IL10 promotes surgical incision healing in sepsis by regulating PI16+ fibroblasts and CCL8+ macrophages Authors: Xiaojun He, Cheng Xu, Yan Zou, Wei Wang, Liwei Wu, Quanhong Zhou Published: 2026 ABSTRACT AbstractSepsis severely impairs surgical incision healing, but the underlying mechanisms and effective interventions remain insufficiently understood. Interleukin-10 (IL10) exerts potent anti-inflammatory activity, yet its role in regulating fibroblast and macrophage dynamics during septic wound repair is unclear. Human peri-incisional skin specimens were collected from septic and non-septic patients. A lipopolysaccharide (LPS)-induced sepsis mouse model with surgical incision was established, followed by IL10 treatment. Single-cell RNA sequencing, cell trajectory analysis, cell–cell communication analysis, gene set variation analysis, and immunofluorescence were used to investigate cellular and molecular alterations. Only mouse samples were used for single-cell sequencing in this study. Single-cell analysis identified 8 cell types, including 3 non-immune cell types and 5 immune cell types. Sepsis increased NK cells and macrophages, and upregulated PI16⁺ fibroblasts and CCL8⁺ macrophages in wound tissue. IL10 treatment elevated mast cells, T cells, and neutrophils, downregulated peptidase inhibitor 16 (Pi16) and chemokine ligand 8 (Ccl8) expression, and attenuated fibroblast–macrophage crosstalk. Mechanistically, IL10 promoted fibroblast proliferation by inducing macrophages to secrete Il1b and Osm. Sepsis delays surgical incision healing via upregulating PI16⁺ fibroblasts and CCL8⁺ macrophages and enhancing inflammatory cell crosstalk. IL10 accelerates wound repair by suppressing these detrimental subsets and activating Il1b/Osm-mediated fibroblast proliferation. PI16 and CCL8 may serve as potential therapeutic targets for septic wound healing.Keywords: Surgical incision healing, Sepsis, IL10, PI16+ fibroblast, CCL8+ macrophage Introduction Sepsis is a severe systemic infectious disease that poses a significant threat to patients’ lives and health [1, 2]. It can lead to multiple organ dysfunction, increase mortality rates, and impose a heavy economic burden on the healthcare system [3]. Notably, in patients with sepsis, the healing of surgical incisions is significantly delayed [4, 5]. Currently, the diagnosis and treatment of sepsis include anti-infective therapy and organ function support [6]. However, these methods have limitations such as complex conditions and difficult treatment. Although IL10 has anti-inflammatory effects, its mechanism of action in the healing of surgical incisions in sepsis patients remains unknown [7]. Fibroblasts and macrophages are of particular interest in this study as they play crucial roles in tissue repair and immune regulation [8, 9]. Fibroblasts are key in the process of skin incision healing, while macrophages are important in the inflammatory response. Their association with the occurrence and development of sepsis makes them potential targets for understanding the mechanism of how sepsis affects incision healing. In this study, we used a variety of experimental techniques and methods, including collecting human skin samples and establishing a sepsis mouse model. We conducted single-cell analysis, cell trajectory analysis, cell communication analysis, enrichment analysis, and immunofluorescence experiments, and employed corresponding statistical methods for data analysis. These methods allow for in-depth exploration of the effects of sepsis on skin tissue cells and the interactions between cells from multiple perspectives. Despite progress in understanding sepsis-related impaired wound healing, critical knowledge gaps remain. The cellular and molecular mechanisms by which sepsis disrupts fibroblast and macrophage homeostasis to delay skin incision repair have not been fully elucidated at single-cell resolution. Furthermore, it remains unclear whether IL10 accelerates septic wound healing by targeting specific fibroblast and macrophage subsets and regulating the underlying intercellular crosstalk networks. This study was designed to define the mechanisms by which IL10 modulates PI16+ fibroblasts and CCL8+ macrophages to rescue impaired surgical incision healing in sepsis. Materials and methods Human Samples The collection of human skin samples was completed in the intensive care unit of the Sixth People’s Hospital of Shanghai. Skin tissue around the surgical incisions was collected from 5 patients who did not develop sepsis post-surgery, and from 5 patients who developed sepsis post-surgery. All 10 patients were male, with an average age of 53.5 years, and all had undergone Dixon surgery (Table 1). All patients included in the study signed informed consent forms. This study was approved by the Ethics Committee of the Sixth People’s Hospital of Shanghai (2024-KY-012). Table 1Baseline characteristics of patients undergone Dixon surgeryCharacteristicNon-sepsis post-surgery (n = 5)Sepsis post-surgery (n = 5)Total (n = 5)Sex, n (%) Male5 (100)5 (100)10 (100) Female0 (0)0 (0)0 (0)Age, years51.355.653.5Incision healing time, days11.622.417Length of hospital stay, days18.227.522.9 Baseline characteristics of patients undergone Dixon surgery Sepsis mouse model Eight-week-old male C57BL/6J mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were administered an intraperitoneal injection of LPS (Beyotime, CAT# ST1470, 12 mg/kg) to establish a sepsis model. Thirty minutes after the model was established, 0.5 ml of physiological saline was injected intraperitoneally every 2 h, for a total of three times. Before establishing the sepsis mouse model, we removed the abdominal hair of the mice, used a sterilized surgical blade to make an incision on the abdomen, and sutured it to simulate the healing process of a surgical wound. IL10 (MCE, CAT# HY-P70517, 45 µg/kg) was administered via tail vein injection for the treatment of sepsis, while the control group received an injection of the same volume of physiological saline. Seventy-two hours after the model was established, the mice were euthanized, and skin tissue surrounding the surgical incision was collected for further experiments. Single-cell analysis workflow Skin samples were collected from three mice in the control group, three LPS‑induced septic mice, and three septic mice treated with IL10. All samples were preserved in tissue storage solution and subjected to single‑cell capture and library preparation using the 10× Genomics Chromium Single Cell 3′ Solution followed by sequencing on an Illumina platform. Upstream data processing was performed by Shanghai OE Company using the Cell Ranger pipeline (version 3.0.1). Raw BCL sequencing data were converted to FASTQ files using Illumina BCL Convert. Reads were aligned to the mouse reference genome (mm10) using the STAR aligner. Cell barcodes and UMIs were processed to correct sequencing errors and remove PCR duplicates. A filtered feature‑barcode expression matrix was generated after excluding low‑quality barcodes and ambient RNA contamination. The expression matrix was then analyzed in R using the Seurat package [10]. Low‑quality cells were excluded using strict QC criteria: nCount_RNA > 1000, nFeature_RNA > 600 and < 5000, percent.mt < 30%. Data were normalized using the LogNormalize method, and highly variable genes were identified. Principal component analysis (PCA) was performed for dimensionality reduction, and batch effects were removed using the Harmony package. Unsupervised clustering was performed with a resolution of 0.4, and visualization was achieved using uniform manifold approximation and projection (UMAP). Cell types were manually annotated based on canonical marker genes with reference to the CellMarker database [11]. Cell trajectory analysis We used the subset function to extract all fibroblasts and macrophages and generate a monocle object. The monocle package was used for trajectory analysis [12–14]. The FindAllMarkers function was employed to identify differentially expressed genes (DEGs) among cells, where genes with an adjusted P value less than 0.05 were considered statistically significant. These DEGs were used to calculate pseudotime scores, and the DDRTree method was applied to order all cells based on their pseudotime scores. Subsequently, we utilized the visualization functions provided by monocle to visualize the data. Cell communication analysis We used the CellChat package to conduct cell communication analysis [15, 16]. The Seurat object was utilized to generate the CellChat object. The mouse receptor-ligand database was designated as the reference database for analysis. Subsequently, we established the cell communication network using the following functions: subsetData, identifyOverExpressedGenes, identifyOverExpressedInteractions, computeCommunProb, computeCommunProbPathway, aggregateNet, and netAnalysis_computeCentrality. After constructing the cell communication network, we used the visualization functions provided by CellChat for data visualization. Enrichment analysis We used Gene Set Variation Analysis (GSVA) for enrichment analysis [17]. GSVA performs enrichment analysis based on gene expression levels of different Seurat clusters, achieving better results than other enrichment analyses. We employed Gene Ontology Biological Process (GOBP) reference data for enrichment analysis, considering enrichment terms with an adjusted P value of less than 0.05 to be statistically significant. We used ggplot2 for data visualization. Immunofluorescence All skin tissue samples were embedded in paraffin and subsequently sectioned. The obtained tissue sections were deparaffinized and hydrated, and ethylenediaminetetraacetic acid (EDTA) was used for antigen retrieval. After antigen retrieval, the sections were blocked with 10-fold diluted donkey serum for one hour. Subsequently, the primary antibodies, diluted appropriately, were added to the sections and incubated overnight at 4 °C. The primary antibodies used were as follows: anti-Pi16 (Invitrogen, CAT# PA5-145657, 1:100), anti-Vimentin (Invitrogen, CAT# MA5-11883, 1:200), anti-CD68 (Invitrogen, CAT# 14-0688-82, 1:200), anti-Ki67 (Proteintech, CAT# 28074-1-AP, 1:500) and anti-CCL8 (Invitrogen, CAT# PA5-95539, 1:200). After primary antibody incubation, the sections were incubated with Goat anti-Rabbit 488 (Invitrogen, CAT# A-11008) and Goat anti-Mouse 594 (Invitrogen, CAT# A-11005) fluorescent secondary antibodies. Finally, we stained the nuclei with DAPI and collected data using a confocal microscope (Leica, STELLARIS). Statistical analysis All statistical analyses were performed using R and Graphpad Prism 9 software. The Kolmogorov-Smirnov test and Shapiro-Wilk test were used to assess the normality of the data. The student’s t-test was used for normally distributed data, and the non-parametric rank-sum test was used for non-normally distributed data. For data from more than two groups that were normally distributed, a One-way ANOVA was conducted, whereas the Kruskal-Wallis H test was used for non-normally distributed data. A P-value less than 0.05 was considered statistically significant. Results Overview of single-cell analysis A total of 21,589 cells were classified into eight cell types. Immune cells include macrophages, neutrophils, T cells, NK cells, and mast cells; non-immune cells include fibroblasts, endothelial cells, and epithelial cells (Fig. 1A). Skin samples from mice in the control group, sepsis group, and IL10 treatment group collectively formed the basis for single-cell analysis (Fig. 1B). The distribution of all cells across these three groups was relatively uniform, underscoring the importance of removing batch effects. We manually annotated the cell types: Col1a1 is a marker gene for fibroblasts, Ctss for macrophages, Cdh5 for endothelial cells, Krt5 for epithelial cells, S100a8 for neutrophils, Camk4 for T cells, Gzma for NK cells, and Mcpt8 for mast cells (Figs. 1C and D). In sepsis induced by LPS stimulation, the proportions of NK cells and macrophages increased, whereas following IL10 treatment, the proportions of mast cells, T cells, and neutrophils increased. The proportions of non-immune cells—fibroblasts, endothelial cells, and epithelial cells—did not change significantly (Fig. 1E). This suggests that in LPS-stimulated sepsis, NK cells and macrophages infiltrate the skin tissue noticeably, potentially affecting incision healing. After IL10 treatment, mast cells, T cells, and neutrophils infiltrate the skin tissue significantly, possibly activating anti-inflammatory signaling pathways. GSVA enrichment analysis for all cell types indicates different functions undertaken by various cells, highlighting the accuracy of our cell annotation (Figs. 1F and 2A). The expression level of KI67 in skin tissue was markedly decreased in septic mice, but significantly restored following IL‑10 administration (Fig. 2B and C). These results further confirm the presence of delayed cutaneous healing in septic mice and demonstrate that IL‑10 can effectively reverse this impairment. Overview of single-cell analysis. A, UMAP dimensionality reduction plot of eight cell types, with different colors representing different cell types. B, UMAP dimensionality reduction plot of three groups of cells, with different colors representing different groups. C, Marker gene expression profile of eight cell types, where the size of the dots represents the percentage of gene expression, red indicates high gene expression, and blue indicates low gene expression. D, Expression profile of representative marker genes for eight cell types. E, Proportion of cells for each of the eight cell types. F, Biological function enrichment of the eight cell types, where red indicates high enrichment scores and blue indicates low enrichment scores Analysis of single-cell populations. A, Top 5 biological functions enriched in eight cell types are shown, with red indicating a high enrichment score and blue indicating a low enrichment score. B, Immunofluorescence of tissue surrounding a mouse skin incision; blue represents nuclei and green represents KI67. C, the quantification of immunofluorescence PI16+ fibroblasts were upregulated in septic skin tissue Fibroblasts are a key cell type in the process of skin incision healing, so we focused on the transcriptional characteristics of fibroblasts in septic conditions and after IL10 anti-inflammatory treatment. A total of 5,620 fibroblasts were divided into 15 cell subpopulations (Fig. 3A). Fibroblasts from the three groups did not show obvious heterogeneity (Fig. 3B). The proportion of C1 and C9 fibroblasts was significantly upregulated in the LPS sepsis group, while the proportion of C2 fibroblasts was significantly upregulated in the IL10 treatment group (Fig. 3C). This suggests that C1 and C9 fibroblasts may play a critical role in skin incision healing under septic conditions. C2 fibroblasts may be activated during IL10 anti-inflammatory treatment, potentially accelerating wound healing. GSVA enrichment analysis showed that C1 fibroblasts are negatively regulated for growth factor stimulation, further suggesting that C1 fibroblasts may be an important cell subpopulation hindering wound healing under septic conditions (Fig. 3D). The marker gene for C1 fibroblasts is Pi16 (Fig. 3E). Studies have confirmed that overexpression of Pi16 in cardiac fibroblasts can significantly inhibit fibroblast proliferation and ultimately inhibit cardiac fibrosis [18]. Therefore, we have reason to infer that fibroblast highly express Pi16 under septic conditions, hindering the healing of surgical incisions. By further analyzing the transcriptional characteristics of Pi16, we found high expression of Pi16 in fibroblasts in the LPS group, and after IL10 anti-inflammatory treatment, the expression level of Pi16 in fibroblasts was significantly reduced (Fig. 3F). This further indicates that high expression of Pi16 in fibroblasts under septic inflammatory stimulation may hinder the healing of surgical incisions. All fibroblasts can be divided into three different pseudo-time states, with state 3 fibroblasts having the highest pseudo-time scores (Fig. 3G). Moreover, Pi16 expression is highest in state 3 fibroblasts, most of which are composed of LPS fibroblasts (Figs. 3H and I). Cells with higher pseudo-time scores have a more mature state, and Pi16 may make fibroblasts more mature, thus losing their proliferative potential. We validated the results of bioinformatics analysis through immunofluorescence in human skin tissue. More PI16-expressing fibroblasts were observed in the tissue surrounding skin incisions in septic patients (Fig. 3J). Figure 4A shows the top 3 biological functions of each fibroblast subpopulation. Furthermore, we validated this in mouse skin tissue using immunofluorescence. More PI16-expressing fibroblasts were observed in the tissue surrounding skin incisions in LPS-treated mice, while treatment with IL10 significantly inhibited fibroblast expression of PI16 (Fig. 4B). This suggests that IL10 may inhibit the expression of PI16 in fibroblasts, thereby promoting postoperative incision healing. Single-cell subgroup analysis of fibroblasts. A, UMAP dimensionality reduction plot of 15 fibroblast subgroups, with different colors representing different subgroups. B, UMAP dimensionality reduction plot of three major fibroblast groups, with different colors representing different groups. C, Proportions of cells in the 15 fibroblast subgroups. D, Biological function enrichment of the 15 fibroblast subgroups, with red representing a high enrichment score and blue representing a low enrichment score. E, Representative marker genes of the 15 fibroblast subgroups, with the size of the dots representing the percentage of gene expression, and red representing high expression levels and green representing low expression levels. F, Expression profile of Pi16 in the three major fibroblast groups. G, Pseudotime branching plot of fibroblasts, with different colors representing cells in different states. H, Pseudotime expression profile of Pi16, with cells colored by different states. I, Pseudotime expression profile of Pi16, with cells colored by different groups. J, Immunofluorescence of human skin incision surrounding tissue, with blue representing nuclei, red representing Vimentin, and green representing PI16. K, the quantification of immunofluorescence Analysis of single-cell subpopulations of fibroblasts. A, Top 3 biological function enrichment of 15 fibroblast subpopulations; red indicates high enrichment scores, blue indicates low enrichment scores. B, Immunofluorescence of tissue surrounding a mouse skin incision; blue represents nuclei, red represents Vimentin, and green represents PI16. C, the quantification of immunofluorescence CCL8+ macrophages were upregulated in septic skin tissue Macrophages play a key role in the inflammatory response, so we explored the transcriptional characteristics of macrophages in the septic state and after IL10 treatment. A total of 2,681 macrophages were divided into 14 cell subpopulations (Fig. 5A). Macrophages from the three groups did not exhibit significant heterogeneity (Fig. 5B). C8 macrophages had the highest proportion in the LPS group, while C9 macrophages had the highest proportion in the IL10 treatment group (Fig. 5C). This suggests that C8 macrophages may play an important role in wound healing post-sepsis, while C9 cells are significantly activated after IL10 treatment. GSVA enrichment analysis revealed that C8 macrophages function in negatively regulating fibroblast growth factors, while C9 macrophages regulate endothelial cell differentiation (Fig. 5D). This indicates that C8 macrophages may impede wound healing in septic skin tissue since incision healing relies on fibroblast proliferation, and C8 macrophages negatively regulate fibroblast growth factors, potentially inhibiting fibroblast proliferation. The marker gene for C8 macrophages is Ccl8 (Fig. 5E). CCL8, as a chemokine, plays a crucial role in immune regulation, involving macrophage recruitment, inflammation promotion, and anti-tumor activities [19–21]. By further exploring Ccl8’s expression profile, we found elevated Ccl8 expression levels in LPS macrophages (Fig. 5F). Therefore, we infer that under septic conditions, macrophage expression of Ccl8 promotes inflammation, which, due to severe inflammation, impedes fibroblast proliferation and ultimately hinders surgical incision healing. All macrophages were categorized into 12 distinct states based on pseudotime scores (Fig. 5G). Ccl8 is primarily expressed in macrophages of state 12, most of which are composed of LPS group macrophages (Fig. 5H and I). This also indirectly confirms Ccl8’s unique expression characteristics. We discovered macrophage infiltration in the skin tissue surrounding human surgical incisions, and septic group macrophages expressed higher levels of CCL8 (Fig. 5J). Figure 6A presents the top 3 biological functions of each macrophage subpopulation. Furthermore, we found more CCL8-expressing macrophages in the skin tissue around incisions in LPS septic mice, while macrophages expressed less CCL8 after IL10 treatment (Fig. 6B). This further suggests IL10’s potential anti-inflammatory effects, which may have a significant therapeutic impact on surgical incision healing in septic patients. Analysis of Single-Cell Subsets of Macrophages. A, UMAP dimensionality reduction plot of 14 macrophage subsets, with different colors representing different subsets. B, UMAP dimensionality reduction plot of macrophages from three groups, with different colors indicating different groups. C, Proportion of cells in the 14 macrophage subsets. D, Biological function enrichment of 14 macrophage subsets, with red indicating high enrichment scores and blue indicating low enrichment scores. E, Representative marker genes of the 14 macrophage subsets, with the size of the dots representing the percentage of gene expression, red indicating high expression, and green indicating low expression. F, Expression profile of Ccl8 in macrophages from the three groups. G, Pseudotime trajectory of macrophages, with different colors representing cells in different states. H, Pseudotime expression profile of Ccl8 with cells colored by different states. I, Pseudotime expression profile of Ccl8 with cells colored by different groups. J, Immunofluorescence of tissue around human skin incisions, with blue representing nuclei, green representing CD68, and red representing CCL8. K, the quantification of immunofluorescence Single-cell subpopulation analysis of macrophages. A, Top 3 biological function enrichments of 14 macrophage subpopulations, with red indicating high enrichment scores and blue indicating low enrichment scores. B, Immunofluorescence of mouse skin incision surrounding tissue, with blue representing nuclei, green representing CD68, and red representing CCL8. C, the quantification of immunofluorescence Ccl8 secreted by macrophages acts on fibroblasts under septic conditions To further explore the connection between macrophages and fibroblasts, we conducted cell communication analysis. Cells in the LPS group exhibited a higher cell communication intensity (Fig. 7A). This suggests that under septic conditions, the enhanced inflammatory response may promote intercellular crosstalk. We found that the crosstalk intensity between fibroblasts and macrophages was the highest (Fig. 7B). This further confirms our previous results that macrophages have a role in regulating fibroblasts. Further analysis of activated receptor-ligand pairs revealed that the LPS group had more activated receptor-ligand pairs (Fig. 7C). Signals from both macrophages and fibroblasts were significantly increased in the LPS group (Figs. 7D and E). Focusing on the CCL pathway, we found that signals from macrophages and fibroblasts were significantly activated in the LPS group (Fig. 7F). We then observed that in the LPS group, not only did CCL8 secreted by macrophages bind to Ackr1 on fibroblasts, but three other receptor-ligand pairs, Ccl7-Ackr1, Ccl5-Ackr1, and Ccl2-Ackr1, were also activated (Fig. 7G). This suggests that Ackr1 may bind with chemokines to inhibit fibroblast proliferation, which in turn inhibits the healing of surgical wounds under septic conditions. Analysis of cell communication in the control group and LPS group. A, the intensity of cell communication in the control group and LPS group. B, the quantity of cell communication between cells, with red indicating a higher quantity and blue indicating a lower quantity. C, signaling pathways enriched in the control group and LPS group. D, the number of cell communications in the control group and LPS group. E, the number of cell communications between fibroblasts and macrophages in both groups. F, the activation status of the CCL signaling pathway between the two groups. G, the activation status of receptor-ligand pairs between fibroblasts and macrophages in both groups; the size of the dots represents statistical significance, with red indicating significant receptor-ligand pair activation and blue indicating non-activation Il1b and Osm secreted by macrophages act on fibroblasts after IL10 treatment To further explore the connection between macrophages and fibroblasts after IL10 treatment, we conducted cell communication analysis separately for the two groups of cells. The cell communication intensity in the IL10 treatment group was lower than that in the LPS group (Fig. 8A). This indicates that the anti-inflammatory effect of IL10 reduces intercellular crosstalk. After IL10 treatment, we found that the crosstalk intensity between macrophages and fibroblasts was no longer the highest (Fig. 8B). This suggests that the treatment effect of IL10 might be mediated through macrophages. In the IL10 treatment group, the IL1 and OSM signaling pathways were significantly activated (Fig. 8C). The cell connectivity diagram more intuitively reveals the reduced cell communication intensity between macrophages and fibroblasts (Fig. 8D). The IL1 signaling pathway is mainly activated in neutrophils, epithelial cells, macrophages, fibroblasts, and mast cells, while the OSM signaling pathway is mainly activated in cells other than T cells and NK cells (Fig. 8E). In the IL10 treatment group, Il1b and Osm secreted by macrophages act on fibroblasts (Figs. 8F and G). Some studies have indicated that Il1b can promote the proliferation of smooth muscle cells, while Osm can promote the proliferation of cardiomyocytes [22, 23]. Therefore, IL10 treatment may activate macrophages to secrete Il1b and Osm to jointly promote the proliferation of fibroblasts, thereby promoting the healing of surgical incisions. This might be a new mechanism by which IL10 promotes surgical incision repair, in addition to its anti-inflammatory effects. Analysis of cell communication in LPS and IL10 treatment groups. A, the strength of cell communication in the LPS and IL10 treatment groups. B, the number of cell communications between cells, with red indicating a higher number and blue indicating a lower number. C, signaling pathways enriched in the LPS and IL10 treatment groups. D, the number of cell communications in the LPS and IL10 treatment groups. E, the activation status of the IL1 and OSM signaling pathways in the IL10 treatment group. F and G, the activation status of receptor-ligand pairs between fibroblasts and macrophages in both groups. The size of the dots represents statistical significance, with red indicating significant activation of receptor-ligand pairs and blue indicating non-activation Discussion Sepsis is a severe systemic infectious disease that poses a significant threat to patients’ lives. It often leads to slow healing of surgical incisions [5]. This study focuses on exploring the impact of sepsis on surgical incision healing and the underlying mechanisms. This research utilizes various techniques, including collecting human skin samples and establishing a sepsis mouse model. Through single-cell analysis, cell trajectory analysis, cell communication analysis, enrichment analysis, and immunofluorescence experiments, we aim to investigate the changes in skin tissue cells under septic conditions and after IL10 treatment. We found a significant infiltration of PI16+ fibroblasts and CCL8+ macrophages in the tissue surrounding skin incisions under septic conditions. After IL10 treatment, the infiltration of PI16+ fibroblasts and CCL8+ macrophages were significantly reduced. PI16 inhibits the proliferation and migration of fibroblasts, which may lead to delayed wound healing. CCL8 is closely related to the inflammatory response and may also cause delayed wound healing through inflammation. The findings of this study are expected to provide new insights into the mechanisms of skin incision healing in sepsis and potential therapeutic targets for clinical treatment. The finding that NK cells and macrophages increase in LPS-stimulated sepsis, while mast cells, T cells, and neutrophils increase after IL10 treatment, provides valuable insights. Further research is needed to elucidate the specific mechanisms by which these cell types contribute to sepsis and the healing process. Additionally, improving the cell classification method to enhance the accuracy of the analysis is crucial. Exploring the relationship between cell proportion changes and clinical symptoms could offer a better understanding of the disease progression. Other factors that may influence these cell type proportions also warrant investigation. Finally, analyzing the significance of the interaction between these cell types in sepsis and treatment is essential for developing more effective therapeutic strategies. Understanding the specific impact of Pi16 on fibroblast function is of great significance. Further validation of Pi16’s role in fibroblasts is necessary to confirm our hypotheses. Exploring the regulatory factors of Pi16 expression could provide additional insights into the mechanisms underlying fibroblast behavior in sepsis. Investigating the effects of IL10 treatment on other characteristics of fibroblasts could help us better understand the therapeutic potential of IL10. Analyzing the correlation between Pi16 expression and the severity of sepsis could aid in prognosis and treatment planning. Elucidating the specific mechanism by which Ccl8 affects macrophage function is crucial for understanding its role in sepsis. Further studies on Ccl8 in sepsis are needed to fully comprehend its implications. Exploring the factors that regulate Ccl8 expression could provide potential targets for therapeutic intervention. Investigating the effects of IL10 treatment on other macrophage characteristics could enhance our understanding of its therapeutic effects. Analyzing the relationship between Ccl8 expression and the prognosis of sepsis could guide clinical decision-making. Identifying the specific roles of these receptor-ligand pairs in the development of sepsis is essential for developing targeted therapeutic strategies. Interventions to modulate this cell-cell interaction could improve the prognosis of sepsis. Exploring whether other cell types are involved in this interaction could provide a more comprehensive understanding of the disease mechanism. Studying the changes in this interaction in different stages of sepsis could help tailor treatment approaches. Analyzing the relationship between this interaction and sepsis complications could inform clinical management. Unraveling the specific mechanisms by which Il1b and Osm promote fibroblast proliferation is crucial for optimizing the therapeutic effects of IL10. Exploring ways to optimize IL10 treatment to enhance its wound-healing promoting effects could lead to improved clinical outcomes. Investigating the impact of IL10 treatment on other cell-cell interactions could provide a broader perspective on its therapeutic potential. Studying the correlation between Il1b and Osm expression and the therapeutic effects of IL10 could guide treatment decisions. Analyzing the application potential of IL10 treatment in different types of wound healing could expand its clinical utility. Collectively, our study identifies a previously unrecognized cellular and molecular mechanism underlying sepsis-impaired surgical incision healing. At single-cell resolution, we demonstrate that sepsis drives the accumulation of PI16+ fibroblasts and CCL8+ macrophages in wounded skin, which mediate fibroblast growth inhibition and excessive inflammatory crosstalk to delay tissue repair. Importantly, we reveal that IL10 reverses this pathological state by suppressing Pi16 expression in fibroblasts and Ccl8 production in macrophages, while redirecting macrophage function to secrete Il1b and Osm that promote fibroblast proliferation. These findings establish PI16+ fibroblasts and CCL8+ macrophages as critical cellular drivers of delayed healing in sepsis, and define IL10-mediated modulation of fibroblast–macrophage dynamics as a novel therapeutic axis for improving surgical wound recovery in septic conditions. In conclusion, this study using human skin samples and sepsis mouse models revealed the impacts of sepsis and IL10 treatment on cell changes and the related mechanisms in the process of skin incision healing. It was found that in sepsis, the proportions of NK cells and macrophages increased, and PI16+ fibroblasts and CCL8+ macrophages were upregulated. Moreover, there were specific interactions between macrophages and fibroblasts. After IL10 treatment, changes in cell proportions and interactions occurred, possibly promoting fibroblast proliferation and surgical incision healing by regulating macrophage secretion of Il1b and Osm. These results provide an important basis for understanding the mechanism of skin incision healing under sepsis and offer ideas for the development of more effective treatment methods. In clinical practice within the intensive care unit, there are cases where postoperative patients develop hospital-acquired infections, leading to sepsis. The surgical incision healing in sepsis patients is often delayed. Therefore, this study offers a unique perspective on the specific role of IL10 in promoting the healing of surgical incisions in sepsis patients. Additionally, the study identifies PI16 and CCL8 as two targets for future drug development, aiming to provide evidence support for clinical practice as much as possible. However, the study has certain limitations and requires further research to improve and validate these findings. Acknowledgements Not applicable. Author contributions Xiaojun He wrote the main manuscript text. Cheng Xu, Yan Zou and Wei Wang prepared Figs. 1, 2, 3, 4, 5, 6, 7 and 8. Liwei Wu and Quanhong Zhou reviewed the manuscript. Funding This work was supported by the Project of National Natural Science Foundation of China (81771933). Data availability The single-cell sequencing data generated in this study have been deposited in the National Genomics Data Center (NGDC), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, under the accession number CRA041981. These data are publicly accessible free of charge via the NGDC official website (https://ngdc.cncb.ac.cn/), following the platform’s access guidelines to ensure reproducibility and transparency of the research. All datasets supporting the findings of this study are available through this accession number, and no restrictions are imposed on their use for academic research purposes. Declarations Ethics approval and consent to participate This study has been approved by the Ethics Committee of Shanghai Sixth People’s Hospital (2024-KY-012), and informed consent has been obtained from the patients or their relative guardians. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Xiaojun He and Cheng Xu contributed equally to this work and shared first authorship. Liwei Wu and Quanhong Zhou contributed equally to this work and shared last authorship. Contributor Information Liwei Wu, Email: docwuliwei@foxmail.com. Quanhong Zhou, Email: zhouanny@hotmail.com. References Associated Data Data Availability Statement The single-cell sequencing data generated in this study have been deposited in the National Genomics Data Center (NGDC), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, under the accession number CRA041981. These data are publicly accessible free of charge via the NGDC official website (https://ngdc.cncb.ac.cn/), following the platform’s access guidelines to ensure reproducibility and transparency of the research. All datasets supporting the findings of this study are available through this accession number, and no restrictions are imposed on their use for academic research purposes.