A Microglia–Astrocyte Signaling Axis Regulates Astrocyte Piezo1 Expression and Inflammatory Responses

ABSTRACT

Structural tissue alterations in numerous brain disorders can initiate mechanosensory signaling pathways and influence neuropathology. Astrocytes are highly mechanosensitive cells that play essential roles in maintaining brain homeostasis; however, the molecular mechanisms underlying astrocyte mechanosensation during pathological conditions remain largely unexplored. In this study, we investigated how the expression of the mechanosensitive ion channel Piezo1 in astrocytes is modulated by inflammatory triggers. We found that direct exposure of primary astrocyte cultures to inflammatory stimuli, including lipopolysaccharide (LPS) or oligomeric amyloid‐β (oAβ), had minimal impact on astrocytic Piezo1 expression. In contrast, when LPS or oAβ were applied to primary microglia cultures, Piezo1 expression was increased in microglia, and conditioned media from these microglia cultures significantly upregulated Piezo1 expression in astrocytes. We further identified that microglia released pro‐inflammatory cytokines (IL‐1α, IL‐1β, and TNF‐α) that can directly enhance Piezo1 expression and Piezo1‐mediated Ca²⁺ signaling in both rodent and human astrocytes. Microglial depletion in 5xFAD mice consistently and substantially reduced astrocytic Piezo1 expression in vivo, supporting the physiological relevance of this microglia–astrocyte signaling axis during pathology. Activation of Piezo1 with Yoda2 did not alter astrocytic inflammatory gene expression under basal conditions but reduced TNF‐α, CCL2, and C3 expression following cytokine pretreatment. Conversely, Piezo1 knockdown increased GFAP expression at baseline and enhanced pro‐inflammatory gene expression under cytokine stimulation, indirectly promoting microglial activation. These findings demonstrate that astrocytic Piezo1 expression is regulated by microglia‐derived inflammatory signals and plays a context‐dependent role in modulating astrocyte reactivity and neuroinflammatory responses.

• Activated microglia‐derived inflammatory signals drive astrocytic Piezo1 expression in both in vitro and in vivo models.

• Piezo1 upregulation modulates astrocytic Ca²⁺ dynamics and inflammatory gene expression, thereby limiting microglial activation.

1. Introduction

Mechanical stimuli are pervasive throughout the body, and most cells possess mechanosensors that enable them to respond to such stimuli (Di et al. 2023). Cells can react not only to changes in their intrinsic mechanical properties, such as stiffness or morphological alterations, but also to external mechanical inputs to the cell membrane, including shear stress, interstitial flow, and compression (Polacheck et al. 2013). Mechanotransduction refers to the activation of mechanosensitive channels in cell membranes by mechanical stress that leads to various responses in cells, thereby transducing mechanical stimuli into cellular biochemical signals (Hamill and Martinac 2001). For example, mechanosensitive ion channels are involved in a growing list of functions, including touch, pain (Woo et al. 2014; Alessandri‐Haber et al. 2005), auditory perception (Zanini and Gopfert 2014), blood pressure regulation (Mohieldin et al. 2016; Luu et al. 2018), and vascular development (Chen and Li 2021). In 2010, Patapoutian's research group reported the discovery of a novel mechanosensitive cation channel they named Piezo which was derived from the Greek word meaning pressure (Coste et al. 2010, 2012). They found Piezo1 to be responsible for the cellular currents triggered by mechanical stimulation. This discovery was the foundation for Patapoutian's Nobel Prize in 2021. Piezo1 is a large membrane protein consisting of 38 transmembrane domains, forming a trimeric structure with a unique propeller‐like architecture that includes a central pore and three extracellular blades (Ge et al. 2015; Fang et al. 2021). Under mechanical forces, this structure transitions from a curved to an expanded state, opening the central pore to allow cation influx (Lin et al. 2019; Wang et al. 2018; Yang et al. 2022). Piezo1 has been reported to play crucial functions in various cell types (Cinar et al. 2015; Liu et al. 2020; Geng et al. 2021). In the central nervous system (CNS), Piezo1 promotes neural stem cell differentiation (Li et al. 2022), modulates axonal regeneration (Song et al. 2019; Li et al. 2021), inhibits oligodendrocyte proliferation and migration (Velasco‐Estevez et al. 2022), and regulates microglia phagocytosis and cytokine release (Jantti et al. 2022; Malko et al. 2023). Furthermore, Piezo1 has been associated with the pathophysiology of several CNS disorders, including Alzheimer's disease (AD) (Jantti et al. 2022; Hu et al. 2023), multiple sclerosis (Velasco‐Estevez, Gadalla, et al. 2020; Jairaman et al. 2021), stroke (Guo et al. 2021; Qu et al. 2023), glioma (Chen et al. 2018), and glaucoma (Uchida et al. 2021).

Astrocytes are the most abundant glial cells in the brain. They occupy non‐overlapping spatial domains throughout the brain where they fulfill many functions such as providing trophic and metabolic support, maintaining ion and water balance, regulating neuroinflammation, and playing pivotal roles in the “tripartite synapse” by modulating synaptic transmission and neurotransmitter recycling (Gradisnik and Velnar 2023). In addition, astrocyte end‐feet surround brain blood vessels, contributing to blood–brain barrier formation and neurovascular coupling (Mishra et al. 2024; Zhao et al. 2015). Activated astrocytes undergo a process called astrogliosis, characterized by the enlargement of their cell bodies and increased complexity of their processes (Pekny and Pekna 2014; Escartin et al. 2019). Intracellular calcium ([Ca²⁺]_int) is a critical second messenger in astrocytes that is widely considered a key indicator of astrocyte activity (Semyanov et al. 2020; Shigetomi et al. 2019).

Astrocytes express the mechanosensitive ion channel Piezo1, which regulates astrocyte intracellular Ca²⁺ signaling (Chi et al. 2022; Velasco‐Estevez, Rolle, et al. 2020; Yu et al. 2023), promotes astrogenesis (Pathak et al. 2014), influences adult neurogenesis (Chi et al. 2022), and modulates astrocyte migration and cytokine release (Velasco‐Estevez, Rolle, et al. 2020). Previous studies have reported that inflammatory stimuli, such as LPS, increased Piezo1 expression and enhanced calcium activities in astrocyte cell cultures (Velasco‐Estevez, Rolle, et al. 2020; Yu et al. 2023). In addition, Piezo1 upregulation has been observed in a rat model of Alzheimer's disease, and this effect was amplified by peripheral bacterial infections (Velasco‐Estevez et al. 2018). More recently, a study reported that neuroinflammation correlates with Piezo1 levels in both human and rodent brains, and TNF‐α can induce Piezo1 upregulation in human astrocytes (Garcia et al. 2023). These findings indicate that astrocytic Piezo1 increases in inflammatory conditions.

Microglia, as the primary immune responders in the brain, play critical roles in producing pro‐inflammatory factors and activating astrocytes (Catalin et al. 2013; Bilbo and Stevens 2017). Based on this, we hypothesized that microglia act as key initiators of Piezo1 upregulation in astrocytes. In support of this hypothesis, we observed increased Piezo1 expression in astrocytes in both LPS‐injected mice and 5xFAD mice compared to wild‐type controls. In vitro, LPS or oAβ treatment significantly upregulated Piezo1 in primary microglia but had minimal effects on astrocytes. However, conditioned media from activated microglia robustly increased Piezo1 expression in astrocytes. We further demonstrated that activated microglia‐derived cytokines, including IL‐1α, IL‐1β, and TNF‐α, can significantly induce Piezo1 upregulation and enhance Ca²⁺ activity in both rodent astrocytes and human iPSC‐derived neurospheres. Depletion of microglia in 5xFAD mice markedly reduced astrocytic Piezo1 expression in vivo. Collectively, these findings identify a microglia–astrocyte signaling axis in which astrocytic Piezo1 expression is broadly regulated under inflammatory conditions by activated microglia and associated pro‐inflammatory cytokines across rodent and human in vitro models, as well as in an in vivo Alzheimer's disease model.

We next examined the functional consequences of Piezo1 upregulation in astrocytes. Under basal conditions, Piezo1 modulation had a limited effect on inflammatory factors, although knockdown increased GFAP expression. In contrast, following cytokine pretreatment, which mimics an inflammatory environment and induces Piezo1 in an upregulated state, Piezo1 activation robustly suppressed TNF‐α, CCL2, and C3 expression, whereas Piezo1 knockdown enhanced TNF‐α and CCL2 levels while consistently reducing C3 expression. IL‐1β and IL‐6 remained largely unchanged in both conditions. Conditioned media from Piezo1‐knockdown astrocytes enhanced pro‐inflammatory cytokine expression in microglia only under cytokine‐pretreated conditions. Together, these results indicate that Piezo1 contributes to maintaining astrocyte homeostasis under resting conditions, whereas it functions as a key anti‐inflammatory regulator in cytokine‐primed reactive astrocytes to suppress pro‐inflammatory gene expression and limit excessive neuroinflammation (Summary Figure 1).

SUMMARY FIGURE 1.

Diagrams depicting the mechanism of Piezo1 regulation in astrocytes. Left diagram: Piezo1 expression in primary astrocytes is elevated due to cytokines released by microglia in response to LPS or oAβ exposure, rather than direct exposure of astrocytes to LPS or oAβ. Astrocytic Piezo1 upregulation is associated with enhanced Ca²⁺ activity and changes in gene expression. Right diagram: Piezo1 upregulation in 5xFAD mice decreases following microglia depletion. Created in BioRender.

2. Methods

2.1. Chemicals and Reagents

Yoda1 (#5586), Yoda2 (#8051), and Gadolinium Chloride (#4741) were purchased from Tocris (Bristol, UK). Lipopolysaccharide (#L2630) was purchased from Sigma‐Aldrich (Burlington, USA). Rat IL‐1α (#400‐01A‐10UG), IL‐1β (#200‐01B‐10UG), TNF‐α (#400‐14‐20UG), and human IL‐1α (#200‐01A‐10UG), IL‐1β (#400‐01B‐10UG), and TNF‐α (#300‐01A‐50UG) were purchased from Gibco (Waltham, USA). PLX3397 (#S7818) used for cell cultures was purchased from Selleck Chemicals (Houston, USA). PLX3397 powder for preparing mouse diets was obtained from AbMole (Houston, USA) and submitted to Research Diets Inc. (New Brunswick, USA) for formulation of the PLX chow (600 mg/kg PLX3397) and the corresponding control chow. Piezo1 DsiRNA and negative control DsiRNA were designed and synthesized by Integrated DNA Technologies (IDT) (Coralville, USA). The Piezo1 DsiRNA duplex sequences were: 5′‐CAAGAAAUACAACCAUCU and 5′‐GAGGUUUAGAUGGUUGU.

2.2. Cell Culture

To obtain primary rat astrocyte cultures, Sprague–Dawley rat pups (P1‐P7) were used. The cortex and hippocampus regions were isolated, triturated with a pestle in warmed astrocyte culture medium (MEM (#11095080, Gibco) supplemented with 10% FBS (#12483020, Gibco), 0.6% glucose (#G8270, Sigma‐Aldrich), 100 U/mL penicillin and 0.1 mg/mL streptomycin (#15140122, Gibco)), and further dissociated by pipetting. The tissue was incubated at 37°C for 15 min in HBSS solution (#14175095, Gibco) containing 0.25% trypsin (#25200072, Gibco). The cell suspension was filtered through a 40 μm sterile cell strainer and centrifuged to collect single cells. Cells were resuspended in astrocyte culture medium and plated in 75 cm³ flasks, allowing them to grow confluent for 1–2 weeks.

For astrocyte passaging, medium was removed, and cells were washed with PBS. 0.25% trypsin was added to cover the cells, followed by incubation at 37°C for approximately 3 min, depending on cell detachment. Detached cells were collected, centrifuged at 1000 rpm for 3 min, and resuspended in fresh astrocyte culture medium. The suspension was passed through a 40 μm cell strainer and plated in new flasks or culture plates. To eliminate microglia, cell cultures were treated with 0.5 μM PLX3397 for 1–3 days prior to each passage. The cells were passaged (1:6 ratio) for 2–4 times until they displayed a pure monolayer of typical astrocyte morphology under a light microscope, indicating their suitability for experimental use.

To obtain primary rat microglia cultures, rat brain cells were isolated using a procedure similar to that for astrocyte cultures and subsequently grown in microglia culture medium (DMEM/F12 (#11320033, Gibco) supplemented with 10% FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin) for over two weeks, with medium changed weekly. Floating microglia were harvested by shaking the culture dishes at 80 rpm for 2–4 h at 37°C using an orbital shaker. The medium was collected, centrifuged at 3000 rpm for 3 min, and the pellet was resuspended and plated on culture plates with coverslips for experiments.

All human iPSC‐derived cell cultures were generated from a healthy erythroid progenitor cell derived female donor cell line which we have previously characterized (Wendt, Lin, et al. 2025; Lee et al. 2023; Frew et al. 2020). The work was approved by the University of British Columbia's Clinical Research Ethics board. Human microglia were differentiated based on Cowley lab's protocol (Washer et al. 2022; Wendt, Lee, et al. 2025). Neural precursor cells (NPCs) were differentiated from iPSCs following our published protocols (Wendt, Lin, et al. 2025; Lee et al. 2023). Astrocytes were differentiated by following the STEMdiff Astrocyte Differentiation Kit (#100–0013, STEMCELL Technologies, Vancouver, Canada) and STEMdiff Astrocyte Maturation Kit (#100–0016, STEMCELL Technologies) according to the manufacturer. Human astrocyte densities for experiments were matched to the experiments conducted on primary rat astrocytes. The same NPCs were used to generate human neurospheres following our recently published protocol (Wendt, Lin, et al. 2025; Wendt, Lee, et al. 2025). Note that for these experiments, optional human iPSC‐derived microglia were not added to the neurospheres and that the neurospheres used here matched our published ctrl hiNS(−) configuration (Wendt, Lin, et al. 2025), containing neurons, astrocytes, and NPCs.

To obtain oAβ‐ or LPS‐treated rat microglia conditioned media (oAβ‐MCM or LPS‐MCM), rat microglia were treated with oAβ or LPS for 6 h, after which the medium was replaced with fresh serum‐free microglia culture medium for an additional 12 h. The conditioned media were then collected and applied to rat astrocytes. The rat microglial culture medium used is suitable for mixed glial cell cultures, including rat astrocytes. This allowed the application of microglia‐conditioned medium directly to astrocytes without switching to astrocyte‐specific medium. Human microglia LPS‐conditioned media (LPS‐hMCM) were generated by treating human microglia with LPS for 48 h. Media were then collected and concentrated at a 1:50 ratio using an Amicon Ultra 3 kDa centrifugal filter (#UFC500324, MilliporeSigma, Burlington, USA). Control conditioned media (Ctrl‐MCM and Ctrl‐hMCM) were prepared using the same procedures but without oAβ or LPS treatment. For human neurosphere treatment, concentrated Ctrl‐hMCM or LPS‐hMCM was diluted 1:50 into neurosphere culture medium.

To obtain Piezo1 knockdown astrocyte‐conditioned media under basal and cytokine‐pretreated conditions (siP_ACM or siP_Cyto ACM), astrocytes were first transfected with Piezo1 siRNA and cultured for three days. For the cytokine‐pretreated conditions, cells were exposed to a cytokine cocktail (IL‐1α, IL‐1β, and TNF‐α) 24 h after transfection for 1 day, followed by replacement with fresh serum‐free medium for an additional 24 h before media collection (siP_Cyto ACM). For the basal conditions, fresh serum‐free medium was added 48 h after siRNA transfection without cytokine treatment, and conditioned media were collected after 24 h (siP_ACM). Conditioned media from control siRNA‐transfected astrocytes, processed in parallel using the same procedures, were collected as control conditioned media (siC_ACM and siC_Cyto ACM).

All cell cultures were maintained at 37°C in an incubator with 95% air and 5% CO₂. For oAβ, LPS, or cytokine treatments in rat primary astrocyte or microglia cultures, serum‐free medium (identical to the culture medium formulation but without FBS) was used. For all treatments in rat primary cells or human astrocytes, samples were collected after 4 h for qPCR, 24 h for western blotting, and 24 h for live calcium imaging. For human neurospheres, all treatments were applied for 48 h prior to conducting experiments (including western blotting and live calcium imaging). The 4 h time point for qPCR analysis was selected because transcriptional changes generally occur earlier than alterations in protein expression, and early time points are commonly used to capture primary gene expression responses following stimulation. This selection was further supported by our pilot experiments, in which Piezo1 mRNA expression was assessed at 2, 4, 8, and 24 h after inflammatory stimulation in microglia. Among these time points, Piezo1 mRNA expression showed the greatest increase at 4 h (Figure S7).

2.3. Cell Transfection

Primary rat astrocytes were transfected with Piezo1 siRNA or negative control using Lipofectamine RNAiMAX (#13778030, Invitrogen, Waltham, USA). Before transfection, cells were seeded in 24‐well plates and cultured until reaching approximately 70% confluence. For each well, siRNA (10 nM final concentration) was diluted in Opti‐MEM (#31985070, Gibco) and mixed with RNAiMAX diluted in Opti‐MEM to generate transfection complexes, following the manufacturer's recommended ratio. The complexes were incubated for 10–15 min at room temperature and then applied to the cells. After 4–6 h of incubation, the transfection medium was replaced with fresh astrocyte culture medium. Cells were maintained for an additional 3 days prior to evaluating Piezo1 knockdown efficiency or conducting live Ca²⁺ imaging experiment.

2.4. Oligomeric Aβ Preparation

Synthetic Aβ1‐42 was purchased from ERI Amyloid Laboratory (Oxford, USA). Oligomer Aβ was prepared following previously published methods (Wendt, Lin, et al. 2025; Um et al. 2012). In brief, 500 μg lyophilized Aβ powder was dissolved in 40 μL of DMSO. The solution was thoroughly mixed and incubated for 20 min at room temperature (RT). Subsequently, 2 mL of F12 media (#11765054, Gibco) was added and incubated overnight at RT to induce oligomerization. The mixture was centrifuged at 14,000 g for 15 min using an Amicon Ultra 3 kDa filter. Due to the 0.5 mL volume limit per filter, the sample was centrifuged four times using the same filter to concentrate the oAβ. The filter was then washed twice with 500 μL of PBS. Finally, the filter was flipped into a collection tube followed by centrifugation at 1000 g for 2 min to collect the concentrated oAβ.

2.5. Animals

All animal procedures were approved by the University of British Columbia Animal Care Committee and carried out in compliance with the Canadian Council for Animal Care guidelines. 5xFAD mice (#034840) and C57BL/6J mice (#000664) were obtained from The Jackson Laboratory (Bar Harbor, USA). For breeding the 5xFAD strain, male heterozygous 5xFAD mice were crossed with female C57BL/6J mice. All mouse strains were housed in an animal facility with controlled temperature and humidity, maintained on a 12‐h light/dark cycle, with food and water provided ad libitum. Both male and female mice were included in the study without sex‐based selection. Brain tissues from one‐year‐old heterozygous 5xFAD mice were used for RT‐qPCR, Western blot, and immunostaining experiments, with age‐ and gender‐matched wild‐type (WT) mice serving as controls. For microglia depletion experiments, heterozygous 5xFAD mice aged 6–7 months were randomly assigned to receive either PLX3397‐containing chow or control chow for three months, with control and experimental groups sex‐matched to ensure a similar proportion of males and females.

2.6. LPS Intracranial Injection

C57BL/6J mice aged 2–4 months were used for LPS intracranial injection. Mice were anesthetized and placed in a stereotaxic frame. The skull was exposed, and the somatosensory cortex was targeted using coordinates from the mouse brain atlas (AP −1.3 mm, ML ±2.0 mm, DV −0.8 mm). A small hole was drilled at the designated site, and 0.5 μL of LPS (5 μg/μL) or sterile saline was injected at a rate of 0.1 μL/min using a Hamilton syringe. Following injection, the needle was left in place briefly to prevent reflux, after which the incision was sutured. Postoperative care included routine monitoring and administration of analgesics as needed. Three days post‐injection, cortical tissue surrounding the injection site (approximately a 2‐mm radius) was collected for RT‐qPCR analysis, or animals were perfused for subsequent immunostaining.

2.7. Astrocyte Isolation and Sorting

At collection, animals were perfused transcardially with HBSS containing transcription and translation inhibitors (27.1 ug/mL of Anisomycin (#15021S, New England Biolabs, Ipswich, USA); 5 ug/mL of Actinomycin D (#15021S, New England Biolabs); 10uM of Triptolide (#97529S, New England Biolabs)) following a previously published method (Marsh et al. 2022). After that, the cortex and hippocampus of the brain were microdissected, and astrocytes were collected as previously described (Agalave et al. 2020). Briefly, tissue was digested for 45 min using the Papain Dissociation System (#LK003178, Worthington Biochemical, Lakewood, USA) made in HBSS with 0.1 mg/mL DNase I (#07900, STEMCELL Technologies) and inhibitors listed above. Mechanical digestion was then performed with Dounce homogenizers and astrocytes were collected via a discontinuous percoll gradient (#GE17‐0891‐02, Sigma‐Aldrich) made using FACS buffer (1xHBSS, 2% BSA, 2 mM EDTA, inhibitors). After washing twice with FACS buffer, cells were blocked with 2% normal donkey serum and TruStain FcX (#156604, 1:100, Biolegend, San Diego, USA) for 20 min on ice, followed by an immunostaining for GLAST(ACSA‐1) primary antibody (#130‐118‐344, 1:100, Miltenyi Biotec, Bergisch Gladbach, Germany) for 30 min at 4°C. After wash with FACS buffer, cells were then stained with Alexa 488‐conjugated secondary antibodies (Invitrogen, 1:100), CD11b‐BV650 (#101259, Biolegend, 1:200), and CD45‐PECy7 (#103114, Biolegend, 1:200) for 30 min at 4°C. GLAST+, CD11b‐, CD45‐ single cells were then sorted and collected and defined as “astrocytes” using a Beckman Coulter CytoFLEX SRT Cell Sorter.

2.8. Mouse Behavioral Tests

In the open field test, mice were placed in a 40 cm × 40 cm chamber for 6 min. Their activity was recorded and analyzed using AnyMaze software. Total distance traveled was used as a measure of locomotor activity, while the number of entries into the center zone and time spent in the center zone (20 cm × 20 cm) were used to assess anxiety‐like behavior. In the Y‐maze test, mice were allowed to freely explore the three arms of the maze for 8 min. Arm entries and sequences were recorded, and spontaneous alternation was calculated as an index of spatial working memory. In the fear conditioning test, mice were habituated to the conditioning chamber on day 1 for 5 min. On day 2, mice received two foot shocks (0.6 mA, 2 s) at 3 and 4 min during a 6‐min session. On day 3, mice were returned to the same chamber for 5 min without foot shock. Mice's total freezing time and freezing episodes on day 3 were measured as indicators of contextual memory.

2.9. RNA Extraction and cDNA Synthesis

Brain tissue (cortex), sorted astrocytes, or cell cultures subjected to various treatments were harvested using 200 μL of Trizol reagent (#15596018, Invitrogen). The samples were then homogenized by pipetting up and down and vortexed intermittently for 10 min to ensure complete homogenization. Following this, 20 μL of 1‐bromo‐3‐chloropropane was added, vortexed for 10 s, and rested at room temperature for 5 min. The mixture was then centrifuged at 12,000 g for 10 min at 4°C. The upper aqueous phase (100 μL) was transferred to a new tube, mixed with 50 μL of isopropanol. Subsequently, 10 μL of nanomag‐D RNA binding beads (#09–01–252, MicroMod, Rostock, Germany) were added to the mixture for magnetic RNA capture. The beads were then washed twice with RNA wash buffer (comprising 80% ethanol, 10 mM KCl, 2 mM Tris, and 0.2 mM EDTA in H₂O). After wash, beads were air‐dried at room temperature, resuspended in TE buffer, and RNA was dissolved. Following RNA concentration determination, equal amounts of mRNA were used to synthesize cDNA by using a High‐Capacity cDNA Reverse Transcription Kit (#4368814, Applied Biosystems, Waltham, USA) as per the manufacturer's protocol.

2.10. Real‐Time Quantitative PCR

The qPCR mixture was comprised of the same amount of cDNA sample, 250 nM of each primer, 5 μL of Luna Universal qPCR Master Mix (#M3003, New England Biolabs), and purified water to reach a final volume of 10 μL. Amplification reactions were conducted using the Bio‐Rad CFX384 Real‐Time System according to the manufacturer's instructions. The thermal cycling program involved an initial denaturation step at 95°C for 5 min, followed by 39 cycles of denaturation at 95°C for 5 s and annealing/extension at 60°C for 30 s. The primer sequences and corresponding references are provided in Table 1.

TABLE 1.

Sequences of mouse (m), rat (r), and human (h) qPCR primers.

Gene	Forward primers	Reverse primers
mPiezo1	ATGGCCGTGAATGTGATTGGGCAGC	CGCCATGGATAGTCAATGCACAGTGC
mGAPDH (Cahyadi et al. 2023)	AGGTCGGTGTGAACGGATTTG	TGTAGACCATGTAGTTGAGGTCA
m/rHPRT (Towriss et al. 2023)	CAGTACAGCCCCAAAATGGTTA	AGTCTGGCCTGTATCCAACA
rPiezo1	TGGCCCTGAAAGAAGTGTCC	AAACCAGTTGGCAGGGTCAA
rIL‐1α (Zhou et al. 2023)	AAGACAAGCCTGTGTTGCTGAAGG	TCCCAGAAGAAAATGAGGTCGGTC
rIL‐1β (Inoue et al. 2002)	GGCTGTCCAGATGAGAGCAT	CCACGGGCAAGACATAGGTA
rTNF‐α (Peinnequin et al. 2004)	AAATGGGCTCCCTCTCATCAGTTC	TCTGCTTGGTGGTTTGCTACGAC
rCCL2 (Kuhne et al. 2017)	ATGCAGTTAATGCCCCACTC	TTCCTTATTGGGGTCAGCAC
rC3 (Aronica et al. 2007)	TCACAGTGATGCAGGACTTC	GGATTTCGATGGTCTGGTAG
rGFAP (Hoppe et al. 2025)	CAAGATGAAACCAACCTGAGGCT	GGCTTGGCCACATCCATCT
hPiezo1 (Zhang et al. 2021)	GGACTCTCGCTGGTCTACCT	GGGCACAATATGCAGGCAGA
hHPRT (Habibi et al. 2020)	CCTGGCGTCGTGATTAGTGAT	AGACGTTCAGTCCTGTCCATAA

2.11. Western Blot

Primary cell cultures or brain tissue samples were harvested and homogenized on ice using RIPA lysis buffer (#89900, Thermo Scientific, Waltham, USA) supplemented with cOmplete ultra protease inhibitor tablets (#5892791001, Sigma‐Aldrich) and PhosSTOP phosphatase inhibitor tablets (#4906845001, Sigma‐Aldrich). The homogenates were centrifuged at 12,000 rpm for 10 min at 4°C to collect the supernatant. Protein concentration was then determined using the DC Protein Assay kit (#5000112, Bio‐Rad, Hercules, USA). Equal protein amounts were mixed with 4× Laemmli Sample Buffer (#1610747, Bio‐Rad) containing β‐mercaptoethanol and vortexed at RT for 1 h before being loaded onto Mini‐Protean TGX gels (#4561096, Bio‐Rad). Subsequently, SDS‐PAGE was conducted, and the separated proteins were transferred onto 0.2 μm pore‐size PVDF membranes using the Trans‐Blot Turbo semi‐dry transfer system (Bio‐Rad). The membranes were then blocked for 1 h at RT with 3% BSA in TBS containing 0.1% Tween‐20 (TBST) and then incubated overnight at 4°C in TBST with primary antibodies. Primary antibodies used were: Piezo1 (#APC‐087, 1:200, Alomone Labs, Jerusalem, Israel), GAPDH (#PA1988, Invitrogen, 1:1000), β‐actin (#MA5‐15739, Invitrogen, 1:1000), iNOS (#ab15323, 1:1000, Abcam, Cambridge, UK), IL‐1β (#ab283818, Abcam, 1:1000), TNF‐α (#60291‐1‐IG, 1:300, Proteintech, Rosemont, USA). After washing, membranes were probed with HRP‐conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, USA) for 2 h at room temperature. Protein bands were visualized using Bio‐Rad Clarity Western ECL substrate and imaged with a ChemiDoc MP imaging system. Representative full‐membrane western blot images demonstrating the specificity of the Piezo1 antibody and the molecular weight of Piezo1 protein in rat cell cultures and mouse tissues are shown in Figure S8.

2.12. Immunocytochemistry Staining

Cells grown on coverslips in a 24‐well plate were washed once with PBS and fixed with 4% paraformaldehyde (PFA) for 15 min at room temperature (RT), followed by three washes with PBS. The cells were then blocked for 30–60 min at RT with a blocking buffer containing 5% BSA in PBS with 0.3% Triton X‐100. After blocking, cells were washed three times with PBS and incubated overnight at 4°C with primary antibodies diluted in the blocking buffer. The primary antibodies used were iNOS (#ab15323, 1:400, Abcam), GFAP (#13‐0300, Invitrogen, 1:1000), and Iba1 (#234009, 1:400, Synaptic Systems, Göttingen, Germany). After primary antibody incubation, cells were washed three times with PBS and then incubated for 1 h at room temperature (protected from light) with Alexa Fluor‐conjugated secondary antibodies (Donkey anti‐rabbit/rat/chicken IgG, Invitrogen, 1:500) diluted in PBS containing 0.5% BSA and 0.03% Triton X‐100. Following secondary incubation, cell nuclei were stained with DAPI (0.5 μg/mL) (#D1306, Invitrogen) in PBS for 10 min. The cells were washed again and mounted with FluorSave Reagent (#345789, MilliporeSigma). The stained coverslips were imaged using confocal microscopy (ZEISS).

2.13. Immunohistochemistry Staining

Mice were euthanized with 30% w/v urethane (#U2500, Sigma‐Aldrich) and perfused transcardially with PBS, followed by 4% paraformaldehyde (PFA). The brains were extracted, immersed in 4% PFA overnight at 4°C, and sectioned into 50 μm slices using a vibratome (Leica VT1200S, Wetzlar, Germany). Brain slices were then incubated overnight at 4°C in blocking solution containing 0.1 M PBS, 20% DMSO, 2% Triton‐X, and 10% Normal Donkey Serum (NDS) (#017‐000‐121, Jackson ImmunoResearch). After blocking, slices were washed four times with wash solution containing 0.1 M PBS, 20% DMSO, and 2% Triton‐X, followed by incubation with primary antibodies diluted in blocking solution for 2 days at 4°C. The primary antibodies used were Piezo1 (Alomone, 1:50), GFAP (Invitrogen, 1:1000), and Iba1 (Synaptic Systems, 1:400). Following primary antibody incubation, slices were washed with wash solution and incubated for 1–2 days at 4°C with secondary antibodies diluted in a solution containing 0.1 M PBS, 20% DMSO, 2% Triton‐X, and 2.5% NDS. Secondary antibodies used were donkey anti‐rabbit Alexa Fluor 568 (1:500), donkey anti‐rat Alexa Fluor 488 (1:500), and donkey anti‐chicken Alexa Fluor 647 (1:500), all from Invitrogen. For additional amyloid plaque staining, slices were incubated with 100 μM Methoxy‐X04 (#4920, Tocris) for 45 min after secondary antibody incubation. After washing, the stained brain slices were imaged using a Zeiss LSM 980 confocal microscope or Zeiss Axio Zoom V16 microscope (Zeiss, Oberkochen, Germany).

2.14. Live Calcium Imaging

Primary astrocyte cultures were incubated with Fluo‐4 AM (5 μg/mL) (#F14201, Invitrogen) for 30 min at 37°C to allow intracellular calcium indicator loading. After loading, cells were perfused with HEPES‐buffered artificial cerebrospinal fluid (HEPES‐ACSF; 140.4 mM NaCl, 10.8 mM HEPES, 2.5 mM KCl, 10 mM glucose, 2.0 mM MgCl₂, 2.0 mM CaCl₂, pH 7.35) at room temperature for 5 min. Live imaging was performed using a Zeiss LSM 980 confocal microscope with an excitation wavelength of 488 nm, capturing time‐series images at 3‐s intervals with a resolution of 512 × 512 pixels. After baseline recording for 5 min, 10 μM Yoda1 or Yoda2 (freshly prepared in HEPES‐ACSF) was applied via perfusion for 10 min, followed by washout with fresh HEPES‐ACSF. To evaluate the effect of Piezo1 inhibition by Gadolinium, 30 μM Gadolinium was added to the HEPES‐ACSF throughout the entire time series recording, including during Yoda1 application. To evaluate calcium activity after Piezo1 knockdown by Piezo1 siRNA, Piezo1 siRNA was transfected into rat astrocytes 3 days before the experimental date as previously mentioned. For calcium‐free experiments, calcium‐free HEPES‐ACSF (140.4 mM NaCl, 10.8 mM HEPES, 2.5 mM KCl, 10 mM glucose, 2.0 mM MgCl₂, 2.0 mM EGTA, pH 7.35) was used during all stages, including baseline recording, Yoda1 application, and washout.

For calcium imaging in human iPSC‐derived neurospheres (hiNS), hiNS were transduced using 0.5 mL of AAV2/php.eB‐gfaABC1D‐jGCaMP8s (titer: 1.3E13 GC/mL) at 40 days in vitro (DIV) in 48‐well glass bottom plates (Mattek, Ashland, USA). On DIV 60, plates were moved to an LSM 880 confocal microscope equipped with a live cell incubator maintaining 37°C and 5% CO₂. Spontaneous calcium activity in hiNS was first recorded for 3 min as a baseline (with or without prior cytokine treatment). Subsequently, 20 μM Yoda2 was added to the culture medium, and an additional 3‐min recording was performed. To assess the effect of Piezo1 inhibition, 100 μM Gadolinium was applied, followed by another 3‐min recording. Calcium activity in hiNS was analyzed using the AQuA (Astrocyte Quantification and Analysis) tool following the published protocol (Wang et al. 2019).

2.15. Elisa

The quantification of oAβ concentration was described in our previously published method (Wendt, Lin, et al. 2025). In brief, oAβ stock solution was 1:5000 diluted and then run in duplicate using the V‐PLEX Aβ Peptide Panel 1 (#K15199E‐2, Meso Scale Diagnostics, Rockville, USA) following the manufacturer's instructions. To quantify proinflammatory cytokines (IL‐1β, IL‐6, and TNF‐α) in 5xFAD mice brain following the PLX3397 diet or in cultured media from primary microglia after oAβ or LPS treatment, the V‐PLEX Mouse Proinflammatory Cytokine Kit (#K152A0H‐1, Meso Scale Diagnostics) was used according to the manufacturer's protocols. Brain tissue was homogenized in RIPA lysis buffer containing protease inhibitors, and total protein concentration was determined using the DC Protein Assay Kit (Bio‐Rad). Microglia media (400 μL per well) was collected from 24‐well plates and centrifuged at 3000 rpm for 3 min to remove cells and debris. A total amount of 325 μg of protein or 30–60 μL of microglia media was loaded into each well of the MSD plate. The plate was analyzed using the MESO QuickPlex SQ 120MM reader. Seven‐point standard curves were prepared using the kit calibrators. Cytokine concentrations were calculated using Discovery Workbench software by interpolating the electrochemiluminescence signal of each sample against the standard curves. All values were normalized to total protein load for brain homogenates or volume for cell culture media.

2.16. Statistical Analysis

Protein bands from Western blot experiments were quantified using ImageJ software. Raw mRNA and protein expression values were first normalized to their corresponding loading controls (e.g., GAPDH/HPRT for qPCR; GAPDH/β‐actin for Western blot) within each experiment prior to downstream analyses. For qPCR and Western blot results of Piezo1 expression, data are presented as relative fold changes compared to the control group. Unless otherwise stated, each data point represents an independent biological replicate. All statistical analyses were performed using GraphPad Prism. Data are shown as mean ± SEM from the indicated number of independent experiments. Normality was evaluated using the Shapiro–Wilk and Kolmogorov–Smirnov tests. For comparisons between two groups, unpaired Welch's t‐test (parametric) or Mann–Whitney test (non‐parametric) was used. For comparisons involving more than two groups, Welch's ANOVA with Dunnett's post hoc test (parametric) or the Kruskal–Wallis test with Dunn's post hoc test (non‐parametric) was applied. For analyses involving multiple measurements across groups, two‐way ANOVA with Sidak's post hoc test or a mixed‐effects model (REML) with Sidak's or uncorrected Fisher's LSD post hoc test was used. Statistical tests were two‐tailed, and significance was defined as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

3. Results

3.1. Piezo1 Expression Is Upregulated in Astrocytes in LPS‐Injected Mice and in 5xFAD Mice

Although multiple studies have reported Piezo1 expression in astrocytes (Chi et al. 2022; Velasco‐Estevez, Rolle, et al. 2020; Yu et al. 2023), how Piezo1 expression changes under pathological conditions in vivo is not fully understood. To investigate this, we first examined Piezo1 expression in C57BL/6J mice following LPS injection. 2.5 μg of LPS was administered into the Somatosensory cortex region of the brain, while a control group received saline injections (Figure 1A). Three days post‐injection, we collected brain tissue for subsequent analyses. Immunostaining of brain slices revealed a marked increase in Piezo1 expression in the LPS‐injected mice compared to the saline‐injected controls (Figure 1B–D). The elevated Piezo1 expression was predominantly observed in cells colocalizing with the astrocyte marker GFAP, indicating that LPS stimulates Piezo1 expression in astrocytes (Figure 1B,D). Additionally, tissue samples surrounding the injection sites were analyzed via RT‐qPCR, which showed a significant elevation in Piezo1 mRNA expression in LPS‐injected mice compared to saline controls (Figure 1E).

FIGURE 1.

Piezo1 expression in astrocytes was increased in LPS‐injected mice and in 5xFAD mice. (A) Illustration of saline and LPS injection site and imaging area in mice brain. Created in BioRender. (B) Immunostaining results showing increased Piezo1 expression in the LPS cortex‐injected C57BL/6J mice that is colocalized with the astrocyte marker GFAP. 2.5 μg LPS was used in this experiment. Samples were collected three days after LPS injection. Scale bar: 20 μm. (C, D) Quantitative analysis of immunostaining results of Ctrl versus LPS‐injected mice. N = 4 from 4 mice. (E) qPCR result showing Piezo1 mRNA expression increased in the LPS‐injected C57BL/6J mice. N = 4. (F) Immunostaining showing astrocyte Piezo1 expression was increased in the 5xFAD mice compared to the control WT mice. The age of 5xFAD mice and control WT mice used in this experiment were about one year old. Scale bar: 50 μm. (G,H) Quantitative analysis data for immunostaining results of WT and 5xFAD mice. N = 4 from 4 mice. (I) qPCR result showed Piezo1 mRNA expression was increased in the 5xFAD mice. N = 5. (J) Western blot experiment showed Piezo1 protein expression increased in the 5xFAD mice compared to the control WT mice. N = 4. (K) In the 5xFAD mice, Piezo1 upregulation appears localized to Aβ plaque enriched areas. Aβ plaques were stained using Methoxy‐X04. *p < 0.05, **p < 0.01, ***p < 0.001. Data are shown as mean ± SEM.

We further assessed Piezo1 expression in the cortex and hippocampus of 5xFAD mice, a widely used model of amyloidosis mimicking key aspects of AD (Jawhar et al. 2012; Oakley et al. 2006). Immunostaining showed minimal Piezo1 expression in wild‐type (WT) mice, whereas 5xFAD mice exhibited a pronounced increase in astrocytic Piezo1 expression (Figure 1F–H). Supporting these findings, RT‐qPCR analysis revealed significantly higher Piezo1 mRNA levels in 5xFAD mice compared to WT mice (Figure 1I). Western blot analysis also confirmed increased Piezo1 protein expression in 5xFAD mice (Figure 1J). Interestingly, we found that astrocytic Piezo1 upregulation was closely associated with Aβ plaques in 5xFAD mice. The increase in Piezo1 expression appeared more pronounced in brain regions containing Aβ plaques (Figure 1K). These results indicate that pathological conditions such as LPS‐induced inflammation and amyloidosis significantly elevate Piezo1 expression in astrocytes.

3.2. Achieving Highly Purified Primary Astrocyte and Microglia Cultures

Given that Piezo1 expression in astrocytes was upregulated by LPS injection and in Aβ plaque burdened 5xFAD mice, we next aimed to investigate the mechanisms underlying Piezo1 upregulation using primary astrocyte and microglia cultures. A critical concern in such studies is microglial contamination in astrocyte cultures (Van Zeller et al. 2022; Saura 2007), as it complicates determining whether the observed effects are purely astrocytic or impacted by the presence of microglia and the release of microglia‐derived factors. To address this, we used PLX3397, a CSF1 receptor inhibitor (Coleman Jr. et al. 2020), to eliminate microglial contamination in astrocyte cultures before conducting experiments (Figure S1A).

To evaluate the efficiency of microglial depletion by PLX3397, we compared the number of microglia in astrocyte cultures treated with or without 0.5 μM PLX3397. Immunostaining for Iba1, a widely used microglial marker, revealed a significant reduction in Iba1‐positive cells in PLX3397‐treated cultures (Figure S1B,C). Since iNOS, a key inflammatory marker, is predominantly expressed in microglia following LPS exposure (Possel et al. 2000), we further analyzed Iba1 and iNOS‐positive cells in astrocyte cultures after LPS stimulation. Immunostaining showed that the number of Iba1‐ and iNOS‐positive microglia was significantly decreased in PLX3397‐treated cultures compared to controls without PLX3397 treatment (Figure S1D,E). Consistent with these findings, Western blot analysis demonstrated markedly reduced iNOS expression in PLX3397‐treated astrocyte cultures compared to controls after cytokine stimulation (Figure S1F,G). These results confirm the successful generation of highly purified primary astrocyte cultures.

We also assessed the purity of the primary microglia cultures and checked for potential astrocyte contamination. Immunostaining for Iba1, GFAP (an astrocyte marker), and DAPI (a nuclear marker) showed no GFAP‐positive cells, and all Iba1‐positive cells colocalized with DAPI, confirming the purity of the primary microglia cultures (Figure S1H). These findings underscore the successful preparation of highly purified primary astrocyte and microglia cultures, ensuring reliable and accurate experimental outcomes.

3.3. LPS Or Aβ Has Minimal Impact on Astrocytic Piezo1 Expression, Whereas LPS or Aβ Pretreated Microglia Conditioned Media Significantly Increased Piezo1 Expression in Astrocyte Cultures

After successfully obtaining highly purified primary astrocyte and microglia cultures, we investigated how astrocytic Piezo1 expression is upregulated by LPS or Aβ. Oligomeric Aβ (oAβ) was prepared following established protocols (Wendt, Lin, et al. 2025; Um et al. 2012). Since microglia are the first responder cells in CNS and Piezo1 mRNA level has been shown to be upregulated by LPS in microglia (Zhu et al. 2023), we first tested whether exposure to LPS or oAβ changes Piezo1 expression in microglia. We found that 100 ng/mL LPS and 0.8 μg/mL oAβ treatment can significantly increase Piezo1 mRNA expression in microglia cultures (Figure 2A,B). The effectiveness of LPS and oAβ on microglia Piezo1 protein upregulation was also confirmed by Western blot analysis (Figure 2C,D).

FIGURE 2.

Microglia Piezo1 is increased by LPS or oligomeric Aβ (oAβ) treatment. (A) Microglia Piezo1 mRNA expression increased following treatment with 100 ng/mL LPS. N = 6. (B) Microglia Piezo1 mRNA expression increased after treatment with 0.8 μg/mL of oAβ. N = 6. (C, D) Microglia Piezo1 protein expression was also increased following treatment with LPS or oAβ. N = 6, N = 5. *p < 0.05, **p < 0.01, ****p < 0.0001. Data are shown as mean ± SEM.

Next, we treated primary astrocyte cultures with 100 ng/mL LPS or 0.8 μg/mL oAβ and examined Piezo1 expression. RT‐qPCR results showed a slight, nonsignificant trend towards an increase in Piezo1 mRNA levels (Figure 3A,B), and Western blot analysis indicated no change in Piezo1 protein expression compared to controls (Figure 3C,D). Even when the oAβ concentration was increased fourfold to 3.2 μg/mL, Piezo1 protein levels in astrocytes remained unchanged (Figure 3D).

FIGURE 3.

Piezo1 expression in astrocyte cultures was increased by LPS‐MCM or oAβ‐MCM instead of directly LPS or oAβ treatment. (A, C) Astrocyte Piezo1 mRNA and protein expression did not increase following LPS treatment. LPS concentration was 100 ng/mL. N = 7, N = 6. (B, D) Astrocyte Piezo1 mRNA and protein expression was not changed by oAβ treatment. oAβ concentration was 0.8 μg/mL. 2× or 4× oAβ means two or four times of oAβ concentration (1.6 or 3.2 μg/mL) in the treatment. N = 7, N = 5. (E) Diagram of LPS or oAβ‐MCM preparation and treatment. Created in BioRender. (F, H) Astrocyte Piezo1 mRNA and protein expression were increased by LPS‐MCM. N = 5, N = 6. (G, I) Astrocyte Piezo1 mRNA and protein expression were increased by oAβ‐MCM. N = 6, N = 5. For all experiments, mRNA expression was tested after 4 h of treatment and protein expression was assessed after 24 h of treatment. *p < 0.05, **p < 0.01 and ns is statistically not significant. Data are shown as mean ± SEM.

Since LPS and oAβ could upregulate Piezo1 expression in microglia but not in astrocytes, we hypothesized that microglia‐derived factors might mediate this effect in astrocytes. To test this, we collected conditioned media from microglia pretreated with either LPS or oAβ (LPS‐MCM or oAβ‐MCM) (Figure 3E). After centrifugation to remove any cell debris, the conditioned media were applied to purified primary astrocyte cultures. As a control, media from untreated microglia (Ctrl‐MCM) were similarly prepared. We then assessed Piezo1 mRNA and protein expression in astrocytes following the MCM treatments. RT‐qPCR analysis revealed that both LPS‐MCM and oAβ‐MCM significantly increased Piezo1 mRNA levels in astrocytes (Figure 3F,G). Similarly, Western blot analysis confirmed a substantial increase in Piezo1 protein expression in astrocytes treated with LPS‐MCM or oAβ‐MCM (Figure 3H,I). These findings strongly suggest that factors released by microglia are responsible for the upregulation of Piezo1 expression in astrocytes.

3.4. Microglia‐Released Cytokines Increase Astrocytic Piezo1 Expression and Enhance Piezo1‐Mediated Ca²⁺ Responses

Studies have demonstrated that microglia‐released proinflammatory cytokines can activate astrocytes and cause morphological changes (Liddelow et al. 2017; Burda and Sofroniew 2017; Hyvärinen et al. 2019), we hypothesized that these cytokines may also serve as a trigger for the upregulation of astrocytic Piezo1 expression. RT‐qPCR analysis revealed that both LPS and oAβ significantly increased the mRNA expression of key proinflammatory cytokines in microglia, including interleukin‐1α (IL‐1α), interleukin‐1β (IL‐1β), and tumor necrosis factor‐α (TNF‐α) (Figure S2A,B). Western blot and ELISA further confirmed that LPS and oAβ enhanced microglial production of IL‐1β and TNF‐α proteins (Figure S2C–F).

To test whether these cytokines directly upregulate Piezo1 expression in astrocytes, we treated primary rat astrocyte cultures with IL‐1α (10 ng/mL), IL‐1β (10 ng/mL), or TNF‐α (20 ng/mL), either individually or in combination. Piezo1 mRNA and protein levels were subsequently assessed after 4 and 24 h of treatment. Cytokine treatment, either individually or in combination, significantly increased Piezo1 expression in primary astrocyte cultures (Figure 4A,B). To determine whether this effect is conserved in human cells, we repeated these experiments using human induced pluripotent stem cell (iPSC)‐derived astrocytes, which are differentiated directly from neural progenitor cells. The result showed cytokine treatment induced significant Piezo1 upregulation in iPSC‐derived astrocytes (Figure 4C).

FIGURE 4.

Cytokine treatment increased Piezo1 expression and Ca²⁺ responses in astrocytes. (A, B) Rat astrocyte Piezo1 mRNA and protein expression increased following cytokine treatment (IL‐1α (IL1α) at 10 ng/mL, IL‐1β (IL1β) at 10 ng/mL, TNF‐α (TNFα) at 20 ng/mL), both separately and in combination (Comb or Cyto Comb). N = 7, N = 6. (C) Piezo1 protein expression increased with cytokine treatment in human IPSC‐derived astrocytes. N = 5. (D) Pretreatment with a cytokine combination for one day (Cyto1d) significantly enhanced Piezo1 activator, Yoda1‐induced Ca²⁺ responses in astrocytes, and this effect was blocked by Piezo1 siRNA (siPiezo1), as indicated by intracellular fluorescence using the Ca²⁺ indicator Fluo‐4 (5 μg/mL). Scale bar: 20 μm. (E) Representative traces of calcium responses in each group following Yoda1 application. (F, G) Summary data of intracellular calcium responses showed that enhanced Ca²⁺ response with Cyto1d treatment is blocked by Piezo1 siRNA (siPz1). N = 18 ROI from three biological replicates in each group. ROI: Region of interest. (H,I) Ca²⁺ response with Cyto1d treatment was also blocked by the Piezo1 channel blocker gadolinium (Gd³⁺; 30 μM) and calcium‐free ACSF (No Ca²⁺). N = 18 ROI from three biological replicates in each group. *p < 0.05, ***p < 0.001, ****p < 0.0001 and ns is statistically not significant. Data are shown as mean ± SEM.

Since Piezo1 functions as a Ca²⁺‐permeable ion channel and Ca²⁺ signaling plays a pivotal role in regulating astrocyte function (Chi et al. 2022; Velasco‐Estevez, Rolle, et al. 2020; Yu et al. 2023), we next investigated the impact of cytokines on Piezo1‐mediated Ca²⁺ responses. Using live Ca²⁺ imaging, we examined the effect of the Piezo1 activator Yoda1 on intracellular Ca²⁺ levels in astrocytes. Astrocyte real time intracellular Ca²⁺ levels were visualized by using Fluo‐4 AM, a cell‐permeable fluorescent Ca²⁺ indicator. We found that without cytokine (IL‐1α at 10 ng/mL, IL‐1β at 10 ng/mL, TNF‐α at 20 ng/mL) treatment, Yoda1 induces a modest, transient Ca²⁺ response characterized by short events (Figure 4E). However, cytokine‐treated astrocytes displayed significantly heightened intracellular Ca²⁺ responses to Yoda1 (Figure 4D,E). Additionally, the pattern of the Ca²⁺ response was also changed to display long‐lasting and more pronounced Ca²⁺ events (Figure 4E). These elevated intracellular Ca²⁺ responses due to cytokine treatment were effectively attenuated by Piezo1 siRNA (Figure 4F,G), Piezo1 channel blocker gadolinium (Gd³⁺), and calcium‐free ACSF (Figure 4H,I), confirming that the cytokine‐induced enhancement of Ca²⁺ signaling was mediated through Piezo1 stimulation and extracellular Ca²⁺ influx. Piezo1 siRNA induced Piezo1 expression knockdown was confirmed in both mRNA and protein level in astrocyte cell cultures before conducting live Ca²⁺ imaging (Figure S3A,B). Yoda2, another Piezo1 activator, with greater selectivity and stability compared to Yoda1 (Kinsella et al. 2024; Parsonage et al. 2023), was also tested in these experiments for its effects on Ca²⁺ responses in astrocytes. Under control conditions, Yoda2 induced larger Ca²⁺ responses than Yoda1, characterized by long‐lasting Ca²⁺ events rather than short Ca²⁺ transients (Figure S3C,D). Similar to Yoda1, Yoda2‐evoked Ca²⁺ responses in astrocyte cultures were significantly enhanced when pretreated with cytokines, and this increase was abolished by Piezo1 siRNA or Gd³⁺(Figure S3C–F).

To determine whether cytokine‐induced upregulation of Piezo1 expression and Ca²⁺ activity observed in rat astrocytes also applies to human cells, we utilized a human iPSC‐derived neurosphere model. This model, which possesses a 3D architecture and multiple cell types (including neurons and astrocytes), thereby more closely mimics the in vivo CNS microenvironment (Wendt, Lin, et al. 2025). Note that these neurospheres did not contain microglia. When neurospheres were exposed to conditioned media derived from LPS‐stimulated human microglia (LPS‐hMCM), Piezo1 expression was significantly upregulated (Figure S3G). Human cytokine combination (IL‐1α, IL‐1β, and TNF‐α) also significantly increased Piezo1 protein expression in neurospheres (Figure S3H). To monitor astrocytic Ca²⁺ dynamics within neurospheres, we employed an AAV‐mediated expression of the calcium indicator GCaMP8s specifically in astrocytes using the GfaABC1D promoter (Figure S3I). Baseline Ca²⁺ activity and Yoda2‐evoked response were then recorded under control and cytokine‐pretreated conditions. The results showed cytokine pretreatment significantly increased baseline Ca²⁺ activity and Yoda2‐evoked Ca²⁺ response in neurosphere astrocytes (Figure S3J,K). In addition, Yoda2‐induced Ca²⁺ response was blocked by Piezo1 inhibitor Gd³⁺(Figure S3L,M). Collectively, these results demonstrate that cytokines released by microglia enhance astrocytic Piezo1 expression and strongly potentiate Piezo1‐mediated Ca²⁺ signaling in both rodent and human astrocytes.

3.5. Microglia Depletion Reduces Astrocytic Piezo1 Expression in 5xFAD Mice

Our experiments in astrocyte cultures demonstrated that astrocytic Piezo1 upregulation is driven by microglia‐released proinflammatory cytokines. Since Piezo1 expression is elevated in 5xFAD mice (see Figure 1), where microglia are also activated, we investigated whether microglia depletion in these mice could diminish astrocytic Piezo1 expression. To test this, we depleted microglia in 5xFAD mice by including PLX3397 in their diet, which has a high efficiency for depleting microglia in many studies (Sosna et al. 2018; Spangenberg et al. 2016; Wendt et al. 2022). Mice fed a control diet without PLX3397 served as controls. In our initial immunostaining in 5xFAD mice, we observed obvious upregulation of Piezo1 starting after 6 months of age (Figure S4A,B), so we chose 6 months old 5xFAD and fed them with the PLX3397 and control diet for 3 months (from about 6 to 9 months of age). Following the PLX3397 treatment, microglia numbers in the brains of 5xFAD mice were significantly reduced (Figure 5A,B). Additionally, brain proinflammatory cytokine levels were markedly decreased in PLX3397‐treated mice compared to controls (Figure 5C,D).

FIGURE 5.

Microglia depletion in 5xFAD mice by feeding PLX3397 chow decreased Piezo1 expression in astrocytes. (A) Immunostaining of Iba1 showed a decrease in the number of microglia in PLX3397 chow‐fed 5xFAD mice. (B) Quantitative analysis of the proportion of microglia depletion. N = 10 from 10 mice. (C) Cytokine mRNA expression in brain tissue was decreased in PLX3397 chow‐fed 5xFAD mice. N = 8. (D) ELISA result showed cytokine protein expression was also decreased in PLX3397 chow‐fed 5xFAD mice brain. N = 5. (E, F) Piezo1 mRNA and protein expression were significantly reduced in the brain tissue of PLX3397 chow‐fed 5xFAD mice compared to control chow‐fed 5xFAD mice. N = 11, N = 7. (G) Immunostaining results showed decreased Piezo1 expression in the brain tissue from PLX3397‐fed 5xFAD mice compared to control chow‐fed 5xFAD mice. (H) Quantitative analysis data showed Piezo1 mean fluorescence intensity was decreased in the brain tissue from PLX3397‐fed 5xFAD mice compared to control chow‐fed 5xFAD mice. N = 13. (I) Piezo1 mRNA expression was decreased in the flow cytometry sorted astrocytes from PLX3397 chow‐fed 5xFAD mice compared to control chow‐fed 5xFAD mice. N = 7. *p < 0.05, **p < 0.01, ****p < 0.0001. Data are shown as mean ± SEM.

We then examined Piezo1 expression in 5xFAD mice with and without microglia depletion. RT‐qPCR and Western blot analysis revealed a significant reduction in Piezo1 mRNA and protein levels in the brain tissues of PLX3397‐fed 5xFAD mice (Figure 5E,F). To assess astrocyte‐specific Piezo1 expression, we performed immunostaining for Piezo1 in brain slices. The mean fluorescence intensity of Piezo1 in GFAP positive astrocytes was significantly decreased in the PLX3397‐fed 5xFAD mice (Figure 5G,H). To more accurately assess astrocyte‐specific Piezo1 expression levels, we isolated astrocytes from 5xFAD mice using fluorescence‐activated cell sorting (FACS) with the astrocyte‐specific GLAST antibody and performed RT‐qPCR analysis on the isolated cell population. Consistent with our immunostaining observations, isolated astrocytes from PLX3397 chow‐fed 5xFAD mice exhibited a significant reduction in Piezo1 mRNA expression compared to controls (Figure 5I). Together, these results demonstrate that microglia are essential drivers of Piezo1 upregulation in the astrocytes of 5xFAD mice, likely through the release of proinflammatory cytokines. Depleting microglia with PLX3397 not only reduced brain proinflammatory cytokine levels but also markedly decreased Piezo1 expression at the whole‐brain and astrocyte‐specific levels. These findings highlight a critical microglia‐to‐astrocyte signaling axis that may contribute to neuroinflammatory regulation and identify Piezo1 as a potential mediator of microglia–astrocyte interactions in AD.

Previous studies show that microglial depletion alters Aβ pathology and behavioral outcomes in AD mouse models (Spangenberg, Lee, et al. 2016; Ferreira et al. 2026; Son et al. 2020). However, most 5xFAD studies focus on either early (~1.5 months) (Spangenberg et al. 2019; Sosna et al. 2018), or late (~10 months) stages (Spangenberg, Lee, et al. 2016; Son et al. 2020), corresponding to pre‐plaque or advanced pathology. In contrast, in the present study, we depleted microglia at 6 months of age, representing an intermediate stage of disease progression when amyloid pathology and neuroinflammatory responses are already established. Therefore, we investigated whether microglial depletion at this stage influences both amyloid pathology and behavioral performance in 5xFAD mice. Our results showed no significant differences in either the number or size of Aβ plaques between PLX3397 and control diet‐fed 5xFAD mice (Figure S5A,B). Behavioral testing showed that total distance traveled and time spent in the center zone in the open field test were unchanged; however, entries into the center zone were significantly decreased in the PLX3397 chow‐fed 5xFAD mice compared to control chow‐fed 5xFAD mice, indicating increased anxiety‐like behavior following microglia depletion (Figure S5C–E). No differences were observed in spontaneous alternation in the Y‐maze test (Figure S5F) or in freezing behavior in the fear conditioning test between the control and PLX3397‐fed 5xFAD mice (Figure S5G,H). These findings indicate that microglia depletion at the 6 month intermediate stage is associated with increased anxiety‐like behavior but does not significantly alter Aβ plaque burden or spatial working memory and contextual fear memory in 5xFAD mice. We observed that microglia depletion reduced astrocytic Piezo1 expression, suggesting that microglia–astrocyte signaling may influence behavioral outcomes independently of detectable changes in Aβ pathology.

3.6. Context‐Dependent Effects of Piezo1 on Astrocyte Reactivity and Astrocyte–Microglia Crosstalk

Astrocytes play a critical role in modulating neuroinflammation under pathological conditions (Colombo and Farina 2016), and a previous study has reported that Piezo1 can regulate astrocytic cytokine and chemokine expression (Velasco‐Estevez, Rolle, et al. 2020). In this study, we investigated how Piezo1 influences astrocyte inflammatory responses under basal conditions and under cytokine pretreated conditions. The expression of IL‐1β, IL‐6, TNF‐α, CCL2, and C3 in astrocytes were assessed following Piezo1 activation by Yoda2 (4 h treatment) or Piezo1 knockdown by using Piezo1 siRNA (3 days post transfection) in both conditions. Piezo1 activation by Yoda2 in normal conditions did not change the expression of TNF‐α, CCL2, and C3, while Yoda2 treatment in cytokine pretreated conditions significantly decreased astrocyte TNF‐α, CCL2, and C3 expression (Figure 6A–C). Similarly, Piezo1 knockdown under basal conditions had no effect on TNF‐α and CCL2 expression. However, in cytokine pretreated astrocytes, Piezo1 knockdown significantly increased TNF‐α and CCL2 expression (Figure 6D,E). Different from the pattern of TNF‐α and CCL2, the expression of C3 was reduced by Piezo1 knockdown under cytokine pretreated conditions (Figure 6F). Interestingly, the expression of IL‐1β and IL‐6 in astrocytes remained unchanged by either Yoda2 application or Piezo1 knockdown in both conditions (Figure 6G–J).

FIGURE 6.

Piezo1 regulates astrocyte reactivity and astrocyte‐microglia crosstalk in a context‐dependent manner. (A–C) In normal conditions, activation of Piezo1 with Yoda2 (1 μM) did not change the expression of TNF‐α, CCL2, or C3 in rat astrocytes. However, when pretreated with cytokine (IL‐1α at 10 ng/mL, IL‐1β at 10 ng/mL, TNF‐α at 20 ng/mL) for 1 day, Yoda2 treatment significantly decreased TNF‐α, CCL2, and C3 expression. N = 6. (D, E) In normal basal states, Piezo1 knockdown in astrocytes via siRNA did not affect TNF‐α and CCL2 expression. In contrast, following 1 day of cytokine pretreatment, Piezo1 knockdown significantly increased both TNF‐α and CCL2 expression in rat astrocytes. N = 5. (F) C3 expression was significantly decreased by Piezo1 knockdown under cytokine pretreated conditions. N = 5. (G–J) IL‐1β and IL‐6 expression levels were not affected by either Yoda2 treatment or Piezo1 knockdown under normal or cytokine pretreated conditions. N = 6, N = 5. (K, L) qPCR and western blot analyses showed that GFAP expression in rat astrocytes was significantly increased by Piezo1 knockdown under basal conditions but not under cytokine pretreated conditions. Notably, cytokine pretreatment alone significantly reduced GFAP expression in rat astrocytes. N = 3, N = 4. (M, N) Conditioned medium from Piezo1 knockdown astrocytes under basal normal conditions (siP_ACM) did not alter pro‐inflammatory cytokine expression in rat microglia. In contrast, conditioned medium from cytokine‐pretreated Piezo1 knockdown astrocytes (siP_Cyto ACM) significantly increased microglial pro‐inflammatory cytokines expression. N = 6, N = 3. *p < 0.05, **p < 0.01, ***p < 0.001 and ns is statistically not significant. Data are shown as mean ± SEM.

We further investigated the role of Piezo1 in regulating astrocyte reactivity by examining GFAP expression following Piezo1 knockdown. Under basal conditions, Piezo1 knockdown significantly increased both GFAP mRNA and protein levels. However, this effect was not observed in cytokine‐pretreated conditions. Notably, cytokine pretreatment itself significantly reduced GFAP expression at both mRNA and protein levels (Figure 6K,L). Given that Piezo1 activation decreased TNF‐α and CCL2 expression, while Piezo1 knockdown increased their expression, we hypothesized that astrocytic Piezo1 may exert anti‐inflammatory effects. To test this, we treated microglia with conditioned medium derived from control or Piezo1 knockdown astrocytes under either basal or cytokine‐pretreated conditions. Piezo1 knockdown in basal astrocytes (siP_ACM) did not alter microglial cytokine expression (Figure 6M), consistent with minimal effects of Piezo1 modulation under normal conditions. In contrast, conditioned media from cytokine‐pretreated Piezo1 knockdown astrocytes (siP_Cyto ACM) significantly increased microglial expression of cytokines (Figure 6N), indicating that Piezo1 activity in reactive astrocytes restrains pro‐inflammatory signaling. Together, these findings demonstrate that Piezo1 differentially regulates astrocyte reactivity and gene expressions depending on the inflammatory context and its expression level. Cytokine priming not only upregulates Piezo1 expression but also appears to enhance its anti‐inflammatory function in astrocytes, thereby contributing to the suppression of neuroinflammatory responses.

4. Discussion

Piezo1 is a vital mechanosensitive ion channel expressed in a wide variety of tissues, reflecting its broad importance in maintaining cellular homeostasis and enabling adaptive responses to changes in the environment (Zong et al. 2023). Despite its established importance in the CNS, there is a noticeable gap in our understanding of how the expression and function of Piezo1 change in brain pathological conditions, such as Alzheimer's disease or during brain inflammation. Investigating how astrocytic Piezo1 responds to such pathological conditions could provide valuable insights into its role in disease progression and highlight its potential as a therapeutic target for conditions involving mechanical dysregulation in the brain.

To investigate the factors contributing to Piezo1 upregulation in vitro, we prepared pure rat primary astrocyte and microglia cultures and treated them with oligomeric Aβ (oAβ) and LPS to assess their effects on Piezo1 expression. Piezo1 mRNA (initially called Fam38A) was first reported in Aβ‐associated astrocytes by Satoh K et al. in 2006 (Satoh et al. 2006). Subsequent work demonstrated that conditioned medium from Aβ‐treated microglia can upregulate Piezo1 expression in astrocytes in vitro (Velasco‐Estevez et al. 2018). In addition, multiple studies have shown that LPS stimulation also induces Piezo1 expression in astrocytes (Velasco‐Estevez, Rolle, et al. 2020; Yu et al. 2023). These findings suggest that astrocytic Piezo1 is responsive to Aβ‐ and LPS‐related inflammatory stimuli in vitro. However, a limitation of these studies is that they did not fully exclude the possibility of microglial contamination in primary astrocyte cultures, which may confound the interpretation of astrocyte‐specific Piezo1 expression. Microglial contamination in primary astrocyte cultures is a well‐documented concern in astrocyte studies (Saura 2007; Kumamaru et al. 2012), In our study, we addressed this issue by pre‐incubating astrocyte cultures with the CSF receptor 1 inhibitor, PLX3397. We found that treating astrocyte cultures with PLX3397 prior to cell passaging significantly reduced microglial contamination (Figure S1). Our iNOS Western blot results revealed a substantial reduction in iNOS expression (~70%) following the removal of microglial contamination from cytokine‐treated astrocyte cultures (Figure S1F). Although both microglia and astrocytes can produce iNOS during inflammation, iNOS is widely recognized as a marker of pro‐inflammatory activation in microglia (Lisi et al. 2017; Miao et al. 2025; Koss et al. 2024). A previous study demonstrated that iNOS expression is predominantly associated with activated microglia, whereas astrocytic iNOS expression is comparatively limited, delayed, and dependent on microglial signaling (Chen et al. 2015). Our findings further support that iNOS expression is primarily linked to activated microglia rather than astrocytes.

Treating rat primary microglia cultures with LPS or oAβ induced microglial activation and upregulation of Piezo1 expression in microglia (Figure 2). In contrast, the same treatments applied to pure rat primary astrocyte cultures had minimal effects on astrocytic Piezo1 expression (Figure 3). In addition to rat primary cultures, we extended our analysis to human iPSC‐derived astrocytes and microglia to assess the translational relevance of our findings. Following treatment with oAβ or LPS, we observed a significant upregulation of Piezo1 expression in human microglia, whereas no comparable change was detected in astrocytes (Figure 6A–D). Microglia, the resident immune cells in the CNS, are the predominant responders to brain potential pathogens (Eyo and Wu 2019; Ousman and Kubes 2012; Ransohoff and Engelhardt 2012). Our results are consistent with this and show that Piezo1 is more robustly responsive to inflammatory or disease‐associated stimuli in microglia than in astrocytes. This cell type–specific regulation may reflect intrinsic differences in mechanosensitivity or immune signaling pathways between microglia and astrocytes. The lack of Piezo1 upregulation in astrocytes under the same conditions may indicate that astrocytic Piezo1 is either less sensitive to these stimuli or requires additional context‐dependent signals for activation. When astrocytes were treated with a higher LPS concentration (500 ng/mL, five‐fold higher), qPCR results revealed a modest 1.3‐fold increase in astrocytic Piezo1 mRNA expression (Figure S6E). This increase was still much lower than the 5.0‐fold increase observed in microglia with 100 ng/mL LPS treatment(Figure 2).

Exposing rat primary astrocyte cultures to conditioned media derived from LPS‐ (100 ng/mL) or oAβ‐pretreated microglia (LPS‐MCM or oAβ‐MCM) significantly increased Piezo1 expression in astrocytes (Figure 3). These findings show that astrocytic Piezo1 expression is strongly driven by factors released from activated microglia and not directly driven by inflammatory stimuli alone at the concentrations tested. A recent study demonstrated that TNF‐α treatment significantly increases Piezo1 expression in human astrocytes, suggesting that pro‐inflammatory cytokines are key upstream regulators of astrocytic Piezo1 (Garcia et al. 2023). Liddelow et al. previously demonstrated that activated microglia release IL‐1α, TNF‐α, and C1q, which together are sufficient to induce reactive astrocytes (Liddelow et al. 2017). Other cytokine combinations, such as IL‐1β with TNF‐α or IFN‐γ with TNF‐α, have also been used to generate distinct reactive astrocyte states (Hyvärinen et al. 2019; Benveniste et al. 1989). The transition to a reactive phenotype is accompanied by pronounced morphological changes (Burda and Sofroniew 2017), which may potentially activate mechanically sensitive Piezo1 channels and drive their upregulation in astrocytes. Consistent with these findings, our results demonstrate that pro‐inflammatory cytokines, including IL‐1α, TNF‐α, and IL‐1β, either individually or in combination, significantly induce Piezo1 expression in astrocytes (Figure 4). This microglia‐astrocyte crosstalk is especially important in the context of neuroinflammatory conditions in neurological disease, where sustained microglia activation and pro‐inflammatory cytokine signaling may drive Piezo1 expression and cause functional changes in astrocytes.

While our findings demonstrated that cytokines increase Piezo1 expression in astrocytes, we also observed that cytokine treatment significantly enhanced Piezo1‐mediated calcium responses. The precise biological mechanisms and consequences of this altered Ca²⁺ activity require further investigation. Interestingly, when we used a calcium‐free perfusion solution, Yoda1 still induced an initial calcium transient, which indicates that Yoda1 could mobilize internal Ca²⁺ stores, such as endoplasmic reticulum (ER) and/or mitochondria, rather than relying solely on extracellular influx. This is consistent with prior studies demonstrating that Piezo1 activation by Yoda1 induces ER calcium release via ER‐localized Piezo1 (Velasco‐Estevez, Rolle, et al. 2020; Kim et al. 2022). Furthermore, previous studies have reported elevated or dysregulated Ca²⁺ signaling in astrocytes of AD mouse models (Kuchibhotla et al. 2009; Shah et al. 2022; Huffels et al. 2022). For instance, in vivo two‐photon imaging of APP/PS1 mice revealed increased resting Ca²⁺ levels, more frequent astrocytic transients, and more intercellular Ca²⁺ waves near amyloid plaques (Kuchibhotla et al. 2009). Since Piezo1 is a calcium‐permeable ion channel and is markedly upregulated in astrocytes in AD mice (Figure 1), further studies are needed to determine whether Piezo1 modulates astrocytic Ca²⁺ signaling in AD.

To clarify the role of microglia and their secreted cytokines in regulating astrocytic Piezo1 expression in vivo, we depleted microglia in 5xFAD mice using the CSF1 receptor specific inhibitor PLX3397 (Elmore et al. 2014; Stoll et al. 2024). PLX3397 administration led to a significant reduction in microglia population and brain pro‐inflammatory cytokine levels, as well as Piezo1 expression in astrocytes (Figure 5). Notably, even at a high dose of 600 mg/kg PLX3397 chow, approximately 20% of activated microglia remained in the brains of 5xFAD mice. These microglia were predominantly plaque‐associated and resistant to PLX3397‐mediated depletion, which is consistent with previous reports showing that Aβ plaque‐associated microglia are relatively refractory to CSF1R inhibition (Lodder et al. 2021; Han et al. 2022; Le et al. 2024). Interestingly, we observed that microglial depletion in 5xFAD mice between 6 and 9 months of age altered behavioral outcomes (Figure S5). Mice treated with PLX3397‐containing chow displayed increased anxiety‐like behavior, while cognitive performance remained largely unchanged. Microglia are known to play critical roles in synaptic pruning and the maintenance of neuronal connectivity. Disruption of these functions through sustained depletion can lead to alterations in circuit integrity and, consequently, behavioral changes (Ferreira et al. 2026). Although some studies have reported improved learning and memory following microglial depletion in AD models (Spangenberg, Lee, et al. 2016; Spangenberg et al. 2019; Dagher et al. 2015), these outcomes are highly variable and may depend on the timing, duration, or degree of depletion. In our study, the lack of significant improvement in spatial or associative memory suggests that chronic PLX3397 administration at a stage of established pathology may not be sufficient to rescue hippocampal‐dependent cognitive deficits.

Finally, our study demonstrated that Piezo1 regulates astrocyte gene expression and astrocyte–microglia interactions in a context‐dependent manner (Figure 6). A previous study reported that pharmacological activation of Piezo1 with Yoda1 significantly reduced IL‐1β, TNF‐α, and CX₃CL1 production in primary mouse astrocytes following 2 days of LPS pretreatment, supporting an anti‐inflammatory role of Piezo1 in reactive astrocytes (Velasco‐Estevez, Rolle, et al. 2020). Consistent with and extending these findings, we show that under basal conditions, Piezo1 knockdown leads to increased GFAP expression. This suggests that Piezo1 contributes to maintaining astrocyte homeostasis and helps restrain low‐level reactive changes. Pharmacological activation of Piezo1 under basal conditions produced minimal effects, likely reflecting the relatively low expression or activity of Piezo1 in resting astrocytes. In the cytokine‐primed state, activation of Piezo1 suppressed the expression of pro‐inflammatory genes, whereas Piezo1 knockdown enhanced their expression. Our data show that both GFAP mRNA and protein levels were reduced in astrocytes following one day of cytokine combination treatment. Previous in vitro studies have also reported a decrease in GFAP expression in astrocytes in response to pro‐inflammatory cytokine stimulation (Hyvärinen et al. 2019; Ronco et al. 2014; John et al. 2005). These findings suggest the existence of a negative feedback mechanism in which inflammatory cytokines upregulate Piezo1 expression, and subsequent Piezo1 activation acts to limit excessive inflammatory signaling in reactive astrocytes.

C3, a key component of the complement pathway and a commonly used marker of A1 reactive astrocytes (Liddelow et al. 2017), exhibited a distinct regulatory pattern: both Piezo1 activation and knockdown reduced C3 expression in cytokine‐primed astrocytes, suggesting that complement pathway genes may require a balanced level of Piezo1‐dependent Ca²⁺ signaling and are regulated through mechanisms that are at least partially independent of classical inflammatory cytokine pathways. However, as C3 alone does not fully define astrocyte phenotype, additional markers will be required to determine whether Piezo1 regulates A1 astrocyte identity.

Overall, our study provides new insights into the regulation of Piezo1 expression in astrocytes and highlights its state‐dependent role in maintaining astrocyte homeostasis and reactivity under both physiological and neuroinflammatory conditions. Several key findings underscore the novelty of this work. First, we clearly demonstrate in vivo upregulation of Piezo1 in astrocytes in pathological contexts, including LPS‐injected mice and the 5xFAD mouse model. Second, by establishing an experimental approach that effectively minimizes microglial contamination in primary astrocyte cultures, we reveal that direct treatment with low concentrations of LPS or oAβ does not significantly increase astrocytic Piezo1 expression, whereas conditioned medium from LPS‐ or oAβ‐treated microglia robustly induces Piezo1, highlighting the importance of microglia–astrocyte signaling in regulating astrocytic Piezo1. Third, we provide in vivo evidence that microglia‐derived factors are key drivers of Piezo1 upregulation in the 5xFAD model, further supporting the pathological relevance of our findings. Importantly, we extend these observations to human systems using neurosphere models. Cytokine pretreatment significantly enhanced both baseline and Yoda2‐induced calcium activity in astrocytes within neurospheres, suggesting that microglia–astrocyte crosstalk and its impact on Piezo1 function are conserved in complex human 3D culture systems. Finally, our study establishes Piezo1 as a context‐dependent regulator of astrocyte function. We show that Piezo1 differentially modulates astrocyte reactivity, inflammatory signaling, and complement pathway gene expression under basal versus inflammatory conditions, highlighting its multifaceted and state‐specific role in shaping astrocyte responses.

Our study does not exclude the possibility that additional microglia‐derived factors, beyond IL‐1α, TNF‐α, and IL‐1β, may influence Piezo1 expression in astrocytes. In vitro cell culture models represent a simplified system that lacks the complexity of the brain microenvironment, such as the presence of blood vessels. Although our qPCR analysis of flow cytometry‐sorted astrocytes revealed a significant reduction in astrocytic Piezo1 expression following microglia depletion in 5xFAD mice (from 100% to 76%), a substantial amount of Piezo1 expression still remains in astrocytes. This suggests that other factors within the brain, beyond microglia and their released cytokines, may also contribute to Piezo1 upregulation in 5xFAD mice. Given the high expression of Piezo1 in astrocytes under inflammatory conditions, further research is needed to fully elucidate its precise role of Piezo1 in maintaining brain function and potential involvement in neurological diseases in vivo.

Author Contributions

Y.B. conceived the study, designed the experiments, performed the majority of the experiments, analyzed and interpreted the data, and wrote the manuscript. H.B.C. prepared and maintained rat mixed primary glial cultures and primary microglial cultures. S.W. and A.J.L. generated and maintained human iPSC‐derived cell cultures and neurospheres, prepared and quantified oligomeric Aβ, and assisted with live calcium imaging experiments in neurospheres. S.W. revised the manuscript. M.T. assisted with fluorescence‐activated cell sorting of astrocytes. I.C. contributed to the acquisition of behavioral data. C.J.G. provided advice on scientific illustration preparation. H.B.N. provided funding. B.A.M. conceived and supervised the study, secured funding, and critically revised the manuscript. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the Canadian Institutes of Health Research (CIHR) grant (F23‐01079), the Canadians for Leading Edge Alzheimer Research Foundation (CLEAR) (F22‐04208), and the Aune McArthur Alzheimer Research Fund (F21‐00358).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Obtaining highly pure primary astrocyte and microglia cell cultures.

Figure S2: LPS or oAβ treatment increased microglia cytokine expression.

Figure S3: Cytokine treatment increased Yoda2 induced Ca²⁺ responses in rat and human astrocytes.

Figure S4: Piezo1 expression in different ages of 5xFAD mice.

Figure S5: PLX3397 diet induced anxiety‐like behavior in 5xFAD mice.

Figure S6: Effects of LPS and oAβ treatment on human iPSC‐derived astrocytes and microglia, and the impact of a higher concentration of LPS on Piezo1 expression in rat astrocytes.

Figure S7: Time‐course analysis of LPS and oAβ treatment in rat primary microglia cultures.

Figure S8: Full‐length western blot of Piezo1 in rat cell cultures and mice brain tissue.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.