A dynamically resolved single-cell architecture of radiation pneumonitis provides insights into acute lung injury

A dynamically resolved single-cell architecture of radiation pneumonitis provides insights into acute lung injury | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A dynamically resolved single-cell architecture of radiation pneumonitis provides insights into acute lung injury Gaoming Liao, Chanjin Liang, Ting Chen, Youqing Zhu, Sihan Tang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7659149/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Radiation pneumonitis (RP) is a deleterious complication of thoracic radiotherapy, yet the cellular mechanisms driving its onset and progression remain unclear. Here we constructed a single-cell dynamic architecture of RP rats with acute lung injury at multiple time points after radiation from 84,865 high-quality cells through single-cell RNA sequencing. Results Endothelial and epithelial cells are damaged within 24 hours after radiation, while epithelial-mesenchymal transitions (EMT) occur in RP lesions at 1–2 weeks. Identification of radiation-induced EMT signature highly correlated with and superior to known EMT signature. Radiation induces oxidative stress and promotes apoptosis in monocytes one week after radiation exposure, and the induced inflammation persists. Macrophage components enhance the pro-inflammatory response following radiation via MIF signaling and exhibit four distinct intercellular communication patterns. The ligand Mif was associated with radiation-induced expression enhancement, and its blockade alleviated pneumonia symptoms. The dynamics and differentiation of lymphocytes reveal that effector and helper T cells activate within 2–4 weeks post-radiation, while tissue-resident memory T cells proliferate at 6 weeks. Conclusions This RP architecture provides a comprehensive view of the cellular architecture and dynamics following radiation exposure, enhancing our understanding of RP’s pathogenesis and offering biomarkers and potential therapeutic targets for early diagnosis and intervention. Molecular Biology Bioinformatics Radiation pneumonitis Single-cell dynamics pro-inflammatory response MIF signaling Acute lung injury Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Radiation pneumonitis (RP) is a potentially severe inflammatory condition of the lungs, primarily affecting patients who undergo radiotherapy for thoracic malignancies such as lung cancer, breast cancer, and esophageal cancer [ 1 – 3 ]. Despite advances in radiation therapy techniques, the development of RP remains a significant clinical challenge, leading to symptoms such as cough, dyspnea, and, in severe cases, progression to pulmonary fibrosis [ 1 , 4 ]. Advanced radiotherapy techniques, such as stereotactic body radiotherapy, have been employed to minimize the volume of healthy lung tissue exposed to radiation and, consequently, reduce the risk of RP [ 5 , 6 ]. Despite the clinical significance of RP, the precise cellular and molecular mechanisms underlying RP are not fully understood. The pathogenesis of RP involves a complex interplay between direct radiation-induced damage to lung tissues and a subsequent inflammatory response. Ionizing radiation leads to cellular damage and the generation of reactive oxygen species (ROS), causing oxidative stress (OS) [ 1 , 7 ]. Damage of alveolar type I and II (AT1/AT2) results in loss of surfactant and exudation of serum proteins into the alveoli, which were not visible under light microscopy, and there were no signs of radiographic or clinical injury [ 1 ]. Activates a cascade of inflammatory cytokines from damaged lung cells, including tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and ligands of the CXCR1/2 , attract inflammatory cells to the alveoli and pulmonary interstitium, inducing acute pneumonia [ 8 , 9 ]. Patients with unilateral radiotherapy had a marked increase in lymphocytes (mainly CD4 + T cells) in both lungs, and this effect was more pronounced in patients with clinically significant RP [ 1 , 10 ]. Furthermore, proliferating and activated CD8 + T cells in radiation mice were more than double those in control, suggesting that lymphocytes may mediate hypersensitivity reactions to RP [ 10 , 11 ]. Understanding the mechanisms underlying RP and identifying effective prevention and early treatment strategies are crucial for improving the quality of life and outcomes for patients undergoing thoracic radiotherapy. The discovery of novel biomarkers and potential therapeutic targets by highlighting genes and pathways uniquely upregulated in radiation condition exposures is a critical step in intervening in RP. Recent advances in single-cell RNA sequencing (scRNA-seq) have revolutionized our understanding of complex tissues by enabling the profiling of gene expression at the resolution of individual cells [ 12 , 13 ]. With the help of scRNA-seq, previous studies explored the mechanisms that lead to radiation-induced pulmonary fibrosis and portrayed cellular senescence in radiation-induced lung injury [ 8 , 11 , 14 ]. Zhou et al. revealed radiosensitivity heterogeneity of lung cell types at the initial of post-radiation but with limited time points and sample sizes [ 15 ]. Nevertheless, to date, no single cell-based efforts have been aimed at studying the multi-point dynamics of RP in the early stages of acute lung injury (ALI). Profiling individual cells from lung tissues at multiple time points post-radiation exposure, enables us to identify key cell types and pathways involved in the initiation and progression of RP, paving the way for the development of a single-cell dynamic landscape of RP, which offers a comprehensive map of cellular states and interactions at different stages of RP. In addition, ligand-receptor interactions and signaling pathways facilitate the mapping of the intercellular communication that orchestrates the immune response, epithelial damage, and tissue remodeling following radiation exposure [ 16 ]. In this study, we provided a comprehensive view of the cellular and molecular architecture, leading to more precise interventions aimed at preventing or mitigating RP. This research systematically explored the status of single cells at multiple time points before and after radiation, and constructed a dynamic architecture of cellular responses, offering unprecedented insights into the temporal changes within the lung microenvironment during RP development in rats with ALI. The development of a single-cell dynamic landscape of RP represents a significant step forward in understanding the cellular architecture and molecular events that underlie this complex condition. Results Single-cell dynamic alterations of radiation pneumonitis in rats Thoracic radiation therapy disrupts the cellular function and molecular levels of lung tissue, leading to acute injury, pneumonia, and possibly irreversible pulmonary fibrosis [ 1 ]. Thus, it is necessary to study the pathogenesis of RP before the development of fibrosis. In this study, we irradiated 20 rats (10 males and 10 females) with a single dose of 20 Gy to construct a pneumonitis rat model (Materials and Methods). Four rats (2 females and 2 males) were selected as the control. To explore dynamic alterations of RP at the cellular and molecular levels, we performed scRNA-seq from 12 rats across the control and IR conditions at multiple time points (Fig. 1 A). A total of 84,865 high-quality cells were obtained, of which 11,719 (13.8%) were from control rats, and 73,146 (86.2%) were from IR rats, including 16,890 (24 hours), 17,801 (1 week), 11,220 (2 weeks), 11,675 (4 weeks) and 15,560 (6 weeks) to the IR rats (Fig. 1 B). UMAP visualization after merging the datasets from per condition displayed similar distributions of cells into clusters, which indicated the absence of obvious batch effects due to sample processing (Fig. 1 B, Additional file 1: Fig. S1A-D). We cataloged all cells into 20 main cell types annotated with canonical marker genes (Fig. 1 B-D, Additional file 1: Fig. S1C), thus identifying 3 epithelial compartments (AT1, AT2 and club cells), 4 stromal compartments (2 fibroblasts and 2 endothelial cells), 8 myeloid compartments (4 macrophages, dendritic cells-DCs, monocytes, eosinophils and mast cells) and 5 lymphocyte compartments (Cd4 + T, Cd8 + T, natural killer-NK, NKT, and B cells). The most abundant cell types were observed to be T lymphocytes and myeloid cells (Additional file 1: Fig. S1E). We presented the dynamics of cell proportions affecting the different compartments after irradiation at various time points. These results pointed to changes in the proportions of myeloid and stromal cells, which tended to increase after irradiation, e.g., alveolar macrophages and lipofibroblasts (Fig. 1 E-F, Additional file 1: Fig. S1F-G, Additional file 2: Table S1). The proportion of T and B lymphocytes decreased within 24 hours after radiation (Fig. 1 E-F, Additional file 1: Fig. S1F-G), indicating that lymphocytes were sensitive to radiation, which was consistent with the previous study [ 15 ]. After that, T and B lymphocytes gradually recover (Fig. 1 F, Additional file 1: Fig. S1F-G). More detailed, aerocyte capillary (aCap) and general capillary (gCap) endothelial cells decreased rapidly within 24 hours after radiation, which may indicate damage to vascular endothelial cells, whereas gCap and two types of fibroblasts (lipofibroblasts and interstitial fibroblasts), were elevated after 1 week of radiation (Additional file 1: Fig. S1F-G). It is noteworthy that the proportion of epithelial cells, especially AT2, increased significantly within 24 hours after radiation, which was related to the decrease caused by the sensitivity of other cells to radiation (Fig. 1 E-F, Additional file 1: Fig. S1F-G, Additional file 2: Table S1), indicating that AT2 resists radiation. By comparing the immune characteristics of different cell types, we found that macrophages, especially M1 macrophages, and monocytes exhibited the strongest inflammatory response and immune effects, which was higher than that of the control (Fig. 1 G, Additional file 1: Fig. S2A-B). The inflammatory response, immune effects, and T cell-mediated cytotoxicity were found to gradually increased after radiation (Fig. 1 H, Additional file 1: Fig. S2C-E). Chest CT image and H&E staining also showed that the inflammatory properties of lung tissue gradually enhanced after radiation (Fig. 1 I, Additional file 1: Fig. S3B). In addition, we evaluated the death properties of cells after radiation. The results showed that the activity of pyroptosis and regulation of apoptosis increased gradually after radiation, and the necroptotic process was strongest at 1 week and 2 weeks (Additional file 1: Fig. S4). These results suggested that radiation stimulates the inflammatory properties and activates immune cells, resulting in a cell death effect. Vascular endothelial cells and alveolar epithelial cells have DNA damage early after 20 Gy radiation Radiation is known to cause damage to cellular DNA. To explore the effect of radiation on the degree of DNA damage in different cell types, we inferred copy number variation (CNV) of 4 major cell types using inferCNV pipeline [ 17 ], and found that stromal cells, especially endothelial cells, and epithelial cells exhibited the highest CNV score (Additional file 1: Fig. S3C-D). Radiation caused epithelial cells to rapidly damage cellular DNA in 24 hours and gradually worsen within 6 weeks (Fig. 2 A-C, Additional file 1: Fig. S5A). Studying the CNV levels of typical characteristics, we found that DNA-dependent protein kinase Prkdc , AT2 marker lamp3 and Etv5 (chromosome 11), and Sftpc (chromosome 15) underwent an increase in CNV burden within 24 hours after radiation (Fig. 2 A). During this period, EMT hallmark, hypoxia signaling, and IL6 JAK STAT3 signaling showed the highest CNV score 24 hours after radiation, indicating early damage (Fig. 2 A). In addition, the anti-proliferation factor Btg2 , the immune genes Cxcr4 and Ptprc , and mTOR signaling on chromosome 13 increased the CNV burden early after radiation (Fig. 2 A). Study have shown that Btg2 encoded protein is involved in the regulation of the G1/S transition of the cell cycle [ 18 ]. Furthermore, the DNA damage repair (DDR) genes Ercc6 , Polb , and Prkdc , which play an important role in DNA replication and repair, also showed an increase in CNV burden in the early post-radiation period (Fig. 2 A and 2 G). Indeed, our multiplex immunofluorescence (mIF) staining also confirmed that epithelial cells underwent DNA damage 24 hours after radiation, showing higher γ-H2AX levels (Fig. 2 H-I, Additional file 1: Fig. S6). Further, we collected 27 DNA damage-related pathways from MSigDB and found that most of these pathways (21/27) increased CNV burden at 24 hours after radiation, and kept elevated at the 4 and 6 weeks (Additional file 1: Fig. S7A). For example, the G2 DNA damage checkpoint serves to prevent the cell from entering mitosis (M-phase) with genomic DNA damage, was significantly increases CNV burden in the early after radiation (Additional file 1: Fig. S7A). Its associated gene, Babam2 , can encode anti-apoptotic, death receptor-associated protein, however, also showed CNV burden in the early after radiation (Fig. 2 G). Indeed, DNA damage and DDR pathways had higher activity within 24 hours after radiation, such as cellular response to DNA damage stimulus and G2 DNA damage checkpoint (Fig. 2 J-K, Additional file 1: Fig. S7B, S7D). At 1–2 weeks after radiation, cells resist damage, probably due to DNA repair (Fig. 2 A, Additional file 1: Fig. S7A). Cells respond to DNA damage by instigating robust DNA repair pathways to remove the damage physically [ 19 ]. It is worth noting that the intrinsic apoptotic signaling pathway in response to DNA damage gradually activated within 6 weeks after radiation (Additional file 1: Fig. S7D), which suggests that epithelial cells are susceptible to involvement in the early stages after 20 Gy radiation and stimulate apoptosis signaling. In addition, cell surface adhesion factors Itga4 , Itgav , Itgb6 , and AT1 marker Pdpn increased CNV score 4 weeks after radiation (Fig. 2 A). Radiation-induced alterations in endothelial cell function were thought to be key factors in organ injury through endothelial cell activation and initiation of the apoptotic pathways [ 20 ]. Our results show that the DNA damage of pulmonary vascular endothelial cells early after radiation and gradually worsened within 6 weeks (Fig. 2 D-F, Additional file 1: Fig. S8). For example, the angiogenesis-related genes Vegfa and Col3a1 , and protein secretion-related genes Napsa and Ap2s1 increased the CNV score at 24 hours after radiation (Fig. 2 D). In addition, the p53 signaling genes Gpx2 and Zfp36l1 , and vascular homeostasis gene Nostrin were reduced the CNV score at 24 hours after radiation (Fig. 2 D). These results suggest that pulmonary vascular endothelial cells exhibit DNA damage early after radiation, which was confirmed by DNA damage pathways (Additional file 1: Fig. S8A-B) and the previous study [ 1 ]. Unlike epithelial cells, in addition to having high activity at 24 hours, endothelial cells also have DNA damage pathways activated at 1 and 2 weeks after radiation (Additional file 1: Fig. S8C-D). Consistent with epithelial cells, the activity of DNA damage-induced apoptosis-related pathways in endothelial cells progressively increased within 6 weeks after radiation (Additional file 1: Fig. S8C-D). These results suggest that pulmonary endothelial and epithelial cells exhibit DNA damage in early after radiation and stimulate apoptosis signaling. Epithelial cells undergo mesenchymal transitions in RP lesions EMT plays a crucial role in chronic inflammation, tissue remodeling, and a variety of fibrotic diseases. It persists under the condition of an activated inflammatory response and eventually causes organ fibrosis [ 21 , 22 ]. To explore the EMT effects of radiation on alveolar cells, we re-clustered epithelial and stromal cells and examined the expression changes of known EMT markers across different time points after radiation (Fig. 3 A, Additional file 1: Fig. S9A-C). The expression of cadherin 1 ( Cdh1 ) was gradually increased in epithelial cells after radiation, while Vim , Cdh11 , and EMT transcription factors, Zeb1 and Zeb2 , were increased in endothelial cells and/or fibroblasts (Fig. 3 A, Additional file 1: Fig. S9B-C), suggesting that epithelial cells may undergo mesenchymal transformation. This phenomenon was also well confirmed by the signature score of known EMT hallmark ( P < 2.2e-16; Fig. 3 B-C). Furthermore, fibroblast proliferation and migration scores in epithelial and stromal cells increased gradually after radiation (Fig. 3 B-C, Additional file 1: Fig. S9D). Indeed, mIF staining also proved that the proportion of Vim + fibroblasts gradually increased after radiation, especially in the 4th and 6th weeks (Fig. 3 D-E). Transcriptional trajectory analysis reveals a dynamic transitional spectrum from AT2 cells to fibroblasts or endothelial cells after radiation (Fig. 3 F-G). The early stage of the pseudotime is dominated by AT2 cells, which are mainly cells 24 hours after radiation (Fig. 3 F, Additional file 1: Fig. S9E). After that, the trajectory extends into two branches. One is dominated by gCap and aCap, most of which from the 1 and 2 weeks after radiation; the other includes lipofibroblasts and interstitial fibroblasts, most of which from the 2 weeks after radiation (Fig. 3 F, Additional file 1: Fig. S9E). Furthermore, RNA velocity analysis showed the transition directions from AT2 cells to two types of fibroblasts (Fig. 3 G), suggesting that epithelial cells undergo mesenchymal transitions after radiation. To systematically study EMT, we sought to detect differentiation potential genes (DPGs) that are associated with radiation. We performed CytoTRACE analysis and found that Vim and Mgp were positively correlated with differentiation scores ( R > 0.6), while Ank3 , Sftpb , and Sftpa1 exhibited negative correlations ( R <-0.6; Fig. 3 H-I, Additional file 1: Fig. S9F). Then, we extracted the top 50 DPGs that positively correlated with the differentiation score as the radiation-induced EMT (EMT-RP) signature (Additional file 2: Table S2). The results showed that the EMT-RP signature score was mainly upregulated in lipofibroblasts and interstitial fibroblasts (Fig. 3 J, Additional file 1: Fig. S10C). Compared with the known hallmark EMT (EMT-H) signature and angiogenesis, we found that although the overlap accounted for less than 6%, the EMT-RP signature score showed a strong positive correlation with the EMT-H and angiogenesis scores in both cellular level ( R = 0.89, P < 2.2e-16) and the individual rat level ( R = 0.97, P = 1.1e-07; Fig. 3 J-K, Additional file 1: Fig. S10A-B). This phenomenon was also observed in all cells (Additional file 1: Fig. S10C-G). Furthermore, we compared EMT-RP and EMT-H signatures in irradiated rats and found that EMT-RP had a higher activity after irradiation compared with EMT-H, both at the single-cell level and rat individual level (Fig. 3 L, Additional file 1: Fig. S10H). To confirm this, we obtained a murine scRNA-seq dataset of the lung responses to radiation injury [ 23 ] and found that the EMT-RP signature had more obvious properties compared to the EMT-H in irradiated mice (Additional file 1: Fig. S10I-K), which is consistent with the results in rats. These findings suggest that the radiation-induced EMT signature is superior to the known hallmark EMT signature in irradiated lungs of murine, which was also well confirmed by bulk expression profile of RP mice [ 8 ] (Fig. 3 M). In summary, this study announced that epithelial cells underwent significant mesenchymal transformation after 20 Gy of radiation, and fibroblast growth, proliferation, and migration gradually increased during this process, suggesting that EMT is an indispensable part of radiation-induced lung injury. Irradiation triggers oxidative stress in monocytes within a week Oxidative stress (OS) is a molecular driver of cellular senescence and a potential contributor to a range of age-related disorders [ 24 ]. After irradiation, we found upregulation of OS defense genes in monocytes (Fig. 4 A). For example, Sod2 , a major antioxidant defense enzyme, and haptoglobin Hp , a protein with a role in OS prevention [ 25 ], were both upregulated at one week after radiation (Fig. 4 A-C, Additional file 2: Table S3). UMAP plot showed that Sod2 was mainly expressed at 1 and 2 weeks after radiation in monocytes (Fig. 4 D). Interestingly, mIF staining showed that Sod2 gradually increased after a week of IR in all cells and AT2 (Fig. 4 E-F, Additional file 1: Fig. S11). Studies have shown that OS can induce protein kinases that phosphorylates C-jun (encoded by Jun ) which combines with C-Fos (encoded by Fos ) forming the AP-1 pathway, triggering transcription involved in inflammation [ 26 , 27 ]. In monocytes, Jun and Fos were found to be significantly upregulated at 24 hours and 6 weeks after radiation (Fig. 4 A-C, Additional file 2: Table S3). In addition, SlpI , a protein affected by reactive products of oxidative metabolism [ 25 ], was also significantly upregulated at 1 week after radiation (Fig. 4 A-C). These results suggest that radiation may trigger OS in monocytes. To confirm this, we performed functional enrichment analysis on irradiated monocytes relative to control cells. The genes upregulated after radiation were found enrich in aging, pulmonary valve artery (such as abnormality of the pulmonary artery), and response to extracellular stimulus and stress-related functions (P < 0.01; Fig. 4 G). In particular, responding to radiation function was enriched by the upregulated genes 1 week after radiation (Fig. 4 G, Additional file 2: Table S4). Further analysis revealed that oxidative stress response and ROS-related functions were activated after radiation. For example, the upregulated genes one week after radiation were enriched in the pathways involved in response to ROS, antioxidant activity, and hydrogen peroxide metabolic process (Fig. 4 G). In addition, ROS metabolic process and response to OS regulation pathways were enriched by the upregulated genes at 1 and 2 weeks after radiation. These results further supported that radiation caused OS in monocytes, which was also confirmed by OS pathway activity, as almost all of them showed upregulation at 1 and 2 weeks after radiation (Additional file 1: Fig. S12A). Indeed, the pathways involved in the cellular response to OS (such as OS1, OS2, and OS4) and superoxide metabolic process (OS5) had significantly higher activity at 1 and 2 weeks after radiation (Fig. 4 D and 4 H). Considering the dynamics of monocytes across different time points after radiation (Fig. 1 F), it is not difficult to speculate that monocytes dropped sharply one week after radiation due to the aggravation of OS, which triggered the cell death program. Indeed, functional enrichment and activity comparison of OS-induced death signaling also confirmed this conjecture. For example, the genes upregulated one week after radiation were involved in the regulation of apoptotic signaling pathway (Fig. 4 G). Cell death in response to OS and apoptotic signaling pathways were also enriched by the upregulated genes after radiation (Fig. 4 G, Additional file 2: Table S4). As expected, intrinsic apoptosis signaling pathways involved in response to OS, such as OS6 and OS3, showed significantly higher activity at 1 and 2 weeks after radiation (Fig. 4 D and 4 H). Furthermore, radiation-induced upregulated genes were also involved in the inflammatory response and interferon signaling-related pathways (Additional file 1: Fig. S12B), which was consistent with the strong inflammatory activity of monocytes (Fig. 1 G). The mIF staining of inflammatory factor Il1r2 confirmed the enhancement of inflammatory response after radiation (Fig. 4 E-F). In particular, myeloid or monocyte differentiation, chemokine or receptor binding activity, and toll-like receptor production-related pathways were enriched by radiation-induced upregulated genes (Additional file 1: Fig. S12B), suggesting that the effects of radiation on monocytes contribute to the development of RP, which supports the fact that monocytes are hypersensitive to ionizing radiation and ROS [ 28 ]. Macrophages enhance the pro-inflammatory phenotype after irradiation Our results revealed that macrophage components, especially M1 macrophages, showed a stronger inflammatory response compared with other cell types (Fig. 1 G), indicating that macrophage are one of the main contributors to RP. To further investigate, we compared the dynamics of macrophage components before and after radiation and found that Bst2 + alveolar macrophages (AMs) as resident macrophages accounted for the majority of the control group, but gradually decreased after radiotherapy (Fig. 5 A, Additional file 1: Fig. S13A-C). Previous study have shown that AMs are the sentinel cells of the alveolar space, orchestrating the initiation and resolution of inflammation during ALI [ 29 ]. In contrast, the proportion of M1 macrophages and interstitial macrophages (IMs) gradually increased after radiation, with the most significant increase in Fabp5 + IMs at week 4 (Fig. 5 A), suggesting that IR chemotactic macrophages to the injury site. Transcriptional trajectory analysis demonstrated a dynamic from Bst2 + AMs to M1 macrophages after radiation (Fig. 5 B). The early stage of the pseudotime was dominated by Bst2 + and Uqcr10 + AMs, which control inflammation and maintain homeostasis [ 29 ]. Thereafter, two branches formed: one dominated by M1 macrophages and Ccr5 + IMs, which were mainly from 1 and 2 weeks after radiation, indicating an enhanced pro-inflammatory response; the other dominated by two types of IMs, mostly from 4 and 6 weeks after radiation (Fig. 5 A-B). CytoTRACE analysis showed that Bst2 + AMs were less differentiation and more developmentally early, while M1 macrophages were more differentiation and developmentally late (Fig. 5 C). Consistent with the cell trajectory, the differentiation potential of cells was highest before radiation and gradually decreased over time after radiation until new AMs ( Uqcr10 + AMs) were formed (Fig. 5 A, Additional file 1: Fig. S13D). Fabp5 plays an important role in fatty acid uptake and metabolism, and was found to be upregulated in all macrophage components after radiation (Fig. 5 D, Additional file 1: Fig. S13E, Additional file 2: Table S3), which means that the fatty acid metabolism by macrophages is increased, suggesting that macrophages expanded after radiation [ 14 ]. Hopx , a crucial marker of specific developmental and differentiation potentials, was found to be upregulated in macrophages after radiation (Fig. 5 D), indicating the development of macrophages. Further analysis revealed that the pro-inflammatory cytokines Il1b and Tlr2 were mainly expressed in M1 macrophages and Ccr5 + IMs and gradually increased after radiation, especially at 1 and 2 weeks (Fig. 5 D-E, Additional file 1: Fig. S13E, Additional file 2: Table S3). Interferon-gamma receptor 2 ( Ifngr2 ) can induce the activation of macrophages [ 30 ] and mainly expressed in M1 macrophages, indicating that M1 macrophages were activated. Furthermore, M1 macrophages and Ccr5 + IMs had stronger inflammatory response scores, and their proportions gradually increased after radiation (Fig. 5 A and 5 F). These results support that radiation enhances the pro-inflammatory response of macrophage components. Indeed, the macrophage components had a stronger inflammatory response at 1 and 2 weeks after radiation, which was confirmed by the activity of immune pathways and receptors (Fig. 5 G-H). Macrophage components exhibit pro-inflammatory properties through MIF signaling To further elucidate the pro-inflammatory properties of macrophages, we performed unsupervised cluster analysis based on immune pathways/receptors activity and classified macrophage components into three categories (Fig. 6 A). The first category includes Fabp5 + IMs, Bst2 + and Uqcr10 + AMs, which activate the cytokines, interleukins, and TNF family members-related pathways (Fig. 6 A). The second category includes Ccr5 + IMs and M1 macrophages, which upregulated immune pathway receptors, such as cytokine receptors, chemokine receptors, and interferon receptors. The last one is Top2a + AMs, which upregulate NK cytotoxicity genes and show activation of immune response (Fig. 6 A, Additional file 1: Fig. S2B). These results suggest functional heterogeneity in the pro-inflammatory of macrophage components, particularly the preference for enrichment of ligand-receptor pairs. For example, the cytokine ligand Ccl6 is mainly upregulated in AMs, while the receptor Cxcr4 is upregulated in M1 macrophages and Ccr5 + IMs (Fig. 6 B). Therefore, it is necessary to dissect the pro-inflammatory mechanisms of macrophages through cell-cell interactions. We employed the CellChat program here to explore potential ligand-receptor interactions and constructed the cell-cell communication network based on 141 murine-secreted signaling pathways (691 unique ligands/receptors), of which 39 signaling pathways significantly interacted in cell types from 6 groups (Table 5). AMs were found to be signal senders (source) in cell-cell interactions, whereas M1 macrophages and IMs tend to be receivers (target, Additional file 1: Fig. S15B), which was consistent with the results of immune receptor (Fig. 6 A-B). In addition, epithelial cells and stromal cells are more likely to be senders, while DCs and other myeloid cells tend to participate in cell interactions as receivers (Additional file 1: Fig. S15B). Moreover, most cell types showed stronger cell interactions after radiation compared with controls, especially macrophage components (Additional file 1: Fig. S14, S15A-B). Afterward, we wondered whether specific signaling pathways contribute to radiation-induced cellular communications. The probability of cell-cell interaction in radiation and control groups was inferred based on predefined signaling pathways. Interestingly, the MIF signaling pathway was found to increase the interaction probability after radiation, especially the cell-cell communications involving myeloid macrophage components (Additional file 1: Fig. S16), which was also observed in other murine scRNA-seq data (Additional file 1: Fig. S17). As a pro-inflammatory mediator, macrophage migration inhibitory factor (MIF) has been shown to be involved in the pathogenesis of acute respiratory distress syndrome, inflammatory and autoimmune diseases [ 31 ]. The expression of MIF signaling ligand Mif , and receptor genes ( Cd74 , Cd44 , and Cxcr4 ) were upregulated in different macrophage components after radiation (Fig. 6 C). Considering the important role of inflammatory factors in RP, we combined MIF signaling with inflammatory signaling (including CCL and CXCL signaling) and summarized four unique radiation-induced communication patterns. The first pattern, at 24 hours after radiation, mainly included epithelial cell interactions with monocytes based on CXCL signaling ( Cxcl3 - Cxcr2 ) and with IMs based on MIF signaling (Fig. 6 D). Cxcl3 was found specifically expressed in epithelial cells and Cxcr2 was upregulated in monocytes (Fig. 6 E, Additional file 1: Fig. S15C). The second pattern, at 1 and 2 weeks, was dominated by the interaction of epithelial and/or endothelial cells with M1 and/or IMs based on MIF signaling (Fig. 6 D). Moreover, the relative intensity of Mif + epithelial cells increased during this period (Fig. 6 D, 6 F, Additional file 1: Fig. S15D), which was inseparable from the recruitment of pro-inflammatory cells to participate in the immune response after EMT occurs (Fig. 1 E-F and Fig. 3 B). Mif was expressed in endothelial and epithelial cells, while Cd74 , Cd44 , and Cxcr4 were mainly upregulated in M1 and/or IMs, and all were elevated after radiation (Fig. 6 E, Additional file 1: Fig. S15E, Additional file 2: Table S3). The third pattern, at 4 weeks, mainly Cd4 + T cells interact with M1 and IMs based on MIF signaling (Fig. 6 D). At this time, the role of lymphocytes and the overall inflammatory response were obviously enhanced, and the T cells gradually recovered (Fig. 1 F and 1 H), which indicates that T cells are involved in the response to RP [ 11 ]. In addition, inflammation can induce DNA damage by releasing inflammatory cytokines [ 32 , 33 ]. Also, macrophages and T-lymphocytes release TNF-α and MIF to exacerbate DNA damage [ 34 , 35 ]. In this study, we indeed found that T cells and macrophages upregulated Mif , while AMs specifically upregulated Tnf , and that both upregulated after radiation, especially at 4 and 6 weeks (Fig. 6 E, 6 G-H, Additional file 1: Fig. S15E), which further explains DNA damage of epithelial cells at 4 and/or 6 weeks after radiation (Fig. 2 A-B). The last pattern, at 6 weeks, was dominated by the interaction of AMs with IMs based on MIF signaling (Fig. 6 D). Indeed, mIF staining also confirmed that Mif + macrophages gradually increased after radiation and reached a maximum at week 6 (Fig. 6 G-H). We also demonstrated that Mif was upregulated after radiation using additional murine datasets [ 8 , 23 ] in both single-cell and bulk data (Fig. 6 I-J). Furthermore, we constructed an IR rat model treated with ISO-1 (MIF inhibitor) and found that the capillary swelling, alveolar exudate, and inflammatory score were reversed (Fig. 6 K-L, Additional file 1: Fig. S15F). In particular, after treatment with ISO-1, Mif + macrophages were significantly reduced (Fig. 6 L-M). Taken together, these results suggest that MIF signaling plays a crucial role in radiation-induced inflammatory responses, with blocking this signaling could reduce the risk of RP. Dynamic changes in lymphocytes during radiation-induced lung injury Lymphocytes play an indispensable role in the development of RP [ 36 , 37 ]. Here, we re-clustered lymphocytes and divided them into 18 subsets, including 7 Cd4 + T subsets (naïve T, Tcm, Tfh, Th2, Treg, Cd69 + Trm and Cd103 + Trm cells), 5 Cd8 + T subsets (naïve T, Tem, Teff, Crtam + Trm and NKT cells), 5 B cell subsets (naïve B, proB, gcB, folB and plasma cells) and 1 NK subset (Additional file 1: Fig. S18A, Fig. S19A). Naïve T and naïve B cells gradually decreased (Additional file 1: Fig. S18B), which indicate that naïve cells differentiated into activated lymphocytes after radiation [ 14 ]. Among Cd4 + T cells, Tcm cells with high differentiation potential increased rapidly at 24 hours after radiation (Additional file 1: Fig. S18B, Fig. S19D-E). In contrast, Tfh cells gradually increased after 24 hours of radiation, and Cd69 + Trm cells proliferated at 6 weeks. This leads us to speculate that radiation causes naïve T cells to differentiate into Tfh and Cd69 + Trm cells via Tcm cells. To confirm this conjecture, we performed transcriptional trajectory analysis for Cd4 + T subset and revealed a dynamic differentiation spectrum from naïve T cells to Tfh and Trm cells (Additional file 1: Fig. S18C-D, Fig. S20B-C). The beginning phase of the pseudotime was dominated by naïve T cells, which were mainly in the control and the early stages (H24 and W1) after radiation (Additional file 1: Fig. S18D-E). After that, naïve T cells differentiate into three functional cell types via Tcm cells. The first one was dominated by Tfh and Th2, most of which belong to cells at 1 to 4 weeks after radiation (Additional file 1: Fig. S18D). During this period, the proportion of Cd8 + Teff and NK cells increased obviously (Additional file 1: Fig. S18B), which supports that inflammatory environment induced activation of Cd8 + T cells [ 11 , 14 ]. The second one, Treg cells, increased slightly within 1 to 4 weeks after radiation (Additional file 1: Fig. S18B, S18D), indicating immune regulation in RP. The last one, Cd69 + Trm, distributed at the end of the differentiation trajectory, and most of which were in the 6 weeks after radiation (Additional file 1: Fig. S18D-E), indicating that radiation transforms T cells into Cd69 + Trm cells with memory and differentiation functions and resides in tissues to play a secondary immune role [ 38 , 39 ]. For Cd8 + T cells, we noticed that the proportion of Teff cells increased from 1 to 2 weeks after radiation (Additional file 1: Fig. S18B, Additional file 2: Table S6), which was consistent with the activity of T cell-mediated cytotoxicity during this period (Additional file 1: Fig. S2E). Transcriptional trajectory analysis reveals the dynamics of differentiation from naïve T cells to NKT cells and Teff cells after radiation (Additional file 1: Fig. S18F-H, Fig. S20D-F). The beginning phase of the pseudotime was dominated by naïve T cells, mainly from the control group (Additional file 1: Fig. S18G, Fig. S20E-F). After that, the trajectory extends into two branches. One was dominated by Teff cells, most of which belong to cells at 1 to 4 weeks after radiation; and the other one was dominated by Tem cells, most of which were from 6 weeks after radiation (Additional file 1: Fig. S18G). In addition, several Cd8 + T cells differentiated into Tem cells with memory function at 6 weeks after radiation (Additional file 1: Fig. S18B, S18G). Analysis of B lymphocytes alone showed an increase in folB and proB cells from 1 to 2 weeks after radiation (Additional file 1: Fig. S18I, Fig. S20G-H), indicating B cell development and differentiation [ 14 ], which is consistent with the strong immune response at this time. Differentiation trajectory and RNA velocity analysis showed that naïve B cells and proB cells dominated in the control and early stage, while they gradually differentiated into folB cells at 1 to 2 weeks after radiation (Additional file 1: Fig. S18J-K, Fig. S20I), further confirming that lymphocyte-mediated immune activity was enhanced 1 to 2 weeks after radiation. Taken together, these results indicate that radiation activates naïve T/B cells to differentiate into functional lymphocytes, participate in immune responses and response to RP, and that several naïve T cells differentiate into T cells with memory functions and reside in tissues. Discussion In this study, we constructed a single-cell dynamic landscape of RP in rats with ALI under 20 Gy dose and characterized in detail the cell type responses to radiation, particularly revealing the damage to vascular endothelial cells and alveolar epithelial cells at the molecular level. Our study systematically delineated epithelial-mesenchymal transitions, monocyte oxidative stress, and pro-inflammatory properties of macrophages at various time points after radiation. The development of a single-cell dynamic architecture of RP provides a unique and highly detailed understanding of the cellular and molecular mechanisms driving the onset and progression of RP. This research offers critical insights into the complex interactions between epithelial, stromal cells, and immune, especially macrophages, following radiation-induced lung injury, contributing to our understanding of how damage, inflammation, and apoptosis are orchestrated at the cellular level. In the early stages of ALI (24 hours) after radiation, epithelial cells underwent obvious damage, both from the single-cell molecular level characterization and γH2AX immunofluorescence verification. This phenomenon was evidenced by genomic aberrations and transcriptomic activities of DNA damage-related pathways, particularly activation of DNA damage repair pathways. The role of AT2 was underscored in the progression of RP, and is known to play a key role in maintaining lung homeostasis and repairing damaged tissue [ 40 , 41 ]. The single-cell architecture revealed that AT2 undergoes significant stress following radiation exposure, with some subsets transitioning into mesenchymal states, contributing to the formation of fibroblasts. In addition to providing valuable mechanistic insights, this single-cell dynamics of RP in rats also highlights potential biomarkers for early detection of the disease. The identification of specific signature in epithelial cells during the early phases of RP may offer new tools for diagnosing radiation-induced lung injury before irreversible damage occurs. We identified a radiation-induced EMT signature that was highly correlated with the hallmark EMT at both the cellular and individual levels in RP rats as well as in lung cancer patients, indicating conservation of irradiated EMT signature. The EMT signature might be used to monitor patients receiving thoracic radiotherapy and enable timely interventions to prevent the progression of RP to chronic fibrosis. One of the key findings from this research is the identification of distinct macrophage subpopulations that appear to play crucial roles in different phases of RP. Pro-inflammatory macrophages, M1 macrophages and Ccr5 + IMs, characterized by high levels of cytokine production, were shown to dominate at 1 and 2 weeks after radiation, contributing to ALI through the promotion of inflammation. This dynamic switch between macrophage phenotypes highlights the importance of immune regulation in both the onset and progress of RP, suggesting that therapeutic interventions aimed at modulating macrophage function could mitigate disease severity. Furthermore, the mapping of ligand-receptor interactions revealed four intercellular communication patterns based on a key signaling pathway, MIF signaling, which mediates pro-inflammatory response (Fig. 6 D). The identification of this pathway opens new avenues for targeted therapies aimed at interrupting the signaling cascades that drive disease progression. For example, inhibitors of MIF signaling could be explored to reduce RP and improve outcomes in patients at risk of developing chronic fibrosis. During this phase, another myeloid cell type, monocytes, showed an activated OS response. A typical feature was that OS genes, such as Sod2 , Hp , and SlpI , were significantly upregulated 1 week after radiation in monocytes. In addition, the upregulated genes at 1 and 2 weeks after radiation were significantly enriched in OS response-related pathways, including regulation of response to oxidative stress pathway, ROS, and cell death in response to oxidative stress (Fig. 4 G). Indeed, these pathways had significantly higher activity at 1 and 2 weeks after radiation than the other groups. These results suggest that future studies could reduce or alleviate radiation pneumonitis by targeting OS, for example in combination with fullerenol therapy [ 42 ]. This finding highlights the progression of RP from an acute damage phase to an inflammatory and points to the potential for targeting myeloid cell stress responses as a therapeutic approach. Lymphocytes, particularly Cd4 + T cells, play a central role in orchestrating the immune response, and their activation, differentiation, and migration are key to understanding the balance between inflammation and tissue repair. Specifically, Cd4 + T cells interact with macrophage components through MIF signaling, enhancing pro-inflammatory properties after radiation (Fig. 6 D-E). Radiation exposure induces DNA damage and OS, triggering the release of cytokines and chemokines that recruit lymphocytes to the site of injury. Among these, Cd4 + Tfh cells and Cd8 + Teff cells are prominently involved in the inflammatory response. As the injury evolves, the balance may shift towards those two types of T cells, which act to promote inflammation and suppress tissue repair. In addition, activated Cd4 + T cells differentiate into various subtypes, such as Th2 and Treg cells, each with distinct roles in modulating the immune response, limiting excessive immune activation [ 43 , 44 ]. Afterwards, some Cd4 + T cells differentiated into Cd69 + Trm cells with memory function and resided in the lung tissues at 6 weeks after radiation. While this study represents a significant advance in our understanding of RP, there are important limitations to consider, such as species-specific differences. Future studies should aim to validate these findings in human tissues, using similar single-cell technologies to create a comprehensive atlas of RP in patients undergoing radiotherapy. Moreover, additional techniques, such as scATAC-seq and spatial transcriptomics, could further enhance our understanding of the RP architecture and spatially organized cellular interactions, separately. T/B cell receptor sequencing could clearly analyze the differentiation and expansion of T/B lymphocytes. In conclusion, this study provides a comprehensive single-cell dynamic landscape of RP in rats with ALI, offering valuable insights into the cellular and molecular mechanisms underlying this complex condition. By identifying key cell populations including immune, epithelial, and stromal involved in RP, as well as the signaling pathways that drive inflammation, this research opens new avenues for devising targeted therapeutic strategies to prevent and treat RP, ultimately improving outcomes for patients undergoing thoracic radiotherapy. Methods Rats and ethics statement Twelve 6-week-old female SD rats (weight: 160 ± 10g) and twelve 6-week-old male SD rats (weight: 190 ± 10g) purchased from Guangzhou Ruige Biotechnology Co. Ltd (Guangzhou, China) were housed in the animal center of Guangzhou Medical University, and all rats were conditionally reared in the animal center for 2 weeks. All rats were raised in a clean environment with a consistent dark-light schedule (lights on from 7 a.m. to 7 p.m.). This study included 24 rats, 20 of which (10 female SD rats and 10 male SD rats), which received a single double lung radiation at a dose of 20 Gy, and the other 4 (two female SD rats and two male SD rats) were randomly selected as negative controls. Four control rats (Crl) and 20 irradiated rats were sacrificed at 24 hours (H24, n = 4), 1 week (W1, n = 4), 2 weeks (W2, n = 4), 4 weeks (W4, n = 4), and 6 weeks (W6, n = 4) after radiation and consequently, lung tissues were harvested intact. The animal experimental were specifically approved by the ethics committee of Guangzhou Medical University (G2023-781) in compliance with the international guidelines. Radiation induced pneumonitis rat model Twenty-four rats (12 female SD rats and 12 male SD rats) were sedated with an intraperitoneal injection of 0.3% sodium pentobarbital (40 mg/kg body weight), fixed in the supine position and placed vertically at 500 mm from the electron beam. The rats were then subjected to bilateral thorax radiation (from the clavicle to the lower margin of the costal arch) with a single dose of 20 Gy at 8-week-old utilizing the 4.5-MeV linear electron accelerator facility (VARIAN Trilogy) to induce lung injury. The lungs were imaged using computed tomography (CT) (Aquilion ONE TSX-301C, Canon, Japan) before being sacrificed. Rats were anesthetized with 0.3% sodium pentobarbital (40 mg/kg body weight) and kept in the supine position. All imaging analyses were performed independently by two radiologists. Twenty-four rats (12 female SD rats and 12 male SD rats) were sedated with an intraperitoneal injection of 0.3% sodium pentobarbital (40 mg/kg body weight), fixed in the supine position and placed vertically at 500 mm from the electron beam. The rats were then subjected to bilateral thorax radiation (from the clavicle to the lower margin of the costal arch) with a single dose of 20 Gy at 8-week-old utilizing the 4.5-MeV linear electron accelerator facility (VARIAN Trilogy) to induce lung injury. The lungs were imaged using computed tomography (CT) (Aquilion ONE TSX-301C, Canon, Japan) before being sacrificed. Rats were anesthetized with 0.3% sodium pentobarbital (40 mg/kg body weight) and kept in the supine position. All imaging analyses were performed independently by two radiologists. Single-cell suspension preparation Ten irradiated rats at 24 hours, 1, 2, 4 and 6 weeks (two rats in each group) after radiation and two control rats were selected for preparation (a total of 12 rats, 1 female and 1 male in each group), and lung tissues were extracted for conducting scRNA-seq. Lung tissues were removed and washed in a 6-well plate containing pre-chilled PBS to remove necrotic tissue, blood, and other impurities. Transfer the tissue to a 1.5 mL centrifuge tube designed for low binding, add 1 mL of digestion solution, and mince the tissue using scissors. Place the tube on a rotary shaker in a 37°C incubator and rotate at approximately 20 rpm for 10 minutes. Remove the centrifuge tube, pipette up and down 30 times, and place it back on the rotary shaker in the 37°C incubator for an additional 10 minutes of digestion. After digestion, remove the centrifuge tube and pipette up and down 30 times before performing AOPI quality control (1:1). Sieve the tissue using a 70 µm cell strainer and centrifuge at 400G for 5 min at 4°C. Discard the filtrate and resuspend the sediment in 1 mL of 1640 medium before performing AOPI quality control (1:1). Add 3 mL of trypan blue solution and incubate for 2 minutes. Terminate trypan blue staining by adding 4 mL of 1640 medium. Centrifuge at 400G for 5 minutes at 4°C. Mix the precipitate in 1640 medium and adjust the concentration to 600–1500 cells/µl. Single-cell suspension preparation Ten irradiated rats at 24 hours, 1, 2, 4 and 6 weeks (two rats in each group) after radiation and two control rats were selected for preparation (a total of 12 rats, 1 female and 1 male in each group), and lung tissues were extracted for conducting scRNA-seq. Lung tissues were removed and washed in a 6-well plate containing pre-chilled PBS to remove necrotic tissue, blood, and other impurities. Transfer the tissue to a 1.5 mL centrifuge tube designed for low binding, add 1 mL of digestion solution, and mince the tissue using scissors. Place the tube on a rotary shaker in a 37°C incubator and rotate at approximately 20 rpm for 10 minutes. Remove the centrifuge tube, pipette up and down 30 times, and place it back on the rotary shaker in the 37°C incubator for an additional 10 minutes of digestion. After digestion, remove the centrifuge tube and pipette up and down 30 times before performing AOPI quality control (1:1). Sieve the tissue using a 70 µm cell strainer and centrifuge at 400G for 5 min at 4°C. Discard the filtrate and resuspend the sediment in 1 mL of 1640 medium before performing AOPI quality control (1:1). Add 3 mL of trypan blue solution and incubate for 2 minutes. Terminate trypan blue staining by adding 4 mL of 1640 medium. Centrifuge at 400G for 5 minutes at 4°C. Mix the precipitate in 1640 medium and adjust the concentration to 600–1500 cells/µl. Library preparation and 10x genomics single-cell RNA sequencing Cells were filtered through a 30um filter and tagged with the 10x Genomic single cell library platform following the manufacturer’s instructions. Briefly, the cell suspension was introduced into a microfluidic chip equipped with 3' chemistry and then barcoded using the 10x Chromium Controller (10x Genomics). Subsequently, RNA from the barcoded cells was reverse transcribed and sequencing libraries were prepared using reagents in the Chromium Single Cell 3' v2 Kit (10x Genomics) and following the manufacturer's guidelines. Purified cDNA libraries were sequenced on an Illumina NovaSeq 6000 platform with paired-end mode at CHI BIOTECH CO., LTD (Shenzhen, China) according to the protocol provided by Illumina. Cell calling and Quality control of scRNA-seq data Raw read files were processed with Cell Ranger 7.1.0 using default mapping arguments. Reads were mapped to the rattus norvegicus (Norway rat) genome assembly and counted with mRatBN7.2 annotations, and counted the unique molecular identifier (UMI) (by using ‘‘cellranger count’’ function). As a result, a digital cell by gene expression matrix was generated, containing the number of UMIs for each gene detected in each cell. In addition, we took some steps to filter out poor quality data. First, we removed cells with high mitochondrial gene expression because dead cells often exhibit extensive mitochondrial contamination [ 45 ]. Specifically, we fit the expression level of mitochondrial genes by using a median-centered median absolute deviation (MAD)-variance normal distribution, and then removed the cells with significantly higher expression levels than expected (determined by Benjamini-Hochberg corrected FDR < 0.01) [ 46 ]. Second, we removed cells for which less than 500 genes were detected. Third, we identified and removed potential doublets by using DoubletFinder, using 92.5th percentile of the doublet score as cutoff [ 47 ]. In the end, we retained a total of 84,865 high-quality single cells for 12 rat samples. Normalization, clustering and visualization The processed whole gene expression matrix with all selected cells was fed to R package Seurat (v5.0.3) for downstream analyses [ 48 ]. Briefly, only genes expressed in more than 5 cells were kept, and the raw UMI count matrix was log-normalized with library sizes of each cell and scaled to 10,000 using the "LogNormalize" function. Based on the normalized gene expression matrix, 2,000 highly variable genes were identified by using the “FindVariableFeatures” function with the “vst” method. “ScaleData” function was then used to scale and center the gene expression matrix after regressing out the heterogeneity associated with the mitochondrial contamination and UMI count. Unsupervised clustering was done by constructing the shared nearest neighbor (SNN) graph by using “FindNeighbors” function from the R package Seurat with Louvain algorithm. The top 30 principal components were considered and the resolution was 1.6 for whole dataset of each cell type. The first two dimensions of uniform manifold approximation and projection (UMAP) was calculated using the "RunUMAP" function. Cell type annotation and differential expression analysis A differential expression test among clusters was then applied with the "FindAllMarkers" function (use default parameters but set min.pct as 0.5) from Seurat. We used two complementary approaches to annotate the identities of different cell clusters: (1) applied canonical markers; (2) we checked whether the well-studied marker genes of different cell types were in the top rank of differential expressed genes of query cluster and then assigned the most likely identity for each cell cluster. Signature score was calculated for selected gene-sets, such as immune pathways and receptors from the ImmPort database ( https://immport.org/shared/home ) and specific signatures from the molecular signatures database (MSigDB) mouse collections ( https://www.gsea-msigdb.org/gsea/msigdb/ ), using “AddModuleScore_UCell” function from “UCell” R package. The UCell score was used to estimate the population of cells in a dataset representing the mixed expression level of the gene-set. Pathway enrichment analysis For a given cell type, we performed pathway enrichment analysis using ActivePathways (v.2.0.3) R package [ 49 ] for all terms from MSigDB (mouse collection M5) based on the differentially expressed genes (DEGs) in radiation rats at different time points compared with normal rats. Terms with a gene set shorter than 10 or longer than 500 will be removed. Terms with BH-corrected significance adjusted P ≤ 0.01 are considered to be significantly enriched by the gene set of interest and will be selected as results. The enrichment results of different radiation groups were used as input to the EnrichmentMap plugin in cytoscape (v.3.9.1) software to draw a network diagram of enrichment analysis at different time points after radiation. Single-cell copy number variation analysis The inferCNV (v.1.19.1) R package ( https://github.com/broadinstitute/inferCNV ) was used to distinguish DNA damage cells by inferring chromosomal CNVs based on the single-cell expression data. The cells from control rats as normal reference cells were used to estimate CNVs for the cell population of radiation rats. A gene ordering file containing chromosomal start and end positions of each gene from the rat mRatBN7.2 assembly was prepared for the input to the “gene_order_file” parameter in the “CreateInfercnvObject” function. The count matrix and annotation file were input to create the infercnv object, which was used to perform infercnv operations, and then run inferCNV with cutoff = 0.1 in “run” function. RNA velocity, CytoTRACE and pseudotime analysis To analyze the transcriptional dynamics in distinct cell subsets, we applied the velocyto python package (v.0.17.17) to estimate the RNA velocity of single cells by distinguishing spliced and un-spliced mRNAs [ 50 ]. We first generated the individual loom file using “velocyto run10x” command based on the output file of CellRanger for each rat sample and then merged all loom files together. Then we fed the merged loom files and the UMAP coordinates of single cells generated by Seurat into velocyto and followed its analysis steps to finally project the RNA velocity vectors onto low-dimension embeddings. The R package CytoTRACE (v.0.3.3) [ 51 ] was applied to predict the differentiation state of cells from the scRNA-seq profiles. Analysis of differentiation trajectories of macrophages, lymphocytes, epithelial and stromal cells was performed using both Monocle 2 and Monocle 3 [ 52 ] by inferring the pseudotemporal ordering of cells according to their transcriptional similarity. To analyze the transcriptional dynamics in distinct cell subsets, we applied the velocyto python package (v.0.17.17) to estimate the RNA velocity of single cells by distinguishing spliced and un-spliced mRNAs [ 50 ]. We first generated the individual loom file using “velocyto run10x” command based on the output file of CellRanger for each rat sample and then merged all loom files together. Then we fed the merged loom files and the UMAP coordinates of single cells generated by Seurat into velocyto and followed its analysis steps to finally project the RNA velocity vectors onto low-dimension embeddings. The R package CytoTRACE (v.0.3.3) [ 51 ] was applied to predict the differentiation state of cells from the scRNA-seq profiles. Analysis of differentiation trajectories of macrophages, lymphocytes, epithelial and stromal cells was performed using both Monocle 2 and Monocle 3 [ 52 ] by inferring the pseudotemporal ordering of cells according to their transcriptional similarity. Analysis of cell-cell communication The CellChat [ 53 ] was used to investigate cell-cell communication between different cell types on the expression matrix. We extracted multiple mouse ligand-receptor resources of secreted signaling based on CellChat database that manual curated from KEGG base and primary literature [ 53 ]. In total, 1,209 ligand-receptor pairs were obtained in this study. Statistical analysis of intercellular communications was done permuting the label of cell type for each cell at 100 times (by default) to test the significance of each interaction pair. The communication probability from one cell type to the other one for a particular ligand-receptor pair were described in the original text [ 53 ]. The significant ligand-receptor pairs with P ≤ 0.05 were determined significant interaction. Acquisition and analysis of radiation murine public datasets The scRNA-seq data of the lung responses to radiation injury (20 mice, including 5 non-IR mice as control, 5 mice after 10 Gy thorax IR and 10 mice after 17 Gy thorax IR) was obtained from a previous study [ 23 ]. The gene expression data from the irradiated lungs of murine (n = 18) was downloaded from the Gene Expression Omnibus (GEO) database (GSE85359) [ 8 ]. Signature scores for epithelial-mesenchymal transition (EMT) and angiogenesis hallmarks from MSigDB [ 54 ] were calculated using single sample gene set enrichment analysis (ssGSEA) method in R package GSVA [ 55 ]. The newly discovered radiation-induced EMT signature in this study was consisted of the top 50 genes that were positively correlated with the differentiation score based on CytoTRACE analysis. The scRNA-seq data of the lung responses to radiation injury (20 mice, including 5 non-IR mice as control, 5 mice after 10 Gy thorax IR and 10 mice after 17 Gy thorax IR) was obtained from a previous study [ 23 ]. The gene expression data from the irradiated lungs of murine (n = 18) was downloaded from the Gene Expression Omnibus (GEO) database (GSE85359) [ 8 ]. Signature scores for epithelial-mesenchymal transition (EMT) and angiogenesis hallmarks from MSigDB [ 54 ] were calculated using single sample gene set enrichment analysis (ssGSEA) method in R package GSVA [ 55 ]. The newly discovered radiation-induced EMT signature in this study was consisted of the top 50 genes that were positively correlated with the differentiation score based on CytoTRACE analysis. H&E and immunofluorescence staining The tissue specimens were obtained from SD rats (n = 24) with the portion of the selected tissue were embedded in paraffin and stained for hematoxylin-eosin (H&E) staining. Briefly, in order to label the nuclear and cytoplasm, tissue sections on glass slides were rehydrated with xylene and alcohol, and then counterstained with hematoxylin (MACKLIN, China) and eosin (SCR, China). Tissue samples were observed under a microscope (BX53, Olympus, Japan). All histological analyses were performed with at least five vision per rat by two pathologists independently. For immunofluorescence staining (IF), a TSAPLus triple fluorescent staining kit (Wuhan Servicebio®, Wuhan, China) was used. Rat lung sections were dewaxed to water, Tris-EDTA (PH 9.0, Sangon, China) was used to antigen repaired, 3% H2O2 was incubated at room temperature in the dark for 25 min to block endogenous peroxidase and reduce non-specific background staining, blocked for 30 min in 5% fetal goat serum and subsequently, incubated with primary antibody (anti-IL1R2 [1/200 dilution; santa cruz; USA], anti-SOD2 [1/500 dilution; proteintech; China], anti-SFTPC [1/500 dilution; proteintech; China], anti-γ-H2AX [1/1000 dilution; abcam; USA], anti-MIF [1/500 dilution; proteintech; China], anti-CD68 [1/500 dilution; proteintech; China], anti-GSN [1/500 dilution; proteintech; China], anti-VIMNETIN [1/500 dilution; proteintech; China]) at 4℃ overnight, (Alexa Fluor® 488 for IL1R2, SFTPC, CD68, VIMNETIN and AlexaFluor594® for SOD2, γ-H2AX, MIF, GSN respectively), and counterstained with secondary antibodies (HRP-Conjugated Goat Anti-Rabbit IgG and HRP-Conjugated Goat Anti-Mouse IgG) for 1 h at room temperature. Nuclear was labeled with 4’6-diamidino-2-phenylindole (DAPI). To further verify the role of MIF in radiation pneumonitis, we constructed an IR (20 Gy) rat model treated with MIF inhibitor (ISO-1, IP 13mg/kg daily, for 2 weeks). Two weeks after the end of radiation, lung tissues were obtained for H&E staining and multiplex immunofluorescence staining. Statistical analysis Comparison between two groups was carried out by two-sided Wilcoxon rank-sum test. Comparison among multiple groups was performed with Kruskal-Wallis test. The correlation between two continuous variables was measured by Pearson correlation coefficient. The significance level of two discrete variables was determined by Fisher's exact test. P value less than 0.05 was considered statistically significant. All statistical analysis were conducted using R software (version 4.2.3, http://www.r-project.org ). Declarations Author’s contributions X.Y.S., Z.S., and G.M.L. conceived and designed the study. X.Y.S., Z.S. and W.W. supervised the study. G.M.L. and C.J.L. performed data curation and investigations. G.M.L. performed single-cell data analysis and result visualization. G.M.L., C.J.L. and T.C. designed and implemented the experimental verification protocol. G.M.L. and C.J.L. designed the statistical analysis plan and G.M.L. performed the statistical analysis. All authors contributed to the acquisition, analysis, verification, or interpretation of data. G.M.L. and C.J.L. drafted the manuscript. All authors revised the manuscript and gave final approval of the version to submission. X.Y.S., Z.S. and W.W. contributed equally to this work and are joint corresponding authors. Funding This study was supported by the National Natural Science Foundation of China (Grant No. 82272996, 82403816), the China Postdoctoral Foundation (Grant No. 2023M740846), the Postdoctoral Fellowship Program of CPSF (Grant No. GZB20230180), the Science and Technology Program of Guangzhou (Grant No. 202206010081), the Key Specialty Construction Project of Guangzhou Medical University (Grant No. LZ202100302), and the Wu Jieping Medical Foundation (Grant No. 202001301). Data availability The scRNA-seq raw data and processed data generated in this study have been deposited in the Gene Expression Omnibus (GEO) repository, with the accession code GSE286896. Other data of this study can be available from the corresponding author upon reasonable request. Standard workflows and open-source R packages and software were used in this study (Materials and methods). No previously unreported custom code was used or developed for the analyses presented in this study. All codes of this study are available from the authors upon reasonable request. Ethics declaration The animal experimental were specifically approved by the ethics committee of Guangzhou Medical University (G2023-781) in compliance with the international guidelines. Consent for publication Not applicable. Conflict of interest The authors declare no competing interests. References Hanania, A.N., et al., Radiation-Induced Lung Injury: Assessment and Management. Chest, 2019. 156 (1): p. 150-162. Voruganti Maddali, I.S., et al., Optimal management of radiation pneumonitis: Findings of an international Delphi consensus study. Lung Cancer, 2024. 192 : p. 107822. Li, F., et al., Risk factors for radiation pneumonitis in lung cancer patients with subclinical interstitial lung disease after thoracic radiation therapy. Radiat Oncol, 2021. 16 (1): p. 70. Kuipers, M.E., et al., Predicting Radiation-Induced Lung Injury in Patients With Lung Cancer: Challenges and Opportunities. Int J Radiat Oncol Biol Phys, 2024. 118 (3): p. 639-649. Weiss, A., et al., Prediction of radiation pneumonitis using the effective alpha/beta of lungs and heart in NSCLC patients treated with proton beam therapy. Radiother Oncol, 2024. 190 : p. 110013. Marks, L.B., et al., Radiation-induced lung injury. Semin Radiat Oncol, 2003. 13 (3): p. 333-45. Pohjoismaki, J.L.O. and S. Goffart, Adaptive and Pathological Outcomes of Radiation Stress-Induced Redox Signaling. Antioxid Redox Signal, 2022. 37 (4-6): p. 336-348. Curras-Alonso, S., et al., An interactive murine single-cell atlas of the lung responses to radiation injury. Nat Commun, 2023. 14 (1): p. 2445. Kasmann, L., et al., Radiation-induced lung toxicity - cellular and molecular mechanisms of pathogenesis, management, and literature review. Radiat Oncol, 2020. 15 (1): p. 214. Roberts, C.M., et al., Radiation pneumonitis: a possible lymphocyte-mediated hypersensitivity reaction. Ann Intern Med, 1993. 118 (9): p. 696-700. Wu, F., et al., Cellular Atlas of Senescent Lineages in Radiation- or Immunotherapy-Induced Lung Injury by Single-Cell RNA-Sequencing Analysis. Int J Radiat Oncol Biol Phys, 2023. 116 (5): p. 1175-1189. Jovic, D., et al., Single-cell RNA sequencing technologies and applications: A brief overview. Clin Transl Med, 2022. 12 (3): p. e694. Travaglini, K.J., et al., A molecular cell atlas of the human lung from single-cell RNA sequencing. Nature, 2020. 587 (7835): p. 619-625. Yang, M., et al., Single-Cell Transcriptome Analysis of Radiation Pneumonitis Mice. Antioxidants (Basel), 2022. 11 (8). Ma, L., et al., A Study on the Radiosensitivity of Radiation-Induced Lung Injury at the Acute Phase Based on Single-Cell Transcriptomics. Front Immunol, 2022. 13 : p. 941976. Armingol, E., et al., Deciphering cell-cell interactions and communication from gene expression. Nat Rev Genet, 2021. 22 (2): p. 71-88. Tirosh, I., et al., Single-cell RNA-seq supports a developmental hierarchy in human oligodendroglioma. Nature, 2016. 539 (7628): p. 309-313. Wu, G.S. and C.H. Bassing, Flip the switch: BTG2-PRMT1 protein complexes antagonize pre-B-cell proliferation to promote B-cell development. Cell Mol Immunol, 2018. 15 (9): p. 808-811. Chatterjee, N. and G.C. Walker, Mechanisms of DNA damage, repair, and mutagenesis. Environ Mol Mutagen, 2017. 58 (5): p. 235-263. Wijerathne, H., et al., Mechanisms of radiation-induced endothelium damage: Emerging models and technologies. Radiother Oncol, 2021. 158 : p. 21-32. Yang, J., et al., Guidelines and definitions for research on epithelial-mesenchymal transition. Nat Rev Mol Cell Biol, 2020. 21 (6): p. 341-352. Youssef, K.K. and M.A. Nieto, Epithelial-mesenchymal transition in tissue repair and degeneration. Nat Rev Mol Cell Biol, 2024. 25 (9): p. 720-739. Jackson, I.L., et al., Gene expression profiles among murine strains segregate with distinct differences in the progression of radiation-induced lung disease. Dis Model Mech, 2017. 10 (4): p. 425-437. Jorgensen, A., et al., Oxidative Stress-Induced Damage to RNA and DNA and Mortality in Individuals with Psychiatric Illness. JAMA Psychiatry, 2024. 81 (5): p. 516-520. Majchrzak-Gorecka, M., et al., Secretory leukocyte protease inhibitor (SLPI), a multifunctional protein in the host defense response. Cytokine Growth Factor Rev, 2016. 28 : p. 79-93. Priya Dharshini, L.C., et al., Oxidative stress responsive transcription factors in cellular signalling transduction mechanisms. Cell Signal, 2020. 72 : p. 109670. Nguyen, T.M. and M. Aragona, Stress-responsive transcription factors train stem cells to remember. Cell Stem Cell, 2021. 28 (10): p. 1679-1680. Bauer, M., et al., Human monocytes are severely impaired in base and DNA double-strand break repair that renders them vulnerable to oxidative stress. Proc Natl Acad Sci U S A, 2011. 108 (52): p. 21105-10. Malainou, C., et al., Alveolar macrophages in tissue homeostasis, inflammation, and infection: evolving concepts of therapeutic targeting. J Clin Invest, 2023. 133 (19). Xu, X., et al., Phosphorylation-Mediated IFN-gammaR2 Membrane Translocation Is Required to Activate Macrophage Innate Response. Cell, 2018. 175 (5): p. 1336-1351 e17. Calandra, T. and T. Roger, Macrophage migration inhibitory factor: a regulator of innate immunity. Nat Rev Immunol, 2003. 3 (10): p. 791-800. Greten, F.R. and S.I. Grivennikov, Inflammation and Cancer: Triggers, Mechanisms, and Consequences. Immunity, 2019. 51 (1): p. 27-41. Gronke, K., et al., Interleukin-22 protects intestinal stem cells against genotoxic stress. Nature, 2019. 566 (7743): p. 249-253. Pollard, J.W., Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer, 2004. 4 (1): p. 71-8. Singh, N., et al., Inflammation and cancer. Ann Afr Med, 2019. 18 (3): p. 121-126. Teng, F., M. Li, and J. Yu, Radiation recall pneumonitis induced by PD-1/PD-L1 blockades: mechanisms and therapeutic implications. BMC Med, 2020. 18 (1): p. 275. Wirsdorfer, F. and V. Jendrossek, The Role of Lymphocytes in Radiotherapy-Induced Adverse Late Effects in the Lung. Front Immunol, 2016. 7 : p. 591. Zheng, M.Z.M. and L.M. Wakim, Tissue resident memory T cells in the respiratory tract. Mucosal Immunol, 2022. 15 (3): p. 379-388. Yenyuwadee, S., et al., The evolving role of tissue-resident memory T cells in infections and cancer. Sci Adv, 2022. 8 (33): p. eabo5871. Schneider, J.L., et al., The aging lung: Physiology, disease, and immunity. Cell, 2021. 184 (8): p. 1990-2019. Parimon, T., et al., Senescence of alveolar epithelial progenitor cells: a critical driver of lung fibrosis. Am J Physiol Cell Physiol, 2023. 325 (2): p. C483-C495. Zhang, T., et al., Fullerenols Mitigate Radiation-Induced Myocardial Injury. Adv Healthc Mater, 2023. 12 (29): p. e2300819. Walker, J.A. and A.N.J. McKenzie, T(H)2 cell development and function. Nat Rev Immunol, 2018. 18 (2): p. 121-133. Alshoubaki, Y.K., et al., Tregs delivered post-myocardial infarction adopt an injury-specific phenotype promoting cardiac repair via macrophages in mice. Nat Commun, 2024. 15 (1): p. 6480. Ilicic, T., et al., Classification of low quality cells from single-cell RNA-seq data. Genome Biol, 2016. 17 : p. 29. Pijuan-Sala, B., et al., A single-cell molecular map of mouse gastrulation and early organogenesis. Nature, 2019. 566 (7745): p. 490-495. McGinnis, C.S., L.M. Murrow, and Z.J. Gartner, DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors. Cell Syst, 2019. 8 (4): p. 329-337 e4. Hao, Y., et al., Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat Biotechnol, 2024. 42 (2): p. 293-304. Paczkowska, M., et al., Integrative pathway enrichment analysis of multivariate omics data. Nat Commun, 2020. 11 (1): p. 735. La Manno, G., et al., RNA velocity of single cells. Nature, 2018. 560 (7719): p. 494-498. Gulati, G.S., et al., Single-cell transcriptional diversity is a hallmark of developmental potential. Science, 2020. 367 (6476): p. 405-411. Cao, J., et al., The single-cell transcriptional landscape of mammalian organogenesis. Nature, 2019. 566 (7745): p. 496-502. Jin, S., et al., Inference and analysis of cell-cell communication using CellChat. Nat Commun, 2021. 12 (1): p. 1088. Liberzon, A., et al., The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst, 2015. 1 (6): p. 417-425. Hanzelmann, S., R. Castelo, and J. Guinney, GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics, 2013. 14 : p. 7. Additional Declarations The authors declare no competing interests. Supplementary Files Additionalfile1.docx Supplementary figures Additionalfile2.xlsx Supplementary tables Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7659149","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":517691052,"identity":"4f751a5f-b321-4d9f-a9bb-c85d7d1f14bb","order_by":0,"name":"Gaoming Liao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+ElEQVRIiWNgGAWjYBACPiA+/KeCTY6f4fzHBx8MbOwIamFjYGA8wHOGz1iy8YCx4YyCtGRitDAf4G2TS9zQfMBMmufDIcYGglr4VycckGwzS9zAdiDZ2MbgADMD++GjG/BqkXi74YDBuTTj7TwHDj7OMbjDx8CTlnYDv5azGw4klB2T3TnjYLNxjsEzZgYJHjPCWg6w/WfccP8xm7SFwWHGBoJa+Hs3HGxoY1PccOAYmzQDUVokeDccZjjDZizZcIbZsMcgLZmNkF/4+c9u/swAjsozjA9+/LGx42c/fAyvFgaJBHR78SoHW3OAoJJRMApGwSgY6QAAViBTcyEwYZ0AAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0005-6414-3670","institution":"The First Hospital of Guangzhou Medical University","correspondingAuthor":true,"prefix":"","firstName":"Gaoming","middleName":"","lastName":"Liao","suffix":""},{"id":517691053,"identity":"2ccbceb8-cc2a-42f8-8e8e-872f6794c25f","order_by":1,"name":"Chanjin Liang","email":"","orcid":"","institution":"The Fifth Hospital of Guangzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chanjin","middleName":"","lastName":"Liang","suffix":""},{"id":517691054,"identity":"244e0888-31f2-42ec-ab00-5fa6661d70c2","order_by":2,"name":"Ting Chen","email":"","orcid":"","institution":"The Fifth Hospital of Guangzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ting","middleName":"","lastName":"Chen","suffix":""},{"id":517691055,"identity":"5c9f4140-30dc-4718-bc4d-c295ac300a6e","order_by":3,"name":"Youqing Zhu","email":"","orcid":"","institution":"Jinan University","correspondingAuthor":false,"prefix":"","firstName":"Youqing","middleName":"","lastName":"Zhu","suffix":""},{"id":517691056,"identity":"2b7d208c-b136-4336-8213-a530b0e4c2ea","order_by":4,"name":"Sihan Tang","email":"","orcid":"","institution":"Shenzhen University","correspondingAuthor":false,"prefix":"","firstName":"Sihan","middleName":"","lastName":"Tang","suffix":""},{"id":517691057,"identity":"06870544-efbe-492c-b03e-16851436f22f","order_by":5,"name":"Hao Nie","email":"","orcid":"","institution":"The Fifth Hospital of Guangzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Nie","suffix":""},{"id":517691058,"identity":"f4862b29-20bb-4d8e-8d6e-4da7eb387027","order_by":6,"name":"Wei Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1UlEQVRIiWNgGAWjYJCCAx//SPDwQ9jMxGk5OLPBRk6ygRQtzLwNacYGB4jVYnAj+eBh3h2HEzefP50mwVBhndjAfvYAAS1pCQfnnjmcuO1G7jYJhjPpiQ08eQl4tZjdzjE48IYNpIV3mwRj2+HEBgkeAwJa8j8c4AFq2dx/FqjlH1FachgO8rYBvc8AdBhjAxFa7O8/Mzg444yNnMSN3M0WCcfSjdt4cvBrkew5/PjDhwpgVPaf3XjjQ421bD/7GfxaUEECELORoH4UjIJRMApGAQ4AACXhTcGPh9+BAAAAAElFTkSuQmCC","orcid":"","institution":"Nanfang Hospital of Southern Medical University","correspondingAuthor":true,"prefix":"","firstName":"Wei","middleName":"","lastName":"Wang","suffix":""},{"id":517691059,"identity":"c67a6995-1dd3-43d6-ad06-fa4cda072559","order_by":7,"name":"Zhi Shi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAz0lEQVRIie3OsQrCMBCA4QuCLpWu59K+QoKL4OCrtAh2qeLkKC1CJsE1oC/h5pjQNXbu4OAj6GYHQa3gmrgJ5t8O7uMOwOX6yTyQJMOAZK+hbQXepP8dged63AgrMvKPStWHQbLarilcFgX428xwBWeR7Gmc5jtNiSgLwJM0EY9Kxp9EpLTV5QVQjAzE11TGHBPyIncrAimVimPUEGJFqpSqnCPLxWSu1mXiYWUgnY3uX2u+DJkY78+3xTDwhYF8YhmAbD61LrRfdblcrn/rAc6rQdD86f+nAAAAAElFTkSuQmCC","orcid":"","institution":"Jinan University","correspondingAuthor":true,"prefix":"","firstName":"Zhi","middleName":"","lastName":"Shi","suffix":""},{"id":517691060,"identity":"e9860ff5-a308-478a-a11f-828b3be8f861","order_by":8,"name":"Xingyuan Shi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDklEQVRIie3RsUrEMBjA8YTAV4Sv3pqjal8hUig3CPcqydLpBkGQG26oHMRJXPsaLsUxUuiUBziog3JwIDhUnG7SXjeHlI7C5T9kCPkR8oUQn+8fNmHMmHb5gxBvDaF5vykGyfReq/fCsrNTkslxRFibJCeaXV2QhRhHyEamEbGAQOz3x/65VHmwLjlZvToFLWQWXS8RgT6Us6ltVI71LSf1zkkYl3VUWI7AwlJc6o7wRcppXjkJcKWjUAsEwJ1QBxJ/DhPEiiWhlgiI7O2lvwWHCQ807YZsEDik9E43icbsZiZrN5lXk7b7SjOPC7b92uvm/DGonjbtyk3+vqtfD4scBbrptSMP+nw+35H1C7foWAcDpO43AAAAAElFTkSuQmCC","orcid":"","institution":"The Fifth Hospital of Guangzhou Medical University","correspondingAuthor":true,"prefix":"","firstName":"Xingyuan","middleName":"","lastName":"Shi","suffix":""}],"badges":[],"createdAt":"2025-09-19 13:45:20","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-7659149/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7659149/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91950221,"identity":"f9d00241-9a39-4f22-a47a-c3c899fe465e","added_by":"auto","created_at":"2025-09-23 06:28:33","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5259568,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/512094081c3a6809055091b4.docx"},{"id":91949010,"identity":"91f27328-2206-43a5-9c80-b83f1d23440d","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7659149.json","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/2ec5d9439589a1e1a3bf3b80.json"},{"id":91949013,"identity":"73f1cdbc-192f-4697-81ec-264bdb2c9792","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":150188,"visible":true,"origin":"","legend":"","description":"","filename":"rs76591490enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/69a35f0f6b413325d7492f18.xml"},{"id":91950219,"identity":"a8042c80-2711-42ca-a606-7d34d7eaae66","added_by":"auto","created_at":"2025-09-23 06:28:33","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":940665,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/4011c0d3aa64ed27a4d9c960.png"},{"id":91949019,"identity":"6d032793-2e25-4587-a15a-78258fe36aed","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":727652,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/1fd32769f2d2d31145c6e5e9.png"},{"id":91949437,"identity":"3bf89022-f9ec-47b6-8d1f-926d9c8439ea","added_by":"auto","created_at":"2025-09-23 06:20:33","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":973056,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/ba21e0a1f6cd6fd27a0fcc6d.png"},{"id":91950220,"identity":"d83c71a9-cad3-4434-a782-bf61669eee17","added_by":"auto","created_at":"2025-09-23 06:28:33","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":713966,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/8ffcd96ee3c913106c4a5450.png"},{"id":91950459,"identity":"d6f8d3b4-6eb0-428b-9871-fe5b91d67d1b","added_by":"auto","created_at":"2025-09-23 06:36:33","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":724536,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/31e9d45c3ced36a95ca9723a.png"},{"id":91949438,"identity":"ef753089-83b7-4914-a6df-5ab99b06e73d","added_by":"auto","created_at":"2025-09-23 06:20:33","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":956298,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/6d6bfef27fd4772f3fd2dcd7.png"},{"id":91949032,"identity":"a9f92d8d-a9ad-4403-a857-7ad9e4a91715","added_by":"auto","created_at":"2025-09-23 06:12:34","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":181960,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/03356191d61ae6b022338197.png"},{"id":91949020,"identity":"f12c9c2c-f3a9-49c8-9775-0a25e55fbf7c","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":142417,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/ae45e58fd72f22c26e085fd2.png"},{"id":91949026,"identity":"e461dc38-383f-4bbd-ba91-4b5eff1adf41","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":217283,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/303b7b15304ca5e357dcda51.png"},{"id":91949034,"identity":"9af2b348-253b-4f27-acbe-fc7ccadd77e8","added_by":"auto","created_at":"2025-09-23 06:12:34","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148304,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/70429bb95abde9fdb87bff9d.png"},{"id":91949024,"identity":"7e21582a-2126-4829-b543-3c83de27b44b","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":179282,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/aade0579d8edb0d20c67a802.png"},{"id":91949031,"identity":"c4837039-a46b-4e8f-ada8-a5f7668e1bee","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":174649,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/889bdc6c2fc400be8113892b.png"},{"id":91949033,"identity":"ac3633fb-f5c5-485a-b713-9ce3c6e1a13d","added_by":"auto","created_at":"2025-09-23 06:12:34","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148088,"visible":true,"origin":"","legend":"","description":"","filename":"rs76591490structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/02d893764f233f27c8acf1cb.xml"},{"id":91949035,"identity":"56880a72-afcf-4900-bd17-aa9bbad68511","added_by":"auto","created_at":"2025-09-23 06:12:34","extension":"html","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":161831,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/4fe36ccb40730ea66db2c120.html"},{"id":91949011,"identity":"762434ac-5d03-443b-9a59-71fa89f80b17","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":623325,"visible":true,"origin":"","legend":"\u003cp\u003eSingle-cell global landscape after irradiation in rats.\u003cstrong\u003e A \u003c/strong\u003eSchematic diagram shows the experimental workflow of this study. \u003cstrong\u003eB\u003c/strong\u003e UMAP visualization of 84,865 high-quality cells from 12 rats (control, 24 hours, 1, 2, 4, and 6 weeks after IR) annotated by cell type. \u003cstrong\u003eC\u003c/strong\u003e The expression of the representative markers. \u003cstrong\u003eD\u003c/strong\u003e The bubble plot shows the expression of the markers used for cell type identification. \u003cstrong\u003eE-F\u003c/strong\u003e Dynamics in cell proportions of the epithelial, myeloid, lymphocytes, and stromal cells (\u003cstrong\u003eE\u003c/strong\u003e), and the cell subsets (\u003cstrong\u003eF\u003c/strong\u003e) across the control (Crl) and IR conditions at the different time points. \u003cstrong\u003eG-H\u003c/strong\u003eViolin diagram of the inflammatory response in different cell subsets (\u003cstrong\u003eG\u003c/strong\u003e) and time points (\u003cstrong\u003eH\u003c/strong\u003e). \u003cstrong\u003eI\u003c/strong\u003e Chest CT image and H\u0026amp;E staining of representative sections at control and different multiple points (24H, 1W, 2W, 4W, 6W) after IR 20Gy.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/2f2635b10eba169ef18b1159.png"},{"id":91949012,"identity":"c1f6537a-2eea-46fc-8197-f61f3453f1f9","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":677815,"visible":true,"origin":"","legend":"\u003cp\u003eDNA damage occurs in alveolar epithelial cells and vascular endothelial cells at 24 hours after IR.\u003cstrong\u003e A-C \u003c/strong\u003eInferred CNVs at control and different time points after IR in epithelial cells, including CNVs of representative genes or/and pathways on chromosomes (\u003cstrong\u003eA\u003c/strong\u003e), average CNVs distribution (\u003cstrong\u003eB\u003c/strong\u003e), and UMAP visualization (\u003cstrong\u003eC\u003c/strong\u003e). \u003cstrong\u003eD-F \u003c/strong\u003eEndothelial cells, like \u003cstrong\u003eA-C\u003c/strong\u003e. \u003cstrong\u003eG\u003c/strong\u003e The CNVs of gene level from representative pathways in epithelial cells. \u003cstrong\u003eH \u003c/strong\u003emIF staining for \u003cem\u003eSftpc\u003c/em\u003e (red), γ-H2AX (green), and DAPI (blue) in rat lung sections from control and radiation groups (24H, 1W, and 2W). The other groups in Figure S6. Arrowheads indicate the colocalization of \u003cem\u003eSftpc\u003c/em\u003e and γ-H2AX. Scale bars, 20 µm. \u003cstrong\u003eI \u003c/strong\u003eQuantification and estimation the relative intensity of the γ-H2AX\u003csup\u003e+ \u003c/sup\u003eAT2 cells in control and irradiated lung tissue sections. To compare two groups, the P value was computed with the student's t-test (two-sided test) from scipy (ns, P\u0026gt;0.05; *, P\u0026lt;0.05; **, P\u0026lt;0.01; ***, P\u0026lt;0.001). Each dot represents one analyzed image (n=5 in each group). \u003cstrong\u003eJ \u003c/strong\u003eUMAP of enrichment scores (UCell scores) of representative pathways in epithelial cells. \u003cstrong\u003eK\u003c/strong\u003e Comparison of enrichment scores distribution of representative pathways in epithelial cells among different groups.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/c6c8fd3d375c7a10d7dd3fd2.png"},{"id":91949434,"identity":"93d10eaa-5055-4b4b-8943-c7ab5606c269","added_by":"auto","created_at":"2025-09-23 06:20:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":610323,"visible":true,"origin":"","legend":"\u003cp\u003eEpithelial cells undergo mesenchymal transitions after IR.\u003cstrong\u003e A \u003c/strong\u003eThe selected genes expression of EMT in epithelial and stromal cells. \u003cstrong\u003eB \u003c/strong\u003eExpression analyses of EMT (left) and fibroblast proliferation (right) signatures. \u003cstrong\u003eC\u003c/strong\u003e UMAP of rat groups and signature scores of representative pathways (EMT and fibroblast proliferation). \u003cstrong\u003eD\u003c/strong\u003e mIF staining for \u003cem\u003eGsn\u003c/em\u003e (red), \u003cem\u003eVim\u003c/em\u003e (green), and DAPI (blue) in rat lung sections from the control and radiation groups. Arrowheads indicate the colocalization of \u003cem\u003eGsn\u003c/em\u003e and \u003cem\u003eVim\u003c/em\u003e. Scale bars, 20 µm. \u003cstrong\u003eE \u003c/strong\u003eStatistical analysis of the relative intensity of vimentin (\u003cem\u003eVim\u003c/em\u003e) in \u003cem\u003eGsn\u003c/em\u003e+ fibroblasts across control and irradiated lung sections. The P value was computed with the student's t-test (two-sided test) from scipy (ns, P\u0026gt;0.05; *, P\u0026lt;0.05; **, P\u0026lt;0.01; ***, P\u0026lt;0.001). Each dot represents one analyzed image. \u003cstrong\u003eF \u003c/strong\u003ePseudotime analysis of epithelial and stromal cells. The branched trajectory was colored by pseudotime (left) and cell subsets (right). The pie chart shows the proportion of rat groups (left) and cell subsets (right) in each branch. \u003cstrong\u003eG\u003c/strong\u003e UMAP of pseudotime (left) and RNA velocity (right). \u003cstrong\u003eH\u003c/strong\u003eThe correlations (left, Pearson method) and expression dynamics (right) of representative DPGs. \u003cstrong\u003eI\u003c/strong\u003e UMAP of representative DPGs. \u003cstrong\u003eJ\u003c/strong\u003e Left, EMT-RP signature score identified by this data. Right,\u003cstrong\u003e \u003c/strong\u003ethe overlap of EMT-RP, EMT-H, and angiogenesis. \u003cstrong\u003eK\u003c/strong\u003e Correlation analysis (Pearson method) of the EMT-RP signature with EMT-H at the single-cell level (left) and rat sample level (right). \u003cstrong\u003eL-M\u003c/strong\u003e Compare the signature score of EMT-RP and EMT-H at the single-cell (\u003cstrong\u003eL\u003c/strong\u003e) and IR rat sample level (\u003cstrong\u003eM\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/db8a2a8c8903a7cd4b39e119.png"},{"id":91949432,"identity":"47c4fbbf-bd59-4791-a702-fedfaff9958c","added_by":"auto","created_at":"2025-09-23 06:20:33","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":538497,"visible":true,"origin":"","legend":"\u003cp\u003eMonocyte cell oxidative stress.\u003cstrong\u003e A \u003c/strong\u003eHeat map shows the expression of top differential expression genes(DEGs) per IR group for monocyte cells. \u003cstrong\u003eB-C \u003c/strong\u003eExpression of OS genes. \u003cstrong\u003eD\u003c/strong\u003eUMAP of rat groups, \u003cem\u003eSod2\u003c/em\u003e expression, and UCell scores of OS-related terms include OS1 (cellular response to ROS) and OS3 (regulation of oxidative stress-induced intrinsic apoptotic signaling pathway). \u003cstrong\u003eE\u003c/strong\u003e mIF staining for \u003cem\u003eSod2\u003c/em\u003e (brown), \u003cem\u003eIl1r2\u003c/em\u003e (green), and DAPI (blue) in rat lung sections from the control and radiation groups (24H, 1W, and 2W). The other groups in Figure S11A. Scale bars, 20 µm. \u003cstrong\u003eF\u003c/strong\u003eStatistical analysis of the \u003cem\u003eSod2\u003c/em\u003e (left) and \u003cem\u003eIl1r2\u003c/em\u003e (right) fluorescence (a.u) in rat lung tissue. The P value was computed with the student's t-test (two-sided test) from scipy (ns, P\u0026gt;0.05; *, P\u0026lt;0.05; **, P\u0026lt;0.01; ***, P\u0026lt;0.001; ****, P\u0026lt;0.0001). Each dot represents one analyzed image (n=5 in each group). \u003cstrong\u003eG\u003c/strong\u003eEnrichment map of top 100 DEGs from per IR group compared with control in monocyte cells. Nodes represent enriched terms, where the node size corresponds to the number of term genes. The color indicates terms that are enriched by IR groups. Similar terms with many common genes are connected and named according to prior knowledge. \u003cstrong\u003eH\u003c/strong\u003e Half violin plot shows UCell scores of OS-related terms include OS1, OS2 (regulation of response to oxidative stress), OS3, OS4 (cellular response to oxidative stress), OS5 (superoxide metabolic process), and OS6 (cell death in response to oxidative stress).\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/6ed8b060731b3efe7de05ca6.png"},{"id":91949017,"identity":"2dacf6ff-4f8b-4e59-ac04-4dcc8113282d","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":616383,"visible":true,"origin":"","legend":"\u003cp\u003eIrradiation enhances the pro-inflammatory phenotype of macrophages.\u003cstrong\u003e A \u003c/strong\u003eLeft, UMAP of 22,423\u003cstrong\u003e \u003c/strong\u003emacrophages. Right, dynamics in cell proportions of macrophage components across the Crl and IR conditions at the different time points. \u003cstrong\u003eB \u003c/strong\u003ePseudotime analysis of macrophage components. The branched trajectory was colored by pseudotime (left) and cell subsets (right). The pie chart shows the proportion of rat groups (left) and cell subsets (right) in each branch. \u003cstrong\u003eC\u003c/strong\u003e CytoTRACE score distribution. P-values were calculated using Kruskal-Wallis statistical test, the same below. \u003cstrong\u003eD \u003c/strong\u003eThe expression of proliferation, differentiation, and pro-inflammatory-related genes in Crl and IR group for \u003cem\u003eCcr5\u003c/em\u003e+ IMs and M1 macrophages. \u003cstrong\u003eE \u003c/strong\u003eUMAP shows the expression of pro-inflammatory genes, \u003cem\u003eIfngr2\u003c/em\u003e and \u003cem\u003eIl1b\u003c/em\u003e. \u003cstrong\u003eF-G\u003c/strong\u003e Violin diagram of inflammatory response signature score in macrophage components (\u003cstrong\u003eF\u003c/strong\u003e) and groups (\u003cstrong\u003eG\u003c/strong\u003e). \u003cstrong\u003eH \u003c/strong\u003eHeat map shows the distribution of activities of immune pathways and receptors from the ImmPort database across Crl and IR groups.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/f39932dd9b7d5cd72596275e.png"},{"id":91949018,"identity":"46e20f7d-8c6f-4386-a425-a258831cfacc","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":570043,"visible":true,"origin":"","legend":"\u003cp\u003eMIF signaling mediates pro-inflammatory properties of macrophage components.\u003cstrong\u003e A \u003c/strong\u003eClustering analysis of immune pathway scores in macrophage components. \u003cstrong\u003eB \u003c/strong\u003eThe expression of chemokine ligand \u003cem\u003eCcl6\u003c/em\u003e and receptor \u003cem\u003eCxcr4\u003c/em\u003e. \u003cstrong\u003eC\u003c/strong\u003e Differential expression of MIF signaling genes in macrophage components in the IR group relative to control. \u003cstrong\u003eD \u003c/strong\u003eChord diagrams show the cell-cell communications based on MIF, CCL, and CXCL signaling across the control and IR conditions at different time points. \u003cstrong\u003eE \u003c/strong\u003eThe expression of MIF signaling ligands/receptors and selected genes in cell types and rats. \u003cstrong\u003eF, H\u003c/strong\u003e Quantification and estimation the relative intensity of \u003cem\u003eMif\u003c/em\u003e in \u003cem\u003eSftpc\u003c/em\u003e positive epithelial cells (\u003cstrong\u003eF\u003c/strong\u003e) and in \u003cem\u003eCd68\u003c/em\u003e positive macrophages (\u003cstrong\u003eH\u003c/strong\u003e) across control and irradiated lung sections. The P value was computed with the student's t-test (two-sided test) from scipy (ns, P\u0026gt;0.05; *, P\u0026lt;0.05; **, P\u0026lt;0.01; ***, P\u0026lt;0.001). Each dot represents one analyzed image (n=5 in each group).\u003cstrong\u003e G \u003c/strong\u003emIF staining for \u003cem\u003eCd68\u003c/em\u003e (red), \u003cem\u003eMif\u003c/em\u003e (green), and DAPI (blue) in rat lung sections from the control and radiation groups. Arrowheads indicate the colocalization of \u003cem\u003eCd68\u003c/em\u003e and \u003cem\u003eMif\u003c/em\u003e. Scale bars, 20 µm. \u003cstrong\u003eI-J \u003c/strong\u003eThe expression of \u003cem\u003eMif\u003c/em\u003e in control and IR murine based on single-cell (\u003cstrong\u003eI\u003c/strong\u003e) and bulk expression data (\u003cstrong\u003eJ\u003c/strong\u003e). \u003cstrong\u003eK \u003c/strong\u003eH\u0026amp;E staining of lung sections across control, IR, and IR + ISO-1 rat groups. \u003cstrong\u003eL\u003c/strong\u003e Quantification and estimation the inflammation score (left) and the relative intensity of \u003cem\u003eMif\u003c/em\u003ein \u003cem\u003eCd68\u003c/em\u003e positive macrophages (right) at control, IR, and IR + ISO-1 rat groups. \u003cstrong\u003eM \u003c/strong\u003emIF staining for \u003cem\u003eCd68\u003c/em\u003e (red), \u003cem\u003eMif\u003c/em\u003e (green), and DAPI (blue) in rat lung sections.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/812b9cb78b7f623a1e4377a7.png"},{"id":91951554,"identity":"d61b1a82-c5e6-4aa0-86ad-19e277644fce","added_by":"auto","created_at":"2025-09-23 06:44:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4969480,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/c5ae6cc7-c911-4079-8d08-40251e479014.pdf"},{"id":91949022,"identity":"d5dea4ab-2a9f-478b-8dd4-142f0160cdf6","added_by":"auto","created_at":"2025-09-23 06:12:33","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":11445189,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary figures\u003c/p\u003e","description":"","filename":"Additionalfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/55f97578c0e4c67eba077deb.docx"},{"id":91949435,"identity":"7832b7d0-428f-4248-91ce-ed11d20914c2","added_by":"auto","created_at":"2025-09-23 06:20:33","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2480049,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary tables\u003c/p\u003e","description":"","filename":"Additionalfile2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7659149/v1/d765712bba2736727d5e01fe.xlsx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eA dynamically resolved single-cell architecture of radiation pneumonitis provides insights into acute lung injury\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eRadiation pneumonitis (RP) is a potentially severe inflammatory condition of the lungs, primarily affecting patients who undergo radiotherapy for thoracic malignancies such as lung cancer, breast cancer, and esophageal cancer [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Despite advances in radiation therapy techniques, the development of RP remains a significant clinical challenge, leading to symptoms such as cough, dyspnea, and, in severe cases, progression to pulmonary fibrosis [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Advanced radiotherapy techniques, such as stereotactic body radiotherapy, have been employed to minimize the volume of healthy lung tissue exposed to radiation and, consequently, reduce the risk of RP [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDespite the clinical significance of RP, the precise cellular and molecular mechanisms underlying RP are not fully understood. The pathogenesis of RP involves a complex interplay between direct radiation-induced damage to lung tissues and a subsequent inflammatory response. Ionizing radiation leads to cellular damage and the generation of reactive oxygen species (ROS), causing oxidative stress (OS) [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Damage of alveolar type I and II (AT1/AT2) results in loss of surfactant and exudation of serum proteins into the alveoli, which were not visible under light microscopy, and there were no signs of radiographic or clinical injury [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Activates a cascade of inflammatory cytokines from damaged lung cells, including tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and ligands of the \u003cem\u003eCXCR1/2\u003c/em\u003e, attract inflammatory cells to the alveoli and pulmonary interstitium, inducing acute pneumonia [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Patients with unilateral radiotherapy had a marked increase in lymphocytes (mainly CD4\u0026thinsp;+\u0026thinsp;T cells) in both lungs, and this effect was more pronounced in patients with clinically significant RP [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Furthermore, proliferating and activated CD8\u0026thinsp;+\u0026thinsp;T cells in radiation mice were more than double those in control, suggesting that lymphocytes may mediate hypersensitivity reactions to RP [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Understanding the mechanisms underlying RP and identifying effective prevention and early treatment strategies are crucial for improving the quality of life and outcomes for patients undergoing thoracic radiotherapy. The discovery of novel biomarkers and potential therapeutic targets by highlighting genes and pathways uniquely upregulated in radiation condition exposures is a critical step in intervening in RP. Recent advances in single-cell RNA sequencing (scRNA-seq) have revolutionized our understanding of complex tissues by enabling the profiling of gene expression at the resolution of individual cells [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. With the help of scRNA-seq, previous studies explored the mechanisms that lead to radiation-induced pulmonary fibrosis and portrayed cellular senescence in radiation-induced lung injury [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Zhou et al. revealed radiosensitivity heterogeneity of lung cell types at the initial of post-radiation but with limited time points and sample sizes [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Nevertheless, to date, no single cell-based efforts have been aimed at studying the multi-point dynamics of RP in the early stages of acute lung injury (ALI). Profiling individual cells from lung tissues at multiple time points post-radiation exposure, enables us to identify key cell types and pathways involved in the initiation and progression of RP, paving the way for the development of a single-cell dynamic landscape of RP, which offers a comprehensive map of cellular states and interactions at different stages of RP. In addition, ligand-receptor interactions and signaling pathways facilitate the mapping of the intercellular communication that orchestrates the immune response, epithelial damage, and tissue remodeling following radiation exposure [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn this study, we provided a comprehensive view of the cellular and molecular architecture, leading to more precise interventions aimed at preventing or mitigating RP. This research systematically explored the status of single cells at multiple time points before and after radiation, and constructed a dynamic architecture of cellular responses, offering unprecedented insights into the temporal changes within the lung microenvironment during RP development in rats with ALI. The development of a single-cell dynamic landscape of RP represents a significant step forward in understanding the cellular architecture and molecular events that underlie this complex condition.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eSingle-cell dynamic alterations of radiation pneumonitis in rats\u003c/h2\u003e\u003cp\u003eThoracic radiation therapy disrupts the cellular function and molecular levels of lung tissue, leading to acute injury, pneumonia, and possibly irreversible pulmonary fibrosis [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Thus, it is necessary to study the pathogenesis of RP before the development of fibrosis. In this study, we irradiated 20 rats (10 males and 10 females) with a single dose of 20 Gy to construct a pneumonitis rat model (Materials and Methods). Four rats (2 females and 2 males) were selected as the control. To explore dynamic alterations of RP at the cellular and molecular levels, we performed scRNA-seq from 12 rats across the control and IR conditions at multiple time points (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). A total of 84,865 high-quality cells were obtained, of which 11,719 (13.8%) were from control rats, and 73,146 (86.2%) were from IR rats, including 16,890 (24 hours), 17,801 (1 week), 11,220 (2 weeks), 11,675 (4 weeks) and 15,560 (6 weeks) to the IR rats (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). UMAP visualization after merging the datasets from per condition displayed similar distributions of cells into clusters, which indicated the absence of obvious batch effects due to sample processing (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB, Additional file 1: Fig. S1A-D). We cataloged all cells into 20 main cell types annotated with canonical marker genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB-D, Additional file 1: Fig. S1C), thus identifying 3 epithelial compartments (AT1, AT2 and club cells), 4 stromal compartments (2 fibroblasts and 2 endothelial cells), 8 myeloid compartments (4 macrophages, dendritic cells-DCs, monocytes, eosinophils and mast cells) and 5 lymphocyte compartments (Cd4\u0026thinsp;+\u0026thinsp;T, Cd8\u0026thinsp;+\u0026thinsp;T, natural killer-NK, NKT, and B cells). The most abundant cell types were observed to be T lymphocytes and myeloid cells (Additional file 1: Fig. S1E).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe presented the dynamics of cell proportions affecting the different compartments after irradiation at various time points. These results pointed to changes in the proportions of myeloid and stromal cells, which tended to increase after irradiation, e.g., alveolar macrophages and lipofibroblasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F, Additional file 1: Fig. S1F-G, Additional file 2: Table S1). The proportion of T and B lymphocytes decreased within 24 hours after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F, Additional file 1: Fig. S1F-G), indicating that lymphocytes were sensitive to radiation, which was consistent with the previous study [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. After that, T and B lymphocytes gradually recover (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF, Additional file 1: Fig. S1F-G). More detailed, aerocyte capillary (aCap) and general capillary (gCap) endothelial cells decreased rapidly within 24 hours after radiation, which may indicate damage to vascular endothelial cells, whereas gCap and two types of fibroblasts (lipofibroblasts and interstitial fibroblasts), were elevated after 1 week of radiation (Additional file 1: Fig. S1F-G). It is noteworthy that the proportion of epithelial cells, especially AT2, increased significantly within 24 hours after radiation, which was related to the decrease caused by the sensitivity of other cells to radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F, Additional file 1: Fig. S1F-G, Additional file 2: Table S1), indicating that AT2 resists radiation. By comparing the immune characteristics of different cell types, we found that macrophages, especially M1 macrophages, and monocytes exhibited the strongest inflammatory response and immune effects, which was higher than that of the control (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG, Additional file 1: Fig. S2A-B). The inflammatory response, immune effects, and T cell-mediated cytotoxicity were found to gradually increased after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH, Additional file 1: Fig. S2C-E). Chest CT image and H\u0026amp;E staining also showed that the inflammatory properties of lung tissue gradually enhanced after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI, Additional file 1: Fig. S3B). In addition, we evaluated the death properties of cells after radiation. The results showed that the activity of pyroptosis and regulation of apoptosis increased gradually after radiation, and the necroptotic process was strongest at 1 week and 2 weeks (Additional file 1: Fig. S4). These results suggested that radiation stimulates the inflammatory properties and activates immune cells, resulting in a cell death effect.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eVascular endothelial cells and alveolar epithelial cells have DNA damage early after 20 Gy radiation\u003c/h3\u003e\n\u003cp\u003eRadiation is known to cause damage to cellular DNA. To explore the effect of radiation on the degree of DNA damage in different cell types, we inferred copy number variation (CNV) of 4 major cell types using inferCNV pipeline [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], and found that stromal cells, especially endothelial cells, and epithelial cells exhibited the highest CNV score (Additional file 1: Fig. S3C-D). Radiation caused epithelial cells to rapidly damage cellular DNA in 24 hours and gradually worsen within 6 weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-C, Additional file 1: Fig. S5A). Studying the CNV levels of typical characteristics, we found that DNA-dependent protein kinase \u003cem\u003ePrkdc\u003c/em\u003e, AT2 marker \u003cem\u003elamp3\u003c/em\u003e and \u003cem\u003eEtv5\u003c/em\u003e (chromosome 11), and \u003cem\u003eSftpc\u003c/em\u003e (chromosome 15) underwent an increase in CNV burden within 24 hours after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). During this period, EMT hallmark, hypoxia signaling, and IL6 JAK STAT3 signaling showed the highest CNV score 24 hours after radiation, indicating early damage (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). In addition, the anti-proliferation factor \u003cem\u003eBtg2\u003c/em\u003e, the immune genes \u003cem\u003eCxcr4\u003c/em\u003e and \u003cem\u003ePtprc\u003c/em\u003e, and mTOR signaling on chromosome 13 increased the CNV burden early after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Study have shown that \u003cem\u003eBtg2\u003c/em\u003e encoded protein is involved in the regulation of the G1/S transition of the cell cycle [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Furthermore, the DNA damage repair (DDR) genes \u003cem\u003eErcc6\u003c/em\u003e, \u003cem\u003ePolb\u003c/em\u003e, and \u003cem\u003ePrkdc\u003c/em\u003e, which play an important role in DNA replication and repair, also showed an increase in CNV burden in the early post-radiation period (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). Indeed, our multiplex immunofluorescence (mIF) staining also confirmed that epithelial cells underwent DNA damage 24 hours after radiation, showing higher γ-H2AX levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH-I, Additional file 1: Fig. S6).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFurther, we collected 27 DNA damage-related pathways from MSigDB and found that most of these pathways (21/27) increased CNV burden at 24 hours after radiation, and kept elevated at the 4 and 6 weeks (Additional file 1: Fig. S7A). For example, the G2 DNA damage checkpoint serves to prevent the cell from entering mitosis (M-phase) with genomic DNA damage, was significantly increases CNV burden in the early after radiation (Additional file 1: Fig. S7A). Its associated gene, \u003cem\u003eBabam2\u003c/em\u003e, can encode anti-apoptotic, death receptor-associated protein, however, also showed CNV burden in the early after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). Indeed, DNA damage and DDR pathways had higher activity within 24 hours after radiation, such as cellular response to DNA damage stimulus and G2 DNA damage checkpoint (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eJ-K, Additional file 1: Fig. S7B, S7D). At 1\u0026ndash;2 weeks after radiation, cells resist damage, probably due to DNA repair (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Additional file 1: Fig. S7A). Cells respond to DNA damage by instigating robust DNA repair pathways to remove the damage physically [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. It is worth noting that the intrinsic apoptotic signaling pathway in response to DNA damage gradually activated within 6 weeks after radiation (Additional file 1: Fig. S7D), which suggests that epithelial cells are susceptible to involvement in the early stages after 20 Gy radiation and stimulate apoptosis signaling. In addition, cell surface adhesion factors \u003cem\u003eItga4\u003c/em\u003e, \u003cem\u003eItgav\u003c/em\u003e, \u003cem\u003eItgb6\u003c/em\u003e, and AT1 marker \u003cem\u003ePdpn\u003c/em\u003e increased CNV score 4 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003eRadiation-induced alterations in endothelial cell function were thought to be key factors in organ injury through endothelial cell activation and initiation of the apoptotic pathways [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Our results show that the DNA damage of pulmonary vascular endothelial cells early after radiation and gradually worsened within 6 weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD-F, Additional file 1: Fig. S8). For example, the angiogenesis-related genes \u003cem\u003eVegfa\u003c/em\u003e and \u003cem\u003eCol3a1\u003c/em\u003e, and protein secretion-related genes \u003cem\u003eNapsa\u003c/em\u003e and \u003cem\u003eAp2s1\u003c/em\u003e increased the CNV score at 24 hours after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). In addition, the p53 signaling genes \u003cem\u003eGpx2\u003c/em\u003e and \u003cem\u003eZfp36l1\u003c/em\u003e, and vascular homeostasis gene \u003cem\u003eNostrin\u003c/em\u003e were reduced the CNV score at 24 hours after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). These results suggest that pulmonary vascular endothelial cells exhibit DNA damage early after radiation, which was confirmed by DNA damage pathways (Additional file 1: Fig. S8A-B) and the previous study [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Unlike epithelial cells, in addition to having high activity at 24 hours, endothelial cells also have DNA damage pathways activated at 1 and 2 weeks after radiation (Additional file 1: Fig. S8C-D). Consistent with epithelial cells, the activity of DNA damage-induced apoptosis-related pathways in endothelial cells progressively increased within 6 weeks after radiation (Additional file 1: Fig. S8C-D). These results suggest that pulmonary endothelial and epithelial cells exhibit DNA damage in early after radiation and stimulate apoptosis signaling.\u003c/p\u003e\n\u003ch3\u003eEpithelial cells undergo mesenchymal transitions in RP lesions\u003c/h3\u003e\n\u003cp\u003eEMT plays a crucial role in chronic inflammation, tissue remodeling, and a variety of fibrotic diseases. It persists under the condition of an activated inflammatory response and eventually causes organ fibrosis [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. To explore the EMT effects of radiation on alveolar cells, we re-clustered epithelial and stromal cells and examined the expression changes of known EMT markers across different time points after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, Additional file 1: Fig. S9A-C). The expression of cadherin 1 (\u003cem\u003eCdh1\u003c/em\u003e) was gradually increased in epithelial cells after radiation, while \u003cem\u003eVim\u003c/em\u003e, \u003cem\u003eCdh11\u003c/em\u003e, and EMT transcription factors, \u003cem\u003eZeb1\u003c/em\u003e and \u003cem\u003eZeb2\u003c/em\u003e, were increased in endothelial cells and/or fibroblasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, Additional file 1: Fig. S9B-C), suggesting that epithelial cells may undergo mesenchymal transformation. This phenomenon was also well confirmed by the signature score of known EMT hallmark (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;2.2e-16; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB-C). Furthermore, fibroblast proliferation and migration scores in epithelial and stromal cells increased gradually after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB-C, Additional file 1: Fig. S9D). Indeed, mIF staining also proved that the proportion of \u003cem\u003eVim\u003c/em\u003e\u0026thinsp;+\u0026thinsp;fibroblasts gradually increased after radiation, especially in the 4th and 6th weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD-E). Transcriptional trajectory analysis reveals a dynamic transitional spectrum from AT2 cells to fibroblasts or endothelial cells after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF-G). The early stage of the pseudotime is dominated by AT2 cells, which are mainly cells 24 hours after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, Additional file 1: Fig. S9E). After that, the trajectory extends into two branches. One is dominated by gCap and aCap, most of which from the 1 and 2 weeks after radiation; the other includes lipofibroblasts and interstitial fibroblasts, most of which from the 2 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, Additional file 1: Fig. S9E). Furthermore, RNA velocity analysis showed the transition directions from AT2 cells to two types of fibroblasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG), suggesting that epithelial cells undergo mesenchymal transitions after radiation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo systematically study EMT, we sought to detect differentiation potential genes (DPGs) that are associated with radiation. We performed CytoTRACE analysis and found that \u003cem\u003eVim\u003c/em\u003e and \u003cem\u003eMgp\u003c/em\u003e were positively correlated with differentiation scores (\u003cem\u003eR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.6), while \u003cem\u003eAnk3\u003c/em\u003e, \u003cem\u003eSftpb\u003c/em\u003e, and \u003cem\u003eSftpa1\u003c/em\u003e exhibited negative correlations (\u003cem\u003eR\u003c/em\u003e\u0026lt;-0.6; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH-I, Additional file 1: Fig. S9F). Then, we extracted the top 50 DPGs that positively correlated with the differentiation score as the radiation-induced EMT (EMT-RP) signature (Additional file 2: Table S2). The results showed that the EMT-RP signature score was mainly upregulated in lipofibroblasts and interstitial fibroblasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eJ, Additional file 1: Fig. S10C). Compared with the known hallmark EMT (EMT-H) signature and angiogenesis, we found that although the overlap accounted for less than 6%, the EMT-RP signature score showed a strong positive correlation with the EMT-H and angiogenesis scores in both cellular level (\u003cem\u003eR\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.89, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;2.2e-16) and the individual rat level (\u003cem\u003eR\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.97, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.1e-07; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eJ-K, Additional file 1: Fig. S10A-B). This phenomenon was also observed in all cells (Additional file 1: Fig. S10C-G). Furthermore, we compared EMT-RP and EMT-H signatures in irradiated rats and found that EMT-RP had a higher activity after irradiation compared with EMT-H, both at the single-cell level and rat individual level (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eL, Additional file 1: Fig. S10H). To confirm this, we obtained a murine scRNA-seq dataset of the lung responses to radiation injury [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] and found that the EMT-RP signature had more obvious properties compared to the EMT-H in irradiated mice (Additional file 1: Fig. S10I-K), which is consistent with the results in rats. These findings suggest that the radiation-induced EMT signature is superior to the known hallmark EMT signature in irradiated lungs of murine, which was also well confirmed by bulk expression profile of RP mice [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eM). In summary, this study announced that epithelial cells underwent significant mesenchymal transformation after 20 Gy of radiation, and fibroblast growth, proliferation, and migration gradually increased during this process, suggesting that EMT is an indispensable part of radiation-induced lung injury.\u003c/p\u003e\n\u003ch3\u003eIrradiation triggers oxidative stress in monocytes within a week\u003c/h3\u003e\n\u003cp\u003eOxidative stress (OS) is a molecular driver of cellular senescence and a potential contributor to a range of age-related disorders [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. After irradiation, we found upregulation of OS defense genes in monocytes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). For example, \u003cem\u003eSod2\u003c/em\u003e, a major antioxidant defense enzyme, and haptoglobin \u003cem\u003eHp\u003c/em\u003e, a protein with a role in OS prevention [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], were both upregulated at one week after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-C, Additional file 2: Table S3). UMAP plot showed that \u003cem\u003eSod2\u003c/em\u003e was mainly expressed at 1 and 2 weeks after radiation in monocytes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Interestingly, mIF staining showed that \u003cem\u003eSod2\u003c/em\u003e gradually increased after a week of IR in all cells and AT2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE-F, Additional file 1: Fig. S11). Studies have shown that OS can induce protein kinases that phosphorylates C-jun (encoded by \u003cem\u003eJun\u003c/em\u003e) which combines with C-Fos (encoded by \u003cem\u003eFos\u003c/em\u003e) forming the AP-1 pathway, triggering transcription involved in inflammation [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In monocytes, \u003cem\u003eJun\u003c/em\u003e and \u003cem\u003eFos\u003c/em\u003e were found to be significantly upregulated at 24 hours and 6 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-C, Additional file 2: Table S3). In addition, \u003cem\u003eSlpI\u003c/em\u003e, a protein affected by reactive products of oxidative metabolism [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], was also significantly upregulated at 1 week after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-C). These results suggest that radiation may trigger OS in monocytes. To confirm this, we performed functional enrichment analysis on irradiated monocytes relative to control cells. The genes upregulated after radiation were found enrich in aging, pulmonary valve artery (such as abnormality of the pulmonary artery), and response to extracellular stimulus and stress-related functions (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). In particular, responding to radiation function was enriched by the upregulated genes 1 week after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG, Additional file 2: Table S4). Further analysis revealed that oxidative stress response and ROS-related functions were activated after radiation. For example, the upregulated genes one week after radiation were enriched in the pathways involved in response to ROS, antioxidant activity, and hydrogen peroxide metabolic process (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). In addition, ROS metabolic process and response to OS regulation pathways were enriched by the upregulated genes at 1 and 2 weeks after radiation. These results further supported that radiation caused OS in monocytes, which was also confirmed by OS pathway activity, as almost all of them showed upregulation at 1 and 2 weeks after radiation (Additional file 1: Fig. S12A). Indeed, the pathways involved in the cellular response to OS (such as OS1, OS2, and OS4) and superoxide metabolic process (OS5) had significantly higher activity at 1 and 2 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eConsidering the dynamics of monocytes across different time points after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF), it is not difficult to speculate that monocytes dropped sharply one week after radiation due to the aggravation of OS, which triggered the cell death program. Indeed, functional enrichment and activity comparison of OS-induced death signaling also confirmed this conjecture. For example, the genes upregulated one week after radiation were involved in the regulation of apoptotic signaling pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). Cell death in response to OS and apoptotic signaling pathways were also enriched by the upregulated genes after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG, Additional file 2: Table S4). As expected, intrinsic apoptosis signaling pathways involved in response to OS, such as OS6 and OS3, showed significantly higher activity at 1 and 2 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH). Furthermore, radiation-induced upregulated genes were also involved in the inflammatory response and interferon signaling-related pathways (Additional file 1: Fig. S12B), which was consistent with the strong inflammatory activity of monocytes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG). The mIF staining of inflammatory factor \u003cem\u003eIl1r2\u003c/em\u003e confirmed the enhancement of inflammatory response after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE-F). In particular, myeloid or monocyte differentiation, chemokine or receptor binding activity, and toll-like receptor production-related pathways were enriched by radiation-induced upregulated genes (Additional file 1: Fig. S12B), suggesting that the effects of radiation on monocytes contribute to the development of RP, which supports the fact that monocytes are hypersensitive to ionizing radiation and ROS [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eMacrophages enhance the pro-inflammatory phenotype after irradiation\u003c/h3\u003e\n\u003cp\u003eOur results revealed that macrophage components, especially M1 macrophages, showed a stronger inflammatory response compared with other cell types (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG), indicating that macrophage are one of the main contributors to RP. To further investigate, we compared the dynamics of macrophage components before and after radiation and found that \u003cem\u003eBst2\u003c/em\u003e\u0026thinsp;+\u0026thinsp;alveolar macrophages (AMs) as resident macrophages accounted for the majority of the control group, but gradually decreased after radiotherapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, Additional file 1: Fig. S13A-C). Previous study have shown that AMs are the sentinel cells of the alveolar space, orchestrating the initiation and resolution of inflammation during ALI [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In contrast, the proportion of M1 macrophages and interstitial macrophages (IMs) gradually increased after radiation, with the most significant increase in \u003cem\u003eFabp5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs at week 4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA), suggesting that IR chemotactic macrophages to the injury site. Transcriptional trajectory analysis demonstrated a dynamic from \u003cem\u003eBst2\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs to M1 macrophages after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The early stage of the pseudotime was dominated by \u003cem\u003eBst2\u003c/em\u003e\u0026thinsp;+\u0026thinsp;and \u003cem\u003eUqcr10\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs, which control inflammation and maintain homeostasis [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Thereafter, two branches formed: one dominated by M1 macrophages and \u003cem\u003eCcr5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs, which were mainly from 1 and 2 weeks after radiation, indicating an enhanced pro-inflammatory response; the other dominated by two types of IMs, mostly from 4 and 6 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-B). CytoTRACE analysis showed that \u003cem\u003eBst2\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs were less differentiation and more developmentally early, while M1 macrophages were more differentiation and developmentally late (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Consistent with the cell trajectory, the differentiation potential of cells was highest before radiation and gradually decreased over time after radiation until new AMs (\u003cem\u003eUqcr10\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs) were formed (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, Additional file 1: Fig. S13D).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eFabp5\u003c/em\u003e plays an important role in fatty acid uptake and metabolism, and was found to be upregulated in all macrophage components after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD, Additional file 1: Fig. S13E, Additional file 2: Table S3), which means that the fatty acid metabolism by macrophages is increased, suggesting that macrophages expanded after radiation [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. \u003cem\u003eHopx\u003c/em\u003e, a crucial marker of specific developmental and differentiation potentials, was found to be upregulated in macrophages after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), indicating the development of macrophages. Further analysis revealed that the pro-inflammatory cytokines \u003cem\u003eIl1b\u003c/em\u003e and \u003cem\u003eTlr2\u003c/em\u003e were mainly expressed in M1 macrophages and \u003cem\u003eCcr5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs and gradually increased after radiation, especially at 1 and 2 weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD-E, Additional file 1: Fig. S13E, Additional file 2: Table S3). Interferon-gamma receptor 2 (\u003cem\u003eIfngr2\u003c/em\u003e) can induce the activation of macrophages [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] and mainly expressed in M1 macrophages, indicating that M1 macrophages were activated. Furthermore, M1 macrophages and \u003cem\u003eCcr5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs had stronger inflammatory response scores, and their proportions gradually increased after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). These results support that radiation enhances the pro-inflammatory response of macrophage components. Indeed, the macrophage components had a stronger inflammatory response at 1 and 2 weeks after radiation, which was confirmed by the activity of immune pathways and receptors (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG-H).\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eMacrophage components exhibit pro-inflammatory properties through MIF signaling\u003c/h2\u003e\u003cp\u003eTo further elucidate the pro-inflammatory properties of macrophages, we performed unsupervised cluster analysis based on immune pathways/receptors activity and classified macrophage components into three categories (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). The first category includes \u003cem\u003eFabp5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs, \u003cem\u003eBst2\u003c/em\u003e\u0026thinsp;+\u0026thinsp;and \u003cem\u003eUqcr10\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs, which activate the cytokines, interleukins, and TNF family members-related pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). The second category includes \u003cem\u003eCcr5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs and M1 macrophages, which upregulated immune pathway receptors, such as cytokine receptors, chemokine receptors, and interferon receptors. The last one is \u003cem\u003eTop2a\u003c/em\u003e\u0026thinsp;+\u0026thinsp;AMs, which upregulate NK cytotoxicity genes and show activation of immune response (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, Additional file 1: Fig. S2B). These results suggest functional heterogeneity in the pro-inflammatory of macrophage components, particularly the preference for enrichment of ligand-receptor pairs. For example, the cytokine ligand \u003cem\u003eCcl6\u003c/em\u003e is mainly upregulated in AMs, while the receptor \u003cem\u003eCxcr4\u003c/em\u003e is upregulated in M1 macrophages and \u003cem\u003eCcr5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;IMs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). Therefore, it is necessary to dissect the pro-inflammatory mechanisms of macrophages through cell-cell interactions. We employed the CellChat program here to explore potential ligand-receptor interactions and constructed the cell-cell communication network based on 141 murine-secreted signaling pathways (691 unique ligands/receptors), of which 39 signaling pathways significantly interacted in cell types from 6 groups (Table\u0026nbsp;5). AMs were found to be signal senders (source) in cell-cell interactions, whereas M1 macrophages and IMs tend to be receivers (target, Additional file 1: Fig. S15B), which was consistent with the results of immune receptor (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-B). In addition, epithelial cells and stromal cells are more likely to be senders, while DCs and other myeloid cells tend to participate in cell interactions as receivers (Additional file 1: Fig. S15B). Moreover, most cell types showed stronger cell interactions after radiation compared with controls, especially macrophage components (Additional file 1: Fig. S14, S15A-B).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAfterward, we wondered whether specific signaling pathways contribute to radiation-induced cellular communications. The probability of cell-cell interaction in radiation and control groups was inferred based on predefined signaling pathways. Interestingly, the MIF signaling pathway was found to increase the interaction probability after radiation, especially the cell-cell communications involving myeloid macrophage components (Additional file 1: Fig. S16), which was also observed in other murine scRNA-seq data (Additional file 1: Fig. S17). As a pro-inflammatory mediator, macrophage migration inhibitory factor (MIF) has been shown to be involved in the pathogenesis of acute respiratory distress syndrome, inflammatory and autoimmune diseases [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. The expression of MIF signaling ligand \u003cem\u003eMif\u003c/em\u003e, and receptor genes (\u003cem\u003eCd74\u003c/em\u003e, \u003cem\u003eCd44\u003c/em\u003e, and \u003cem\u003eCxcr4\u003c/em\u003e) were upregulated in different macrophage components after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). Considering the important role of inflammatory factors in RP, we combined MIF signaling with inflammatory signaling (including CCL and CXCL signaling) and summarized four unique radiation-induced communication patterns. The first pattern, at 24 hours after radiation, mainly included epithelial cell interactions with monocytes based on CXCL signaling (\u003cem\u003eCxcl3\u003c/em\u003e-\u003cem\u003eCxcr2\u003c/em\u003e) and with IMs based on MIF signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). \u003cem\u003eCxcl3\u003c/em\u003e was found specifically expressed in epithelial cells and \u003cem\u003eCxcr2\u003c/em\u003e was upregulated in monocytes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, Additional file 1: Fig. S15C). The second pattern, at 1 and 2 weeks, was dominated by the interaction of epithelial and/or endothelial cells with M1 and/or IMs based on MIF signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). Moreover, the relative intensity of \u003cem\u003eMif\u003c/em\u003e\u0026thinsp;+\u0026thinsp;epithelial cells increased during this period (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF, Additional file 1: Fig. S15D), which was inseparable from the recruitment of pro-inflammatory cells to participate in the immune response after EMT occurs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F and Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). \u003cem\u003eMif\u003c/em\u003e was expressed in endothelial and epithelial cells, while \u003cem\u003eCd74\u003c/em\u003e, \u003cem\u003eCd44\u003c/em\u003e, and \u003cem\u003eCxcr4\u003c/em\u003e were mainly upregulated in M1 and/or IMs, and all were elevated after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, Additional file 1: Fig. S15E, Additional file 2: Table S3). The third pattern, at 4 weeks, mainly Cd4\u0026thinsp;+\u0026thinsp;T cells interact with M1 and IMs based on MIF signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). At this time, the role of lymphocytes and the overall inflammatory response were obviously enhanced, and the T cells gradually recovered (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH), which indicates that T cells are involved in the response to RP [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In addition, inflammation can induce DNA damage by releasing inflammatory cytokines [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Also, macrophages and T-lymphocytes release TNF-α and \u003cem\u003eMIF\u003c/em\u003e to exacerbate DNA damage [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. In this study, we indeed found that T cells and macrophages upregulated \u003cem\u003eMif\u003c/em\u003e, while AMs specifically upregulated \u003cem\u003eTnf\u003c/em\u003e, and that both upregulated after radiation, especially at 4 and 6 weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG-H, Additional file 1: Fig. S15E), which further explains DNA damage of epithelial cells at 4 and/or 6 weeks after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-B). The last pattern, at 6 weeks, was dominated by the interaction of AMs with IMs based on MIF signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). Indeed, mIF staining also confirmed that \u003cem\u003eMif\u003c/em\u003e\u0026thinsp;+\u0026thinsp;macrophages gradually increased after radiation and reached a maximum at week 6 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG-H). We also demonstrated that \u003cem\u003eMif\u003c/em\u003e was upregulated after radiation using additional murine datasets [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] in both single-cell and bulk data (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eI-J). Furthermore, we constructed an IR rat model treated with ISO-1 (MIF inhibitor) and found that the capillary swelling, alveolar exudate, and inflammatory score were reversed (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eK-L, Additional file 1: Fig. S15F). In particular, after treatment with ISO-1, \u003cem\u003eMif\u003c/em\u003e\u0026thinsp;+\u0026thinsp;macrophages were significantly reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eL-M). Taken together, these results suggest that MIF signaling plays a crucial role in radiation-induced inflammatory responses, with blocking this signaling could reduce the risk of RP.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eDynamic changes in lymphocytes during radiation-induced lung injury\u003c/h3\u003e\n\u003cp\u003eLymphocytes play an indispensable role in the development of RP [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Here, we re-clustered lymphocytes and divided them into 18 subsets, including 7 Cd4\u0026thinsp;+\u0026thinsp;T subsets (na\u0026iuml;ve T, Tcm, Tfh, Th2, Treg, Cd69\u0026thinsp;+\u0026thinsp;Trm and Cd103\u0026thinsp;+\u0026thinsp;Trm cells), 5 Cd8\u0026thinsp;+\u0026thinsp;T subsets (na\u0026iuml;ve T, Tem, Teff, \u003cem\u003eCrtam\u003c/em\u003e\u0026thinsp;+\u0026thinsp;Trm and NKT cells), 5 B cell subsets (na\u0026iuml;ve B, proB, gcB, folB and plasma cells) and 1 NK subset (Additional file 1: Fig. S18A, Fig. S19A). Na\u0026iuml;ve T and na\u0026iuml;ve B cells gradually decreased (Additional file 1: Fig. S18B), which indicate that na\u0026iuml;ve cells differentiated into activated lymphocytes after radiation [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Among Cd4\u0026thinsp;+\u0026thinsp;T cells, Tcm cells with high differentiation potential increased rapidly at 24 hours after radiation (Additional file 1: Fig. S18B, Fig. S19D-E). In contrast, Tfh cells gradually increased after 24 hours of radiation, and Cd69\u0026thinsp;+\u0026thinsp;Trm cells proliferated at 6 weeks. This leads us to speculate that radiation causes na\u0026iuml;ve T cells to differentiate into Tfh and Cd69\u0026thinsp;+\u0026thinsp;Trm cells via Tcm cells. To confirm this conjecture, we performed transcriptional trajectory analysis for Cd4\u0026thinsp;+\u0026thinsp;T subset and revealed a dynamic differentiation spectrum from na\u0026iuml;ve T cells to Tfh and Trm cells (Additional file 1: Fig. S18C-D, Fig. S20B-C). The beginning phase of the pseudotime was dominated by na\u0026iuml;ve T cells, which were mainly in the control and the early stages (H24 and W1) after radiation (Additional file 1: Fig. S18D-E). After that, na\u0026iuml;ve T cells differentiate into three functional cell types via Tcm cells. The first one was dominated by Tfh and Th2, most of which belong to cells at 1 to 4 weeks after radiation (Additional file 1: Fig. S18D). During this period, the proportion of Cd8\u0026thinsp;+\u0026thinsp;Teff and NK cells increased obviously (Additional file 1: Fig. S18B), which supports that inflammatory environment induced activation of Cd8\u0026thinsp;+\u0026thinsp;T cells [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. The second one, Treg cells, increased slightly within 1 to 4 weeks after radiation (Additional file 1: Fig. S18B, S18D), indicating immune regulation in RP. The last one, Cd69\u0026thinsp;+\u0026thinsp;Trm, distributed at the end of the differentiation trajectory, and most of which were in the 6 weeks after radiation (Additional file 1: Fig. S18D-E), indicating that radiation transforms T cells into Cd69\u0026thinsp;+\u0026thinsp;Trm cells with memory and differentiation functions and resides in tissues to play a secondary immune role [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eFor Cd8\u0026thinsp;+\u0026thinsp;T cells, we noticed that the proportion of Teff cells increased from 1 to 2 weeks after radiation (Additional file 1: Fig. S18B, Additional file 2: Table S6), which was consistent with the activity of T cell-mediated cytotoxicity during this period (Additional file 1: Fig. S2E). Transcriptional trajectory analysis reveals the dynamics of differentiation from na\u0026iuml;ve T cells to NKT cells and Teff cells after radiation (Additional file 1: Fig. S18F-H, Fig. S20D-F). The beginning phase of the pseudotime was dominated by na\u0026iuml;ve T cells, mainly from the control group (Additional file 1: Fig. S18G, Fig. S20E-F). After that, the trajectory extends into two branches. One was dominated by Teff cells, most of which belong to cells at 1 to 4 weeks after radiation; and the other one was dominated by Tem cells, most of which were from 6 weeks after radiation (Additional file 1: Fig. S18G). In addition, several Cd8\u0026thinsp;+\u0026thinsp;T cells differentiated into Tem cells with memory function at 6 weeks after radiation (Additional file 1: Fig. S18B, S18G). Analysis of B lymphocytes alone showed an increase in folB and proB cells from 1 to 2 weeks after radiation (Additional file 1: Fig. S18I, Fig. S20G-H), indicating B cell development and differentiation [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], which is consistent with the strong immune response at this time. Differentiation trajectory and RNA velocity analysis showed that na\u0026iuml;ve B cells and proB cells dominated in the control and early stage, while they gradually differentiated into folB cells at 1 to 2 weeks after radiation (Additional file 1: Fig. S18J-K, Fig. S20I), further confirming that lymphocyte-mediated immune activity was enhanced 1 to 2 weeks after radiation. Taken together, these results indicate that radiation activates na\u0026iuml;ve T/B cells to differentiate into functional lymphocytes, participate in immune responses and response to RP, and that several na\u0026iuml;ve T cells differentiate into T cells with memory functions and reside in tissues.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we constructed a single-cell dynamic landscape of RP in rats with ALI under 20 Gy dose and characterized in detail the cell type responses to radiation, particularly revealing the damage to vascular endothelial cells and alveolar epithelial cells at the molecular level. Our study systematically delineated epithelial-mesenchymal transitions, monocyte oxidative stress, and pro-inflammatory properties of macrophages at various time points after radiation. The development of a single-cell dynamic architecture of RP provides a unique and highly detailed understanding of the cellular and molecular mechanisms driving the onset and progression of RP. This research offers critical insights into the complex interactions between epithelial, stromal cells, and immune, especially macrophages, following radiation-induced lung injury, contributing to our understanding of how damage, inflammation, and apoptosis are orchestrated at the cellular level.\u003c/p\u003e\u003cp\u003eIn the early stages of ALI (24 hours) after radiation, epithelial cells underwent obvious damage, both from the single-cell molecular level characterization and γH2AX immunofluorescence verification. This phenomenon was evidenced by genomic aberrations and transcriptomic activities of DNA damage-related pathways, particularly activation of DNA damage repair pathways. The role of AT2 was underscored in the progression of RP, and is known to play a key role in maintaining lung homeostasis and repairing damaged tissue [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The single-cell architecture revealed that AT2 undergoes significant stress following radiation exposure, with some subsets transitioning into mesenchymal states, contributing to the formation of fibroblasts. In addition to providing valuable mechanistic insights, this single-cell dynamics of RP in rats also highlights potential biomarkers for early detection of the disease. The identification of specific signature in epithelial cells during the early phases of RP may offer new tools for diagnosing radiation-induced lung injury before irreversible damage occurs. We identified a radiation-induced EMT signature that was highly correlated with the hallmark EMT at both the cellular and individual levels in RP rats as well as in lung cancer patients, indicating conservation of irradiated EMT signature. The EMT signature might be used to monitor patients receiving thoracic radiotherapy and enable timely interventions to prevent the progression of RP to chronic fibrosis. One of the key findings from this research is the identification of distinct macrophage subpopulations that appear to play crucial roles in different phases of RP.\u003c/p\u003e\u003cp\u003ePro-inflammatory macrophages, M1 macrophages and \u003cem\u003eCcr5\u003c/em\u003e + IMs, characterized by high levels of cytokine production, were shown to dominate at 1 and 2 weeks after radiation, contributing to ALI through the promotion of inflammation. This dynamic switch between macrophage phenotypes highlights the importance of immune regulation in both the onset and progress of RP, suggesting that therapeutic interventions aimed at modulating macrophage function could mitigate disease severity. Furthermore, the mapping of ligand-receptor interactions revealed four intercellular communication patterns based on a key signaling pathway, MIF signaling, which mediates pro-inflammatory response (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). The identification of this pathway opens new avenues for targeted therapies aimed at interrupting the signaling cascades that drive disease progression. For example, inhibitors of MIF signaling could be explored to reduce RP and improve outcomes in patients at risk of developing chronic fibrosis. During this phase, another myeloid cell type, monocytes, showed an activated OS response. A typical feature was that OS genes, such as \u003cem\u003eSod2\u003c/em\u003e, \u003cem\u003eHp\u003c/em\u003e, and \u003cem\u003eSlpI\u003c/em\u003e, were significantly upregulated 1 week after radiation in monocytes. In addition, the upregulated genes at 1 and 2 weeks after radiation were significantly enriched in OS response-related pathways, including regulation of response to oxidative stress pathway, ROS, and cell death in response to oxidative stress (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). Indeed, these pathways had significantly higher activity at 1 and 2 weeks after radiation than the other groups. These results suggest that future studies could reduce or alleviate radiation pneumonitis by targeting OS, for example in combination with fullerenol therapy [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. This finding highlights the progression of RP from an acute damage phase to an inflammatory and points to the potential for targeting myeloid cell stress responses as a therapeutic approach. Lymphocytes, particularly Cd4 + T cells, play a central role in orchestrating the immune response, and their activation, differentiation, and migration are key to understanding the balance between inflammation and tissue repair. Specifically, Cd4 + T cells interact with macrophage components through MIF signaling, enhancing pro-inflammatory properties after radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD-E). Radiation exposure induces DNA damage and OS, triggering the release of cytokines and chemokines that recruit lymphocytes to the site of injury. Among these, Cd4 + Tfh cells and Cd8 + Teff cells are prominently involved in the inflammatory response. As the injury evolves, the balance may shift towards those two types of T cells, which act to promote inflammation and suppress tissue repair. In addition, activated Cd4 + T cells differentiate into various subtypes, such as Th2 and Treg cells, each with distinct roles in modulating the immune response, limiting excessive immune activation [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Afterwards, some Cd4 + T cells differentiated into Cd69 + Trm cells with memory function and resided in the lung tissues at 6 weeks after radiation.\u003c/p\u003e\u003cp\u003eWhile this study represents a significant advance in our understanding of RP, there are important limitations to consider, such as species-specific differences. Future studies should aim to validate these findings in human tissues, using similar single-cell technologies to create a comprehensive atlas of RP in patients undergoing radiotherapy. Moreover, additional techniques, such as scATAC-seq and spatial transcriptomics, could further enhance our understanding of the RP architecture and spatially organized cellular interactions, separately. T/B cell receptor sequencing could clearly analyze the differentiation and expansion of T/B lymphocytes. In conclusion, this study provides a comprehensive single-cell dynamic landscape of RP in rats with ALI, offering valuable insights into the cellular and molecular mechanisms underlying this complex condition. By identifying key cell populations including immune, epithelial, and stromal involved in RP, as well as the signaling pathways that drive inflammation, this research opens new avenues for devising targeted therapeutic strategies to prevent and treat RP, ultimately improving outcomes for patients undergoing thoracic radiotherapy.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Methods","content":"\u003ch2\u003eRats and ethics statement\u003c/h2\u003e\u003cp\u003eTwelve 6-week-old female SD rats (weight: 160 ± 10g) and twelve 6-week-old male SD rats (weight: 190 ± 10g) purchased from Guangzhou Ruige Biotechnology Co. Ltd (Guangzhou, China) were housed in the animal center of Guangzhou Medical University, and all rats were conditionally reared in the animal center for 2 weeks. All rats were raised in a clean environment with a consistent dark-light schedule (lights on from 7 a.m. to 7 p.m.). This study included 24 rats, 20 of which (10 female SD rats and 10 male SD rats), which received a single double lung radiation at a dose of 20 Gy, and the other 4 (two female SD rats and two male SD rats) were randomly selected as negative controls. Four control rats (Crl) and 20 irradiated rats were sacrificed at 24 hours (H24, n = 4), 1 week (W1, n = 4), 2 weeks (W2, n = 4), 4 weeks (W4, n = 4), and 6 weeks (W6, n = 4) after radiation and consequently, lung tissues were harvested intact. The animal experimental were specifically approved by the ethics committee of Guangzhou Medical University (G2023-781) in compliance with the international guidelines.\u003c/p\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eRadiation induced pneumonitis rat model\u003c/h2\u003e\u003cp\u003eTwenty-four rats (12 female SD rats and 12 male SD rats) were sedated with an intraperitoneal injection of 0.3% sodium pentobarbital (40 mg/kg body weight), fixed in the supine position and placed vertically at 500 mm from the electron beam. The rats were then subjected to bilateral thorax radiation (from the clavicle to the lower margin of the costal arch) with a single dose of 20 Gy at 8-week-old utilizing the 4.5-MeV linear electron accelerator facility (VARIAN Trilogy) to induce lung injury. The lungs were imaged using computed tomography (CT) (Aquilion ONE TSX-301C, Canon, Japan) before being sacrificed. Rats were anesthetized with 0.3% sodium pentobarbital (40 mg/kg body weight) and kept in the supine position. All imaging analyses were performed independently by two radiologists.\u003c/p\u003e\u003c/div\u003e\u003cp\u003eTwenty-four rats (12 female SD rats and 12 male SD rats) were sedated with an intraperitoneal injection of 0.3% sodium pentobarbital (40 mg/kg body weight), fixed in the supine position and placed vertically at 500 mm from the electron beam. The rats were then subjected to bilateral thorax radiation (from the clavicle to the lower margin of the costal arch) with a single dose of 20 Gy at 8-week-old utilizing the 4.5-MeV linear electron accelerator facility (VARIAN Trilogy) to induce lung injury. The lungs were imaged using computed tomography (CT) (Aquilion ONE TSX-301C, Canon, Japan) before being sacrificed. Rats were anesthetized with 0.3% sodium pentobarbital (40 mg/kg body weight) and kept in the supine position. All imaging analyses were performed independently by two radiologists.\u003c/p\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eSingle-cell suspension preparation\u003c/h2\u003e\u003cp\u003eTen irradiated rats at 24 hours, 1, 2, 4 and 6 weeks (two rats in each group) after radiation and two control rats were selected for preparation (a total of 12 rats, 1 female and 1 male in each group), and lung tissues were extracted for conducting scRNA-seq.\u0026nbsp;Lung tissues were removed and washed in a 6-well plate containing pre-chilled PBS to remove necrotic tissue, blood, and other impurities. Transfer the tissue to a 1.5 mL centrifuge tube designed for low binding, add 1 mL of digestion solution, and mince the tissue using scissors. Place the tube on a rotary shaker in a 37°C incubator and rotate at approximately 20 rpm for 10 minutes. Remove the centrifuge tube, pipette up and down 30 times, and place it back on the rotary shaker in the 37°C incubator for an additional 10 minutes of digestion. After digestion, remove the centrifuge tube and pipette up and down 30 times before performing AOPI quality control (1:1). Sieve the tissue using a 70 µm cell strainer and centrifuge at 400G for 5 min at 4°C. Discard the filtrate and resuspend the sediment in 1 mL of 1640 medium before performing AOPI quality control (1:1). Add 3 mL of trypan blue solution and incubate for 2 minutes. Terminate trypan blue staining by adding 4 mL of 1640 medium. Centrifuge at 400G for 5 minutes at 4°C. Mix the precipitate in 1640 medium and adjust the concentration to 600–1500 cells/µl.\u003c/p\u003e\u003c/div\u003e\u003ch2\u003eSingle-cell suspension preparation\u003c/h2\u003e\u003cp\u003eTen irradiated rats at 24 hours, 1, 2, 4 and 6 weeks (two rats in each group) after radiation and two control rats were selected for preparation (a total of 12 rats, 1 female and 1 male in each group), and lung tissues were extracted for conducting scRNA-seq.\u0026nbsp;Lung tissues were removed and washed in a 6-well plate containing pre-chilled PBS to remove necrotic tissue, blood, and other impurities. Transfer the tissue to a 1.5 mL centrifuge tube designed for low binding, add 1 mL of digestion solution, and mince the tissue using scissors. Place the tube on a rotary shaker in a 37°C incubator and rotate at approximately 20 rpm for 10 minutes. Remove the centrifuge tube, pipette up and down 30 times, and place it back on the rotary shaker in the 37°C incubator for an additional 10 minutes of digestion. After digestion, remove the centrifuge tube and pipette up and down 30 times before performing AOPI quality control (1:1). Sieve the tissue using a 70 µm cell strainer and centrifuge at 400G for 5 min at 4°C. Discard the filtrate and resuspend the sediment in 1 mL of 1640 medium before performing AOPI quality control (1:1). Add 3 mL of trypan blue solution and incubate for 2 minutes. Terminate trypan blue staining by adding 4 mL of 1640 medium. Centrifuge at 400G for 5 minutes at 4°C. Mix the precipitate in 1640 medium and adjust the concentration to 600–1500 cells/µl.\u003c/p\u003e\u003ch2\u003eLibrary preparation and 10x genomics single-cell RNA sequencing\u003c/h2\u003e\u003cp\u003eCells were filtered through a 30um filter and tagged with the 10x Genomic single cell library platform following the manufacturer’s instructions. Briefly, the cell suspension was introduced into a microfluidic chip equipped with 3' chemistry and then barcoded using the 10x Chromium Controller (10x Genomics). Subsequently, RNA from the barcoded cells was reverse transcribed and sequencing libraries were prepared using reagents in the Chromium Single Cell 3' v2 Kit (10x Genomics) and following the manufacturer's guidelines. Purified cDNA libraries were sequenced on an Illumina NovaSeq 6000 platform with paired-end mode at CHI BIOTECH CO., LTD (Shenzhen, China) according to the protocol provided by Illumina.\u003c/p\u003e\u003ch2\u003eCell calling and Quality control of scRNA-seq data\u003c/h2\u003e\u003cp\u003eRaw read files were processed with Cell Ranger 7.1.0 using default mapping arguments. Reads were mapped to the rattus norvegicus (Norway rat) genome assembly and counted with mRatBN7.2 annotations, and counted the unique molecular identifier (UMI) (by using ‘‘cellranger count’’ function). As a result, a digital cell by gene expression matrix was generated, containing the number of UMIs for each gene detected in each cell. In addition, we took some steps to filter out poor quality data. First, we removed cells with high mitochondrial gene expression because dead cells often exhibit extensive mitochondrial contamination [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Specifically, we fit the expression level of mitochondrial genes by using a median-centered median absolute deviation (MAD)-variance normal distribution, and then removed the cells with significantly higher expression levels than expected (determined by Benjamini-Hochberg corrected FDR \u0026lt; 0.01) [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Second, we removed cells for which less than 500 genes were detected. Third, we identified and removed potential doublets by using DoubletFinder, using 92.5th percentile of the doublet score as cutoff [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. In the end, we retained a total of 84,865 high-quality single cells for 12 rat samples.\u003c/p\u003e\u003ch2\u003eNormalization, clustering and visualization\u003c/h2\u003e\u003cp\u003eThe processed whole gene expression matrix with all selected cells was fed to R package Seurat (v5.0.3) for downstream analyses [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. Briefly, only genes expressed in more than 5 cells were kept, and the raw UMI count matrix was log-normalized with library sizes of each cell and scaled to 10,000 using the \"LogNormalize\" function. Based on the normalized gene expression matrix, 2,000 highly variable genes were identified by using the “FindVariableFeatures” function with the “vst” method. “ScaleData” function was then used to scale and center the gene expression matrix after regressing out the heterogeneity associated with the mitochondrial contamination and UMI count. Unsupervised clustering was done by constructing the shared nearest neighbor (SNN) graph by using “FindNeighbors” function from the R package Seurat with Louvain algorithm. The top 30 principal components were considered and the resolution was 1.6 for whole dataset of each cell type. The first two dimensions of uniform manifold approximation and projection (UMAP) was calculated using the \"RunUMAP\" function.\u003c/p\u003e\u003ch2\u003eCell type annotation and differential expression analysis\u003c/h2\u003e\u003cp\u003eA differential expression test among clusters was then applied with the \"FindAllMarkers\" function (use default parameters but set min.pct as 0.5) from Seurat. We used two complementary approaches to annotate the identities of different cell clusters: (1) applied canonical markers; (2) we checked whether the well-studied marker genes of different cell types were in the top rank of differential expressed genes of query cluster and then assigned the most likely identity for each cell cluster. Signature score was calculated for selected gene-sets, such as immune pathways and receptors from the ImmPort database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://immport.org/shared/home\u003c/span\u003e\u003cspan address=\"https://immport.org/shared/home\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and specific signatures from the molecular signatures database (MSigDB) mouse collections (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/msigdb/\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/msigdb/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), using “AddModuleScore_UCell” function from “UCell” R package. The UCell score was used to estimate the population of cells in a dataset representing the mixed expression level of the gene-set.\u003c/p\u003e\u003ch2\u003ePathway enrichment analysis\u003c/h2\u003e\u003cp\u003eFor a given cell type, we performed pathway enrichment analysis using ActivePathways (v.2.0.3) R package [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e] for all terms from MSigDB (mouse collection M5) based on the differentially expressed genes (DEGs) in radiation rats at different time points compared with normal rats. Terms with a gene set shorter than 10 or longer than 500 will be removed. Terms with BH-corrected significance adjusted P ≤ 0.01 are considered to be significantly enriched by the gene set of interest and will be selected as results. The enrichment results of different radiation groups were used as input to the EnrichmentMap plugin in cytoscape (v.3.9.1) software to draw a network diagram of enrichment analysis at different time points after radiation.\u003c/p\u003e\u003ch2\u003eSingle-cell copy number variation analysis\u003c/h2\u003e\u003cp\u003eThe inferCNV (v.1.19.1) R package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/broadinstitute/inferCNV\u003c/span\u003e\u003cspan address=\"https://github.com/broadinstitute/inferCNV\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used to distinguish DNA damage cells by inferring chromosomal CNVs based on the single-cell expression data. The cells from control rats as normal reference cells were used to estimate CNVs for the cell population of radiation rats. A gene ordering file containing chromosomal start and end positions of each gene from the rat mRatBN7.2 assembly was prepared for the input to the “gene_order_file” parameter in the “CreateInfercnvObject” function. The count matrix and annotation file were input to create the infercnv object, which was used to perform infercnv operations, and then run inferCNV with cutoff = 0.1 in “run” function.\u003c/p\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003eRNA velocity, CytoTRACE and pseudotime analysis\u003c/h2\u003e\u003cp\u003eTo analyze the transcriptional dynamics in distinct cell subsets, we applied the velocyto python package (v.0.17.17) to estimate the RNA velocity of single cells by distinguishing spliced and un-spliced mRNAs [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. We first generated the individual loom file using “velocyto run10x” command based on the output file of CellRanger for each rat sample and then merged all loom files together. Then we fed the merged loom files and the UMAP coordinates of single cells generated by Seurat into velocyto and followed its analysis steps to finally project the RNA velocity vectors onto low-dimension embeddings. The R package CytoTRACE (v.0.3.3) [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e] was applied to predict the differentiation state of cells from the scRNA-seq profiles. Analysis of differentiation trajectories of macrophages, lymphocytes, epithelial and stromal cells was performed using both Monocle 2 and Monocle 3 [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e] by inferring the pseudotemporal ordering of cells according to their transcriptional similarity.\u003c/p\u003e\u003c/div\u003e\u003cp\u003eTo analyze the transcriptional dynamics in distinct cell subsets, we applied the velocyto python package (v.0.17.17) to estimate the RNA velocity of single cells by distinguishing spliced and un-spliced mRNAs [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. We first generated the individual loom file using “velocyto run10x” command based on the output file of CellRanger for each rat sample and then merged all loom files together. Then we fed the merged loom files and the UMAP coordinates of single cells generated by Seurat into velocyto and followed its analysis steps to finally project the RNA velocity vectors onto low-dimension embeddings. The R package CytoTRACE (v.0.3.3) [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e] was applied to predict the differentiation state of cells from the scRNA-seq profiles. Analysis of differentiation trajectories of macrophages, lymphocytes, epithelial and stromal cells was performed using both Monocle 2 and Monocle 3 [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e] by inferring the pseudotemporal ordering of cells according to their transcriptional similarity.\u003c/p\u003e\u003ch2\u003eAnalysis of cell-cell communication\u003c/h2\u003e\u003cp\u003eThe CellChat [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e] was used to investigate cell-cell communication between different cell types on the expression matrix. We extracted multiple mouse ligand-receptor resources of secreted signaling based on CellChat database that manual curated from KEGG base and primary literature [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. In total, 1,209 ligand-receptor pairs were obtained in this study. Statistical analysis of intercellular communications was done permuting the label of cell type for each cell at 100 times (by default) to test the significance of each interaction pair. The communication probability from one cell type to the other one for a particular ligand-receptor pair were described in the original text [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. The significant ligand-receptor pairs with P ≤ 0.05 were determined significant interaction.\u003c/p\u003e\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\u003ch2\u003eAcquisition and analysis of radiation murine public datasets\u003c/h2\u003e\u003cp\u003eThe scRNA-seq data of the lung responses to radiation injury (20 mice, including 5 non-IR mice as control, 5 mice after 10 Gy thorax IR and 10 mice after 17 Gy thorax IR) was obtained from a previous study [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. The gene expression data from the irradiated lungs of murine (n = 18) was downloaded from the Gene Expression Omnibus (GEO) database (GSE85359) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Signature scores for epithelial-mesenchymal transition (EMT) and angiogenesis hallmarks from MSigDB [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] were calculated using single sample gene set enrichment analysis (ssGSEA) method in R package GSVA [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. The newly discovered radiation-induced EMT signature in this study was consisted of the top 50 genes that were positively correlated with the differentiation score based on CytoTRACE analysis.\u003c/p\u003e\u003c/div\u003e\u003cp\u003eThe scRNA-seq data of the lung responses to radiation injury (20 mice, including 5 non-IR mice as control, 5 mice after 10 Gy thorax IR and 10 mice after 17 Gy thorax IR) was obtained from a previous study [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. The gene expression data from the irradiated lungs of murine (n = 18) was downloaded from the Gene Expression Omnibus (GEO) database (GSE85359) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Signature scores for epithelial-mesenchymal transition (EMT) and angiogenesis hallmarks from MSigDB [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] were calculated using single sample gene set enrichment analysis (ssGSEA) method in R package GSVA [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. The newly discovered radiation-induced EMT signature in this study was consisted of the top 50 genes that were positively correlated with the differentiation score based on CytoTRACE analysis.\u003c/p\u003e\u003ch2\u003eH\u0026amp;E and immunofluorescence staining\u003c/h2\u003e\u003cp\u003eThe tissue specimens were obtained from SD rats (n = 24) with the portion of the selected tissue were embedded in paraffin and stained for hematoxylin-eosin (H\u0026amp;E) staining. Briefly, in order to label the nuclear and cytoplasm, tissue sections on glass slides were rehydrated with xylene and alcohol, and then counterstained with hematoxylin (MACKLIN, China) and eosin (SCR, China). Tissue samples were observed under a microscope (BX53, Olympus, Japan). All histological analyses were performed with at least five vision per rat by two pathologists independently. For immunofluorescence staining (IF), a TSAPLus triple fluorescent staining kit (Wuhan Servicebio®, Wuhan, China) was used. Rat lung sections were dewaxed to water, Tris-EDTA (PH 9.0, Sangon, China) was used to antigen repaired, 3% H2O2 was incubated at room temperature in the dark for 25 min to block endogenous peroxidase and reduce non-specific background staining, blocked for 30 min in 5% fetal goat serum and subsequently, incubated with primary antibody (anti-IL1R2 [1/200 dilution; santa cruz; USA], anti-SOD2 [1/500 dilution; proteintech; China], anti-SFTPC [1/500 dilution; proteintech; China], anti-γ-H2AX [1/1000 dilution; abcam; USA], anti-MIF [1/500 dilution; proteintech; China], anti-CD68 [1/500 dilution; proteintech; China], anti-GSN [1/500 dilution; proteintech; China], anti-VIMNETIN [1/500 dilution; proteintech; China]) at 4℃ overnight, (Alexa Fluor® 488 for IL1R2, SFTPC, CD68, VIMNETIN and AlexaFluor594® for SOD2, γ-H2AX, MIF, GSN respectively), and counterstained with secondary antibodies (HRP-Conjugated Goat Anti-Rabbit IgG and HRP-Conjugated Goat Anti-Mouse IgG) for 1 h at room temperature. Nuclear was labeled with 4’6-diamidino-2-phenylindole (DAPI). To further verify the role of MIF in radiation pneumonitis, we constructed an IR (20 Gy) rat model treated with MIF inhibitor (ISO-1, IP 13mg/kg daily, for 2 weeks). Two weeks after the end of radiation, lung tissues were obtained for H\u0026amp;E staining and multiplex immunofluorescence staining.\u003c/p\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eComparison between two groups was carried out by two-sided Wilcoxon rank-sum test. Comparison among multiple groups was performed with Kruskal-Wallis test. The correlation between two continuous variables was measured by Pearson correlation coefficient. The significance level of two discrete variables was determined by Fisher's exact test. P value less than 0.05 was considered statistically significant. All statistical analysis were conducted using R software (version 4.2.3, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.r-project.org\u003c/span\u003e\u003cspan address=\"http://www.r-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eX.Y.S., Z.S., and G.M.L. conceived and designed the study. X.Y.S., Z.S. and W.W. supervised the study. G.M.L. and C.J.L. performed data curation and investigations. G.M.L. performed single-cell data analysis and result\u0026nbsp;visualization. G.M.L., C.J.L. and T.C. designed and implemented the experimental verification protocol. G.M.L. and C.J.L. designed the statistical analysis plan and G.M.L. performed the statistical analysis. All authors contributed to the acquisition, analysis, verification, or interpretation of data. G.M.L. and C.J.L. drafted the manuscript. All authors revised the manuscript and gave final approval of the version to submission. X.Y.S., Z.S. and W.W. contributed equally to this work and are joint corresponding authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the National Natural Science Foundation of China (Grant No. 82272996, 82403816), the China Postdoctoral Foundation (Grant No. 2023M740846), the Postdoctoral Fellowship Program of CPSF (Grant No. GZB20230180), the Science and Technology Program of Guangzhou (Grant No. 202206010081), the Key Specialty Construction Project of Guangzhou Medical University (Grant No. LZ202100302), and the Wu Jieping Medical Foundation (Grant No. 202001301).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe scRNA-seq raw data and processed data generated in this study have been deposited in the Gene Expression Omnibus (GEO) repository, with the accession code GSE286896. Other data of this study can be available from the corresponding author upon reasonable request.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eStandard workflows and open-source R packages and software were used in this study (Materials and methods). No previously unreported custom code was used or developed for the analyses presented in this study. All codes of this study are available from the authors upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe animal experimental were specifically approved by the ethics committee of Guangzhou Medical University (G2023-781) in compliance with the international guidelines.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003cbr\u003e\u003c/strong\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eHanania, A.N., et al., \u003cem\u003eRadiation-Induced Lung Injury: Assessment and Management.\u003c/em\u003e Chest, 2019. \u003cstrong\u003e156\u003c/strong\u003e(1): p. 150-162.\u003c/li\u003e\n\u003cli\u003eVoruganti Maddali, I.S., et al., \u003cem\u003eOptimal management of radiation pneumonitis: Findings of an international Delphi consensus study.\u003c/em\u003e Lung Cancer, 2024. \u003cstrong\u003e192\u003c/strong\u003e: p. 107822.\u003c/li\u003e\n\u003cli\u003eLi, F., et al., \u003cem\u003eRisk factors for radiation pneumonitis in lung cancer patients with subclinical interstitial lung disease after thoracic radiation therapy.\u003c/em\u003e Radiat Oncol, 2021. \u003cstrong\u003e16\u003c/strong\u003e(1): p. 70.\u003c/li\u003e\n\u003cli\u003eKuipers, M.E., et al., \u003cem\u003ePredicting Radiation-Induced Lung Injury in Patients With Lung Cancer: Challenges and Opportunities.\u003c/em\u003e Int J Radiat Oncol Biol Phys, 2024. \u003cstrong\u003e118\u003c/strong\u003e(3): p. 639-649.\u003c/li\u003e\n\u003cli\u003eWeiss, A., et al., \u003cem\u003ePrediction of radiation pneumonitis using the effective alpha/beta of lungs and heart in NSCLC patients treated with proton beam therapy.\u003c/em\u003e Radiother Oncol, 2024. \u003cstrong\u003e190\u003c/strong\u003e: p. 110013.\u003c/li\u003e\n\u003cli\u003eMarks, L.B., et al., \u003cem\u003eRadiation-induced lung injury.\u003c/em\u003e Semin Radiat Oncol, 2003. \u003cstrong\u003e13\u003c/strong\u003e(3): p. 333-45.\u003c/li\u003e\n\u003cli\u003ePohjoismaki, J.L.O. and S. Goffart, \u003cem\u003eAdaptive and Pathological Outcomes of Radiation Stress-Induced Redox Signaling.\u003c/em\u003e Antioxid Redox Signal, 2022. \u003cstrong\u003e37\u003c/strong\u003e(4-6): p. 336-348.\u003c/li\u003e\n\u003cli\u003eCurras-Alonso, S., et al., \u003cem\u003eAn interactive murine single-cell atlas of the lung responses to radiation injury.\u003c/em\u003e Nat Commun, 2023. \u003cstrong\u003e14\u003c/strong\u003e(1): p. 2445.\u003c/li\u003e\n\u003cli\u003eKasmann, L., et al., \u003cem\u003eRadiation-induced lung toxicity - cellular and molecular mechanisms of pathogenesis, management, and literature review.\u003c/em\u003e Radiat Oncol, 2020. \u003cstrong\u003e15\u003c/strong\u003e(1): p. 214.\u003c/li\u003e\n\u003cli\u003eRoberts, C.M., et al., \u003cem\u003eRadiation pneumonitis: a possible lymphocyte-mediated hypersensitivity reaction.\u003c/em\u003e Ann Intern Med, 1993. \u003cstrong\u003e118\u003c/strong\u003e(9): p. 696-700.\u003c/li\u003e\n\u003cli\u003eWu, F., et al., \u003cem\u003eCellular Atlas of Senescent Lineages in Radiation- or Immunotherapy-Induced Lung Injury by Single-Cell RNA-Sequencing Analysis.\u003c/em\u003e Int J Radiat Oncol Biol Phys, 2023. \u003cstrong\u003e116\u003c/strong\u003e(5): p. 1175-1189.\u003c/li\u003e\n\u003cli\u003eJovic, D., et al., \u003cem\u003eSingle-cell RNA sequencing technologies and applications: A brief overview.\u003c/em\u003e Clin Transl Med, 2022. \u003cstrong\u003e12\u003c/strong\u003e(3): p. e694.\u003c/li\u003e\n\u003cli\u003eTravaglini, K.J., et al., \u003cem\u003eA molecular cell atlas of the human lung from single-cell RNA sequencing.\u003c/em\u003e Nature, 2020. \u003cstrong\u003e587\u003c/strong\u003e(7835): p. 619-625.\u003c/li\u003e\n\u003cli\u003eYang, M., et al., \u003cem\u003eSingle-Cell Transcriptome Analysis of Radiation Pneumonitis Mice.\u003c/em\u003e Antioxidants (Basel), 2022. \u003cstrong\u003e11\u003c/strong\u003e(8).\u003c/li\u003e\n\u003cli\u003eMa, L., et al., \u003cem\u003eA Study on the Radiosensitivity of Radiation-Induced Lung Injury at the Acute Phase Based on Single-Cell Transcriptomics.\u003c/em\u003e Front Immunol, 2022. \u003cstrong\u003e13\u003c/strong\u003e: p. 941976.\u003c/li\u003e\n\u003cli\u003eArmingol, E., et al., \u003cem\u003eDeciphering cell-cell interactions and communication from gene expression.\u003c/em\u003e Nat Rev Genet, 2021. \u003cstrong\u003e22\u003c/strong\u003e(2): p. 71-88.\u003c/li\u003e\n\u003cli\u003eTirosh, I., et al., \u003cem\u003eSingle-cell RNA-seq supports a developmental hierarchy in human oligodendroglioma.\u003c/em\u003e Nature, 2016. \u003cstrong\u003e539\u003c/strong\u003e(7628): p. 309-313.\u003c/li\u003e\n\u003cli\u003eWu, G.S. and C.H. Bassing, \u003cem\u003eFlip the switch: BTG2-PRMT1 protein complexes antagonize pre-B-cell proliferation to promote B-cell development.\u003c/em\u003e Cell Mol Immunol, 2018. \u003cstrong\u003e15\u003c/strong\u003e(9): p. 808-811.\u003c/li\u003e\n\u003cli\u003eChatterjee, N. and G.C. Walker, \u003cem\u003eMechanisms of DNA damage, repair, and mutagenesis.\u003c/em\u003e Environ Mol Mutagen, 2017. \u003cstrong\u003e58\u003c/strong\u003e(5): p. 235-263.\u003c/li\u003e\n\u003cli\u003eWijerathne, H., et al., \u003cem\u003eMechanisms of radiation-induced endothelium damage: Emerging models and technologies.\u003c/em\u003e Radiother Oncol, 2021. \u003cstrong\u003e158\u003c/strong\u003e: p. 21-32.\u003c/li\u003e\n\u003cli\u003eYang, J., et al., \u003cem\u003eGuidelines and definitions for research on epithelial-mesenchymal transition.\u003c/em\u003e Nat Rev Mol Cell Biol, 2020. \u003cstrong\u003e21\u003c/strong\u003e(6): p. 341-352.\u003c/li\u003e\n\u003cli\u003eYoussef, K.K. and M.A. Nieto, \u003cem\u003eEpithelial-mesenchymal transition in tissue repair and degeneration.\u003c/em\u003e Nat Rev Mol Cell Biol, 2024. \u003cstrong\u003e25\u003c/strong\u003e(9): p. 720-739.\u003c/li\u003e\n\u003cli\u003eJackson, I.L., et al., \u003cem\u003eGene expression profiles among murine strains segregate with distinct differences in the progression of radiation-induced lung disease.\u003c/em\u003e Dis Model Mech, 2017. \u003cstrong\u003e10\u003c/strong\u003e(4): p. 425-437.\u003c/li\u003e\n\u003cli\u003eJorgensen, A., et al., \u003cem\u003eOxidative Stress-Induced Damage to RNA and DNA and Mortality in Individuals with Psychiatric Illness.\u003c/em\u003e JAMA Psychiatry, 2024. \u003cstrong\u003e81\u003c/strong\u003e(5): p. 516-520.\u003c/li\u003e\n\u003cli\u003eMajchrzak-Gorecka, M., et al., \u003cem\u003eSecretory leukocyte protease inhibitor (SLPI), a multifunctional protein in the host defense response.\u003c/em\u003e Cytokine Growth Factor Rev, 2016. \u003cstrong\u003e28\u003c/strong\u003e: p. 79-93.\u003c/li\u003e\n\u003cli\u003ePriya Dharshini, L.C., et al., \u003cem\u003eOxidative stress responsive transcription factors in cellular signalling transduction mechanisms.\u003c/em\u003e Cell Signal, 2020. \u003cstrong\u003e72\u003c/strong\u003e: p. 109670.\u003c/li\u003e\n\u003cli\u003eNguyen, T.M. and M. Aragona, \u003cem\u003eStress-responsive transcription factors train stem cells to remember.\u003c/em\u003e Cell Stem Cell, 2021. \u003cstrong\u003e28\u003c/strong\u003e(10): p. 1679-1680.\u003c/li\u003e\n\u003cli\u003eBauer, M., et al., \u003cem\u003eHuman monocytes are severely impaired in base and DNA double-strand break repair that renders them vulnerable to oxidative stress.\u003c/em\u003e Proc Natl Acad Sci U S A, 2011. \u003cstrong\u003e108\u003c/strong\u003e(52): p. 21105-10.\u003c/li\u003e\n\u003cli\u003eMalainou, C., et al., \u003cem\u003eAlveolar macrophages in tissue homeostasis, inflammation, and infection: evolving concepts of therapeutic targeting.\u003c/em\u003e J Clin Invest, 2023. \u003cstrong\u003e133\u003c/strong\u003e(19).\u003c/li\u003e\n\u003cli\u003eXu, X., et al., \u003cem\u003ePhosphorylation-Mediated IFN-gammaR2 Membrane Translocation Is Required to Activate Macrophage Innate Response.\u003c/em\u003e Cell, 2018. \u003cstrong\u003e175\u003c/strong\u003e(5): p. 1336-1351 e17.\u003c/li\u003e\n\u003cli\u003eCalandra, T. and T. Roger, \u003cem\u003eMacrophage migration inhibitory factor: a regulator of innate immunity.\u003c/em\u003e Nat Rev Immunol, 2003. \u003cstrong\u003e3\u003c/strong\u003e(10): p. 791-800.\u003c/li\u003e\n\u003cli\u003eGreten, F.R. and S.I. Grivennikov, \u003cem\u003eInflammation and Cancer: Triggers, Mechanisms, and Consequences.\u003c/em\u003e Immunity, 2019. \u003cstrong\u003e51\u003c/strong\u003e(1): p. 27-41.\u003c/li\u003e\n\u003cli\u003eGronke, K., et al., \u003cem\u003eInterleukin-22 protects intestinal stem cells against genotoxic stress.\u003c/em\u003e Nature, 2019. \u003cstrong\u003e566\u003c/strong\u003e(7743): p. 249-253.\u003c/li\u003e\n\u003cli\u003ePollard, J.W., \u003cem\u003eTumour-educated macrophages promote tumour progression and metastasis.\u003c/em\u003e Nat Rev Cancer, 2004. \u003cstrong\u003e4\u003c/strong\u003e(1): p. 71-8.\u003c/li\u003e\n\u003cli\u003eSingh, N., et al., \u003cem\u003eInflammation and cancer.\u003c/em\u003e Ann Afr Med, 2019. \u003cstrong\u003e18\u003c/strong\u003e(3): p. 121-126.\u003c/li\u003e\n\u003cli\u003eTeng, F., M. Li, and J. Yu, \u003cem\u003eRadiation recall pneumonitis induced by PD-1/PD-L1 blockades: mechanisms and therapeutic implications.\u003c/em\u003e BMC Med, 2020. \u003cstrong\u003e18\u003c/strong\u003e(1): p. 275.\u003c/li\u003e\n\u003cli\u003eWirsdorfer, F. and V. Jendrossek, \u003cem\u003eThe Role of Lymphocytes in Radiotherapy-Induced Adverse Late Effects in the Lung.\u003c/em\u003e Front Immunol, 2016. \u003cstrong\u003e7\u003c/strong\u003e: p. 591.\u003c/li\u003e\n\u003cli\u003eZheng, M.Z.M. and L.M. Wakim, \u003cem\u003eTissue resident memory T cells in the respiratory tract.\u003c/em\u003e Mucosal Immunol, 2022. \u003cstrong\u003e15\u003c/strong\u003e(3): p. 379-388.\u003c/li\u003e\n\u003cli\u003eYenyuwadee, S., et al., \u003cem\u003eThe evolving role of tissue-resident memory T cells in infections and cancer.\u003c/em\u003e Sci Adv, 2022. \u003cstrong\u003e8\u003c/strong\u003e(33): p. eabo5871.\u003c/li\u003e\n\u003cli\u003eSchneider, J.L., et al., \u003cem\u003eThe aging lung: Physiology, disease, and immunity.\u003c/em\u003e Cell, 2021. \u003cstrong\u003e184\u003c/strong\u003e(8): p. 1990-2019.\u003c/li\u003e\n\u003cli\u003eParimon, T., et al., \u003cem\u003eSenescence of alveolar epithelial progenitor cells: a critical driver of lung fibrosis.\u003c/em\u003e Am J Physiol Cell Physiol, 2023. \u003cstrong\u003e325\u003c/strong\u003e(2): p. C483-C495.\u003c/li\u003e\n\u003cli\u003eZhang, T., et al., \u003cem\u003eFullerenols Mitigate Radiation-Induced Myocardial Injury.\u003c/em\u003e Adv Healthc Mater, 2023. \u003cstrong\u003e12\u003c/strong\u003e(29): p. e2300819.\u003c/li\u003e\n\u003cli\u003eWalker, J.A. and A.N.J. McKenzie, \u003cem\u003eT(H)2 cell development and function.\u003c/em\u003e Nat Rev Immunol, 2018. \u003cstrong\u003e18\u003c/strong\u003e(2): p. 121-133.\u003c/li\u003e\n\u003cli\u003eAlshoubaki, Y.K., et al., \u003cem\u003eTregs delivered post-myocardial infarction adopt an injury-specific phenotype promoting cardiac repair via macrophages in mice.\u003c/em\u003e Nat Commun, 2024. \u003cstrong\u003e15\u003c/strong\u003e(1): p. 6480.\u003c/li\u003e\n\u003cli\u003eIlicic, T., et al., \u003cem\u003eClassification of low quality cells from single-cell RNA-seq data.\u003c/em\u003e Genome Biol, 2016. \u003cstrong\u003e17\u003c/strong\u003e: p. 29.\u003c/li\u003e\n\u003cli\u003ePijuan-Sala, B., et al., \u003cem\u003eA single-cell molecular map of mouse gastrulation and early organogenesis.\u003c/em\u003e Nature, 2019. \u003cstrong\u003e566\u003c/strong\u003e(7745): p. 490-495.\u003c/li\u003e\n\u003cli\u003eMcGinnis, C.S., L.M. Murrow, and Z.J. Gartner, \u003cem\u003eDoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors.\u003c/em\u003e Cell Syst, 2019. \u003cstrong\u003e8\u003c/strong\u003e(4): p. 329-337 e4.\u003c/li\u003e\n\u003cli\u003eHao, Y., et al., \u003cem\u003eDictionary learning for integrative, multimodal and scalable single-cell analysis.\u003c/em\u003e Nat Biotechnol, 2024. \u003cstrong\u003e42\u003c/strong\u003e(2): p. 293-304.\u003c/li\u003e\n\u003cli\u003ePaczkowska, M., et al., \u003cem\u003eIntegrative pathway enrichment analysis of multivariate omics data.\u003c/em\u003e Nat Commun, 2020. \u003cstrong\u003e11\u003c/strong\u003e(1): p. 735.\u003c/li\u003e\n\u003cli\u003eLa Manno, G., et al., \u003cem\u003eRNA velocity of single cells.\u003c/em\u003e Nature, 2018. \u003cstrong\u003e560\u003c/strong\u003e(7719): p. 494-498.\u003c/li\u003e\n\u003cli\u003eGulati, G.S., et al., \u003cem\u003eSingle-cell transcriptional diversity is a hallmark of developmental potential.\u003c/em\u003e Science, 2020. \u003cstrong\u003e367\u003c/strong\u003e(6476): p. 405-411.\u003c/li\u003e\n\u003cli\u003eCao, J., et al., \u003cem\u003eThe single-cell transcriptional landscape of mammalian organogenesis.\u003c/em\u003e Nature, 2019. \u003cstrong\u003e566\u003c/strong\u003e(7745): p. 496-502.\u003c/li\u003e\n\u003cli\u003eJin, S., et al., \u003cem\u003eInference and analysis of cell-cell communication using CellChat.\u003c/em\u003e Nat Commun, 2021. \u003cstrong\u003e12\u003c/strong\u003e(1): p. 1088.\u003c/li\u003e\n\u003cli\u003eLiberzon, A., et al., \u003cem\u003eThe Molecular Signatures Database (MSigDB) hallmark gene set collection.\u003c/em\u003e Cell Syst, 2015. \u003cstrong\u003e1\u003c/strong\u003e(6): p. 417-425.\u003c/li\u003e\n\u003cli\u003eHanzelmann, S., R. Castelo, and J. Guinney, \u003cem\u003eGSVA: gene set variation analysis for microarray and RNA-seq data.\u003c/em\u003e BMC Bioinformatics, 2013. \u003cstrong\u003e14\u003c/strong\u003e: p. 7.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"The Fifth Hospital of Guangzhou Medical University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Radiation pneumonitis, Single-cell dynamics, pro-inflammatory response, MIF signaling, Acute lung injury","lastPublishedDoi":"10.21203/rs.3.rs-7659149/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7659149/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eRadiation pneumonitis (RP) is a deleterious complication of thoracic radiotherapy, yet the cellular mechanisms driving its onset and progression remain unclear. Here we constructed a single-cell dynamic architecture of RP rats with acute lung injury at multiple time points after radiation from 84,865 high-quality cells through single-cell RNA sequencing.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eEndothelial and epithelial cells are damaged within 24 hours after radiation, while epithelial-mesenchymal transitions (EMT) occur in RP lesions at 1\u0026ndash;2 weeks. Identification of radiation-induced EMT signature highly correlated with and superior to known EMT signature. Radiation induces oxidative stress and promotes apoptosis in monocytes one week after radiation exposure, and the induced inflammation persists. Macrophage components enhance the pro-inflammatory response following radiation via MIF signaling and exhibit four distinct intercellular communication patterns. The ligand \u003cem\u003eMif\u003c/em\u003e was associated with radiation-induced expression enhancement, and its blockade alleviated pneumonia symptoms. The dynamics and differentiation of lymphocytes reveal that effector and helper T cells activate within 2\u0026ndash;4 weeks post-radiation, while tissue-resident memory T cells proliferate at 6 weeks.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eThis RP architecture provides a comprehensive view of the cellular architecture and dynamics following radiation exposure, enhancing our understanding of RP\u0026rsquo;s pathogenesis and offering biomarkers and potential therapeutic targets for early diagnosis and intervention.\u003c/p\u003e","manuscriptTitle":"A dynamically resolved single-cell architecture of radiation pneumonitis provides insights into acute lung injury","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-23 06:12:28","doi":"10.21203/rs.3.rs-7659149/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"348189cf-7b9e-4806-b827-d4b83d1a0ee8","owner":[],"postedDate":"September 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55018214,"name":"Molecular Biology"},{"id":55018215,"name":"Bioinformatics"}],"tags":[],"updatedAt":"2025-09-23T06:12:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-23 06:12:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7659149","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7659149","identity":"rs-7659149","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}