IL-33-Driven Macrophage Reprogramming as a Potential Immunometabolic Strategy for Herpes Simplex Keratitis

Abstract

Background: Herpes simplex keratitis (HSK), caused by herpes simplex virus type 1 (HSV-1), is a major cause of infectious blindness. Macrophages are key antiviral effector cells, yet the metabolic mechanisms driving their protective responses remain poorly defined. This study aimed to determine whether interleukin-33 (IL-33) modulates macrophage metabolism and function to enhance antiviral protection in HSK. Methods: Bone marrow-derived macrophages (BMDMs) were stimulated with IL-33, followed by phenotypic and functional characterization using qRT-PCR, flow cytometry, and immunofluorescence. Integrated transcriptomic and non-targeted LC-MS metabolomic profiling was performed to uncover regulatory pathways. For in vivo validation, differently treated BMDMs were adoptively transferred subconjunctivally into a mouse HSK model. Clinical scoring, fluorescein staining, TCID₅₀ quantification of tear samples, and corneal viral gene detection were used to evaluate disease severity and viral burden. Results: IL-33 stimulation increased CD169 and MHC-II expression, expanded the CD169⁺ macrophage subset, and suppressed HSV-1 replication in vitro. Multi-omics integration identified 616 differentially expressed genes and 417 differentially expressed metabolites, revealing substantial remodeling of lipid and amino acid metabolism and suggesting a critical IL-33–lipoprotein lipase (LPL)–palmitoylcarnitine (L-PC) metabolic axis. In vivo, prophylactic adoptive transfer of IL-33-treated BMDMs significantly reduced corneal opacity, epithelial injury, tear viral titers, and virogene expression. LPL inhibition eliminated these benefits, whereas L-PC supplementation partially restored antiviral and clinical improvements. Conclusions: IL-33 reprograms macrophages toward a CD169⁺ antiviral phenotype through an LPL-dependent metabolic pathway, establishing an LPL–L-PC axis essential for enhanced antiviral function and protection against HSK. These findings highlight metabolic tuning of macrophages as a potential preventive immunomodulatory approach for HSV-1-induced ocular disease.

1. Introduction

Herpes simplex keratitis (HSK) is a severe infectious corneal disease caused by herpes simplex virus type 1 (HSV-1) and remains the leading cause of corneal blindness worldwide [1]. Current therapeutic strategies face several challenges, including increasing antiviral drug resistance, corticosteroid-related adverse effects, and high recurrence rates following corneal transplantation [2,3]. These limitations underscore the urgent need for novel therapeutic approaches to effectively manage and control HSK progression.

Interleukin-33 (IL-33), a member of the IL-1 cytokine family, functions as an immune-modulatory alarmin [4,5]. Through its receptor, IL-33 orchestrates a wide range of innate and adaptive immune responses across multiple immune cell types [6,7]. It has shown immunoprotective effects in various viral infections. For instance, IL-33 overexpression in a rabies virus mouse model enhanced the recruitment of CD4⁺ and CD8⁺ T cells [8]. During early lymphocytic choriomeningitis virus infection, IL-33 promoted the expansion of stem-like CD8⁺ T cells by modulating chromatin accessibility and balancing interferon (IFN) signaling [9]. In coxsackievirus B3-induced myocarditis, IL-33 activated F4/80⁺ macrophages [10]. However, whether IL-33 modulates macrophage antiviral responses in the context of HSV-1 infection remains unclear.

Macrophages play a crucial role in antiviral immunity and corneal viral clearance during HSK [11], but relying solely on the traditional M1/M2 phenotype is insufficient for effective HSK control [11,12]. Therefore, this study focuses on a distinct macrophage subset—CD169⁺ macrophages. CD169, also known as sialoadhesin (siglec-1), mediates nonspecific antiviral defense by recognizing the sialic acid structures on HSV-1, facilitating viral phagocytosis and cytokine secretion, and promoting robust antigen presentation to elicit adaptive antiviral immunity [13,14]. Our previous work developed a ganglioside GM1 liposomal vaccine targeting CD169⁺ macrophages [15]; however, this vaccine functioned only as a prophylactic approach and was limited by its short half-life, high cost, and complex preparation process.

Recent research has highlighted the importance of phenotypes and functions [16]. Beyond serving as sources of energy and biosynthetic precursors, metabolic intermediates also regulate gene expression and signaling pathways, thereby influencing immune responses [17,18,19]. While the role of metabolism in macrophage function is well established, its specific contribution to antiviral responses during HSV-1 infection remains poorly explored.

In this study, we observed that IL-33 promoted macrophage differentiation toward the CD169 subtype and impaired their ability to reduce viral burden. Then, we employed an integrated multi-omics approach—combining transcriptomic and metabolomic analyses—to systematically characterize the molecular alterations induced by IL-33 in bone marrow-derived macrophages (BMDMs) during HSV-1 infection. Compared to single-omics approaches, this integrative strategy provided a more comprehensive view of cellular state transitions, revealing gene–metabolite–function relationships and key immunometabolic pathways. Furthermore, we validated our observations in an in vivo adoptive transfer model using BMDMs, providing supportive evidence for the involvement of metabolic mechanisms in HSK pathogenesis and indicating possible directions for future immunometabolic research.

2. Results

2.1. IL-33 Regulated Macrophage Differentiation and Function During HSV-1 Infection

After successful differentiation of BMDMs, cells were infected with HSV-1 and treated with IL-33. IL-33 induced a time- and dose-dependent increase in CD169 mRNA expression (Figure 1A) and increased the proportion of CD169⁺ cells as assessed by flow cytometry (FC) (Figure 1B). Meanwhile, MHC-II expression was significantly upregulated (Figure 1C), suggesting enhanced antigen-presenting capacity and maturation. Western blot (WB) further confirmed increased CD169 protein levels upon IL-33 treatment (Figure 1D), which was attenuated by ST2 blockade, indicating an ST2-dependent effect. These findings were supported by immunofluorescence staining (Figure 1E) and the upregulation of CD86 (Figure 1F). The gating strategy is presented in Supplementary Figure S1A.

Next, we investigated whether IL-33 affects HSV-1 burden in macrophages. IL-33 treatment significantly reduced infectious viral titers in the culture supernatant as measured by TCID₅₀ (Figure 2A). In parallel, intracellular viral burden was decreased, as indicated by reduced GFP signals (Figure 2B), a lower percentage and MFI of GFP⁺ cells by flow cytometry (Figure 2C–E), and decreased viral protein and transcript levels (Figure 2F,G, Supplementary Figure S2A). Importantly, bafilomycin A1 (BafA1) partially attenuated the IL-33-mediated reduction in intracellular HSV-1 (Figure 2C–F), suggesting that this effect is at least partly dependent on lysosomal pathways. Collectively, these results indicate that IL-33 reduces HSV-1 burden in macrophages and may facilitate lysosome-associated intracellular viral degradation.

2.2. Adoptive Transfer of IL-33-Treated Macrophages Alleviated HSK

To evaluate the prophylactic protective effect of IL-33-treated macrophages in vivo, ocular adoptive transfers were performed (Figure 3A). Mice received subconjunctival injections one day prior to corneal HSV-1 inoculation. Adoptive transfer of IL-33-treated BMDMs alleviated HSK symptoms, showing reduced corneal opacity and epithelial damage. TCID₅₀ assays of tear samples revealed lower viral titers in the IL-33 group (Figure 3B,C). In addition, corneal viral burden was quantified at 3, 6, 9, and 12 d.p.i., demonstrating a time-dependent reduction in HSV-1 levels in the IL-33 group (Figure 3D, Supplementary Figure S2B). Histological analysis further showed preserved epithelial integrity, accompanied by increased leukocyte infiltration (Figure 3E). Immunofluorescence staining identified enhanced infiltration of F4/80⁺CD169⁺ cells in the corneas of IL-33-treated mice (Figure 3F).

2.3. Transcriptomic Analysis of IL-33-Treated MACROPHAGES

To investigate the effect of IL-33 on the transcriptomic profiles of macrophages, we performed RNA sequencing analysis. RNA integrity was confirmed by an Agilent 5300 Bioanalyzer, showing high-quality RNA profiles, and RNA concentrations were measured by a NanoDrop spectrophotometer (Supplementary Figure S4 and Table S1). The data quality was robust, with a stable base error rate ranging between 0.01% and 0.015% across all samples (Supplementary Figure S4). Base distribution was uniform, showing no 5′ or 3′ bias, and data dispersion among groups was low (Supplementary Figure S4). The correlation of samples within groups was high, with a Pearson correlation coefficient r > 0.99 (Figure 4A). Principal component analysis (PCA) showed clear separation between groups (Figure 4B), confirming that the data quality met the requirements for subsequent analysis.

A total of 11,338 genes were identified in the PBS group and 11,414 genes in the IL-33 group, with 10,979 genes co-expressed in both groups (Figure 4C). Using the criteria of |log2FC| ≥ 1 and P adj < 0.05, 616 genes were identified as significantly differentially expressed following IL-33 treatment. Among these, 300 genes were upregulated and 316 were downregulated (Figure 4D,E). Hierarchical clustering of the top 50 DEGs confirmed a distinct transcriptional profile between the groups (Figure 4F).

2.4. DEGs Functional Enrichment Analysis

To further investigate the molecular mechanisms underlying IL-33-mediated regulation of macrophage immune response, GO enrichment analysis was conducted on the 616 DEGs (Figure 5A–C). In the biological process (BP) category, the DEGs were significantly enriched in processes such as positive regulation of defense response, the acylglycerol metabolic process, positive regulation of cell differentiation, the fatty acid metabolic process, and the cytokine-mediated signaling pathway. In the cellular components (CCs) category, the DEGs were enriched in the cell surface, extracellular matrix, extracellular space, membrane-bounded organelle, and MHC protein complex. In terms of molecular function (MF), enrichment was observed in immune receptor activity, enzyme binding, GTPase regulator activity, nucleoside-triphosphatase regulator activity, and MHC class I receptor activity. The chordal correlation of GO enrichment results with key DEGs indicates that genes such as lipase (LPL), G protein-coupled receptor C family member 5B (GPRC5B), Toll-like receptor 2 (TLR2), chemokine receptor 7 (CCR7), calreticulin (CALR), and plasminogen activator inhibitor (PAI-1) may play important roles in IL-33 regulation (Figure 5D).

KEGG pathway enrichment revealed significant involvement in the PPAR signaling pathway, cytokine–cytokine receptor interaction, TGF-β signaling pathway, phagosome, antigen processing and presentation, and fatty acid biosynthesis, among other processes (Figure 5E). Key genes in these pathways include LPL, long-chain lipoyl coenzyme A synthetase (ACSL), CALR, and various MHC molecules including H2-M2, H2-Oa, H2-Q6, H2-Q7, H2-Eb2, and H2-Ob (Figure 5F). These findings collectively highlight LPL as a critical target for IL-33 in modulating macrophage function.

Additionally, gene set enrichment analysis (GSEA) of the entire transcriptome revealed that IL-33 treatment significantly upregulated several metabolic and immune-related pathways, including PI3K-AKT-mTOR signaling, cholesterol homeostasis, smooth endoplasmic reticulum, cholesterol biosynthesis, response to pheromones, mTORC1 signaling, and cytoplasmic ribosomal proteins. Conversely, IL-33 treatment downregulated TNFα signaling through NFκB signaling and TGFβ signaling pathways, suggesting a role for metabolic reprogramming in IL-33′s regulation of macrophage function (Figure 6A). Protein–protein interaction (PPI) network analysis identified fatty acid synthase (FASN), CALR, heat shock protein 90b1 (HSP90b1), stearoyl-CoA desaturase 1 (SCD1), and LPL as hub genes central to the IL-33 regulatory network (Figure 6B).

2.5. Metabolomic Analysis of IL-33-Treated Macrophages

To assess the impact of IL-33 on the metabolic status of infected BMDMs, we performed non-targeted LC-MS metabolomic analysis. PCA under both positive and negative ion modes revealed a clear separation between the PBS and IL-33 groups, indicating that IL-33 significantly altered the metabolic profiles of BMDMs (Figure 7A). OPLS-DA showed group separation with favorable R²Y/Q² (Figure 7B). Moreover, QC samples showed tight clustering in PCA, and OPLS-DA permutation testing indicated no obvious overfitting (Supplementary Figure S5).

A total of 1082 metabolites were identified in the PBS group, while 1134 metabolites were identified in the IL-33 group, with 1075 shared metabolites (Figure 7C). Based on the criteria of VIP > 1 and p value < 0.05, we identified 417 DEMs in response to IL-33 treatment, of which 401 were upregulated and 16 were downregulated (Figure 7D,E).

2.6. DEMs Annotation and Enrichment Analysis

Hierarchical clustering of the top 30 DEMs (ranked by VIP scores) revealed a marked increase in overall metabolite abundance in the IL-33 group (Figure 8A). HMDB classification showed that these DEMs mainly belonged to three major classes: carboxylic acids and derivatives (28.41%), organooxygen compounds (8.12%), and fatty acyl (6.67%) (Figure 8B).

KEGG pathway enrichment and differential scoring analyses showed that IL-33 treatment significantly upregulated several metabolic pathways, including aminoacyl tRNA biosynthesis, D-amino acid metabolism, nucleotide metabolism, β-alanine metabolism, tyrosine metabolism, pyrimidine metabolism, biosynthesis of phenylalanine, tyrosine, tryptophan, glutathione metabolism, glycerophospholipid metabolism, purine metabolism, alanine, aspartic acid, glutamate metabolism, etc. (Figure 8C,D). These findings highlight the extensive metabolic reprogramming induced by IL-33 in infected BMDMs.

2.7. Joint Analysis of the Transcriptome and Metabolome

To further reveal the synergistic regulatory mechanism of IL-33 at both the transcriptional and metabolic levels in macrophages, we conducted correlation analysis between DEGs and DEMs. Heatmaps were generated using the top 20 DEGs and DEMs, and Pearson correlation analysis showed several gene–metabolite pairs with strong positive or negative correlations (Figure 9A). Notably, genes such as LPL, CALR, Fc gamma receptor III (FCGR3), and interferon-induced transmembrane protein 2 (IFITM2) were significantly positively correlated with metabolites like L-PC, arachidonic acid, and various phospholipids (e.g., PE and PC). In contrast, genes such as C-X-C motif chemokine 16 (CXCL16), T-cell-specific GTPase 1 (TGTP1), and transglutaminase (TGM) showed negative correlations with metabolites like L-isoleucine, L-phenylalanine, and tyramine.

The gene–metabolite interaction network highlighted a central role for lipid metabolites, suggesting their importance as critical mediators of IL-33-induced macrophage regulation (Figure 9B). The nine-quadrant plot further categorized the gene–metabolite relationships, identifying positive correlations for subsequent validation (Figure 9C). The O2PLS model analysis demonstrated a clear separation between groups in the latent variable space, with a goodness-of-fit score (R²X) of 0.8207, indicating strong concordance between transcriptomic and metabolomic datasets (Figure 9D). KEGG co-annotation analysis showed that both DEGs and DEMs were enriched in pathways associated with lipid metabolism, transmembrane transport, nucleotide metabolism, and amino acid metabolism (Figure 9E).

To directly assess lipid droplet levels, we stained cells with BODIPY 493/503 methyl bromide and analyzed them by fluorescence microscopy (Figure 9F). Intracellular lipid droplet content was low in the PBS group, but it increased after IL-33 treatment. Taken together, IL-33 stimulation markedly enhances lipid droplet formation in macrophages, consistent with activation of lipid metabolic processes. Moreover, IL-33 treatment significantly upregulated the expression of key molecules involved in lipid utilization and fatty acid pathways, including LPL, SLC25A20, and ACADVL (Figure 9G), further supporting that IL-33 promotes a metabolic program associated with lipid metabolism and fatty acid oxidation in macrophages.

2.8. IL-33 Activated the LPL-L-PC Axis to Alleviate HSK Lesions

Based on the above results, we hypothesized that IL-33 alleviated HSK by activating the LPL-L-PC axis. To test this hypothesis, BMDMs receiving different treatments were adoptively transferred via subconjunctival injection one day prior to HSV-1 infection, including the LPLi group (IL-33 + LPLi) and the L-PC group (IL-33 + LPLi + L-PC) (Figure 10A), and their effects were compared with the IL-33 group described above.

At baseline (0 d.p.i.), none of the treatments caused ocular surface or epithelial damage (Supplementary Figure S6). Following HSV-1 challenge, mice in the LPLi group developed pronounced corneal opacity and edema, whereas those in the IL-33 and L-PC groups exhibited markedly milder symptoms, showing only minor punctate lesions. Corneal fluorescein staining confirmed reduced epithelial injury in the IL-33 and L-PC groups (Figure 10B,C).

Consistent with these clinical findings, TCID₅₀ assays of tear samples revealed significantly lower viral titers in the IL-33 and L-PC groups. Similarly, qRT-PCR analysis of corneal tissues showed markedly decreased expression of HSV-gB and HSV-gD (Figure 10D). Together, these data demonstrated that IL-33-treated macrophages mitigate HSV-1 infection and corneal pathology by activating the LPL–L-PC axis, underscoring its potential therapeutic relevance in HSK.

3. Discussion

This study provided evidence that IL-33 orchestrated macrophage differentiation and metabolic reprogramming to enhance antiviral defense during HSV-1 infection. Through integrated transcriptomic and metabolomic analyses, we found that IL-33 induces a distinct CD169⁺ macrophage phenotype characterized by enhanced antiviral and antigen-presenting functions. Mechanistically, IL-33 activates the LPL-L-PC axis, promoting metabolic remodeling toward enhanced lipid utilization and immune activation. These effects ultimately translate into reduced viral burden and alleviated corneal pathology when macrophages are transferred prior to infection.

Notably, CD169 (Siglec-1) has been widely recognized as a virus-responsive interferon-stimulated gene (ISG) in myeloid cells, and its induction often reflects activation of a broader antiviral ISG program. Therefore, the elevated CD169 expression observed following IL-33 stimulation may, at least in part, indicate an IFN-associated antiviral state in macrophages, although the upstream regulatory mechanisms require further clarification. Future studies examining canonical ISG signatures and interferon pathway activation will help define whether CD169 induction in this model is directly driven by IL-33 signaling or occurs secondary to IFN-related antiviral programs.

Our findings revealed that IL-33 exerts dual regulatory effects on macrophages—reshaping transcriptional programs while rewiring key metabolic pathways. Transcriptomic profiling identified 616 DEGs enriched in lipid metabolism, cytokine signaling, and antigen presentation pathways, highlighting LPL, CALR, and TLR2 as core regulatory hubs. Previous studies have shown that LPL facilitates lipid uptake and supports phagocytic and inflammatory functions of macrophages, while CALR acts as an endoplasmic reticulum chaperone essential for antigen presentation and immune homeostasis [20,21,22]. The upregulation of these genes following IL-33 stimulation suggested that IL-33 transformed macrophages into a metabolically active, immunocompetent state optimized for antiviral defense.

Metabolomic profiling further demonstrated extensive remodeling of lipid and amino acid metabolism. IL-33 markedly increased levels of long-chain fatty acyl metabolites, particularly L-PC, alongside nucleotide and amino acid intermediates. Integration of multi-omics data pinpointed the LPL–L-PC axis as a central regulatory pathway linking metabolic activation with immune enhancement. Given that LPL catalyzes the hydrolysis of triglycerides to release free fatty acids, which can be converted to L-PC for mitochondrial β-oxidation [23], these findings suggested that IL-33 promoted oxidative metabolism to sustain macrophage effector functions. Enhanced mitochondrial activity may also contribute to antiviral defense by generating reactive oxygen species (ROS) and supporting the production of type I interferons, both of which are critical for viral clearance [24,25].

In addition, although our inhibitor experiments and the partial rescue effect of L-PC provide supportive evidence for the involvement of the IL-33–LPL–L-PC axis, the current study does not fully establish strict causal dependency of this metabolic pathway. Specifically, genetic loss-of-function approaches targeting LPL, direct measurements of LPL enzymatic activity, and targeted quantification of L-PC or fatty acid oxidation (FAO) flux would further strengthen the mechanistic conclusions. Future work will therefore consider genetic strategies (e.g., Lpl knockdown/knockout or macrophage-specific deletion), combined with enzymatic activity assays and targeted metabolomics/FAO profiling, to more precisely define how IL-33-driven lipid utilization programs contribute to macrophage antiviral function.

The functional relevance of this IL-33–LPL–L-PC axis was validated through prophylactic adoptive transfer experiments. IL-33-treated macrophages significantly mitigated corneal lesions and reduced viral titers following HSV-1 infection. In contrast, inhibition of LPL abrogated these protective effects, while supplementation with L-PC partially restored antiviral activity. These results underscore that IL-33-driven lipid metabolic remodeling is not merely correlative but functionally essential for macrophage-mediated viral control. Although immune cell infiltration can contribute to immunopathology in HSK, our results suggest that the increased infiltration of CD169⁺ macrophages in this model is more likely associated with protective antiviral control.

Collectively, our study expands current understanding of IL-33 as an immunometabolic regulator beyond its classical role as an alarmin. The discovery of the IL-33–LPL–L-PC axis provides mechanistic insight into how cytokine signaling integrates with metabolic pathways to dictate macrophage function. Given the limitations of current antiviral and corticosteroid-based HSK therapies, targeting IL-33 signaling or modulating macrophage metabolism may represent a promising immunomodulatory strategy.

4. Materials and Methods

4.1. BMDMs Culture, Stimulation and Infection

BMDMs were obtained as previously described [26]. Briefly, bone marrow cells were flushed from the tibias and femurs of 6-week-old male C57BL/6 mice. The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, 11965092, Gibco, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Gibco, USA) and 1% penicillin-streptomycin (Gibco, USA). Differentiation was induced by adding 20 ng/mL mouse macrophage colony-stimulating factor (M-CSF; AbClonal, Wuhan, China) for 6 days.

HSV-1 McKrae was prepared according to previously outlined procedures [27]. Following confirmation of BMDM identity, cells were stimulated with either 10 ng/mL recombinant IL-33 (MedChemExpress, Shanghai, China) or PBS. The concentration of IL-33 was selected based on previous studies reporting effective stimulation of macrophages at this level [28]. Cells were subsequently infected with HSV-1 McKrae at a multiplicity of infection (MOI) of 1.

4.2. Quantitative Real-Time PCR Analysis (qRT-PCR)

Total RNA was extracted using TRIzol Reagent according to manufacturer’s protocol. One microgram of RNA was reverse transcribed into cDNA. qRT-PCR was performed on an ABl QuantStudio Six Flex (Invitrogen, Carlsbad, CA, USA) using the ChamQ Universal SYBR qPCR Kit (Vazyme, Nanjing, China). β-actin was used as the internal control. Relative gene expression levels were calculated using the 2^−ΔΔCT method. Primer sequences are listed in Supplementary Table S3.

4.3. Flow Cytometry (FC) Analysis

Cell viability was assessed by incubating the cells with Fixable Viability Stain 780 (BD, San Jose, CA, USA) for 10 min at room temperature prior to subsequent antibody staining. BMDMs were stained with antibodies against F4/80 (E-AB-F0995C, Elabscience, Wuhan, China), CD169 (12-5755-80, eBioscience, San Diego, CA, USA), MHC-II (107625, Biolegend, San Diego, CA, USA) and CD86 (A25824, Abclonal, Wuhan, China). FC analysis was carried out using a BD Accuri C6 Plus (BD Biosciences, San Jose, CA, USA) and Aria II (BD Biosciences, San Jose, CA, USA), and the acquired data were processed with FlowJo software (version 10.8).

4.4. Western Blot Analysis

Cells were lysed in ice-cold RIPA buffer (Solarbio, Beijing, China), and total proteins were collected after centrifugation (12,000 rpm, 15 min, 4 °C). Proteins were separated by SDS–PAGE and transferred to PVDF membranes (Millipore, Burlington, MA, USA). Membranes were blocked with rapid blocking buffer (Yoche, Shanghai, China) and incubated overnight at 4 °C with primary antibodies against CD169, HSV-1 gD and ACTB, followed by HRP-conjugated secondary antibodies. Signals were detected using an ECL kit (Abbkine, Wuhan, China), and band intensities were quantified with ImageJ (v1.53).

4.5. Immunofluorescence (IF) Staining

BMDMs were fixed with 4% paraformaldehyde for 30 min, followed by permeabilization with 0.5% Triton X-100 (P0096, Beyotime, Shanghai, China) for 10 min. Afterward, nonspecific binding was blocked using 5% bovine serum albumin (BSA; ST023, Beyotime, China) for 1 h. The cells were incubated with antibodies against F4/80 (E-AB-F0995C, Elabscience, Wuhan, China) and CD169 (12-5755-80, eBioscience, San Diego, CA, USA) overnight at 4 °C. Nuclei were stained with a DAPI mounting medium (ab104139, Abcam, Cambridge, UK).

IF of corneal tissues were conducted following a previously reported protocol. Freshly isolated mouse corneas were fixed in 4% paraformaldehyde and subsequently treated with 20 mM EDTA for 30 min. Tissue permeabilization was achieved using 0.5% Triton X-100 for 30 min, followed by blocking with 5% donkey serum for 1 h to reduce nonspecific binding. The corneas were then incubated with antibodies against F4/80 and CD169 at 4 °C overnight. After thorough washing, samples were mounted using a DAPI-containing mounting medium for nuclear counterstaining.

4.6. HSV-1 GFP Infection Readout Assay

BMDMs were incubated with HSV-1 GFP at a multiplicity of infection (MOI) of 1 for 2 h to allow viral adsorption, followed by washing twice with PBS to remove unbound virus. Fresh culture medium was then added, and the cells were further incubated for 12 h. Macrophage-mediated viral burden reduction was assessed under a fluorescence microscope by quantifying the number of GFP-positive cells.

4.7. Animals, HSK Mouse Model and Adoptive Transfer of BMDMs

All animal experiments were approved by the Ethics Committee of Nanjing Drum Tower Hospital (Approval No. 20231103) and conducted in accordance with institutional and provincial guidelines. Six-week-old male C57BL/6 mice with normal eye development were purchased from Yangzhou university. The mice were housed under specific pathogen-free (SPF) conditions (25 °C, 40–50% humidity, 12 h light/dark cycle) at the Animal Experiment Centre of Nanjing Drum Tower Hospital.

BMDMs were differently treated with IL-33, IL-33 + lipoprotein lipase inhibitor (LPLi; GSK264220A, MCE, Shanghai, China), or IL-33 + LPLi + L-palmitoylcarnitine (L-PC; TMIM-00043, TargetMol, Shanghai, China), then adoptively transferred to the ocular surface via subconjunctival injection (10⁴ cells/eye). Mice were divided into four groups based on BMDM treatment: Ctrl (vehicle), IL-33, LPLi (IL-33 + LPLi), and L-PC (IL-33 + LPLi + L-PC). Mice were randomly assigned to experimental groups prior to adoptive transfer. One day post-transfer, mice were anesthetized with intraperitoneal injection of 1% sodium pentobarbital (80 mg/kg). HSK was induced by gently cross-hatching the right cornea with a 30-gauge needle, followed by unilateral topical application of 5 μL HSV-1 McKrae (1 × 10⁶ PFU/mL).

4.8. Assessment of HSK Clinical Symptoms

Clinical scoring and fluorescein staining evaluation were performed by two independent investigators blinded to group allocation. Corneal lesions were scored on a 0–4 scale based on epithelial damage and stromal opacity, as previously described [29]. Representative images were obtained via stereomicroscopy. Corneal fluorescein staining (CFS) was performed following standard protocols. Four corneal quadrants were scored (0–3), and total CFS scores were calculated.

4.9. Tears Swab Collection and Median Tissue Culture Infective Dose (TCID₅₀) Assay

Tear samples were collected by instilling 10 µL sterile PBS onto the ocular surface and retrieving it with a sterile swab [29]. Serial dilutions of tear samples were added to Vero cells in 96-well plates. After 72 h of incubation, cytopathic effects were recorded, and viral titers were calculated using the Reed–Muench method.

4.10. Hematoxylin–Eosin (H&E) Staining

Eyeballs were fixed in 4% paraformaldehyde for 24 h, followed by graded ethanol dehydration and paraffin embedding. Serial paraffin sections (5 μm) were prepared, deparaffinized in xylene, and rehydrated through decreasing concentrations of ethanol. Sections were subsequently stained with hematoxylin and eosin using a commercial kit (Servicebio, Wuhan, China) in accordance with the manufacturer’s protocol.

4.11. Transcriptome Sequencing

Total RNA was extracted from BMDMs using TRIzol Reagent (R711, Vazyme, Nanjing, China) according to the manufacturer’s protocol. Each group included four biological replicates. Transcriptome sequencing was performed by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). RNA quality was assessed using the Agilent 5300 Bioanalyzer (Santa Clara, CA, USA), and RNA concentrations were measured using the ND-2000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Sequencing libraries were prepared and sequenced on the NovaSeq X Plus platform (PE150) using the NovaSeq Reagent Kit (Illumina, San Diego, CA, USA). Raw reads were filtered to generate high-quality clean reads, which were aligned to the reference genome using HISAT2 software (v2.2.1). The alignment quality was evaluated (Supplementary Table S2). The raw transcriptome data of this study has been uploaded to the NCBI Sequence Read Archive (SRA) database, with the project number PRJNA1332403.

4.12. Transcriptome Data Analysis

Transcript expression levels were calculated as transcripts per million (TPMs) for visualization purposes. Differential expression analysis was performed on the raw count matrix using DESeq2, and genes with |log2(FC)| ≥ 1 and Benjamini–Hochberg adjusted p value (Padj) < 0.05 were defined as differentially expressed genes (DEGs). Functional enrichment analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), were conducted, and terms with false discovery rate (FDR) < 0.05 were included. Gene set enrichment analysis (GSEA) was performed using MSigDB gene sets with the Signal2Noise ranking metric; pathways with FDR q value < 0.25 were considered significantly enriched.

4.13. Metabolite Extraction

Each group contained four or five biological replicates. Each sample was thoroughly mixed with 300 μL of extraction solution (methanol:water = 4:1, v:v) containing 0.02 mg/mL internal standard (L-2-chlorophenylalanine). Samples were homogenized using the Wonbio-96c frozen tissue grinder (Shanghai Wanbo Biotechnology Co., Ltd., Shanghai, China) for 6 min (−10 °C, 50 Hz), followed by low-temperature ultrasonic extraction for 30 min (5 °C, 40 kHz). After standing at −20 °C for 30 min, samples were centrifuged for 15 min (4 °C, 13,000× g). The supernatant was transferred to the injection vial for liquid chromatography–mass spectrometry (LC-MS) analysis. Quality control (QC) samples were prepared by pooling equal volumes of supernatant from all samples.

4.14. LC-MS Analysis

LC-MS analysis was conducted using a SCIEX UPLC-Triple TOF 5600 system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Milford, CT, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The mobile phases consisted of 0.1% formic acid in water: acetonitrile (95:5, v/v) (solvent A) and 0.1% formic acid in acetonitrile: isopropanol: water (47.5:47.5:5, v/v) (solvent B). The flow rate was set at 0.40 mL/min, and the column temperature was maintained at 40 °C. Mass spectrometry was performed on a Triple TOFTM5600+ mass spectrometer (Sciex, Framingham, MA, USA) using electrospray ionization in both positive and negative ion modes. All untargeted metabolomic data used in this publication have been deposited to the EMBL-EBI MetaboLights database with the identifier MTBLS13068.

4.15. Metabolite Data Analysis

Metabolites were identified by comparison with the Human Metabolome Database (HMDB), Metlin, and Majorbio in-house databases. Metabolites with a variable importance in projection (VIP) > 1 and p value < 0.05 were identified as differentially expressed metabolites (DEMs), using orthogonal partial least squares discriminant analysis (OPLS-DA) modeling and Student’s t-test. DEMs between groups were subsequently subjected to metabolic pathway enrichment and analysis using the KEGG database.

4.16. Integrated Transcriptome and Metabolome Analysis

Correlation analysis between DEGs and DEMs was performed using Pearson’s correlation coefficients. Correlation heatmaps were constructed using genes and metabolites with |correlation coefficient| > 0.8 and p value < 0.05. DEGs and DEMs were simultaneously mapped to KEGG pathways to identify shared functional networks, and correlation networks were visualized in Cytoscape (v3.9.1). Cross-omics integration was further performed using two-way orthogonal partial least squares (O2PLS).

4.17. Lipid Droplets Staining

Cells were rinsed with PBS and incubated with BODIPY 493/503 (1μM, HY-D1614, MCE, China) at 37 °C for 30 min to label intracellular lipid droplets. Nuclear staining was performed using DAPI. Fluorescence images were captured using a Leica THUNDER imaging system (Leica, Wetzlar, Germany).

4.18. Statistical Analysis

Statistical analyses were performed using GraphPad Prism version 10.0 (GraphPad Software, San Diego, CA, USA). Data are presented as mean ± SD. All statistical tests were two-tailed. Normality was assessed using the Shapiro–Wilk test. For comparisons between two groups, an unpaired two-tailed Student’s t-test was used. For comparisons involving more than two groups, one-way or two-way ANOVA was applied as appropriate, followed by multiple comparisons correction. When data did not meet normality assumptions, non-parametric tests were used (Mann–Whitney U test for two groups and Kruskal–Wallis test with Dunn’s multiple comparisons test for multiple groups). Differences were considered statistically significant at p < 0.05.

5. Conclusions

In conclusion, IL-33 reprograms macrophage metabolism and function through the LPL–L-PC axis, leading to enhanced antiviral activity and alleviation of HSK pathology. These findings highlight the prophylactic potential of immunometabolic modulation and establish IL-33 as a key bridge linking cytokine signaling, metabolism, and antiviral immunity.