SLPI-Loaded Liposomes Targeting Kupffer Cells Modulate Macrophage Polarization and Mitigate Radiation-Induced Liver Damage

Abstract

Kupffer cells (KCs) make up the predominant population of resident innate immune cells in the liver, serving as key immune sentinels that maintain local immune surveillance and immunoregulatory homeostasis. However, their functional involvement and phenotypic dynamics during radiation-induced liver damage (RILD) remain insufficiently explored. Therefore, we established a mouse model of RILD and, through systematic single-cell-level profiling of hepatic immune cell populations, found that KCs play a critical role in hepatic immune responses and undergo a pronounced radiation-induced shift toward a pro-inflammatory M1 phenotype. Further KC depletion/reconstitution, molecular assays, and coculture experiments consistently demonstrated that M1-polarized KCs exacerbate liver damage, with secretory leukocyte protease inhibitor (SLPI) being identified as a key molecular mediator driving this polarization and its pathogenic effects. To further substantiate these findings, we designed a liposome-based delivery strategy to selectively inhibit SLPI in KCs, which effectively suppressed M1 polarization and alleviated radiation-induced liver damage, underscoring the therapeutic relevance and translational potential of this approach in RILD. Overall, these findings demonstrate that radiation drives KCs toward an SLPI-dependent pro-inflammatory M1 state, thereby exacerbating liver injury. Moreover, targeted liposomal suppression of SLPI effectively reverses this polarization and protects against RILD, highlighting SLPI-modulated KC reprogramming as a promising therapeutic approach.

1. Introduction

The liver is highly sensitive to ionizing radiation [1] and is prone to both acute and chronic radiation damage under various circumstances [2], including radiation therapy (RT) [3], diagnostic imaging, experimental studies, occupational exposure in nuclear-related professions, and accidental radiation incidents [4]. In patients undergoing RT for hepatocellular carcinoma or other malignancies in the upper abdomen and lower thoracic region, radiation-induced liver damage (RILD) has emerged as a common treatment-associated complication. Although radiotherapy techniques have evolved from conventional two-dimensional radiotherapy (2D-RT) to advanced modalities such as three-dimensional conformal radiotherapy (3D-CRT), intensity-modulated radiotherapy (IMRT), stereotactic body radiotherapy (SBRT), volumetric modulated arc therapy (VMAT), and particle therapy with protons or heavy ions (PHIRT) [5], the risk of hepatic radiation toxicity persists due to the inevitable exposure of normal liver tissue. Recent clinical studies have underscored the clinical relevance of RILD. For instance, Li et al. reported that the incidence of non-classic RILD reached 22.7% within three months following IMRT in patients with locally advanced hepatocellular carcinoma [6]. Furthermore, Du et al. identified a tumor number ≥ 2, a mean liver dose (MLD) ≥ 1371.4 cGy, and a normal liver volume (NLV) ≤ 700 mL as independent risk factors for non-classic RILD (ncRILD) [7]. Collectively, these findings emphasize the urgency of elucidating the mechanistic underpinnings of RILD and developing effective therapeutic strategies.

To date, multiple hypotheses have been proposed to elucidate the pathogenesis of RILD, primarily encompassing oxidative stress [8,9,10], DNA damage [11,12], dysregulated lipid metabolism [13,14,15], and inflammatory responses. Ionizing radiation significantly enhances the production of reactive oxygen species (ROS) and reactive nitrogen species (RNS) in hepatocytes, which leads to oxidative stress, secondary inflammation, and subsequent structural and functional damage to hepatic cells [16]. In parallel, radiation-induced damage to both nuclear and mitochondrial DNA activates the cyclic GMP-AMP synthase–stimulator of interferon genes (cGAS–STING) signaling pathway, which exacerbates cellular stress responses and promotes senescence, ultimately contributing to the onset and progression of RILD [17]. Radiation exposure can induce hepatic steatosis and triglyceride accumulation. At later stages, it may trigger a reprogramming of lipid metabolic pathways via activation of peroxisome proliferator-activated receptor alpha (PPARα), thereby altering hepatic lipid homeostasis [13]. In addition, radiation exposure upregulates Toll-like receptor 4 (TLR4) signaling and downstream nuclear factor-κB (NF-κB) activation, thereby enhancing the production of pro-inflammatory cytokines and chemokines [18,19,20,21].

Kupffer cells (KCs), the predominant population of liver-resident macrophages, are rapidly activated upon exposure to ionizing radiation and contribute to hepatic injury through inflammatory cascades [22,23]. Moreover, KCs secrete transforming growth factor-beta 1 (TGF-β1), which activates hepatic stellate cells (HSCs) via c-Jun N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK) pathways, promoting extracellular matrix deposition and accelerating fibrotic progression in RILD [22]. Despite recognition of their importance, the molecular mechanisms underlying KC activation and functional reprogramming following irradiation remain poorly defined. Elucidating the dynamic alterations in KCs’ phenotypes and their regulatory signaling networks under radiation stress is of great significance for understanding the immunopathology of RILD and developing targeted therapeutic strategies.

Accumulating evidence indicates that KCs play critical roles in shaping the hepatic immune microenvironment during RILD. Radiation exposure profoundly remodels hepatic immune signaling; however, the molecular programs governing KC polarization remain incompletely understood. Notably, secretory leukocyte protease inhibitor (SLPI) expression has been reported to be markedly elevated in pro-inflammatory M1-polarized KCs in murine models of RILD, suggesting a potential link between SLPI signaling and macrophage-driven inflammatory responses. However, whether SLPI directly regulates KC polarization and thereby contributes to the development and progression of RILD is not known. Therefore, this study aims to determine whether SLPI is aberrantly upregulated in KCs in a murine RILD model and to assess its contribution to RILD progression through the promotion of M1-type polarization. Moreover, we seek to elucidate the molecular mechanisms by which SLPI drives radiation-induced functional reprogramming of KCs and to establish a Kupffer cell-targeted liposome-based delivery platform that enables selective modulation of SLPI activity in vivo. By integrating mechanistic investigation with a targeted nanodelivery strategy, this study further evaluates the therapeutic feasibility and translational relevance of liposomal SLPI regulation for mitigating RILD.

2. Results

2.1. Single-Cell RNA-Seq Analysis Reveals Increased Proportion of M1-Polarized KCs in RILD

Compared with the control group (ALT = 33.77 U/L, AST = 158.8 U/L), the 1-day post-irradiation group showed elevated serum ALT (67.69 U/L) and AST (232.3 U/L) levels, which are indicative of RILD. By 14 days post-irradiation, ALT and AST levels further increased to 202.1 U/L and 310.5 U/L, respectively, suggesting progressive exacerbation of liver injury (Figure 1A). Histopathological examination revealed distinct temporal features of radiation-induced hepatic injury. On day 1 post-irradiation, liver sections exhibited acute injury characterized by marked sinusoidal dilation, hepatocellular swelling, and nuclear pyknosis, which are indicative of early radiation-induced cytotoxicity. By day 14, more pronounced structural alterations were observed, including hepatocellular atrophy, cytoplasmic condensation, and focal cellular fragmentation. Collectively, these histomorphological findings confirm the successful establishment of the RILD mouse model (Figure 1B). Moreover, 10×Genomics single-cell RNA-sequencing with pseudotime analysis revealed a shift of KCs toward an M1-like pro-inflammatory phenotype. Gene dynamics analysis further identified two keys differentially expressed genes potentially involved in the functional reprogramming of KCs during RILD: HMOX1 (heme oxygenase 1) and SLPI (Figure 1C–E) [24]. Flow cytometry analysis of primary liver macrophages confirmed a significant increase in the proportion of M1-polarized KCs after 30 Gy X-ray exposure (Figure 1F). These results collectively demonstrate an elevated proportion of M1-polarized KCs during RILD progression.

2.2. Radiation-Induced M1-Type Reprogramming of KCs Contributes to the Regulation of RILD

To investigate the functional role of KCs in RILD, a KC-depletion mouse model was established using gadolinium chloride (GdCl₃). Mice were divided into four groups: PBS, GdCl₃, IR + PBS, and IR + GdCl₃. HE staining at 24 h post-30 Gy X-ray whole-liver irradiation revealed intact hepatic architecture in the PBS group. Mild hepatocyte separation and slight cellular disorganization were observed in the GdCl₃ group. In contrast, the IR + PBS group exhibited hepatocellular atrophy, sinusoidal dilation, and cytoplasmic vacuolization—histopathological features consistent with radiation-induced liver injury. Notably, the IR + GdCl₃ group showed relatively preserved hepatic architecture with markedly attenuated histopathological damage (Figure 2A). Immunohistochemical analysis confirmed a marked reduction in KC following GdCl₃ administration (Figure S1A). qPCR analysis demonstrated that pro-inflammatory cytokine transcripts, including TNF-α, IL-6, and IL-1β, were significantly elevated in the IR + PBS group compared to the PBS group, whereas their levels were markedly decreased in the IR + GdCl₃ group (Figure S1C). Serum levels of ALT and AST followed similar trends (Figure 2C). Western blot analysis further validated that protein levels of pro-inflammatory cytokines and cleaved caspase-3 were elevated in the IR + PBS group but substantially downregulated in the IR + GdCl₃ group (Figure 2E). These results collectively indicate that KCs are critically involved in the development of RILD, largely via their pro-inflammatory cytokine secretion. To further dissect the subtype-specific responses of KCs to radiation, multicolor immunofluorescence staining was performed using markers for M1 macrophages (iNOS), M2 macrophages (CD206), DNA damage marker γH2A.X, and DAPI. Quantitative analysis showed no significant differences in γH2A.X intensity among M2-type KCs across groups (p = 0.9316, p = 0.8054, and p = 0.9014), whereas M1-type KCs exhibited more pronounced radiation-induced damage (Figure S1E).

To directly evaluate the role of M1-type KCs in RILD pathogenesis, a model of M1 macrophage adoptive transfer was constructed via tail vein injection. Mice were divided into four groups: IR + PBS, IR + GdCl₃, IR + M1, and IR + GdCl₃ + M1. HE staining demonstrated aggravated hepatocellular injury in the IR + M1 group compared with the IR + PBS group, characterized by pronounced hepatocellular atrophy and marked sinusoidal dilation. Conversely, the IR + GdCl₃ + M1 group exhibited partially preserved hepatic architecture with alleviated histopathological damage, although mild architectural disorganization remained evident (Figure 2B). Immunohistochemistry confirmed decreased M1 macrophage infiltration in the IR + GdCl₃ group, while the IR + GdCl₃ + M1 group showed a moderate increase compared to IR + GdCl₃, yet remained lower than the IR + M1 group (Figure S1B). Serum biochemical analysis revealed that ALT and AST levels were significantly elevated in the IR + M1 group compared to IR + PBS (Figure 2D), indicating exacerbated liver dysfunction. Correspondingly, mRNA and protein levels of TNF-α, IL-6, and IL-1β were significantly upregulated in the IR + M1 group (Figure S1D and Figure 2F). Collectively, these findings demonstrate that M1-polarized KCs represent a key effector subset that exacerbates inflammation and promotes hepatic injury during the progression of RILD.

Figure 2. Radiation-induced reprogramming of M1-type KCs exacerbates liver injury in RILD. (A) Representative H&E staining of liver tissues from each group (n = 5). Tissue sections were fixed, paraffin-embedded, and examined under brightfield illumination using a fluorescence microscope system. Blue arrow indicates hepatocytes with normal morphology. Gray arrows indicate shrunken hepatocytes; green arrows denote disorganized and condensed hepatocytes; black arrows highlight disrupted cytoplasm and dilated hepatic sinusoids. Scale bar: 100 μm. (B) H&E staining of liver tissues from mice in the M1 macrophage adoptive transfer mode (n = 5). Morphological changes were visualized as in (A). Scale bar: 100 μm. (C) Serum ALT and AST levels measured 24 h after 30 Gy whole-liver X-ray irradiation (n = 5 per group). Radiation significantly increased ALT levels compared to PBS controls (* p < 0.05), while KC depletion reduced ALT levels compared to the IR + PBS group (** p < 0.01). Similarly, AST levels were elevated post-irradiation (*** p < 0.001) and attenuated upon KC depletion (*** p < 0.001). (D) Serum ALT and AST levels following adoptive transfer of M1 macrophages (n = 5 per group). M1 macrophage infusion exacerbated liver injury, with ALT significantly increased compared to IR + PBS (* p < 0.05) and decreased after KC depletion (** p < 0.01). AST levels showed a similar trend, with significant elevation in the IR + M1 group (** p < 0.01) and marked reduction following depletion (**** p < 0.0001). (E) Western blot analysis of liver tissues (n = 3) showing expression levels of TNF-α, IL-6, IL-1β, and cleaved caspase-3 in KC-depleted mice. Quantification using ImageJ indicated significant increases in inflammatory proteins in the IR + PBS group compared to the PBS group (** p < 0.01 and *** p < 0.001) and significant reductions following KC depletion (* p < 0.05). (F) Western blot analysis of inflammatory cytokines and cleaved caspase-3 in the M1 macrophage adoptive transfer mode (n = 3). Compared to IR + PBS, the IR + M1 group showed significantly higher protein expression (* p < 0.05, ** p < 0.01, and *** p < 0.001). KC depletion mitigated these effects, with markedly reduced protein levels compared to the IR + M1 group (** p < 0.01, *** p < 0.001, and **** p < 0.0001).

Figure 2. Radiation-induced reprogramming of M1-type KCs exacerbates liver injury in RILD. (A) Representative H&E staining of liver tissues from each group (n = 5). Tissue sections were fixed, paraffin-embedded, and examined under brightfield illumination using a fluorescence microscope system. Blue arrow indicates hepatocytes with normal morphology. Gray arrows indicate shrunken hepatocytes; green arrows denote disorganized and condensed hepatocytes; black arrows highlight disrupted cytoplasm and dilated hepatic sinusoids. Scale bar: 100 μm. (B) H&E staining of liver tissues from mice in the M1 macrophage adoptive transfer mode (n = 5). Morphological changes were visualized as in (A). Scale bar: 100 μm. (C) Serum ALT and AST levels measured 24 h after 30 Gy whole-liver X-ray irradiation (n = 5 per group). Radiation significantly increased ALT levels compared to PBS controls (* p < 0.05), while KC depletion reduced ALT levels compared to the IR + PBS group (** p < 0.01). Similarly, AST levels were elevated post-irradiation (*** p < 0.001) and attenuated upon KC depletion (*** p < 0.001). (D) Serum ALT and AST levels following adoptive transfer of M1 macrophages (n = 5 per group). M1 macrophage infusion exacerbated liver injury, with ALT significantly increased compared to IR + PBS (* p < 0.05) and decreased after KC depletion (** p < 0.01). AST levels showed a similar trend, with significant elevation in the IR + M1 group (** p < 0.01) and marked reduction following depletion (**** p < 0.0001). (E) Western blot analysis of liver tissues (n = 3) showing expression levels of TNF-α, IL-6, IL-1β, and cleaved caspase-3 in KC-depleted mice. Quantification using ImageJ indicated significant increases in inflammatory proteins in the IR + PBS group compared to the PBS group (** p < 0.01 and *** p < 0.001) and significant reductions following KC depletion (* p < 0.05). (F) Western blot analysis of inflammatory cytokines and cleaved caspase-3 in the M1 macrophage adoptive transfer mode (n = 3). Compared to IR + PBS, the IR + M1 group showed significantly higher protein expression (* p < 0.05, ** p < 0.01, and *** p < 0.001). KC depletion mitigated these effects, with markedly reduced protein levels compared to the IR + M1 group (** p < 0.01, *** p < 0.001, and **** p < 0.0001).

2.3. SLPI-Mediated Reprogramming of KCs

To further investigate the effects of different doses of ionizing radiation on macrophage polarization and SLPI expression, flow cytometry was performed to analyze the M1/M2 polarization ratios of Raw264.7 cells following X-ray exposure at doses of 0, 5, 10, and 20 Gy. These results were then compared with those obtained for F4/80⁺ KCs and F4/80⁺CD11b⁺ monocyte-derived macrophages (MDMs). As shown in Figure 3A, the M1/M2 ratio of KCs increased progressively from 0 to 10 Gy (p = 0.004), while a decline was observed from 10 to 20 Gy (p = 0.0001). In contrast, 10 Gy irradiation significantly elevated the M1/M2 ratio in MDMs (p = 0.0019). qRT-PCR analysis revealed that the mRNA expression of SLPI, similar to HMOX1 in Raw264.7 cells under different radiation doses and in liver tissues at various time points post-30 Gy whole-liver irradiation, followed a pattern consistent with the changes in the M1/M2 ratio (Figure 3B,C). To further explore and confirm the regulatory role of SLPI, stable SLPI-overexpressing and SLPI-knockdown Raw264.7 cell lines were established (Figure S2A,B). Western blot analysis demonstrated that SLPI knockdown significantly suppressed the expression of M1-associated pro-inflammatory cytokines TNF-α, IL-6, and IL-1β after radiation exposure (Figure 3D), whereas SLPI overexpression upregulated the mRNA levels of M1 signature markers (Figure S2C). Additionally, SLPI expression was markedly higher in M1-polarized macrophages than in M2 cells, as confirmed by both immunoblotting and qRT-PCR (Figure S2D). To assess the radiosensitivity of different macrophage subtypes, apoptosis rates of M1 and M2 macrophages were analyzed by flow cytometry following exposure to varying doses of X-rays. The results indicated no significant differences in apoptosis between M2 macrophages across groups compared to their M1 counterparts (p = 0.2044, p = 0.3900, p = 0.6150, and p = 0.2555) (Figure S2E). Collectively, these findings suggest that SLPI facilitates M1 polarization and enhances the pro-inflammatory phenotype of macrophages, implicating its potential role in the pathogenesis of RILD.

2.4. SLPI-Mediated Reprogramming of KCs Exacerbates RILD

To investigate the role of SLPI-mediated macrophage polarization in hepatocyte radiation injury, a co-culture system of primary mouse hepatocytes and macrophages was established (Figure 4A and Figure S3A). Compared with irradiated hepatocytes cultured alone (5 Gy or 10 Gy), co-culture with irradiated Raw264.7 cells increased hepatocyte apoptosis by 3.267% ± 1.397% and 7.000% ± 0.6968%, respectively, accompanied by significantly elevated secretion of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α (Figure 4B,C). In subsequent experiments, SLPI-silenced Raw264.7 cells (transfected with siSLPI) were co-cultured with irradiated hepatocytes. The results demonstrated that hepatocyte apoptosis decreased by 15.17% ± 4.787% (5 Gy) and 15.31% ± 3.416% (10 Gy), with a concomitant reduction in the levels of IL-1β, IL-6, and TNF-α (Figure 4D,E). Similarly, co-culture with irradiated primary KCs, as compared to irradiated hepatocytes cultured alone, markedly increased hepatocyte apoptosis by 9.400% ± 0.8380% and 6.967% ± 1.518%, along with a significant rise in inflammatory cytokine levels (Figure S3B,C). These results suggest that SLPI promotes the polarization of KCs toward the M1 phenotype, enhances their pro-inflammatory profile, and aggravates radiation-induced hepatocyte injury. Collectively, SLPI functions as a critical regulatory factor in the pathogenesis of RILD.

2.5. Targeted Intervention of KC Reprogramming to Ameliorate RILD

To evaluate the therapeutic potential of targeting SLPI-mediated KC reprogramming in RILD, liposomes encapsulating siSLPI were synthesized using microfluidic technology. Transmission electron microscopy (TF20) revealed uniform spherical structures (Figure 5A). The average particle size of blank liposomes was 79 nm, while siSLPI-loaded liposomes measured 78 nm with an encapsulation efficiency of 90% (Figure 5B and Figure S4A), and the excitation spectrum of DiD-labeled liposomes confirmed successful fluorescent tagging (Figure 5C). Cellular uptake experiments showed that liposomes began entering Raw264.7 cells within 0.5 h and achieved complete internalization by 1 h (Figure 5D). In vivo imaging demonstrated hepatic enrichment of liposomes within 1 h post-injection, with peak accumulation at 4 h and subsequent renal clearance after 8 h (Figure 5E). CCK-8 assays confirmed that liposomes did not significantly affect cell viability (Figure 5F), and H&E staining showed no adverse histological changes in major organs, indicating good biocompatibility (Figure 5G).

To validate the establishment of the in vivo SLPI knockdown models, SLPI expression in mouse liver tissues was assessed by Western blotting and qPCR following intravenous administration of siSLPI-liposomes prior to whole-liver irradiation (30 Gy). (Figure S4B,C). At 24 h post-irradiation, H&E staining revealed that mice in the IR group exhibited histopathological features consistent with radiation-induced liver injury, including hepatocellular atrophy, sinusoidal dilation, and cytoplasmic vacuolization. In contrast, the IR + siSLPI group demonstrated partial preservation of hepatic architecture with alleviated hepatocellular damage, although mild disorganization of hepatic cords remained evident (Figure 5H). IHC and ELISA confirmed significantly reduced hepatic levels of TNF-α, IL-6, and IL-1β in the IR + siSLPI group (p = 0.0131; 0.0076; 0.0101, respectively) (Figure 5I,J and Figure S5A). Serum biochemical analysis showed marked reductions in ALT and AST levels (ALT: 45.21 U/L; AST: 213.4 U/L) compared to the IR group, indicating alleviation of liver damage (Figure S4D). In addition, AAV8-shSLPI vectors were utilized to construct a second KC-specific SLPI knockdown model in vivo, which was similarly confirmed by hepatic Western blot and qPCR analyses (Figure S4E,F). Under the same irradiation protocol, H&E staining showed that IR + shSLPI mice exhibited improved hepatic architecture compared to the IR group (Figure S4G). IHC and ELISA results demonstrated significantly decreased cytokine levels in the IR + shSLPI group (IL-1β, IL-6, and TNF-α: p = 0.0082, 0.0113, and 0.0047, respectively) (Figures S4H, S5B and S4I), accompanied by reduced ALT (22.12 U/L) and AST (122.5 U/L) levels (Figure S4J). Collectively, these findings demonstrate that targeted inhibition of SLPI in KCs effectively attenuates radiation-induced hepatic inflammation and injury, offering a promising therapeutic strategy for RILD.

Figure 5. Therapeutic strategy of targeting SLPI-mediated KC reprogramming to alleviate RILD. (A) Transmission electron microscopy (TEM) image of liposomes stained with phosphotungstic acid, revealing uniform spherical morphology. Scale bar: 50 nm. (B) Dynamic light scattering analysis showing the particle size of blank liposomes (left, 79 nm) and siSLPI-loaded liposomes (right, 78 nm). (C) Excitation spectrum of DiD-labeled liposomes. (D) Time-dependent intracellular uptake of liposomes by Raw264.7 cells. Nuclei were stained with Hoechst (blue), liposomes were labeled with DiD (red), and cytoplasm was stained with Calcein AM (green). Scale bar = 50 μm. (E) In vivo fluorescence imaging showing hepatic accumulation of liposomes at 1 h post-injection (* p < 0.05 and ** p < 0.01 vs. 1 h). (F) CCK-8 assay evaluating the effect of liposomes on Raw264.7 cell proliferation (** p < 0.01, *** p < 0.001, and **** p < 0.0001 vs. control). (G) H&E staining of heart, liver, spleen, lung, and kidney tissues from treated mice (n = 5), showing no histopathological abnormalities. Scale bar: 100 μm. (H) H&E staining of liver tissues from different treatment groups (n = 5). Blue arrows: normal hepatocytes; gray arrows: shrunken hepatocytes; green arrows: intact hepatocytes; black arrows: cytoplasmic disruption and sinusoidal dilation. Scale bar: 100 μm. (I) Immunohistochemistry analysis of hepatic IL-1β, IL-6, and TNF-α expression in RILD mice treated with siSLPI-loaded liposomes. Scale bar: 100 μm. (J) ELISA quantification of hepatic IL-1β, IL-6, and TNF-α levels in RILD mice treated with siSLPI-liposomes (* p < 0.05, ** p < 0.01, and **** p < 0.0001 vs. NC group).

Figure 5. Therapeutic strategy of targeting SLPI-mediated KC reprogramming to alleviate RILD. (A) Transmission electron microscopy (TEM) image of liposomes stained with phosphotungstic acid, revealing uniform spherical morphology. Scale bar: 50 nm. (B) Dynamic light scattering analysis showing the particle size of blank liposomes (left, 79 nm) and siSLPI-loaded liposomes (right, 78 nm). (C) Excitation spectrum of DiD-labeled liposomes. (D) Time-dependent intracellular uptake of liposomes by Raw264.7 cells. Nuclei were stained with Hoechst (blue), liposomes were labeled with DiD (red), and cytoplasm was stained with Calcein AM (green). Scale bar = 50 μm. (E) In vivo fluorescence imaging showing hepatic accumulation of liposomes at 1 h post-injection (* p < 0.05 and ** p < 0.01 vs. 1 h). (F) CCK-8 assay evaluating the effect of liposomes on Raw264.7 cell proliferation (** p < 0.01, *** p < 0.001, and **** p < 0.0001 vs. control). (G) H&E staining of heart, liver, spleen, lung, and kidney tissues from treated mice (n = 5), showing no histopathological abnormalities. Scale bar: 100 μm. (H) H&E staining of liver tissues from different treatment groups (n = 5). Blue arrows: normal hepatocytes; gray arrows: shrunken hepatocytes; green arrows: intact hepatocytes; black arrows: cytoplasmic disruption and sinusoidal dilation. Scale bar: 100 μm. (I) Immunohistochemistry analysis of hepatic IL-1β, IL-6, and TNF-α expression in RILD mice treated with siSLPI-loaded liposomes. Scale bar: 100 μm. (J) ELISA quantification of hepatic IL-1β, IL-6, and TNF-α levels in RILD mice treated with siSLPI-liposomes (* p < 0.05, ** p < 0.01, and **** p < 0.0001 vs. NC group).

3. Discussion

RILD is a consequence of complex and dynamic interactions among various hepatic cell types. Following radiation exposure, radiosensitive hepatic cell populations—including KCs, HSCs, and sinusoidal endothelial cells (SECs)—rapidly secrete pro-inflammatory cytokines, triggering fibrotic remodeling, disrupting hepatic microcirculation, and activating hypoxia-related signaling, all of which culminate in progressive liver injury. However, traditional radiobiological approaches often lack the resolution to capture dynamic changes in cellular function and intercellular communication in real time, thus limiting our understanding of RILD pathogenesis at the single-cell level. Single-cell transcriptomic profiling has enabled deeper insight into these processes. In our previous scRNA-seq study, we delineated irradiation-induced transcriptional remodeling across hepatic cell populations and identified KCs as a key cell type undergoing substantial phenotypic and functional alterations during RILD progression [24]. While phenotypic and functional dysregulation of KCs has long been implicated in the pathogenesis of RILD, the specific phenotypic transitions and molecular mechanisms underlying these changes remain poorly understood. In the present study, we demonstrate that radiation significantly increases the proportion of M1-polarized KCs in the liver and markedly upregulates the expression of SLPI in KCs. These findings suggest that SLPI may play a pivotal role in modulating macrophage polarization in response to radiation stress. This provides novel insights into the mechanistic basis of RILD and highlights SLPI as a promising therapeutic target for intervention in radiation-induced liver injury.

KCs, the predominant immune cells in the liver, account for approximately 35% of all non-parenchymal cells and 80–90% of the total hepatic macrophage population [25]. The liver is widely recognized as an “innate immunity-dominant organ [26],” owing to the abundance of immune cells, particularly KCs, which are predominantly localized within the hepatic sinusoids and intravascular spaces of sinusoidal endothelial cells [27]. In addition to KCs, the liver sinusoids also harbor other innate immune cell subsets, including natural killer (NK) cells, natural killer T (NKT) cells, and dendritic cells (DCs) [28], collectively forming a robust immunological barrier. The intimate interactions between KCs and both parenchymal and non-parenchymal cells provide frontline defense against exogenous insults. KCs play central roles in immune surveillance and homeostasis. They phagocytose apoptotic debris and metabolic waste, mediate erythrophagocytosis and iron recycling, regulate cholesterol trafficking through cholesteryl ester transfer protein (CETP) secretion, eliminate blood-borne pathogens, and maintain hepatic immune tolerance through antigen presentation [29,30,31,32,33,34,35]. Beyond physiological roles, recent studies highlight their pathological relevance. KCs can promote hepatic lipid accumulation via apolipoprotein secretion and metabolic reprogramming, while transcriptional regulators such as factor-1α (HIF-1α) and CCAAT/enhancer-binding protein β (C/EBPβ) modulate inflammatory and metabolic responses in steatotic liver disease [36,37]. Recent studies have further demonstrated that ionizing radiation elevates N6-methyladenosine (m6A) RNA modifications in KCs. Specifically, the methyltransferase METTL3 enhances m6A activity and activates the TEAD1–STING–NLRP3 signaling cascade, thereby amplifying immune–inflammatory responses during RILD. This highlights the crucial role of KCs in mediating radiation-induced hepatic inflammation [38]. To functionally validate the pathogenic contribution of KCs in RILD, we established a KC-depletion model using GdCl₃ in mice, followed by 30 Gy X-ray whole-liver precision irradiation. Our results showed that radiation markedly elevated serum ALT and AST levels, induced significant histopathological changes in liver tissue, and upregulated the expression of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β). In contrast, these injury markers were significantly alleviated in GdCl₃-treated mice, underscoring the pivotal role of KCs in RILD pathogenesis. Furthermore, to investigate the specific contribution of M1-polarized macrophages, we constructed a mouse model via adoptive transfer of in vitro-induced M1-type macrophages, followed by identical irradiation protocols. This model exhibited significant increases in ALT and AST, as well as prominent histological features of liver damage, including sinusoidal dilation, hepatocyte membrane rupture, and nucleolar condensation. Concomitantly, both mRNA and protein levels of TNF-α, IL-6, and IL-1β were markedly upregulated, confirming the pro-inflammatory and pro-injury role of M1-like KCs in the context of RILD. Notably, a discrepancy was observed between the in vivo and in vitro findings regarding MDM polarization. While in vitro irradiation at 10 Gy induced a significant shift toward M1 polarization, no marked alteration in the M1/M2 ratio of recruited macrophages was detected in the 30 Gy RILD mouse model. This difference likely reflects the complexity of the hepatic microenvironment in vivo. In contrast to isolated culture conditions, infiltrating macrophages in irradiated liver tissue are subject to multifaceted regulatory cues, including cytokines released by KCs, hepatocytes, and hepatic stellate cells, as well as extracellular matrix components and sinusoidal structural constraints. These signals may buffer or reshape macrophage polarization responses. Additionally, radiation-induced recruitment dynamics and temporal heterogeneity in macrophage differentiation may further contribute to the observed stability of MDM polarization in vivo. Collectively, these findings underscore the importance of microenvironmental regulation in shaping macrophage responses following irradiation.

SLPI is a secreted protein containing two whey acidic protein domains and is widely expressed in immune cells and body fluids [39,40,41,42,43,44]. Its expression can be induced by inflammatory stimuli, including microbial components and viral analogs, and it participates in immune defense, inflammatory regulation, and tissue repair [45,46,47]. SLPI plays a pivotal role in innate immune responses. It has been reported to be expressed in macrophages [48] and can be induced in pulmonary macrophages via the TLR2 signaling pathway during Mycobacterium tuberculosis infection [47]. In a model of acetaminophen-induced acute liver failure, SLPI expression was elevated in biliary epithelial cells and necrotic areas rich in KCs, suggesting a potential association with hepatic necrosis [45]. Although SLPI is known to be dynamically regulated under various inflammatory and tumor-related conditions, its precise role in macrophage polarization and inflammation regulation remains incompletely understood. In this study, we constructed RAW264.7 macrophage cell models with stable overexpression or knockdown of SLPI and demonstrated that SLPI significantly promotes the expression of M1 macrophage markers, such as TNF-α and IL-6, suggesting its crucial regulatory role in M1 polarization of KCs. Furthermore, we established an in vitro co-culture system mimicking the irradiated hepatic microenvironment, where KCs (with or without SLPI modulation) were co-cultured with hepatocytes under irradiation conditions. Notably, SLPI-expressing macrophages exposed to irradiation significantly enhanced pro-inflammatory cytokine expression and apoptosis in hepatocytes. In contrast, SLPI knockdown in co-cultured macrophages attenuated these responses. These findings support the notion that SLPI acts as a key mediator of radiation-induced M1 polarization in KCs, contributing to the amplification of inflammation and hepatocyte injury during RILD. Collectively, these results highlight the pro-inflammatory and pro-damage function of SLPI in the context of radiation-induced liver injury. Future studies will focus on in vivo-targeted intervention experiments to validate SLPI as a potential molecular target for the prevention and treatment of RILD and to promote the translational development of intervention strategies in this field.

Liposomes are spherical vesicles composed of one or more phospholipid bilayers and are widely utilized in drug delivery systems due to their excellent biocompatibility, low toxicity, and lack of immunosuppressive effects. Liposomal formulations have been used as therapy [49]. In recent years, macrophage-targeted gene therapy has emerged as a promising therapeutic strategy. Among various delivery systems, cationic lipid-based nanoparticles have proven to be one of the most effective non-viral vectors for delivering RNA molecules specifically to KCs [50]. These cationic carriers encapsulate negatively charged RNA molecules via electrostatic interactions, and the incorporation of neutral lipids helps to reduce cytotoxicity while promoting efficient cytoplasmic delivery [51]. Previous studies have confirmed that KCs can efficiently internalize liposomes with particle sizes around 100 nm via endocytosis [52]. For instance, a research team from the University of Virginia successfully delivered PD-L1 siRNA to KCs using liposomal carriers, resulting in a significant enhancement of antiviral immune responses in the liver [53,54].

In this study, we developed a liposome-based siSLPI delivery system that effectively achieved targeted knockdown of SLPI expression in KCs. Subsequent biochemical and histopathological analyses revealed that silencing SLPI significantly reduced serum ALT and AST levels in a murine model of RILD, improved hepatic tissue architecture, and alleviated nuclear pyknosis and disorganized hepatic cords. ELISA and immunohistochemistry further demonstrated that SLPI inhibition markedly decreased the expression of pro-inflammatory cytokines, including TNF-α, IL-6, and IL-1β, confirming the pivotal role of SLPI in driving M1 polarization of KCs and thereby promoting RILD progression. This study established a KC-targeted delivery platform and provided evidence supporting its therapeutic potential in RILD. While these findings offer new perspectives for macrophage-targeted precision therapy, the molecular mechanisms underlying SLPI-induced M1 polarization of KCs remain to be fully elucidated. Moreover, the signaling pathways through which M1-type KCs exacerbate RILD require further mechanistic investigation. Future studies will focus on elucidating the mechanistic roles of SLPI in RILD and evaluating its feasibility and translational potential as a therapeutic target.

This study has several limitations that should be acknowledged. Although our murine irradiation model recapitulates key pathological features of RILD, interspecies differences and variations in radiation dose, fractionation regimens, and injury kinetics may limit direct clinical extrapolation. Validation in fractionated irradiation settings and human liver samples will therefore be necessary to enhance translational relevance. KC depletion using GdCl₃ provides functional insight but does not fully replicate genetic or lineage-specific ablation. Incomplete depletion, temporal variability, and compensatory recruitment of monocyte-derived macrophages may confound KC-specific interpretations, warranting confirmation using genetic targeting or lineage-tracing approaches. In addition, while liposome-mediated SLPI inhibition demonstrated therapeutic efficacy, nanoparticle biodistribution remains influenced by the mononuclear phagocyte system, protein corona formation, and physicochemical properties, which may limit absolute cell-type specificity. Further pharmacokinetic, biosafety, and multi-dose evaluations will be required to support clinical translation.

Addressing these limitations will further refine our understanding of SLPI-mediated Kupffer cell reprogramming and facilitate the development of macrophage-targeted interventions for RILD. Despite these constraints, our integrative multi-level evidence establishes an SLPI-dependent pro-inflammatory KC axis as a central driver of RILD.

4. Materials and Methods

4.1. Experimental Animals

Male C57BL/6J mice (6–8 weeks old) were purchased from GemPharmatech Co., Ltd. (Nanjing, Jiangsu, China) and housed in the SPF-grade animal facility of the Medical College of Soochow University. The animals were maintained under controlled conditions with a temperature of 20–25 °C, relative humidity of 40–70%, and air cleanliness of Class 10,000. Mice had free access to standard chow and autoclaved drinking water provided by the facility. All animal procedures strictly complied with the guidelines of the Regulations on the Administration of Laboratory Animals and the Guide for the Care and Use of Laboratory Animals. The experimental protocols were reviewed and approved by the Animal Ethics Committee of Soochow University, and all procedures complied with institutional ethical standards and animal welfare regulations.

4.2. Murine Model of RILD

RILD was induced in C57BL/6J mice using the X-RAD SmART small-animal image-guided irradiation system (Precision X-Ray Inc., North Branford, CT, USA). Mice underwent precise whole-liver irradiation with the following parameters: tube voltage of 225 kV, tube current of 13 mA, dose rate of 3 Gy/min, total dose of 30 Gy, exposure time of 590 s, and source-to-surface distance (SSD) of 306 mm. Animals were randomly divided into three groups: control group (non-irradiated, n = 5), 1-day post-irradiation group (n = 5), and 2-week post-irradiation group (n = 5).

4.3. Kupffer Cell-Depleted RILD Mouse Model

To investigate the role of KCs in RILD, C57BL/6J mice were randomly assigned into four groups (n = 5 per group): in the control group, mice received an equal volume of phosphate-buffered saline (PBS) via tail vein injection; those in the GdCl₃ group received intravenous injection of gadolinium chloride (GdCl₃, 10 mg/kg, Solarbio, Beijing, CAS No. 10138-52-0); those in the IR + PBS group were injected with PBS 24 h prior to liver irradiation; and those in the IR + GdCl₃ group were injected with GdCl₃ 24 h prior to irradiation.

4.4. RILD Mouse Model with Adoptive Transfer of M1 Macrophages

To evaluate the functional role of M1-polarized macrophages in RILD, C57BL/6J mice were randomly divided into four groups (n = 5 per group) as follows: in the IR + PBS group, mice received intravenous injection of PBS 24 h prior to irradiation; those in the IR + GdCl₃ group were injected with GdCl₃ (10 mg/kg) via the tail vein 24 h before irradiation; those in the IR + GdCl₃ + M1 group were pretreated with GdCl₃ 24 h before irradiation, followed by tail vein infusion of 5 × 10⁶ in vitro-polarized M1 macrophages 4 h prior to irradiation; and those in the IR + M1 group received intravenous injection of 5 × 10⁶ M1 macrophages 4 h prior to irradiation without prior GdCl₃ depletion.

4.5. Serum Biochemical Analysis of Liver Function

Whole blood was collected from each mouse via retro-orbital bleeding. Samples were allowed to clot at room temperature for 1 h and subsequently centrifuged at 5000 rpm for 15 min at 4 °C to obtain serum. The levels of ALT and AST in the serum were measured using an XN-1000V automated analyzer (Sysmex Corporation, Kobe, Japan). The results were expressed in units per liter (U/L) and served as biochemical indicators of liver function injury.

4.6. Hematoxylin–Eosin (H&E) Staining

Mouse tissues, including the heart, liver, spleen, lungs, and kidneys, were fixed in 4% paraformaldehyde solution (pH 7.4) for 24 h, followed by dehydration, clearing, paraffin infiltration, and embedding. Paraffin-embedded sections were deparaffinized in xylene and rehydrated through a graded ethanol series. H&E staining was performed using a commercial staining kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer’s instructions. After staining, the sections were dehydrated, cleared, and mounted. Histological changes in liver tissues were observed and imaged using a fluorescence microscope in bright-field mode. (Leica Microsystems, Wetzlar, Germany) to evaluate the pathological features of RILD.

4.7. scRNA-Seq Analysis Using 10×Genomics

Liver tissues were collected from mice in the control group and from those 2 weeks after 30 Gy irradiation. scRNA-seq was performed using the 10×Genomics Chromium platform, following protocols described in our previous study. Briefly, single-cell suspensions were loaded onto a 10×Genomics microfluidic chip (Chromium Single Cell A Chip) and encapsulated into oil-in-water emulsion droplets along with barcoded gel beads to generate Gel Bead-In-Emulsions (GEMs). Reverse transcription was carried out within droplets to synthesize barcoded cDNA. After library amplification and purification, sequencing was performed on an Illumina high-throughput platform to generate single-cell transcriptomic profiles. Single-cell RNA sequencing (scRNA-seq) data were analyzed using the Seurat package (version 5.4) in R (version 4.4.2). Standard preprocessing steps were performed, including quality control, data normalization, identification of highly variable genes, scaling, principal component analysis (PCA), and cell clustering. To correct batch effects and integrate datasets across samples, Harmony (version 1.2.4) was applied following PCA. The Harmony-corrected embeddings were subsequently used for downstream dimensionality reduction and clustering analyses. Pseudotime trajectory analysis was conducted using Monocle (version 2.34) to characterize dynamic gene expression changes across macrophage subpopulations. Highly variable genes were selected to construct cell trajectories and visualize lineage progression. Cell type annotation was performed based on canonical marker genes in combination with publicly available cell marker databases.

4.8. Isolation of Hepatic Parenchymal Cells, Non-Parenchymal Cells, and KCs

Primary liver cells were isolated from male C57BL/6J mice using the classical two-step collagenase perfusion method [55]. The procedure for isolating hepatocytes was as follows: the liver was first perfused via the portal vein with a pre-perfusion solution containing 4–6 mg/L heparin sodium until the effluent became clear and colorless. This was followed by perfusion with a digestion buffer containing 0.5 g/L collagenase to enzymatically dissociate the liver tissue. The digested tissue was filtered through a 40 μm cell strainer, and the filtrate was centrifuged at 50× g for 5 min. The centrifugation step was repeated three times, and the resulting cell pellet was collected as the hepatocyte fraction. Cells were resuspended in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin and cultured under standard conditions.

To isolate hepatic non-parenchymal cells (NPCs), the supernatant obtained from the hepatocyte fraction was collected and centrifuged at 800× g for 10 min. The resulting pellet, containing a mixed population of non-parenchymal cells, was resuspended and subjected to density gradient centrifugation using 70% and 30% Percoll (Cat. No. 17089101, Cytiva, Marlborough, MA, USA). The cell suspension was layered between the two Percoll gradients and centrifuged at 1500× g for 24 min. The interphase between the 70% and 30% Percoll layers, which appeared as a hazy band, was carefully collected and centrifuged again at 800× g for 5 min. The final cell pellet was identified as the KC population.

4.9. Immunohistochemical (IHC) Staining

As previously described, paraffin-embedded liver tissue sections were subjected to routine deparaffinization and rehydration. Immunohistochemistry was performed to detect macrophages and inflammatory cytokines in liver tissue. The sections were incubated overnight at 4 °C with the following primary antibodies: recombinant rabbit anti-F4/80 antibody (ab300421, Abcam, Cambridge, UK), anti-iNOS antibody (ab15323, Abcam, UK), anti-TNF-α antibody (#41504, Signalway Antibody, College Park, MD, USA), anti-IL-6 antibody (#41739, Signalway Antibody, USA), and anti-IL-1β antibody (#41059, Signalway Antibody, USA). On the following day, sections were incubated with the corresponding HRP-conjugated secondary antibodies. Color development was carried out using a DAB Substrate Kit (Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China), and nuclei were counterstained with hematoxylin. After dehydration and mounting, the stained sections were examined under an inverted fluorescence microscope (Leica Microsystems, Wetzlar, Germany) to assess the distribution of F4/80⁺ Kupffer cells and the expression levels of iNOS, TNF-α, IL-6, and IL-1β.

4.10. Cell Culture, Irradiation, and Induction of M1/M2 Macrophage Polarization

The murine macrophage cell line RAW264.7 was obtained from Xinrun Biotechnology Co., Ltd. (Wuxi, China). Cells were cultured in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Waltham, MA, USA) and 1% penicillin–streptomycin solution (Procell Life Science & Technology Co., Ltd., Wuhan, China). All cultures were maintained in a humidified incubator at 37 °C with 5%CO₂. Irradiation of RAW264.7 cells was performed using the RAD Source RS2000 X-ray irradiator (Rad Source Technologies, Suwanee, GA, USA) at a dose rate of 1.214 Gy/min. Cells were exposed to radiation doses of 0 Gy, 5 Gy, 10 Gy, or 20 Gy as indicated. For macrophage polarization, RAW264.7 cells were seeded in complete medium supplemented with either 20 ng/mL interferon-γ (IFN-γ; Cat. No. 315-05-20UG, Peprotech, Rocky Hill, NJ, USA) for M1 polarization or 20 ng/mL interleukin-4 (IL-4; Cat. No. 214-14-5UG, Peprotech, USA) for M2 polarization. Cells were cultured for 48 h to induce M1- or M2-type phenotypes.

4.11. Western Blot Analysis

Total protein was extracted from mouse liver tissues and RAW264.7 macrophages using RIPA lysis buffer supplemented with a protease inhibitor cocktail. After lysis, the supernatants were collected by centrifugation and protein concentrations were determined using a BCA Protein Assay Kit (Cat. No. P0010, Beyotime Biotechnology, China). Equal amounts of protein from each sample were separated by SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were blocked and incubated overnight at 4 °C with the following primary antibodies: anti-TNF-α, anti-IL-6, anti-IL-1β, anti-cleaved caspase-3 (Cat. No. #29034, Signalway Antibody, USA), anti-SLPI (Cat. No. #38316, Signalway Antibody, USA), anti-ARG1 (Cat. No. #32468, Signalway Antibody, USA), anti-Tubulin (Beyotime Biotechnology, China), and anti-GAPDH (Beyotime Biotechnology, China). After washing with TBS containing 0.1% Tween-20 (TBST), the membranes were incubated with appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (Beyotime Biotechnology, China) at room temperature for 1 h. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection kit (Cat. No. P10300, NCM Biotech, Suzhou, China) and imaged using a multi-color chemiluminescence imaging system (Alpha Innotech, San Leandro, CA, USA).

4.12. RNA Extraction and Quantitative Real-Time PCR (qRT-PCR)

Total RNA was extracted from mouse liver tissues and RAW264.7 macrophages using a Fast RNA Extraction Kit for Tissue/Cells (Cat. No. U10018A, UU Biotech, Suzhou, China), following the manufacturer’s instructions. The RNA concentration and purity were determined using a NanoDrop spectrophotometer (Thermo Scientific, USA). Complementary DNA (cDNA) was synthesized from mRNA using the All-In-One 5× RT MasterMix (Cat. No. G592, ABMGood, Richmond, BC, China). Quantitative real-time PCR was performed using NovoStart SYBR qPCR SuperMix Plus (Cat. No. E096-01A, Novoprotein, Shanghai, China) on a fluorescence-based real-time PCR detection system. Primer sequences used for qRT-PCR are listed in

Table 1. Primer sequences used for quantitative real-time PCR (qRT-PCR).

Gene	Species	Forward Primer (5′–3′)	Reverse Primer (5′–3′)
SLPI	Mouse	AAGCCACAATGCCGTACTGACTG	ACAGGATTCACGCACTTGGAACC
HMOX1	Mouse	ACCGCCTTCCTGCTCAACATTG	CTCTGACGAAGTGACGCCATCTG
IL-6	Mouse	CTTCTTGGGACTGATGCTGGTGAC	TCTGTTGGGAGTGGTATCCTCTGTG
IL-1β	Mouse	CACTACAGGCTCCGAGATGAACAAC	TGTCGTTGCTTGGTTCTCCTTGTAC
TNFα	Mouse	GGACTAGCCAGGAGGGAGAACAG	GCCAGTGAGTGAAAGGGACAGAAC

. Relative gene expression was calculated using the 2^−ΔΔCt method, with GAPDH as the internal control.

4.13. Cell Transfection

Small interfering RNAs (siRNAs) targeting the mouse Slpi gene were synthesized by Shanghai GenePharma Co., Ltd, Shanghai, China. The Slpi overexpression plasmid was constructed by cloning the full-length mouse Slpi cDNA into the pcDNA3.1(+) eukaryotic expression vector (Weizhen Biotech, Shandong, Jinan, China). Both gene knockdown and overexpression strategies were applied to modulate SLPI expression levels in macrophages. The siRNA sequences were as follows: Negative control siRNA (siNC): Sense: 5′-UUCUCCGAACGUGUCACGUdTdT-3′ Antisense: 5′-ACGUGACACGUUCGGAGAAdTdT-3′ SLPI-targeting siRNA (siSLPI): Sense: 5′-GCUGUUGGAGAGCAAAUAATT-3′ Antisense: 5′-UUAUUUGCUCUCCAACAGCTT-3′. For overexpression, the SLPI cDNA was subcloned into the pcDNA3.1(+) vector, and the empty vector was used as a negative control. Transfection of siRNAs and plasmids into RAW264.7 macrophages was performed using the jetPRIME^® DNA/siRNA Transfection Reagent (Polyplus-transfection^®, Illkirch, France) according to the manufacturer’s protocol. Successful transfection was confirmed by qRT-PCR and Western blotting.

4.14. Co-Culture Model of Hepatocytes and Macrophages

A Transwell-based indirect co-culture system was established to study the interaction between hepatocytes and macrophages. Primary hepatocytes were seeded in the lower chamber (basal compartment) of a Transwell system (pore size: 0.4 μm; Corning, NY, USA). After cell attachment, primary KCs, RAW264.7 cells, or siRNA-transfected RAW264.7 cells in the logarithmic growth phase were added to the upper chamber (apical insert) to initiate co-culture. The ratio of hepatocytes to RAW264.7 cells was maintained at 6:1 [56]. The experimental groups were designed as follows. Irradiated hepatocytes + non-irradiated KCs or RAW264.7 cells: hepatocytes were irradiated and co-cultured with untreated KCs or RAW264.7 cells. Non-irradiated hepatocytes + irradiated KCs or RAW264.7 cells: KCs or RAW264.7 cells were irradiated prior to co-culture with untreated hepatocytes. Irradiated hepatocytes (monoculture): hepatocytes were irradiated and cultured alone as damage control. Irradiated hepatocytes + irradiated KCs or RAW264.7 cells: both hepatocytes and macrophages were irradiated before co-culture. Irradiated hepatocytes + irradiated and SLPI-siRNA-transfected RAW264.7 cells: hepatocytes and RAW264.7 cells (transfected with SLPI-targeting siRNA) were both irradiated prior to co-culture. After 24 h of co-culture, hepatocytes from the lower chamber were collected for subsequent molecular and phenotypic analyses.

4.15. Flow Cytometry

4.15.1. Phenotypic Analysis of Hepatic Macrophages

Non-parenchymal liver cells were isolated from control mice and from mice on day 1 and week 2 post-irradiation (30 Gy X-ray). Cells were stained with the following fluorophore-conjugated monoclonal antibodies: APC anti-mouse F4/80 (T45-2342, #566787, BD Pharmingen, San Diego, CA, USA), FITC anti-CD11b (M1/70, #557396, BD Pharmingen, San Diego, CA, USA), and PE-Cy7 anti-mouse CD86 (GL1, #560582, BD Pharmingen, San Diego, CA, USA). After surface staining, cells were fixed and permeabilized using the Fixation/Permeabilization Kit (#554714, BD Pharmingen, USA), followed by intracellular staining with PE anti-mouse CD206 (MMR, #141706, BioLegend, San Diego, CA, USA). Samples were analyzed using a FACSVerse flow cytometer (BD Biosciences, San Jose, CA, USA). Data were processed with FlowJo software (version 10) (Tree Star Inc., Ashland, OR, USA) to assess macrophage polarization status based on CD86 (M1 marker) and CD206 (M2 marker) expression.

4.15.2. Apoptosis Analysis of Hepatocytes in Co-Culture Analysis of Hepatocyte Apoptosis in the Co-Culture System

To assess apoptosis, primary hepatocytes were co-cultured with RAW264.7 cells, siSLPI-transfected RAW264.7 cells, or primary Kupffer cells for 24 h, according to the previously described Transwell co-culture model. These cultures were then subjected to 0, 5, or 10 Gy of X-ray irradiation. After an additional 24 h, hepatocytes were harvested, resuspended in PBS, and stained using the Annexin V-PE/7-AAD Apoptosis Detection Kit (A213-01, Novoprotein Scientific Inc., Shanghai, China), following the manufacturer’s instructions. Flow cytometric analysis was performed on a FACSVerse instrument (BD Biosciences, San Jose, CA, USA), and the percentages of apoptotic cells (Annexin V⁺/7-AAD⁻ for early apoptosis and Annexin V⁺/7-AAD⁺ for late apoptosis) were quantified.

4.16. Lipid Nanoparticle (LNP) Preparation, Characterization, and Encapsulation Efficiency Analysis

4.16.1. LNP Synthesis via Microfluidic Mixing Synthesis of LNPs via Microfluidic Mixing

Lipid nanoparticles (LNPs) were synthesized using a microfluidic-based NanoGenerator Flex-M system (PreciGenome LLC, San Jose, CA, USA). A lipid mixture consisting of SM-102, DSPC, cholesterol, and PEG-2000-DMG at a molar ratio of 50:11.3:60:1.5 was dissolved in ethanol. The fluorescent dye DiD was added at a volume ratio of 1:500 (DiD: ethanol). Separately, siRNA was dissolved in acetate buffer (pH 4.5). Both solutions were injected into a Y-shaped microfluidic chip at a constant flow rate of 3 mL/min, with a flow rate ratio of lipids to siRNA = 1:3. The resulting LNPs were dialyzed against PBS for 24 h to remove ethanol and then filtered using a 0.22 μm membrane and stored at 4 °C until further use.

4.16.2. Particle Size and Zeta Potential Measurement

The hydrodynamic diameter and zeta potential of LNPs were measured using dynamic light scattering (DLS) on a Zetasizer Nano ZS90 analyzer (Malvern Panalytical, Malvern, UK) with disposable cuvettes. Measurements were performed at 25 °C in triplicate, and data were reported as mean ± SD.

4.16.3. TEM

LNP morphology was visualized using field-emission transmission electron microscopy (FE-TEM). A 10 μL aliquot of LNP suspension was dropped onto a carbon-coated copper grid, followed by negative staining with 10 μL of 2% phosphotungstic acid for contrast enhancement. After air drying for 6 h, the samples were imaged using a TF20 microscope (Thermo Fisher Scientific, USA).

4.16.4. Encapsulation Efficiency (EE%) of siRNA

The encapsulation efficiency (EE%) of siRNA within LNPs was determined using the Quant-iT™ RiboGreen™ RNA assay kit (Thermo Fisher Scientific, USA). LNP samples were divided into two groups: the Triton X-100-treated (lysis group) and untreated (non-lysis group) groups. Fluorescence intensity was measured using a Synergy 3 microplate reader (BioTek Instruments, Winooski, VT, USA) at excitation/emission wavelengths of 480/520 nm. The total siRNA content was determined from the lysis group, while free (non-encapsulated) siRNA was assessed from the non-lysis group. The encapsulation efficiency was calculated using the following formula:

$$\mathrm{Encapsulation} \mathrm{Efficiency} (\%)=\left(1-{\displaystyle \frac{F_{non\text{-}lysis}}{F_{lysis}}}\right)\times 100$$

where F_non-lysis and F_lysis represent the fluorescence intensity of non-lysed and lysed LNPs, respectively, after subtracting background values.

4.17. Cellular Uptake and Fluorescent Imaging

To assess the uptake of lipid nanoparticles (LNPs), RAW264.7 murine macrophages were incubated with DiD-labeled LNPs at designated time points. After incubation, cells were stained with Calcein AM (C2012-0.1 ml, Beyotime Biotechnology, China) to visualize live cytoplasm and Hoechst 33342 (C1028, Beyotime Biotechnology, China) for nuclear counterstaining. Following staining, the cells were observed under an FV1200 laser scanning confocal microscope (Olympus, Japan) to evaluate the intracellular distribution and fluorescence signal intensity of the LNPs.

4.18. Cell Viability Assay (CCK-8)

Cell viability was evaluated using the Cell Counting Kit-8 (CCK-8) assay (K1018, APExBIO Technology LLC, Houston, TX, USA). RAW264.7 cells were seeded into 96-well plates and divided into the following groups: medium-only control, blank control, and liposome-treated groups (10 μL, 20 μL, 50 μL, and 100 μL). Liposomes were diluted in complete culture medium and added to each group accordingly. After 24 h of incubation, 100 μL of a mixture of CCK-8 reagent and DMEM medium (1:9, v/v) was added to each well. The plates were incubated at 37 °C in the dark for 1 h, and the absorbance was measured at 450 nm using a microplate reader.

4.19. In Vivo Fluorescence Imaging (IVIS)

To evaluate the biodistribution of liposomes, DID-labeled liposomes (containing DID dye, CAS: 127274-91-3; MedChemExpress LLC, Monmouth Junction, NJ, USA) were intravenously injected via the tail vein into C57BL/6J mice. At designated time points post-injection, whole-body fluorescence imaging was performed using the IVIS Lumina III in vivo imaging system (PerkinElmer, Waltham, MA, USA). The distribution of fluorescent signals was captured and analyzed to assess the in vivo trafficking of liposomes.

4.20. In Vivo Mouse Models for SLPI Knockdown via AAV8-shSLPI and siSLPI-Loaded Liposomes

(1)

AAV8-shSLPI adenoviral vector model: male C57BL/6J mice aged 6–8 weeks were randomly assigned into four groups (n = 5 per group): the NC group (negative control, no irradiation), shSLPI group (AAV8-shSLPI injection alone), IR group (irradiation only), and IR + shSLPI group (AAV8-shSLPI injection + irradiation). Mice in the shSLPI groups received AAV8-shSLPI viral particles (5 × 10¹¹ vg per mouse) via intravenous tail vein injection. Two weeks post-injection, the mice were subjected to subsequent experimental procedures.

(2)

siSLPI liposome delivery model: similarly, male C57BL/6J mice (6–8 weeks old) were randomly divided into four groups: the NC group (n = 5), siSLPI group (siSLPI liposome only, n = 5), IR group (irradiation only, n = 5), and IR + siSLPI group (siSLPI liposome + irradiation, n = 5). Mice in the siSLPI groups were administered siSLPI-loaded liposomes at a dose of 2 mg/kg via tail vein injection. Twenty-four hours after liposome administration, the mice were subjected to further experimentation.

4.21. Enzyme-Linked Immunosorbent Assay (ELISA)

The cell culture supernatants were collected from the co-culture system of hepatic parenchymal cells and macrophages and the in vivo siSLPI-liposome-mediated SLPI knockdown mouse model, according to their respective experimental groupings. The levels of inflammatory cytokines—IL-6, TNF-α, and IL-1β—in the supernatants were quantified using mouse-specific ELISA kits (Mouse IL-6, Mouse TNF-α, and Mouse IL-1β ELISA kits, Youxuan Bio, Shanghai, China), strictly following the manufacturer’s instructions. Absorbance was measured at 450 nm using a microplate reader, and cytokine concentrations were calculated based on standard curves generated for each target protein.

4.22. Statistical Analysis

All experiments were independently repeated at least three times. Data are presented as mean ± standard deviation (SD). Flow cytometry data were analyzed using FlowJo software (BD Biosciences, San Jose, CA, USA). Image quantification was performed using ImageJ software (version 1.53; National Institutes of Health, Bethesda, MD, USA). Statistical analyses were conducted using GraphPad Prism software (version 8.0; GraphPad Software, San Diego, CA, USA). For comparisons between two groups, a two-tailed unpaired Student’s t-test was applied. For comparisons among multiple groups, one-way analysis of variance (ANOVA) was used, followed by appropriate post hoc tests when necessary. p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****).

5. Conclusions

KCs are central drivers of RILD, with radiation-induced M1 polarization representing a key pathological mechanism underlying disease progression. Using single-cell transcriptomic analysis combined with functional assays, we demonstrated that ionizing radiation markedly shifts KCs toward a pro-inflammatory M1 phenotype and induces pronounced upregulation of SLPI. As a key molecular regulator of KC polarization, elevated SLPI amplifies inflammatory cytokine production, exacerbates hepatic tissue damage, and promotes the progression of RILD, whereas gain- and loss-of-function models confirmed that SLPI overexpression potentiates M1 polarization and inflammation, while its suppression mitigates liver dysfunction and histopathological injury.

To mechanistically target the SLPI–KC axis in RILD, we designed a liposome-based targeted delivery system to selectively modulate SLPI expression within KCs. This nanocarrier effectively attenuated radiation-induced SLPI upregulation, thereby suppressing M1 polarization at its cellular origin and substantially alleviating hepatic inflammatory responses and tissue injury in vivo. Therefore, these findings establish the SLPI–KC axis as a pivotal pathogenic pathway driving immune dysregulation and liver damage following irradiation and demonstrate that liposome-mediated targeting of SLPI enables functional reprogramming of KCs to restore immune homeostasis. This strategy not only provides new mechanistic insight into RILD pathogenesis but also highlights a rational and translatable nanodelivery approach for precision intervention in radiation-induced liver injury.