The ZNF148‐ZEB1‐AS1‐IGF2BP2‐NOD2 Axis Drives Microglial Antipneumococcal Immunity in Bacterial Meningitis

ABSTRACT

Streptococcus pneumoniae (Spn) meningitis remains a lethal central nervous system (CNS) infection with limited therapies. This study identifies the lncRNA ZEB1‐AS1 as a central coordinator of microglial immunity against Spn through a multi‐tiered regulatory cascade. Transcriptomic analysis revealed Spn‐induced ZEB1‐AS1 upregulation in human microglia, driven by ZNF148, which directly binds its promoter. Functional interrogation demonstrated that ZEB1‐AS1 knockdown impairs bacterial clearance and pro‐inflammatory cytokine production (IL‐1β, IL‐6, TNF‐α, p < 0.01), while its overexpression amplifies these responses. Crucially, ZEB1‐AS1 recruits the m6A reader IGF2BP2 to stabilize NOD2 mRNA in cytoplasmic complexes, extending transcript stability. This molecular scaffolding enables NOD2‐dependent antimicrobial functions, as evidenced by rescue experiments in which IGF2BP2 overexpression reversed ZEB1‐AS1 deficiency phenotypes. In vivo, microglial manipulation of the murine homolog Zeb1‐os1 regulated cerebral Spn burdens, NOD2 expression, and infection‐induced cognitive outcomes in both directions. The tripartite ZEB1‐AS1/IGF2BP2/NOD2 interaction was validated by RNA pulldown and co‐immunoprecipitation, establishing a linear pathway from ZNF148‐mediated transcriptional activation to IGF2BP2‐dependent mRNA stabilization. Collectively, this ZNF148 to ZEB1‐AS1 to IGF2BP2 to NOD2 axis bridges the gap between transcriptional and post‐transcriptional immune regulation, proposing IGF2BP2's RNA‐binding domain as a therapeutic target against drug‐resistant Spn meningitis.

ZNF148 upregulates ZEB1‐AS1, which scaffolds IGF2BP2 to stabilize NOD2 mRNA, enhancing microglial anti‐pneumococcal immunity via cytokine production and bacterial clearance in bacterial meningitis.

1. Introduction

Bacterial meningitis remains one of the most life‐threatening infectious diseases. In some underdeveloped regions, its mortality rate can reach as high as 54%, and up to 24% of survivors suffer from chronic neurological sequelae (Hasbun 2022). Among its subtypes, pneumococcal meningitis caused by Spn represents the most prevalent form, underscoring the urgent need for novel therapeutic strategies (Kim 2010; Gil et al. 2022; Mook‐Kanamori et al. 2011; Fiore et al. 2000). Within the central nervous system (CNS), microglia—constituting 5%–12% of brain cells—serve as specialized resident macrophages orchestrating frontline immune surveillance and pathogen clearance (Borst et al. 2021; Prinz et al. 2019; Escoubas and Molofsky 2024). Upon Spn invasion, microglia deploy an arsenal of pattern recognition receptors (PRRs), including Toll‐like receptors (TLRs), nucleotide‐binding oligomerization domain (NOD)‐like receptors (NLRs), and intracellular sensors, to detect pathogen‐associated molecular patterns. These interactions trigger phagocytosis and cytokine signaling cascades, initiating a coordinated immune defense (Escoubas and Molofsky 2024; Castro‐Gomez and Heneka 2024). Notably, NOD2—an NLR family member that recognizes bacterial muramyl dipeptide (MDP)—has emerged as a critical mediator that links innate immunity with inflammatory regulation by sensing bacterial components and activating downstream signaling pathways. Our prior work revealed NOD2's multifaceted roles in Spn meningitis, spanning bacterial clearance and immune homeostasis, yet its precise regulatory mechanisms in microglia remain enigmatic (Wang, Zhang, et al. 2022; Wang et al. 2020; Liu et al. 2010).

Recent advances highlight the pivotal role of long non‐coding RNAs (lncRNAs) in fine‐tuning immune responses. Among these, ZEB1‐antisense 1 (ZEB1‐AS1), initially characterized as an oncogenic driver through its regulation of epithelial‐mesenchymal transition (EMT) (Chen et al. 2018; Ali and Grote 2020; Ferrer and Dimitrova 2024; Li et al. 2018; Chen and Shen 2020; Chen et al. 2025), exhibits broader functional relevance. Beyond cancer, ZEB1‐AS1 stabilizes target mRNAs via RNA‐binding proteins (RBPs), modulates inflammatory pathways, and resists microbial invasion, as evidenced in Chlamydia trachomatis infection and atherosclerosis models (Luo et al. 2022; Zhong et al. 2021; Chen et al. 2021; Xu et al. 2019). Crucially, Xu et al. demonstrated ZEB1‐AS1's capacity to stabilize NOD2 mRNA in endothelial cells, hinting at its potential immunoregulatory roles in infection (Xu et al. 2019). However, whether this axis operates in microglia during pneumococcal meningitis, and how ZEB1‐AS1 itself is regulated under infectious conditions, remain unexplored.

Our current work unravels a hierarchical regulatory axis centered on ZEB1‐AS1. This axis integrates upstream transcriptional control with downstream mRNA stabilization. We identified ZNF148, a zinc‐finger transcription factor previously associated with DNA damage responses and EMT (Chua et al. 2023; Kim et al. 2022; Fang et al. 2017; Hahn and Hermeking 2014), as the master switch driving ZEB1‐AS1 induction during Spn infection. Mechanistically, Spn‐derived signals activate ZNF148, which binds directly to the ZEB1‐AS1 promoter, triggering its transcriptional upregulation—a novel pathway linking bacterial detection to lncRNA reprogramming. This discovery expands ZNF148's functional repertoire beyond cancer biology, positioning it as a transcription factor acting as a CNS‐specific stress sensor.

Insulin‐like growth factor 2 mRNA‐binding protein 2 (IGF2BP2) is an RNA‐binding protein that governs post‐transcriptional regulation by stabilizing target mRNAs through recognition of N6‐methyladenosine (m6A) modifications and AU‐rich elements (Lederer et al. 2014; Li et al. 2024; Nan et al. 2024). In inflammatory diseases, IGF2BP2 exhibits context‐dependent roles: it exacerbates rheumatoid arthritis by stabilizing GSTM5 and STAT1 mRNAs (Nan et al. 2024; Yang et al. 2023), yet ameliorates colitis through PPARγ‐mediated anti‐inflammatory signaling (Wang, Ji, et al. 2021). Notably, IGF2BP2 collaborates with lncRNAs like ZEB1‐AS1 to amplify pathogen‐sensing pathways—mechanisms implicated in cancer but unexplored in neuroinflammation (Liu et al. 2022; Wang et al. 2019; Bian et al. 2024).

The induced ZEB1‐AS1 translocates to the cytoplasm, where it acts as a scaffold for IGF2BP2 to stabilize NOD2 mRNA, thereby amplifying microglial phagocytosis and cytokine production.

Our findings bridge critical gaps in understanding how non‐coding RNAs coordinate microglial immunity. First, we establish ZNF148 as the upstream activator of ZEB1‐AS1 in response to bacterial challenge, thereby resolving the long‐standing question of how pathogen sensing connects to lncRNA induction. Second, we demonstrate the cytoplasmic repurposing of ZEB1‐AS1—traditionally studied for nuclear roles in cancer—as an mRNA stability enhancer in neuroinfection (Yang et al. 2024). Third, by elucidating the spatiotemporal dynamics of this axis (nuclear ZNF148 cytoplasmic ZEB1‐AS1/IGF2BP2 complexes), we provide a paradigm for lncRNA‐mediated regulation across cellular compartments.

2. Materials and Methods

2.1. Cell Culture

The human microglial cell line HMC3 (Wuhan Procell Life Science & Technology Co. Ltd., Catalog No. CL‐0620) was cultured in MEM‐α medium (Gibco) supplemented with 10% fetal bovine serum (FBS, Kangyuan Biology) at 37°C under 5% CO₂ humidified conditions. For experimental standardization, cells were routinely passaged using 0.25% trypsin–EDTA upon reaching 80%–90% confluence and reseeded at a 1:3 to 1:4 split ratio. In all experiments, cells were seeded at a density of 2–5 × 10^5 cells/mL to achieve 60%–70% confluence at intervention timepoints; medium was refreshed every 48 h until endpoint analysis.

2.2. Bacterial Culture and Inoculum Preparation

Streptococcus pneumoniae strains (BNCC337114, BNCC) were cultured on Columbia blood agar plates (HopeBio) at 37°C under 5% CO₂ for 18–24 h. Single colonies with typical α‐hemolytic morphology were selected and inoculated into Todd‐Hewitt broth supplemented with 0.5% yeast extract (THY medium, Sigma‐Aldrich). Primary cultures were incubated aerobically at 37°C until reaching mid‐exponential phase (OD₆₀₀ = 0.4–0.6, corresponding to ~10⁸ CFU/mL, as calibrated by serial dilution plating).

For experimental infections or molecular analyses, bacterial cells were harvested by centrifugation (4000 × g, 10 min, 4°C), washed twice with sterile phosphate‐buffered saline (PBS, pH 7.4), and resuspended in PBS. The final bacterial concentration was quantified using a spectrophotometer (OD600) and verified by plating serial dilutions on Columbia blood agar plates for colony‐forming unit (CFU) enumeration. Aliquots were stored at −80°C in THY medium containing 15% (v/v) glycerol for long‐term preservation.

2.3. Stereotaxic Intracerebroventricular Delivery of AAV in C57BL/6 Mice

2.3.1. Animals

C57BL/6 mice (male, 4 weeks old, 12–18 g) were housed under specific pathogen‐free conditions (22°C ± 1°C, 12 h light/12 h dark cycle) with ad libitum access to food and water. All procedures were approved by Qilu Hospital of Shandong University Animal Care Committee, Protocol QWLL‐2‐24‐134.

2.3.2. Viral Preparation

Recombinant AAV9 vectors (1 × 10¹³ vg/mL, [GeneChem]) or control AAV9‐CAG empty vectors were diluted in sterile artificial cerebrospinal fluid (aCSF: 148 mM NaCl, 3 mM KCl, pH 7.4) and stored at −80°C until use.

2.3.3. Surgical Procedure

Mice were anesthetized by 3% isoflurane for induction, followed by 1.5% for maintenance via a nose cone. Body temperature was maintained at 37°C using a heating pad. The head was fixed in a digital stereotaxic frame (RWD Life Science) with ear bars. After disinfecting the scalp with 75% ethanol, a midline incision was made to expose the skull. The left lateral ventricle was localized using the following coordinates relative to bregma: Anteroposterior (AP), −0.5 mm; Mediolateral (ML), ±1.0 mm; Dorsoventral (DV), −2.3 mm. A 33‐gauge Hamilton syringe (701 N, Hamilton Company) connected to a microinjection pump (UMP3, World Precision Instruments) delivered 2 μL of viral suspension at 100 nL/min. The needle was left in place for 5 min post‐injection to prevent reflux.

Postoperative analgesia was administered via oral ibuprofen (30 mg/kg/day) in drinking water for 24 h. Neurological deficits were assessed every 12 h using a standardized scoring system: 0 (normal), 1 (reduced mobility), 2 (circling behavior), 3 (unilateral paralysis), 4 (lethargy), and 5 (moribund/death).

2.4. Western Blot Analysis

Cellular and tissue proteins were extracted using high‐efficiency R lysis buffer (Solarbio Life Sciences). Protein concentrations were quantified via the BCA assay (Beyotime Biotechnology), followed by mixing with 5× protein loading buffer (Solarbio). Equal amounts of protein (30 μg) were separated on 10% SDS‐PAGE gels (Abbine Scientific) and transferred onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked with 5% skim milk for 1 h at room temperature and then incubated overnight at 4°C with the following primary antibodies: Anti‐NOD2 (1:1000, Cat# 12125, Affinity Biosciences), Anti‐IGF2BP2 (1:1000, Cat# A2749, ABclonal), Anti‐ZNF148 (1:1000, Cat# A7701, ABclonal), and Anti‐GAPDH (1:10,000, Cat# AC033, ABclonal). After washing, membranes were probed with horseradish peroxidase (HRP)‐conjugated secondary antibody (1:5000, Cat# AS014, ABclonal) for 1 h at room temperature. Protein bands were visualized using the SuperFemto ECL Chemiluminescence Kit (Vazyme Biotech) according to the manufacturer's protocol.

2.5. RNA Extraction, Reverse Transcription (RT), and Quantitative Real‐Time PCR (qRT‐PCR)

Total RNA from cells and tissues was isolated using the RNA‐easy Isolation Reagent (Vazyme). Cytoplasmic and nuclear RNA fractions were separated using the Cytoplasmic and Nuclear RNA Purification Kit (Norgen, Cat# 21000). RNA quality was assessed by measuring the purity as A260/A280 ratio between 1.8 and 2.0, and concentration greater than 200 μg/mL using a spectrophotometer. Subsequently, 1 μg of total RNA was reverse‐transcribed into cDNA in a 20 μL reaction volume using HiScript II Q Select RT SuperMix for qPCR (+gDNA wiper) (Vazyme). The reaction was performed under the following conditions: 50°C for 15 min and 85°C for 5 s. qRT‐PCR was performed in a 20 μL reaction system containing 2 μL cDNA template and AceQ Universal SYBR qPCR Master Mix (Vazyme). Thermal cycling parameters included an initial denaturation at 95°C for 5 min, followed by 40 cycles of 95°C for 10 s and 60°C for 30 s. Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as a housekeeping gene. Relative gene expression levels were normalized and calculated using the 2 − ΔΔCt method.

2.6. Immunofluorescence Staining

HMC3 cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.5% Triton X‐100 for 20 min, and blocked with 5% bovine serum albumin (BSA) for 30 min at room temperature. After PBS washes, cells were incubated overnight at 4°C with the primary antibody, Anti‐IGF2BP2 (1:1000, Cat# A2749, ABclonal). Subsequently, cells were incubated with fluorophore‐conjugated secondary antibodies (species‐specific) for 1 h at room temperature. After PBS washes, nuclei were counterstained with DAPI (from Solarbio) for 5 min. Fluorescence images were acquired using a laser scanning confocal microscope.

2.7. RNA Pull‐Down Assay

RNA‐protein interactions were analyzed using a magnetic RNA‐protein pull‐down kit (BersinBio). Briefly, lysates from 2 × 10^7 HMC3 cells were incubated at 37°C for 4 h with a mixture of two separate biotinylated probes targeting ZEB1‐AS1 and NOD2 mRNA. RNA‐bound protein complexes were then captured using streptavidin‐conjugated magnetic beads, and the retrieved proteins were subsequently analyzed by mass spectrometry for proteomic profiling and validated via Western blotting.

2.8. RNA Fluorescence In Situ Hybridization (FISH)

HMC3 cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X‐100. To reduce nonspecific probe binding, pre‐hybridization was performed. Subsequently, cells were hybridized overnight (~16 h) at 42°C with a Cy3‐labeled ZEB1‐AS1 probe (5 μM; synthesized by BersinBio) in hybridization buffer (FISH Kit, BersinBio). Nuclei were counterstained with DAPI (Solarbio), and images were captured using a confocal fluorescence microscope (Leica DMi8).

2.9. RNA Immunoprecipitation (RIP) Assay

RNA‐protein interactions were analyzed using a Magna RIP RNA Immunoprecipitation kit (BersinBio). Briefly, 1 × 10^7 HMC3 cells were lysed in RIP lysis buffer containing protease and RNase inhibitors. Cell lysates were immunoprecipitated overnight at 4°C with 2 μg of anti‐IGF2BP2 antibody (rabbit monoclonal; Cat# A2749, ABclonal) conjugated to Protein G magnetic beads. Normal rabbit IgG (2 μg; Cat# AC005, ABclonal) conjugated to Protein G magnetic beads served as a negative control. RNA‐protein complexes were washed three times with high‐stringency buffer (150 mM NaCl, 0.1% SDS). Co‐precipitated RNAs were extracted using TRIzol reagent (Invitrogen) and quantified via qRT‐PCR with ZEB1‐AS1‐specific primers.

2.10. Enzyme‐Linked Immunosorbent Assay (ELISA)

Supernatants from HMC3 cells stimulated with Streptococcus pneumoniae (Spn; multiplicity of infection [MOI] = 100) for 12 h were collected by centrifugation at 1000 × g for 10 min at 4°C. For cytokine quantification, we used human IL‐1β (SEKH‐0002, Solarbio), IL‐6 (SEKH‐0013, Solarbio), and TNF‐α (SEKH‐0047, Solarbio) ELISA kits. Following the manufacturer's protocols, we incubated standards and samples in pre‐coated 96‐well plates. After terminating the TMB substrate reaction with 2 M H2SO4, optical density (OD) was measured at 450 nm using a microplate reader (Tecan Infinite M200). Sample concentrations were calculated against standard curves generated using 4‐parameter logistic regression.

2.11. Multiplex Fluorescence Immunohistochemistry (mIHC)

Formalin‐fixed paraffin‐embedded (FFPE) tissue sections were processed using the TSAPlus Multiplex Fluorescence Staining Kit (Cat# G1226, Servicebio). Briefly, sections underwent deparaffinization, antigen retrieval (Citrate buffer, 95°C, 20 min), and blocking with 3% BSA for 30 min. Sequential staining was performed as follows: Sections were incubated with rabbit anti‐IBA1 (1:1000, Cat# A19776, ABclonal) at 4°C overnight. Then, HRP‐conjugated goat anti‐rabbit IgG (1:200, Cat# AS014, ABclonal) was incubated at 25°C for 1 h. Signal amplification was achieved by applying iF488‐TSA working solution (Servicebio) for 10 min while protected from light. Afterward, antibody stripping was performed using heat‐mediated antigen elution in 65°C glycine‐HCl buffer (pH 2.0) for 20 min. Following stripping, sections were incubated with rabbit anti‐NOD2 (1:1000, Cat# DF12125, Affinity) at 4°C overnight. Subsequently, HRP‐conjugated goat anti‐rabbit IgG (1:200, Cat# AS014, ABclonal) was incubated at 25°C for 1 h. Signal amplification was then performed by applying iF555‐TSA working solution (Servicebio) for 10 min while protected from light. Nuclei were counterstained with DAPI (1 μg/mL for 5 min), and slides were mounted with anti‐fade medium (Servicebio). Multispectral images were acquired using the VS200 automated slide scanning system (Olympus) with consistent exposure settings applied across all channels during image acquisition.

2.12. In Vitro Bacterial Load and Phagocytosis Assays

2.12.1. Bacterial Load Quantification

HMC3 cells in different treatment groups were infected with Streptococcus pneumoniae (Spn) at a multiplicity of infection (MOI) of 100 for 12 h. Supernatants (1 mL) were collected and serially diluted 10^4‐fold in sterile PBS. Aliquots (10 μL) of each diluted sample were plated in triplicate on Columbia blood agar and incubated at 37°C with 5% CO₂ for 24 h. Colony‐forming units (CFU/mL) were counted to quantify extracellular bacterial load.

2.12.2. Phagocytosis Assay

HMC3 cells were incubated with fluorescein isothiocyanate (FITC)‐labeled Streptococcus pneumoniae (Spn) at an MOI of 100 for 1 h. Phagocytosis was terminated by washing cells with ice‐cold PBS. Extracellular bacterial fluorescence was quenched using 0.04% (w/v) trypan blue for 5 min at 25°C. After three washes with PBS, intracellular fluorescence intensity was measured at 485/538 nm (excitation/emission) using a microplate reader (Tecan Infinite M200). Data were normalized to cell count.

2.13. In Vivo Bacterial Load Quantification in Mice

2.13.1. Intracerebroventricular Infection

C57BL/6 mice were anesthetized and secured in a stereotaxic apparatus (RWD Life Science). A 2 μL suspension containing 1 × 10^5 CFU of Streptococcus pneumoniae (Spn) in sterile PBS was injected into the right lateral ventricle (coordinates: AP −0.5 mm, ML +1.0 mm, and DV −2.3 mm, relative to bregma) at a rate of 0.5 μL/min, using a Hamilton syringe.

2.13.2. Tissue Processing

At 72 h post‐infection, whole brains were harvested under aseptic conditions. Each brain was homogenized individually in 0.5 mL ice‐cold sterile PBS using a sterile glass homogenizer with 20 vertical strokes. The homogenates were then centrifuged at 300 × g for 10 min at 4°C.

2.13.3. Bacterial Enumeration

Following centrifugation, the supernatants were collected and serially diluted in sterile PBS. Aliquots (100 μL) of each dilution were plated on Columbia blood agar and incubated at 37°C with 5% CO₂ for 24 h. Colony‐forming units (CFU/mL) were then quantified.

2.14. Y‐Maze Spontaneous Alternation Test

2.14.1. Apparatus Configuration

The Y‐maze consisted of three opaque polyethylene arms (40 cm long × 8 cm wide × 15 cm high) arranged at 120° angles and labeled as arms A, B, and C for clarity.

2.14.2. Pre‐Test Adaptation

Mice were individually housed in the testing room (23°C ± 1°C, 45% ± 5% relative humidity) for 24 h prior to experiments. For maze adaptation, each mouse was placed in the central area and allowed to explore freely for 10 min without recording. The maze was thoroughly cleaned with 70% ethanol between trials.

2.14.3. Testing Procedure

Mice were randomly placed at the end of one arm, facing the central area. Exploration behavior was recorded for 5 min using the ANY‐maze tracking system (v7.33). An arm entry was recorded when all four paws of the mouse crossed the defined arm threshold. A spontaneous alternation was scored when the mouse made consecutive entries into three different arms (e.g., A → B → C), reflecting its spatial working memory performance.

2.15. Chromatin Immunoprecipitation (ChIP) Assay

The ChIP assay was performed using the ABclonal ChIP Kit (Cat# RK20258) according to the manufacturer's protocol with the following modifications:

2.15.1. Crosslinking and Chromatin Preparation

Cells were fixed with 1% formaldehyde (Sigma) for 10 min at 37°C, followed by quenching with 125 mM glycine.

Chromatin was sheared using a Covaris M220 sonicator (peak power: 75 W, duty factor: 5%, cycles of 200 × 0.5 s pulses) to achieve 200–500 bp fragments.

2.15.2. Immunoprecipitation

Following chromatin preparation, pre‐cleared chromatin (50 μg) was incubated with 5 μg of anti‐Target Protein antibody (Abcam, Cat# ab69933) or IgG isotype control (ABclonal, Cat# AC005) overnight at 4°C.

Protein A/G magnetic beads (20 μL/sample) were used for immune complex capture.

2.15.3. DNA Purification and Analysis

Crosslinks were reversed at 65°C for 6 h, followed by DNA purification using silica‐column purification.

Enriched DNA was quantified by qPCR targeting the specific genomic regions of interest.

2.16. Paraffin Sectioning of Mouse Brain Tissue

2.16.1. Tissue Harvesting and Fixation

Mice were transcardially perfused with ice‐cold PBS (pH 7.4) followed by 4% paraformaldehyde (PFA, Sigma, Cat# P6148). Brains were dissected and post‐fixed in 4% PFA for 24 h at 4°C; then they were transferred to 70% ethanol for long‐term storage.

2.16.2. Tissue Processing and Embedding

Dehydration was performed through an ethanol series (70%, 85%, 95%, 100%; 1 h each). Clearing was done using xylene (3 × 1 h) to remove ethanol. Samples were immersed in molten paraffin (Leica, Cat# 39601006) at 60°C for paraffin infiltration (3 × 1 h). Oriented embedding was carried out by coronally positioning brains in paraffin molds with the rostral–caudal axis aligned to target the striatum (Bregma: +0.5 to −0.2 mm).

2.16.3. Sectioning and Mounting

Sections of 4 μm thickness were cut using a rotary microtome (Leica RM2255). To validate the ribbon quality and verify striatal localization, every 10th section was stained with hematoxylin and eosin (H&E). Sections were mounted on Superfrost Plus slides (Thermo, Cat# J1800AMNZ) and dried overnight at 42°C.

2.17. Statistical Analysis

All datasets were analyzed using GraphPad Prism 9.0 (GraphPad Software). Continuous variables are expressed as mean ± SEM derived from three or more independent biological replicates (n ≥ 3). Intergroup differences for two‐group comparisons were assessed via unpaired Student's t‐test (parametric) or Mann–Whitney U test (non‐parametric). Multiple‐group comparisons were performed using one‐way ANOVA with Tukey's post hoc test or Kruskal–Wallis test with Dunn's correction. Prior to these analyses, normality was verified using the Shapiro–Wilk test (α = 0.05), and homogeneity of variance was confirmed via the Brown‐Forsythe test. Statistical significance was defined as p < 0.05 (two‐tailed).

3. Results

3.1. Spn Infection Induces Dose‐ and Time‐Dependent Upregulation of ZEB1‐AS1 in Human Microglia

As resident immune cells of the CNS, microglia critically mediate neuroinflammatory responses to bacterial invasion. To systematically identify lncRNAs modulating microglial antibacterial defense, we performed transcriptome profiling of HMC3 cells challenged with Spn (MOI = 100 for 12 h) versus saline‐treated controls. RNA‐seq analysis (a fold change > 2, p < 0.05) identified 217 differentially expressed lncRNAs (107 upregulated, 110 downregulated), with ZEB1‐AS1 ranking within the top 20 most significantly induced transcripts (Figure 1A,B).

FIGURE 1.

ZEB1‐AS1 expression increases with Spn infection multiplicity and duration in HMC3 cells. (A) Volcano plot of differentially expressed lncRNAs in Spn‐infected vs. saline‐treated groups (ZEB1‐AS1 highlighted). (B) Top 20 most significantly dysregulated lncRNAs (up/downregulated) between Spn‐infected and control groups (ZEB1‐AS1 marked in red). (C) qRT‐PCR analysis of ZEB1‐AS1 expression in HMC3 cells infected with Spn (MOI = 100) at 0 h, 3 h, 6 h, and 12 h (n = 3). (D) ZEB1‐AS1 expression in HMC3 cells infected with Spn at MOI = 0, 25, 50, and 100 for 12 h (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

Notably, ZEB1‐AS1 was previously reported to participate in host defense via immunomodulatory pathways against phylogenetically distinct intracellular pathogens such as Chlamydia trachomatis (Luo et al. 2022). However, its specific role in extracellular bacterial infections in the CNS remained unexplored. Our discovery of robust ZEB1‐AS1 induction in Spn‐challenged microglia prompted a focused investigation of its function in this clinically relevant model of bacterial meningitis.

In vitro validation confirmed a dual‐phase induction pattern. First, there was a time‐dependent upregulation peaking at 12 h post‐infection (MOI = 100; Figure 1C), and second, a dose‐dependent elevation proportional to bacterial load (12 h; Figure 1D), establishing ZEB1‐AS1 as a potential regulator of microglial antibacterial immunity.

3.2. ZNF148 Drives Transcriptional Upregulation of ZEB1‐AS1 During Spn Infection

In silico promoter analysis using JASPAR predicted two conserved binding motifs for ZNF148 within the ZEB1‐AS1 promoter region (Figure 2A). Consistent with these computational predictions, infection with Streptococcus pneumoniae (Spn) at a multiplicity of infection (MOI) of 100 for 12 h triggered significant upregulation of ZNF148 protein expression in HMC3 cells (Figure 2B). To functionally validate these findings, we demonstrated that ZNF148 knockdown substantially reduced ZEB1‐AS1 transcript levels (Figure 2C). Conversely, ZNF148 overexpression significantly increased ZEB1‐AS1 transcript levels (Figure 2D), establishing a causal relationship between ZNF148 expression and ZEB1‐AS1 transcription.

FIGURE 2.

ZNF148 transcriptionally upregulates ZEB1‐AS1 in microglia during Spn infection. (A) JASPAR‐predicted ZNF148 binding sites on the ZEB1‐AS1 promoter. (B) Western blot showing increased ZNF148 protein levels in HMC3 cells infected with Spn (MOI = 100) for 12 h (n = 3). (C, D) qRT‐PCR analysis of ZEB1‐AS1 expression upon ZNF148 knockdown or overexpression (n = 3). (E) ChIP assay demonstrating enriched ZNF148 binding at two predicted ZEB1‐AS1 promoter regions in Spn‐infected vs. control HMC3 cells (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

ChIP assays with ZNF148‐specific antibodies confirmed direct binding of ZNF148 to both predicted promoter sites. Notably, this binding was significantly enhanced following Spn exposure (Figure 2E). Collectively, these data demonstrated that Spn‐induced ZNF148 directly activates ZEB1‐AS1 transcription through promoter binding.

3.3. ZEB1‐AS1 Potentiates Microglial Antimicrobial Functions During Spn Infection

Functional characterization revealed that ZEB1‐AS1 critically regulates microglial responses to Spn infection. Knockdown of ZEB1‐AS1 significantly reduced pro‐inflammatory cytokine expression (IL‐1β, IL‐6, TNF‐α) specifically in Spn‐infected microglia compared to non‐targeting controls (Figure 3A), whereas no effect was observed in uninfected cells. This was accompanied by impaired phagocytic capacity, evidenced by decreased intracellular fluorescence of FITC‐labeled Spn (Figure 3D), and by concurrently increased bacterial load in culture supernatants (Figure 3C). Conversely, ZEB1‐AS1 overexpression enhanced cytokine production exclusively in infected cells (Figure 3B), while improving bacterial clearance through increased phagocytosis (Figure 3F) and reduced extracellular bacterial load (Figure 3E). These coordinated findings establish ZEB1‐AS1 as a key regulator of microglial antimicrobial defense.

FIGURE 3.

ZEB1‐AS1 enhances Spn‐induced inflammatory cytokine production and phagocytosis in HMC3 cells. (A, B) qRT‐PCR and ELISA analysis of IL‐1β, IL‐6, and TNF‐α mRNA/protein levels in ZEB1‐AS1‐knockdown or ‐overexpressing HMC3 cells post‐Spn infection (MOI = 100, 12 h; n = 3). (C) Bacterial load in supernatants of ZEB1‐AS1‐knockdown cells 12 h post‐infection (n = 3). (D) Mean fluorescence intensity (FITC‐labeled Spn) at 525 nm in ZEB1‐AS1‐knockdown cells 1 h post‐infection (n = 6). (E) Bacterial load in supernatants of ZEB1‐AS1‐overexpressing cells 12 h post‐infection (n = 3). (F) Mean fluorescence intensity (FITC‐labeled Spn) at 525 nm in ZEB1‐AS1‐overexpressing cells 1 h post‐infection (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

3.4. ZEB1‐AS1 Governs Microglial Immune Functions Through NOD2 Signaling

Building on our prior evidence of NOD2's essential role in anti‐pneumococcal immunity (Wang, Zhang, et al. 2022; Wang et al. 2020; Liu et al. 2010), Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis demonstrated significant enrichment of the NOD‐like receptor signaling pathway (Figure 4A). Inspired by reports that ZEB1‐AS1 stabilizes NOD2 mRNA in endothelial cells under low‐density lipoprotein stimulation (Xu et al. 2019), we hypothesized a similar regulatory mechanism in microglia during Spn infection. Mechanistic studies demonstrated that ZEB1‐AS1 knockdown substantially reduced NOD2 expression in infected cells at both the transcriptional and translational levels. Conversely, its overexpression enhanced NOD2 production in vitro (Figure 4B,C). Crucially, ZEB1‐AS1 reconstitution in NOD2‐knockdown cells partially rescued NOD2 expression and alleviated the suppression of cytokine production caused by NOD2 deficiency (Figure 4D,E). Functional rescue assays confirmed that ZEB1‐AS1 overexpression restored bacterial clearance impaired by NOD2 knockdown (Figure 4F,G), establishing NOD2 as the essential downstream mediator of ZEB1‐AS1's antimicrobial actions.

FIGURE 4.

ZEB1‐AS1 potentiates Spn‐induced inflammation and phagocytosis via NOD2. (A) KEGG enrichment analysis of pathways in Spn‐infected vs. control HMC3 cells (NOD‐like receptor signaling highlighted). (B) qRT‐PCR and Western blot showing reduced NOD2 mRNA/protein levels in ZEB1‐AS1‐silenced cells (n = 3). (C) qRT‐PCR and Western blot showing upregulated NOD2 expression in ZEB1‐AS1‐overexpressing cells (n = 3). (D) Restoration of NOD2 expression by ZEB1‐AS1 overexpression in NOD2‐silenced cells (n = 3). (E) qRT‐PCR and ELISA confirming rescued cytokine production (IL‐1β, IL‐6, TNF‐α) in ZEB1‐AS1‐overexpressing cells (n = 3). (F) Increased mean fluorescence intensity (FITC‐Spn) at 525 nm in ZEB1‐AS1‐overexpressing cells (n = 6). (G) Reduced bacterial load in supernatants of ZEB1‐AS1‐overexpressing cells (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

3.5. Functional Conservation of Zeb1‐os1 in Modulating Neuroinflammation and Cognitive Outcomes

Despite limited sequence conservation between species, the murine homolog Zeb1‐os1 recapitulated the functional properties of human ZEB1‐AS1 in vivo. Microglia‐specific knockdown of Zeb1‐os1 significantly increased cerebral Spn loads, while reducing NOD2 expression within these cells (Figure 5A,B,D). Conversely, Zeb1‐os1 overexpression enhanced bacterial clearance and amplified NOD2 signals specifically within microglial populations (Figure 5A,C,D). Critically, in Spn‐infected mice, Zeb1‐os1 overexpression significantly improved spontaneous alternation rates at 14 days post‐infection, whereas knockdown substantially reduced this cognitive metric (Figure 5E), demonstrating conserved neuroprotective functions across species.

FIGURE 5.

Microglial Zeb1‐as1 regulates post‐infection bacterial load, modulates NOD2, and induces long‐term behavioral effects. (A) Spn load in brain homogenates of microglia‐specific Zeb1‐as1‐knockdown/overexpression mice at 72 h post‐infection (n = 3). (B) Altered NOD2 mRNA/protein levels in brain tissues of Zeb1‐as1‐knockdown mice (n = 3). (C) NOD2 expression changes in brain tissues of Zeb1‐as1‐overexpressing mice (n = 3). (D) Multiplex IHC of brain sections showing NOD2 (red) colocalization with Iba‐1⁺ microglia (green), with quantitative MFI analysis (n = 3). (E) Y‐maze spontaneous alternation rates at 14 days post‐Spn infection (n = 10). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

3.6. Cytoplasmic ZEB1‐AS1 Stabilizes NOD2 mRNA to Potentiate Microglial Immunity

The functional mechanisms of lncRNAs are intrinsically linked to their subcellular localization. Fractionation assays in HMC3 cells revealed predominant cytoplasmic enrichment of ZEB1‐AS1 (Figure 6A), and fluorescence in situ hybridization (FISH) confirmed this spatial distribution along with a marked infection‐induced upregulation of ZEB1‐AS1 expression (Figure 6B). This cytoplasmic localization prompted investigation into post‐transcriptional regulatory mechanisms. Extending Xu et al.'s foundational work (Xu et al. 2019), we employed actinomycin D (ActD) chase assays to quantify NOD2 mRNA decay kinetics. Strikingly, ZEB1‐AS1 knockdown accelerated NOD2 mRNA degradation, while its overexpression significantly prolonged transcript stability (Figure 6C,D). These results establish mRNA stabilization as the core mechanism through which ZEB1‐AS1 amplifies NOD2‐dependent immune responses.

FIGURE 6.

ZEB1‐AS1 regulates NOD2 mRNA stability. (A) Subcellular fractionation and (B) RNA‐FISH confirming cytoplasmic localization of ZEB1‐AS1 in HMC3 cells, with enhanced signal post‐Spn infection (n = 3). (C, D) Actinomycin D chase assay showing accelerated NOD2 mRNA decay in ZEB1‐AS1‐silenced cells and delayed decay in ZEB1‐AS1‐overexpressing cells (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

3.7. ZEB1‐AS1‐IGF2BP2‐NOD2 Axis Mediates Microglial Immune Surveillance

To elucidate how ZEB1‐AS1 stabilizes NOD2 mRNA, we investigated the mediating RNA‐binding proteins (RBPs). Since ZEB1‐AS1 is localized in the cytoplasm (Figure 6A,B), we performed RNA pull‐down assays followed by mass spectrometry, which identified eight candidate RBPs (Table 1). Among these, IGF2BP2, an m6A‐binding protein that enhances mRNA stability, was prioritized for further study. The immunofluorescence assay confirmed the cytoplasmic distribution of IGF2BP2 in HMC3 cells (Figure 7C).

TABLE 1.

In HMC3 cells, an RNA pull‐down assay utilizing a biotin‐labeled ZEB1‐AS1 probe was conducted. The eluted proteins were subsequently analyzed by mass spectrometry, revealing the identification of RNA‐binding proteins.

Protein	Coverage (%)	Unique peptide
MEX3D	2	1
IGF2BP2	7	3
RO60	2	1
RBM14	11	5
RBMX	7	2
FXR1	3	1
RBM39	3	1
FXR2	3	1

Note: The IGF2BP2 is emphasized and bolded.

FIGURE 7.

IGF2BP2 interacts with ZEB1‐AS1 and stabilizes NOD2 mRNA. (A) RNA pull‐down/Western blot confirming IGF2BP2 binding to ZEB1‐AS1 and NOD2 mRNA (n = 3). (B) RIP‐qPCR showing specific enrichment of ZEB1‐AS1 and NOD2 mRNA in IGF2BP2 complexes (n = 3). (C) Immunofluorescence revealing cytoplasmic localization of IGF2BP2. (D, E) Western blot/qRT‐PCR showing reduced NOD2 expression in IGF2BP2‐silenced cells and elevated levels in IGF2BP2‐overexpressing cells (n = 3). (F, G) Actinomycin D chase assay indicating accelerated NOD2 mRNA decay in IGF2BP2‐silenced cells and prolonged stability in IGF2BP2‐overexpressing cells (n = 3). (*p < 0.05; **p < 0.01; ****p < 0.0001).

Biotinylated RNA pull‐down and RNA immunoprecipitation (RIP) assays demonstrated that IGF2BP2 directly binds both ZEB1‐AS1 and NOD2 mRNA (Figure 7A,B), confirming the formation of a tripartite complex. Functionally, IGF2BP2 knockdown mimicked the effect of ZEB1‐AS1 silencing by reducing NOD2 expression (Figure 7D), whereas its overexpression dose‐dependently increased NOD2 levels (Figure 7E). Actinomycin D chase assays showed that IGF2BP2 knockdown accelerated NOD2 mRNA degradation, while overexpression slowed degradation kinetics (Figure 7F,G).

Manipulation of IGF2BP2 bidirectionally regulated immune functions: knockdown suppressed cytokine secretion (IL‐1β, IL‐6, TNF‐α; Figure 8A) and impaired bacterial clearance (Figure 8C,D), whereas overexpression enhanced both responses (Figure 8B,E,F). Importantly, IGF2BP2 overexpression in ZEB1‐AS1‐silenced cells rescued NOD2 expression (Figure 9A), reversed the accelerated mRNA degradation (Figure 9B), and restored cytokine release (Figure 9C) as well as bacterial clearance capacity (Figure 9D,E).

FIGURE 8.

IGF2BP2 modulates Spn‐induced phagocytosis and cytokine secretion. (A, B) qRT‐PCR/ELISA detection of IL‐1β, IL‐6, and TNF‐α levels in IGF2BP2‐silenced or ‐overexpressing HMC3 cells (n = 3). (C) Increased bacterial load in IGF2BP2‐knockdown cells (n = 3). (D) Reduced mean fluorescence intensity (FITC‐Spn) in IGF2BP2‐knockdown cells (n = 6). (E) Decreased bacterial load in IGF2BP2‐overexpressing cells (n = 3). (F) Elevated mean fluorescence intensity in IGF2BP2‐overexpressing cells (n = 6). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

FIGURE 9.

IGF2BP2 overexpression rescues ZEB1‐AS1 knockdown phenotypes. (A) Restoration of NOD2 expression in ZEB1‐AS1‐silenced cells by IGF2BP2 overexpression. (B) Partial rescue of NOD2 mRNA stability by IGF2BP2 overexpression in ZEB1‐AS1‐silenced cells (n = 3). (C) Restored inflammatory cytokine production in rescue experiments. (D) Increased mean fluorescence intensity (FITC‐Spn) at 525 nm (n = 6). (E) Reduced bacterial load in supernatants post‐Spn infection (n = 3). (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

Together with our ZNF148 findings (Figure 2), these results delineate a complete regulatory cascade: ZNF148 transcriptionally activates ZEB1‐AS1, which scaffolds IGF2BP2 to stabilize NOD2 mRNA, thereby coordinating microglial antimicrobial responses against S. pneumoniae .

4. Discussion

Spn‐induced bacterial meningitis continues to pose significant clinical challenges due to high mortality and neurological sequelae despite available antibiotic therapies (Brouwer et al. 2010; Wall et al. 2021; Jaber and Beahm 2023). The limited efficacy of existing treatments underscores the critical need to delineate host‐pathogen interactions within the CNS and identify novel therapeutic targets (Sudo et al. 2024; Zhang et al. 2021; Li et al. 2023). Our study unveils the ZNF148/ZEB1‐AS1/IGF2BP2/NOD2 axis as a regulator of microglial innate immunity during Spn infection, bridging transcriptional activation, post‐transcriptional mRNA stabilization, and functional immune responses. This represents the first comprehensive demonstration of a lncRNA scaffolding complex that integrates extracellular bacterial sensing with effector functions in CNS immunity, advancing our understanding of neuroinflammatory pathogenesis while revealing RNA‐centric therapeutic opportunities.

The discovery of ZNF148‐mediated transcriptional activation of ZEB1‐AS1 establishes a direct mechanistic link between microbial challenge and lncRNA‐driven immunity. ZNF148—a zinc‐finger transcription factor previously associated with stress responses and oncogenesis—emerges as a sentinel for Spn detection in microglia, rapidly induced upon infection to bind the ZEB1‐AS1 promoter. Crucially, ZEB1‐AS1 operates across regulatory layers: transcriptionally controlled by ZNF148 while post‐transcriptionally stabilizing NOD2 mRNA via cytoplasmic scaffolding of the m6A‐reader IGF2BP2. This dual functionality distinguishes it from most lncRNAs, which typically function within single regulatory paradigms just like chromatin remodeling or miRNA sponging (Luo et al. 2022; Ma et al. 2019; Qian et al. 2019; Gu et al. 2024; Bure and Nemtsova 2023). The formation of a direct tripartite ZEB1‐AS1/IGF2BP2/NOD2 mRNA complex represents a novel mechanism for mRNA stabilization. The evolutionary conservation of IGF2BP2's RNA‐stabilizing role, now extended from cancer and viral infections (Yao et al. 2022; Huang et al. 2018; Hou et al. 2021; Yu et al. 2022; Wang, Chen, and Qiang 2021; Wang, Wang, et al. 2022; Mazeaud et al. 2024) to bacterial neuroinflammation, underscores its versatility in cellular stress adaptation.

Microglia, as CNS‐resident immune sentinels, must precisely balance pathogen clearance with neuroprotection (Jeong et al. 2022; Hegdekar et al. 2023; Filgueira et al. 2021; Chen et al. 2019; Waltl and Kalinke 2022). Our work demonstrates that the ZEB1‐AS1/IGF2BP2/NOD2 axis promotes microglial polarization toward an M1 phenotype, evidenced by enhanced NOD2‐dependent phagocytosis and amplified secretion of pro‐inflammatory cytokines including IL‐1β, IL‐6, and TNF‐α. NOD2—a cytosolic sensor of bacterial peptidoglycan—initiates canonical NF‐κB/MAPK signaling upon Spn detection, although mechanisms sustaining its expression during CNS infections remain poorly characterized (Wang et al. 2020; Ashton et al. 2022). ZEB1‐AS1‐mediated stabilization of NOD2 mRNA resolves this gap, ensuring robust immune activation. The translational relevance is further substantiated by in vivo conservation: microglial‐specific manipulation of murine Zeb1‐os1 bidirectionally modulated cerebral bacterial loads and NOD2 expression, directly impacting infection‐associated cognitive outcomes.

Therapeutically, augmenting ZEB1‐AS1 activity using strategies such as RNA mimics, CRISPRa technology, or small‐molecule agonists could bolster microglial defenses in immunocompromised hosts or antibiotic‐resistant Spn infections, thereby enhancing bacterial clearance capacity. Achieving this precision requires temporal control, as ZEB1‐AS1 concomitantly drives beneficial bacterial clearance outcomes and detrimental neurotoxic cytokine release outcomes. The established role of IGF2BP2 in oncology provides immediate translational leverage—its inhibitors are already in cancer clinical development (Cai et al. 2024; Dahlem et al. 2022; Feng et al. 2022; Chanda et al. 2024; Sa et al. 2022; Shen and Ding 2025) and could be repurposed for CNS infections. However, blood–brain barrier penetration and microglia‐specific targeting remain challenges (Gao and Hernandes 2021; Broux et al. 2015; Huang et al. 2023; Choi and Shusta 2023; Sommonte et al. 2022). Emerging solutions include lipid nanoparticles conjugated with microglia‐homing peptides (Gao et al. 2024; Shin et al. 2024; Han et al. 2023; Terashima et al. 2018; Zen et al. 2025) and CRISPR‐Cas13 systems engineered for RNA stabilization (Abudayyeh et al. 2017; Wu et al. 2022), which may enable spatially and temporally controlled modulation of this newly uncovered.

Several limitations warrant consideration in interpreting our findings. First, while we demonstrate ZEB1‐AS1/IGF2BP2‐driven microglial polarization toward the M1 phenotype, a detailed characterization of anti‐inflammatory responses and the dynamic shifts between M1 and M2 phenotypes remains incomplete. Second, clinical validation using human meningitis specimens, such as correlation analyses between ZNF148/ZEB1‐AS1 expression gradients and patient outcomes, would strengthen translational relevance. Third, our proposed m6A‐dependent stabilization mechanism underlying the IGF2BP2‐NOD2 mRNA interaction requires direct validation. This should be achieved through m6A‐RIP sequencing and functional rescue experiments using m6A‐binding‐deficient mutants. Fourth, although NOD2 elevation enhanced phagocytosis, the subcellular mechanisms coordinating cytoskeletal remodeling remain unexamined. Fifth, while this study focused on quantifying the phagocytic uptake of Spn using FITC‐labeled bacteria, this method does not evaluate post‐internalization killing events. Our attempts to directly measure intracellular bacterial burden using antibiotic protection assays were confounded by technical hurdles, precluding reliable assessment of microglial bactericidal capacity. Sixth, while ZEB1‐AS1's role is validated in Spn infection models, its functional conservation across other neurotropic pathogens remains undetermined and warrants systematic comparative investigation.

5. Conclusion

Collectively, this study establishes that Spn infection activates a ZNF148/ZEB1‐AS1/IGF2BP2/NOD2 signaling axis to orchestrate microglial–antimicrobial immunity. Mechanistically, Spn‐induced ZNF148 directly binds the ZEB1‐AS1 promoter, driving ZEB1‐AS1 transcription. Meanwhile, cytoplasmic ZEB1‐AS1 acts as a scaffold for the m6A reader IGF2BP2 to stabilize NOD2 mRNA, thereby amplifying NOD2‐dependent pro‐inflammatory cytokine production (IL‐1β, IL‐6, and TNF‐α) and bacterial clearance via phagocytosis. The functional conservation of the murine Zeb1‐os1 in regulating cerebral bacterial loads, neuroinflammation, and infection‐induced cognitive impairment, together with the rescue of immune functions upon pathway modulation, highlights this axis as both a fundamental mediator of CNS host defense and a potential therapeutic target for bacterial meningitis.

Author Contributions

Xiufu Hu and Fang Jiang: designed and performed cellular and animal experiments, including bacterial challenge assays, phagocytosis quantification, and cytokine profiling. Ling Li, Ruimei Hu, and Meng Dong: conducted bioinformatics analyses, statistical modeling, and data visualization. Aihua Cao: conceptualized the study, designed experimental workflows, and supervised methodology validation. Critically revised the manuscript, interpreted results, and finalized intellectual content.

Funding

This work was supported by the National Natural Science Foundation of China, 82171352, 82371366. and Key Research and Development Program of Shandong Province, 2025KIHZ03.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Original picture of Western Blot.

Figure S2: Efficiency of gene knockdown and overexpression.

Figure S3: Changes in HMC3 cell viability under different infection times and MOI (Trypan Blue staining method).

Hu, X. , Jiang F., Liu X., et al. 2026. “The ZNF148‐ZEB1‐AS1‐IGF2BP2‐NOD2 Axis Drives Microglial Antipneumococcal Immunity in Bacterial Meningitis.” Glia 74, no. 2: e70125. 10.1002/glia.70125.

Data Availability Statement

The data needed to evaluate the conclusions in the paper are present in the paper and the Supporting Information. The data can be provided by the corresponding author pending scientific review and a completed material transfer agreement.