IL-32 drives PR3-dependent NET formation to promote portal vein tumor thrombosis and radioresistance in HCC

Highlights

• •IL-32 is upregulated in HCC and further increased in PVTT.
• •IL-32 expression is associated with NET formation in HCC and PVTT.
• •Tumor-derived IL-32 promotes NET formation, with PR3 acting as a functional mediator.
• •NETs promote PVTT progression and are associated with radiotherapy resistance.
• •The IL-32–PR3–NET axis represents a potential therapeutic target in HCC.

Abstract

Background

IL-32 is a pro-inflammatory cytokine associated with chronic inflammation. Neutrophil extracellular traps (NETs) contribute to thrombosis, tumor invasion, and metastasis. This study investigated whether IL-32 derived from hepatocellular carcinoma (HCC) cells induces NET formation, promotes portal vein tumor thrombus (PVTT), and contributes to radiotherapy resistance.

Methods

Immunofluorescence was used to assess co-localization of IL-32 and NET markers in tumor tissues. Mass cytometry analyzed immune cell composition in the tumor microenvironment. Neutrophils from healthy donors were used to evaluate NET formation. PAD4⁻/⁻ mice and DNase I treatment were used to assess NET function in vivo.

Results

IL-32 and NETs were significantly elevated and positively correlated in HCC, particularly in patients with PVTT. Tumor-derived IL-32 induced robust NET formation in a dose-dependent manner, which was markedly attenuated by PR3 inhibition, identifying PR3 as a critical mediator. In HCC and PVTT tissues, IL-32 co-localized with NET markers and was associated with increased neutrophil infiltration. In vivo, NET depletion significantly suppressed intrahepatic metastasis. Notably, radiotherapy further enhanced NET formation, and elevated IL-32–NET levels were associated with radioresistance in patients.

Conclusion

Tumor-derived IL-32 drives PR3-dependent NET formation, promoting PVTT progression and contributing to radiotherapy resistance in HCC. Targeting the IL-32–PR3–NET axis may represent a potential therapeutic strategy.

Background

Liver cancer remains a major global health challenge and continues to impose a substantial burden on public health. Liver cancer ranks fourth in incidence among digestive system malignancies and third in cancer-related mortality, underscoring its high lethality and poor clinical outcomes [1]. Although the prevalence of chronic alcoholic hepatitis is increasing, chronic viral hepatitis caused by HBV and HCV remains the leading cause of hepatocellular carcinoma, with disease progression typically advancing from persistent hepatitis to cirrhosis and ultimately malignant transformation [2,3]. IL-32, initially identified as natural killer cell transcript 4, can stimulate NK cells and multiple immune cell types to produce inflammatory mediators and is recognized as an important pro-inflammatory cytokine [4,5]. IL-32 has been closely associated with chronic inflammatory diseases such as rheumatoid arthritis and systemic lupus erythematosus, and is considered a key driver of sustained inflammatory responses [6,7]. In HCC, IL-32 also plays a significant role, with its mRNA and protein expression markedly elevated in tumor tissues, particularly in regions adjacent to vascular invasion. Serum IL-32 levels progressively increase from healthy controls to acute hepatitis, chronic hepatitis, cirrhosis, and HCC, suggesting a strong association between IL-32–mediated inflammation and the severity of liver injury [8,9]. Although a specific receptor for IL-32 has not yet been identified, previous studies have shown that IL-32 interacts with proteinase 3 (PR3) on neutrophils, and PR3-mediated proteolytic processing can enhance IL-32 bioactivity [10]. However, whether IL-32 is capable of activating neutrophils to induce neutrophil extracellular trap (NET) formation remains unclear.

In recent years, neutrophils have gained increasing attention for their tumor-promoting functions [11]. Rather than serving solely as innate immune effector cells, they have been shown to facilitate tumor progression through multiple mechanisms[[11], [12], [13]]. NETs, released by activated neutrophils and composed of chromatin fibers and granular enzymes such as myeloperoxidase, were initially described as antimicrobial structures [12,14]. In HCC, however, NET formation is markedly increased and contributes to tumor initiation, recurrence, and metastasis by amplifying inflammatory signaling, enhancing immunosuppression, disrupting endothelial barriers, and capturing circulating tumor cells [15,16]. Given the close association between NETs and thrombosis, whether NETs contribute to the formation of portal vein tumor thrombus (PVTT)—and thereby promote the invasive and metastatic behavior of HCC—remains unknown [17]. Therefore, we focused on determining whether IL-32 can activate neutrophils and promote NET formation, thereby facilitating PVTT development and promoting distant metastasis.

Patients with HCC generally have poor outcomes, particularly those with PVTT. Radiotherapy is an important therapeutic option for HCC with or without PVTT [18,19]. However, despite advancements in radiotherapy technologies and treatment strategies, radioresistance and disease recurrence remain major obstacles [20]. Although previous studies have shown that tumor-intrinsic mechanisms, including enhanced DNA damage repair and impaired ferroptosis, contribute to HCC radioresistance [21,22], the mechanisms linking innate immune responses to radioresistance remain poorly understood. In particular, whether neutrophils and neutrophil extracellular traps (NETs) contribute to the development of radioresistance has not been clearly established. Given the important role of NETs in tumor progression and metastasis, we sought to investigate their potential contribution to HCC radioresistance.

Methods

Cell lines and clinical sample collection

Primary human neutrophils were cultured in RPMI-1640 medium supplemented with 1–2% fetal bovine serum and used within 24 hours of isolation. All cell lines were routinely tested for mycoplasma contamination and confirmed to be negative. Peripheral blood samples from healthy donors, patients with hepatocellular carcinoma (HCC), and patients with HCC accompanied by portal vein tumor thrombus (PVTT), as well as fresh tumor tissues, were collected at Qingdao Municipal Hospital. Written informed consent was obtained from all participants, and the study protocol was approved by the Institutional Ethics Committee of Qingdao Municipal Hospital. Inclusion criteria were as follows: (1) a confirmed diagnosis of primary HCC; (2) a confirmed diagnosis of HCC with PVTT; (3) no prior chemotherapy or radiotherapy; and (4) availability of complete clinical records. Patients with concurrent extrahepatic malignancies were excluded.

Immunofluorescence staining of tissue sections

Formalin-fixed, paraffin-embedded sections (4 μm) from HCC, adjacent peritumoral tissues, and portal vein tumor thrombi (PVTT) were subjected to standard immunofluorescence staining. Sections were deparaffinized, rehydrated through graded ethanol, and underwent antigen retrieval in citrate buffer (pH 6.0) using microwave heating for 15 min. After cooling, slides were blocked with 3% BSA or 5% normal serum for 1 h to reduce background fluorescence. Primary antibodies against IL-32 (1:500, Proteintech), MPO (1:300, Proteintech), and H3Cit (1 μg/mL, Abcam) were applied overnight at 4°C in a humidified chamber. After washing with PBS, Alexa Fluor–conjugated secondary antibodies (488, 594, or 647; 1:500, Thermo Fisher Scientific) were added for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI (1 μg/mL) for 5 min. Finally, slides were mounted with antifade medium and visualized using a Leica confocal fluorescence microscope. Quantification was performed by analyzing ≥5 randomly selected high-power fields (400×) per sample using ImageJ software.

NET formation

Human neutrophils were freshly isolated from peripheral blood of healthy donors by density-gradient centrifugation and hypotonic erythrocyte lysis. Purity (>90%) and viability (>95%) were confirmed by flow cytometry (CD66b⁺/CD16⁺) and trypan blue exclusion. Cells were suspended in RPMI-1640 medium containing 1% heat-inactivated FBS at 5 × 10⁵ cells/mL. For NET induction, 200 µL of neutrophil suspension was seeded on poly-L-lysine–coated 24-well plates and stimulated for 3–4 h under indicated conditions: (i) 10–20% (v/v) plasma from healthy donors, HCC, or HCC-PVTT patients; (ii) recombinant IL-32 (rIL-32, 200–1400 pg/mL, Abcam); or (iii) PBS, rIL-32 (1000 pg/mL), rIL-32 + PR3-neutralizing antibody (clone WGM2, Abcam, ab91181), or phorbol 12-myristate 13-acetate (PMA) (20 nM, positive control, Sigma). Cells were then fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 5% BSA. After overnight incubation with primary antibodies against MPO and H3Cit (both 1:200, Abcam) at 4°C, Alexa Fluor-conjugated secondary antibodies (1:500, Thermo Fisher) were applied for 1 h in the dark. DNA was counterstained with DAPI or SYTOX Green (1:10,000). Images were acquired using a Leica confocal microscope, and MPO⁺/H3Cit⁺ extracellular structures were quantified from ≥5 random fields using ImageJ.

Measurement of serum IL-32 and NET level

Plasma samples from HCC patients, HCC patients with PVTT, and healthy donors were analyzed by ELISA. Circulating IL-32 levels were measured using a commercial human IL-32 ELISA kit according to the manufacturer’s protocol. NET-associated MPO–DNA complexes were quantified using a modified capture ELISA. Briefly, 96-well plates were coated with anti-MPO antibody (5 μg/mL) overnight at 4°C, blocked with 1% BSA, and incubated with 100 μL diluted plasma for 2 h at room temperature. After washing with PBST, bound DNA was detected with PicoGreen® dsDNA reagent, and fluorescence intensity (480/520 nm) was measured using a plate fluorometer. Results were normalized to the mean value of healthy controls. Correlation between IL-32 and MPO–DNA levels was analyzed using Spearman’s rank test.

Establishment of the portal vein tumor thrombus (PVTT) mouse model

Male C57BL/6 wild-type (WT) and Pad4⁻/⁻ mice (6–8 weeks, 18–22 g) were maintained under specific pathogen-free (SPF) conditions with ad libitum access to food and water. All animal experiments were approved by the Institutional Animal Care and Use Committee of Shandong University, Qingdao Campus, and were conducted in accordance with the ARRIVE guidelines. For PVTT induction, mice were anesthetized with 2% isoflurane, and a midline abdominal incision was made to expose the portal vein under a stereomicroscope. Hepa1-6 hepatocellular carcinoma cells (1 × 10⁶ in 100 μL PBS) were slowly injected into the main portal vein using a 30-gauge insulin syringe. Hemostasis was achieved by gentle compression for 1–2 min, followed by two-layer closure with 5-0 absorbable sutures. Postoperatively, mice received buprenorphine (0.05 mg/kg, s.c.) for analgesia and were monitored daily for recovery, body weight, and general condition. For interventional studies, recombinant IL-32 (rIL-32, 1 μg/mouse, i.v.) or DNase I (100 U/mouse, i.p.) was administered every other day starting 24 h post-surgery. Tumor burden and intrahepatic spread were evaluated on days 14–21 post-injection by in vivo bioluminescence imaging (IVIS Spectrum, PerkinElmer). At endpoint, livers were harvested for histopathology (H&E).

Eligible C57BL/6 mice were randomly assigned to the indicated groups, whereas Pad4⁻/⁻ mice constituted a prespecified genotype group (6 male mice per group, 6 weeks old). Tumor take was assessed by bioluminescence imaging, and PVTT was confirmed histologically by the presence of tumor-cell aggregates within the portal-vein lumen. Imaging and histological analyses were performed by investigators blinded to group allocation.

Establishment of the orthotopic hepatocellular carcinoma patient-derived xenograft model

Six-week-old NCG mice were used to establish the orthotopic hepatocellular carcinoma patient-derived xenograft (PDX) model. Surgically resected tumor tissues obtained from patients with hepatocellular carcinoma (HCC) were cut into approximately 1 × 1 × 1 mm³ fragments and implanted subcutaneously into the flanks of recipient NCG mice. When the subcutaneous tumors reached approximately 10 mm in diameter, they were excised and cut into 1 × 1 × 1 mm³ fragments for orthotopic implantation. Recipient mice were anesthetized by intraperitoneal injection of 4% chloral hydrate at 0.1 mL per 10 g body weight. Approximately 5 min after administration, adequate anesthetic depth was confirmed, and the upper abdomen was disinfected with 75% ethanol. A longitudinal upper-abdominal incision of approximately 1 cm was made through the skin and subcutaneous tissues. Gentle pressure was applied to expose the left hepatic lobe. A blind-ended implantation tract was created in the left hepatic lobe using a 25-gauge needle. After hemostasis was achieved and residual blood was removed, each tumor fragment was coated with Matrigel and inserted into the implantation tract. A gelatin sponge was applied over the implantation site to seal the tract and control bleeding. The abdominal incision was then closed. Tumor growth was monitored after implantation. Approximately 4 weeks later, when the orthotopic tumors reached approximately 6 mm in diameter, the mice underwent irradiation. Mice were euthanized 1 weeks after irradiation. The orthotopic xenografts were harvested, weighed, and processed for analysis.

Immunohistochemistry (IHC) staining

Immunohistochemical staining was performed on paraffin-embedded sections using a two-step protocol. After deparaffinization, rehydration, and microwave antigen retrieval (15 min), endogenous peroxidase was quenched with 3% H₂O₂ and sections were blocked with normal serum. Slides were incubated with primary antibodies against IL-32 (Proteintech, 1:500) and H3Cit (Abcam, 1 µg/mL) at 37°C for 2 h, followed by species-appropriate secondary antibodies. Signals were visualized with DAB and counterstained with hematoxylin. Staining intensity and the percentage of positive cells were evaluated in five randomly selected high-power fields (400×).

The modified response evaluation criteria in solid tumors (mRECIST)

Tumor response was evaluated using contrast-enhanced CT or MRI at 8–12 weeks after radiotherapy according to the modified Response Evaluation Criteria in Solid Tumors (mRECIST) for hepatocellular carcinoma. Complete response was defined as the disappearance of arterial enhancement in all target lesions, whereas partial response was defined as a decrease of at least 30% in the sum of the diameters of viable, arterially enhancing target lesions relative to baseline. Progressive disease was defined as an increase of at least 20% in the sum of the diameters of viable target lesions relative to the smallest value recorded or the appearance of new lesions. Cases meeting neither partial-response nor progressive-disease criteria were classified as stable disease. Patients with complete or partial response were categorized as the radiotherapy-responsive group, whereas those with stable or progressive disease were categorized as the radiotherapy-non-responsive group. The two groups comprised 3 patients each. Treatment response was independently evaluated by two investigators who were blinded to the experimental data, with disagreements resolved by consensus.

Mass cytometry

Fresh tissue samples were minced and enzymatically dissociated in RPMI 1640 containing collagenase IV (1 mg/mL), hyaluronidase (250 μg/mL), and DNase I (20 μL/mL) for 60 min at 37°C. Cell suspensions were filtered through a 70-μm strainer, subjected to ACK red blood cell lysis, and washed with FACS buffer (PBS containing 0.5% BSA). Up to 3 × 10^6 cells per sample were stained with metal-conjugated surface and intracellular antibodies following Fc-blocking, fixation, and permeabilization. Antibodies were conjugated using the MaxPAR Antibody Labeling Kit (Fluidigm), and clones are listed in Supplementary Table 1. Data were acquired using a mass cytometer (PTL-MC601, PLTtech) with EQ Four Element Calibration Beads (Fluidigm). FCS files were bead-normalized and analyzed using FlowJo. After exclusion of debris, dead cells, and doublets, live singlets were subjected to PARC clustering and t-SNE dimensionality reduction. Cell populations were annotated based on marker expression, and population frequencies were compared between groups using Student’s -test.

Statistical analysis

All data are presented as mean ± SEM unless otherwise indicated. Differences between two groups were assessed using unpaired two-tailed Student’s t test, and comparisons among multiple groups (healthy donors, HCC, and HCC with PVTT) were analyzed by one-way ANOVA followed by Tukey’s post hoc test. The correlation between plasma IL-32 levels and MPO–DNA complexes was evaluated using Spearman’s rank correlation. Survival analyses were performed using the Kaplan–Meier method, and statistical significance between curves was determined by the log-rank test. All statistical analyses were conducted using GraphPad Prism software (version 9.0). A P value < 0.05 was considered statistically significant.

Results

Serum IL-32 levels positively correlate with NETs formation in HCC and HCC-PVTT patients, and IL-32 promotes neutrophil NETosis

It is reported that plasma levels of NETs were markedly elevated in patients with hepatocellular carcinoma (HCC) compared with healthy controls [15]. To determine whether circulating IL-32 is associated with systemic NET formation, we measured plasma concentrations of IL-32 and MPO–DNA complexes by ELISA in healthy donors, patients with HCC, and patients with HCC complicated by portal vein tumor thrombus (PVTT). Plasma levels of both IL-32 and MPO–DNA complexes were significantly higher in patients with HCC than in healthy donors and were further increased in patients with PVTT. Notably, both IL-32 and MPO–DNA levels exhibited a stepwise increase from healthy donors to patients with HCC and further to patients with PVTT (Fig. 1A-B). Correlation analysis revealed a significant positive association between plasma IL-32 and MPO–DNA concentrations in patients with HCC and in those with PVTT (Fig. 1C-D). These findings suggest that circulating IL-32 levels are associated with systemic NET formation in patients with HCC and PVTT.

Fig. 1.

Elevated plasma IL-32 levels correlate with NETs formation and promote NETosis in patients with HCC and HCC with PVTT. A-B Plasma levels of IL-32 and MPO–DNA complexes were measured by ELISA in healthy donors (n=22), patients with hepatocellular carcinoma (HCC) (n=26), and patients with HCC with portal vein tumor thrombus (PVTT) (n=22), with MPO–DNA complexes used to assess NETs formation. C Correlation analyses were performed between plasma IL-32 protein levels and MPO–DNA complexes in patients with HCC. D Correlation analyses were performed between plasma IL-32 protein levels and MPO–DNA complexes in patients with HCC with PVTT. E Neutrophils isolated from healthy donors were stimulated with plasma from healthy donors, patients with HCC, or patients with HCC with PVTT to assess NETs formation (n=3 per group). The percentage of NET-releasing cells was calculated as the ratio of NET-releasing neutrophils to the total number of neutrophils. H3Cit was used as a marker of NETs (green), myeloperoxidase (MPO) as a neutrophil marker (red), and nuclei were counterstained with DAPI (blue). The percentage of NET-releasing cells was calculated as the ratio of NET-releasing neutrophils to the total number of neutrophils. Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

To further determine whether the plasma from healthy donors, patients with HCC, and patients with PVTT differed in their capacity to induce NET formation, plasma samples from the three groups in three independent experiments. were collected and used to stimulate neutrophils isolated from healthy donors. Confocal immunofluorescence analysis showed that plasma from patients with HCC induced significantly greater NET formation than plasma from healthy donors, whereas plasma from patients with PVTT induced the most extensive NET formation, significantly exceeding that induced by plasma from patients with HCC (Fig. 1E). In contrast, plasma from healthy donors induced minimal NET formation. Together, these findings demonstrate that circulating IL-32 levels are positively associated with NET formation and that plasma from patients with PVTT exhibits the greatest NET-inducing capacity among the three groups.

IL-32 expression correlates with NETs formation in both HCC and PVTT tissues

Previous studies have demonstrated that neutrophils from HCC patients exhibit an enhanced capacity to release extracellular DNA and spontaneously form neutrophil extracellular traps (NETs), characterized by elevated H3Cit expression and extensive chromatin decondensation, compared with those from healthy donors [15]. These findings indicate that the tumor microenvironment of HCC is in a highly pro-NETotic state. Building upon these observations, we first examined whether NETs were differentially present in HCC tissues with high versus low IL-32 expression. Immunofluorescence staining revealed markedly stronger co-localized signals of myeloperoxidase (MPO) and citrullinated histone H3 (H3Cit) in IL-32–high HCC tissues compared with IL-32–low tumors, indicating enhanced NETs formation in IL-32–high HCC samples (Fig. 2A). Given the close association between IL-32 expression and NETs formation observed in HCC tissues, we next investigated whether NETs accumulation was similarly increased in portal vein tumor thrombus (PVTT) with high IL-32 expression. Immunofluorescence co-localization analysis of IL-32 with MPO/H3Cit demonstrated that PVTT lesions with high IL-32 expression exhibited significantly stronger MPO and H3Cit signals than IL-32–low PVTT lesions (Fig. 2B). Previous studies have demonstrated that IL-32 regulates hepatocellular carcinoma (HCC) cell proliferation and migration through multiple mechanisms [23]. To further investigate which cells IL-32 is derived from in our study, we performed immunofluorescence co-localization analysis of IL-32 with HepPar-1, a well-established marker of hepatocellular origin. Confocal imaging revealed clear co-localization of HepPar-1 and IL-32 within PVTT lesions (Fig. 2C), indicating that IL-32 is likely derived from HCC cells. Above all, these results indicate a strong association between elevated IL-32 expression and increased NETs formation in both HCC and PVTT tissues and suggest that HCC cells represent a major cellular source of IL-32.

Fig. 2.

IL-32–associated NETs formation in HCC and portal vein tumor thrombus tissues. A Representative immunofluorescence images showing differential NETs formation in IL-32–high and IL-32–low HCC tissues (n=3 per group). B Representative immunofluorescence images illustrating NETs formation in IL-32–high and IL-32–low portal vein tumor thrombus (PVTT) tissues (n=3 per group). C Representative immunofluorescence images showing the colocalization of HepPar-1 and IL-32 in PVTT tissues. IL-32 is shown in purple, citrullinated histone H3 (H3Cit) was used as a marker of NETs (green), myeloperoxidase (MPO) as a neutrophil marker (red), and nuclei were counterstained with DAPI (blue). Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

Neutrophils are enriched in the PVTT microenvironment revealed by mass cytometry

Immune cells play pivotal roles in the progression of inflammation and the remodeling of the tumor microenvironment [24]. To characterize the immune cell composition of hepatocellular carcinoma (HCC) and portal vein tumor thrombus (PVTT), we performed mass cytometry (CyTOF) profiling of adjacent non-tumor liver tissue, HCC, and PVTT tissues. Based on lineage-specific surface markers, 11 major immune cell populations were identified, and the expression patterns of 42 immune markers were profiled across these populations (Fig. 3A Supplementary Table 1). t-SNE analysis revealed distinct immune cell populations and demonstrated differences in their distribution among adjacent non-tumor liver tissue, HCC, and PVTT tissues. Notably, the immune cell composition of PVTT differed from that of both primary HCC and adjacent non-tumor liver tissues, with prominent alterations in neutrophil and macrophage populations (Fig. 3B-C). Quantitative analysis further showed that the relative abundance of neutrophils was significantly higher in PVTT tissues than in HCC and adjacent non-tumor liver tissues (Fig. 3D). These findings identify neutrophil enrichment as a prominent feature of the PVTT immune microenvironment and suggest that neutrophils may be involved in PVTT formation and progression.

Fig. 3.

Mass cytometry profiling of the immune cell composition of portal vein tumor thrombi. Mass cytometry (CyTOF) was performed on hepatocellular carcinoma (C), normal liver (N), and portal vein tumor thrombus (TT) tissues (n = 3 independent samples per group). A Heatmap showing the normalized expression of 38 immune markers across 11 immune cell clusters identified by CyTOF. B t-distributed stochastic neighbor embedding (t-SNE) visualization of the major immune cell populations, color-coded according to cell identity defined by lineage-specific markers. C Distribution of immune cell populations among C, N, and TT tissues. D Groupwise quantification of the relative abundance of each immune cell population. Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

IL-32 induces dose-dependent NET formation through PR3-dependent signaling in neutrophils

Given that plasma IL-32 levels were positively correlated with circulating MPO–DNA concentrations and that plasma from patients with HCC and PVTT exhibited strong NET-inducing activity, we hypothesized that IL-32 may contribute to NET formation in the HCC-PVTT microenvironment. To test this hypothesis, neutrophils isolated from healthy donors were stimulated with recombinant human IL-32 (rIL-32) at concentrations of 200, 600, 1000, and 1400 pg/mL in three independent experiments. Confocal immunofluorescence analysis demonstrated a concentration-dependent increase in NET formation with increasing concentrations of rIL-32 (P < 0.05 Fig. 4A). These findings indicate that IL-32 is sufficient to promote NET formation in human neutrophils in vitro.

Fig. 4.

Recombinant IL-32 induces NETs formation in neutrophils from healthy donors in a dose-dependent manner, which is significantly attenuated by PR3 blockade. A Neutrophils isolated from healthy donors were stimulated with recombinant IL-32 at concentrations of 200 pg/mL (A), 600 pg/mL (B), 1,000 pg/mL (C), and 1,400 pg/mL (D) (n=3 per group). B Neutrophils isolated from healthy donors were stimulated with phosphate-buffered saline (PBS) (A), 1,000 pg/mL recombinant IL-32 (B), 1,000 pg/mL recombinant IL-32 in the presence of a PR3-blocking antibody (C), phorbol 12-myristate 13-acetate (PMA) as a positive control (D) (n=3 per group). Citrullinated histone H3 (H3Cit) was used as a marker of NETs (green), myeloperoxidase (MPO) as a neutrophil marker (red), and nuclei were counterstained with DAPI (blue). Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

Previous studies have identified proteinase 3 (PR3) as a specific binding receptor for interleukin-32 (IL-32), mediating its biological activity independent of enzymatic function [10]. Based on this established interaction, we next investigated whether PR3 mediates IL-32-induced NET formation in neutrophils. Because rIL-32 at 1000 pg/mL induced robust NET formation in the dose-response experiment, this concentration was selected for subsequent mechanistic experiments. Neutrophils were pretreated with a neutralizing antibody against PR3 and subsequently stimulated with rIL-32 (1000 pg/mL). PBS-treated cells were used as the negative control, whereas PMA was used as a positive control for NET induction. Confocal immunofluorescence analysis showed that stimulation with rIL-32 (1000 pg/mL) markedly increased MPO and H3Cit-positive fluorescence intensity and area compared with PBS treatment, with levels comparable to those observed following PMA stimulation (Fig. 4B). Notably, pretreatment with the PR3-neutralizing antibody substantially attenuated the IL-32-induced increase in MPO and H3Cit-positive fluorescence intensity and area. These findings indicate that blockade of PR3 markedly reduces IL-32-induced NET formation, suggesting that PR3 is an important mediator of IL-32-induced NETosis in human neutrophils. Collectively, these results support a functional role for the IL-32/PR3 axis in promoting NET formation.

5.IL-32 and NETs promote PVTT formation and intrahepatic metastasis in vivo

We next sought to determine whether IL-32 and NETs contribute to the metastatic progression of HCC in vivo. A PVTT mouse model was established by portal vein injection of Hepa1-6 cells into C57BL/6 wild-type and PAD4⁻/⁻ mice (Fig. 5A). Bioluminescence imaging and hematoxylin and eosin (H&E) staining revealed that treatment with recombinant IL-32 markedly increased intrahepatic metastatic burden and promoted more extensive PVTT formation compared with PBS-treated controls (Fig. 5B), supporting a role for IL-32 in promoting HCC progression and metastatic dissemination. In contrast, administration of DNase I or genetic deficiency of PAD4 markedly reduced both the metastatic burden and PVTT size (Fig. 5B), indicating that disruption of NET formation or degradation of extracellular DNA attenuates tumor dissemination and PVTT development. Consistent with these findings, analysis of metastatic lesions on the liver surface showed that PAD4⁻/⁻ mice exhibited the fewest metastatic lesions, followed by DNase I–treated mice, whereas IL-32–treated mice displayed the most extensive metastatic dissemination (Fig. 5C). Collectively, these findings demonstrate that IL-32 promotes PVTT formation and intrahepatic and distant metastatic dissemination, at least in part through NET formation, in vivo.

Fig. 5.

In vivo validation of the roles of IL-32 and NETs in portal vein tumor thrombus formation and intrahepatic metastasis. A Schematic illustration of the portal vein injection method used to establish a mouse model of portal vein tumor thrombus (n=6 per group). B Representative in vivo bioluminescence imaging, gross liver specimens, and hematoxylin and eosin (H&E) staining of liver tissues. C Quantification of intrahepatic metastatic nodules in the four experimental groups.

Radiotherapy induces dose-dependent NETs formation in an orthotopic HCC PDX model

Previous studies in a genetically engineered mouse model of bladder cancer demonstrated that radiotherapy increases NETs formation in vivo [25]. Given the potential link between radiotherapy and neutrophil activation, we established an orthotopic patient-derived xenograft (PDX) model in NCG mice to assess whether irradiation enhances NETs formation in PVTT tissues. Mice were irradiated with single or fractionated doses (2 Gy, 2 × 6 Gy, or 12 Gy), and tumor samples were collected at 72 hours and 1 week post-irradiation (Fig. 6A). Immunofluorescence of NETs markers (MPO and H3Cit) in tumor sections revealed a pronounced increase in NETs expression with prolonged irradiation time and escalating radiation dose compared with non-irradiated controls (Fig. 6B). These observations demonstrate that radiotherapy robustly promotes NETs formation within the tumor microenvironment, which may contribute to radiation-induced microenvironment remodeling and the establishment of radioresistance.

Fig. 6.

Radiotherapy promotes NETs formation in tumor tissues in an orthotopic patient-derived xenograft model. A Schematic illustration of the establishment of an orthotopic patient-derived xenograft (PDX) model (n= 3 per group). B Representative confocal images showing NETs formation in tumor tissues from PDX mice following radiotherapy. Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

IL-32 and NETs expression are elevated in radioresistant HCC with PVTT tissues

In a murine bladder cancer model, elevated NETs levels contribute to the development of radioresistance [25]. Consistent with these findings, our in vivo experiments showed that radiotherapy induces a marked increase in NETs formation. To further examine whether NETs expression differs between radiosensitive and radioresistant tumors, clinical specimens were obtained from HCC patients with PVTT who underwent radiotherapy followed by surgical resection. Patients were categorized into radiosensitive and radioresistant groups according to the mRECIST. Representative cases illustrated that in patient 2, PVTT lesions exhibited minimal regression after radiotherapy compared with patient 1, indicating a radioresistant phenotype. Immunohistochemical staining of postoperative tumor tissues revealed markedly elevated IL-32 and H3Cit expression in radioresistant samples compared with radiosensitive ones (Fig. 7A). Consistent with our in vivo observations—where radiotherapy enhanced NETs formation and NETs promoted tumor progression—these clinical data support a strong association between NETs accumulation and radioresistance. Quantitative analysis confirmed significantly higher IL-32 and H3Cit expression in radioresistant versus radiosensitive tumors (Fig. 7B–D). Moreover, correlation analysis of post-radiotherapy plasma samples demonstrated a positive relationship between IL-32 and H3Cit levels (Fig. 7E). Together, these data indicate that elevated IL-32 expression and enhanced NETs formation are closely associated with the development of radioresistance in HCC with PVTT.

Fig. 7.

Elevated IL-32 and NET formation in radioresistant hepatocellular carcinoma. A Representative post-radiotherapy computed tomography (CT) images from a radiosensitive HCC patient (Patient 1) and a radioresistant HCC patient (Patient 2), showing the radiological response to radiotherapy. Representative immunohistochemical staining of IL-32 and citrullinated histone H3 (H3Cit) in tumor tissues from radiosensitive and radioresistant HCC patients (n=3 per group). B Correlation analysis between IL-32 and H3Cit expression in radiosensitive and radioresistant HCC tissues (n=3 per group). C Plasma IL-32 levels in HCC patients before and after radiotherapy (n=10 per group). D Plasma MPO–DNA levels in HCC patients before and after radiotherapy (n=10 per group). E Correlation analysis between plasma IL-32 and MPO–DNA levels in HCC patients before and after radiotherapy. Data are presented as means ± SD. ns, not significant; *P < 0.05; ***P < 0.001.

Discussion

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer‐related mortality worldwide, and many patients are diagnosed at advanced stages when curative surgical options are no longer feasible [19,26]. Intrahepatic hematogenous dissemination represents the predominant route of HCC progression; therefore, preventing intrahepatic spread is crucial for improving clinical outcomes [2]. IL-32 has been implicated in various chronic inflammatory diseases and is involved in the transition from chronic hepatitis to HCC [8]. IL-32α has been reported to promote hematogenous metastasis of hepatocellular carcinoma cells via activation of the VEGF–STAT3 signaling pathway [23]. Earlier studies reported that IL-32 can bind to proteinase-3 (PR3) on neutrophils, and PR3-mediated proteolytic processing enhances IL-32 bioactivity [10]. Nevertheless, it remained unclear whether IL-32 is capable of activating neutrophils to induce NETosis. In our study, serum samples from patients with HCC and HCC with PVTT demonstrated a significant positive correlation between IL-32 and NETs levels. When neutrophils from healthy donors were stimulated with patient serum, NETs formation increased markedly, with PVTT serum inducing higher NETs levels than serum from patients with HCC alone. Recombinant IL-32 stimulation produced a similar dose-dependent increase in NETosis, further supporting a direct role for IL-32 in promoting NETs formation. Importantly, blockade of PR3 significantly reduced NET formation, supporting PR3 as a functional mediator of IL-32-associated NETosis. Although NETs are traditionally viewed as host defense structures released in response to microbial stimulation, our findings suggest that, within the inflammatory milieu of HCC, IL-32 may promote pathological NET formation, with PR3 contributing to this process. These observations refine the mechanistic understanding of NETosis in tumor-associated inflammation and provide new insights into how IL-32–mediated NETs formation may contribute to HCC progression and PVTT development.

NETs play a critical role in the development of deep vein thrombosis, where they not only promote thrombus formation but may also function as a third structural scaffold in addition to fibrin and von Willebrand factor [27]. Given this pronounced pro-thrombotic capacity and the propensity of hepatocellular carcinoma (HCC) for intrahepatic hematogenous dissemination, it is plausible that NETs may similarly contribute to the formation of portal vein tumor thrombus (PVTT) in HCC patients. In our study, functional evidence from mouse models supports this hypothesis. Both PAD4-deficient mice, which have impaired NETs formation, and mice treated with DNase I, which enzymatically degrades NETs, exhibited markedly reduced intrahepatic distant metastasis. These findings indicate that disrupting NETs formation can significantly limit HCC progression and suggest that targeting NETs may represent a promising therapeutic strategy for managing advanced HCC.

Radiotherapy is widely used as a major therapeutic modality for patients with advanced malignancies [18]. However, a subset of patients exhibits poor responsiveness and eventually develops radioresistance. Previous studies in bladder cancer models have shown that radiotherapy induces a marked increase in NETs, and this accumulation contributes to reduced sensitivity to radiation [25]. Our findings extend these observations to HCC, demonstrating that NETs formation progressively increases with prolonged radiation exposure and higher radiation doses in a murine HCC model, indicating a close association between NETs and radiotherapy response. To further explore the clinical relevance of this association, we examined postoperative tumor specimens from radiosensitive and radioresistant HCC patients. Immunohistochemical analysis revealed significantly higher expression of IL-32 and NETs markers in radioresistant tumors compared with radiosensitive ones. These results suggest that aberrant NETs accumulation may contribute to the development of radioresistance in HCC. Accordingly, targeting NETs formation may represent a promising therapeutic strategy for overcoming radioresistance and improving radiotherapy outcomes in HCC patients.

Although this study supports a role for the IL-32–PR3–NETs axis in HCC and PVTT progression and radioresistance, several limitations should be acknowledged. First, the limited number of radiosensitive and radioresistant patients warrants validation in larger independent cohorts. Second, the mechanisms linking NETs to radioresistance remain unclear. Finally, whether radiotherapy-induced NET formation is mediated through the IL-32–PR3 axis remains to be determined. Further studies using larger cohorts and specific genetic or pharmacological models are needed to clarify the mechanistic and therapeutic relevance of this axis.

In summary, our findings identify the tumor-derived IL-32/PR3/NETs axis as a contributor to portal vein tumor thrombus formation and suggest its involvement in radiotherapy resistance in hepatocellular carcinoma. These findings highlight IL-32–driven NET formation as a potential therapeutic vulnerability and support further investigation of strategies targeting IL-32 signaling or NETs, particularly in combination with radiotherapy. IL-32 may also serve as a biomarker of radiotherapy response, with potential implications for treatment stratification in patients with HCC and PVTT.

Ethics approval and consent to participate

Human studies were conducted in accordance with the Declaration of Helsinki and its later amendments. The study protocol was approved by the Ethics Committee of Qingdao Municipal Hospital. Written informed consent was obtained from all participants prior to inclusion in the study, and consent for publication was obtained from all participants.

All animal experiments were conducted in accordance with the ARRIVE guidelines and relevant institutional and national guidelines for the care and use of laboratory animals. The study protocol was approved by the Animal Ethics Committee of the Institutional Animal Care and Use Committee of Shandong University, Qingdao Campus.

Consent for publication

Not applicable.

Funding

This study was funded by the Natural Science Foundation of Shandong Province (ZR2025MS1363).

Data availability

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

CRediT authorship contribution statement

Liming Yao: Visualization, Validation. Ying Pan: Data curation. Guangjun Shi: Writing – review & editing. Pengcheng Kang: Data curation, Writing – review & editing. Shaohai Luan: Writing – original draft. Kaiming Leng: Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.