives mitochondrial damage via aberrant hyper-activation of the AMPK pathway.(A) Western blot analysis and quantification (n = 3). (B) Western blot analysis and quantification of mitophagy markers (n = 3). (C) Representative JC-1 staining images and quantification (n = 6). (D) Cellular ATP levels were measured in the indicated groups (n = 6). (E) Western blot analysis and quantification of AMPK/mTOR pathway proteins (n = 3). All data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

3.9. STAT3 transcriptionally upregulates AHA1 expression under OGD/R stress

To investigate upstream regulation of AHA1, SH-SY5Y cells were subjected to OGD/R. Consistent with in vivo findings, OGD/R markedly upregulated AHA1 (Fig. 10A). Bioinformatic analysis identified STAT3 as a potential transcriptional regulator of AHA1 (Fig. 10B and C), supported by positive correlation in clinical datasets (Fig. 10D and E). STAT3 silencing with specific siRNA significantly attenuated OGD/R-induced AHA1 upregulation at both protein and mRNA levels (Fig. 10F–H). These results establish AHA1 as a direct transcriptional target of STAT3 under ischemic stress.

Fig. 10.

STAT3 directly regulates AHA1 transcription under OGD/R. (A) Western blot and quantification of AHA1 after OGD/R (n = 3).(B) Venn diagram of predicted transcription factors from six databases; STAT3 identified as core candidate. (C) Predicted STAT3 binding motif and high-confidence site (789–799) in the AHA1 promoter (JASPAR). (D–E) Correlation of STAT3 and AHA1 mRNA in two clinical stroke datasets. (F) qRT-PCR showing STAT3 knockdown efficiency with three siRNAs under OGD/R (n = 3). (G) Western blot and quantification of STAT3 and AHA1 protein levels after knockdown (n = 3). (H) qRT-PCR of STAT3 and AHA1 mRNA levels (n = 3). Data are mean ± SD; ∗p < 0.05, ∗p < 0.01; ns, not significant.

3.10. AHA1 functions via an Hsp90-Independent mechanism to Exacerbate Neuronal Injury

Although AHA1 is classically an Hsp90 co-chaperone, we investigated whether its pro-mitophagic role under OGD/R is Hsp90-dependent. AHA1 overexpression in SH-SY5Y cells exacerbated OGD/R-induced pathology, including increased LC3B-II/I ratio, enhanced p62 degradation, ATP depletion, and elevated ROS. Co-treatment with the Hsp90 inhibitor 17-AAG failed to reverse these effects (Fig. 11A–D). These findings indicate that AHA1 drives excessive mitophagy and neuronal injury via an Hsp90-independent mechanism, supporting a non-canonical model in which AHA1 directly interacts with ATP5A1 to reprogram mitochondrial bioenergetics and mitophagic flux during ischemic stress.

Fig. 11.

AHA1 Functions via an Hsp90-Independent Mechanism to Exacerbate Neuronal Injury.

(A) Western blot analysis and quantification of mitophagy markers LC3BII/I and p62 (n = 3). (B) Cellular ATP levels were measured in the indicated groups(n = 6). (C) Representative fluorescence images of intracellular ROS. (D) Quantification of the ROS fluorescence ratio (n = 6). All data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

4. Discussion

Autophagy, a conserved eukaryotic degradation pathway, is critical for maintaining intracellular homeostasis [17,18]. In CIRI, its role is context-dependent: moderate autophagy during early ischemia removes damaged organelles, limits apoptosis and necrosis [19]. For example, autophagy can timely degrade damaged organelles such as mitochondria due to ischemia and hypoxia, preventing them from releasing harmful substances and causing cell damage [20]. On the other hand, preserves neuronal function, whereas excessive autophagy promotes autophagosome accumulation, degrades essential cellular components, and exacerbates neuronal damage and infarct expansion [21,22]. Pharmacological studies support this duality, with metformin conferring neuroprotection via autophagy inhibition, while Salidroside mitigates injury through ERβ/BNIP3-mediated mitochondrial autophagy activation [23,24]. Here, we report that cerebral ischemia-reperfusion can excessively activate the mitochondrial autophagy pathway, and silencing AHA1 can inhibit autophagy, indicating that AHA1 plays a negative role in the injury.

AHA1, a co-chaperone of Hsp90, is traditionally known for its role in maintaining protein homeostasis [[25], [26], [27]], particularly in chronic neurodegenerative diseases [[28], [29], [30]]. However, its role in acute brain injury has remained unclear. Here, we demonstrate that ischemia-reperfusion stress markedly upregulates AHA1 expression in both OGD/R and MCAO models. Consistent with these findings, Clinical data from the GEO database, in which serum samples from stroke patients exhibited a pronounced increase in AHA1 expression compared with healthy controls. Notably, elevated AHA1 levels were also positively correlated with clinical indices of stroke severity and unfavorable neurological outcomes, suggesting that AHA1 may serve as a potential circulating biomarker for mitochondrial stress and disease progression in ischemic stroke. This clinical observation underscores the translational relevance of our experimental findings. AHA1 upregulation was accompanied by enhanced mitophagy, evidenced by increased PINK1 and Parkin expression and an elevated LC3B-II/I ratio. Silencing AHA1 reversed these changes, restored ATP levels, and reduced ROS accumulation, indicating that AHA1 functions as a critical mediator of pathological mitophagy and mitochondrial dysfunction during ischemia-reperfusion injury.

To clarify the functional role of AHA1, we silenced its expression in vitro using AHA1-siRNA. AHA1 knockdown reversed the aberrant alterations of autophagy markers, restored mitochondrial membrane potential (ΔΨm), and markedly reduced intracellular ROS levels, indicating that AHA1 suppression mitigates ischemia-reperfusion–induced mitochondrial dysfunction and oxidative injury. Consistently, intracerebroventricular delivery of AHA1-shRNA in MCAO mice improved neurological outcomes, reduced infarct volume, and preserved neuronal morphology, extending the protective effects observed in vitro to the in vivo setting. A central finding of this study lies in the use of Bafilomycin A1 (Baf A1) to functionally distinguish autophagic activity from autophagic flux, thereby delineating the specific role of AHA1 in this process. We observed that under OGD/R-induced injury, co-treatment with Baf A1 resulted in a marked accumulation of LC3B-II protein. This observation indicates that ischemia–reperfusion stress effectively activates autophagic flux, with an active “throughput” from autophagosome formation to lysosomal degradation. Critically, AHA1 knockdown significantly attenuated the Baf A1-induced LC3B-II accumulation. This result provides a mechanistic explanation for our previous observations: the reduction in basal LC3B-II/I levels upon AHA1 knockdown is not due to accelerated autophagosome degradation, but rather reflects an inhibition of autophagic flux at the initiation or formation stage. Serving as a “molecular probe,” Baf A1 blocks downstream autophagic flux, allowing inference of upstream autophagic activity. Our data clearly indicate that AHA1 is a key positive regulator of autophagosome biogenesis or maturation, whose function is essential for maintaining robust OGD/R-induced autophagic flux. Consequently, we conclude that AHA1 orchestrates the early steps of autophagic flux, thereby playing a critical role in “reprogramming” autophagy during ischemic neuronal injury—a function that aligns closely with its newly identified role in binding and regulating the mitochondrial ATP synthase subunit ATP5A1.

Mechanistically, co-immunoprecipitation and mass spectrometry identified ATP5A1, a key structural protein of mitochondrial complex V and is responsible for energy conversion in the catalytic process of ATP synthesis [31,32]. AHA1 knockdown restored ATP levels and reduced AMPK activation, whereas ATP5A1 overexpression partially reversed the autophagy suppression caused by AHA1 silencing, confirming that AHA1 regulates AMPK-dependent autophagy through ATP5A1-mediated mitochondrial energy metabolism. STAT3, a redox-sensitive transcription factor that regulates autophagy and oxidative stress responses in the nervous system, was confirmed to be activated under ischemic conditions and to promote AHA1 expression. STAT3 activation is known to respond to inflammatory and oxidative stimuli, and previous work has shown that ROS can further activate STAT3 signaling in a positive feedback manner. Thus, we established a STAT3–AHA1–ATP5A1 axis that links cellular stress signaling to mitochondrial dysfunction. Nevertheless, our investigation focused primarily on verifying this upstream-downstream relationship, and the precise transcriptional mechanisms require further study. Traditionally, AHA1 was considered to mainly act as a co-chaperone of HSP90, functioning by regulating the ATPase activity of HSP90 [33]. Our results demonstrate that AHA1 directly binds to ATP5A1, reprogramming mitochondrial bioenergetics and promoting mitophagy during ischemic stress. Inhibition of HSP90 with 17-AAG did not affect these processes, confirming the autonomous role of AHA1. These findings redefine AHA1 as a novel regulator of mitochondrial homeostasis and autophagic signaling, offering new mechanistic insight into ischemic neuronal injury. Further studies are needed to delineate the molecular basis of this non-canonical pathway. However, the precise mechanisms underlying this process remain to be further explored and validated.

Finally, our study provides compelling evidence that the AHA1-ATP5A1 axis serves as a critical determinant of pathological mitophagy, and that its inhibition confers substantial neuroprotection against ischemic injury. Nevertheless, several limitations warrant consideration. The maintenance of mitochondrial homeostasis relies on a delicate equilibrium between mitophagy and mitochondrial biogenesis, both governed by highly coordinated regulatory networks. The present study primarily focused on elucidating the role of the AHA1-ATP5A1 axis in modulating autophagic flux. Although multiple lines of evidence—including LC3B lipidation, p62 degradation, and ultrastructural analyses-consistently indicate that AHA1 knockdown alleviates excessive mitophagy, the precise in vivo quantification of autophagic flux remains technically challenging. Future investigations should therefore aim to provide a more comprehensive assessment of mitochondrial dynamics by incorporating biogenesis markers such as PGC-1α and employing advanced flux analysis methodologies. Despite these limitations, the pronounced neuroprotective effects observed following AHA1 suppression strongly support our conclusion that targeting this pathological mitophagy pathway represents a promising therapeutic strategy for ischemic brain injury.

In summary, we propose a novel mechanistic model where I/R stress, via STAT3, upregulates AHA1, which then interacts with ATP5A1 to trigger a detrimental cascade of AMPK activation and excessive mitophagy, ultimately leading to neuronal death. Our in vivo experiments confirm that interrupting this pathological process by silencing AHA1 alleviates reperfusion injury, offering a promising new therapeutic avenue for ischemic stroke.

5. Conclusion

This study reveals a novel, non-canonical function for AHA1 in ischemic stroke. We demonstrate that AHA1 is upregulated, translocates to mitochondria, and directly interacts with ATP5A1, promoting excessive and detrimental mitophagy by aberrantly activating the AMPK/mTOR pathway. Critically, silencing AHA1 in vivo confers potent neuroprotection, reducing brain damage and improving functional recovery. Our findings identify AHA1 as a key pathological driver and establish the AHA1-ATP5A1 axis as a promising therapeutic target to mitigate neuronal injury in stroke.

Author statement

All data were generated in-house, and no paper mill was used. All authors agree to be accountable for all aspects of their work and to ensure integrity and accuracy.

CRediT authorship contribution statement

Duo Jin: Investigation, Methodology, Software, Supervision. Xinran Ge: Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft. Li Liu: Methodology, Supervision, Validation. Yujie Jia: Methodology, Supervision, Validation. Chang Liu: Methodology, Supervision, Validation. Ke Meng: Methodology, Supervision, Validation. Xiaowei Zhang: Conceptualization, Funding acquisition, Project administration, Writing – review & editing.

Declaration of competing interest

All authors disclosed no relevant relationships.

Appendix A. Supplementary data

The following is the Supplementary data to this article: