MST1, a novel therapeutic target for Alzheimer's disease, regulates mitochondrial homeostasis by mediating mitochondrial DNA transcription and PI3K-Akt-ROS pathway.

doi:10.21203/rs.3.rs-4610030/v1

Download PDF

Research Article

MST1, a novel therapeutic target for Alzheimer's disease, regulates mitochondrial homeostasis by mediating mitochondrial DNA transcription and PI3K-Akt-ROS pathway.

https://doi.org/10.21203/rs.3.rs-4610030/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Alzheimer's disease (AD) stands as one of the prevalent irreversible neurodegenerative conditions marked by gradual cognitive deterioration and neuronal loss. The mammalian Ste20-like kinase (MST1)-Hippo pathway is pivotal in regulating cell apoptosis, immune response, mitochondrial function, and oxidative stress. However, the association between MST1 and mitochondrial function in AD remains unknown. Therefore, this study aims to investigate the effect of MST1 on neuronal damage and cognitive impairment by regulating mitochondrial homeostasis in AD.

Methods

In this study, we selected 4- and 7-month-old 5xFAD mice to simulate the early and middle stages of AD, we assessed cognitive function, detected neuronal damage indicators, and evaluated mitochondrial morphology, dynamics, oxidative stress, ATP, and mitochondrial apoptosis-related indicators. We employed RNA-seq technology to explore potential mechanisms of action. In vitro studies were conducted to investigate the effects of MST1 on the viability and mitochondrial function of SH-SY5Y model cells, aiming to validate the potential molecular mechanisms of MST1.

Results

Overexpression of MST1 accelerated neuronal degeneration and cognitive deficits, alongside promoting oxidative stress and mitochondrial damage. Similarly, in cell models, MST1 overexpression facilitated apoptosis and mitochondrial dysfunction. Knockdown of MST1 expression and chemical inactivation of MST1 improved cognitive decline, mitochondrial dysfunction while reducing neuronal degeneration. In terms of mechanism, MST1 was found to regulate the transcription of mitochondrial genes, including MT-Nd4L, MT-ATP6, and MT-CO2, by binding to PGC1α. It influenced cellular oxidative stress through the PI3K-Akt-ROS pathway, ultimately disrupting mitochondrial homeostasis and mediating cell damage.

Conclusion

Overall, our results showed that MST1 primarily regulates mitochondrial DNA transcription levels by interacting with PGC1α and modulates cellular oxidative stress through the PI3K-Akt-ROS pathway, thereby disrupting mitochondrial homeostasis. This discovery can potentially enhance mitochondrial energy metabolism pathways by targeting MST1, thereby offering novel potential therapeutic targets for treating AD.

MST1

Alzheimer’s disease

Mitochondrial homeostasis

oxidative phosphorylation (OXPHOS)

PI3K-Akt signaling pathway

Alzheimer's disease (AD) stands as the predominant neurodegenerative disease and prevalent form of dementia among the elderly population. With the aging demographic on a consistent rise globally [1], age-related neurodegenerative ailments emerge as a pressing public health concern. AD is characterized by the deposition of Aβ plaques and the formation of tau protein neurofibrillary tangles, internationally acknowledged as its key pathological features [2]. However, the precise pathogenesis of AD remains unclear, while certain methods exist to ameliorate symptoms. a definite cure remains elusive [3, 4]. Therefore, it remains imperative to investigate the pathological and physiological mechanisms underlying AD to identify novel therapeutic targets.

Mitochondria—crucial organelles for energy metabolism—play a vital role in meeting the high-energy demands of neurons [5]. In addition, mitochondria contribute over 90% of the energy required for synapses [5, 6]. Mitochondria stand as the sole organelles, apart from the nucleus, that possess their individual DNA (mtDNA) along with distinct transcription and translation mechanisms. mtDNA mainly encodes subunits that form the mitochondrial oxidative respiratory chain complex, which is important for mitochondrial energy synthesis and mitochondrial homeostasis. Mitochondrial homeostasis is maintained through dynamic regulatory processes, including mitochondrial dynamics (fusion and fission), quality control (biogenesis and mitophagy) and energy metabolism (oxidative phosphorylation) [7, 8]. Recent research indicates that disruptions in mitochondrial homeostasis occur in the initial stages of AD progression [9, 10]. Dysfunction of mitochondrial respiratory chain complex, breakdown in Metabolic Reprogramming, and a decreased mitochondrial content are deemed pivotal factors contributing to the onset of Alzheimer's disease. Moreover, mitochondrial dysfunction leads to electron leakage and a decrease in ATP production in the mitochondrial respiratory chain, consequently elevating oxidative stress and neuronal damage [11, 12]. However, mitochondrial oxidative stress can induce abnormal oxidative phosphorylation, triggering excessive reactive oxygen species (ROS) production within the mitochondria. This pivotal factor contributes to the dysregulation of mitochondrial homeostasis in AD [13]. In recent years, it has garnered significant attention that targeting mitochondria and brain bioenergetics metabolism could potentially pave a novel therapeutic pathway towards mitigating the progression of Alzheimer's disease.

Studies have reported activation of the Hippo signaling pathway in patients and animals with AD [14–16], but the precise mechanism remains unclear. The Hippo pathway represents a highly conserved kinase-signaling pathway [17]. Mammalian Ste20-like kinase 1 (MST1), also referred to as STK4, serves as an upstream serine/threonine kinase within Hippo signaling pathway. In the canonical Hippo pathway, MST1 triggers the activation of downstream LATS1/2 and YAP/TAZ kinases, consequently regulating organ size, tissue regeneration, and tumor suppression [17, 18]. In noncanonical Hippo signaling, MST1 has been reported to regulate various cellular functions such as apoptosis, immune responses, and oxidative stress [19]. However, the physiological functions and intracellular signaling mechanisms of MST1 remain unclear. In the nervous system, previous studies showed that MST-mediated FoxO3a-Bim respond to Aβ induced neuronal death [20]. Phosphorylated MST1 stimulates microglia activation, contributing to neuronal cell death in brain ischemic-reperfusion injury or stroke [21]. Recent studies highlight the involvement of MST1/2 in mitochondrial function. The MST1/2-BNIP3 axis plays a role in inducing mitochondrial autophagy and regulating neuronal viability during mitochondrial stress [22]. MST1 is involved in regulating mitochondrial dynamics [23–25]. However, the role of MST1 in AD progression through the modulation of mitochondrial homeostasis and the underlying regulatory mechanisms await experimental verification.

However, the role of MST1 in regulating mitochondrial homeostasis in AD remains unclear. Therefore, this study aims to investigate whether MST1 influences neuronal damage and cognitive impairment by regulating mitochondrial homeostasis in AD. Our findings reveal a correlation between MST1 activation and pathological progression of AD. MST1 overexpression exacerbates mitochondrial dysfunction, neuronal damage, and cognitive impairment. Conversely, downregulation of MST1 expression shows potential for symptom improvement. Regarding the mechanism, MST1 influenced transcriptomic features, particularly genes associated with the oxidative phosphorylation pathway. It modulates the transcription level of mitochondrial DNA. It influences the activity of respiratory chain complex enzymes through interaction with PGC1α. Furthermore, MST1 regulates mitochondrial oxidative stress through the PI3K-Akt-ROS pathway. This study elucidates a novel role of MST1 in the onset and progression of AD, laying a solid theoretical foundation for identifying potential therapeutic targets for the condition.

Animals and drug treatment

All animal procedures adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and received approval from the Ethics Committee of Shandong University Second Hospital (Approval No. KYLL2024362). Our aim was to minimize animal usage and suffering while ensuring experiment reproducibility. The experimental mice we used in this study were all male mice. 5xFAD and wild-type (WT) mice in the same C57BL/6J background were purchased from Beijing Hua fu kang Biotechnology Co., Ltd. 5xFAD transgenic mice contain five FAD mutations (APPSwF1Lon, PSEN1*M146L*L286V) under the Thy1 promoter. All mice were housed in a specific pathogen-free facility with a 12 h light/dark cycle, temperature maintained at 24 ± 2°C, and free access to food and water.

XMU-MP-1 (HY-100526, Med Chem Express), a reversible and selective MST1/2 inhibitor, was dissolved in DMSO to 5 mg/mL and administered to the mice through intraperitoneal injection at a daily dose of 1 mg/kg for 1 month [26].

Cell cultures and modeling

Human neuroblastoma cells (SH-SY5Y) were procured from the Shanghai Cell Bank of the Chinese Academy of Science (Shanghai, China) and cultured in Dulbecco's Modified Eagle Medium (DMEM, Gibco, California, USA) supplemented with 10% fetal bovine serum (FBS, Lonsera) and 1% penicillin-streptomycin (Solarbio, Beijing, China) under standard conditions of 5% CO2 and 37 ℃.

β-Amyloid (1–42) oligomer was obtained from Shanghai Qiangyao Biotechnology Co., Ltd. 1 mg of β-Amyloid (1–42) oligomer powder was dissolved in 100 µL of dimethyl sulfoxide (DMSO) and subsequently mixed with 140 µL of PBS to prepare the working solution. SH-SY5Y cells were treated with Aβ_1-42 (20 µM) for 24 h to establish an in vitro AD model. 740 Y-P and LY294002 (MedChemExpress, USA) were used as activator and inhibitor of the PI3K-Akt pathway, respectively.

Stereotactic AAV injection

AAV-GFP-MST1 and AAV-shRNA-MST1 were procured from the Shanghai GeneChem Corporation (Shanghai, China) and Hanbio Tech (Shanghai, China), respectively. The targeting sequence of AAV-shRNA-MST1 was 5′-GCCAGATTGTTGCAATCAAGC-3′. Stereotaxic injections were administered following established protocols [27]. Mice were anesthetized with pentobarbital sodium (50 mg/kg) and positioned in the stereotaxic apparatus. Small holes were drilled at each injection site using the following coordinates: anterior fontanelle as the zero point, anterioposterior = -2 mm, lateral = ± 1.5 mm, depth = 2.0 mm. Subsequently,1 µL of AAV was injected into the left and right hippocampi of the mice using a microinjector (0.2 µL/min). The needle remained in place for 5 min before and after injection to ensure viral distribution. After completing the injection, the scalps of the mice were sutured. A water maze test was conducted 1 month later.

Plasmids and transfection

MST1 overexpression plasmids were procured from Jiman Biotechnology Co., Ltd. (Shanghai, China), while MST1 knockdown plasmids were obtained from Tsingke Biotechnology Co., Ltd. (Beijing, China). The targeting sequence of siRNA-MST1 is 5′-CAGAAGUGAUUCAGGAAAUTT-3′. PGC1α overexpressing plasmids were purchased from Jipeng Biotechnology Co., Ltd. (Shandong, China). SH-SY5Y cells were seeded into 6-well plates. When the cells reached approximately 70–90% confluence, they were transfected with the respective plasmids using Lipofectamine™ 3000 Transfection Reagent (L3000-015, Invitrogen) according to the manufacturer’s instructions.

Morris water maze(MWM)

The Morris water maze (MWM) test comprises place navigation and spatial probe tests and assesses the spatial learning and memory abilities of mice [28]. Reference objects surrounding the pool remained consistent throughout the experiment, and the water temperature in the pool was maintained between 20 and 22 ℃. During the place navigation tests, a circular opaque platform was positioned 1 cm below the water surface at the center of the target quadrant. Mice were gently placed into the water, facing the wall, from four random quadrants for 5 days. They were given 60 s to locate the hidden platform. Time and swimming trajectories were recorded using an automatic camera system. If the mice failed to reach the hidden platform within 60 s, they were guided to it and allowed to remain there for 20 s. On the final day, the platform was removed for the spatial probe experiment. The number of crossings over the original hidden platform and time spent swimming in the target quadrant within 60 s were recorded.

Transmission electron microscopy (TEM)

The fresh mouse hippocampal tissue was cut into small pieces (1 mm × 1 mm × 3 mm) and promptly immersed in a 3% glutaraldehyde fixing solution. The TEM samples were prepared using a conventional method. All samples were rinsed, fixed with 1% osmium tetroxide, dehydrated, embedded in Epon, and sectioned into ultrathin slices (70–100 nm). Subsequently, they were stained with lead citrate and uranium acetate. Finally, the ultrastructure and morphology of the mitochondria were visualized using a transmission electron microscope (JEOL-1200E).

Immunofluorescence (IF)

After behavioral testing, mice underwent cardiac perfusion with precooled 4% paraformaldehyde (PFA), Subsequently, the brains were removed and immersed in 4% PFA at 4°C for 24 h. Brain tissue samples were dehydrated using a sucrose gradient and embedded in an optimal cutting temperature (OCT) compound (SAKURA, Japan). The brain tissues were sectioned into 20 µm slices using a freezing microtome (Leica, USA). Brain slices were fixed by 4% PFA for 20 min, washed with PBS thrice, and subjected to antigen retrieval with an antigen repair solution for 5 min at room temperature (20–25 ℃). Following incubation with 1% TritonX-100 and 5% goat serum, brain slices were incubated with primary antibodies overnight at 4°C. This included an anti-p-MST1 antibody (1:100, Bioss), anti-PSD95 antibody (1:200, Abclonal), and anti-SYP antibody (1:200, Abcam). The next day, slices were incubated with a secondary antibody (goat antirabbit IgG Alexa Fluor Plus 488, 1:500, Abcam) for 1 h at room temperature (20–25℃). DAPI-containing anti-fluorescence quencher (Sigma) was utilized for nuclear staining. Images were captured with a laser scanning confocal microscope (Zeiss, LSM 800, Germany) and multispectral imaging system (PerkinElmer, Mantra, America) and further analyzed using ImageJ.

Thioflavine S

The 30 µm brain slices were rinsed with PBS thrice and mounted onto glass slides. Brain sections were stained with 1% thioflavin solution (Solarbio) at room temperature (20–25 ℃) for 5 min. The slices were then rinsed with 50% alcohol twice for 5 min each. Images were captured using a multispectral imaging system (PerkinElmer, MA, USA).

Immunohistochemistry (IHC)

Immunohistochemical techniques were employed to assess the expression level of p-MST1 protein. Brain tissues were embedded in paraffin and sectioned into 5 µm slices. The sections underwent dewaxing, rehydration, antigen repair using EDTA antigen repair solution (G1203, Servicebio, Wuhan), and inactivation of endogenous peroxidase with 3% hydrogen peroxide solution. Sections were incubated with 3% bovine serum albumin (BSA) at room temperature (20–25 ℃) for 30 min. they were then incubated overnight at 4°C with anti-p-MST1 antibody (4635R, Bioss, 1:100). The next day, sections were incubated with HRP-labeled goat antirabbit IgG (GB23303, Servicebio, 1:200) at room temperature (20–25 ℃) for 50 min. Sections were stained with diaminobenzidine (DAB) and hematoxylin for 3 min. Images were visualized with a digital pathology scanner (NanoZoomer S60, Japan).

Hematoxylin eosin staining (HE)

After dewaxing and rehydration, the 5-µm paraffin sections underwent treatment with hematoxylin-eosin (HD) constant staining pretreatment solution for 1 min. They were then immersed in Hematoxylin solution for 3–5 min. Subsequently, the sections were sequentially treated with hematoxylin differentiation solution and hematoxylin blue solution. The samples were rinsed with tap water after each dyeing step. The sections were immersed in 95% ethanol for 1 min and in eosin dye for 15 s. They were then dehydrated in absolute alcohol and rendered transparent using xylene, and sealed with neutral gum. Images were captured using a digital pathology scanner (Nano Zoomer S60, Japan).

Nissl staining

Nissl staining was conducted to observe surviving neurons. Dewaxed and hydrated 5 µm tissue sections were treated with Nissl's stain for 2–5 min, followed by immersion in 1% glacial acetic acid for slight differentiation. The reaction was terminated by washing with running water. Subsequently, the slices were rendered transparent with xylene for 10 min and sealed with gum.

Western blotting

Total protein samples were extracted from the cultured SH-SY5Y cells or hippocampal tissues subjected to various treatments using cool radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime, Shanghai, China). Protein quantification was performed using a BCA kit (Solarbio). Protein samples were heated with 5 × loading buffer (Beyotime) at 95 ℃ for 10 min. The protein samples (30–60 µg) and protein ladder (26616, Thermo Fisher Scientific, USA) were loaded and electrophoresed on SDS-PAGE gel (10%, 12.5%, 15%). Proteins were transferred to polyvinylidene fluoride (PVDF) membranes (Merck Millipore, Darmstadt, Germany) for 1–2 h and blocked with 5% nonfat milk solution for 1.5 h. The membranes were incubated with primary antibodies overnight at 4 ℃ and then immersed in a secondary antibody solution for 1 h. Finally, protein expression was detected using enhanced chemiluminescence (ECL) western blotting (WB) detection reagents (Merck Millipore), and images were captured by a chemiluminescence imager (Tanon4800, Shanghai, China). Table S1 presents the information for all antibodies used.

Co-Immunoprecipitation (Co-IP)

SH-SY5Y cells were transfected with a FLAG-tagged MST1 plasmid. and resuspended in cell lysis buffer for WB and IP (Beyotime, Shanghai, China) supplemented with 1% protease inhibitors (Beyotime) and 1% phosphatase inhibitors (Beyotime). The samples were centrifuged at 13,000×g for 10 min at 4 ℃, and the supernatant containing protein was collected. Subsequently, the total protein concentration was determined using a BCA assay kit. One hundred microliters of the samples were allocated for use as the input group. The remaining samples were mixed with flag magnetic beads (HY-K0207, MedChemExpress, USA) and incubated overnight at 4℃ on a rotary mixer. The following day, the magnetic beads were retrieved using a magnetic rack and washed with TBST. Subsequently, the input group sample and magnetic beads were mixed in a loading buffer and heated at 95℃ for 10 min. Potential interactors of MST1 were then detected using SDS-PAGE.

Real-time quantitative polymerase chain reaction (RT-qPCR)

Total RNA from the cells and brain tissues was isolated using the RNA-Quick Purification Kit (ES Science Biotech, Shanghai) following the instructions of the manufacturer. One microgram of RNA was utilized to synthesize cDNA using an Evo M-MLV RT Kit (Accurate Biotechnology Co., Ltd.). SYBR qPCR Master Mix (Vazyme) and specific primers were employed to quantify the mRNA levels of the genes through RT-qPCR. Table S2 lists the primer sequences.

RNA-seq

Total RNA was extracted from the hippocampal tissue of mice using TRIzol™ reagent (Novogene) and utilized as input material for RNA Library preparation. To ensure library quality, RT-qPCR was conducted to accurately quantify the effective concentration of the library (> 1.5 nM). Different libraries were sequenced with Novogene using an Illumina NovaSeq 6000 Sequencing System. For differential gene analysis, padj (FDR-adjusted p-values) ≤ 0.05 and log2 (foldchange) ≥ 1.0 were set as the threshold for significant differential expression.

Enzyme linked immunosorbent assay (ELISA)

The Aβ₁₋₄₂ level mice's hippocampal tissue was assessed using an ELISA kit (Elabscience Biotechnology Co., Ltd.), following the instructions of the manufacturer for measurements and analyses.

Cell-counting-kit 8 (CCK-8) assay

SH-SY5Y cells were seeded in 96-well plates at the density of 8000 cells/well. Following incubation, CCK8 reagent (MedChemExpress) was added to the medium at a 1:100 ratio and incubated for 2–4 h to assess cell viability. The optical density at 450 nm was measured using a microplate reader (Biotek, Winooski, Vermont, USA).

Flow cytometry

SH-SY5Y cells were seeded into 6-well plates, following transfection, modeling, and treatment with the 740 Y-P drugs, the cells were incubated in a 10 µM DCFH-DA solution (Elabscience Biotechnology Co., Ltd.) for 60 min at 37 ℃ in the dark. Subsequently, the cells were collected and resuspended in a serum-free medium. The intensity of green fluorescence, indicative of intracellular ROS levels, was measured using flow cytometry.

An Annexin V-FITC/PI apoptosis detection kit (Vazyme) was utilized to assess cell apoptosis levels. Cells were harvested and washed twice with precooled PBS. Subsequently, they were gently resuspended in 1× Annexin V-FITC/PI binding buffer and incubated with annexin-V-FITC and PI at room temperature (20–25℃) for 10 min in the dark. Apoptotic cells were detected by flow cytometry (Beckman).

Oxidative stress assay

Mouse hippocampal tissue was homogenized in normal saline (mass ratio of 1:9) using a tissue homogenizer. The levels of oxidative stress indicators (total superoxide dismutase (T-SOD), glutathione peroxidase (GSH-PX), and malondialdehyde (MDA)) were measured with commercial reagent kits (Nanjing Jian cheng Bioengineering Institute) according to the instructions of the manufacturer.

Probes for mitochondrial detection

SH-SY5Y cells were cultured in confocal dishes and treated according to experimental requirements. To assess mitochondrial membrane potential (MMP), tetramethylrhodamine methyl ester (TMRM) staining solution (Invitrogen, USA) was introduced to the cell growth medium at a final concentration of 50 nM and incubated for 20 min at 37°C. To assess mitochondrial oxidative stress, MitoSOX™ Red reagent (Invitrogen, USA), a mitochondrial superoxide indicator, was added to cells at a final concentration of 5 µM and incubated for 20 min at 37°C in the dark. To evaluate mitochondrial morphology, the MitoTracker probe (Invitrogen, USA) was introduced to the cell medium at a final concentration of 100 nM and incubated for 30 min at 37°C in the dark. Hoechst 33342 nuclear staining dye and DAPI were utilized for nuclei staining. The fluorescence intensity of the cells was observed using laser scanning confocal microscope (LSM 800, Zeiss, Germany).

ATP Levels Measurement

ATP levels in cells or tissues were assessed using an Enhanced ATP Assay Kit (S0027, Biotechnology, Shanghai, China), following the instructions of the manufacturer. Cell or tissue samples were lysed in ATP detection lysate and centrifuged to obtain the supernatant. The cell or tissue supernatants were promptly added to the ATP detection working solution. The relative light unit (RLU) value of the sample was immediately measured using a cell Imaging Multi-Mode reader (Biotic, Cytation5, USA).

Measurement of mitochondrial respiratory capacity

The oxygen consumption rate (OCR) representing the mitochondrial respiratory capacity of SH-SY5Y cells was assessed using the XF Cell Mito Stress Test Kit on a Seahorse XFe24 Analyzer (Seahorse Bioscience/Agilent Technologies, North Billerica, MA), following a standard protocol. SH-SY5Y cells were seeded in Seahorse Xe24 plates (80,000 cells/per well). The mitochondrial stress test experiment measured the OCR of SH-SY5Y cells by sequentially introducing targeted drugs to the mitochondrial electron transport chain (ETC), including 1.5 µM oligomycin (ATP synthase inhibitor), 2 µM phenylhydrazone (FCCP; mitochondrial respiration uncoupler), and 0.5 µM rotenone/antimycin A (complete respiratory inhibitor). Key parameters reflecting mitochondrial function were obtained, including basal respiration, ATP-linked respiration, proton leak, and maximum respiration. Subsequently, cells were lysed in 1 × RIPA buffer and total protein quantification was conducted on each well sample. The data were then normalized.

Mitochondria isolation and complex activity

Mitochondria were isolated from cells using a Cell Mitochondria Separation and Extraction Kit (KTP4003, Abbkine, Wuhan, China). Initially, cells were harvested and washed by suspending the cell pellet with ice-cold PBS, followed by centrifugation at 500 × g for 3 min, after which the PBS was discarded. The cell pellets were resuspended in cold lysis buffers A, B, and C, and the mixture was centrifuged at 600 × g for 10 min. Finally, the supernatant was collected and centrifuged at 11000g for 10 min at 4℃. The resulting sediment contained the isolated mitochondria.

According to the instructions of the manufacturer, the activity of mitochondrial complexes I-V was measured using the mitochondrial respiratory chain complex activity detection kit (Abbkine, Wuhan, China).

Statistical analysis

Statistical analyses were conducted using GraphPad Prism 8.0 software, and results were presented as mean ± standard error of the mean (SEM). Statistical differences between two groups were performed using unpaired two-tailed Student’s t-test. One-way analysis of variance (ANOVA) was used to compare multiple exposure groups, followed by Tukey’s multiple comparison test. Data from the MWM test was analyzed using two-way ANOVA with the Bonferroni multiple comparison test. RNA-seq data were analyzed by R (4.1.1). All experiments were repeated at least three times. Statistical significance was defined at p < 0.05.

MST1 activation is associated with disease progression in 5xFAD mice

Labeling the content and distribution of amyloids in the brain parenchyma is a crucial indicator for evaluating the pathological status of AD. In 5xFAD mice, the presence of Aβ₁₋₄₂ emerges at an early age and progressively increases thereafter, facilitating the early formation of amyloid deposition plaques [29]. Therefore, we assessed amyloid plaque deposition in the brains of 5xFAD mice at 1-, 3-, 6-, and 9-months-old using thioflavin staining. Mature amyloid plaques were present in the hippocampus and cortex beginning form 3 months, with deposition gradually intensifying as the mice aged (Fig. 1A, B).

To explore the activation of MST1 in 5xFAD, we assessed its activated form p-MST1 (Thr183). Our WB results showed a gradual increase in the ratio of p-MST1 to MST1 as the age of 5xFAD mice increased, aligning with the observed changes in hippocampal neuronal apoptotic protein Bax and amyloid plaque deposition (Fig. 1C, D). However, these results did not differ significantly in age-matched WT mice (Fig. 1C, D). Therefore, we suggest that the activation of MST1 to p-MST1 in 5xFAD mice correlates with the pathological advancement of the disease. To further validate p-MST1 activity and distribution, we conducted IHC assays, revealing a significant increase in p-MST1 levels within the hippocampus (Fig. 1F, G) and cortex (Fig. 1H, I) of 6-month-old 5xFAD mice compared with the WT group, consistent with the WB results.

MST1 promotes cognitive deficits and neuronal damage in 5-month 5xFAD mice

To assess the effect of MST1 on cognitive and memory impairment in AD mice, we randomly partitioned 4-month-old C57 mice and 5xFAD mice into four groups. We administered AAV-GFP vehicle and the AAV-GFP MST1 into the DG region of mice through hippocampal stereotaxic injection to upregulate MST1 (Fig. 2A). Four weeks post-injection, spontaneous GFP fluorescence from the virus was observed in the hippocampus, indicating successful injection (Fig. 2B). We also evaluated the expression efficiency of MST1 through WB and RT-qPCR. The results showed significantly higher levels of MST1 expression at the protein and mRNA levels than those in the control group (Fig. 2C-E).

Four weeks post-injection, the MWM test was performed to assess the spatial learning and memory abilities of the mice. During the training period, the mice did not exhibit significant differences in swimming speed (Fig. 2F). Compared with the C57 + AAV-vehicle group, the 5xFAD + AAV-vehicle group exhibited a significantly prolonged escape latency (time to find the hidden platform) starting from the third day of training (Fig. 2G, J). Moreover, the escape latency of the 5xFAD + AAV-MST1 group during the training was significantly longer than that of the 5xFAD + AAV-vehicle group (Fig. 2G, J). The number of platform crossings and total time spent in the target quadrant were reduced considerably in the 5xFAD + AAV-MST1 group than in the 5xFAD + AAV-vehicle group (Fig. 2H, I). These measures were also significantly reduced in the 5xFAD + AAV-vehicle group compared with the C57 + AAV-vehicle group (Fig. 2H, I). These MWM results suggested that MST1 overexpression exacerbated spatial learning and memory impairments in 5xFAD mice.

Due to the pivotal role of neuronal loss in cognitive decline, we examined the effect of MST1 expression on neurons in the CA1, CA3 and DG regions of the hippocampus using Nissl and HE staining. Results from Nissl staining showed a rise in the count of damaged neurons in the 5xFAD + AAV vehicle group, with MST1 overexpression further exacerbating neuronal damage (Fig. 3A, B). HE staining revealed a compact arrangement and normal morphology of neuronal cells in the normal group. In contrast, the 5xFAD + AAV vehicle group exhibited sparse neuron arrangement with some pyknotic nuclei and deepening chromatin. MST1 overexpression further aggravated this increase in abnormal neuron abundance (Fig. 3C). Additionally, we assessed the expression of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase 9, Cleaved Caspase 3, and Cyt-C) through immunoblotting. The Bax /Bcl-2 ratio and expression levels of Cleaved Caspase 9, Cleaved Caspase 3, and Cyt-C in the 5xFAD + AAV vehicle group were significantly higher than in the C57 + AAV vehicle group. In the 5xFAD + AAV-MST1 group, the expression levels of apoptosis-related proteins were higher than in the 5xFAD + AAV vehicle group (Fig. 3D, E). These data support that MST1 promoted neuronal apoptosis. Finally, WB revealed reduced expression levels of the synaptic marker proteins PSD95 and SYP in the 5xFAD + AAV-MST1 group compared with the 5xFAD group (Figure S1A, B). IF of PSD95 and SYP validation yielded similar results (Fig. 3F-H), suggesting that MST1 overexpression also impaired the synaptic structure of neurons. ELISA results further revealed that Aβ deposition did not differ significantly between the 5xFAD + ad-MST1 and 5xFAD + ad-vehicle groups, implying that overexpression of MST1 did not significantly increase Aβ₁₋₄₂ content (Fig. 3I). In conclusion, this part of the study suggested that MST1 promoted neuronal apoptosis and exacerbated the pathological process in 5xFAD mice, but not by affecting the levels of Aβ.

MST1 exacerbates mitochondrial dysfunction and oxidative stress levels in 5-month-old 5xFAD mice

TEM was employed to monitor mitochondrial ultrastructural changes in mouse hippocampal neurons. The mitochondrial perimeter is indicative of morphological alterations [30]. The mitochondrial morphology (round or oval) appeared normal, exhibiting clear and intact outer membranes and cristae in the C57 + AAV vehicle group. We also observed that mitochondrial perimeter was relatively longer in the C57 + AAV vehicle group. Although the mitochondria showed abnormal morphology and smaller perimeter in the C57 + AAV-MST1 group, there was no statistically significant difference when compared to the C57 + AAV vehicle group (Fig. 4A, B). Meanwhile, in the 5xFAD + AAV-vehicle and 5xFAD + AAV-MST1 groups, certain mitochondria exhibited shrinkage, swelling, and incomplete cristae, accompanied by decreased matrix density and shortened perimeters (Fig. 4A, B). The mitochondrial morphology in the 5xFAD + AAV-MST1 group deteriorated further than in the 5xFAD + AAV-vehicle group (Fig. 4A, B). The WB results showed significant variations in mitochondrial dynamics-related proteins (OPA1, MFN2, Drp1, and Fis1) and mitochondrial biogenic proteins (PGC1α and Nrf1) between the C57 + AAV-vehicle and 5xFAD + AAV-vehicle groups, as well as between the 5xFAD + AAV-vehicle and 5xFAD + AAV-MST1 groups (Fig. 4C, D). These differences suggest that overexpression of MST1 in 5xFAD mice stimulates mitochondrial division while inhibiting mitochondrial fusion and biogenesis.

Our findings suggest that the level of mitochondrial oxidative stress was increased in the 5xFAD + AAV-MST1 group compared to the 5xFAD + AAV-vehicle group, as evidenced by MitoSOX red fluorescence staining (Fig. 4E, F). Levels of ATP were reduced in the 5xFAD + AAV-vehicle group compared to the C57 + AAV-vehicle. Importantly, the reduction in ATP levels was more pronounced in 5xFAD mice overexpressing MST1 (Fig. 4G). Our data support that overexpression of MST1 interferes with energy metabolic processes in 5xFAD mice. The activities of SOD, GSH, and MDA serve as indicators of cellular oxidative stress levels. In the 5xFAD + AAV-vehicle group, the levels of SOD and GSH decreased (Fig. 4H, I), while MDA increased (Fig. 4J) than those in the control group, compared with the control group. Meanwhile, the 5xFAD + AAV-MST1 group exhibited significantly reduced SOD and GSH levels, along with a significant increase in MDA activity, compared those of the 5xFAD + AAV-vehicle group (Fig. 3H-J). Hence, MST1 promotes oxidative stress in 5xFAD mice.

Taken together, our findings imply that overexpression of MST1 in the hippocampus of 5xFAD mice results in impaired mitochondrial morphology and function, leading to oxidative stress and imbalance in energy metabolism, which leads to hippocampal damage and exacerbation of cognitive deficits in mice.

Down-regulation of MST1 alleviates cognitive impairment and mitochondrial dysfunction in 8-month-old mice

To verify the effect of MST1 downregulation on cognition, mitochondrial function, and neurons in AD mice, we administered AAVs into the DG region of 7-month-old C57 and 5xFAD mice (Fig. 5A). Four weeks later, we observed a significant decrease in MST1 protein and mRNA expression levels compared with the control group (Fig. 4B, C, D). In addition, the abundance of p-MST1 protein was significantly lower than that observed in the control group (Fig. 4B, C).

Subsequently, the MWM was employed to assess the spatial learning and memory of each mouse group. the mice did not exhibit significant differences in swimming speed During the training period (Fig. 4E). The 5xFAD + AAV-vehicle group exhibited a significantly prolonged escape latency (time to find the hidden platform) starting from the third day of training compared with the C57 + AAV-vehicle group (Fig. 4F, I). Moreover, the escape latency was significantly shorter in the 5xFAD + AAV-shMST1 group during the training compared to the 5xFAD + AAV-vehicle group (Fig. 4F, I). The number of platform crossings and total time spent in the target quadrant were reduced considerably in the 5xFAD + AAV-vehicle than in the C57 + AAV-vehicle group (Fig. 2H, I). These measures were significantly increased in the 5xFAD + AAV-shMST1 compared with the 5xFAD + AAV-vehicle group (Fig. 2H, I). The MWM results showed that MST1 knockdown improved the ability to locate the hidden platforms and enhanced the cognitive function of 8-month-old 5xFAD mice. To observe the effect of knocking down MST1 on neuronal synapses, we performed immunofluorescence staining of synaptic-related markers (PSD95 and SYP) in the DG region of the hippocampus. Our data showed a significant reduction of synapse-related proteins in 8-month-old 5xFAD mice, but knockdown of MST1 significantly increased these proteins in 5xFAD mice (Figure A-C). The above results indicate that 8-month-old 5xFAD mice present significant synaptic damage, but MST1 Knockdown could prevent synaptic impairments.

Compared with the 5xFAD + AAV-vehicle group, The 5xFAD + AAV-shMST1 group exhibited significantly elevated expression of mitochondrial fusion proteins (OPA1 and MFN2) and mitochondrial biogenic proteins (PGC1α and Nrf1), along with a significant decrease in the expression of mitochondrial fission proteins (Drp1 and Fis1) (Fig. 4H and Figure S3G). MitoSOX red staining results showed a reduction in mitochondrial ROS levels following MST1 downregulation (Fig. 4I and Figure S3G).

Transcriptomic analysis of the regulatory effects of MST1 in 8-month-old 5xFAD mice

To elucidate the molecular mechanism underlying the influence of MST1 on AD, we employed RNA-seq technology to conduct transcriptome sequencing on hippocampal tissues obtained from mice in the 5xFAD + AAV-vehicle and 5xFAD + AAV-shMST1 group. The volcano plot displayed the upregulated and downregulated genes within the co-expressed DEGs (Figure S3A). Overall, 359 differently expressed genes (DEGs, p ≤ 0.01) were assessed, with 162 downregulated (log2FC ≤ − 0.5) and 197 upregulated (log2FC ≥ 1.0). The Venn diagram illustrates the gene counts detected in each group while overlapping areas indicate co-expressed genes between the two groups (Fig. 6A). The heatmap presents clustering analysis results of the DEGs, revealing the expression of an identical gene across various samples, and confirming the consistency of biological replicates (Fig. 6B). To elucidate the functional roles of the identified DEGs, we conducted Gene Ontology (GO) enrichment analyses and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway. GO enrichment of DEGs was primarily focused on cellular metabolism (NAD(P) + nucleosidase activity, NAD + nucleotidase, cyclic ADP-ribose generation, NAD + nucleosidase activity) and immune inflammation regulation (positive regulation of immune effector processes, negative regulation of immune system processes, interleukin-6 production, and regulation of immune effector processes) (Figure S3B). KEGG analysis revealed a significant enrichment of unigenes encoding enzymes in the oxidative phosphorylation pathway. We also found that the PI3K-Akt signaling pathway being the most significantly enriched (Fig. 6C). In addition, enrichment was observed in antioxidant resistance and chemical carcinogenic ROS pathways. Representative GSEA results aligned with the KEGG pathway analysis findings (Fig. 6D, E). We observed significant upregulation of mitochondrial respiratory chain-related genes following MST1 knockdown, including NADH dehydrogenase subunit 4 (MT-ND4L), ATP synthase F0 subunit 6 (MT-ATP6), cytochrome b oxidase (MT-CO2), ATP synthase F0 subunit 8 (MT-ATP8), Gm28439, and BC002163 (Figure S3C).

MST1 effects on SH-SY5Y cell model induced by β-Amyloid in vitro

To investigate the effects and potential mechanisms of MST1 on cell models, we selected SH-SY5Y cells induced by Aβ₁₋₄₂ as the AD in vitro model. Initially, we exposed SH-SY5Y cells to various concentrations of Aβ₁₋₄₂ (0, 5, 10, 20, 40 µM) for 24 h. Subsequent WB results revealed the level of MST1 activation (p-MST1) and the ratios of phosphorylated MST1 to total MST1 (p-MST1/MST1) were gradual increased, as the concentration of Aβ₁₋₄₂ increased. Statistical significance was observed when induced with 20 µM Aβ₁₋₄₂ (Fig. 7A, B). Therefore, we selected 20 µM Aβ₁₋₄₂ for subsequent experiments to establish the cell model. We assessed the activation of MST1 in the model and control groups using IF, and we found that Aβ caused a significant increase in the fluorescence intensity of p-MST1, consistent with the WB results (Fig. 7C, D).

We constructed an MST1 overexpression plasmid and MST1-specific small interfering RNA (siRNA), along with their respective vehicles, for cell transfection. WB and RT-qPCR results showed successful upregulation (Figure S4A-C) or downregulation (Figure S4D-F) of MST1, respectively. Next, we assessed cell viability using the CCK8 assay and observed a decrease in SH-SY5Y cell viability upon treatment with Aβ₁₋₄₂. Moreover, overexpression of MST1 further reduced cell viability (Fig. 7E). However, loss of MST1 expression promoted cell survival (Fig. 7F). Additionally, the abundance of mitochondria-dependent apoptosis-related proteins (Bax, Cleaved Caspase 3, and Cyt-C) was increased in the AD cell model, with a more pronounced increase following transfection with the MST1 overexpression plasmid (Figure S4I, J). However, transfection with siRNA reversed this effect (Figure S4G, H). The trend in the abundance of the antiapoptotic protein Bcl-2 was opposite to that of the apoptotic proteins in each group (Figure S4G-J). To further quantify the effect of MST1 on apoptosis, we conducted a flow cytometry analysis. Revealing increased apoptosis in the Aβ and ad-MST1 + Aβ groups compared to the control group. The apoptosis rate of the ad-MST1 + Aβ group was significantly higher than in the Aβ group. MST1 knockdown significantly decreased the apoptosis rate (Fig. 6F). The above results indicated that Aβ-treated cells led to the activation of MST1, which further promoted SH-SY5Y cell apoptosis.

We investigated the effects of MST1 activation on mitochondrial morphology and function in SH-SY5Y cells. From MitoTracker staining data, we observed more mitochondrial fragmentation in the Aβ and ad-MST1 + Aβ group. The TMRM and Mitosox Red staining results showed that mitochondrial membrane potential was damaged and mitochondrial ROS were elevated in the ad-MST1 + Aβ group compared to the control and Aβ group, whereas MST1 knockdown reversed these manifestations. These findings support the potential of MST1 to regulate mitochondrial functions.

MST1 regulates the transcription of mitochondrial genes and affects mitochondrial oxidative phosphorylation by binding PGC1α

In our RNA-seq analysis, we performed KEGG enrichment analysis and identified enrichment of the oxidative phosphorylation pathway. Among all the DEGs, the three genes with significant differences were MT-ND4L, MT-ATP6, and MT-CO2, all associated with the oxidative phosphorylation pathway. To verify whether MST1 regulates mitochondrial gene transcription, we validated the expression of the identified candidate genes via RT-qPCR. The results showed a significant reduction in the mRNA expression of MT-ND4L, MT-ATP6, and MT-CO2 in the ad-MST1 + Aβ group and a significant elevation in the si-MST1 + Aβ group, compared with the Aβ group (Fig. 8A, B). Additionally, the protein expression levels of MT-ND4L, MT-ATP6 and MT-CO2 were significantly decreased within the Aβ and si-Ctrl + Aβ groups, whereas knocking down MST1 effectively reversed this decline (Figure S5A, B). Furthermore, upon overexpression of MST1, the abundance of these proteins significantly decreased ((Figure S5C, D). These results suggest that MST1 modulates mitochondrial DNA transcription and the expression of ECT proteins in an AD cell model. Here, the subcellular localization of the activated form of p-MST1 at baseline was primarily concentrated in the cytoplasm and nucleus. However, in the AD cell model, the co-localization of p-MST1 and MitoTracker increased in the Aβ group (Fig. 8C, D). This suggests that Aβ treatment of SH-SY5Y cells promotes the activation of MST1 to the p-MST1 form and causes p-MST1 to accumulate more on mitochondria.

To further explore the downstream molecular mechanism of MST1, we performed Co-Immunoprecipitation (Co-IP) experiments. Interestingly, Co-IP results showed that PGC1α could potentially bind to MST1 (Fig. 8E). PGC1α is known to be a key transcriptional co-activator for inducing gene expression under physiological and pathological stress conditions and can be involved in the expression of relevant target genes. More importantly, one of the main functions of PGC1a is to activate mitochondrial biosynthesis and oxidative phosphorylation. In view of this we proposed the hypothesis that the effect of MST1 on mitochondrial oxidative phosphorylation-related genes is related to that of PGC1a. Firstly, we performed WB experiments, which showed that the reduction of MT-ND4L, MT-ATP6, and MT-CO2 proteins after overexpression of MST1 was reversed to some extent after overexpression of PGC1a (Fig. 8F, G). Furthermore, our analysis of cellular OCR revealed that MST1 overexpression reduced the maximum respiratory capacity and ATP production in cells while also increasing proton leakage within the ETC. However, overexpression of PGC1α reversed the OCR impairment caused by MST1 (Fig. 8H-J). Besides, we assessed the activity of mitochondrial respiratory chain complexes I-V. Overexpression of MST1 worsened the decline in complex enzyme activity induced by Aβ; PGC1α overexpression mitigated this damage caused by MST1 (Fig. 8K). Taking together, our data support the idea that PGC1α is an important downstream molecule of MST1 affecting mitochondrial oxidative phosphorylation in Aβ-induced SH-SY5Y cells.

MST1 regulation of oxidative stress through PI3K-Akt signaling in SY5Y cell

Analysis of the RNA-seq results revealed enrichment of the PI3K-Akt pathway, suggesting that MST1 may be involved in AD regulation through this signalling pathway. To investigate the potential mechanism, we initially examined changes in the expression of PI3K, Akt, and p-Akt proteins upon upregulation or downregulation of MST1. WB revealed that p-Akt protein was suppressed in the Aβ group and further decreased in the group overexpressing MST1 (Fig. 9A, B). However, the inhibitory effect observed in the Aβ group was reversed when MST1 was knocked down (Fig. 9C, D). Total PI3K and Akt expression remained unchanged in each group. To further evaluate the influence of the PI3K-Akt pathway on oxidative stress and mitochondrial function, we treated MST1-overexpressing cells with 740Y-P (25 µM) for 24 h to activate the PI3K-Akt pathway. Our findings suggest that the expression of mitochondrial apoptosis proteins (Bax, Cleaved Caspase-3, and Cyt-C), initially induced by MST1 overexpression, was significantly reduced by 740Y-P. Hence, MST1-induced apoptosis was reduced by 740Y-P (Fig. 9A, B). Treatment of MST1-overexpressing cells with 740Y-P significantly suppressed ROS levels (Fig. 9E, F), facilitating the restoration of MMP (Fig. 9I, J).

Subsequently, MST1-knockdown cells were treated with LY294002 (20 µM), a PI3K-Akt pathway inhibitor. Our findings revealed increased expression of mitochondrial apoptosis-related proteins namely Bax, Caspase-3, and Cyt-C. In contrast, the antiapoptotic protein Bcl-2 was decreased in the si-MST1 + Aβ + LY294002 group compared with the si-MST1 + Aβ group (Fig. 9C, D). This indicates that the administration of LY294002 counteracted the protective effect of MST1 knockdown. We also observed that MST1 knockdown significantly alleviated oxidative stress and improved MMP compared to the Aβ group, while LY294002 treatment exacerbated oxidative stress (Fig. 9G, H) and reduced MMP (Fig. 9K, L). These findings suggest that LY294002 reverses the inhibitory effects on oxidative stress and the enhancement of mitochondrial function induced by MST1 knockdown. As expected, MST1 might serve as a major regulator of the PI3K-Akt-ROS signaling pathway in Aβ-induced SH-SY5Y cells.

XMU-MP-1 relieves AD symptoms by inhibiting MST1 activity

XMU-MP-1, a novel Hippo kinase inhibitor, can effectively suppress MST1 expression [26]. 7-month-old WT and 5xFAD mice were intraperitoneally injected with XMU-MP-1 or DMSO for 1 month (Fig. 10A). As shown in the MWM results, the spatial cognitive and memory abilities of the 8-month-old 5xFAD mice were significantly impaired. However, following 1 month of treatment with XMU-MP-1, the 5xFAD mice exhibited shortened search times for hidden platforms, prolonged time spent in the target quadrant, and increased platform crossings (Fig. 10B-E). These results show that XMU-MP-1 rescued cognitive function impairments in 8-month-old 5xFAD mice.

Following 1 month of treatment, we evaluated certain proteins using WB. The expression of the p-MST1 protein was significantly suppressed, and p-MST1/MST1 levels decreased in the 5xFAD + XMU-MP-1 group (Fig. 10F-G). Moreover, synaptic-related proteins (PSD95 and SYP) were increased in the 5xFAD + XMU-MP-1 group compared to the 5xFAD + DMSO group (Fig. 10H-I). Levels of apoptosis-related proteins (Bax, Cleaved Caspase 9, Cleaved Caspase 3, and Cyt-C) decreased in the 5xFAD + XMU-MP-1 group (Fig. 10J-K). In contrast, the expression of the antiapoptotic protein Bcl-2 and mitochondrial biogenesis-related proteins (PGC1α and Nrf1) increased in the 5xFAD + XMU-MP-1 group (Fig. 10J-M). These results show that inhibiting MST1 expression through chemical methods yields effects consistent with gene knockout, suggesting that blocking MST1 activation improves cognitive, improves mitochondrial function, and reduces the incidence of mitochondrial apoptosis of 5xFAD mice.

Finally, the mechanism was confirmed. We observed elevated expression of mitochondrial respiratory chain-related proteins (MT-ND4L, MT-ATP6, and MT-CO2) and activation of the PI3K-Akt pathway when MST1 activation was inhibited by XMU-MP-1 (Fig. 10N-Q). This suggests that XMU-MP-1 affects mitochondrial genes expression and regulates the PI3K-Akt pathway in AD mice by inhibiting MST1.

Early AD is characterized by widespread mitochondrial dysfunction and oxidative damage [31, 32]. The mitochondrial genome is crucial to maintaining homeostasis [31, 33, 34]. Previous studies have reported proteomic analyses of various brain regions in patients with AD after death, revealing widespread dysregulation of the Hippo pathway in severely affected regions of the AD brain [15]. MST1, as the core kinase of the Hippo pathway, exhibits abnormal expression in the peripheral blood of patients with AD [35]. While recent studies have shown that MST1 mediates neuronal loss and cognitive impairment in AD mice [14, 36]. In this study, we discovered that MST1 modulates mitochondrial DNA transcription (i.e., mitochondrial oxidative phosphorylation-related genes) and ROS levels to maintain mitochondrial homeostasis. This regulatory process significantly influences cognitive impairment and neuronal damage in AD mice, suggesting that MST1 could serve as a novel therapeutic target for AD.

Previous studies have highlighted that 5xFAD mice exhibit amyloid plaque deposition at an unusually early stage [37, 38]. Here, we also observed a significant quantity of amyloid plaques in mice as young as 3 months old, with a steady accumulation as the mice aged. Moreover, a significant association was observed between MST1 activation and the pathological advancement of AD. The expression levels of p-MST1 in the hippocampus and cortex of 6-month-old 5XFAD mice were markedly higher than in WT mice. This underscores the activation of MST1 during amyloid-mediated neurodegeneration [14, 36]. Although most of the current studies have focused on interventions on Aβ, unfortunately, most of these therapies have not been successful. Our study found that MST1 activation similarly led to an AD-like phenotype without increasing the burden of Aβ. As previously studied, Aβ may not lead to neurological damage immediately, however, it is more closely associated with cognitive decline than Aβ accumulation when MST1 activation occurs [36].

Information transmission between neurons occurs through synapses, which are crucial for maintaining normal physiological functions. Proteins such as PSD95 and SYP are pivotal in neuronal synapses, contributing to synapse formation and plasticity regulation. Meanwhile synaptic destruction serves as an early indicator of neuronal degeneration, with progressive neuronal and synaptic loss reflecting AD progression [39, 40]. According, we overexpressed the MST1 protein in 4-month-old mice (early AD stage) and subsequently, knocked it down MST1 in 7-month-old mice (middle stage AD). Upregulation of MST1 exacerbated the AD phenotype and worsened neurodegeneration. This was evidenced by decreased spatial learning and memory ability, increased abundance of damaged and necrotic neurons, and impaired neuron synaptic plasticity. However, MST1 knockdown improved cognitive and neuronal functions and increased expression of synaptic proteins PSD95 and SYP, suggesting a reduction in synaptic loss. Similar results were observed following MST1 inhibition by XMU-MP-1 in 7-month-old mice. Overall, our findings confirm the involvement of MST1 in regulating cognitive, neuronal, and synaptic plasticity in AD.

Neurons, as high energy-demanding cells, depend on mitochondria as their energy factories. Within neurons are distinct structures, including cell bodies, dendrites, axons, and synapses, each with unique functions and specific requirements for mitochondrial storage pools [41, 42]. To sustain normal neuronal function, mitochondria must continually move, fuse, divide, and replicate, ensuring a dynamic balance. However, disruption in mitochondrial homeostasis can lead to neuronal damage [43]. While research on the relationship between MST1 and mitochondria in neurodegenerative diseases is relatively limited, studies have examined how MST1 influences mitochondrial division, fusion, and autophagy processes through mitochondrial dynamic proteins [44, 45]. MST1 can reportedly induce alterations in mitochondrial ultrastructure, dysfunction, and ATP reduction in dilated cardiomyopathy [24], aligning with our findings in AD. Additionally, MST1 induces mitochondrial dysfunction by regulating mitochondrial ROS and oxidative stress [46, 47]. In our study, we employed MitoSOX staining to assess mitochondrial oxidative stress levels and assessed SOD, GSH, and MDA levels as indicators of intracellular oxidative stress. MST1 was found to modulate mitochondria and exacerbate cellular oxidative stress, consistent with previous findings.

Assessing the influence of gene expression on signaling pathways provides insights into the significance of phenotypes, as the upregulation or downregulation of a single gene may have limited effects on complex phenotypes. In our study, we analyzed differential genes using two datasets (5xFAD + AAV vehicle vs. 5xFAD + AAV shMST1), including GO, KEGG, and GSEA. Many DEGS were associated with cellular functions involving cellular energy metabolism and regulating immune inflammation. Similarly, previous studies have highlighted the influence of MST1 on cellular mitochondrial metabolism and oxidative stress[47], and its involvement in mammalian inflammation and regulation of innate and adaptive immunity [48, 49]. Among the top DEGs were those belonging to the mitochondrial oxidative phosphorylation pathway, including MT-ND4L, MT-ATP6, and MT-CO2. The mitochondrial respiratory chain complex I, also known as nicotinamide adenine dinucleotide (NADH) dehydrogenase, serves as the entry point for electrons into the respiratory process. It is the largest protein complex, comprising 40 subunits. Among these, MT-ND4L is one of the seven subunits encoded by mtDNA. Previous studies have linked MT-ND4L mutations and AD through whole-exome sequencing [50]. Studies have also shown a significant deficiency in complex I and IV enzymes in AD [51]. MT-ATP6 is a subunit of mitochondrial complex V, contributing to mitochondrial ATP synthesis alongside proton transport. MT-CO2, a subunit of mitochondrial cytochrome c oxidase (Complex IV), plays a crucial role, and its impairment can result in mitochondrial respiratory chain dysfunction.

To further explore the role of MST1 on AD, we employed the SH-SY5Y cell model for in vitro assessment. Initially, we observed a concurrent activation of MST1 with the escalating concentration of Aβ. This finding aligns with the results observed in animal models and further confirms the conclusions from previous studies [36]. In addition, MST1 influenced cell viability, facilitated apoptosis, and affected mitochondrial morphology and function. This is consistent with the in vivo results.

In our study, pathway analysis revealed the influence of MST1 on various genes related to mitochondrial energy metabolism, particularly the mitochondrial respiratory chain-related genes MT-ND4L, MT-CO2, and MT-ATP6. We validated the enrichment results through RT-qPCR and WB, yielding highly consistent results. Mitochondria contain mtDNA, which encodes 13 subunits of the oxidative phosphorylation (OXPHOS) system [52, 53]. Mitochondria primarily generate ATP through oxidative phosphorylation, highlighting the significance of regulating oxidative phosphorylation subunits for mitochondrial homeostasis. To further explore the downstream molecular mechanism of MST1, we performed Co-Immunoprecipitation (Co-IP) experiments. Interestingly, Co-IP results showed that PGC1α could potentially bind to MST1. Peroxisome-proliferator-activated γ co-activator 1α (PGC1α), is the primary regulatory factor for mitochondrial biogenesis [53]. Previous studies have reported that PGC1α overexpression in myotubes enhances mtDNA levels and modulates its transcription [54]. PGC1α can stimulate mitochondrial respiration by inducing the expression of OXPHOS subunits [55]. To investigate whether the influence of MST1 on mtDNA transcription and related-protein expression is related to downstream PGC1a. So, we performed WB experiments and verified that PGC1a affects MST1 on downstream MT-ND4L, MT-CO2, and MT-ATP6 protein expression. In addition, oxidative phosphorylation subunits comprise mitochondrial respiratory chain complexes I–V, crucial for maintaining mitochondrial respiration. In our study, we assessed the activity of the mitochondrial respiratory chain complexes I–V and observed mitochondrial respiration through a mitochondrial stress test. MST1 reduced the activity of the five mitochondrial respiratory chain complexes, increased proton leakage, reduced maximum respiratory capacity, and impair ATP production. However, PGC1α appeared capable of mitigating the impairment of respiratory chain complex enzymes and mitochondrial respiration induced by MST1 overexpression in cells. Overall, we hypothesize that MST1 modulates the transcription and translation of MT-ND4L, MT-CO2, and MT-ATP6 through PGC1α, influencing the OXPHOS pathway and disrupting mitochondrial homeostasis.

We also observed that the PI3K-Akt signaling pathway ranked highest in the KEGG pathway analysis, suggesting its association with the role of MST1 in AD. The PI3K-Akt signaling pathway is crucial in regulating fundamental cellular processes and metabolism [56]. This pathway is involved in various mechanisms of AD, such as neuronal loss, inflammation, oxidative stress, and glucose metabolism [57]. Other studies have reported that PI3K-Akt signal transduction is involved in mitochondrial aerobic respiration and exerts broad effects on oxidative stress [58]. In addition, PI3K-Akt activation directly stimulates NADPH oxidase (NOx) and promotes ROS production [58]. Several studies have suggested that the Hippo pathway negatively modulates Akt [59]. In this study, similar to previous studies, we found MST1 prompted elevated oxidative stress, heightened levels of mitochondria-related apoptotic proteins, and mitochondrial dysfunction by suppressing the PI3K-Akt pathway. In addition, we observed that chemical inhibitors or activators targeting the PI3K-Akt pathway counteracted the effects of MST1 in various treatments, thus affecting oxidative stress and mitochondrial function. Collectively, our data support the involvement of MST1 in AD development through regulation of the PI3K-Akt-ROS pathway.

Finally, XMU-MP-1 was intraperitoneally injected in 7-month-old 5xFAD mice to inhibit MST1 expression. We found that XMU-MP-1 intervention alleviated cognitive impairment in AD mice. WB validation revealed that chemically modulating MST1 expression reduced apoptosis and synaptic damage. Additionally, we noted variations in PGC1a, mtDNA-encoded proteins, and proteins involved in the PI3K-AKT pathway. Furthermore, in vivo testing confirmed that MST1 indeed influences oxidative phosphorylation and PI3K-AKT pathway, aligning with previous in vitro findings.

While this study explored a novel mechanism involving MST1 in AD, it has some limitations. Although this study primarily focused on the impact of MST1 on mitochondrial function and oxidative stress in AD, exploring its immunoregulatory role could provide a more comprehensive understanding of its involvement in the disease. Investigating how MST1 influences immune responses, neuroinflammation, and microglial activation in AD models may uncover additional therapeutic targets. Overall, our results validate that MST1 modulates the transcription and translation level of mtDNA through PGC1α, culminating in dysfunction of the mitochondrial respiratory chain. It enhances oxidative stress through the PI3K-Akt pathway. These mechanisms can ultimately disrupt mitochondrial homeostasis in AD, thus accelerating cognitive impairment and neuronal damage in 5xFAD mice. In conclusion, this study presents evidence supporting the potential therapeutic benefits of targeting MST1. Additionally, translating findings from preclinical models to clinical settings is essential for developing effective therapeutic strategies for AD. Future studies should explore the correlation between MST1 expression levels, mitochondrial dysfunction, and cognitive decline in human AD patients. Investigating MST1 as a potential biomarker for disease progression and therapeutic response could guide personalized treatment approaches.

AAV

adeno-associated virus

Alzheimer's disease

Aβ

β-amyloid

ANOVA

analysis of variance

BSA

bovine serum albumin

CCK8

Cell-Counting-Kit-8

Co-IP

Co-Immunoprecipitation

DAB

diaminobenzidine

DEG

differently expressed gene

DRP1

dynamin-related protein 1

ECL

enhanced chemiluminescence

ECT

electron transport chain

ELISA

enzyme-linked immunosorbent assay

FCCP

phenylhydrazone

FIS1

fission 1 protein

Hematoxylin-eosin

immunofluorescence

IHC

immunohistochemistry

MFN2

mitofusion 2

MWM

Morris water maze

MST1

mammalian Ste20-like kinase

NRF1

recombinant nuclear respiratory factor 1

OCR

oxygen consumption rate

OCT

optimal cutting temperature

OPA1

optic atrophy 1

OXPHOS

oxidative phosphorylation

PFA

paraformaldehyde

PGC1α

peroxisome-proliferator-activated γ co-activator-1α

PSD95

postsynaptic density-95

PVDF

polyvinylidene difluoride

RIPA

radio immunoprecipitation assay

ROS

reactive oxygen species

SDS

sodium dodecyl sulfate-polyacrylamide

SEM

standard error of the mean

SYP

synaptophysin

TEM

transmission electron microscopy

western blotting

Supplementary Material

Supplementary figures and tables are provided in the attached document.

Competing Interests

The authors have declared that no competing interest exists.

Funding

This study was supported by grants from the National Natural Science Foundation of China (82171410, 81870848), the Natural Science Foundation of Shandong Provincial (ZR2023MH340), Cultivation Fund of the Second Hospital of Shandong University (2022YP93), Jinan Clinical Medical Science and Technology Innovation Program (202134018), Shandong University Multidisciplinary Research and Innovation Team of Young Scholars (2020QNQT019), and Clinical Research Center of Shandong University (No. 2020SDUCRCA001).

Author contributions

Dongqing Cui, Ping Wang, Lili Cao, and Jianzhong Bi conceived and designed the research. Dongqing Cui, Haixia Liu, Xiaowei Du, and Dingxin Liu performed the experiments. Dongqing Cui analyzed the data and drafted the manuscript. Zhiping Liu and Tong Wang participated in RNA-seq data analysis and drafted relevant content for the manuscript. Jianzhong Bi, Ping Wang, Lili Cao, Shunliang Xu, Zhaohong Xie, Xiaolei Zheng, and Hui Yang are responsible for data supervision and manuscript review. All authors read and approved the final manuscript.

Data availability

The data in this study are available from the corresponding author upon reasonable request.

Scheltens P, De Strooper B, Kivipelto M, Holstege H, Chételat G, Teunissen CE, et al. Alzheimer's disease. Lancet. 2021;397:1577–90.
Hardy JA, Higgins GA. Alzheimer's disease: the amyloid cascade hypothesis. Science. 1992;256(5054):184–5.
Breijyeh Z, Karaman R. Comprehensive Review on Alzheimer’s Disease: Causes and Treatment. Molecules. 2020; 25.
Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396:413–46.
Reiss AB, Gulkarov S, Jacob B, Srivastava A, Pinkhasov A, Gomolin IH et al. Mitochondria in Alzheimer’s Disease Pathogenesis. Life. 2024; 14.
Harris Julia J, Jolivet R, Attwell D. Synaptic Energy Use and Supply. Neuron. 2012;75:762–77.
Dhapola R, Sarma P, Medhi B, Prakash A, Reddy DH. Recent Advances in Molecular Pathways and Therapeutic Implications Targeting Mitochondrial Dysfunction for Alzheimer’s Disease. Mol Neurobiol. 2021;59:535–55.
Han Y, Liu D, Cheng Y, Ji Q, Liu M, Zhang B, Zhou S. Maintenance of mitochondrial homeostasis for Alzheimer's disease: Strategies and challenges. Redox Biol. 2023; 63.
Lee A, Kondapalli C, Virga DM, Lewis TL, Koo SY, Ashok A et al. Aβ42 oligomers trigger synaptic loss through CAMKK2-AMPK-dependent effectors coordinating mitochondrial fission and mitophagy. Nat Commun. 2022; 13.
Dhapola R, Beura SK, Sharma P, Singh SK, HariKrishnaReddy D. Oxidative stress in Alzheimer’s disease: current knowledge of signaling pathways and therapeutics. Mol Biol Rep. 2024; 51.
Morán M, Moreno-Lastres D, Marín-Buera L, Arenas J, Martín MA, Ugalde C. Mitochondrial respiratory chain dysfunction: Implications in neurodegeneration. Free Radic Biol Med. 2012;53:595–609.
Alzheimer's Disease Puzzle. Delving into Pathogenesis Hypotheses. aging and disease. 2024.
Lanzillotta C, Di Domenico F, Perluigi M, Butterfield DA. Targeting Mitochondria in Alzheimer Disease: Rationale and Perspectives. CNS Drugs. 2019;33:957–69.
Irwin M, Tare M, Singh A, Puli OR, Gogia N, Riccetti M et al. A Positive Feedback Loop of Hippo- and c-Jun-Amino-Terminal Kinase Signaling Pathways Regulates Amyloid-Beta-Mediated Neurodegeneration. Front Cell Dev Biology. 2020; 8.
Xu J, Patassini S, Rustogi N, Riba-Garcia I, Hale BD, Phillips AM et al. Regional protein expression in human Alzheimer’s brain correlates with disease severity. Commun Biology. 2019; 2.
Tasker R, Rowlands J, Ahmed Z, Di Pietro V. Co-Expression Network Analysis of Micro-RNAs and Proteins in the Alzheimer’s Brain: A Systematic Review of Studies in the Last 10 Years. Cells. 2021; 10.
Sahu MR, Mondal AC. The emerging role of Hippo signaling in neurodegeneration. J Neurosci Res. 2019;98:796–814.
Meng Z, Moroishi T, Guan K-L. Mechanisms of Hippo pathway regulation. Genes Dev. 2016;30:1–17.
Yu L, Liu Y, Jin Y, Cao X, Chen J, Jin J, et al. Lentivirus-Mediated HDAC3 Inhibition Attenuates Oxidative Stress in APPswe/PS1dE9 Mice. J Alzheimers Dis. 2018;61:1411–24.
Sanphui P, Biswas SC. FoxO3a is activated and executes neuron death via Bim in response to β-amyloid. Cell Death Dis. 2013;4:e625–e.
Li D, Ni H, Rui Q, Gao R, Chen G. Mst1: Function and Mechanism in Brain and Myocardial Ischemia Reperfusion Injury. Curr Neuropharmacol. 2018;16:1358–64.
Jeong DJ, Um J-H, Kim YY, Shin DJ, Im S, Lee K-M, et al. The Mst1/2-BNIP3 axis is required for mitophagy induction and neuronal viability under mitochondrial stress. Exp Mol Med. 2024;56:674–85.
Zhang J, Sun L, Li W, Wang Y, Li X, Liu Y. Overexpression of macrophage stimulating 1 enhances the anti-tumor effects of IL-24 in esophageal cancer via inhibiting ERK-Mfn2 signaling-dependent mitophagy. Biomed Pharmacother. 2019; 114.
Wu W, Ziemann M, Huynh K, She G, Pang Z-D, Zhang Y, et al. Activation of Hippo signaling pathway mediates mitochondria dysfunction and dilated cardiomyopathy in mice. Theranostics. 2021;11:8993–9008.
Ma C, Fan L, Wang J, Hao L, He J. Hippo/Mst1 overexpression induces mitochondrial death in head and neck squamous cell carcinoma via activating β-catenin/Drp1 pathway. Cell Stress Chaperones. 2019;24:807–16.
Fan F, He Z, Kong LL, Chen Q, Yuan Q, Zhang S, et al. Pharmacological targeting of kinases MST1 and MST2 augments tissue repair and regeneration. Sci Transl Med. 2016;8(352):352ra108.
Guo N, Soden ME, Herber C, Kim MT, Besnard A, Lin P, et al. Dentate granule cell recruitment of feedforward inhibition governs engram maintenance and remote memory generalization. Nat Med. 2018;24:438–49.
Bromley-Brits K, Deng Y, Song W. Morris Water Maze Test for Learning and Memory Deficits in Alzheimer's Disease Model Mice. J Visualized Experiments. 2011.
Oakley H, Cole SL, Logan S, Maus E, Shao P, Craft J, et al. Intraneuronal β-Amyloid Aggregates, Neurodegeneration, and Neuron Loss in Transgenic Mice with Five Familial Alzheimer's Disease Mutations: Potential Factors in Amyloid Plaque Formation. J Neurosci. 2006;26:10129–40.
Lam J, Katti P, Biete M, Mungai M, AshShareef S, Neikirk K et al. A Universal Approach to Analyzing Transmission Electron Microscopy with ImageJ. Cells. 2021; 10.
Lin MT, Beal MF. Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature. 2006;443:787–95.
Weidling I, Swerdlow RH. Mitochondrial Dysfunction and Stress Responses in Alzheimer’s Disease. Biology. 2019; 8.
Shadel GS. Coupling the mitochondrial transcription machinery to human disease. Trends Genet. 2004;20:513–9.
Calame DG, Emrick LT. Functional genomics and small molecules in mitochondrial neurodevelopmental disorders. Neurotherapeutics. 2024; 21.
Huo C, Chen M-H, Hour T-C, Huang L-C, Fong Y-O, Kuo Y-Y et al. Application of Micro-Western Array for Identifying Different Serum Protein Expression Profile among Healthy Control, Alzheimer’s Disease Patients and Patients’ Adult Children. Brain Sci. 2022; 12.
Wang H, Shang Y, Wang E, Xu X, Zhang Q, Qian C et al. MST1 mediates neuronal loss and cognitive deficits: A novel therapeutic target for Alzheimer’s disease. Prog Neurobiol. 2022; 214.
Lee S, Jang KI, Lee H, Jo YS, Kwon D, Park G et al. Multi-proteomic analyses of 5xFAD mice reveal new molecular signatures of early‐stage Alzheimer's disease. Aging Cell. 2024.
Korkmaz OT, Ay H, Aytan N, Carreras I, Kowall NW, Dedeoglu A, Tuncel N. Vasoactive Intestinal Peptide Decreases β-Amyloid Accumulation and Prevents Brain Atrophy in the 5xFAD Mouse Model of Alzheimer’s Disease. J Mol Neurosci. 2018;68:389–96.
Jack CR, Knopman DS, Jagust WJ, Petersen RC, Weiner MW, Aisen PS, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 2013;12:207–16.
Chang J, Li Y, Shan X, Chen X, Yan X, Liu J, Zhao L. Neural stem cells promote neuroplasticity: a promising therapeutic strategy for the treatment of Alzheimer’s disease. Neural Regeneration Res. 2024;19:619–28.
Kiebish MA, Han X, Cheng H, Lunceford A, Clarke CF, Moon H, et al. Lipidomic analysis and electron transport chain activities in C57BL/6J mouse brain mitochondria. J Neurochem. 2008;106:299–312.
Stauch KL, Purnell PR, Fox HS. Quantitative Proteomics of Synaptic and Nonsynaptic Mitochondria: Insights for Synaptic Mitochondrial Vulnerability. J Proteome Res. 2014;13:2620–36.
Cardanho-Ramos C, Morais VA. Mitochondrial Biogenesis in Neurons: How and Where. Int J Mol Sci. 2021; 22.
Shang H, VanDusseldorp TA, Ma R, Zhao Y, Cholewa J, Zanchi NE, Xia Z. Role of MST1 in the regulation of autophagy and mitophagy: implications for aging-related diseases. J Physiol Biochem. 2022;78:709–19.
Feng X, Wang S, Yang X, Lin J, Man W, Dong Y et al. Mst1 Knockout Alleviates Mitochondrial Fission and Mitigates Left Ventricular Remodeling in the Development of Diabetic Cardiomyopathy. Front Cell Dev Biology. 2021; 8.
Wang P, Geng J, Gao J, Zhao H, Li J, Shi Y et al. Macrophage achieves self-protection against oxidative stress-induced ageing through the Mst-Nrf2 axis. Nat Commun. 2019; 10.
Wang Y, Yang Q, Shen S, Zhang L, Xiang Y, Weng X. Mst1 promotes mitochondrial dysfunction and apoptosis in oxidative stress-induced rheumatoid arthritis synoviocytes. Aging. 2020;12(16):16211–23.
Ueda Y, Kondo N, Kinashi T. MST1/2 Balance Immune Activation and Tolerance by Orchestrating Adhesion, Transcription, and Organelle Dynamics in Lymphocytes. Front Immunol. 2020; 11.
Li T, Wen Y, Lu Q, Hua S, Hou Y, Du X, et al. MST1/2 in inflammation and immunity. Cell Adhes Migr. 2023;17:1–15.
Zhang X, Farrell JJ, Tong T, Hu J, Zhu C, Wang LS, et al. Association of mitochondrial variants and haplogroups identified by whole exome sequencing with Alzheimer's disease. Alzheimer's Dement. 2021;18:294–306.
Holper L, Ben-Shachar D, Mann JJ. Multivariate meta-analyses of mitochondrial complex I and IV in major depressive disorder, bipolar disorder, schizophrenia, Alzheimer disease, and Parkinson disease. Neuropsychopharmacology. 2018;44:837–49.
Lim K. Mitochondrial genome editing: strategies, challenges, and applications. BMB Rep. 2024;57:19–29.
Kaarniranta K, Uusitalo H, Blasiak J, Felszeghy S, Kannan R, Kauppinen A et al. Mechanisms of mitochondrial dysfunction and their impact on age-related macular degeneration. Prog Retin Eye Res. 2020; 79.
Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, Mootha V, et al. Mechanisms Controlling Mitochondrial Biogenesis and Respiration through the Thermogenic Coactivator PGC-1. Cell. 1999;98:115–24.
Lehman JJ, Barger PM, Kovacs A, Saffitz JE, Medeiros DM, Kelly DP. Peroxisome proliferator-activated receptor gamma coactivator-1 promotes cardiac mitochondrial biogenesis. J Clin Invest. 2000;106(7):847–56.
Chi M, Liu J, Mei C, Shi Y, Liu N, Jiang X et al. TEAD4 functions as a prognostic biomarker and triggers EMT via PI3K/AKT pathway in bladder cancer. J Experimental Clin Cancer Res. 2022; 41.
Kumar M, Bansal N. Implications of Phosphoinositide 3-Kinase-Akt (PI3K-Akt) Pathway in the Pathogenesis of Alzheimer’s Disease. Mol Neurobiol. 2021;59:354–85.
Liu C-M, Ma J-Q, Sun Y-Z. Puerarin protects rat kidney from lead-induced apoptosis by modulating the PI3K/Akt/eNOS pathway. Toxicol Appl Pharmcol. 2012;258:330–42.
Ye X, Deng Y, Lai Z-C. Akt is negatively regulated by Hippo signaling for growth inhibition in Drosophila. Dev Biol. 2012;369:115–23.

Download PDF

Reviewers agreed at journal
28 Jun, 2024
Reviewers invited by journal
28 Jun, 2024
Editor assigned by journal
22 Jun, 2024
First submitted to journal
19 Jun, 2024

You are reading this latest preprint version

MST1, a novel therapeutic target for Alzheimer's disease, regulates mitochondrial homeostasis by mediating mitochondrial DNA transcription and PI3K-Akt-ROS pathway.

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and methods

Results

Discussion

Abbreviations

Declarations

Supplementary Material

Competing Interests

Funding

Author contributions

Data availability

References

Supplementary Files

Status:

Version 1