PIM1 Inhibition Alleviates Aspergillus fumigatus Keratitis by Regulating DDX41-Mediated STING Signaling Pathway

Abstract

Purpose

Fungal keratitis (FK) is a severe, sight-threatening infection with limited therapeutic options. Although the proto-oncogene PIM1 has been implicated in various inflammatory diseases, its role and underlying mechanisms in FK remain unknown.

Methods

We investigated the expression and function of PIM1 in human corneal epithelial cells (HCECs) and a mouse model of Aspergillus fumigatus keratitis using genetic and pharmacologic approaches. Protein–protein interactions and phosphorylation events were analyzed by co-immunoprecipitation, GST pull-down, and in vitro kinase assays. Inflammatory cytokine production and signaling pathway activation were assessed using ELISA, quantitative RT-PCR, and Western blot.

Results

A. fumigatus infection significantly upregulated PIM1 expression in both HCECs and mouse corneas. PIM1 overexpression enhanced the fungus-induced inflammatory response, whereas PIM1 knockdown or pharmacologic inhibition attenuated the production of proinflammatory cytokines (TNF-α, IL-6, IL-1β, IFN-β). Mechanistically, PIM1 directly interacted with and promoted serine phosphorylation of the DNA sensor DDX41. This PIM1–DDX41 interaction was essential for activating the downstream STING–TBK1–IRF3 signaling pathway. The proinflammatory effects of PIM1 were abolished by DDX41 knockdown or STING inhibition. Importantly, both genetic silencing and pharmacologic inhibition of PIM1 alleviated disease severity, reduced fungal load, and suppressed the DDX41/STING-mediated inflammatory cascade in the mouse model of FK.

Conclusions

Our findings reveal a novel mechanism by which PIM1 aggravates A. fumigatus keratitis through activation of the DDX41–STING signaling axis, highlighting PIM1 as a promising therapeutic target for FK.

Fungal keratitis (FK) is a severe, sight-threatening corneal infection commonly caused by pathogens including Aspergillus, Fusarium, and Candida species.¹ It is particularly prevalent in tropical and subtropical regions, where ocular trauma from plant or vegetable matter is a major risk factor. Other significant predisposing factors include contact lens wear, misuse of topical corticosteroids, and underlying ocular surface diseases.^2,3 Standard treatment involves topical antifungal agents such as natamycin or voriconazole. However, challenges such as poor corneal drug penetration and rising antifungal resistance often complicate management, frequently requiring surgical interventions such as penetrating keratoplasty in advanced cases.⁴ These challenges underscore the urgent need to develop novel and effective antifungal strategies to overcome drug resistance and improve therapeutic outcomes in FK.

Human corneal epithelial cells (HCECs), which form the outermost layer of the cornea, serve as the primary physical and immunologic barrier against environmental pathogens.^5,6 Beyond functioning as a passive barrier, HCECs act as active sentinels of the innate immune system.⁷ They express an array of pattern recognition receptors, including Toll-like receptors (TLRs) and C-type lectin receptors, which detect conserved fungal pathogen–associated molecular patterns such as β-glucans and mannans. Upon ligand binding, HCECs initiate robust intracellular signaling cascades that lead to the production of diverse proinflammatory cytokines (IL-1β, IL-6, TNF-α, IFN-β), chemokines (IL-8, CXCL8), and antimicrobial peptides (defensins, LL-37).⁸ This cytokine milieu plays a vital role in recruiting innate immune cells—particularly neutrophils—from the limbal vasculature and tears to the infection site.⁹ While an appropriate inflammatory response is essential for effective pathogen clearance, excessive or dysregulated inflammation can lead to immunopathologic tissue damage, corneal melting, and even perforation. Therefore, precise modulation of the immune response is crucial to achieving both effective antimicrobial defense and tissue preservation.

The provirus integration site for Moloney murine leukemia virus (PIM) kinases is a family of highly conserved serine/threonine kinases that consists of PIM1, PIM2, and PIM3.¹⁰ The proto-oncogene PIM1, as the earliest identified and most extensively studied member, has been associated with tumor progression, metastasis, and chemotherapy resistance.¹¹ Although traditionally associated with cancer development, emerging evidence indicates that PIM1 also plays a significant role in immune-inflammatory pathologies.¹² PIM1 exerts its effects through the phosphorylation of a diverse array of substrates directly implicated in inflammatory pathways, such as NF-κB, JAK/STAT, and AKT. Inhibition of PIM1 has been shown to reduce disease severity in animal models of rheumatoid arthritis,^13,14 inflammatory bowel disease,^15,16 Alzheimer's disease,¹⁷ and pulmonary fibrosis,¹⁸ underscoring its potential as a therapeutic target for inflammatory diseases. Recent studies further revealed that aberrant PIM1 expression contributes to Th17/Treg imbalance in autoimmune uveitis, a pathology that can be ameliorated through PIM1 inhibition.^19,20 However, the role of PIM1 in FK remains unknown. Elucidating its function and underlying mechanisms in this context is urgently needed and may reveal novel therapeutic strategies for the treatment of FK.

In this study, we explore the functional involvement of PIM1 in the inflammatory cascade associated with FK and clarify its mechanistic basis. Our results indicate that suppression of PIM1 activity significantly reduces Aspergillus fumigatus–triggered inflammatory responses in both HCECs and experimental mouse models. Moreover, we establish that PIM1 exerts its pathogenic influence by directly binding to DDX41 and orchestrating downstream STING signaling. These findings uncover previously unrecognized mechanisms driving immune pathology in FK and support the therapeutic potential of PIM1-targeted strategies for mitigating this vision-impairing disorder.

Materials and Methods

Cell Culture

HCECs, immortalized with SV40 and generously provided by Professor Fu-Shin Yu (Wayne State University), were cultured in Dulbecco's modified Eagle's medium (DMEM)/F12 (11320033; Gibco, Grand Island, NY, USA) supplemented with 50% defined keratinocyte–serum-free medium (K-SFM; 10744019, Gibco) and 1% penicillin/streptomycin (15070063; Gibco). Primary human corneal epithelial cells (PHCECs) were purchased from Sunncell (SNP-H134; Wuhan, China) and cultured in EpiLife medium (MEPI500CA; Gibco). The HEK293T cell line, obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China), was maintained in high-glucose DMEM (11965092, Gibco) containing 10% fetal bovine serum (FBS). Cell line authenticity was verified through short tandem repeat analysis. All cells were incubated at 37°C in a humidified atmosphere containing 5% CO₂.

A. fumigatus Preparation

A. fumigatus (strain 93024), obtained from the China Center for Type Culture Collection (CCTCC, Beijing, China), was processed to yield conidia and hyphae following established protocols.²¹ In brief, the fungus was cultured on Sabouraud dextrose agar (213400; BD Bioscience, Franklin Lakes, NJ, USA) at 37°C for 7 days. Subsequently, conidia and hyphae were collected and incubated in liquid Sabouraud medium with agitation at 500 rpm for 18 hours. The hyphae were mechanically fragmented into 20- to 40-µm segments, washed in phosphate-buffered saline (PBS), and pelleted by centrifugation. For in vitro assays, hyphal fragments were inactivated by heating at 56°C for 60 minutes and adjusted to a concentration of 1 × 10⁶ hyphal fragments/mL. For in vivo experiments, viable hyphal fragments were resuspended in PBS to achieve a final concentration of 1 × 10⁸ hyphal fragments/mL.

FK Mouse Model Construction

All animal procedures received approval from the Ethics Committee of Qilu Hospital of Shandong University and adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Male C57BL/6 mice, aged 7 to 8 weeks, were acquired from Gempharmatech (Nanjing, China). Under pentobarbital anesthesia, the central corneal surface was gently abraded with a 26-gauge needle. A 5-µL hyphal suspension was then applied to the ocular surface, followed by coverage with a tailored parafilm mold. After eyelid suturing, an additional 5 µL of the hyphal suspension was delivered into the interface between the parafilm and the cornea. At an indicated time postprocedure, the parafilm was removed, and the corneal assessments were performed. Clinical scores were calculated according to the methodology described by Wu et al.²²

Quantitative RT-PCR

Total RNA from HCECs was extracted using TRIzol reagent (15596018; Invitrogen, Carlsbad, CA, USA), while corneal RNA was isolated with the RNeasy Micro Kit (74004; Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. Reverse transcription was performed to synthesize cDNA using the PrimeScript RT reagent kit (RR037A; TaKaRa, Kyoto, Japan). Quantitative PCR was carried out with SYBR Premix Ex Taq (RR420A; TaKaRa) on the 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Primer sequences are provided in Supplementary Table S1. Gene expression was quantified using the 2^−ΔΔCt method, with β-actin as the reference gene for normalization.

Antibodies and Reagents

Antibodies targeting PIM2 (25865-1-AP, 1:1000), PIM3 (33202-1-AP, 1:1000), β-actin (20536-1-AP, 1:50,000), GST (10000-0-AP, 1:2000), MYC (16286-1-AP, 1:2000), Flag (66008-4-Ig, 1:2000), His (66005-1-Ig, 1:2000), STING (19851-1-AP, 1:2000), IRF3 (11312-1-AP, 1:1000), and TBK1 (28397-1-AP, 1:1000) were sourced from Proteintech (Wuhan, China). Antibodies recognizing phosphorylated STING (Ser366, 19781; Ser365, 72971; 1:1000), phosphorylated TBK1 (5483, 1:1000), phosphorylated IRF3 (4947, 1:1000), and DDX41 (15076, 1:1000) were obtained from Cell Signaling Technology (Danvers, MA, USA). Antibodies against PIM1 (ab300453, 1:1000) were acquired from Abcam (Cambridge, UK). Antibodies for phosphorylated threonine (sc-5267), serine (sc-81514), and tyrosine (sc-7020) were procured from Santa Cruz (Dallas, TX, USA). The compounds SGI-1776 (S2198) and C176 (S6575) were purchased from Selleck Chemicals (Houston, TX, USA).

RNA Interference

Small interfering RNAs (siRNAs) targeting PIM1 and DDX41 were designed and produced by GenePharma (Shanghai, China). HCECs were plated in 6-cm dishes and incubated overnight. The next day, transfection was performed with 80 nM siRNA or negative control (NC) siRNA for 24 hours using Lipofectamine 2000 Reagent (11668-019; Invitrogen). For in vivo experiments, PIM1 siRNA was delivered via subconjunctival injection into the cornea before infection with A. fumigatus. The sequences of the siRNAs are provided in Supplementary Table S2.

Plasmid Transfection

The Flag-PIM1 and Myc-DDX41 expression vectors were generated by cloning the coding sequences of PIM1 and DDX41 into pcDNA3.1 plasmids containing Flag or Myc tags, respectively (Addgene, Watertown, MA, USA). Cells were seeded in 6-cm dishes and allowed to adhere overnight. Subsequently, transfection was performed with 4 µg of plasmid DNA using Lipofectamine 2000 Reagent. Following 24 hours of transfection, the cells were subjected to A. fumigatus infection.

Western Blot

Western blot analysis was conducted according to established methods.²³ Mouse corneal tissues or HCECs were lysed with RIPA buffer (P0013K; Beyotime, Shanghai, China) containing protease inhibitor cocktail (P1005; Beyotime). Protein concentrations were determined using a BCA protein assay kit (P0010S; Beyotime). Equal amounts of protein were separated by SDS-PAGE and electrophoretically transferred to PVDF membranes (Merck Millipore, Burlington, MA, USA). After blocking with nonfat milk, the membranes were probed with specific primary antibodies and corresponding horseradish peroxidase (HRP)–conjugated secondary antibodies. Protein signals were visualized using a GE Amersham Imager 600 (GE, Chicago, IL, USA), and band intensities were quantified with ImageJ software (National Institutes of Health, Bethesda, MD, USA), normalized to β-actin as the loading control.

ELISA Analysis

The supernatant of HCEC cultures was harvested and centrifuged at 1000g for 5 minutes. Mouse corneal tissues were homogenized via ultrasonic disruption in PBS supplemented with a protease inhibitor cocktail, followed by centrifugation to harvest the soluble fraction. Concentrations of IL-1β, IL-6, TNF-α, and IFN-β in the samples were quantified using commercial ELISA kits according to the manufacturer's protocols. Absorbance was measured at 450 nm using a microplate reader. Information on the ELISA kits utilized is provided in Supplementary Table S3.

Fungal Load Determination

Corneas infected with A. fumigatus were processed using a Micro Tissue Grinder (OSE-Y30; TIANGEN BIOTECH, Beijing, China) in PBS to create a homogenate. A 30-µL portion of this homogenate was spread onto Sabouraud dextrose agar plates and incubated at 37°C for 24 hours to facilitate fungal colony growth. Fungal burden was quantified by counting the number of colony-forming units.

GST Pull-Down Assay

The coding sequences of DDX41 and PIM1 were inserted into pGEX-4T-1 and pET-22b(+) vectors to generate GST-DDX41 and His-PIM1 recombinant constructs, respectively. Recombinant proteins were expressed in Escherichia coli BL21 strains induced with 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and purified using commercial protein purification kits (Beyotime). For GST pull-down assays, purified GST-DDX41 and His-PIM1 were incubated with GST-tag Purification Resin (P2253; Beyotime) in protein binding buffer for 2 hours. The resin was subsequently washed twice with wash buffer, eluted by boiling in SDS loading buffer, and subjected to Western blot analysis.

Co-Immunoprecipitation

Cells were lysed in Western and IP Lysis Buffer (P0013B; Beyotime) containing protease inhibitor cocktail. Following centrifugation, the supernatant was harvested, and 800 µg of total protein was incubated overnight with 5 µL of primary antibody under constant rotation. Subsequently, magnetic beads were introduced, and the mixture was further incubated for 2 hours at 4°C. The beads were washed three times with lysis buffer, and bound proteins were eluted via boiling in SDS loading buffer prior to immunoblotting analysis. To minimize detection of antibody-derived signals, the following reagents were utilized: anti-Flag (KTSM1338; AlpaLifeBio, Shenzhen, China) Nanobody Magarose Beads, anti-Myc (KTSM1336; AlpaLifeBio) Beads, and HRP-conjugated Veriblot for IP Detection Reagent (ab131366; Abcam).

Mass Spectrometry Assay

The pcDNA3.1 (Flag) or pcDNA3.1 PIM1 (Flag-PIM1) plasmids were introduced into HCECs using Lipofectamine 2000 Reagent. After 48 hours, cells were lysed with Western and IP Lysis Buffer, and the supernatant was obtained following centrifugation. Cleared lysates were obtained by centrifugation and subsequently incubated overnight with anti-Flag beads. Following three washes with lysis buffer, the beads were boiled in SDS loading buffer to elute bound proteins. The eluates were separated by SDS-PAGE, and proteins were visualized using silver staining. Differential protein bands displaying notable intensity variations were excised and subjected to liquid chromatography/tandem mass spectrometry analysis (Qinglian Biotech Co., Ltd, Beijing, China) for protein identification.

In Vitro Kinase Assay

A 50-µL kinase reaction system was prepared, consisting of reaction buffer (25 mM Tris pH 7.5, 5 mM β-glycerophosphate, 2 mM DTT, 0.1 mM Na₃VO₄, 10 mM MgCl₂), 200 µM ATP, 25 ng His-PIM1, and 40 ng GST-DDX41. The mixture was incubated at 30°C for 30 minutes. Phosphorylation levels of DDX41 were assessed by immunoblotting using p-Ser specific antibody.

Hematoxylin and Eosin Staining

Formalin-fixed, paraffin-embedded corneal sections (4 µm) were deparaffinized in xylene and rehydrated through graded ethanol. Nuclei were stained with Mayer's hematoxylin for 8 minutes, followed by bluing in tap water. Cytoplasm was counterstained with eosin Y for 2 minutes. Then, sections were dehydrated, cleared in xylene, and mounted. Stained sections were imaged with an Olympus microscope (Tokyo, Japan).

Periodic Acid–Schiff Staining

Following deparaffinization and rehydration, corneal sections (4 µm) were oxidized with 1.0% aqueous periodic acid for 10 minutes. After rinsing, sections were incubated in Schiff's reagent for 20 minutes in the dark, allowing the aldehyde groups to react and produce a magenta color. Slides were then washed in running tap water for 10 minutes to develop the color. A Mayer's hematoxylin counterstain was applied for nuclear visualization (1 minute). Finally, sections were dehydrated, cleared in xylene, and mounted with a resinous medium. Stained sections were captured using an Olympus microscope.

Statistical Analysis

Statistical analyses were performed using GraphPad Prism (GraphPad Software, La Jolla, CA, USA), and image processing was carried out with Adobe Photoshop CC (Adobe, San Jose, CA, USA). Data are presented as mean ± SD from at least three independent experiments. For comparisons between two groups, a Student's t-test was applied, whereas one-way ANOVA was employed for evaluating differences among three or more groups. A P value of less than 0.05 was considered statistically significant.

Results

PIM1 Promotes A. fumigatus–Induced Inflammatory Response in HCECs

To determine whether A. fumigatus infection influences PIM1 expression, HCECs were exposed to heat-inactivated hyphae for 0 to 24 hours. Western blot analysis demonstrated a marked upregulation of PIM1 protein levels upon A. fumigatus stimulation, peaking at 12 hours. In contrast, expression of other PIM kinase family members, PIM2 and PIM3, remained unchanged (Figs. 1A–D). A similar induction of PIM1 was confirmed in PHCECs (Supplementary Fig. S1). To further explore the functional role of PIM1 in the A. fumigatus–induced inflammatory response, HCECs were transfected with PIM1 overexpression plasmid (pcDNA3.1-PIM1) prior to A. fumigatus challenge. Overexpression efficiency was validated by Western blot (Figs. 1E, 1F). As previously reported,²³ A. fumigatus stimulation significantly enhanced the transcription of proinflammatory cytokines TNF-α, IL-6, IL-1β, and IFN-β. This effect was further amplified by PIM1 overexpression (Fig. 1G). Similarly, ELISA results confirmed that PIM1 overexpression augmented the secretion of these cytokines in A. fumigatus–stimulated cells (Fig. 1H). Collectively, these findings indicate that PIM1 exacerbates the inflammatory response in HCECs following A. fumigatus infection.

Figure 1.

PIM1 enhances A. fumigatus–induced inflammatory response in HCECs. HCECs were stimulated with A. fumigatus hyphae (1 × 10⁶ hyphal fragments/mL) for 0 (control), 1, 3, 6, 12, and 24 hours. (A) Protein levels of PIM1, PIM2, PIM3, and β-actin were analyzed by Western blot. (B–D) Quantification of PIM1 (B), PIM2 (C), and PIM3 (D) expression normalized to β-actin. For overexpression studies, HCECs were transfected with either empty vector pcDNA3.1 (Vector) or PIM1-expressing plasmid pcDNA3.1-PIM1 (PIM1) for 24 hours, followed by stimulation with A. fumigatus hyphae for 12 hours. (E) Western blot was used to detect the protein levels of PIM1 and β-actin. (F) Quantification of PIM1 protein levels in (E). (G) mRNA expression levels of TNF-α, IL-6, IL-1β, and IFN-β were measured using qRT-PCR. (H) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in culture supernatants were assessed by ELISA. Data are presented as the mean ± SD; ^#P > 0.05, *P < 0.05, **P < 0.01; n = 3.

PIM1 Inhibition Attenuates the Inflammatory Response Induced by A. fumigatus Infection in HCECs

Given that PIM1 overexpression exacerbates inflammatory responses, we next sought to determine whether inhibition of PIM1 could attenuate A. fumigatus–induced inflammation. HCECs were transfected with either negative control siRNA or two independent PIM1-targeting siRNAs (siPIM1-1 and siPIM1-2), followed by stimulation with A. fumigatus hyphae. Western blot analysis confirmed efficient knockdown of PIM1 protein (Fig. 2A). Quantitative RT-PCR (qRT-PCR) results indicated that PIM1 silencing significantly reduced the mRNA expression of TNF-α, IL-6, IL-1β, and IFN-β in A. fumigatus–stimulated cells (Fig. 2B). Consistent with this, ELISA assays showed that PIM1 depletion markedly suppressed the secretion of these inflammatory cytokines (Fig. 2C). These anti-inflammatory effects were corroborated in PHCECs, where PIM1 knockdown similarly suppressed cytokine expression (Supplementary Figs. S2C–E). To further validate these findings pharmacologically, we employed SGI-1776,^24,25 a selective PIM1 inhibitor (Fig. 2D). Both qRT-PCR and ELISA analyses demonstrated that SGI-1776 treatment significantly downregulated the expression and secretion of TNF-α, IL-6, IL-1β, and IFN-β in A. fumigatus–challenged HCECs (Figs. 2E, 2F). Together, these genetic and pharmacologic interventions indicate that inhibition of PIM1 effectively mitigates the inflammatory response triggered by A. fumigatus in HCECs.

Figure 2.

PIM1 inhibition mitigates A. fumigatus–induced inflammatory response. HCECs were transfected with either negative control siRNA (siNC) or two specific PIM1-targeting siRNAs (siPIM1-1 and siPIM1-2) for 24 hours and then stimulated with A. fumigatus hyphae (1 × 10⁶ hyphal fragments/mL) for an additional 12 hours. (A) Western blot analysis of PIM1 and β-actin protein levels. Quantitative analysis of PIM1 is provided in Supplementary Figure S2A. (B) mRNA expression levels of TNF-α, IL-6, IL-1β, and IFN-β were evaluated by qRT-PCR. (C) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in culture supernatants were measured using ELISA. HCECs were pretreated with the PIM1 inhibitor SGI-1776 (0, 2, and 4 µM) for 12 hours prior to 12-hour stimulation with A. fumigatus hyphae. (D) Western blot analysis of PIM1 and β-actin protein levels. Quantitative data are presented in Supplementary Figure S2B. (E) qRT-PCR analysis of TNF-α, IL-6, IL-1β, and IFN-β mRNA expression. (F) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in culture supernatants were assessed by ELISA. Data are presented as the mean ± SD; *P < 0.05, **P < 0.01; n = 3.

PIM1 Phosphorylates DDX41 Through a Direct Interaction In Vitro

To investigate the mechanism underlying PIM1-mediated regulation of A. fumigatus keratitis, we conducted a mass spectrometry–based proteomic screen using PIM1 as bait. Among the identified interacting proteins, DEAD-box helicase 41 (DDX41) received the highest binding score and was selected for further validation (Fig. 3A). Direct binding between PIM1 and DDX41 was confirmed by GST pull-down assays using purified recombinant His-PIM1 and GST-DDX41 proteins (Fig. 3B). Furthermore, co-immunoprecipitation (co-IP) experiments demonstrated an endogenous interaction between PIM1 and DDX41 in HCECs (Fig. 3C). To provide additional evidence, exogenous Flag-PIM1 and Myc-DDX41 were coexpressed in HEK293T cells, and reciprocal co-IP confirmed their specific association (Figs. 3D, 3E).

Figure 3.

PIM1 directly interacts with DDX41 and regulates its phosphorylation. (A) HCECs transfected with either pcDNA3.1 (Flag) or pcDNA3.1 PIM1 (Flag-PIM1) were stimulated with A. fumigatus hyphae (1 × 10⁶ hyphal fragments/mL) for 12 hours. Cell lysates were immunoprecipitated using anti-Flag beads, and bound proteins were separated by SDS-PAGE and visualized by silver staining. (B) A GST pull-down assay was performed to examine direct binding between PIM1 and DDX41. (C) Endogenous interaction between PIM1 and DDX41 in HCECs was assessed by co-IP. HEK293T cells were cotransfected with Flag-PIM1 and Myc-DDX41 for 48 hours. (D) Co-IP was carried out with anti-Flag beads, and Myc-DDX41 was detected with anti-Myc antibody. (E) Reciprocal co-IP was performed using anti-Myc beads, and Flag-PIM1 was probed with anti-Flag antibody. (F) HCECs were transfected with NC siRNA (siNC) or PIM1 siRNAs (siPIM1-1 and siPIM1-2) for 24 hours, then exposed to A. fumigatus hyphae for 12 hours. DDX41 was immunoprecipitated and immunoblotted with anti–p-Ser and DDX41 antibodies. (G) HCECs were transfected with pcDNA3.1 (Vector) or pcDNA3.1-PIM1 (PIM1) for 24 hours, then exposed to A. fumigatus hyphae for 12 hours, followed by DDX41 immunoprecipitation and immunoblotting with indicated antibodies. (H) HCECs were transfected with NC siRNA (siNC), PIM1 siRNA (siPIM1-2), and/or pcDNA3.1-PIM1 (PIM1) for 24 hours, then treated with A. fumigatus hyphae for 12 hours. Co-IP of DDX41 was performed, and associated proteins were analyzed by Western blot. (I) Quantification of protein levels in (H). (J) HCECs were pretreated with SGI-1776 (0 and 4 µM) for 12 hours prior to 12-hour stimulation with A. fumigatus hyphae. Co-IP assay was performed using anti-DDX41 antibody and immunoblotted with the antibodies indicated. (K) Quantification of protein levels in (J). (L) Recombinant His-PIM1 and GST-DDX41 were incubated in reaction buffer with or without phosphatase for 30 minutes. The protein levels of p-Ser, His-PIM1, and GST-DDX41 were detected by Western blot. Quantification of protein levels in Figures 3F, G, and L is shown in Supplementary Figures S3A–C. Data are presented as the mean ± SD; **P < 0.01; n = 3.

As a serine/threonine kinase, PIM1 modulates downstream signaling through phosphorylation of target substrates.¹² We therefore hypothesized that PIM1 may regulate DDX41 via phosphorylation during fungal infection. Due to the lack of commercially available phosphospecific antibodies against DDX41, we performed immunoprecipitation of DDX41 and subsequently detected changes in its phosphorylation status using pan–anti-phosphoserine, anti-phosphothreonine, and anti-phosphotyrosine antibodies. Stimulation with A. fumigatus significantly enhanced serine phosphorylation of DDX41 but did not affect threonine or tyrosine phosphorylation (Fig. 3F). Genetic knockdown of PIM1 attenuated this serine phosphorylation (Fig. 3F), whereas PIM1 overexpression increased it (Fig. 3G). The impaired phosphorylation in PIM1-silenced cells was rescued by PIM1 reconstitution in HCECs with A. fumigatus infection (Figs. 3H, 3I). Consistent with these results, pharmacologic inhibition of PIM1 with SGI-1776 markedly suppressed A. fumigatus–induced serine phosphorylation of DDX41 (Figs. 3J, 3K). Finally, an in vitro kinase assay confirmed that PIM1 directly phosphorylates DDX41 at serine residues (Fig. 3L). Thus, these findings demonstrate that PIM1 physically interacts with DDX41 and promotes its serine phosphorylation in response to A. fumigatus infection.

PIM1 Regulates STING Signaling by Interacting With DDX41 in HCECs

Prior studies have shown that DDX41 activates STING and its downstream inflammatory signaling cascades during infection.^26,27 Activated STING orchestrates immune-inflammatory responses primarily through phosphorylation of downstream mediators such as TBK1 and IRF3.²⁸ To examine whether PIM1 influences STING pathway activation in A. fumigatus–stimulated HCECs, we evaluated phosphorylation of key STING signaling molecules. Overexpression of PIM1 markedly increased the protein levels of p-STING, p-TBK1, and p-IRF3 (Figs. 4A, 4C), indicating enhanced STING pathway activity. Furthermore, A. fumigatus treatment alone promoted phosphorylation of STING, TBK1, and IRF3, while PIM1 knockdown effectively attenuated this response (Figs. 4B, 4D). Reconstitution of PIM1 in knockdown cells restored the phosphorylation levels (Figs. 4E, 4H), and pharmacologic inhibition of PIM1 with SGI-1776 significantly suppressed the phosphorylation of STING, TBK1, and IRF3 (Figs. 4F, 4I). Given that PIM1 did not interact with STING (Supplementary Fig. S4A), we determined whether DDX41 is required for PIM1-mediated regulation of STING signaling. Western blot analysis revealed that PIM1 failed to elevate p-STING, p-TBK1, and p-IRF3 levels in DDX41-silenced cells, whereas DDX41 reconstitution restored the ability of PIM1 to enhance phosphorylation of these proteins (Fig. 4G). Additionally, PIM1 had no influence on cGAS expression in response to A. fumigatus infection (Supplementary Figs. S4B, S4C). In summary, these results demonstrate that PIM1 activates the STING signaling pathway in a DDX41-dependent manner during A. fumigatus stimulation.

Figure 4.

PIM1 regulates STING signaling through DDX41. (A) HCECs were transfected with pcDNA3.1 (Vector) or pcDNA3.1-PIM1 (PIM1) for 24 hours, followed by stimulation with A. fumigatus hyphae (1 × 10⁶ hyphal fragments/mL) for 12 hours. Protein levels of p-STING, STING, p-TBK1, TBK1, p-IRF3, IRF3, PIM1, and β-actin were assessed by Western blot. (B) HCECs were transfected with NC siRNA (siNC) or PIM1 siRNAs (siPIM1-1 and siPIM1-2) for 24 hours, then treated with A. fumigatus hyphae for 12 hours. Protein expression was analyzed via Western blot. (C) Quantification of protein levels in (A). (D) Quantification of protein levels in (B). (E) HCECs were transfected with NC siRNA (siNC), PIM1 siRNA-2 (siPIM1-2), and/or pcDNA3.1-PIM1 (PIM1) for 24 hours, followed by A. fumigatus hyphae treatment for 12 hours. Protein levels were detected by Western blot. (F) HCECs were pretreated with SGI-1776 (0, 2, 4 µM) for 12 hours, then exposed to A. fumigatus hyphae for 12 hours. Protein levels were detected by Western blot. (G) HCECs were transfected with NC siRNA (siNC), PIM1 siRNA-2 (siPIM1-2), DDX41 siRNA (siDDX41), or pcDNA3.1-DDX41 (DDX41) for 24 hours, followed by A. fumigatus hyphae treatment for 12 hours. Protein levels were detected by Western blot. (H) Quantification of protein levels in (E). (I) Quantification of protein levels in (F). Quantification of protein levels in Figure 4G is provided in Supplementary Figure S4D. Data are presented as the mean ± SD; **P < 0.01; n = 3.

PIM1 Modulates A. fumigatus–Induced Inflammatory Response Through DDX41/STING Signaling in HCECs

Building upon the observed interaction between PIM1 and DDX41 and their regulatory influence on STING signaling, we next sought to determine whether DDX41/STING contributes to PIM1-mediated inflammatory responses in A. fumigatus keratitis. After confirming efficient knockdown and subsequent reconstitution of DDX41 (Fig. 5A), we evaluated cytokine expression under these conditions. qRT-PCR and ELISA analyses revealed that PIM1 overexpression failed to enhance the mRNA and protein levels of TNF-α, IL-6, IL-1β, and IFN-β in DDX41-deficient cells. However, reintroduction of DDX41 restored the ability of PIM1 to promote proinflammatory cytokine production (Figs. 5B, 5C). To further examine the functional role of STING signaling in this regulatory axis, we inhibited STING activity using the specific inhibitor C176 (Fig. 5D). Both transcriptional and protein analyses showed that C176 treatment significantly attenuated the upregulation of TNF-α, IL-6, IL-1β, and IFN-β induced by PIM1 overexpression (Figs. 5E, 5F). These results indicate that PIM1 enhances the inflammatory response to A. fumigatus in HCECs through a mechanism that involves DDX41 and requires functional STING signaling.

Figure 5.

PIM1 regulates A. fumigatus–induced inflammatory response by interacting with DDX41. HCECs were transfected with NC siRNA (siNC), PIM1 siRNA-2 (siPIM1-2), DDX41 siRNA (siDDX41), and/or pcDNA3.1-DDX41 (DDX41) for 24 hours, followed by stimulation with A. fumigatus hyphae (1 × 10⁶ hyphal fragments/mL) for 12 hours. (A) Protein levels of DDX41, PIM1, and β-actin were analyzed by Western blot. Quantitative data are shown in Supplementary Figure S5A. (B) mRNA expression levels of TNF-α, IL-6, IL-1β, and IFN-β were evaluated by qRT-PCR. (C) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in culture supernatants were measured using ELISA. HCECs were transfected with pcDNA3.1 (Vector) or pcDNA3.1-PIM1 (PIM1) for 24 hours and then treated with A. fumigatus hyphae for 12 hours. C176 (1 µM) was added to the medium for 24 hours before harvest. (D) Protein levels of PIM1 and β-actin were analyzed by Western blot. Quantitative analysis is shown in Supplementary Figure S5B. (E) mRNA expression levels of TNF-α, IL-6, IL-1β, and IFN-β were evaluated by qRT-PCR. (F) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in culture supernatants were measured using ELISA. Data are presented as the mean ± SD; *P < 0.05, **P < 0.01; n = 3.

PIM1 Knockdown Attenuates A. fumigatus Keratitis in a Mouse Model

Mice represent a well-established model system for investigating corneal infections worldwide.^29,30 In the present study, we evaluated the functional significance of PIM1 using a mouse model of A. fumigatus keratitis. Following infection with live fungal hyphae for durations of 0.5, 1, and 3 days, mouse corneas were collected for further analysis. Western blot analysis showed that A. fumigatus infection significantly increased the protein levels of PIM1 and p-STING, with maximal expression observed at 1 day postinfection (Fig. 6A). This time point was therefore selected for subsequent experiments. Concurrently, we also detected a significant increase in serine phosphorylation of DDX41 in response to infection (Fig. 6B). To knock down PIM1 expression in vivo, PIM1-specific siRNAs were administered via subconjunctival injection 1 day prior to fungal challenge. Efficient knockdown of PIM1 was confirmed by Western blot (Fig. 6F). Mice treated with PIM1 siRNA exhibited reduced corneal ulceration, diminished opacity, lower clinical scores, and decreased fungal load compared to control animals (Figs. 6C–E). Furthermore, hematoxylin and eosin and periodic acid–Schiff staining revealed that fungal infection induced significant inflammatory cell infiltration, stromal disruption, and fungal invasion in the corneal stroma. These pathologic changes were markedly alleviated after PIM1 knockdown (Supplementary Fig. S6). PIM1 knockdown also attenuated the A. fumigatus–induced increases in phosphorylation levels of STING and DDX41 (Figs. 6F, 6G). Both mRNA and protein levels of the inflammatory cytokines TNF-α, IL-6, IL-1β, and IFN-β were also reduced following PIM1 silencing, as measured by qRT-PCR and ELISA (Figs. 6H, 6I). Collectively, these results demonstrate that PIM1 knockdown alleviates the inflammatory response and disease severity in a mouse model of A. fumigatus keratitis.

Figure 6.

Knockdown of PIM1 ameliorates A. fumigatus keratitis in mice. (A) Mouse corneas were infected with 5 µL of live A. fumigatus hyphae (1 × 10⁸ hyphal fragments/mL) for 0.5, 1, and 3 days, then excised for analysis. Protein levels of p-STING, STING, PIM1, and β-actin were assessed by Western blot. Quantitative data are provided in Supplementary Figure S7A. (B) DDX41 was immunoprecipitated from corneal lysates and immunoblotted with anti–p-Ser and anti-DDX41 antibodies. Quantification is shown in Supplementary Figure S7B. Mice received subconjunctival injection of 5 µL control siRNA (siNC, 10 µM) or PIM1 siRNAs (siPIM1-1/siPIM1-2, 10 µM). After 24 hours, corneas were infected with 5 µL of live hyphae for an additional 24 hours before harvesting. (C) Keratitis severity was evaluated by slit-lamp examination. (D) Clinical scores were calculated based on slit-lamp observations. (E) Fungal burden was quantified by colony-forming unit (CFU) assays. (F) Protein expression of p-STING, STING, PIM1, and β-actin in mouse corneas was detected by Western blot. Quantitative results are shown in Supplementary Figure S7C. (G) DDX41 was immunoprecipitated and immunoblotted with anti–p-Ser and anti-DDX41 antibodies. Quantification is presented in Supplementary Figure S7D. (H) mRNA expression of TNF-α, IL-6, IL-1β, and IFN-β was measured by qRT-PCR. (I) Secreted levels of TNF-α, IL-6, IL-1β, and IFN-β in corneal homogenates were detected using ELISA. Data are presented as the mean ± SD; **P < 0.01; n = 6.

PIM1 Inhibitor Ameliorates A. fumigatus Keratitis In Vivo

To further evaluate the therapeutic potential of targeting PIM1 in FK, the specific inhibitor SGI-1776 was administered to inhibit PIM1 kinase activity in a mouse model of A. fumigatus infection. Slit-lamp examination revealed that SGI-1776 treatment markedly attenuated corneal damage and inflammatory responses induced by A. fumigatus (Fig. 7A). Clinical scoring and fungal burden assays further supported these findings, demonstrating significantly reduced disease severity and microbial load in inhibitor-treated corneas (Figs. 7B, 7C). Western blot analysis indicated that SGI-1776 effectively suppressed the activation of both DDX41 and STING signaling (Figs. 7D, 7E), aligning with previous observations in HCECs. Moreover, transcriptional and cytokine profiling via qRT-PCR and ELISA confirmed that PIM1 inhibition significantly downregulated key proinflammatory mediators, including TNF-α, IL-6, IL-1β, and IFN-β (Figs. 7F, 7G). Taken together, these results demonstrate that pharmacologic inhibition of PIM1 alleviates A. fumigatus–induced keratitis by modulating the DDX41–STING signaling axis.

Figure 7.

PIM1 inhibitor inhibits A. fumigatus keratitis in vivo. Mouse corneas were infected with 5 µL of live A. fumigatus hyphae (1 × 10⁸ hyphal fragments/mL) for 24 hours. Two hours prior to infection, mice received a subconjunctival injection of 5 µL SGI-1776 (1 µg/µL or 2 µg/µL). Corneas were harvested 24 hours postinfection. (A) Keratitis severity was evaluated by slit-lamp examination. (B) Clinical scores were calculated based on slit-lamp observations. (C) Fungal burden was quantified by CFU assays. (D) Protein expression of p-STING, STING, PIM1, and β-actin in mouse corneas was analyzed by Western blot. Quantification is presented in Supplementary Figure S8A. (E) DDX41 was immunoprecipitated from corneal lysates and immunoblotted with anti–p-Ser and anti-DDX41 antibodies. Quantification is provided in Supplementary Figure S8B. (F) mRNA expression levels of TNF-α, IL-6, IL-1β, and IFN-β were measured by qRT-PCR. (G) Protein levels of TNF-α, IL-6, IL-1β, and IFN-β in corneal homogenates were determined using ELISA. Data are presented as the mean ± SD; *P < 0.05, **P < 0.01; n = 6.

Discussion

The three members of the PIM kinase family (PIM1, PIM2, and PIM3) share high amino acid sequence similarity, suggesting that they may perform both overlapping and distinct functions.³¹ In this study, we found that all three PIM kinases were expressed at low basal levels in normal HCECs. However, upon stimulation with A. fumigatus, only PIM1 showed a substantial increase in protein expression. This upregulation of PIM1 was further confirmed in A. fumigatus–infected corneal tissues. These results align with previous reports indicating elevated PIM1 levels in various immune-inflammatory diseases, including lupus nephritis,³² allergic asthma,³³ and rheumatoid arthritis.³⁴ Nevertheless, the upstream regulatory mechanisms controlling PIM1 expression require further in-depth investigation. PIM1 serves as a critical mediator in cytokine signal transduction. Its expression is rapidly induced by inflammatory stimuli such as lipopolysaccharide, TNF-α, and interleukins.³⁵ By phosphorylating and activating downstream substrates, PIM1 amplifies inflammatory responses across multiple disease contexts.³⁶ Accordingly, our study demonstrated that aberrantly elevated PIM1 exacerbated the inflammatory response triggered by A. fumigatus infection. Conversely, either knockdown of PIM1 or treatment with a PIM1 inhibitor significantly attenuated A. fumigatus–induced inflammation. The characteristic overexpression and proinflammatory role of PIM1 highlight its potential as a therapeutic target for FK.

PIM1 is constitutively active and functions primarily by phosphorylating a wide range of substrate proteins, thereby modulating their stability, activity, and subcellular localization. To date, numerous PIM1 substrates have been identified, which are involved in diverse biological processes, including apoptosis, cell cycle progression, gene transcription, cellular metabolism, and inflammatory responses.³⁶ For instance, PIM1 suppression has been shown to modulate c-Myc via phosphorylation of histone H3 at serine 10 (S10) and c-Myc at serine 62 (S62).^37,38 These phosphorylation events are critical for c-MYC–mediated transcriptional activation and contribute to oncogenic transformation.³⁹ Additionally, PIM1 phosphorylates the p65/RelA subunit of NF-κB, leading to enhanced NF-κB activity and subsequent upregulation of IL-6 expression. This increase in IL-6 further promotes PIM1 expression, forming a self-reinforcing inflammatory feedback loop.^40,41 In this study, we identified DDX41 as a novel interacting protein of PIM1. We further demonstrated that PIM1 deletion significantly suppresses DDX41 serine phosphorylation, establishing PIM1 as a key kinase mediating this posttranslational modification. Importantly, our results indicate that PIM1 enhances A. fumigatus–induced inflammatory responses via its interaction with DDX41. These findings uncover a previously unrecognized mechanism through which PIM1 regulates innate immune signaling in FK. However, the specific domain or exact phosphorylation sites within DDX41 targeted by PIM1 remain to be precisely mapped. Studies to elucidate these molecular details are currently ongoing.

DDX41, a member of the DEAD-box RNA helicase family, participates in pre-mRNA splicing as a core component of the spliceosome. It also functions as a key sensor for both microbial and endogenous nucleic acids, linking nucleic acid detection to the activation of downstream inflammatory pathways.⁴² Central to this role is its functional partnership with the adaptor protein STING. Following ligand binding, DDX41 undergoes conformational changes and recruits STING through direct protein–protein interactions. Activated STING then facilitates the phosphorylation and activation of TBK1, which in turn phosphorylates the transcription factor IRF3.⁴³ The DDX41–STING–TBK1–IRF3 axis forms a critical signaling cascade that orchestrates type I interferon and proinflammatory cytokine production in response to invading pathogens.²⁷ Bruton's tyrosine kinase has been reported to phosphorylate DDX41 at Tyr364 and Tyr414, modifications essential for DNA recognition and the subsequent binding of DDX41 to STING.²⁶ In the present study, we identify a novel mechanism for the activation of the DDX41–STING pathway, demonstrating that PIM1 directly interacts with and phosphorylates DDX41 at serine residues. This phosphorylation then promotes downstream STING signaling activation and enhances the production of proinflammatory cytokines upon A. fumigatus infection. While our previous work revealed cGAS–STING pathway activation in response to fungal stimulation,⁴⁴ the current study shows that PIM1 knockdown does not affect cGAS expression. These findings suggest that the cGAS and PIM1 pathways may operate in parallel, each contributing distinctly to A. fumigatus–induced inflammation. The PIM1–DDX41 axis likely represents a parallel signaling route that activates STING independently of cGAS.

Given the prevalent overexpression of PIM1 in diverse malignancies, numerous PIM1-targeted inhibitors have been developed, showing substantial therapeutic efficacy in cancer treatment.¹¹ Furthermore, PIM1 inhibition has emerged as a promising approach to modulate immune cell activation and cytokine production in inflammatory disorders.¹² For instance, SGI-1776 was shown to ameliorate airway inflammation and mucus hypersecretion in asthma by upregulating Runx3.⁴⁵ Similarly, the PIM1 inhibitor SMI-4a alleviates osteoarthritis by suppressing NLRP3 inflammasome activation in macrophages,⁴⁶ and the novel inhibitor KMU-470 attenuates inflammatory signaling via the TLR4–NF-κB–NLRP3 axis in RAW 264.7 cells.⁴⁷ In this study, we demonstrate that SGI-1776 significantly reduces inflammatory responses and corneal tissue damage induced by A. fumigatus infection. Mechanistically, SGI-1776 suppresses DDX41 activation, leading to inhibition of the downstream STING signaling pathway and subsequent attenuation of the inflammatory cascade. These findings suggest that pharmacologic targeting of PIM1 may offer a novel therapeutic approach for FK. However, its pharmacokinetic properties and long-term safety profile for ocular application require further comprehensive evaluation before clinical translation can be considered.

This study identifies PIM1 as a critical regulator of inflammatory responses in corneal epithelial cells via the DDX41–STING signaling pathway. Nevertheless, several limitations should be acknowledged. Although the corneal epithelium serves as the primary barrier against fungal pathogens, stromal fibroblasts and infiltrating immune cells also contribute substantially to the inflammatory response observed in fungal keratitis. The role of PIM1 in these additional cell populations remains unclear and warrants further investigation. Addressing these questions in future studies will help to more comprehensively define the function of PIM1 in the pathogenesis of fungal keratitis.

In summary (Fig. 8), our results demonstrated that A. fumigatus infection upregulates PIM1 expression, leading to phosphorylation and activation of its binding partner DDX41. This activation initiates downstream STING signaling, thereby amplifying the inflammatory response to A. fumigatus challenge. Notably, pharmacologic inhibition of PIM1 attenuated A. fumigatus keratitis by suppressing DDX41 activation. These findings reveal the role of the PIM1/DDX41/STING axis in the pathogenesis of FK and support the therapeutic targeting of PIM1 for managing this condition.

Figure 8.

Working model of the PIM1-DDX41-STING axis in A. fumigatus keratitis. A. fumigatus infection upregulates PIM1 expression. PIM1 then directly binds to and phosphorylates the innate immune sensor DDX41. Phosphorylated DDX41 activates the STING signaling pathway, leading to the phosphorylation of TBK1 and IRF3. This cascade ultimately promotes the transcription and secretion of proinflammatory cytokines (TNF-α, IL-6, IL-1β, IFN-β), contributing to the immunopathology of fungal keratitis. Pharmacologic inhibition of PIM1 by SGI-1776 disrupts this axis, attenuating DDX41/STING activation and the subsequent inflammatory response.