Abstract
The incidence of nonalcoholic steatohepatitis (NASH) is increasing worldwide, and effective treatment is urgently needed. To understand the molecular mechanisms behind the effectiveness of bariatric surgery in treating NASH, we integrated single-cell and bulk RNA sequencing data to identify the role of liver macrophage polarization in alleviating NASH and screen possible drugs for treatment. Analysis revealed that bariatric surgery alleviates NASH by inhibiting liver M1 macrophage polarization with 12 differentially expressed M1 macrophage-related genes. Additionally, 56 potentially effective drugs were predicted for NASH treatment. These findings shed light on the effectiveness of bariatric surgery in treating NASH and offer potential drug candidates for further exploration.
Similar content being viewed by others
Introduction
Nonalcoholic fatty liver disease (NAFLD) has become a serious disease affecting human health, with a global prevalence rate of approximately 25% [1]. NAFLD is highly heterogeneous and can be divided into nonalcoholic fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH) according to the different histological types [2]. NAFL is characterized by steatosis of hepatocytes due to lipid accumulation, on the basis of which NASH necrotizes inflammation and hepatocyte injury [3]. NASH may progress to cirrhosis and eventually develop into hepatocellular carcinoma or end-stage liver disease [4]. In addition, as the hepatic manifestation of metabolic syndrome, NASH is always accompanied by severe extrahepatic diseases, including dyslipidemia, hypertension, obesity, and type 2 diabetes mellitus (T2DM) [3]. The pathogenesis of NASH is incompletely understood, and the only drug currently approved by the U.S. Food and Drug Administration for the treatment of NASH is Rezdiffra [5].
Liver macrophages, including inherent liver macrophages (Kupffer cells) and monocyte-derived macrophages, are involved in the regulation of liver immune homeostasis and the development of a variety of liver diseases (such as NASH, chronic viral hepatitis, and liver cancer) [6]. Moreover, liver macrophages are characterized by heterogeneity and plasticity, resulting in dynamic changes in the occurrence and development of NASH[7]. Macrophages are usually polarized into two subtypes: M1 macrophages, which are “proinflammatory”, and M2 macrophages, which are “anti-inflammatory” [8]. Liver macrophages can be activated by lipopolysaccharide (LPS) or excessive free fatty acids (FFAs) and release many proinflammatory cytokines [9]. Through paracrine signaling, Kupffer cells aggravate lipid accumulation and oxidative stress in hepatocytes and cause liver progression from simple steatosis to NASH and even fibrosis [9]. During this process, the number of M1 macrophages increases significantly, whereas the number of M2 macrophages decreases substantially. There is a dynamic balance in the proportion of M1/M2 macrophages, and an imbalance in M1/M2 macrophages may be the key to the pathogenesis of NASH [10]. A recent study revealed that proinflammatory macrophages in adipose tissue are associated with liver fibrosis in NASH and that targeted macrophage therapy can effectively reverse fibrosis [11]. The development of drugs that target macrophages may be an effective strategy for treating NASH.
Bariatric surgery (BS) is one of the most common treatments for morbid obesity and NASH, and it is more effective than nonoperative treatment. Laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy (LSG) are the two most common bariatric operations. The metabolic function of adipose tissue has improved and changed from a proinflammatory state to an anti-inflammatory state after bariatric surgery [12]. In addition, BS can reduce liver fat and improve inflammation and fibrosis, thus reversing pathological changes in the liver of NASH patients [13]. A meta-analysis included paired liver biopsy samples from 766 NAFLD patients from 15 studies. The results revealed that steatosis improved in 91.6% of patients, inflammation was relieved in 81.3%, fibrosis was reversed in 65.5%, and NAFLD disappeared completely in 69.5% of patients [14]. Studies have shown that the alleviation of chronic liver inflammation caused by lifestyle intervention and BS can be attributed in part to the phenotypic transformation of macrophages from M1 to M2 [15]. Clarifying the relationship between the remission of liver inflammation or the reversal of fibrosis and the polarization of macrophages after BS is crucial. The underlying biological mechanisms deserve further exploration.
With the development of transcriptome sequencing technology, we can easily obtain high-throughput data from tissues. Bioinformatics analysis can be used to accurately deduce the molecular mechanisms underlying diseases. In this study, RNA-seq data from NASH liver tissues were acquired from the GEO database. We analyzed the scRNA-seq data from NASH and normal liver tissues. After filtering and dimension reduction, the cell types were annotated manually. The macrophages of the two groups were compared. To further understand the changes in the proportion of macrophages in NASH patients and the effect of BS on macrophages, we performed deconvolution on bulk RNA-seq data via the CIBERSORT algorithm and calculated the immune cell abundances. Afterward, by calculating the correlation coefficient and performing differential gene analysis, we identified the genes associated with the changes in M1 macrophages after BS. We used the CMAP database to screen candidate drugs that may downregulate the expression of key genes. Finally, we validated the expression of the aforementioned genes in different datasets and constructed mouse models of NASH and sleeve gastrectomy, which were validated in vivo at the animal level. Understanding the relationship between macrophages and BS is expected to contribute to understanding the underlying mechanisms and provide a new perspective on therapeutic strategies for NASH.
Materials and methods
Database
All of the RNA-seq data for NAFLD patients were downloaded from the NCBI Gene Ontology Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). ScRNA-seq data from mouse liver tissue, including 3 normal and 3 NASH samples, were incorporated into the GSE129516 dataset. The bulk RNA-seq datasets from the human liver were included in GSE48452 (12 normal controls and 17 NASH patients). GSE83452 and GSE106737 contained liver biopsies at baseline (NASH, no NASH) and after 1 year of follow-up (diet restriction, bariatric surgery). GSE83452 included bulk RNA-seq data from 152 patients at baseline and 79 patients at the 1-year follow-up. GSE106737 included data from 21 patients with histologically proven NASH one year after Roux-en-Y gastric bypass. The above data sets do not clearly distinguish between fibrotic and non-fibrotic NASH.
scRNA-seq data processing
After the original data were read in R, we used the “Seurat” package to process and filter the data. We first create the Seurat object and add grouping information to the meta. data, and then the percentage feature set function was used to evaluate the expression proportions of the mitochondrial genes in each cell. To remove low-quality cells, we set the criteria of cell filtration to nFeature_RNA > 200, nFeature_RNA < 6000, and percent. MT < 20. The findVariableFeatures function was used to identify the top 3000 variable genes. We then integrated the data and reduced the dimension via principal component analysis (PCA). The top 20 principal components (PCs) were used for UMAP. The FindClusters function was used to identify the main cell clusters with the resolution set to 0.3.
In addition, the marker genes of each cell cluster were identified via the FindAllMarkers function. Manual annotation was performed based on marker genes provided in the GSE129516 dataset.
Identification of differentially expressed genes (DEGs)
The DEGs between NASH macrophages and normal liver macrophages were identified via the FindMarkers function in the “Seurat” package. The R package “Limma” was used to distinguish DEGs between NASH and normal liver tissue. DEGs were defined as those whose logFC filter was > 0.485 and adjusted p value was < 0.05.
GO / KEGG enrichment analysis
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed via the R package “clusterProfiler”.
CIBERSORT analysis
CIBERSORT is an important deconvolution tool based on linear support vector regression [16]. The tool was first published in Nature Methods in 2015. After the LM22 signature (the expression matrix of human immune cells) was downloaded from the CIBERSORT website (https://CIBERSORTx.stanford.edu/), we counted 22 immune cell types from each liver tissue sample on the basis of the LM22 signature and bulk RNA-seq data.
Connectivity map and statistical analysis
Connectivity map analysis was conducted via the CLUE web tool (https://clue.io/) [17]. After the differentially expressed M1 macrophage-related genes (DEMRGs) identified in NASH patients were uploaded, we selected the present perturbations (compounds, CPs) and listed them by connectivity score (median tau score). The 3D chemical structures of the top four compounds were searched in PubChem (https://pubchem.ncbi.nml.gov) [18]. R (version 4.3.0), and all the above packages were utilized for statistical analysis. The criterion for statistical significance was p < 0.05 (two-tailed).
Animals and diet
From the age of 8 weeks, male C57BL/6 J mice were fed a high-fat diet (HFD) with 60% kcal derived from fat (D12492, 60% kcal fat diet, Biopike, China) or a normal chow diet (NCD) as a control for 16 weeks (HFD group n = 18, NCD group n = 6). The mice (Jackson Laboratory, Bar Harbor, Maine, USA) were housed under a 12 h light–12 h dark cycle at 24–26 °C and had ad libitum access to water during the entire experiment. All animal protocols adhered to the regulations of the Animal Welfare Ethics Committee of the Air Force Military Medical University and were performed according to the Guidelines for the Care and Use of Laboratory Animals (license number IACUC-20190107).
Sleeve gastrectomy
After 16 weeks of HFD, the mice underwent sleeve gastrectomy (SG) or a sham operation (n = 6 per group). The mice were fasted for 12 h preoperatively and had free access to water. Anesthesia was induced via 3% isoflurane and oxygen (2 L/min). The isoflurane volatile tank was set at 2%, and the oxygen flow rate was 1 L/min during the maintenance period. The skin was prepared, and the surgical area was sterilized with iodine vapor. A 1–2 cm incision was made along the midline of the abdomen under the xiphoid process to expose the stomach. The stomach is lifted out of the abdomen via ophthalmic forceps, and the loose perigastric connective tissue is carefully separated to free the stomach. Approximately 80% of the gastric tissue was excised from 3 to 5 mm from the gastroesophageal junction to 3–5 mm from the pylorus, leaving the narrow gastric tissue continuous with the gastroesophageal junction and the pylorus. The gastric contents were removed, the stump was sterilized with iodine, and the incision was closed with interrupted vertical inversion sutures with a 9–0 suture. The stomach was gently returned to its anatomic position, and the muscles and skin around the surgical site were adequately irrigated. The skin was closed via 6–0 absorbable sutures and resterilized with iodine vapor. On the first postoperative day, the mice were fasted, watered freely, and injected subcutaneously with 20 mL/kg of sugar saline; they were fed a liquid diet on the second and third postoperative days; and they were put back on an HFD on the fourth postoperative day in each group. All mice in the sham group were released from the stomach tissue only after the abdominal opening and were anesthetized for the same amount of time as those in the SG group. Twelve weeks after BS, the mice were euthanized for sampling.
Histological analysis
Liver samples were fixed in 4% formalin buffer, embedded in paraffin, and then stained with H&E on 4 μm thick serial sections to evaluate hepatocyte morphology. Masson staining to distinguish non-fibrotic and fibrotic NASH. Additionally, frozen sections of fixed liver samples were stained with oil red O to assess hepatic steatosis.
Quantitative PCR analysis
Total RNA was extracted with TRIzol reagent (DP419. Tiangen, Beijing, China) and reverse-transcribed into cDNA by using a PrimeScript RT reagent Kit (RR037A, Takara, Dalian, China). TB Green (RR820A, Takara, Dalian, China) was used to quantify the PCR amplification products. The qPCR assays were performed on a CFX Connect real-time PCR detection system (1855201, Bio-Rad, USA) according to the manufacturer’s instructions. The mRNA expression levels of the related genes were normalized to that of the housekeeping gene β-actin.
Enzyme-linked immunosorbent assay (ELISA)
CXCL10 expression levels were quantified by ELISA kit (E – EL -M0021, Elabscience), and FAT1 expression levels were quantified by ELISA kit (MM - 48515M1, Meimian).Each well was coated with 50–100 μL of capture antibody at a pre-optimized concentration. The plate was sealed and incubated at 4 °C overnight or at room temperature (RT) for 2 h. Wells were washed three times with 0.05% PBST buffer (phosphate-buffered saline containing 0.05% Tween-20) using an ELISA plate washer. Residual liquid was removed by aspiration or plate inversion. The plate was blocked with 100–200 μL per well of 3% BSA in PBS, sealed, and incubated at RT for 1 h. Washing was repeated 50–100 μL of serially diluted standards and experimental samples were added to designated wells. Each dilution was tested in triplicate. The plate was sealed and incubated at RT for 2 h. Liquid was aspirated, and wells were washed six times with 0.05% PBST buffer using the ELISA plate washer. A second specific detection antibody was diluted in 0.05% PBST with 0.5% BSA and added to wells (50–100 μL per well). The plate was sealed and incubated at RT for 1 h. A horseradish peroxidase (HRP)-conjugated secondary antibody was diluted in 0.05% PBST with 0.5% BSA and added to wells (50–100 μL per well). The plate was sealed and incubated at RT for 1 h. Liquid was aspirated, and wells were washed six times with 0.05% PBST buffer. TMB substrate solution was added to each well and incubated at RT for 7 min or until a clear gradient of color intensity emerged in standard wells. The enzymatic reaction was terminated by adding 50 μL of 2 M H2SO4 per well. Absorbance was immediately measured at 450 nm using a microplate reader.
Statistical analysis
Data were expressed as mean ± SEM. All data was firstly analyzed for the Normality test and Homogeneity of variance test. Students’ t tests were used to analyze the differences between the two groups, and p value less than 0.05 was considered statistically significant. Graph Pad Prism version 9.5 was used for statistical analysis.
Results
Single-cell RNA-seq analysis revealed the immune landscape of NASH
Figure 1 shows the workflow of this study. After acquisition, processing, and integration of single-cell transcriptomics data from 3 normal and 3 NASH mouse livers, T-distributed stochastic neighbor embedding (t-SNE) visualization of the combined chow and NASH data revealed ten major clusters, which correspond to endothelial cells, macrophages, T cells, B cells, cholangiocytes, plasma B cells, dendritic cells (DCs), HSCs, hepatocytes, and a cluster representing dividing cells, based on marker gene expression (Fig. 2A, B). Next, we counted the number of all types of cells in the NASH group and control group (Fig. 2C). The number of macrophages in the NASH group was greater than that in the control group. To further analyze the differences in macrophages between NASH patients and control individuals, the “FindMarkers” function was used to identify the differentially expressed genes (DEGs) between NASH macrophages and normal liver macrophages (Fig. 3A). A total of 1060 DEGs were identified in NASH macrophages, including 418 upregulated and 642 downregulated genes. The probability density curve of the fold difference of up-or down-regulated genes, which can be more intuitive to see the number of up-or down-regulated genes (Fig. 3B). GO analysis revealed that the upregulated DEGs were most positively correlated with “proinflammatory”, such as “enzyme inhibitor activity”, and “cytokine activity” (Fig. 3C). In the KEGG analysis, signaling pathways related to M1 macrophage polarization, including the Cytokine-cytokine receptor interaction and the Complement and coagulation cascades, were the main pathways affected (Fig. 3D).
DEGs differentially expressed genes, DEMRGs differentially expressed M1 macrophage-related genes, MRGs macrophage-related genes, NASH nonalcoholic steatohepatitis.
A t-SNE visualization of liver cell clusters based on 33110 single-cell transcriptomes. Cell counts for endothelial cells (Endo), macrophages, T cells, B cells, DCs, cholangiocytes (Chol), hepatocytes (Hep), dividing cells, plasma B cells, and HSCs are indicated in parentheses. B Violin plots showing representative marker gene expression for each cluster. C Comparison of the proportions of cell clusters in NASH and normal tissues.
A Volcano plot of the DEGs. B Probability density plot of fold difference of up-or down-regulated genes. C GO analysis of the DEGs. D KEGG analysis of the DEGs.
Bulk RNA-seq analysis revealed the high proportion of M1 macrophages in NASH patients
Macrophages are usually polarized into two subtypes: M1 macrophages, which are “proinflammatory”, and M2 macrophages, which are “anti-inflammatory” [8]. Previous studies have shown that the number of M1 macrophages increases significantly due to the progression of liver disease from simple steatosis to NASH [19, 20]. To further confirm the changes in macrophage content in NASH tissue via bioinformatics analysis, we downloaded bulk RNA-seq datasets from human livers, including GSE61260 (containing 38 normal and 24 NASH samples) and GSE48452 (containing 12 normal and 17 NASH samples). CIBERSORT is a tool used to count 22 immune cell types (including M0 macrophages, M1 macrophages, and M2 macrophages) on the basis of bulk RNA-seq data [16]. The proportion of macrophages in each liver sample is shown in a stacked column plot (Fig. 4A). The proportion of M1 macrophages in NASH tissues was much greater than that in normal liver tissues (Fig. 4B).
A The proportion of macrophages in each sample. B Comparison of immune cells between NASH and normal liver tissues.
Bulk RNA-seq analysis revealed the downregulation of M1 macrophages in NASH patients by bariatric surgery (BS)
Bariatric surgery (BS) is one of the most common treatments for morbid obesity and NASH, and it is more effective than nonoperative treatment [21]. It is not clear whether the proportion of M1 macrophages in liver tissue changes after BS and whether this change is greater than that associated with nonsurgical treatment. Thus, we downloaded two datasets that included bulk RNA-seq data of pre- and post-BS liver tissue. GSE106737 was incorporated into the baseline and 1-year follow-up data from obese patients whose treatment strategies included BS (n = 21) and lifestyle intervention (n = 20). The CIBERSORT results revealed that the proportion of M1 macrophages decreased after BS (Fig. 5C, D); however, it did not change after lifestyle intervention (Fig. 5A, B). The GSE83452 data were collected from 152 patients at baseline and 79 patients at the 1-year follow-up (38 diet restrictions and 41 bariatric surgeries). Notably, the CIBERSORT results of GSE83452 were similar to those of GSE10673, in which the number of M1 macrophages decreased after BS (Fig. 5G, H) but did not change after diet restriction (Fig. 5E, F).
A, B The proportions of immune cells in each sample and comparisons between the lifestyle intervention and control groups in the GSE106737 cohort. C, D The proportion of immune cells in each sample and comparisons between the bariatric surgery and control groups in the GSE106737 cohort. E, F The proportions of immune cells in each sample and comparisons between the lifestyle intervention and control groups in the GSE83452 cohort. G, H The proportion of immune cells in each sample and comparisons between the bariatric surgery and control groups in the GSE83452 cohort.
Identification of M1 macrophage-related genes (MRGs) in NASH patients
Combining the CIBERSORT results and the expression matrix from liver samples in GSE83452, we calculated the genes co-expressed with M1 macrophages. The thresholds were set as P < 0.001 and R2 > 0.16. A total of 83 genes were correlated with M1 macrophage infiltration. Figure 6A shows the top ten M1 macrophage-related genes (MRGs) with the strongest correlation. The network diagram illustrates the correlations of 83 MRGs with M1 macrophages, 77 of which were positively correlated and six of which were negatively correlated (Fig. 6B). Moreover, the heatmap shows the co-expression relationships among 83 MRGs (Fig. 6C). Next, we performed GO and KEGG analyses to analyze the possible mechanism based on 83 MRGs (Fig. 6D, E). The results revealed that MRGs were enriched mainly in immune/inflammatory-related pathways.
A Correlation maps of the top 10 genes associated with M1 macrophage infiltration. B Co-expression network diagram of MRGs. C Heatmap of the correlations between the M1 macrophage proportion and MRGs. D GO analysis of the MRGs. E KEGG analysis of the MRGs.
Identification of differentially expressed M1 macrophage-related genes (DEMRGs) between pre- and post-BS NASH patients
The above results indicate that BS is the key to the remission of hepatitis and the reduction in the number of proinflammatory M1 macrophages. We identified the differentially expressed genes (DEGs) between pre- and post-BS. The volcano plot revealed that 55 DEGs were upregulated and 137 were downregulated post-BS (Fig. 7A). In total, 12 differentially expressed M1 macrophage-related genes (DEMRGs) were obtained through overlapping 192 DEGs with 83 MRGs (Fig. 7B). All of the DEMRGs were downregulated post-BS (Fig. 7C, D).
A Volcano plot of the DEGs between pre- and post-bariatric surgery. B Venn diagram showing the DEMRGs that overlap between DEGs and MRGs. C Heatmap of the DEMRGs before and after bariatric surgery. D Boxplots of the DEMRGs before and after bariatric surgery.
CMAP screened the potential therapeutic agents for NASH
CMAP analysis was used to deduce potential compounds for NASH by comparing the 12 DEMRGs with a compound-based expression matrix. The expression of 12 DEMRGs was upregulated in NASH patients and downregulated post-BS; thus, we selected the compounds that downregulated the DEMRGs with the highest negative scores. In total, 56 compounds were screened as candidates to treat NASH, with median tau score criteria below −90 (Fig. 8A). In terms of the median tau score, the top ten compounds were damnacanthal, BRD-K13872703, alpha-estradiol, verapamil, an IKK-2 inhibitor, caffeic acid, atorvastatin, cimetidine, dihydroergotamine and BRD-K88742110 (Fig. 8B). The descriptions and median tau scores of the top ten compounds are listed in Table 1. In addition, the 3D chemical structures of the top four compounds were obtained from PubChem (Fig. 8C).
A The significant perturbation genes with the lowest CMap connectivity score; (B) the top 10 perturbed compounds; (C) 3D chemical structures of damnacanthal, BRD-K13872703, alpha-estradiol and verapamil. (Red represents oxygen atoms, yellow represents sulfur atoms, blue represents nitrogen atoms, gray represents carbon atoms and white represents hydrogen atoms).
Validation of M1 macrophage-related genes in mouse model
To verify the expression of DEMRGs in vivo, we constructed NASH and BS (sleeve gastrectomy) mouse models (Fig. 9A). Masson staining was used to detect the degree of liver fibrosis in both normal chow diet and high-fat diet fed mice, and the staining results showed that all the NASH mouse models used were fibrotic NASH (Fig. 9B). HE staining and oil red O staining were performed on the livers (Fig. 9C), which revealed balloon-like hepatocytes, inflammatory cell infiltration, and a large amount of lipid droplet aggregation in the livers of HFD-fed mice. The infiltration of inflammatory cells in the liver of mice after BS was reduced and the degree of steatosis was alleviated. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured in NASH mice after BS, and we found that the liver function of the mice was significantly improved after BS (Fig. 9D).
A Flow chart of the construction of the NASH and sleeve gastrectomy mouse model. B Masson staining of mouse liver tissue. Scale bar: 50 μm or 20 μm. C H&E staining and oil red O staining of mouse liver tissue. Scale bar: 200 μm. D Serum ALT and AST levels. E Relative mRNA expression levels of M1 macrophages markers and M2 macrophages markers. F Relative mRNA expression levels of DEMRGs. G Elisa results of FAT1 and CXCL10 in HFD mice and SG mice.
To explore whether BS surgery alleviate M1 macrophages infiltration, we then extracted RNA from mouse liver tissues and used real-time quantitative PCR to quantify M1 macrophages by means of CD80, CD64, and iNOS and M2 macrophages by means of CD163, CD206, and Arg1. We found that the expression of M1 macrophages was reduced after BS, with no significant change in M2 macrophage expression (Fig. 9E). The above results indicate that BS surgery alleviate both M1 macrophages infiltration and liver damage. We further validated 12 DEMGRs. However, the results revealed that only FAT1 and CXCL10 presented the same trend as expected (Fig. 9F). Then, we have detected the protein levels of FAT1 and CXCL10 by ELISA, and the results showed that the protein levels of FAT1 and CXCL10 decreased after BS (Fig. 9G). Based on our data, we proposed a hypothetical model as illustrated in Fig. 10.
Schematic representation of bariatric surgery, illustrating the removal of a portion of the stomach. Molecular characterization of proinflammatory M1 macrophages, highlighting upregulated markers associated with tissue damage.The model suggests that bariatric surgery reduces NASH progression by suppressing M1 macrophage activation.
Discussion
With the increasing number of obese people annually, NAFLD has become a global threat to human health. NASH is a severe inflammatory subtype of NAFLD that is prone to fibrosis and progresses to liver cirrhosis or even hepatocellular carcinoma. The mechanisms underlying the occurrence and development of NASH remain undefined, and the only drug currently approved by the U.S. Food and Drug Administration for the treatment of NASH is Rezdiffra [5]. Hepatic macrophages are highly heterogeneous and plastic and change dynamically during the development of NASH [7]. During NASH, the number of M1 macrophages increases significantly, resulting in lipid accumulation, oxidative stress, and inflammation in the liver [9]. The results of animal experiments have shown that macrophage-targeted therapy can effectively reverse fibrosis [11]. Macrophage-targeted drugs have promising potential in the treatment of NASH.
Bariatric surgery (BS) is an effective approach for NASH. It can reduce liver fat and ameliorate the inflammatory response and fibrosis [21].
The pathophysiology of NASH is multifactorial and not yet completely understood, innate immunity is a major contributing factor in which liver- resident macrophages (Kupffer cells) and recruited macrophages play a central part in disease progression. In children with NAFLD, numerous activated macrophages were found located in the spaces between damaged hepatocytes [22]. Portal infiltration of macrophages also seems to be an early event in human NAFLD, occurring already at the stage of steatosis before inflammation or fibrosis develops, but predicting progressive disease [23]. Liver-resident Kupffer cells initiate inflammation and help recruit blood-derived monocytes; both differentiate into pro-inflammatory macrophages and further promote NAFLD progression. BS can lead to immune modulation, strong evidence suggests, by affecting macrophage infiltration and promoting a phenotypic switch from M1 to M2-like macrophages. Several studies observed significant reductions in macrophage number up to a year after surgery using the CD68 marker [24,25,26,27,28]. BS was also found to alter the phenotype of macrophages favoring a shift towards M2-like macrophages [29,30,31,32,33,34].BS decreases expression of proinflammatory cytokines and chemotactic signals. In addition to revealing the molecular biological mechanisms by which BS improves NASH from the perspective of macrophage polarization, we can also identify genes related to the downregulation of proinflammatory M1 macrophages and design potential therapeutic drugs.
In this study, we downloaded, processed, and integrated single-cell transcriptomics data from normal and NASH livers. After counting the number of different cell types, we found that NASH livers had a greater proportion of macrophages. Compared with those in normal liver macrophages, there were 36 upregulated and 55 downregulated differentially expressed genes (DEGs) in NASH liver macrophages. GO analysis revealed that 36 upregulated DEGs were positively correlated with “Major Histocompatibility Complex class II”. Major Histocompatibility Complex class II molecules are expressed by antigen-presenting cells (APCs), including macrophages. According to the KEGG analysis, signaling pathways related to the immune/inflammatory response were the main pathways affected. These results illustrated that macrophages in the livers of patients with NASH tend to play a proinflammatory role.
CIBERSORT was used to count 22 immune cell types in each liver tissue sample on the basis of bulk RNA-seq data [16]. As expected, the number of M1 macrophages in the liver increased significantly in NASH patients. We also revealed that the proportion of M1 macrophages from the liver and adipose tissue decreased 1 year after bariatric surgery. Interestingly, after lifestyle intervention for 1 year, although the patients lost weight, the proportion of liver M1 macrophages did not change. Once again, the superiority of the curative effect of BS has been confirmed from the view of macrophage polarization.
To explore the mechanisms by which the number of M1 macrophages decreases after BS, we screened 83 M1 macrophage-related genes (MRGs) and 192 differentially expressed genes (DEGs) before BS and after BS. In total, 12 differentially expressed M1 macrophage-related genes (DEMRGs) were obtained through the overlapping of 192 DEGs with 83 MRGs. The 12 DEMRGs are SRGN, CXCL9, CXCL10, GPNMB, IL32, MSN, FAT1, FGL2, BIRC3, TYROBP, ALOX5AP and SLAMF6. SRGN is highly expressed in immune cells and encodes a protein called Serglycin, which is a proteoglycan. A study revealed an increased proportion of proinflammatory M1 macrophages in the adipose tissue of Srgn + /+ mice compared with that in the adipose tissue of Srgn–/– mice [35]. The level of the SRGN mRNA decreased after BS [35]. SRGN has proven to be a biomarker of NASH and has been verified in clinical samples and mouse models [36]. The proteins encoded by C-X-C motif chemokine ligand 9 (CXCL9) and C-X-C motif chemokine ligand 10(CXCL10) are chemokines, which are proinflammatory factors derived from M1 macrophages [37]. CXCL9 and CXCL10 are biomarkers of NASH and can exacerbate the progression of NAFLD [38, 39]. Glycoprotein NMB (GPNMB) is an endogenous glycoprotein that is highly expressed in macrophages and has a proinflammatory effect [40]. The concentration of serum soluble GPNMB in NASH patients is greater than that in patients with simple steatosis [41]. Moreover, GPNMB is a marker of lipid-associated macrophages (LAMs) and a promising therapeutic target for NAFLD [42]. Interleukin-32 (IL-32) is a cytokine that induces proinflammatory cytokines and chemokines via p38-MAPK and NF-κB. In the liver, IL-32 induces the differentiation of monocytes into macrophage-like cells [43]. NASH livers produce increased IL32, and increased circulating levels of IL32 are independently related to systolic blood pressure [44]. Moesin (membrane-organizing extension spike protein) is the protein product of MSNs. It facilitates lipopolysaccharide recognition to activate the secretion of proinflammatory cytokines by macrophages (PMID: 20546116). The protein product of FAT atypical cadherin 1(FAT1) is an atypical cadherin. The expression of FAT1 increases in NASH and cirrhotic livers, suggesting that FAT1 may be a new therapeutic target for NASH [45]. Fibrinogen-like protein 2 (FGL2) induces macrophages to produce proinflammatory cytokines by activating the NF-κB and p38-MAPK pathways and thus aggravates the progression of NASH [46]. Moreover, in addition to promoting the polarization of M1 macrophages, Tao R et al. reported that FGL2 promotes oxidative stress and leads to mitochondrial dysfunction [47]. BIRC3 is a member of the human apoptosis protein (IAP) family [48]. Prior studies have indicated that BIRC3 is upregulated in NASH and is an important biomarker for NASH [49, 50]. TYRO protein tyrosine kinase-binding protein (TYROBP) is a key factor in the NASH liver and is downregulated after Roux-en-Y gastric bypass surgery [51]. Dang D et al. reported that TYROBP is highly expressed in the macrophages of humans and mice [52]. Arachidonate 5-lipoxygenase-activating protein (ALOX5AP) is present in adipose tissue, and high expression of ALOX5AP can lead to weight gain and insulin resistance [53]. Chen et al. reported that ALOX5AP is associated with macrophages and that potential carcinogenetic factors are suppressed by BS [54]. Signaling lymphocyte activation molecule family member 6 (SLAMF6) is a T-cell coreceptor. SLAMF6 is a newly discovered biomarker of NASH that has not yet been studied.
Based on the 12 DEMRGs identified in NASH patients, we then conducted CMAP analysis via the CLUE web tool (https://clue.io/). A total of 56 compounds were screened as candidates for the treatment of NASH, with median tau score criteria below −90. Arranged by the median tau score, the top four compounds were damnacanthal, BRD-K13872703, alpha-estradiol, and verapamil. Damnacanthal is a type of medicinal anthraquinone [55]. The main effects of damnacanthal are antinociceptive and anti-inflammatory [56]. Damniotic substances can inhibit NF-kB signaling, which is an important signaling pathway of macrophage polarization [57]. BRD-K13872703 is a chlorine/bicarbonate exchanger inhibitor and purinergic receptor antagonist. Chandrashekaran et al. reported that the purinergic receptor X7 modulates the induction and translocation of Glut4 in hepatic stellate cells and thus accelerates the progression of NASH [58]. The purinergic receptor antagonist BRD-K13872703 may play an important role in the treatment of NASH. Alpha-estradiol is an estrogen receptor agonist. In the absence of estrogen, macrophages polarize to the M1 phenotype and promote the development of obese ovariectomized mice into NASH [59].
A study revealed that hepatic steatosis in mice can be relieved after receiving an estrogen receptor α agonist [60]. Verapamil is an L-type calcium channel blocker. In addition to being hypotensive, verapamil can improve liver regeneration in NASH patients [61]. The 3D chemical structures of damnacanthal, BRD-K13872703, alpha-estradiol, and verapamil were obtained from PubChem. One study revealed that macrophage-targeted therapy can effectively reverse fibrosis in NAFLD patients [11]. The compounds that we screened on the basis of the 12 DEMRGs may be promising drugs for NASH treatment.
To verify the expression of DEMRGs in vivo, we generated a NASH mouse model by feeding the mice a high-fat diet (HFD). After 16 weeks of HFD, the mice underwent sleeve gastrectomy (SG). However, the results revealed that only FAT1 and CXCL10 exhibited the expected trend. There may be two reasons for this. First, the targets of bioinformatics analysis and qPCR are quite distinct. In the bioinformatics analysis, all the databases were derived from patients, whereas the qPCR results were obtained from experimental mice. Second, the time points of the bioinformatics analysis and qPCR are quite different. qPCR analysis was conducted 12 weeks after SG surgery. Unlike the extended recovery period of patients, the livers of mice still showed signs of inflammation after 3 months of SG, which might explain the absence of a decrease in the expression of some MRGs.
Our study has several limitations. First, because there is a lack of single-cell transcriptome data of liver tissue after bariatric surgery, we are unable to use pseudotime analysis to describe the dynamic changes in macrophages at the single-cell level. Second, while relevant studies have reported the effects of DEMRGs on the polarization of macrophages in NASH, the molecular biological mechanisms still require further exploration, which is also the focus of our future work. Third, the specific efficacy of the compounds we screened for the treatment of NASH needs to be supported by more in vivo and in vitro experimental data.
Conclusions
In summary, our study revealed the potential mechanism of bariatric surgery in facilitating NASH by focusing on the polarization of macrophages. Following bariatric surgery, there was a decrease in the proportion of M1 macrophages in the liver and subcutaneous adipose tissues. We identified 12 differentially expressed genes related to M1 macrophages (DEMRGs) in NASH, including SRGN, CXCL9, CXCL10, GPNMB, IL32, MSN, FAT1, FGL2, BIRC3, TYROBP, ALOX5AP, and SLAMF6. Finally, we screened 56 compounds as potential drugs for treating NASH.
Data availability
The RNA-seq data used in this study can be found in online repositories. The names of the repositories/repositories and accession numbers are GSE129516, GSE48452, GSE106737, and GSE83452.
References
Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease-meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology. 2016;64:73–84.
Wong VW-S, Ekstedt M, Wong GL-H, Hagström H. Changing epidemiology, global trends and implications for outcomes of NAFLD. J Hepatol. 2023;79:842–852.
Tacke F, Puengel T, Loomba R, Friedman. SLJJoH. An integrated view of anti-inflammatory and antifibrotic targets for the treatment of NASH. J Hepatol. 2023;79:552–66.
Pinter M, Pinato DJ, Ramadori P, Heikenwalder MJCCR. NASH and hepatocellular carcinoma: immunology and immunotherapy. Clin Cancer Res: J Am Assoc Cancer Res. 2023;29:513–20.
Rong L, Zou J, Ran W, Qi X, Chen Y, Cui H, et al. Advancements in the treatment of non-alcoholic fatty liver disease (NAFLD). Front Endocrinol (Lausanne). 2023;13:1087260.
Guilliams M, Scott CLJI. Liver macrophages in health and disease. Immunity. 2022;55:1515–29.
Vonderlin J, Chavakis T, Sieweke M, Tacke FJC. Gastroenterology M, Hepatology. The multifaceted roles of macrophages in NAFLD pathogenesis. Cell Mol Gastroenterol Hepatol. 2023;15:1311–24.
Yunna C, Mengru H, Lei W, Weidong. CJEjop. Macrophage M1/M2 polarization. Eur J Pharm. 2020;877:173090.
Dixon LJ, Barnes M, Tang H, Pritchard MT, Nagy LEJCP. Kupffer cells in the liver. Compr Physiol. 2013;3:785–97.
Wang B, Li X, Hu W, Zhou Y, Din. YJIl. Silencing of lncRNA SNHG20 delays the progression of nonalcoholic fatty liver disease to hepatocellular carcinoma via regulating liver Kupffer cells polarization. IUBMB Life. 2019;71:1952–61.
Martinez P, Nault G, Steiner J, Wempe MF, Pierce A, Brunt B, et al. Therapeutic targeting of adipose tissue macrophages ameliorates liver fibrosis in non-alcoholic fatty liver disease. Neuro-Oncol Adv. 2023;5:111.
Liu Y, Jin J, Chen Y, Chen C, Chen Z, Xu LJA. Integrative analyses of biomarkers and pathways for adipose tissue after bariatric surgery. Adipocyte. 2020;9:384–400.
Laursen TL, Hagemann CA, Wei C, Kazankov K, Thomsen KL, Knop FK, et al. Bariatric surgery in patients with non-alcoholic fatty liver disease-from pathophysiology to clinical effects. World J Hepatol. 2019;11:138–49.
Mummadi RR, Kasturi KS, Chennareddygari S, Sood. GKJCG, Hepatology. Effect of bariatric surgery on nonalcoholic fatty liver disease: systematic review and meta-analysis. Clin Gastroenterol Hepatol: Clin Pr J Am Gastroenterol Assoc. 2008;6:1396–402.
Turner L, Santosa. SJAiN. Putting ATM to BED: how adipose tissue macrophages are affected by bariatric surgery, exercise, and dietary fatty acids. Adv Nutr (Bethesda, Md). 2021;12:1893–910.
Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat methods. 2015;12:453–7.
Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ, et al. The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Sci (N Y, N Y). 2006;313:1929–35.
Wang Y, Xiao J, Suzek TO, Zhang J, Wang J, Zhou Z, et al. PubChem’s BioAssay database. Nucleic Acids Res. 2012;40:D400–D12.
Alabdulaali B, Al-Rashed F, Al-Onaizi M, Kandari A, Razafiarison J, Tonui D, et al. Macrophages and the development and progression of non-alcoholic fatty liver disease. Front Immunol. 2023;14:1195699.
Barreby E, Chen P, Aouadi MJNRE. Macrophage functional diversity in NAFLD—more than inflammation. Nat Rev Endocrinol. 2022;18:461–72.
Chauhan M, Singh K, Thuluvath. PJJDd, sciences. Bariatric surgery in NAFLD. Digest Dis Sci. 2022;67:408–22.
Lotowska JM, Sobaniec-Lotowska ME, DMJSJoG Lebensztejn. The role of Kupffer cells in the morphogenesis of nonalcoholic steatohepatitis–ultrastructural findings. The first report in pediatric patients. Scand J Gastroenterol. 2013;48:352–7.
Gadd VL, Skoien R, Powell EE, Fagan KJ, Winterford C, Horsfall L, et al. The portal inflammatory infiltrate and ductular reaction in human nonalcoholic fatty liver disease. Hepatol (Balt, Md). 2014;59:1393–405.
Cancello R, Zulian A, Gentilini D, Mencarelli M, Della Barba A, Maffei M, et al. Permanence of molecular features of obesity in subcutaneous adipose tissue of ex-obese subjects. Int J Obes (2005). 2013;37:867–73.
Trachta P, Dostálová I, Haluzíková D, Kasalický M, Kaválková P, Drápalová J, et al. Laparoscopic sleeve gastrectomy ameliorates mRNA expression of inflammation-related genes in subcutaneous adipose tissue but not in peripheral monocytes of obese patients. Mol Cell Endocrinol. 2014;383:96–102.
Haluzíková D, Lacinová Z, Kaválková P, Drápalová J, Křížová J, Bártlová M, et al. Laparoscopic sleeve gastrectomy differentially affects serum concentrations of FGF-19 and FGF-21 in morbidly obese subjects. Obes (Silver Spring, Md). 2013;21:1335–42.
Aghamohammadzadeh R, Greenstein AS, Yadav R, Jeziorska M, Hama SF, et al. Effects of bariatric surgery on human small artery function: evidence for reduction in perivascular adipocyte inflammation, and the restoration of normal anticontractile activity despite persistent obesity. J Am Coll Cardiol. 2013;62:128–35.
Cancello R, Henegar C, Viguerie N, Taleb S, Poitou C, Rouault C, et al. Reduction of macrophage infiltration and chemoattractant gene expression changes in white adipose tissue of morbidly obese subjects after surgery-induced weight loss. Diabetes. 2005;54:2277–86.
Hess DA, Trac JZ, Glazer SA, Terenzi DC, Quan A, Teoh H, et al. Vascular risk reduction in obesity through reduced granulocyte burden and improved angiogenic monocyte content following bariatric surgery. Cell Rep Med. 2020;1:100018.
Hagman DK, Larson I, Kuzma JN, Cromer G, Makar K, Rubinow KB, et al. The short-term and long-term effects of bariatric/metabolic surgery on subcutaneous adipose tissue inflammation in humans. Metab: Clin Exp. 2017;70:12–22.
Liu Y, Aron-Wisnewsky J, Marcelin G, Genser L, Le Naour G, Torcivia A, et al. Accumulation and changes in composition of collagens in subcutaneous adipose tissue after bariatric surgery. J Clin Endocrinol Metab. 2016;101:293–304.
García-Rubio J, León J, Redruello-Romero A, Pavón E, Cozar A, Tamayo F, et al. Cytometric analysis of adipose tissue reveals increments of adipocyte progenitor cells after weight loss induced by bariatric surgery. Sci Rep. 2018;8:15203.
Cinkajzlová A, Lacinová Z, Kloučková J, Kaválková P, Trachta P, Kosák M, et al. An alternatively activated macrophage marker CD163 in severely obese patients: the influence of very low-calorie diet and bariatric surgery. Physiol Res. 2017;66:641–52.
Aron-Wisnewsky J, Tordjman J, Poitou C, Darakhshan F, Hugol D, Basdevant A, et al. Human adipose tissue macrophages: m1 and m2 cell surface markers in subcutaneous and omental depots and after weight loss. J Clin Endocrinol Metab. 2009;94:4619–23.
Doncheva AI, Norheim FA, Hjorth M, Grujic M, Paivandy A, Dankel SN, et al. Serglycin is involved in adipose tissue inflammation in obesity. J Immunol (Balt, Md : 1950). 2022;208:121–32.
Zhang J-j, Shen Y, Chen X-y, Jiang M-l, Yuan F-h, Xie S-l, et al. Integrative network-based analysis on multiple Gene Expression Omnibus datasets identifies novel immune molecular markers implicated in non-alcoholic steatohepatitis. Front Endocrinol (Lausanne). 2023;14:1115890.
House IG, Savas P, Lai J, Chen A, Oliver AJ, Teo ZL, et al. Macrophage-derived CXCL9 and CXCL10 are required for antitumor immune responses following immune checkpoint blockade. Clin Cancer Res: J Am Assoc Cancer Res. 2020;26:487–504.
Zhang X, Shen J, Man K, Chu ES, Yau TO, Sung JC, et al. CXCL10 plays a key role as an inflammatory mediator and a non-invasive biomarker of non-alcoholic steatohepatitis. J Hepatol. 2014;61:1365–75.
Li L, Xia Y, Ji X, Wang H, Zhang Z, Lu P, et al. MIG/CXCL9 exacerbates the progression of metabolic-associated fatty liver disease by disrupting Treg/Th17 balance. Exp cell Res. 2021;407:112801.
Saade M, Araujo de Souza G, Scavone C, Kinoshita. PFJFii. The role of GPNMB in inflammation. Front Immunol. 2021;12:674739.
Katayama A, Nakatsuka A, Eguchi J, Murakami K, Teshigawara S, Kanzaki M, et al. Beneficial impact of Gpnmb and its significance as a biomarker in nonalcoholic steatohepatitis. Sci Rep. 2015;5:16920.
Liu H, Yerevanian A, Westerhoff M, Hastings MH, Guerra JRB, Zhao M, et al. Roles of Activin A and Gpnmb in metabolic dysfunction-associated steatotic liver disease (MASLD). Diabetes. 2024;73:260–79.
Netea MG, Lewis EC, Azam T, Joosten LA, Jaekal J, Bae S-Y, et al. Interleukin-32 induces the differentiation of monocytes into macrophage-like cells. Proc Natl Acad Sci USA. 2008;105:3515–20.
Tomasi M, Cherubini A, Pelusi S, Margarita S, Bianco C, Malvestiti F, et al. Circulating Interlukin-32 and altered blood pressure control in individuals with metabolic dysfunction. Int J Mol Sci. 2023;24:7465.
Valletta D, Czech B, Thasler WE, Müller M, Bosserhoff A-K, Hellerbrand C. Expression and function of the atypical cadherin FAT1 in chronic liver disease. Biochem Biophys Res Commun. 2012;426:404–8.
Hu J, Wang H, Li X, Liu Y, Mi Y, Kong H, et al. Fibrinogen-like protein 2 aggravates nonalcoholic steatohepatitis via interaction with TLR4, eliciting inflammation in macrophages and inducing hepatic lipid metabolism disorder. Theranostics. 2020;10:9702–20.
Tao R, Han M, Yuan W, Xiao F, Huang J, Wang X, et al. Fibrinogen-like protein 2 promotes proinflammatory macrophage polarization and mitochondrial dysfunction in liver fibrosis. Int Immunopharmacol. 2023;117:109631.
Frazzi RJC, bioscience. BIRC3 and BIRC5: multi-faceted inhibitors in cancer. 2021;11:1-14.
Wang Z, Huang Y, Zhu M, Cao J, Xiong ZJB. Communications BR. TLR2 and CASP7 as the biomarkers associated with non-alcoholic fatty liver disease and chronic kidney disease. Biochem Biophys Res Commun. 2023;667:50–7.
Nguyen L, Masouminia M, Mendoza A, Samadzadeh S, Tillman B, Morgan T, et al. Alcoholic hepatitis versus non-alcoholic steatohepatitis: Levels of expression of some proteins involved in tumorigenesis. Exp Mol Pathol. 2018;104:45–9.
Chen F, Zhou Y, Wu Z, Li Y, Zhou W, Wang Y Integrated analysis of key genes and pathways involved in nonalcoholic steatohepatitis improvement after Roux-en-Y gastric bypass surgery. 2021;11:611213.
Dang D, Taheri S, Das S, Ghosh P, Prince LS, Sahoo. DJFip. Computational approach to identifying universal macrophage biomarkers. Front Physiol. 2020;11:275.
Kaaman M, Rydén M, Axelsson T, Nordström E, Sicard A, Bouloumié A, et al. ALOX5AP expression, but not gene haplotypes, is associated with obesity and insulin resistance. Int J Obes (2005). 2006;30:447–52.
Chen S, Tang L, Guillot A, Liu HJM. Bariatric surgery associates with nonalcoholic steatohepatitis/hepatocellular carcinoma amelioration via SPP1 suppression. Metabolites. 2022;13:11.
Abu N, Mohd Ali N, Yong Ho W, Keong Yeap S, Abdul Aziz MY. Banu Alitheen NJA-CAiMC. Damnacanthal: a Promis Compd a medicinal anthraquinone. 2014;14:750–5.
Okusada K, Nakamoto K, Nishida M, Fujita-Hamabe W, Kamiya K, Mizushina Y, et al. The antinociceptive and anti-inflammatory action of the CHCl3-soluble phase and its main active component, damnacanthal, isolated from the root of Morinda citrifolia. Biol pharm bull. 2011;34:103–7.
Kim M-H, Jeong H-JJI. Damnacanthal inhibits the NF-κB/RIP-2/caspase-1 signal pathway by inhibiting p56lck tyrosine kinase. Immunotoxicology. 2014;36:355–63.
Chandrashekaran V, Das S, Seth RK, Dattaroy D, Alhasson F, Michelotti G, et al. Purinergic receptor X7 mediates leptin induced GLUT4 function in stellate cells in nonalcoholic steatohepatitis. Biochimica et biophysica acta. 2016;1862:32–45.
Shu Z, Zhang G, Zhu X, Xiong WJB. Communications BR. Estrogen receptor α mediated M1/M2 macrophages polarization plays a critical role in NASH of female mice. Biochem Biophys Res Commun. 2022;596:63–70.
Chow J, Jones M, Prelle K, Simpson ER, Boon. WCJTJoe. A selective estrogen receptor α agonist ameliorates hepatic steatosis in the male aromatase knockout mouse. J Endocrinol. 2011;210:323–34.
Lai J-L, Lian Y-E, Wu J-Y, Wang Y-D, Bai Y-NJA. Verapamil induces autophagy to improve liver regeneration in non-alcoholic fatty liver mice. Adipocyte. 2021;10:532–45.
Acknowledgements
We sincerely thank the researchers who uploaded their data to the GEO database. We thank Dr. Jianming Zeng (University of Macau) and all the members of his bioinformatics team, biotrainee, for generously sharing their experience and codes.
Funding
This study was supported by the Xijing Hospital Boost Project (XJZTCY15), Shaanxi Province Natural Science Basic Research Program (2022JQ-820), and Youth Programs of the National Natural Science Foundation of China (82201627).
Author information
Authors and Affiliations
Contributions
QWZ: writing-original draft and editing; SZD: Drawing figures and conceptualization; XYC: Data analysis; YYW: Funding acquisition and project administration; YLY: Funding acquisition; All authors reviewed the manuscript. The author(s) read and approved the final manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics
The animal study was reviewed and approved by the Animal Welfare Ethics Committee of the Air Force Military Medical University.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Zheng, Q., Deng, S., Chen, X. et al. Macrophage inhibition in the alleviation of nonalcoholic steatohepatitis caused by bariatric surgery. Genes Immun (2025). https://doi.org/10.1038/s41435-025-00334-6
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41435-025-00334-6
