Abstract
Neuroinflammation and ferroptosis significantly contribute to neuronal death in Alzheimer’s disease (AD) and other neurodegenerative disorders. MicroRNAs (miRNAs) are crucial regulators of these pathological processes. We employed transcriptomic analysis in an APP/PSEN1 Tg AD mouse model to identify dysregulated miRNAs and construct a miRNA–mRNA pathway network. We discovered increased miR-7a expression in the AD brain, targeting Krüppel-like factor 4 (Klf4), a transcriptional factor implicated in amyloid-β (Aβ) oligomer-induced neuroinflammation and RSL3-induced neuronal ferroptosis. Elevated Klf4 levels in AD mice brains suggest its involvement in AD pathology. The miR-7a mediated silencing of Klf4 alleviates neuroinflammation by modulating NF-κB, iNOS, and NLRP3 pathways, and inhibition of ferroptosis by targeting labile iron levels, GPX4, Nrf2 pathway, and mitochondrial damage. These findings highlight the neuroprotective role of miR-7a and its potential as RNA therapeutic. Pharmacological targeting of the miR-7a–Klf4 axis with blood–brain barrier (BBB)-permeable compound effectively mitigates neuroinflammation and ferroptosis, suggesting the miR-7a–Klf4 axis as a novel therapeutic target for AD.
Graphical Abstract
Graphical Abstract.
Introduction
Alzheimer’s disease (AD) is a prevalent progressive neurodegenerative condition, characterized by the accumulation of amyloid-β (Aβ) and tau protein aggregation species [1, 2]. These pathological events are responsible for the observed impairments in learning, memory, and overall cognitive function [1, 2]. The multifaceted etiology of AD, with its intricate pathological indicators and clinical symptoms, presents formidable challenges in the development of effective treatments [3]. Currently available therapeutic measures are confined to providing only temporary and symptomatic relief [4]. The stagnation in therapeutic advancements can be largely attributed to the insufficient exploration of viable therapeutic targets and a dearth of innovative strategies aimed at accurately targeting the regulators of disease pathology [5]. There is an urgent and imperative need to identify new therapeutic targets that are intrinsic to the neurodegenerative processes, to improve the therapeutic management of AD.
Recent studies have expanded the understanding of AD pathogenesis beyond Aβ and tau to include the pivotal role of neuroinflammation in the disease onset and progression [6]. Microglial cells, the brain resident macrophages, serve a paradoxical role. While microglia are essential for the function and homeostasis of central nervous system (CNS), they also exacerbate neurodegeneration by releasing proinflammatory mediators [7]. Microglia exist in homeostatic and disease-associated states that regulates its dual role [8, 9]. However, the molecular mechanisms governing these states remain elusive. Unraveling the key regulators of microglial state and neuroinflammation is crucial for developing therapeutic interventions [10, 11]. Aβ peptides misfold and aggregate to form Aβ oligomers (AβO) and fibrils, with former being the most toxic species that triggers microglial activation and neuroinflammation. AβO activates microglial cells and induces the production and release of proinflammatory cytokines, resulting in neuroinflammation [12]. Oligomers bind to toll-like receptors (TLR4), receptors for advanced glycation end products (RAGEs), Fc receptors, and nod-like receptors, leading to the activation of various pathways. Oligomers activate NF-κB, kinase pathways (MAPK and tyrosine kinases: Lyn and Syk), and inflammasomes downstream, triggering the expression of various proinflammatory molecules such as TNFα, IL6, IL1β, and caspase 1. Oligomers are directly taken up by microglial cells, inducing mitochondrial damage and the generation of reactive oxygen species (ROS) and NO, which further aggravates neuroinflammation. Activated microglia drives neuroinflammation via proinflammatory mediators, modulated by pathways such as NF-κB, ROS, and NLRP3 inflammasome [13]. Ferroptosis/oxytosis, a new form of cell death triggered by iron-mediated lipid peroxidation, plays a role in various disease conditions [14]. It is characterized by elevated labile iron pool (LIP), lipid peroxidation, ROS production, oxidative stress, impaired glutathione peroxidase 4 (GPX4) function, disrupted Nrf2 signaling, mitochondrial damage, and subsequent cell death implicated in various neurodegenerative disorders [14–19]. Recent studies have established the connection between ferroptosis and various neurodegenerative diseases, including AD [15, 20]. Proteins involved in the ferroptosis pathway are significantly altered in AD, implicating its potential contribution to neurodegeneration [21–23]. The increased labile iron levels and lipid peroxidation products in the AD brain suggest its pathological nexus with ferroptosis [15, 24]. RAS-selective lethal 3 (RSL3) is a small molecule ferroptosis inducer in various cell types, including neuronal cells [14]. RSL3 covalently binds to the selenocysteine in the GPX4 active site and inhibits the enzyme responsible for removing lipid hydroperoxides. Inhibition of GPX4 results in the accumulation of lipid hydroperoxides, leading to lipid peroxidation and ROS accumulation [25]. Lipid peroxidation and redox imbalance further activate ferritinophagy and alter iron homeostasis, elevating Fe2+ levels and resulting in ferroptosis. Evidence of ferroptosis in aging human microglia and Aβ-induced ferroptotic signatures in neuronal cells underscore its significance in cell death mechanisms [15, 24, 26]. A deeper understanding of the molecular underpinnings and regulators of ferroptosis may offer new avenues for mitigating neurodegeneration in AD.
MicroRNAs (miRNAs) play functional and regulatory roles in cellular physiology and pathophysiology. They are the master regulators of gene expression at the post-transcriptional level, orchestrating the degradation of messenger RNAs (mRNAs) or suppressing their translation [27]. As master regulators, miRNAs modulate a multitude of mRNA targets, influencing physiological and pathological pathways [28–31]. They orchestrate various neuronal functions, from signal transduction to memory consolidation. Despite their significance and critical role, comprehensive identification and understanding of miRNAs in AD pathology are still limited. miRNAs have been shown to target APP, BACE1, AChE, tau, GSK3β, and phosphatases [30]. These insights underscore the potential of miRNA and miRNA-based therapeutics in targeting AD pathological mechanisms. Discovering novel miRNAs that govern other pathways contributing to neurodegeneration, such as neuroinflammation and ferroptosis, herald a new therapeutic prospect.
In this study, we embarked on a comprehensive analysis of miRNA profiles and their target mRNAs to establish a miRNA–mRNA pathway network employing the APP/PSEN1 transgenic (Tg) AD mouse model. The study uncovered a significant upregulation of miR-7a in the AD brains, highlighting its regulatory function in AD pathology. Notably, Klf4 was identified as a principal mRNA target of miR-7a and Klf4 protein levels was significantly elevated in AD. Klf4 was upregulated in AβO induced neuroinflammation and miR-7a–Klf4 axis was found to regulate microglial activation and neuroinflammation. Klf4 is a pivotal factor in neuronal ferroptosis and regulate ferroptotic neuronal death. The inhibition of Klf4 by miR-7a effectively reduced AβO-induced microglial activation, neuroinflammation, and RSL3-triggered ferroptotic cell death. The pharmacological modulation of the miR7a–Klf4 axis by honokiol is a promising approach to alleviate neuroinflammation and ferroptosis in AD, offering potential RNA-based and small molecule interventions.
Materials and methods
Chemicals and reagents
All the chemicals and reagents were purchased and used without purification unless mentioned. The RNA isolation reagent Trizol, DEPC treated water (#AM9922), nuclease-free water (NFW) (#AM9932), RNaseZap, Verso cDNA Synthesis Kit (#AB1453A), DyNAmo ColorFlash SYBR Green qPCR Kit (#F416L), Lipofectamine™ RNAiMAX Transfection Reagent (#13778075) and MicroAmp™ Optical 96-Well Reaction Plate with Barcode & Optical Adhesive Films (#4314320), Calcein AM (#C3099), 4',6-diamidino-2-phenylindole (DAPI) (#62248), BODIPY™ 581/591 C11 (Lipid Peroxidation Sensor) (#D3861) MitoTracker™ Green FM Dye (#M46750) Rhodamine 123 (#R302), and Human beta Amyloid (1-42) PTD Recombinant Protein (#03-112) were purchased from Invitrogen, ThermoFisher Scientific. miRNA mimic, inhibitor (Assay Id-MH10047), and negative control (#4464058) were purchased from Invitrogen, ThermoFisher Scientific, and Eurogenetec. iTaq Universal SYBR Green Supermix 2X (#1725121) was purchased from BioRad. miScript II RT Kit (#218161) was purchased from QIAGEN Pvt. Ltd. The custom DNA oligos (primers) were synthesized by Eurofins and directly used without further purification. Atto-647 5′ modified miR-7a RNA sequence and Klf4 mRNA fragment RNA oligonucleotides were purchased from IDT (Supplementary Table S10). 2′,7′-Dichlorofluorescin diacetate (DCFDA #287 810), Griess reagent (#G4410), 1S,3R-RSL 3 (#SML2234), Escherichia coli lipopolysaccharide (LPS) (#L4391), 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyl tetrazolium bromide (MTT) (#M5655), quercetin (#Q4951), honokiol (#H4914), and nicorandil (#N3539) from Merck, Sigma–Aldrich. The cell media Dulbecco’s Modified Eagle Medium (DMEM-F12), fetal bovine serum (FBS), goat serum (GS), penicillin–streptomycin (PS), and Roswell Park Memorial Institute (RPMI) were purchased from Gibco, ThermoFischer Scientific. Sterile T25 flasks, 96-, 12-, and 6-well plates (Thermos Fischer Scientific), and confocal dishes (SPL life Sciences) were purchased and used without further sterilization. The antibodies Klf4 (6C5) (#BSM-52850R), Nrf2 (#PA5-88084), iNOS (PA1-036), Anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488 (#A-11034), and Anti-Chicken IgY (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 568 (A-11041) were purchased from Invitrogen. Antibodies Iba1(E404W) (#17198S) and TNFα (D2D4) (#11948S) were purchased from cell signaling technologies (CST). The antibodies NF-κB (ARC0084) (#A19605) and NLRP3 (#A12694) were from Abclonal. The antibodies Iba1 chicken (#234 009) and GPX4 (#E-AB-64550) were purchased from synaptic systems and Elabscience, respectively. The antibodies were used as per the manufacturer’s instructions with suitable dilutions (Supplementary Table S1). The list of antibodies and their details used in the study is completely detailed in Supplementary Table S1. The sandwich enzyme-linked immunosorbent assay (ELISA) kits for mouse IL1β (#RK00006), TNFα (#RK00027), and IL6 (#RK00008) were purchased from ABclonal and used as per the manufacture’s protocol. The complimentary DNA (cDNA) synthesis was performed on Eppendorf thermocycler and quantitative real time - polymerase chain reaction (qRT-PCR) was performed on QuantaStudio 5, Applied Biosystems real time PCR instrument. Circular dichroism (CD) and fluorescence spectra were recorded on J1500 (Jasco) and Agilent Cary Eclipse Fluorescence Spectrophotometer, respectively. Temperature-dependent absorbance kinetics was performed in JASCO V-750 UV spectrophotometer. Well plate absorbance and fluorescence measurement were performed in Spectramax i3 (Molecular devices) plate reader instrument. Live cell fluorescence imaging was carried out in Leica DMi8 fluorescence microscope equipped with live cell set up. The confocal imaging was performed using Olympus Fluoview-3000 confocal laser scanning microscope and analyzed by ImageJ software.
Animals and maintenance
APP/PSEN1 double transgenic mice [B6.Cg-Tg(APPswe,PSEN1dE9)85Dbo/Mmjax] were obtained from the Jackson Laboratory (MMRRC stock no. 34 832) and maintained in Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR) Animal Facility under 12-h light/dark cycle. All the animal breeding, maintenance, and studies were performed according to the protocols of the Institutional Animal Ethics Committee (IAEC). The protocol (TG004) was approved by the IAEC.
RNA extraction and NanoString miRNA expression assay
Total RNA including miRNA was isolated from mouse cortex and hippocampus tissue using Qiagen miRNAeasy mini kit (Cat#217004). Tissue was homogenized in TOMY homogenizer in presence of liquid nitrogen, for three cycles. Seven hundred microliters of QIAzol Lysis reagent was added to the tissue powder and mixed well. Lysate was incubated at room temperature for 5 min, and 140 μl of chloroform was added for phase separation. Rest of the protocol was followed as per manufacturer’s guidelines. RNA was eluted in 25 μl of NFW (Ambion, Cat #AM9932). Quantitation was performed using Qubit RNA HS assay (Invitrogen, Cat #Q32855) kit and qualitatively analyzed on Agilent 2100 bioanalyzer nano chip (Agilent, Cat #5067-1511). With NanoString Mouse v1.5 -miRNA assay kit (CSO-MMIR15-12), miRNA (3 μl) was ligated to mir-Tag with ligation buffer and ligase supplied with the kit. Ligated product was diluted with 15 μl of NFW and denatured at 85°C for 5 min. Five microliters of this was hybridized overnight at 65°C with reporter and capture probes. Protocol followed as per nCounter miRNA expression assay user manual (MAN-C0009-07). Post hybridization, samples were analyzed on NanoString nCounter SPRINT machine.
NanoString nCounter miRNA profiling and data analysis
Analysis was performed on all samples using the nCounter Analysis System (NanoString Technologies) and the nCounter Mouse v1 miRNA Panel (NS_M_miR_v1.5) having 599 unique miRNA barcodes for endogenous miRNA. The housekeeping genes used in the panel includes, β-actin (Actb), β-2-microglobulin (B2m), glyceraldehyde 3-phosphate dehydrogenase (Gapdh), and ribosomal protein L19 (Rpl19). The panel also includes the SpikeIn mIRNAs, Arabidopsis thaliana miR159a (ath-miR159a), Caenorhabditis elegans (cel)-miR-248 and miR254, and Oryza sativa (osa)-miR 414 along with the positive and negative controls for assessing overall assay efficiency as well as for monitoring the ligation efficiency.
The raw miRNA data in.RCC (Reporter Code Count) format was further analyzed using nSolver analysis software (NanoString technologies) version 4.0. QC metrics of imaging, binding density, positive control, limit of detection, and ligation were checked before proceeding with further analysis. Normalization of raw data was performed using the geometric mean of positive controls and top 100 highly expressed miRNAs. The differential expression between the groups was calculated using the build ratio utility present within the nSolver (foldchange), which uses Student’s unpaired t-test for the P-value calculation. The thresholds considered for a significant differentially expressed miRNA were, foldchange ≥ 1.2 or ≤ −1.2, P-value ≤ 0.05 and either of the two groups (test/control) should have their geometric mean expression ≥ average count for negative control probes in the panel. Unsupervised hierarchical clustering analysis followed by heatmap visualization and principal component analysis (PCA) was performed with the normalized and log2 transformed data using ClustVis (https://biit.cs.ut.ee/clust vis/) [32].
MiRNAs target prediction and functional over-representation analysis
For the statistically significant miRNA features (P-value < 0.05), mRNA target prediction was performed in a dual approach of experimentally validated mRNA target prediction and in silico target prediction. Experimentally validated targets were predicted using miRTarBase v.9.0 database, miRecords v.4, and TarBase v.8. The union of all interactions retrieved from the three queried sources was considered as the set of validated miRNA–target gene interactions in our study. To search for additional miRNA targets, in silico target predictions were performed for selected miRNAs using TargetScan v.8 and miRDB v.6.0 queried using miRWalk webserver, where the score for target interaction probability threshold was set to 0.9. Targets where the miRNA-binding sites ranged from cds, 5′- and/or 3′-UTR regions were considered for analysis. Only the intersection of all interactions retrieved from the two queried in silico sources were considered as the set of in silico predicted miRNA–target gene interactions in our study. Further the Gene Ontology (GO) biological process, cellular component, and molecular function along with REACTOME pathways were queried for both the targets of up or downregulated miRNAs, separately using ClusterProfiler [33] and ReactomePA R package [34]. Functions/pathways with a P-value < 0.05 were considered strongly enriched for the target genes targeted by selected miRNAs.
MiRNA–mRNA pathway network analysis
Validated target collection of selected individual miRNAs (P-value < 0.05) were enriched for Reactome Pathways using DAVID webserver (https://david.ncifcrf.gov/) [35]. The top significant (P-value < 0.05) enriched pathways and their corresponding genes were selected for further visualization as miRNA–mRNA pathway network using Cytoscape v3.10 [36]. The miRNA node size has been scaled based on the total number of validated interactions shared with target mRNAs.
MiRNA disease association and miRNA–mRNA disease network
In order to associate or enrich the differentially regulated miRNAs between AD relative to healthy controls (P-value < 0.05), both the association of miRNA with disease as well as validated targets of miRNA with disease were analyzed. The miRNAs and disease association was established using PhenomiR v.2.0 and miEAA (querying MNDR database), while the miRNA-validated target and disease association was analyzed through MNDR v.3.0 and DisGeNET cytoscape application [37]. Diseases with a P-value < 0.05 were considered strongly enriched for the selected collection of miRNAs. The miRNA–Gene Disease network was visualized using Cytoscape v3.10.
MiRNA expression validation by qPCR assay
The total RNA used for NanoString was further utilized for validation by qPCR. The miRNA cDNA was synthesized by miScript II RT Kit by using manufacturer’s protocol. The primers for the target miRNA were designed by using miRPrimer software [38] and custom synthesized from Eurofins. The assays were optimized and the qPCR was performed in Applied Biosystems QuantaStudio 5 system and assessed the miRNA expression with snoRNA as control by using comparative Ct (ΔΔCT) method.
Immunofluorescence of mouse brain
The brain tissue was fixed in a 4% paraformaldehyde solution, and then dehydrated in a 25% sucrose solution. The tissues were sectioned using a cryotome with a thickness of 25 μm. The sections were washed twice with phosphate buffered saline (PBS) to remove the antifreeze media and permeabilized with PBS containing 0.01% Triton X-100 for 10 min. After two additional washes with PBS, the tissues were incubated with a blocking solution (PBST with 3% goat serum). The tissues were then treated with the Klf4 primary antibody overnight at 4°C, washed three times with PBS, and subsequently treated with an Alexa-488 fluorophore-conjugated secondary antibody at room temperature. After a 1-h incubation, the samples were washed three times with PBS, mounted with DAPI-containing mounting media for nuclear staining, and subjected to confocal microscopy.
Western blot
Twelve-month-old WT and AD mice were sacrificed, and brain samples were collected. The brain tissues were homogenized in RIPA lysis buffer containing a protease inhibitor cocktail using a Dounce homogenizer. After complete homogenization, the lysate was centrifuged at 12 000 rpm for 10 min at 4°C, and the supernatant was collected. Total protein concentration was quantified using the Bradford assay with BSA as a standard. Forty micrograms of total protein was subjected to SDS–PAGE followed by western blotting. The blots were blocked with 5% skim milk powder, incubated with Klf4 primary antibody (1:1000) overnight at 4°C, and then treated with HRP-conjugated secondary antibody (1:5000) for 1 h at room temperature. The blots were developed on X-ray films using Clarity Western Enhanced Chemiluminescence (ECL) reagent (BIORAD). Band intensities were quantified using Fiji.
Circular dichroism and fluorescence spectroscopy
To understand the binding of miR-7a with the 3′-UTR region of Klf4 mRNA, we performed CD spectroscopy. Custom-synthesized miR-7a and the complementary region of Klf4 mRNA sequences were purchased and dissolved in Tris–HCl buffer (10 mM, pH 8.0). The miR-7a (25 μM) and Klf4 sequences (25 μM) were annealed at 65°C in 10 mM Tris buffer, and their CD spectra were measured individually. The sequences were then mixed, annealed at 65°C to promote binding at the complementary region, and allowed to cool slowly. The mixture was subjected to CD spectroscopy, and the data were processed and plotted using Origin software. To further determine the binding ability of miR-7a with Klf4, fluorescence-tagged miR-7a (miR-7a-Atto647) was titrated with increasing concentrations of the Klf4 sequence. A fixed concentration of miR-7a-Atto647 (500 nM) was used, and Klf4 sequence was added incrementally (250 nM to 10 μM), with Atto647 fluorescence measured after a 5-min incubation period following each addition. The complementary base pairing between miR-7a and Klf4 resulted in changes in Atto647 fluorescence, attributed to alterations in the local environment of the fluorescent dye due to base stacking. The change in fluorescence was plotted against the concentration of the Klf4 sequence to assess the binding behavior of miR-7a and Klf4.
UV absorbance RNA thermal melting (Tm) experiment
To understand the complementarity binding of miR-7a with the 3′-UTR region of Klf4 mRNA, we have performed UV absorbance RNA thermal melting experiment. The miR-7a and Klf4 sequences were mixed at an equimolar concentration of 2 μM in Tris buffer (10 mM, pH 7.4). The samples were annealed by heating at 90°C for 10 min and slow cooling to room temperature over 10 h and kept at 4°C overnight. To perform thermal melting 500 μl of 2 μM annealed samples were taken in quartz cuvette of 1 cm path length. The samples were kept at the starting temperature for 5 min to stabilize, and the temperature-dependent absorbance was recorded at 260 nm wavelength from 15°C to 90°C with an increment of 0.5°C. The normalized absorbance at 260 nm as function of temperature was plotted and analyzed using Origin software. The melting temperature (Tm) value was determined by using sigmoidal fitting with Boltzmann function with good fit and R2 value of 0.99. First derivative of normalized absorbance as a function of temperature yielded Tm with accurate to ±1.0°C.
General cell culture
SH-SY5Y neuroblastoma cells were cultured in DMEM-F12 medium supplemented with 10% FBS and 1% PS and incubated in a humidified incubator at 37°C with 5% CO2. The cells were grown in T25 flasks, trypsinized at 80% confluency, and reseeded as needed. N9 microglial cells were cultured in RPMI medium with 10% FBS and 1% PS and incubated under the same conditions. For experiments, 96-well, 12-well, and 6-well plates were seeded with cell densities of 15 000, 80 000, and 200 000 cells per well, respectively, and cultured for 24 h before proceeding with experiments.
RT-PCR and immunofluorescence
MiR-7a, Klf4 mRNA, and protein expression in SH-SY5Y and N9 microglial cells were assessed using qPCR and immunofluorescence. For qPCR, cells were cultured in a 6-well plate under standard conditions. Total RNA was isolated using Trizol reagent and quantified using a nano spectrophotometer to assess quantity and quality. For miR-7a expression analysis, cDNA was synthesized using the miScript II RT Kit according to the manufacturer’s protocol. Expression levels were analyzed by qPCR with snoRNA as the housekeeping gene, and the ratio of Ct values was plotted to compare expression between the two cell types. Similarly, for Klf4 mRNA expression, cDNA was prepared using the Verso cDNA Synthesis Kit, and mRNA expression was assessed by qPCR using β-actin as the control gene. For Klf4 protein expression, cells were cultured in confocal dishes for 24 h and then subjected to immunofluorescence. Cells were washed twice with PBS for 5 min each, fixed with 4% paraformaldehyde for 15 min, and washed three times with PBS for 5 min each. The cells were then permeabilized for 5 min, washed twice with PBS, and blocked for 30 min. Cells were treated with the Klf4 primary antibody (1:100) overnight at 4°C, washed three times with PBS for 5 min each, and then incubated with the Alex-488 fluorophore-conjugated secondary antibody (1:500) for 1 h at room temperature. After washing twice with PBS for 5 min each, DAPI solution in PBS was added for 15 min to stain the nuclei, and unbound probe was washed away with PBS. Finally, the samples were subjected to confocal imaging.
MiR-7a mimic and inhibitor transfection and immunofluorescence experiments
The miR-7a mimic, inhibitor, and negative control (NC) were purchased from Invitrogen and Eurogentec used as per the manufacturers protocol. The miR-7a mimic is a double-stranded RNA with a 2′-deoxy modification designed to preferentially load guide strands onto the RISC complex and silence target mRNAs. The miR-7a inhibitor is a single-stranded RNA with a 2′-O-Me modification for stability that specifically binds to miR-7a and inhibits its activity. The ability of miR-7a to target Klf4 and alter its levels was directly assessed at the protein level by transfecting neuronal cells and performing immunofluorescence to measure the level of Klf4. The products were constituted in NFW and used for the experiments. For transfection, the cells were cultured in confocal dishes and transfected with miR-7a mimics and inhibitor using Lipofectamine™ RNAiMAX transfection reagent as per the standard procedure. Cells were incubated for 9 h and then added low serum media (2.5% FBS) and after 30 h further subjected immunofluorescence for protein expression.
MiR-7a inhibitor - 5′ [A*A*AAACAACAAAAUCACUAGUCUUCCA*C*A*C*A] 3′ with O-Me modification.
Mi-7a mimic - Sense strand: 5′ [ACAACAAAAUCACUAGUCUUCCAdTdT] 3′
Antisense strand: 5′ [UGGAAGACUAGUGAUUUUGUUGUdTdT] 3′ with 2′ -deoxy modification.
Inflammation experiments—qPCR and immunofluorescence
N9 microglial cells were seeded in confocal dishes and 12-well plates and cultured under standard conditions for 24 h. The cells were treated with different concentrations of LPS (100, 250, and 500 ng/ml) and incubated for an additional 24 h. Total RNA was isolated and subjected to qPCR for mRNA expression analysis. Cells in the confocal dishes were fixed, subjected to Klf4 immunofluorescence, and imaged using a confocal microscope. For AβO-induced inflammation, cells were primed with LPS (500 ng/ml) for 4 h, and then treated with AβO oligomers (2.5, 5, and 10 μM). After treatment, the cells were subjected to qPCR for mRNA expression analysis and Klf4 immunofluorescence for protein expression.
To understand the inhibitory effect of microglial activation by miR-7a mimics, the microglial cells were transfected with miR-7a mimic, after 9 h the media was replaced with low serum media. The cells were primed with LPS (500 ng/ml), followed by treatment with AβO (5 μM) for 24 h. Then cells were subjected to Iba1 immunofluorescence to analyses the inhibitory effect of miR-7a on microglial activation.
DCFDA assay for ROS measurement
To assess ROS generation and the rescuing effect of miR-7a, microglial cells were transfected with miR-7a mimic (100 and 250 nM). After 9 h, the medium was replaced with low serum medium. The cells were primed with LPS (500 ng/ml), followed by treatment with AβO (5 μM) for 24 h. Subsequently, the cells were subjected to the DCFDA assay to measure intracellular ROS. Cells were washed twice with PBS, then treated with 20 μM DCFDA reagent in serum-free RPMI medium. After 45-min incubation, the cells were washed twice with PBS, placed in fresh medium, and analyzed using a multiplate reader in well scan mode to measure the mean fluorescence (λex = 485 and λex = 525) at from each well. A minimum of 100-point readings was set in a circular pattern, and the mean DCF fluorescence was used to determine cellular ROS levels.
Griess assay for NO measurement
The microglial cells were transfected with miR-7a mimic, after 9 h the media was replaced with low serum and phenol red free media. The cells were primed with LPS (500 ng/ml) for 4 h, followed by treatment with AβO (5 μM) for 24 h. Cell media was subjected to Griess assay to detect the relative NO released by activated microglial cells. Equal volume of cell media and Griess reagent (100 μl) were mixed and incubated for 10 min. The absorbance of the chromophore formed during diazotization of nitrite with sulfanilamide and subsequent coupling with naphthylethylenediamine was recorded by absorbance at 546 nm. The relative NO levels were calculated by normalizing against control cell media.
Neuroinflammatory mRNA and protein expression
To understand the inhibitory effect of microglial activation by miR-7a mimics, the microglial cells were transfected with miR-7a mimic, after 9 h the media was replaced with low serum media. The cells were primed with LPS (500 ng/ml), followed by treatment with AβO (5 μM) for 24 h. Then cells were subjected to qPCR for mRNA expression quantification of various neuroinflammatory mediators and immunofluorescence for protein expression analysis. The cell media was subjected to sandwich ELISA for protein quantification of TNFα, IL6, and IL1β using Abclonal ELISA kit as per the manufacturer’s protocol.
Ferroptosis experiments—qPCR and immunofluorescence
SH-SY5Y cells were cultured in 12 well plate and confocal dishes and treated with different concentrations of RSL3 (0.5, 1, and 2.5 μM) and vehicle to control at 60% confluency and incubated for 24 h. Then the cells were subjected to qPCR and immunofluorescence of Klf4 to understand the Klf4 levels in RSL3 induced ferroptosis condition.
Cells were transfected with miR-7a mimic at 50% confluency, after 9 h the media was replaced with low serum media. After 6-h incubation, the cells were treated with RSL3 (2 μM) and incubated for 24 h. The samples were subjected to GPX4 and Nrf2 immunofluorescence. Mean fluorescence intensity was calculated for relative protein levels and the ratio of nuclear and cytoplasmic mean fluorescence used to measure the extent of nuclear translocation of Nrf2.
Lipid peroxidation assay—BODIPY sensor
The extent of lipid peroxidation in the cells was imaged using the BODIPY C11 fluorescent probe. The probe shows red emission and upon reacting with lipid peroxides display green fluorescence. The increase in green fluorescence shows increased lipid peroxidation. The cells were cultured and transfected with miR-7a mimic and then treated with RSL3 (2 μM) and incubated for 24 h. Then, cells were treated with BODIPY sensor (5 μM) and incubated for 30 min followed by washing with PBS and imaging in live cell set up. The induction of ferroptosis is marked by increased lipid peroxides and the increased green fluorescence and the rescue has reduced green emission. To quantify the extent of lipid peroxidation we have performed assay in 24-well plate. Fluorescence in green (λex= 488 and λem = 510 nm) and red (λex= 581 and λem = 591 nm) emission were recorded in plate reader with well scan mode and plotted ratio of green and red to calculate relative lipid peroxidation.
Labile iron pool measurements—Calcein-AM imaging
Cells were seeded in confocal dishes and cultured in standard conditions. At 50% confluency, the cells were transfected with miR-7a mimic using RNAiMAX transfection reagent, and then after 9 h the media was replaced with low serum media. After 12 h, the cells were treated with RSL3 (2 μM) to induce ferroptosis for 24 h. Then cells were treated with calcein-AM (500 nM) for 45 min and washed with PBS and imaged in live cell imaging system. Calcein-AM intracellularly binds to Fe2+ and exhibits reduced fluorescence indicating increased LIP. The cells were imaged in green channel with live cell set up. The mean fluorescence intensity was quantified using ImageJ Fiji and normalized against control for relative iron levels in cells.
Mitochondrial imaging and MMP measurement
To analyze the protective effect of miR-7a targeting Klf4 on the ferroptosis mediated mitochondrial damage we performed mitochondrial imaging and estimation of mitochondrial membrane potential (MMP). To image the mitochondrial damage the cells transfected with miR-7a (100 nM and 250 nM) and challenged with RSL3 were incubated for 24 h. Then the cells were treated with Mitotracker green GM (250 nM) and incubated for 30 min. The cells were washed twice with PBS and imaged in fluorescence microscope with green channel in live cell set up. To measure the MMP cells cultured in 24-well plate were transfected with miR-7a (50, 100, and 250 nM) and then challenged with RSL3 and incubated for 24 h. Cells were treated with Rho123 (250 nM) a potential dependent mitochondrial probe and incubated for 30 min. Washed with PBS and fluorescence (λex = 507 and λem = 527 nm) was measured in multiplate reader with well scan mode with 80 reading points per well and the mean fluorescence intensity was used to measure the MMP.
Cell rescue by miR-7a and compounds
SH-SY5Y cells were seeded and cultured in 96-well plates and transfected with miR-7a mimic using RNAi transfection reagent. After 6 h, the media was removed and incubated with low serum DMEMF12 media. Six hours later the cells were treated with 2 μM of RSL3 to induce ferroptosis. After 30 h of incubation, RSL3-induced ferroptotic cell death and rescue by miR-7a was assessed by MTT assay. Briefly, cells were treated with 10 μl MTT reagent (5 mg/ml) for 3 h. Then the formazan crystals were dissolved in DMSO:Methanol (1:1) solvent mixture. The absorbance was measured at 570 nm to calculate the percentage cell viability against the control with vehicle treatment.
The cells were seeded and cultured in 96-well plates for 24 h and at 70% confluency the cells were treated with AβO (5 μM) for N9 microglial cells for inflammation or RSL3 (2 μM) for ferroptosis in SH-SY5Y cells and with different concentration of compounds. After 30 h of incubation, the cell viability was assessed by MTT assay. Percentage cell viability was calculated with respect to control using the absorbance at 570 nm.
RT-PCR and immunofluorescence with honokiol treatment
SH-SY5Y and N9 cells were cultured in confocal dishes and at 60% confluency the cells were treated with AβO alone and with honokiol (2.5 and 5 μM) to induce microglial activation and ferroptosis. After 24 h incubation, the cells were subjected to Iba1 immunofluorescence. The microglial activation and rescue by honokiol was assessed using Iba1 positivity and rescue by honokiol treatment. The effect of honokiol on Klf4 expression was assessed by RT-PCR and immunofluorescence. SH-SY5Y cells were cultured in 12-well plate and confocal dishes and then treated with honokiol (2.5 μM). Total RNA was isolated from well plate and subjected to RT-PCR for Klf4 mRNA expression. The dishes were subjected to immunofluorescence and the mean fluorescence intensity of Klf4 was calculated for relative protein levels.
Mouse cortical primary neuron-glia mixed culture
To evaluate anti-inflammatory and neuronal rescue effect of honokiol we have employed mouse primary cortical neuron-glia mixed culture using reported protocol with slight modifications [39]. Day one pups (P1) were sacrificed and dissected to get the cortical tissue and subjected to trituration. The tissue was triturated sequentially with serological pipette, 1 ml, 200 μl, and 20 μl tips to get homogeneous cell suspension. Then, the cells were strained through 100-μm mesh size filter and pelleted by centrifugation. The cells were seeded onto poly-D-lysine coated (20 μg/ml) dishes and plates. Then, the cells were maintained in neuron-glia maintenance media for 8 days. Cells were treated with AβO (2.5 μM) alone and with honokiol (2.5 and 5 μM) for 24 h and subjected to cell viability and immunofluorescence assay. Neuronal cells and activated microglia were probed with NeuN (Rabbit) and Iba1 (Chick) primary antibody (1:200 dilution) respectively. Alex-488 conjugated anti-rabbit and Alex-568 conjugated anti-chicken secondary antibodies (1:500 dilution) were used and subjected to confocal imaging.
Statistical analysis
All statistical analysis was performed by using R (version 4.0.2, https://www.r-project.org/) for transcriptomic analysis. P-value <0.05 was considered statistically significant. Pairwise comparisons were used to compare between the WT and AD groups. We have used GraphPad and Origin for plotting and analysis of the data. The statistical tests were applied using GraphPad. Used student t test for comparing two groups with * is P> 0.05 and one-way ANOVA to compare multiple groups with Bonferroni post hoc test * is P> 0.05 and ns indicate no significance difference.
Results
Differentially expressed miRNAs and altered miR–mRNA pathway/disease network associated with AD
A multitude of miRNAs are dysregulated in AD, and miRNA panels have been widely utilized as diagnostic biomarkers. The extensive regulatory roles of miRNAs within the CNS have prompted an in-depth investigation into their pathogenic and neuroprotective implications in AD. Utilizing the APP/PSEN1 transgenic AD mouse model, we assessed the expression of miRNAs in AD-afflicted brains in comparison to wild-type (WT) mice, aiming to establish altered miRNAs, mRNAs, and pathway network (Fig. 1A). The transcriptomic profiling of miRNAs in the cortex and hippocampus regions was carried out using the NanoString platform (Supplementary Tables S2 and S3). Transcriptome analysis revealed that 21 out of 577 analyzed miRNAs were differentially expressed (nCounter counts threshold of 40 and P< 0.005, t-test) in the cortex of AD compared to WT brain (Fig. 1B and D). Notably, five miRNAs (miR-7a, miR-696, miR-34a, miR-302b, and miR-151-3p) exhibits >1.5-fold change in expression levels. MiR-7a being significantly upregulated with a fold change of 2.05, whereas miR-696, showed pronounced downregulation by 2.55-fold. In the hippocampus, 11 miRNAs showed substantial expression changes, with a mix of both upregulation (5) and downregulation (6) observed (Fig. 1C and Supplementary Fig. S1C). The analysis of 577 miRNAs highlighted consistent upregulation of miR-7a in both cortex and hippocampus, while miR-148 and miR-301a were specifically upregulated in the hippocampus. Conversely, miR-696 and miR-34a were notably downregulated in the cortex, with miR-153 and miR-128 downregulated in the hippocampus. Multidimensional plots distinctly illustrated the separation of WT and AD miRNA profiles within both brain regions, indicating the potential regulatory impact of these dysregulated miRNAs in AD pathology (Supplementary Fig. S1A and B).
Figure 1.
Differentially expressed miRNAs and altered miRNA–mRNA pathway network. (A) Schematic of experimental and bioinformatics design of miRNA transcriptomics analysis. Volcano plots showing differentially expressed miRNAs in cortex (B) and hippocampus (C) of 12-month-old APP/PSEN1 Tg mouse model (n = 3 mice for both WT and AD group, P < 0.05). Differentially expressed miRNAs were identified using NanoString analysis (n = 3 mice per group, P < 0.05). (D) Heatmap of the top variable miRNAs of cortex. Molecular pathway of functional overrepresentation analysis for identified mRNA targets of upregulated (E) and downregulated (F) miRNA in cortex (P-value < 0.05). (G) MiRNA–mRNA pathway network constructed for dysregulated miRNA of cortex by using significantly (P-value < 0.05) enriched pathways and their corresponding genes.
Next, we performed a functional over-representation analysis on the altered mRNAs, employing target mRNA predictions for dysregulated miRNAs (Supplementary Tables S4 and S5). Further, we interrogated the GO categories: biological process, cellular component, and molecular function and REACTOME pathways for targets of both upregulated and downregulated miRNAs, utilizing ClusterProfiler and ReactomePA tools. The functional over-representation analysis of downregulated miRNAs in the cortex, using predicted mRNAs, identified pathways associated with AD pathology, including cellular stress response, transcriptional regulation, nervous system development, antigen processing, death receptor signaling, cell cycle signaling, transforming growth factor-β (TGF-β) signaling, and lipid metabolism (Fig. 1E and Supplementary Fig. S2). The death receptor signaling and cell stress response pathways are involved in triggering neuronal cell death due to amyloid toxicity and the intrinsic response to combating neurodegeneration and brain atrophy. Altered antigen processing, TGF-β signaling, and lipid metabolism are intrinsic pathways of microglia activation in AD, resulting in neuroinflammation. The analysis of upregulated miRNA targets indicated involvement in stress response, antigen processing, proteasomal degradation, cargo trafficking, immune cell signaling, TLR cascade, and Rho GTPase cycle pathways (Fig. 1F and Supplementary Fig. S3). Altered pathways such as antigen processing, proteasomal degradation, immune cell signaling, and the TLR cascade suggest their role in the interaction of Aβ with microglial cells. These pathways are involved in the uptake and clearance of Aβ through degradation, and the neuroinflammatory pathways linked to AD are regulated by miRNAs. In the hippocampus, downregulated miRNA pathways encompassed Rho GTPase cycle, cytokine signaling, antigen processing, proteasomal degradation, nervous system development, Golgi trafficking, transcriptional regulation, SUMOylation, and lipid metabolism (Supplementary Fig. S4). The pathways for upregulated miRNAs included Rho GTPase, transcriptional regulation, TLR cascades, mitogen-activated protein kinase (MAPK) activation, and signaling by the Hippo pathway (Supplementary Fig. S5). Collectively, the functional over-representation analyses elucidated that dysregulated miRNA targeted mRNAs are involved in various processes such as amyloid clearance, neuronal stress response and death, microglial activation, antigen presentation, and neuroinflammation-related pathways. These processes are potentially implicated in the development and progression of AD. This underscores the critical role of miRNAs in deciphering the intricate mechanisms of neurodegenerative diseases such as AD.
The validated mRNA targets of selected individual miRNAs (P-value < 0.05) were enriched for reactome pathways using DAVID webserver. The most significantly enriched pathways and their corresponding genes were subsequently visualized within a miRNA–mRNA pathway network. This network demonstrated the regulatory functions of miRNAs across various pathways, including transcription regulation, chromatin organization, Rho GTPase cycle, post-translational modifications (PTMs), membrane trafficking, cell–cell communication, immune cell signaling, cellular stress response, cell cycle pathways, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor activity, and synaptic plasticity (Fig. 1G, Supplementary Fig. S6A and Supplementary Data S1 and S2). These pathways have known associations with AD and other neurodegenerative diseases. Further investigations were undertaken to delineate the connections between miRNAs, validated mRNAs, and diseases, resulting in the construction of a miRNA–mRNA disease network. The relationships between miRNAs and diseases were established using PhenomiR v.2.0 and miEAA (querying the MNDR database). Further, the associations between miRNA–mRNA and diseases were analyzed using MNDR v.3.0 and the DisGeNET Cytoscape (Supplementary Tables S6–S9). The miRNA–mRNA disease network distinctly illustrated the links between dysregulated miRNAs and various neurodegenerative disorders, including AD (Supplementary Fig. S6B and C). These comprehensive transcriptomic and bioinformatic analyses have solidified the association of identified miRNAs with pathological pathways implicated in AD and other neurodegenerative diseases, emphasizing their potential as diagnostic and therapeutic targets.
MiR-7a is significantly induced and directly target Klf4 in AD
After identifying the altered miRNAs in the cortex and hippocampus of AD brains and the establishment of their mRNA pathway–disease network, we validated significantly altered miRNAs (miR-7a, miR-148a, miR-34a, miR-128a, and miR-153a) expression in the AD cortex and hippocampus using qRT-PCR. The expression patterns aligned with our comprehensive transcriptome analysis, revealing a significant upregulation of miR-7a and miR-148, while miR-34a and miR-128a were notably downregulated (Fig. 2A). Among these, miR-7a was markedly upregulated in both brain regions. Considering the neuroprotective role of miR-7a suggested by various studies, we hypothesize that miR-7a is potentially induced under disease conditions to mitigate AD pathology. The pathway network analysis vide supra revealed altered transcription regulation, immune, cell stress, and cell death response pathways. Our literature analysis, target prediction, and sequence alignment suggest Klf4 as a target for miR-7a bearing potential role in AD pathology. We observed an increase in Klf4 levels in the AD brain by immunofluorescence and western blot, suggesting its potential role in AD onset and progression (Fig. 2B and C, and Supplementary Data S3). The conserved sequence of miR-7a aligns complementarily with the 3′ UTR region of Klf4 mRNA, indicating a potential binding interaction that is reported to attenuate Klf4 expression (Fig. 2D). To investigate the complementary base pairing between miR-7a and Klf4, we utilized CD spectroscopy, which differentiates between single-stranded and double-stranded RNA structures. The interaction between miR-7a and Klf4 resulted in a partial double-stranded RNA formation, as evidenced by characteristic CD spectral bands at 270 (+), 240 (–), 220 (+), and 210 (–) nm (Fig. 2E). The annealing of miR-7a with the complementary Klf4 fragment led to a shift in the CD spectral band, indicative of base pairing which resulted in shifting of 270 nm band to lower wavelength and lowered intensity at 210 (–) nm with increased intensity. Fluorescence spectroscopy experiments with Atto-647 conjugated miR-7a (Atto647-miR-7a) and increasing concentrations of the complementary Klf4 mRNA fragment demonstrated a significant dose-dependent reduction in fluorescence (Fig. 2F and Supplementary Fig. S7A). The fluorophore conjugation to RNA potentially hinders the binding activity, although conjugation chemistry uses a linker to keep the fluorophore apart and minimize this effect. To confirm further binding, we performed a UV thermal melting experiment to study the complementary binding of the miR-7a and Klf4 mRNA sequences. The results show that the miR-7a and Klf4 mRNA sequences exhibit a Tm value of ∼21°C (theoretically calculated Tm ∼ 22°C), indicating complementary sequence binding (Supplementary Fig. S7B). These in vitro studies confirmed the binding of miR-7a to the 3′ UTR region of Klf4 mRNA, thereby corroborating its direct targeting mechanism.
Figure 2.
MiR-7a directly regulates increased Klf4 in AD. (A) Expression profile of dysregulated miRNAs identified by transcriptomic analysis in cortex and hippocampus of AD mouse brain validated by qRT-PCR (n = 3 biologically independent animals for WT and AD). (B) Representative immunofluorescence images of Klf4 in WT and APP/PSEN1 Tg AD mouse brain and its quantification (n = 4 animals for each WT and AD group with 2–3 section per animal; scale bar: 2 mm). (C) Representative western blot images of Klf4 in WT and APP/PSEN1 Tg AD mouse brain and its quantification (n = 3 animals for each WT and AD group). (D) Schematic representation of Klf4 gene and its mRNA with complementary region of human (hsa-miR-7a) and mouse (mmu-miR-7a) miR-7a. (E) Circular dichroism spectra of miR-7a, Klf4 and miR-7a-Klf4 annealed in Tris–HCl buffer (10 mM, pH 7.4) performed at 25°C. (F) Dose–response curve of change in fluorescence of Atto-647 labeled miR-7a with increasing concentration of Klf4 mRNA complementary sequence. (G) Relative expression of miR-7a and Klf4 mRNA in SH-SY5Y neuronal and N9 microglial cells by qRT-PCR and represented CtTarget /CtControl (n = 3). (H) Quantification of Klf4 in SH-SY5Y and N9 cells by immunofluorescence (n > 500 cells quantified for each group; scale bar: 200 μm). (I) Schematic representation of transfection experiment. Representative confocal images Klf4 immunofluorescence in SH-SY5Y transfected with miR-7a mimic (J) and inhibitor (K) with negative control (NC) and its quantification (n > 500 cells quantified for each group). Data are represented as mean ± s.e.m. (A–C), (G), and (H) Unpaired, two-tailed t-test *P < 0.05). (J and K) One-way ANOVA with Bonferroni post-hoc test *P < 0.05, **P < 0.01 and ****P < 0.0001; scale bar: 200 μm).
We analyzed the expression profiles of miR-7a and Klf4 through RT-PCR and immunofluorescence in SH-SY5Y neuronal and N9 microglial cell models. The results indicated higher relative expression of miR-7a in N9 microglial cells. In contrast, Klf4 expression was less pronounced in N9 microglial cells compared to SH-SY5Y cells (Fig. 2G). Subsequent immunofluorescence analysis showed higher expression of Klf4 protein in SH-SY5Y cells compared to N9, suggesting potential regulation by miR-7a (Fig. 2H and Supplementary Fig. S7C). Klf4, known for its transcription regulation function, is predominantly located in the nucleus. To confirm the direct regulatory effect of miR-7a on Klf4, SH-SY5Y cells were transfected with a miR-7a mimic and inhibitor, along with a negative control, and subjected to immunofluorescence to assess Klf4 levels (Fig. 2I). Results showed a notable decrease in Klf4 expression in cells transfected with the miR-7a mimic (Fig. 2J), whereas cells with the miR-7a inhibitor exhibited a significant increase in Klf4 protein levels (Fig. 2K). These findings affirm the direct regulatory role of miR-7a on Klf4, emphasizing its potential as an intrinsic ncRNA regulator of Klf4.
MiR-7a–Klf4 axis regulates AβO-mediated microglial activation and neuroinflammation
Alterations in the miR-7a–Klf4 axis observed under AD conditions prompted an investigation into the role of Klf4 in AD-related pathologies. The miRNA–mRNA pathway network revealed miRNAs regulatory effect on immune cell pathways within the AD mouse brain. Reports have implicated the role of Klf4 in inflammatory pathways, and AβO are known to trigger microglial activation and neuroinflammation. We examined potential involvement of Klf4 in these processes by monitoring its levels following AβO treatment. N9 microglial cells exhibited dose-dependent increases in Klf4 expression at both mRNA and protein levels upon AβO and lipopolysaccharide (LPS) treatment (Fig. 3A–C and Supplementary Fig. S8A). Klf4, as a transcription factor, likely governs the expression of proteins that influence microglial state and proinflammatory mediators. To confirm the role of Klf4 in microglial activation and its modulation, we utilized immunofluorescence with Iba1, an activated microglia marker. Post-transfection with a miR-7a mimic and subsequent priming with LPS (500 ng/ml) and AβO (5 μM) challenge, we observed a reduction in microglial activation, indicated by fewer Iba1 positive cells and altered morphology (Fig. 3D and Supplementary Fig. S8B). Activated microglial cells secrete various cytokines that provide feedback signals to the microglia, resulting in oxidative stress. Specifically, AβO exposure significantly increased intracellular ROS levels, which were mitigated by miR-7a treatment, suggesting a protective effect against ROS production (Fig. 3E).
Figure 3.
MiR-7a–Klf4 axis regulates microglial activation and neuroinflammation. (A) Relative Klf4 mRNA expression by qRT-PCR of N9 cells treated with increasing concentration of AβO (2.5, 5, and 10 μM) (n = 5 replicates). (B) Representative confocal images of Klf4 immunofluorescence in N9 cells treated with increasing concentration of AβO (2.5, 5, and 10 μM) and its quantification (n > 500 cells quantified for each group; scale bar: 200 μm). (C) Quantification of Klf4 protein expression in N9 cells treated with increasing concentration of LPS (100, 250, and 500 ng/ml) by immunofluorescence (n > 500 cells quantified for each group; scale bar: 200 μm). (D) Representative confocal images of Iba1 immunofluorescence for N9 cells transfected with miR-7a mimic and then challenged with AβO (scale bar: 200 μm). (E) Relative intracellular ROS levels detected by DCFDA fluorescence in N9 cells treated with miR-7a mimic and AβO (the bar denotes the normalized mean DCFDA fluorescence detected by well scan with >80 data points (n = 4 biological replicates). (F) qRT-PCR for NF-κB expression of N9 cells treated with miR-7a and AβO (n = 6). (G) Quantification of NF-κB protein expression by immunofluorescence (n > 500 cells quantified for each group). RT-PCR of IL6 and TNFα expression (H) and iNOS (I) of N9 cells treated with miR-7a and AβO (n = 3). (J) Quantification of iNOS protein expression by immunofluorescence (n > 500 cells quantified for each group). (K) Relative NO levels released to media of N9 cells treated with miR-7a and challenged by AβO, quantified by Griess assay (the bar denotes normalized mean absorbance at 540 nm; n = 4 replicates). (L) RT-PCR of NLRP3 expression of LPS primed N9 cells treated with miR-7a and challenged by AβO (n = 3 replicates). (M) Quantification of NLRP3 protein expression by immunofluorescence (n > 500 cells quantified for each group). (N) RT-qPCR of IL1β and Casp1 expression of LPS primed N9 cells treated with miR-7a and challenged AβO (n = 3 and 6 replicates). (O) Schematic of sandwich ELISA and quantification of TNFα, IL6, and IL1β released in the media of N9 cells treated with miR-7a and challenged with AβO [n = 6 biological replicates. Created in BioRender. T, G. (2025), https://BioRender.com/5ikum4c. Data are represented as mean ± s.e.m. All the statistics were performed using one-way ANOVA with Bonferroni post-hoc test *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001].
AβO is known to induce the expression of inflammatory mediators through NF-κB pathway. Further, we explored the regulatory impact of miR-7a–Klf4 axis on the NF-κB levels (Fig. 3F, G, and Supplementary Fig. S8C). MiR-7a-mediated Klf4 silencing resulted in decreased NF-κB expression, corroborated by the downregulation of proinflammatory cytokines IL6 and TNFα (Fig. 3H). We assessed the expression levels of iNOS and found that miR-7a treatment to target Klf4 has effectively reduced iNOS mRNA and protein expression in AβO treated microglial cells (Fig. 3I, J, and Supplementary Fig. S8D). Further determination of relative NO by Griess assay has demonstrated the significant reduction of NO levels (Fig. 3K). We also assessed NLRP3 expression, a key player in neuroinflammatory pathways, and found it significantly elevated following AβO treatment (Fig. 3L, M, and Supplementary Fig. S8E). However, miR-7a treatment substantially decreased NLRP3 expression and its downstream cytokines IL1β and caspase 1 (Fig. 3N). The mitigation of neuroinflammation was further confirmed by quantifying neuroinflammatory mediators TNFα, IL6, and IL1β via sandwich ELISA. The results demonstrated that miR-7a treatment targeting Klf4 effectively inhibited the secretion of proinflammatory mediators induced by AβO (Fig. 3O). Overall, our studies indicate that miR-7a mimic treatment targeting Klf4 significantly reduces microglial activation and attenuates various neuroinflammatory pathways, including NF-κB, iNOS, and NLRP3 inflammasome, thereby alleviating neuroinflammation. The disruption of Klf4-mediated neuroinflammation by miR-7a suggests its potential as an RNA therapeutic targeting neuroinflammation in AD.
MiR-7a–Klf4 axis regulates neuronal ferroptosis
Ferroptosis, characterized by iron-mediated lipid peroxidation and cell death, is implicated in neurodegenerative disease pathogenesis. Klf4, a mediator of apoptotic cell death and regulator of iron homeostasis, led us to hypothesize its role in ferroptosis. We investigated this hypothesis using RSL3-induced ferroptosis in SH-SY5Y neuroblastoma cells, to understand the role of miR-7a–Klf4 axis. RSL3 inhibits GPX4, essential for ferroptosis, causing neuronal cell death. Upon ferroptosis induction by RSL3, Klf4 mRNA, and protein levels significantly increased, suggesting the regulatory role of Klf4 in ferroptosis (Fig. 4A and B). As Klf4 being transcription factor involved in various biological processes, its upregulation potentially mediate ferroptotic cell death. If Klf4 influences ferroptosis, its suppression by miR-7a mimic treatment could protect cells from ferroptotic cell death. To confirm this, we assessed different hallmarks of ferroptosis in RSL3-induced conditions and treatment with a miR-7a mimic. Neuronal cells were transfected with a miR-7a mimic, then challenged with RSL3 and analyzed for ferroptosis hallmarks to observe the rescue effect.
Figure 4.
MiR-7a–Klf4 axis regulates neuronal ferroptosis. (A) Relative Klf4 mRNA expression by qRT-PCR of SH-SY5Y cells treated with increasing concentration of RSL3 (0.5, 1, and 2.5 μM) (n = 6 replicates). (B) Representative confocal images Klf4 immunofluorescence in SH-SY5Y cells treated with increasing concentration of RSL3 (0.5, 1, and 2.5 μM) and its quantification (n > 500 cells quantified for each group; scale bar: 200 μm). (C) Quantification of GPX4 protein expression by immunofluorescence (n > 500 cells quantified for each group). (D) Fluorescence images of C11- Bodipy lipid peroxidation probe (scale bar: 25 μm). (E) Relative Fe+3 levels quantified by Calcein-AM fluorescence imaging (n > 500 cells quantified for each group). Representative confocal images of Nrf2 immunofluorescence (F) and quantification of its translocation to nucleus (G) (the bar denotes mean of nuclear/cytoplasmic fluorescence. n > 500 cells quantified for each group). (H) Mitochondrial imaging using Mitotracker green (scale bar: 25 μm). (I) Relative MMP quantified using Rho123 potential dependent fluorescent dye (normalized mean Rho123 fluorescence detected by well scan with > 80 data points. n = 5 replicates). (J) Cell viability quantified by MTT assay (n = 8 biological replicates). Data are represented as mean ± s.e.m. All the statistics were performed using one-way ANOVA with Bonferroni post-hoc test, *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001).
We assessed GPX4 levels, a marker known to reduce during RSL3-induced ferroptotic cell death, via immunofluorescence assay. Results revealed a significant reduction in GPX4 levels following RSL3 treatment, which was dose-dependently rescued in cells treated with a miR-7a mimic (Fig. 4C and Supplementary Fig. S9A). The lipid peroxidation, a ferroptosis hallmark, was monitored using a red fluorescent C11-Bodipy probe, which emits green fluorescence upon oxidation by lipid peroxides. The BODIPY dye fluorescence imaging showed increased lipid peroxide levels in RSL3-treated cells, while miR-7a treatment prevents lipid peroxidation (Fig. 4D and Supplementary Fig. S9B). Quantitative analysis in well plate assay confirmed a significant rise in lipid peroxidation, reduced significantly by miR-7a treatment. Ferroptosis is initiated by an intracellular LIP through Fenton-type reactions causing lipid peroxidation. Employing calcein-AM fluorescence imaging, we analyzed LIP under ferroptosis condition and its mitigation by miR-7a mimics. A marked reduction in calcein-AM fluorescence indicated increased LIP in RSL3-treated cells, whereas miR-7a treated cells retained fluorescence, implying inhibition of iron overload (Fig. 4E and Supplementary Fig. S9C). Iron-triggered Fenton-type reactions generate ROS, leading to oxidative stress and Nrf2 activation. Measuring intracellular ROS using 2′-7′-dichlorodihydrofluorescein diacetate (DCFDA), we found miR-7a significantly reduced ROS in RSL3-challenged neuronal cells (Supplementary Fig. S9D). Under oxidative stress, Nrf2 translocate to the nucleus, inducing antioxidant gene expression. The reduction of LIP and ROS, coupled with GPX4 rescue by miR-7a, potentially prevents oxidative stress and Nrf2 activation (Fig. 4F, G, and Supplementary Fig. S9E). Excessive oxidative stress and lipid peroxidation can cause mitochondrial damage in ferroptosis. We evaluated the effect of miR-7a-mediated Klf4 silencing in preventing mitochondrial structural and functional damage. Mitotracker green imaging revealed mitochondrial fragmentation, as observed by the fragmented and bulged mitochondria in ferroptotic cells, while miR-7a-treated cells retained normal long tubular mitochondrial morphology, indicating prevention of mitochondrial damage (Fig. 4H). To ascertain functional impairment, we quantified the relative MMP utilizing rhodamine 123 (Rho123), a fluorescent probe whose signal intensity is dependent on MMP. In the context of RSL3-induced ferroptosis, the membrane potential reduced to 75% compared to the control. Contrastingly, miR-7a-treated cells exhibited a restoration of membrane potential in a dose-dependent manner, achieving full recovery at a concentration of 250 nM (Fig. 4I).
The attenuation of Klf4 expression via miR-7a mimic intervention was instrumental in reversing the hallmarks of ferroptosis, underscoring the pivotal role of Klf4 in neuronal ferroptosis. Encouraged by the protective effect of miR-7a, we proceeded to examine its efficacy in safeguarding neuronal cells from ferroptotic death. The cell viability assays showed a significant amelioration of ferroptotic cell death attributed to miR-7a-facilitated Klf4 suppression (Fig. 4J). These findings highlight the significant role of Klf4 in ferroptosis and demonstrate the therapeutic promise of miR-7a as an RNA-based agent targeting ferroptosis in AD and other neurodegenerative conditions.
Pharmacological targeting of miR-7a–Klf4 axis using small molecules to rescue neuroinflammation and ferroptosis
The delivery of RNA therapeutics to the brain for treating CNS disorders is a formidable challenge, emphasizing the necessity for crafting small molecule modulators that can precisely manipulate the regulatory axis to enhance therapeutic outcomes. The strategic inhibition of Klf4 by miR-7a has demonstrated efficacy in mitigating microglial activation, neuroinflammation, and ferroptotic neuronal death, which are pivotal in the AD pathology. Development of RNA therapeutics or small molecules that can modulate the miR-7a–Klf4 axis either by bolstering endogenous miR-7a levels or by directly targeting Klf4 mRNA and protein represents a promising therapeutic avenue for AD and other neurodegenerative disorders (Fig. 5A).
Figure 5.
Pharmacological targeting of miR-7a–Klf4 axis rescue neuroinflammation and ferroptosis. (A) Schematic representation of chemical targeting of miR-7a–Klf4 axis to curb neuroinflammation and ferroptosis [Created in BioRender. T, G. (2025) https://BioRender.com/adyeblf]. (B) Chemical structures of small molecules used to modulate miR-7a–Klf4 axis. (C) N9 microglial cell rescue from AβO induced inflammatory cell death by honokiol (HO) in dose-dependent manner (n = 8). (D) SH-SY5Y neuronal cell rescue from RSL3 induced ferroptotic cell death by honokiol in dose dependent manner (n = 8). (E) Representative confocal images of Iba1 immunofluorescence of N9 cells treated with AβO and honokiol (scale bar: 200 μm). (F) Relative Klf4 mRNA expression by qRT-PCR of SH-SY5Y cells treated with honokiol (2.5 μM) (n = 5, unpaired, two-tailedt-test, P < 0.05). (G) Representative confocal images of Klf4 immunofluorescence in SH-SY5Y cells treated with honokiol (2.5 μM) and its quantification (n > 500 cells quantified for each group; scale bar: 200 μm). (H) Schematic showing the protocol for mouse primary neuron-glia mixed culture [Created in BioRender. T, G. (2025) https://BioRender.com/vvwzz98]. (I) Representative confocal images of primary neuron-glia mixed culture treated with AβO (2.5 μM) alone and in presence of honokiol (2.5 and 5 μM) and probed for NeuN and Iba1 markers for neuronal cells and activated microglial cells (scale bar: 200 μm). (J) Cell viability of primary neuron-glia mixed culture treated with AβO (2.5 μM) and honokiol (2.5 and 5 μM) measured by MTT assay (n = 4). Data are represented as mean ± s.e.m. Statistics used for (F) and (G) is unpaired, two-tailed t-test P < 0.05. For (C), (D), and (J), one-way ANOVA with Bonferroni post-hoc test, *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.
A few small molecules have been reported to elevate miR-7a levels and suppress Klf4 expression. We evaluated such molecules namely nicorandil, quercetin, and honokiol, potentially modulate the miR-7a–Klf4 axis (Fig. 5B). We assessed their efficacy in mitigating neuroinflammatory and ferroptotic cell death via cell viability assays. Notably, honokiol demonstrated a concentration-dependent effect in rescuing cells from neuroinflammation and ferroptosis compared to others (Fig. 5C and D, and Supplementary Fig. S10A–D). The BBB permeability of honokiol and its minimal cytotoxicity at concentrations up to 20 μM in SH-SY5Y cells and 10 μM in N9 cells underscore its therapeutic potential (Supplementary Fig. S10E and F). Honokiol was effective at lower concentrations in alleviating neuroinflammation and ferroptosis. Further honokiol effectively rescued N9 microglial cells from AβO-mediated activation as evident from Iba1 immunofluorescence (Fig. 5E). The expression analysis of Klf4 mRNA and protein revealed that honokiol significantly reduced the Klf4 expression at both levels (Fig. 5F and G). Targeting Klf4 rescued neuroinflammation and ferroptosis, that suggest Klf4 as a potential therapeutic target for AD and other neurodegenerative disorders.
Honokiol is a biphenolic natural product that displays neuroprotective and anti-inflammatory activity with effective BBB permeability. Structurally, honokiol is a good radical scavenging agent and prevents oxidative stress. It is reported to target Klf4 in inflammatory conditions and exhibit rescue benefits. The rescue studies in our work demonstrate effective reduction of inflammation and ferroptosis at low concentrations (250 nM to 5 μM). To confirm the rescue effect of honokiol on neuroinflammation and neuronal death, we have employed mice primary neuron-glia mixed culture model that mimics in vivo conditions. We have isolated cells from day 1 pup (P1) brain and cultured for 8 days followed by treatments (Fig. 5H). Primary mixed cell culture was primed with LPS (250 ng/ml) for 4 h and then challenged with AβO (2.5 μM) for 24 h. Cells were subjected to NeuN and Iba1 immunostaining to probe neuron and activated microglial cells, respectively. The confocal images show the activation of microglia in AβO treatment, which was decreased by honokiol treatment (Fig. 5I). The activated microglia have resulted in decreased neuronal population indicating the neuronal death that was rescued by honokiol. We have performed cell viability assay to confirm the rescue of AβO induced primary neuronal cell death by honokiol. The results show concentration dependent rescue of cell death suggesting its therapeutic effect (Fig. 5J). Further research is necessary to explore the potential of honokiol and its derivatives to effectively modulate AD and other neurodegenerative conditions.
Discussion
AD is a complex neurodegenerative disorder and the key regulatory mechanisms driving its onset and progression are not fully understood. As upstream transcription regulators, miRNA target multiple mRNAs across cellular pathways, orchestrating cellular homeostasis and disease pathways. The multitargeting capability of miRNAs establishes a miRNA–mRNA regulatory network, enabling the precise regulation of cellular pathways. Despite the critical regulatory role of miRNAs, their involvement in AD and the potential miRNA–mRNA regulatory networks remain underexplored. The failure to develop effective therapeutics for AD underscores the need to decipher the miRNA–mRNA pathway network implicated in AD pathology, which could reveal new diagnostic and therapeutic targets. In this study, we utilized an APP/PSEN1 Tg AD mouse model to profile miRNA expression alterations in the AD brain, identifying significant changes in the cortex and hippocampus. Independent clustering of miRNA profiles from WT and AD mouse brains highlighted a substantial dysregulation, with 21 and 11 miRNAs significantly altered in the cortex and hippocampus, respectively. This transcriptomic landscape revealed both upregulated and downregulated miRNAs, reprogramming the miRNA–mRNA pathway network and contributing to AD pathology. By constructing a miRNA–mRNA pathway and disease network based on validated mRNA target predictions, we observed that miRNAs regulate various altered disease pathways in AD. Modulating the miRNA–mRNA axis may offer a strategy to reprogram this network, leading to improved disease-modifying treatments. GO and functional over-representation analyses identified pathways related to stress response, immune function, cell death and cycle, neuronal and synaptic regulation, protein homeostasis, and other processes linked to CNS functions, innate immunity, and neurodegeneration. Further examination of the miRNA–mRNA disease network revealed the influence of altered miRNAs in various CNS and neurodegenerative disorders, including AD. Our analysis underscores the significance of the miRNA regulatory network in disease pathology and neuroprotective role of miRNAs in AD and other neurodegenerative diseases.
The transcriptomic analysis, corroborated by qPCR experiments, indicates an upregulation of miR-7a in the cortex and hippocampus of AD mouse brain. MiR-7a, an abundantly expressed miRNA in the brain and evolutionarily conserved, is recognized for its neuroprotective properties and influence on CNS functions [40]. It targets mRNAs associated with AD and Parkinson disease (PD), such as insulin-degrading enzyme (IDE) and a-Syn [41, 42]. Reports suggest the regulatory effect of miR-7a on Klf4, a pleotropic transcription factor implicated in cancer and AD. Elevated Klf4 expression in AD brains suggests its potential involvement in disease onset and progression. We propose that elevated Klf4 in AD brains serves as a principal transcriptional regulator of AD-related pathological pathways and miR-7a targeting it for neuroprotection. MiR-7a expression aims to counterbalance overexpression of Klf4 in AD, establishing a miR-7a–Klf4 axis that plays role in neurodegeneration. We have shown the direct interaction of miR-7a with Klf4 at the 3′ UTR region via CD and fluorescence spectroscopy. Klf4 mRNA contains two miR-7a binding sites, one conserved among mammals and another across vertebrates [40]. The inverse expression patterns of miR-7a and Klf4 mRNA in SH-SY5Y neuroblastoma and N9 microglial cells imply the direct regulatory role of miR-7a. Transfection assays with miR-7a mimic and inhibitor affirm the regulation of Klf4 by miR-7a and the significance of miR-7a–Klf4 axis in AD pathology. The role of the miR-7a–Klf4 axis has been studied in cancer cells and its role in regulating various pathways linked to cancer development, progression and metastasis, and resistance [43–46]. The direct regulation of Klf4 by miR-7a and its role in controlling various pathways and cancer pathobiology suggest role of miR-7a–Klf4 axis in neurons and microglial pathology linked to AD and other neurodegenerative disorders.
The miRNA–mRNA pathway analysis has revealed alterations in immune cell pathways within the brain affected by AD. Notably, Klf4 has been implicated in microglial activation and neuroinflammation. Our study demonstrated that Klf4 expression is induced at both mRNA and protein levels in LPS and AβO treated microglial cells. The activation states of microglia and various neuroinflammatory pathways are ameliorated when cells are transfected with miR-7a mimic and challenged with AβO. These findings underscore the pivotal role of the miR-7a–Klf4 axis in microglial activation and neuroinflammation, particularly in regulating NF-κB, iNOS, and NLRP3 pathways. There are reports suggesting the induction of Klf4 expression in LPS and AβO treated microglial cells [47–49]. Elevated Klf4 expression has been observed in activated microglial cells, both in vitro and in vivo, using the 5xFAD mouse model [50]. However, there is a dearth of information regarding the intrinsic regulator of Klf4 under neuroinflammatory conditions and the specific downstream pathways it regulates. Our observations indicate that Klf4 targets NF-κB expression and downstream proinflammatory mediators such as TNFα and IL6. Similar observations were reported earlier wherein Klf4 induces NF-κB signaling [49] and Klf4 directly interacts with NF-κB and regulates its activity [51]. Contradictory reports exist, with some suggesting that Klf4 is downregulated in activated glial cells exerts an anti-inflammatory effect by suppressing the NF-κB pathway in macrophages, endothelial, microglial cells and astrocytes [52–56]. The inhibition of MEG3, a long non-coding RNA (lncRNA)-mediated suppression of Klf4 induced inflammation in microglial cells of cerebral ischemia–reperfusion injury model [54]. It is noteworthy that Klf4, as a pleiotropic transcription factor, can regulate NF-κB signaling both positively and negatively, depending on the cell type and disease context. Klf4 directly regulates iNOS expression, leading to an increase in NO production [47]. Suppression of Klf4 reduces NLRP3 expression and downstream mediators such as IL1β and caspase 1 [57]. Additionally, NF-κB is also known to regulate NLRP3 expression and Klf4 regulates both NF-κB and NLRP3 that accounts for significant reduction of NLRP3 and downstream IL1β and Caspase 1 levels. NLRP3 is a direct target for miR-7a and inhibited microglial NLRP3 inflammasome activation in PD model [58]. Further, Klf4 binds to the NLRP3 promoter and positively regulates its expression. Other miRNAs, including miR-146a-5p and miR-92a, have been reported to target Klf4, suggesting the regulatory role of the miR–Klf4 axis in neuroinflammation [59, 60].
Klf4 is recognized for its key role in cell death mechanisms, including apoptosis, oxidative stress, and iron homeostasis [61]. The miR–mRNA pathway analysis indicates significant alterations in cellular stress responses, as well as cell cycle and death pathways. We have observed a marked induction of Klf4 in neuronal ferroptosis and upregulation of Klf4 appears to orchestrate several processes involved in ferroptosis. Inhibition of Klf4 in ferroptotic cells using miR-7a mimics leads to a reduction in ferroptotic markers such as Fe3+, GPX4, lipid peroxidation, oxidative stress, Nrf2 signaling, and mitochondrial damage, thereby enhancing cell survival. Previous studies have highlighted the role of the miR-7a–Klf4 axis in regulating oxidative stress and apoptosis mediated neuronal cell death [62, 63]. Our findings reveal that miR-7a mimic treatment reduces ROS and oxidative stress, suggesting the involvement of Klf4 in redox homeostasis. It also prevents the accumulation of Fe3+ in neuronal cells challenged with RSL3, possibly through the regulation of heme uptake mediated by heme carrier protein 1 (HCP1) [64]. Moreover, miR-7-5p has been reported to downregulate the expression of iron-related genes such as transferrin receptor (TfR), ferritin, and iron regulatory protein 2 (IRP2), as well as ferroptosis markers like lipid peroxidation (ALOX12) [65]. Our study reveals that miR-7a modulates iron levels and lipid peroxidation in ferroptosis by targeting Klf4, suggesting that the miR-7a–Klf4 axis regulates ferroptosis by regulating LIP, ROS-oxidative stress, mitochondrial integrity, and lipid peroxidation in neuronal cells.
Investigations into the involvement of the miR-7a–Klf4 axis in neuroinflammation and ferroptosis have led us to identification of small molecule modulators to counteract these pathological pathways. Identification of small molecules that potentially induce miR-7a is one of the approaches to target miR7a–Klf4 axis [66]. Nicorandil has been shown to significantly increase miR-7a levels, offering relief from astrocytic inflammation [67]. Honokiol, on the other hand, diminishes Klf4 protein levels and NF-κB, curtailing the expression of iNOS, Cox-2, MCP-1, IL-6, and TNF-α in both LPS-treated microglia and mouse brains [68]. Similarly, quercetin targets Klf4, safeguarding SH-SY5Y cells from H2O2-induced oxidative stress and apoptosis [62]. We employed these molecules to pharmacologically modulate the miR-7a–Klf4 axis, assessing their effects on neuroinflammatory-driven microglial cell death and RSL3-induced neuronal ferroptosis. Our results indicate that quercetin effectively alleviates neuroinflammation, while honokiol rescues cells from both neuroinflammation and ferroptosis at lower doses due to its ability to reduce Klf4 expression. Notably, higher concentrations of honokiol have been associated with the induction of ferroptosis, a phenomenon warranting further exploration to understand its concentration-dependent dual role [69, 70]. Comprehensive study evaluating the therapeutic potential of honokiol in mouse primary mixed culture have confirmed the role of miR-7a–Klf4 axis in neuroinflammation and neurodegeneration. Studies have demonstrated the beneficial effects of honokiol by ameliorating oxidative stress, neuroinflammation, apoptosis, and ferroptosis cell death in AD and cancer models [71, 72]. The evaluation of Klf4 expression at the mRNA and protein levels in cells treated with honokiol in our study demonstrated a significant decrease, suggesting that Klf4 is one of the targets for honokiol. Earlier reports indicated that honokiol targets Klf4 protein, possibly by direct binding or by regulating its expression or activation, which requires further investigation [68]. Klf4 is known to regulate its expression at the transcription level by negative autoregulation, potentially resulting in decreased Klf4 mRNA levels [73]. Other possibilities include direct binding of honokiol to Klf4 mRNA, resulting in structural modification and directing it for degradation, which leads to decreased mRNA levels [74, 75]. The promising results in our study suggest further in vivo studies to evaluate the therapeutic effects of honokiol in alleviating neuroinflammation and ferroptosis to treat AD and other neurodegenerative conditions. Detailed preclinical studies and lead optimization potentially warrant this molecule for clinical evaluation as a drug candidate for AD. Recent studies suggest that neuroinflammation is a major etiopathology in various neurodegenerative conditions and that ferroptosis is linked to AD and potentially plays a role in other CNS diseases. Further detailed study of the role of the miR-7a-Klf4 axis in various other neurodegenerative diseases could potentially uncover new therapeutic targets and opportunities for drug development. Identifying and assessing new miRNA and small molecule therapeutics that modulate this axis is crucial for developing improved treatments for AD and other neurodegenerative disorders [66, 76]. The recent approvals of various RNA therapeutics and their safety profiles in clinical studies suggest further exploration for the development of miRNA-based therapeutics. Small molecules targeting the miR-7a–Klf4 axis modulate neuroinflammation, oxidative stress, and ferroptosis, serving as multipronged approaches for managing complex multifactorial AD. These findings highlight the potential for novel therapeutic interventions that target miR-7a–Klf4 axis, enhancing the diseases management strategies.
In summary our study has unveiled a spectrum of miRNAs that exhibit altered expression patterns in AD, leading to the establishment of a reprogrammed miRNA–mRNA pathway network pertinent to AD. Notably, we have discovered an elevated level of Klf4 and induction of miR-7a in the AD brain to target Klf4 for neuroprotection. The miR-7a-mediated attenuation of Klf4 elucidates its pivotal role in orchestrating neuroinflammation via diverse molecular conduits, including NF-κB signaling, iNOS, and the NLRP3 inflammasome cascade. Furthermore, our study highlighted the significant role of the miR-7a–Klf4 axis in ferroptosis-a distinct modality of cell death characterized by LIP overload, lipid peroxidation, ROS, oxidative stress, and mitochondrial impairment (Fig. 6). The strategic pharmacological manipulation of the miR-7a–Klf4 axis utilizing small molecule has demonstrated efficacy in ameliorating neuroinflammation and ferroptosis, positing the miR-7a–Klf4 axis as a viable pharmacological target for AD and potentially other neurodegenerative disorders. These insights pave the way for the inception of novel therapeutic modalities aimed at mitigating the impact of these chronic diseases.
Figure 6.
Schematic representation of miR-7a–Klf4 axis regulating neuroinflammation and ferroptosis resulting in neurodegeneration through various pathways and its targeting by using miR7a and honokiol [Created in BioRender. T, G. (2025), https://BioRender.com/dzh3fu8].
Supplementary Material
Acknowledgements
We thank Theracues Innovations Pvt Ltd, Bangalore, India, a service provider for NanoString platforms and for Bioinformatic analysis. Graphical abstract Created in BioRender. T, G. (2025) https://BioRender.com/mbq53tp. We thank Shivprasad P.S. for help in confocal imaging, lab members and Prathamesh for discussion. M.R. thanks UGC and JNCASR for the student fellowships.
Author contributions: Madhu Ramesh (Conceptualization [equal], Data curation [lead], Formal Analysis [equal], Investigation [lead], Methodology [equal], Validation [equal], Writing—original draft [equal], Writing—review & editing [equal]), and T. Govindaraju (Conceptualization [equal], Formal Analysis [equal], Funding acquisition [lead], Methodology [equal], Project administration [lead], Resources [lead], Supervision [lead], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]).
Contributor Information
Madhu Ramesh, Bioorganic Chemistry Laboratory, New Chemistry Unit, Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Jakkur P.O., Bengaluru 560064 Karnataka, India.
Thimmaiah Govindaraju, Bioorganic Chemistry Laboratory, New Chemistry Unit, Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Jakkur P.O., Bengaluru 560064 Karnataka, India.
Supplementary data
Supplementary data is available at NAR Molecular Medicine online.
Conflict of interest
All the authors declare they have no competing interests.
Funding
The authors thank JNCASR, DST (CEFPRA grant: IFCPAR/CEFIPRA-62T10-3), core grant (CRG/2020/004594), Science and Engineering Research Board (SERB), New Delhi, India, for the funding.
Data availability
NanoString miRNA transcriptomic data is deposited in NCBI GEO repository with accession number GSE271276. All data supporting the findings of this study are available within the paper and Supplementary files. All other data generated in this study are available from the corresponding author on reasonable request.
References
- 1.Blennow K, de Leon MJ, Zetterberg H. Alzheimer’s disease. The Lancet. 2006; 368:387–403. 10.1016/S0140-6736(06)69113-7. [DOI] [PubMed] [Google Scholar]
- 2.Knopman DS, Amieva H, Petersen RCet al.. Alzheimer disease. Nat Rev Dis Primers. 2021; 7:33. 10.1038/s41572-021-00269-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Long JM, Holtzman DM. Alzheimer disease: an update on pathobiology and treatment strategies. Cell. 2019; 179:312–39. 10.1016/j.cell.2019.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cummings J, Zhou Y, Lee Get al.. Alzheimer’s disease drug development pipeline: 2023. A&D Transl Res Clin Interv. 2023; 9:e12385. 10.1002/trc2.12385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Govindaraju T Alzheimer's Disease: Recent Finding in Pathophysiology, Diagnostic and Therapeutic Modalities. 2022; London: Royal Society of Chemistry; 10.1039/9781839162732. [DOI] [Google Scholar]
- 6.Heneka MT, Carson MJ, El Khoury Jet al.. Neuroinflammation in Alzheimer’s disease. Lancet Neurol. 2015; 14:388–405. 10.1016/S1474-4422(15)70016-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hickman S, Izzy S, Sen Pet al.. Microglia in neurodegeneration. Nat Neurosci. 2018; 21:1359–69. 10.1038/s41593-018-0242-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hansen DV, Hanson JE, Sheng M. Microglia in Alzheimer’s disease. J Cell Biol. 2018; 217:459–72. 10.1083/jcb.201709069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Deczkowska A, Keren-Shaul H, Weiner Aet al.. Disease-associated microglia: a universal immune sensor of neurodegeneration. Cell. 2018; 173:1073–81. 10.1016/j.cell.2018.05.003. [DOI] [PubMed] [Google Scholar]
- 10.Sarlus H, Heneka MT. Microglia in Alzheimer’s disease. J Clin Investig. 2017; 127:3240–9. 10.1172/JCI90606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Leng F, Edison P. Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here?. Nat Rev Neurol. 2021; 17:157–72. 10.1038/s41582-020-00435-y. [DOI] [PubMed] [Google Scholar]
- 12.Fontana IC, Zimmer AR, Rocha ASet al.. Amyloid-β oligomers in cellular models of Alzheimer’s disease. J Neurochem. 2020; 155:348–69. 10.1111/jnc.15030. [DOI] [PubMed] [Google Scholar]
- 13.Gao C, Jiang J, Tan Yet al.. Microglia in neurodegenerative diseases: mechanism and potential therapeutic targets. Sig Transduct Target Ther. 2023; 8:359. 10.1038/s41392-023-01588-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dixon SJ, Lemberg KM, Lamprecht MRet al.. Ferroptosis: an iron-dependent form of nonapoptotic cell death. Cell. 2012; 149:1060–72. 10.1016/j.cell.2012.03.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tan S, Schubert D, Maher P. Oxytosis: a novel form of programmed cell death. Curr Top Med Chem. 2001; 1:497–506. [DOI] [PubMed] [Google Scholar]
- 16.Jiang X, Stockwell BR, Conrad M. Ferroptosis: mechanisms, biology and role in disease. Nat Rev Mol Cell Biol. 2021; 22:266–82. 10.1038/s41580-020-00324-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Padhi D, Baruah P, Ramesh Met al.. Hybrid molecules synergistically mitigate ferroptosis and amyloid-associated toxicities in Alzheimer’s disease. Redox Biol. 2024; 71:103119. 10.1016/j.redox.2024.103119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Baruah P, Moorthy H, Ramesh Met al.. A natural polyphenol activates and enhances GPX4 to mitigate amyloid-β induced ferroptosis in Alzheimer’s disease. Chem Sci. 2023; 14:9427–38. 10.1039/D3SC02350H. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dai E, Chen X, Linkermann Aet al.. A guideline on the molecular ecosystem regulating ferroptosis. Nat Cell Biol. 2024; 26:1447–57. 10.1038/s41556-024-01360-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Smith MA, Harris PL, Sayre LMet al.. Iron accumulation in Alzheimer disease is a source of redox-generated free radicals. Proc Natl Acad Sci USA. 1997; 94:9866–8. 10.1073/pnas.94.18.9866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hambright WS, Fonseca RS, Chen Let al.. Ablation of ferroptosis regulator glutathione peroxidase 4 in forebrain neurons promotes cognitive impairment and neurodegeneration. Redox Biol. 2017; 12:8–17. 10.1016/j.redox.2017.01.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bao WD, Pang P, Zhou XTet al.. Loss of ferroportin induces memory impairment by promoting ferroptosis in Alzheimer’s disease. Cell Death Differ. 2021; 28:1548–62. 10.1038/s41418-020-00685-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Greenough MA, Lane DJR, Balez Ret al.. Selective ferroptosis vulnerability due to familial Alzheimer’s disease presenilin mutations. Cell Death Differ. 2022; 29:2123–36. 10.1038/s41418-022-01003-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Adeniyi PA, Gong X, MacGregor Eet al.. Ferroptosis of microglia in aging human white matter injury. Ann Neurol. 2023; 94:1048–66. 10.1002/ana.26770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sui X, Zhang R, Liu Set al.. RSL3 Drives ferroptosis through GPX4 inactivation and ROS production in colorectal cancer. Front Pharmacol. 2018; 9:1371. 10.3389/fphar.2018.01371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Huang L, McClatchy DB, Maher Pet al.. Intracellular amyloid toxicity induces oxytosis/ferroptosis regulated cell death. Cell Death Dis. 2020; 11:828. 10.1038/s41419-020-03020-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bartel DP MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004; 116:281–97. 10.1016/S0092-8674(04)00045-5. [DOI] [PubMed] [Google Scholar]
- 28.Lau P, Bossers K, Janky Ret al.. Alteration of the microRNA network during the progression of Alzheimer’s disease. EMBO Mol Med. 2013; 5:1613–34. 10.1002/emmm.201201974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Islam MR, Kaurani L, Berulava Tet al.. A microRNA signature that correlates with cognition and is a target against cognitive decline. EMBO Mol Med. 2021; 13:e13659. 10.15252/emmm.202013659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Walgrave H, Zhou L, De Strooper Bet al.. The promise of microRNA-based therapies in Alzheimer’s disease: challenges and perspectives. Mol Neurodegeneration. 2021; 16:76. 10.1186/s13024-021-00496-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Shang R, Lee S, Senavirathne Get al.. MicroRNAs in action: biogenesis, function and regulation. Nat Rev Genet. 2023; 24:816–33. 10.1038/s41576-023-00611-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Metsalu T, Vilo J. ClustVis: a web tool for visualizing clustering of multivariate data using principal component analysis and heatmap. Nucleic Acids Res. 2015; 43:W566–70. 10.1093/nar/gkv468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Wu T, Hu E, Xu Set al.. ClusterProfiler 4.0: a universal enrichment tool for interpreting omics data. Innovation. 2021; 2:100141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Yu G, He QY. ReactomePA: an R/bioconductor package for reactome pathway analysis and visualization. Mol BioSyst. 2016; 12:477–9. 10.1039/C5MB00663E. [DOI] [PubMed] [Google Scholar]
- 35.Sherman BT, Hao M, Qiu Jet al.. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022; 50:W216–21. 10.1093/nar/gkac194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Shannon P, Markiel A, Ozier Oet al.. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003; 13:2498–504. 10.1101/gr.1239303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Piñero J, Saüch J, Sanz Fet al.. The DisGeNET cytoscape app: exploring and visualizing disease genomics data. Comput Struct Biotechnol J. 2021; 19:2960–7. 10.1016/j.csbj.2021.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Busk PK A tool for design of primers for microRNA-specific quantitative RT-qPCR. BMC Bioinf. 2014; 15:29. 10.1186/1471-2105-15-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Chen SH, Oyarzabal EA, Hong JS. Preparation of rodent primary cultures for neuron-glia, mixed glia, enriched microglia, and reconstituted cultures with microglia. Methods Mol Biol. 2013; 1041:231–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zhao J, Zhou Y, Guo Met al.. MicroRNA-7: expression and function in brain physiological and pathological processes. Cell Biosci. 2020; 10:77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fernández-de Frutos M, Galán-Chilet I, Goedeke Let al.. MicroRNA-7 impairs insulin signaling and regulates aβ levels through posttranscriptional regulation of the insulin receptor substrate 2, insulin receptor, insulin-degrading enzyme, and liver X receptor pathway. Mol Cell Biol. 2019; 39:e00170-19. 10.1128/MCB.00170-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.McMillan KJ, Murray TK, Bengoa-Vergniory Net al.. Loss of MicroRNA-7 regulation leads to α-synuclein accumulation and dopaminergic neuronal loss in vivo. Mol Ther. 2017; 25:2404–14. 10.1016/j.ymthe.2017.08.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Okuda H, Xing F, Pandey PRet al.. MiR-7 suppresses brain metastasis of breast cancer stem-like cells by modulating KLF4. Cancer Res. 2013; 73:1434–44. 10.1158/0008-5472.CAN-12-2037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Chang YL, Zhou PJ, Wei Let al.. MicroRNA-7 inhibits the stemness of prostate cancer stem-like cells and tumorigenesis by repressing KLF4/PI3K/akt/p21 pathway. Oncotarget. 2015; 6:24017–31. 10.18632/oncotarget.4447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Meza-Sosa KF, Pérez-García EI, Camacho-Concha Net al.. MiR-7 promotes epithelial cell transformation by targeting the tumor suppressor KLF4. PLoS One. 2014; 9:e103987. 10.1371/journal.pone.0103987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Morales-Martinez M, Vega GG, Neri Net al.. MicroRNA-7 regulates migration and chemoresistance in non-hodgkin lymphoma cells through regulation of KLF4 and YY1. Front Oncol. 2020; 10:588893. 10.3389/fonc.2020.588893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Kaushik DK, Thounaojam MC, Kumawat KLet al.. Interleukin-1β orchestrates underlying inflammatory responses in microglia via Krüppel-like factor 4. J Neurochem. 2013; 127:233–44. 10.1111/jnc.12382. [DOI] [PubMed] [Google Scholar]
- 48.Li L, Zi X, Hou Det al.. Krüppel-like factor 4 regulates amyloid-β (Aβ)-induced neuroinflammation in Alzheimer's disease. Neurosci Lett. 2017; 643:131–7. 10.1016/j.neulet.2017.02.017. [DOI] [PubMed] [Google Scholar]
- 49.You Z, Yang Z, Cao Set al.. The novel KLF4/BIG1 regulates LPS-mediated neuro-inflammation and migration in BV2 cells via PI3K/Akt/NF-kB signaling pathway. Neuroscience. 2022; 488:102–11. 10.1016/j.neuroscience.2022.01.014. [DOI] [PubMed] [Google Scholar]
- 50.Ennerfelt H, Holliday C, Shapiro DAet al.. CARD9 attenuates aβ pathology and modifies microglial responses in an Alzheimer’s disease mouse model. Proc Natl Acad Sci USA. 2023; 120:e2303760120. 10.1073/pnas.2303760120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Shaverdashvili K, Padlo J, Weinblatt D.et al.. KLF4 activates NF-κb signaling and esophageal epithelial inflammation via the Rho-related GTP-binding protein RHOF. PLoS One. 2019; 14:e0215746. 10.1371/journal.pone.0215746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Kaushik DK, Gupta M, Das Set al.. Krüppel-like factor 4, a novel transcription factor regulates microglial activation and subsequent neuroinflammation. J Neuroinflammation. 2010; 7:68. 10.1186/1742-2094-7-68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Hamik A, Lin Z, Kumar Aet al.. Kruppel-like factor 4 regulates endothelial inflammation. J Biol Chem. 2007; 282:13769–79. 10.1074/jbc.M700078200. [DOI] [PubMed] [Google Scholar]
- 54.Li Z, Jia Y, Han Set al.. Klf4 alleviates lipopolysaccharide-induced inflammation by inducing expression of MCP-1 induced protein 1 to deubiquitinate TRAF6. Cell Physiol Biochem. 2018; 47:2278–90. 10.1159/000491538. [DOI] [PubMed] [Google Scholar]
- 55.Zhang X, Wang L, Han Zet al.. KLF4 alleviates cerebral vascular injury by ameliorating vascular endothelial inflammation and regulating tight junction protein expression following ischemic stroke. J Neuroinflammation. 2020; 17:107. 10.1186/s12974-020-01780-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Wang C, Li L. The critical role of KLF4 in regulating the activation of A1/A2 reactive astrocytes following ischemic stroke. J Neuroinflammation. 2023; 20:44. 10.1186/s12974-023-02742-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Yao J, Kong Q, Wang Yet al.. Mechanism of Kruppel-Like factor 4 in pyroptosis of nasal mucosal epithelial cells in mice with allergic rhinitis. Am J Rhinol Allergy. 2023; 37:337–47. 10.1177/19458924221148568. [DOI] [PubMed] [Google Scholar]
- 58.Zhou Y, Lu M, Du RHet al.. MicroRNA-7 targets nod-like receptor protein 3 inflammasome to modulate neuroinflammation in the pathogenesis of Parkinson’s disease. Mol Neurodegeneration. 2016; 11:28. 10.1186/s13024-016-0094-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Fan C, Li Y, Lan Tet al.. Microglia secrete miR-146a-5p-containing exosomes to regulate neurogenesis in depression. Mol Ther. 2022; 30:1300–14. 10.1016/j.ymthe.2021.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Wen Z, Lin S, Li Cet al.. MiR-92a/KLF4/p110δ regulates titanium particles-induced macrophages inflammation and osteolysis. Cell Death Discov. 2022; 8:197. 10.1038/s41420-022-00999-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Cheng Z, Zou X, Jin Yet al.. The role of KLF(4) in Alzheimer’s disease. Front Cell Neurosci. 2018; 12:325. 10.3389/fncel.2018.00325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Xi J, Zhang B, Luo Fet al.. Quercetin protects neuroblastoma SH-SY5Y cells against oxidative stress by inhibiting expression of Krüppel-like factor 4. Neurosci Lett. 2012; 527:115–20. 10.1016/j.neulet.2012.08.082. [DOI] [PubMed] [Google Scholar]
- 63.Kong B, Wu PC, Chen Let al.. MicroRNA-7 protects against 1-methyl-4-phenylpyridinium iodide-induced cell apoptosis in SH-SY5Y cells by directly targeting Krüpple-Like Factor 4. DNA Cell Biol. 2016; 35:217–25. 10.1089/dna.2015.3097. [DOI] [PubMed] [Google Scholar]
- 64.Li H, Zhang C, Shen Het al.. Physiological stress-induced corticosterone increases heme uptake via KLF4–HCP1 signaling pathway in hippocampus neurons. Sci Rep. 2017; 7:5745. 10.1038/s41598-017-06058-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Tomita K, Nagasawa T, Kuwahara Yet al.. MiR-7-5p is involved in ferroptosis signaling and radioresistance thru the generation of ROS in radioresistant HeLa and SAS cell lines. Int J Mol Sci. 2021; 22:8300. 10.3390/ijms22158300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Nguyen LD, Wei Z, Silva MCet al.. Small molecule regulators of microRNAs identified by high-throughput screen coupled with high-throughput sequencing. Nat Commun. 2023; 14:7575. 10.1038/s41467-023-43293-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Dong YF, Chen ZZ, Zhao Zet al.. Potential role of microRNA-7 in the anti-neuroinflammation effects of nicorandil in astrocytes induced by oxygen-glucose deprivation. J Neuroinflammation. 2016; 13:60. 10.1186/s12974-016-0527-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Kaushik DK, Mukhopadhyay R, Kumawat KLet al.. Therapeutic targeting of Krüppel-like factor 4 abrogates microglial activation. J Neuroinflammation. 2012; 9:57. 10.1186/1742-2094-9-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69.Guo C, Liu P, Deng Get al.. Honokiol induces ferroptosis in colon cancer cells by regulating GPX4 activity. Am J Cancer Res. 2021; 11:3039–54. [PMC free article] [PubMed] [Google Scholar]
- 70.Liu F, Zhang Y, Xia Xet al.. Honokiol induces ferroptosis in ovarian cancer cells through the regulation of YAP by OTUB2. J Obstet Gynaecol. 2024; 50:864–72. 10.1111/jog.15922. [DOI] [PubMed] [Google Scholar]
- 71.Faysal M, Khan J, Zehravi Met al.. Neuropharmacological potential of honokiol and its derivatives from Chinese herb Magnolia species: understandings from therapeutic viewpoint. Chin Med. 2023; 18:154. 10.1186/s13020-023-00846-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72.Guo Y, Liu X, Xu Qet al.. Revealing the role of honokiol in human glioma cells by RNA-seq analysis. Biocell. 2024; 48:945–58. 10.32604/biocell.2024.049748. [DOI] [Google Scholar]
- 73.Mahatan CS, Kaestner KH, Geiman DEet al.. Characterization of the structure and regulation of the murine gene encoding gut-enriched Krüppel-like factor (Krüppel-like factor 4). Nucleic Acids Res. 1999; 27:4562–9. 10.1093/nar/27.23.4562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Disney MD The Druggable Transcriptome Project: from chemical probes to precision medicines. Biochemistry. 2025; 64:1647–61. 10.1021/acs.biochem.5c00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75.Taghavi A, Springer NA, Zanon PRAet al.. The evolution and application of RNA-focused small molecule libraries. RSC Chem Biol. 2025; 6:510–27. 10.1039/D4CB00272E. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76.Corey DR It is time to revisit miRNA therapeutics. Nucleic Acid Ther. 2025; 35:1–5. 10.1089/nat.2024.0069. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
NanoString miRNA transcriptomic data is deposited in NCBI GEO repository with accession number GSE271276. All data supporting the findings of this study are available within the paper and Supplementary files. All other data generated in this study are available from the corresponding author on reasonable request.