BACKGROUND:Acetaminophen (APAP) is a widely used analgesic and antipyretic agent that is safe at therapeutic doses. However, due to its extensive misuse, APAP-induced liver injury has become a major public health concern. Although mesenchymal stem/stromal cells (MSCs) represent a promising emerging therapy for APAP-induced liver injury, considerable research has focused on enhancing their efficacy, notably through genetic modification. Fibroblast growth factor 21 (FGF21), an endocrine hormone activated by metabolic stress, is known to regulate energy homeostasis, glucose and lipid metabolism, and to promote the homing of MSCs to sites of injury. Consequently, this study was designed to determine whether genetically engineering MSCs to overexpress FGF21 (FGF21-MSCs) augments their therapeutic potential against APAP-induced acute liver injury (ALI). METHODS:In this investigation, MSCs were employed as a platform for FGF21 gene delivery. The MSCs were transduced with lentiviral vectors encoding the FGF21 gene to facilitate sustained FGF21 overexpression. We subsequently assessed the therapeutic potential of these FGF21-MSCs in a murine model of APAP-induced ALI. The extent of liver injury was comprehensively evaluated. Furthermore, the underlying mechanisms were elucidated using techniques including immunohistochemistry, immunofluorescence, and flow cytometry. RESULTS:Our results demonstrated that FGF21-MSCs significantly enhanced the therapeutic efficacy of conventional MSCs against APAP-induced ALI via a biphasic mechanism: attenuating oxidative stress and inflammation during the acute injury phase, while actively fostering tissue repair during the subsequent regenerative phase. This protective effect is primarily mediated through the enhancement of macrophage phagocytic capacity, thereby accelerating tissue repair and regeneration. CONCLUSIONS:Our findings demonstrate that FGF21-MSCs significantly augment therapeutic efficacy against APAP-induced ALI in mice, thereby yielding critical insights and revealing novel therapeutic targets for ALI prevention and treatment.

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by persistent synovial inflammation, progressive bone erosion, and persistent pain, with limited therapeutic options for comprehensive disease control. Growth differentiation factor 15 (GDF15), a stress-responsive cytokine, has been implicated in inflammation and pain regulation, but its role in RA is not fully understood. Here, we investigated the expression, function, and underlying mechanisms of GDF15 in RA. We found that GDF15 was significantly upregulated in the serum and synovial tissues of patients with RA and in collagen-induced arthritis (CIA) mice. Functional studies showed that Gdf15 deficiency exacerbated arthritis severity, synovial inflammation, pannus formation, bone destruction, and mechanical hypersensitivity in collagen antibody-induced arthritis (CAIA) mice, whereas administration of recombinant GDF15 alleviated disease progression, reduced joint inflammation and bone loss, and improved pain-related behaviors in CIA mice. Mechanistically, GDF15 inhibited osteoclast differentiation and resorptive activity without affecting precursor viability, through suppression of NF-κB signaling and attenuation of calcium-dependent NFATc1 activation. In addition, GDF15 decreased osteoclast-derived Netrin-1 expression and reduced CGRP+ sensory nerve fiber density in the synovium. Rescue experiments demonstrated that exogenous Netrin-1 supplementation partially counteracted the beneficial effects of GDF15 on pain-related behaviors and CGRP+ sensory nerve remodeling in CAIA mice, further supporting the involvement of the osteoclast-derived Netrin-1/CGRP+ sensory nerve axis in GDF15-mediated pain regulation. Collectively, these findings identify GDF15 as an endogenous protective factor in inflammatory arthritis that limits osteoclastogenesis, bone destruction, and pain-related nerve remodeling, highlighting its potential as a therapeutic target for RA.

Objective To analyze the influencing factors for lower extremity deep vein thrombosis(DVT),a common complication after total hip arthroplasty(TH A),in elderly patients.Methods A retrospective analysis was conducted on 378 patients who underwent THA in Northern Jiangsu People's Hospital from September 2022 to September 2023.The patients were divided into a DVT group(n=34)and a non-DVT group(n=344)according to whether DVT was happened or not.Univariate analysis was performed to compare preoperative,intraoperative and postoperative clinical data between the two groups.Multivariate logistic regression analysis was employed to identify independent influencing factors for DVT following THA and to construct a predictive model.Results Univariate analysis showed there were significant differences between the two groups(P＜0.05)in the following factors:age ≥65 years(x2=8.765),diabetes(x2=7.462),intraoperative blood loss(x2=5.451),blood transfusion(x2=69.232),preoperative activated partial thromboplastin time(APTT)(Z=3.317),preoperative albumin(Z=3.974),preoperative hemoglobin(Z=3.341),postoperative hemoglobin(t=2.030),and postoperative albumin(Z=4.268).Multivariate logistic regression analysis indicated that age,diabetes,intraoperative blood loss,blood transfusion,preoperative APTT,and postoperative albumin were independent influencing factors for lower extremity DVT after THA(P＜0.05).A nomogram risk prediction model was developed,and the calibration curve demonstrated good fit.The predictive performance of the nomogram was evaluated using the concordance index,which indicated that the model had favorable predictive ability.Conclusion Age,diabetes,intraoperative blood loss,intraoperative blood transfusion,preoperative APTT,and postoperative albumin are independent influencing factors for DVT following THA.The constructed nomogram model exhibits good predictive value.

Bladder cancer (BCa) mortality is mainly driven by metastatic dissemination and an immunosuppressive tumor microenvironment. Here, we identify ELN (tropoelastin), an extracellular matrix protein abundantly secreted by cancer-associated fibroblasts (CAFs), as a critical determinant of these processes and a marker of poor prognosis. ELN promotes epithelial-mesenchymal transition (EMT), facilitates lymphatic spread, and induces immune dysfunction characterized by macrophage polarization toward an M2 phenotype and T cell exhaustion. Mechanistically, ELN functions as a binding partner of TGF-β receptor 2 (TGFBR2), thereby triggering SMAD2/3-dependent TGF-β1 secretion and establishing a feed forward signaling loop. This ELN/TGFBR2/TGF-β1 axis amplifies metastatic capacity and immunosuppressive signaling, ultimately accelerating disease progression and diminishing responsiveness to immune checkpoint blockade. Functional studies in BCa organoids and murine models demonstrated that pharmacologic blockade of the ELN-TGFBR2 interaction effectively suppressed tumor metastasis and restored antitumor immunity. Collectively, our findings establish ELN as a CAF-derived driver of metastasis and immune evasion in BCa. Targeting the ELN-TGFBR2 interaction offers a promising therapeutic strategy to limit metastatic progression and enhance the efficacy of immunotherapy in this lethal disease.

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a liver manifestation of metabolic syndrome characterized by excessive hepatic lipid accumulation and lipid metabolism disorders. It has become the most common chronic liver disease worldwide. β-arrestin2 is a multifunctional scaffold protein that is among the most important regulatory molecules, and it exerts key roles in regulating various cellular processes, such as immune response, cellular collagen production, and inflammation. In the current study, we aimed to explore the function of β-arrestin2 in the development and progression of MASLD. Firstly, we observed that the expression of β-arrestin2 was upregulated in liver samples from patients with MASLD. Then, the western diet (WD) combined with CCl4 injection-induced MASLD was established in wild-type mice, and showed that liver β-arrestin2 expression was also gradually increased, and positively correlated with the degree of lipid metabolism disorder during MASLD progression. Ulteriorly, β-arrestin2 knockout (Arrb2 KO) mice were utilized to induce the MASLD model and found that β-arrestin2 deficiency significantly ameliorated lipid accumulation and inflammatory response in the liver of MASLD mice. Furthermore, the in vitro depletion and overexpression experiments showed that increased β-arrestin2 aggravated lipid accumulation via inhibiting the activation of the TAK1/AMPK pathway, which may be mediated by competitively binding to TAB1 with TAK1. These findings suggest that β-arrestin2 is essential to regulate intrahepatic lipid metabolism. Here, we provide a novel insight in understanding of the expression and function of β-arrestin2 in MASLD, demonstrating that it may be a potential therapeutic target for MASLD treatment.

The dome-shaped cornea is a transparent, non-vascularized, and epithelialized highly organized tissue. Physical and chemical injuries may trigger corneal wound healing (CWH) response and result in neovascularization that impairs the visual function. CWH involves not only migration, proliferation, and differentiation of the cells in different layers of cornea, but also the mobilization of immune cells. We demonstrated here that human adipose-derived mesenchymal stromal cells (ADSCs) could effectively inhibit neovascularization during ethanol-induced injury in mouse cornea. Importantly, we found that while neutrophils are essential for CWH, excessive and prolonged neutrophil retention during the granulation stage contributes to neovascularization. ADSCs were found to promote the clearance of neutrophils in the cornea during the granulation stage, likely via increasing the reverse transendothelial cell migration of CXCR4high neutrophils from cornea to the lung. Our results demonstrate that ADSCs are effective in treating CWH-induced neovascularization and modulation of neutrophil clearance could be novel strategies for better vision recovery after injury.

BackgroundSystemic lupus erythematosus (SLE) is a persistent autoimmune disorder marked by dysregulation of the immune system, resulting in extensive tissue inflammation and subsequent damage. Fibroblasts are essential contributors to the pathogenesis of SLE, particularly in driving the progression of tissue fibrosis and inflammation. Recent research has proposed that the GEM gene may regulate fibroblast activity in SLE. However, the precise molecular mechanisms through which GEM modulates fibroblast functions in the context of SLE are yet to be fully elucidated. Gaining insight into these mechanisms is crucial for uncovering potential therapeutic targets aimed at addressing fibrosis and inflammation associated with SLE.MethodsSingle-cell RNA sequencing was integrated with cell-based assays, such as quantitative reverse transcription PCR (qRT-PCR) and functional cellular experiments, to investigate the underlying mechanisms. The regulatory mechanisms of GEM in fibroblasts were analyzed through functional cell assays.ResultsDifferential gene expression in fibroblast subpopulations was identified through single-cell RNA sequencing, with GEM emerging as a key gene implicated in these alterations. Trajectory analysis indicated that GEM expression correlated with fibroblast proliferation and migration. Subsequent experiments confirmed that GEM regulates fibroblast viability and influences SLE disease progression through modulation of cell proliferation, migration, and apoptosis.ConclusionsGEM is highly differentially expressed in fibroblast subpopulations within SLE, and its altered expression impacts fibroblast proliferation and migration. GEM may regulate fibroblast activity and apoptosis, potentially contributing to the progression of SLE.

Mucosal-associated invariant T (MAIT) cells are a highly conserved population of immune cells that can be activated via the major histocompatibility complex class I-related protein pathway or cytokine pathways, playing a central role in immune surveillance. This review provides comprehensive information on their thymic developmental origin, tissue-specific distribution, and microbial regulatory networks, with a focus on analyzing the bidirectional regulatory mechanisms in diseases. In infectious diseases, MAIT cells eliminate pathogens through the rapid release of cytokines; however, sustained antigen exposure leads to functional exhaustion. In autoimmune diseases, their migration disorders and proinflammatory cytokine secretion of MAIT cells exacerbate tissue damage. In the tumor microenvironment, they play a paradoxical role, being capable of mediating antitumor effects while also being reprogrammed into a protumor phenotype. Based on their tissue targeting ability and functional plasticity, we discuss novel strategies for targeted therapy, including engineering chimeric antigen receptor-MAIT cells to enhance tumor killing, blocking exhaustion pathways to reverse functional impairment, and regulating the microbiota-metabolic axis to reprogram cell activity. This review integrates cutting-edge evidence, reveals the translational potential of MAIT cells as a cross-disease regulatory hub, and provides a theoretical framework for precision immunotherapy.

Background & Aims: Treatments directly targeting fibrosis remain limited. Given the unique intrinsic features of macrophages and their capacity to engraft in the liver, we genetically engineered bone marrow -derived macrophages with a chimeric antigen receptor (CAR) to direct their phagocytic activity against hepatic stellate cells (HSCs) in multiple mouse models. This study aimed to demonstrate the therapeutic efficacy of CAR macrophages (CAR -Ms) in mouse models of fibrosis and cirrhosis and to elucidate the underlying mechanisms. Methods: uPAR expression was studied in patients with fibrosis/cirrhosis and in murine models of liver fibrosis, including mice treated with carbon tetrachloride, a 5-diethoxycarbonyl-1, 4-dihydrocollidine diet, or a high-fat/cholesterol/fructose diet. The safety and efficacy of CAR -Ms were evaluated in vitro and in vivo . Results: Adoptive transfer of CAR -Ms resulted in a significant reduction in liver fibrosis and the restoration of function in murine models of liver fibrosis. CAR -Ms modulated the hepatic immune microenvironment to recruit and modify the activation of endogenous immune cells to drive fibrosis regression. These CAR -Ms were able to recruit and present antigens to T cells and mount specific antifibrotic T -cell responses to reduce fibroblasts and liver fibrosis in mice. Conclusion: Collectively, our findings demonstrate the potential of using macrophages as a platform for CAR technology to provide an effective treatment option for liver fibrosis. CAR -Ms might be developed for treatment of patients with liver fibrosis. (c) 2024 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

BACKGROUND:Poor neoplastic differentiation contributes to the rapid progression of uterine corpus endometrial carcinoma (UCEC). Thus, it is essential to identify candidate genes, clarifying the carcinogenesis and progression of UCEC. METHODS:We screened genes that affect differentiation and prognosis in UCEC. Least absolute selection and shrinkage operator (LASSO) regression, univariate Cox, and multivariate Cox proportional risk regression analyses were performed to screen out γ-glutamyl hydrolase (GGH) as the candidate gene. The clinical value of GGH on prognosis was evaluated. The relationship between GGH and immune infiltration was assessed by CIBERSORT, EPIC, ssGSEA, unsupervised clustering and immunohistochemistry (IHC). Additionally, we investigated the effect of GGH knockdown in vitro. RESULTS:Among the GGH, CDKN2A, and SIX1 genes, the impact of GGH was predominant on immune infiltration in UCEC. A nomogram containing GGH and other clinical features showed good predictive performance via curve analysis (DCA). In the functional analysis, GGH affected differentiation, tumour proliferation, and immune regulation. The immunosuppressive components were enriched in the GGH-high group, with poor immunotherapy efficacy. The study suggests that GGH may influence the progression of UCEC by regulating the glycolytic process. CONCLUSIONS:GGH is closely associated with various immune cell infiltrations. Our study demonstrates the prognostic role of GGH in carcinogenesis in UCEC.

INTRODUCTION:Osteoarthritis (OA) is a prevalent clinical chronic degenerative condition characterized by the degeneration of articular cartilage. Currently, drug treatments for OA come with varying degrees of side effects, making the development of new therapeutic approaches for OA imperative. Mesenchymal stem cells (MSCs) are known to mitigate the progression of OA primarily through paracrine effects. The conditioned medium (CM) derived from MSCs encapsulates a variety of paracrine factors secreted by these cells. METHODS:In this study, we investigated the effect of the CM of infrapatellar fat pad-derived MSCs (IPFSCs) on OA in vitro and in vivo, as well as and the potential underlying mechanisms. We established three experimental groups: the normal group, the OA group, and the CM intervention group. In vitro experiments, we used methods such as qPCR, Western blot, immunofluorescence, and flow cytometry to detect the impact of CM on OA chondrocytes. In vivo experiments, we evaluated the changes in the knee joints of OA rats after intra-articular injection of CM treatment. RESULTS:The results showed that injection of CM into the knee joint inhibited OA development in a rat model induced by destabilization of the medial meniscus and anterior cruciate ligament transection. The CM increased the deposition of extracellular matrix-related components (type II collagen and Proteoglycan). The activation of PI3K/AKT/NF-κB signaling pathway was induced by IL-1β in chondrocytes, which was finally inhibited by CM-IPFSCs treatment. CONCLUSION:In summary, IPFSCs-CM may have therapeutic potential for OA.

BACKGROUND:Alcohol-associated liver disease (ALD) is a major health care challenge worldwide with limited therapeutic options. Although mesenchymal stem/stromal cells (MSCs) represent a newly emerging therapeutic approach to treat ALD, thus far, there have been extensive efforts to try and enhance their efficacy, including genetically engineering MSCs. FGF21, an endocrine stress-responsive hormone, has been shown to regulate energy balance, glucose, and lipid metabolism and to enhance the homing of MSCs toward injured sites. Therefore, the purpose of this study was to investigate whether MSCs that overexpress FGF21 (FGF21-MSCs) improve the therapeutic effect of MSCs in treating ALD. METHODS:Human umbilical cord-derived MSCs served as the gene delivery vehicle for the FGF21 gene. Human umbilical cord-derived MSCs were transduced with the FGF21 gene using lentiviral vectors to mediate FGF21 overexpression. We utilized both chronic Lieber-DeCarli and Gao-binge models of ethanol-induced liver injury to observe the therapeutic effect of FGF21-MSCs. Liver injury was phenotypically evaluated by performing biochemical methods, histology, and inflammatory cytokine levels. RESULTS:Compared with MSCs alone, administration of MSCs overexpressing FGF21(FGF21-MSCs) treatment significantly enhanced the therapeutic effect of ALD in mice, as indicated by the alleviation of liver injury with reduced steatosis, inflammatory infiltration, oxidative stress, and hepatic apoptosis, and the promotion of liver regeneration. Mechanistically, FGF21 could facilitate the immunomodulatory function of MSCs on macrophages by setting metabolic commitment for oxidative phosphorylation, which enables macrophages to exhibit anti-inflammatory inclination. CONCLUSIONS:Our data elucidate that MSC modification by FGF21 could enhance their therapeutic effect in ALD and may help in the exploration of effective MSCs-based cell therapies for the treatment of ALD.

The potent immunomodulatory function of mesenchymal stem/stromal cells (MSCs) elicited by proinflammatory cytokines IFN-γ and TNF-α (IT) is critical to resolve inflammation and promote tissue repair. However, little is known about how the immunomodulatory capability of MSCs is related to their differentiation competency in the inflammatory microenvironment. In this study, we demonstrate that the adipocyte differentiation and immunomodulatory function of human adipose tissue-derived MSCs (MSC(AD)s) are mutually exclusive. Mitochondrial reactive oxygen species (mtROS), which promote adipocyte differentiation, were decreased in MSC(AD)s due to IT-induced upregulation of superoxide dismutase 2 (SOD2). Furthermore, knockdown of SOD2 led to enhanced adipogenic differentiation but reduced immunosuppression capability of MSC(AD)s. Interestingly, the adipogenic differentiation was associated with increased mitochondrial biogenesis and upregulation of peroxisome proliferator-activated receptor gamma coactivator-1 alpha (PPARGC1A/PGC-1α) expression. IT inhibited PGC-1α expression and decreased mitochondrial mass but promoted glycolysis in an SOD2-dependent manner. MSC(AD)s lacking SOD2 were compromised in their therapeutic efficacy in DSS-induced colitis in mice. Taken together, these findings indicate that the adipogenic differentiation and immunomodulation of MSC(AD)s may compete for resources in fulfilling the respective biosynthetic needs. Blocking of adipogenic differentiation by mitochondrial antioxidant may represent a novel strategy to enhance the immunosuppressive activity of MSCs in the inflammatory microenvironment.

Osteoarthritis (OA) is a degenerative disease characterized by cartilage wear and degradation. Ginkgolide K (GK) is a natural compound extracted from Ginkgo biloba leaves and possesses anti-inflammatory and anti-apoptotic effects. We found that the biological characteristics of GK were highly consistent with those of OA medications. This study aimed to determine and verify the therapeutic effect of GK on OA and mechanism of its therapeutic effect. For the in vivo experiment, OA rats were regularly injected in the articular cavity with GK, and the curative effects were observed after 4 and 8 weeks. For the in vitro experiment, we treated OA chondrocytes with different concentrations of GK and then detected the related indices of OA. Through the in vivo and in vitro experiments, we found that GK could promote the production of major components of the cartilage extracellular matrix. Transcriptome sequencing revealed that GK may activate hypoxia-inducible factor 1 alpha via the hypoxia signaling pathway, which, in turn, activates yes-associated protein and inhibits apoptosis of OA chondrocytes. GK has a therapeutic effect on OA and, therefore, has the potential to be developed into a new drug for OA treatment.

Knee arthrofibrosis is one of the most serious complications of knee surgery; however, its pathogenesis is unclear, and current treatment methods have not achieved satisfactory results. Mesenchymal stem cells (MSCs) have good anti-inflammatory and antifibrotic properties, and studies have reported that human infrapatellar fat pad-derived MSCs (IPFSCs) have the advantages of strong proliferative and differentiating ability, ease of acquisition, and minimal harm to the donor. Increasing evidence has shown that MSCs function through their paracrine extracellular vesicles (EVs). Our study is aimed at exploring the effects of human IPFSC-derived EVs (IPFSC-EVs) on knee arthrofibrosis and the underlying mechanisms in vivo and in vitro. In the in vivo study, injecting IPFSC-EVs into the knee joint cavity effectively reduced surgery-induced knee arthrofibrosis in rats. In the in vitro study, IPFSC-EVs were found to inhibit the proliferation of fibroblasts in the inflammatory environment. Additionally, we screened a potential IPFSC-EV molecular target, metallothionein 2A (MT2A), using RNA sequencing. We found that silencing MT2A partially reversed the inhibitory effect of IPFSC-EVs on fibroblast proliferation in the inflammatory environment. In conclusion, IPFSC-EVs inhibit the progression of knee arthrofibrosis by regulating MT2A, which inhibits fibroblast proliferation in the inflammatory environment.

The thymus is required for T cell development and the formation of the adaptive immunity. Stromal cells, which include thymic epithelial cells (TECs) and mesenchymal stromal cells (MSCs), are essential for thymic function. However, the immunomodulatory function of thymus-derived MSCs (T-MSCs) has not been fully explored. MSCs were isolated from mouse thymus and their general characteristics including surface markers and multi-differentiation potential were characterized. The immunomodulatory function of T-MSCs stimulated by IFN-γ and TNF-α was evaluated in vitro and in vivo. Furthermore, the spatial distribution of MSCs in the thymus was interrogated by using tdTomato-flox mice corssed to various MSC lineage Cre recombinase lines. A subset of T-MSCs express Nestin, and are mainly distributed in the thymic medulla region and cortical-medulla junction, but not in the capsule. The Nestin-positive T-MSCs exhibit typical immunophenotypic characteristics and differentiation potential. Additionally, when stimulated with IFN-γ and TNF-α, they can inhibit activated T lymphocytes as efficiently as BM-MSCs, and this function is dependent on the production of nitric oxide (NO). Additionally, the T-MSCs exhibit a remarkable therapeutic efficacy in acute liver injury and inflammatory bowel disease (IBD). Nestin-positive MSCs are mainly distributed in medulla and cortical-medulla junction in thymus and possess immunosuppressive ability upon stimulation by inflammatory cytokines. The findings have implications in understanding the physiological function of MSCs in thymus.

Additional File 1: Supplementary Table 1

BACKGROUND & AIMS:Liver fibrosis/cirrhosis is significant health burden worldwide, resulting in liver failure or cancer and accounting for many deaths each year. The pathogenesis of liver fibrosis is very complex, which makes treatment challenging. Growth differentiation factor 15 (GDF15), a cysteine knot protein belonging to the transforming growth factor β (TGF-β) superfamily, has been shown to play a protective role after tissue injury and to promote a negative energy balance during obesity and diabetes. However, paucity of literature is available about GDF15 function in liver fibrosis. This study aimed to investigate the immunomodulatory role and therapeutic potential of GDF15 in progression of hepatic fibrosis. METHODS:GDF15 expression was studied in patients with fibrosis/cirrhosis and in 2 murine models of liver fibrosis, including mice treated with CCl4 or DDC diet. GDF15 involvement in the pathogenesis of liver fibrosis was assessed in Gdf15 knockout mouse using both CCl4 and DDC diet experimental models. We used the CCl4 and/or DDC diet-induced liver fibrosis model to examine the antifibrotic and anti-inflammatory effects of AAV8-mediated GDF15 overexpression in hepatocytes or recombinant mouse GDF15. RESULTS:GDF15 expression is decreased in the liver of animal models and patients with liver fibrosis/cirrhosis compared with those without liver disease. In vivo studies showed that GDF15 deficiency aggravated CCl4 and DDC diet-induced liver fibrosis, while GDF15 overexpression mediated by AAV8 or its recombinant protein alleviated CCl4 and/or DDC diet-induced liver fibrosis. In Gdf15 knockout mice, the intrahepatic microenvironment that developed during fibrosis showed relatively more inflammation, as demonstrated by enhanced infiltration of monocytes and neutrophils and increased expression of proinflammatory factors, which could be diminished by AAV8-mediated GDF15 overexpression in hepatocytes. Intriguingly, GDF15 exerts its effects by reprogramming the metabolic pathways of macrophages to acquire an oxidative phosphorylation-dependent anti-inflammatory functional fate. Furthermore, adoptive transfer of GDF15-preprogrammed macrophages to mouse models of liver fibrosis induced by CCl4 attenuated inflammation and alleviated the progression of liver fibrosis. CONCLUSION:GDF15 ameliorates liver fibrosis via modulation of liver macrophages. Our data implicate the importance of the liver microenvironment in macrophage programming during liver fibrosis and suggest that GDF15 is a potentially attractive therapeutic target for the treatment of patients with liver fibrosis.

Dear Editor, Immune-related cell (IRCs)-based machine learning (ML) models, including random forest (RF), multilayer perceptron (MLP), generalised linear model (GLM) and gradient boosting machine (GBM), have shown great performance in the estimation of alcoholic liver disease (ALD). Pathological biopsy of the liver is considered the most reliable method for determining the diagnosis and evaluating the staging and prognosis of liver disease, but liver biopsy can cause complications and therefore lacks specific noninvasive diagnostic biomarkers.1 IRCs are biomarkers of systemic inflammation, and identifying changes in IRCs associated with ALD will contribute to improving the diagnosis of ALD.2, 3 However, current biomarker discoveries are usually focused on an individual biomolecule, resulting in low clinical applicability,4, 5 and few studies have investigated diagnostic value of IRCs in ALD. Therefore, there is urgent ongoing research exploring an accurate and sensitive noninvasive test that is low cost and low risk. Recently, ML has been widely used in biomedical research and disease diagnosis, and ML is a promising effective method in the identification of hepatic fibrosis and cirrhosis.3, 4 However, no studies have investigated the role of ML model, especially IRC-based ML model, in the diagnosis, disease progression and prognosis of ALD. In this study, we developed IRC-based ML models to assess ALD progression and prognosis. Overall, 207 ALD patients (including alcoholic fatty liver (AFL) and alcoholic cirrhosis (ALC)) and 234 healthy controls (HCs) were included (Figure 1A and Table 1). ML had great performance in the estimation of ALD. To explore the role of IRCs in disease onset, we compared the differences between ALD and HCs. Gender, age, MCV, platelets, RBCs, neutrophil/lymphocyte ratio (NLR) and monocyte/neutrophil ratio (MNR) were included in the nomogram using LASSO regression (Figure 1B–D). Decision curve analysis (DCA) curve found that the net benefit of the nomogram was 0.03 to 0.99 (Figure 1C), suggesting that the predictive ability and accuracy of model fitting are high. Here, we introduced the findings of the RF model, as it had the greatest performance (Figure 1E–H and Table S1). AUCs of PR and ROC were 0.9982 and 0.9975 in the training set, and 0.9984 and 0.9978 in the testing set, respectively (Figure 1E and F and Table 1). In addition, the evaluation of ML (i.e., root mean square error (RMSE) and mean square error (MSE)) is shown in Table S2, indicating that the ML models are robust and reliable. These results suggested that ML models, especially the RF model, not only have high precision but also have high predictability. We conducted additional ML models for ALD using variables not screened by least absolute shrinkage and selection operator (LASSO) regression as sensitivity analyses (Figure 1I–L). Results were consistent with regression screening for most outcomes, which means that our results are robust and that we can obtain good predictions by using several important variables instead of all variables. To further investigate the role of IRCs in the disease progression of ALD, we conducted subgroup analyses in AFL and ALC patients. First, 13 variables may be potential risk factors for AFL patients, and 8 variables including gender, basophils, MCV, lymphocytes, neutrophils, platelets, NLR and MNR were included. The RF model had the greatest performance, and AUCs of PR and ROC were 0.9984 and 0.9990 in the training set, and 1.0000 and 1.0000 in the testing set, respectively (Figure 2A–G and Table 1). Second, 17 variables may be related to ALC patients, and 10 variables were eventually left in the LASSO regression and nomogram (Figure 2H–N), that is, age, gender, MCV, lymphocyte, platelet, RBC, platelet/lymphocyte ratio (PLR), platelet/neutrophil ratio (PNR), MNR and NLR. The RF model had the greatest performance (Figure 2M and N and Table 1). Finally, 13 variables may be related to disease progression when comparing ALC with AFL, and 11 variables were eventually left in the LASSO regression and nomogram. The RF model had the greatest performance (Table 1 and Figure S1). To further explore the role of IRCs in the prognosis of ALD, MELD (model for end stage liver disease) and MDF (Maddrey's discriminant function) score-based ML were used to predict disease severity. MELD-associated LASSO regression found that basophils, MCV, neutrophils, RBC and white blood cells (WBCs) were potential prognostic factors (Figure 3A–G). The RF model had high performance (Figure 3F andG and Table 1). In addition, MDF-associated LASSO regression found that eosinophils, platelets, RBCs, PMR and PNR were potential prognostic factors (Figure 3H–N). The MLP model had high performance (Figure 3M andN and Table 1). These significant IRCs and the best model of MDF differ from the MELD-based model. One possible explanation is that the evaluation method is different between the two scores. ML represents the future of clinical medicine and will be implemented in healthcare systems to facilitate early identification of target populations. We used four ML models to predict ALD, which were constructed based on IRCs that have significant roles in liver injury, fibrogenesis and regeneration.6 Different IRCs play different leading roles in disease progression and prognosis, which means that appropriate cells are needed to assess the different stages of the disease. Neutrophils, the most common components of circulating WBCs, are implicated in the immune pathogenesis of ALD with complex and multifaceted properties, that is, neutrophils directly cause hepatocyte injury and liver inflammation in ALD, while neutrophil-induced cytokines have a protective role in inflammation and liver repair.7, 8 In addition to neutrophils alone, the effects of neutrophil and immune cell interaction (e.g., PNR, MNR and NLR) were also relevant features of the ML models. Platelet interactions with IRCs, especially neutrophils, have been extensively investigated in a variety of diseases6, 9 except liver diseases. Future studies are highly recommended to explore the diversity and complexity of neutrophil functions to formulate targeted interventions and treatments for ALD. In addition, systemic inflammation is often accompanied by changes in RBCs. For example, RBCs increase the risk of liver disease, and RBC count is related to fatty liver index and disease progression.10 In conclusion, IRCs and their interactions are of great importance in ALD, and IRC-based ML models, especially the RF model, are accurate and inexpensive tools for identifying the progression and prognosis of ALD. This study could be used as a basis for the development of ML models for disease prediction. We thank all the subjects who participated in this study. The authors declare no conflicts interests. XZ is supported by the National Natural Science Foundation of China (82100628), the Natural Science Foundation of Anhui Province (2108085QH313), the Postdoctoral Research Foundation of China (2021M700183B496). YS is supported by the National Natural Science Foundation of China (82100627) and the Natural Science Foundation of Anhui Province (2108085QH311). Raw data are available on reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Objective To investigate the effect of IFN-α-2b in preventing postoperative arthrofibrosis in rats, its antiproliferation effect on fibroblasts in vitro, and its molecular mechanism. Methods The rat model of arthrofibrosis was established and treated with different concentrations of drugs. Knee specimens were collected for histological and immunohistochemical staining to observe the effect of IFN-α-2b on arthrofibrosis in rats. The biological information was further mined according to the database data, and the possible regulatory mechanism of IFN-α-2b on fibroblasts was analyzed. The inhibitory effect of IFN-α-2b on fibroblast proliferation and migration in vitro was detected by cell counting kit-8 (CCK-8), immunofluorescence analysis, cell cycle test, EdU assay, wound healing test, and Transwell method, and the analysis results were verified by Western blotting method. Results The test results of rat knee joint specimens showed that IFN-α-2b significantly inhibited the degree of fibrosis after knee joint surgery, the number of fibroblasts in the operation area was less than that of the control group, and the expression of collagen and proliferation-related proteins decreased. In vitro experimental results show that IFN-α-2b can inhibit the proliferation and migration of fibroblasts. According to the results of database analysis, it is suggested that the STAT1/P21 pathway may be involved, and it has been verified and confirmed by Western blotting and other related methods. Conclusion IFN-α-2b can reduce surgery-induced arthrofibrosis by inhibiting fibroblast proliferation and migration, which may be related to the regulation of STAT1/p21 signaling pathway.