The mRNA micro-array datasets used in biological information analysis are GSE85359 and GSE41789. These data were based on the Agilent GPL1261 platform ([mouse430_2] Affymetrix mouse genome 430 2.0 array), and a total of 50 radiation pneumonitis samples and 22 normal control samples were included. Differentially expressed genes (DEGs) in radiation pneumonitis were identified using Gene Expression Omnibus (GEO) database. DAVID database was used for gene ontology (GO) and Kyoto Encyclopedia of Genes and Genome (KEGG) enrichment analysis, and then Gene Set Enrichment Analysis (GSEA) was used to analyze abnormal expressions. Protein–protein interaction (PPI) networks were constructed using STRING (String: http://string.embl.de/) and Cytoscape. Syk inhibitors were analyzed using Discovery Studio 4.5 software.

C57BL/6N mice (female, 6–8 weeks) purchased from Charles River Laboratories (China, Beijing) were raised in specific pathogen-free (SPF) barrier environment in the Laboratory Animal Center of Jilin University with suitable temperature, humidity, and light control. Animal experiments were carried out in accordance with the ethical approval issued by the Animal Ethics Committee at the First Bethune Hospital of Jilin University (IACUC approval number 2022-0034).

Syk inhibitor fostamatinib was purchased from Invitrogen™ (Carlsbad, CA, USA), interleukin-1β (IL-1β), interleukin-6 (IL-6), phospho-p38 (p-p38), p53, α-smooth muscle actin (α-SMA), collagen type 1-alpha1 (COL1A1), transforming growth factor (TGF)-β, and Syk antibodies were purchased from ZEN BioScience (Chengdu, Sichuan, China), and the p-Syk antibody was from the β-tubulin antibody purchased from Abcam (Cambridge, MA, USA). Hydroxyproline content assay kit was purchased from Solarbio (cat# BC0255). All other reagents were obtained from Sigma Chemical Co. (St. Louis, MO, USA).

The female C57BL/6N mice were anesthetized by an intraperitoneal injection of pentobarbital (60 mg/kg). A total of 30 were fixed and placed in the supine position for radiation. The mouse head and the abdomen were shielded by using lead plates. Thoracic radiation was performed with a single dose of 18 Gy in an X-ray irradiator (Faxitron, USA). The mice were randomly allocated into group R (n = 10) and group R+S (intraperitoneal injection of 20 mg/kg·day fostamatinib, n = 10; fostamatinib was dissolved in 0.5% DMSO). The other 10 untreated mice served as controls (group C). The mice in groups C and R received an equal volume of 0.5% DMSO simultaneously. All mice received intraperitoneal injection 1 day before radiation and after 7 days. Five mice in each group were randomly selected and sacrificed 2 weeks after radiation. The remaining mice were sacrificed at 10 weeks after radiation by cervical dislocation.

The mice were sacrificed, and both lungs were harvested. Part of the lungs were snap-frozen in liquid nitrogen for western blot analysis. Some lungs were fixed in 10% neutral-buffered formalin at 4°C for 12 h, processed, embedded in paraffin wax, sliced into 5-μm sections, and observed by using a light microscope after hematoxylin/eosin (HE) and Masson’s trichrome staining. The number of inflammatory cells per high-power field, width of blood vessels, and width of interalveolar septa were quantitatively determined from five random fields in high magnification. Smith scale was applied to evaluate the degree of lung injury at 2 weeks after radiation, according to the presence of pneumonedema, inflammation, hemorrhage, atelectasis, and hyaline membrane formation. Each of these features was scored as follows: no damage (0), mild (1), moderate (2), and severe (3) (14, 15). Each mouse was observed at five high-power fields, and then the average was taken. Ashcroft score was used to assess the grade of lung fibrosis at 10 weeks after radiation (16, 17).

The immunohistochemistry of Syk was performed as described previously (18). The slides were treated overnight with primary antibody SYK (1:50) at 4°C and incubated with secondary antibody (1:500) (HRP-conjugated anti-rabbit IgG).

Total cellular protein was extracted as previously described. Equal amounts of proteins were mixed with an equal volume of reducing SDS sample buffer and boiled at 95°C for 5 min. Protein samples were resolved by 10% SDS-PAGE and electroblotted on nitrocellulose membranes (BioRad, Hercules, CA, USA). After electroblotting, nonspecific binding was blocked with 5% bovine serum albumin (BSA). The membrane was incubated overnight with primary antibodies at 4°C and incubated with horseradish-peroxidase-conjugated secondary antibodies for 1 h at room temperature. Primary antibodies against the following proteins were used at 1:1,000 dilutions unless otherwise indicated: p-SYK, IL-1β, IL-6, TGF-β1 p-p38, p53, α-SMA, COL1A1, and GAPDH antibodies (1:5,000 or 1:10,000). During the western blot experiments, some blots were cut prior to hybridization with antibodies. Immunoreactive bands were visualized using enhanced chemiluminescence (Amersham, Arlington, IL, USA) as previously described. Western blot was repeated at least five times.

All experiments were conducted at least three times. Data depicted in graphs represent the mean ± SE for each group. Brown–Forsythe and Bartlett tests were employed to assess the homogeneity of variances. One-way analysis of variance (ANOVA) was conducted for comparison among multiple groups, and Tukey multiple comparison test was used to evaluate the significance between groups. In all analyses, P-values less than 0.05 were considered statistically significant.

The Venn diagram in Figure 1A shows the DEGs between radioactive pneumonitis and normal lung tissue, and the heatmap of core genes is shown in Figure 1B . We constructed a PPI network and then performed molecular complex detection (MCODE) to screen out PPI network modules with MCODE score >10 and number of nodes >11. The core driver genes of radiation pneumonitis were screened and are listed in Table 1 . Among these genes, Syk was highly expressed, with a node degree of 48 in radiation pneumonitis. By analyzing the PPI network of all DEGs and using MCODE to screen out the three most important modules (as shown in Figure 1C ), we annotated the functions of these module genes and their abundances (as shown in Supplementary Table S1 ), including leukocyte chemotaxis, cell cycle, integral component of membrane, mitotic cytokinesis, ATP binding, and P53 signaling pathway.

The GO analysis showed that the DEGs between radioactive pneumonitis and normal lung tissue were enriched in B-cell receptor signaling pathway, transmembrane receptor protein tyrosine kinase signaling pathway, integrin-mediated signaling pathway, positive regulation of cytokinesis, and mobile behavior. The KEGG analysis revealed enrichment of DEGs in chemokine signaling pathway, cytokine–cytokine receptor interaction, P53 signaling pathway, primary immune deficiency, and hematogenesis cell lines ( Supplementary Figure S1 , Supplementary Table S2 ). The GSEA of DEGs showed that there was a significant abnormal regulation of mechanical ventilation pathways in radiation pneumonitis ( Supplementary Figure S2 ).

We conducted drug molecular docking analysis targeting the inhibition of Syk gene in the drug database. The criteria of drug screening include the following: (1) it has the ability to specifically kill the cells with abnormal expression of Syk and has already entered in clinical trials, (2) minor toxicity and side effects to the human body, and (3) good solubility and high absorption. The top 20 drug molecules scored out of 7,049 drugs that specifically act on the SYK transcriptional protein were selected and are presented in Supplementary Table S3 , among which ZINC43131420 (fostamatinib) had the most favorable pharmacokinetic profile and less hepatotoxicity according to ADMET analysis ( Table 2 ). Fostamatinib binds to SYK transcriptional protein primarily through hydrogen bonding and van der Waals forces ( Figure 2A ). The intermolecular potential energy curves revealed the potential energy changes between the drug and target protein molecules ( Supplementary Figure S3 ). The interaction energy curve and RMSD curve ( Figures 2B, C ) indicated that fostamatinib is very stable and not prone to be allosteric. The toxicity analysis results are shown in Figure 2D . Fostamatinib has no potential of mutagenic and developmental toxicity, which could be an ideal drug.

We observed the pathological changes in the lung tissue of mice at 2 weeks after radiation. Compared with group C, the alveolar walls were significantly widened, the capillaries were dilated and congested (yellow arrows), and there were inflammatory cell infiltration around bronchioles and pulmonary interstitial inflammation (black arrows) in group R ( Figures 3A–D ). The application of Syk inhibitor significantly reduced pulmonary interstitial inflammation, slightly widened alveolar walls, and reduced telangiectasia and inflammatory cell exudation ( Figures 3A–D ). Lung injury score confirmed that there was a statistical decrease in group R+S compared with group R ( Figure 3E ). Therefore, it can be inferred that Syk inhibitors have a protective effect on radiation pneumonitis.

Consistent with the results of the bioinformatics analysis, we found that the protein level and phosphorylation of Syk were highly expressed in the lung tissue of irradiated mice and significantly decreased in mice administered with the Syk inhibitor ( Figures 4A–C ).

Western blot analysis demonstrated that the expression of p-p38 and p53 in lung tissue increased at 2 weeks after radiation compared with that observed in the control group, which was also consistent with the findings in the bioinformatics analysis. p-p38 and p53 were downstream sites of the Syk signaling pathway, and their expressions decreased in the lung tissue of the group administered with the Syk inhibitor ( Figures 4D, E ).

IL-1β and IL-6 are the major cytokines in radiation pneumonitis (19, 20). To determine whether inflammation was suppressed by Syk inhibitor, the expression of IL-1β and IL-6 at 2 weeks after radiation was evaluated by Western blot analysis. The levels of IL-6 and IL-1β expressions in lung tissue were significantly increased in irradiated lung tissue and were downregulated in the Syk-inhibitor-administered group, as shown in Figures 5A–C .

The experimental results discussed above indicated that Syk inhibitors can alleviate radiation pneumonitis. We further investigated whether pulmonary fibrosis was influenced by Syk inhibition. Mice were sacrificed at 10 weeks after radiation. H&E and Masson’s trichrome staining showed that, in group R, the alveolar walls were widened, the capillaries were moderately dilated and congested (green arrows), there was some lymphocyte and monocyte infiltration ( Figures 6A–D ), and mild to moderate fibrous and collagen hyperplasia was observed, with alveolar collapse and poor aeration (black arrows). Compared with group R, the degree of lung interstitial inflammation and fibrous and collagen hyperplasia (black arrows) was significantly reduced in the Syk-inhibitor-administered group ( Figures 6A–D ). The Ashcroft score and collagen volume fraction (CVF) confirmed changes in pulmonary fibrosis in these three groups ( Figures 6E, F ). The content of hydroxyproline in the lung tissue of group R was significantly higher (yellow arrows) than that of the control group, suggesting collagen hyperplasia, which was significantly reduced by Syk inhibitor ( Figure 6G ).

TGF-β is an important cytokine that promotes the pathogenesis of pulmonary fibrosis by inducing the production of collagen and the loss of lung elasticity (21). α-SMA and COL1A1 are usually used as biomarkers to evaluate the degree of pulmonary fibrosis (22, 23). We demonstrated that the expressions of TGF-β1, SMA, and COL1A1 in the lung tissue were significantly increased in group R and were inhibited by Syk inhibitor in group R+S ( Figures 6H–K ). The results showed that Syk inhibitor could attenuate the development of pulmonary fibrosis.