Alternative activation of macrophages by prostacyclin synthase ameliorates alcohol induced liver injury

Alcoholic liver disease (ALD) is a major cause of chronic liver disease worldwide. Macrophages exhibit different functional states and are classified as classically activated (M1) and alternatively activated (M2) macrophages. However, the mechanisms that govern M1/M2 polarization in chronic ALD remain to be elucidated. Prostacyclin (PGI2) synthase (PTGIS) is an enzyme of the prostaglandin pathway which catalyzes the conversion of Prostaglandin H2 (PGH2) to PGI2. PTGIS has anti-inflammatory properties. However, the function of PTGIS in ALD has not yet been determined. In this study, we demonstrated that PTGIS was downregulated in ALD and forced PTGIS expression in vivo using recombinant adeno-associated viral vector-packed PTGIS overexpression plasmid, which alleviated the inflammatory response and suppressed the macrophage M1 phenotype in mice. Loss- and gain-of function-experiments demonstrated that forced PTGIS expression inhibited the macrophage switch to the M1 phenotype and promoted M2 polarization. Furthermore, we identified the genes regulated by PTGIS through RNA-sequencing (RNA-seq) analysis. Gene ontology and KEGG pathway analyses showed that PTGIS regulates many genes involved in the immune response and is enriched in the Janus kinase/signal transducers and activators of transcription (JAK/STAT) signal transduction pathway, which plays an important role in regulating macrophage polarization. The proteins interacting with JAKs were predicted using the STRING database. The overlap between the RNA-seq and the STRING database was interleukin-6; this indicated that it was involved in macrophage polarization regulated by JAK/STAT signaling. We further explored the microRNAs that could regulate the expression of PTGIS through TargetScan. The results of luciferase assay illustrated that the expression of PTGIS was regulated by miR-140-3p.1. These results imply that PTGIS plays a pivotal role in ALD, partly by influencing macrophage polarization.


Introduction
Alcohol consumption is a leading risk factor for global disease burden and causes substantial health loss [1].
Alcoholic liver disease (ALD) is the leading cause of liverrelated disorders and contributes to 25% of alcohol-related deaths worldwide [2]. ALD is characterized by progressive liver steatosis, fibrosis, and cirrhosis. Consumption of alcohol increases gut permeability, which leads to the translocation of both microbes and microbial products, including lipopolysaccharide (LPS), β-glucan, and other pathogen-associated molecular patterns, into the portal circulation, activating innate immune responses in the liver [3,4]. Macrophages are highly heterogeneous, plastic populations that undergo pleiotropic coordinated responses to tissue damage through distinct programs of activation, known as classical activation (M1) or alternative activation (M2) [5,6]. M1 activation is mainly induced by Th1 cytokines, such as interferon-gamma (IFNγ), or bacterial molecular patterns, including endotoxin/LPS.
The polarization phenotype of macrophages in response to ethanol is complex, with increased expression of both M1 and M2 polarization markers as well as that of both pro-and anti-inflammatory mediators [15]. Louvet et al. showed that activation of cannabinoid receptor 2 expressed in Kupffer cells inhibits M1 polarization and favors M2 response, thereby exerting anti-steatogenic roles by paracrine interactions with hepatocytes [15]. Limiting M1 macrophage polarization attenuates alcohol-induced liver injury, and IL-10 released from M2 polarized macrophages promotes M1 apoptosis [9]. Interestingly, ethanol-induced M1 polarization is modifiable; M1 macrophages can be shifted to an M2 phenotype [16]. However, little is currently known regarding the mechanisms of hyperpolarization and the impact of these complex phenotypes on the development of chronic ethanol-induced inflammation in the liver.
MicroRNAs (miRNAs), endogenous small non-coding RNAs, can inhibit target gene expression by directly binding to complementary sequences in the 3′ untranslated region (3′UTR) of mRNAs and thus are involved in many physiological processes, including cellular differentiation and immune cell activation [17]. miRNAs may play pivotal roles in modulating macrophage activation, polarization, tissue infiltration, and resolution of inflammation [18][19][20].
Prostaglandins are key lipid mediators that regulate physiological functions and inflammatory responses in health and disease [21]. They are synthesized from arachidonic acid via cyclooxygenases (1 and 2) to generate unstable prostaglandin H 2 (PGH 2 ), which is further processed by terminal synthases into the major prostaglandins (PGI 2 , PGE 2 , PGD 2 , and PGF2α) or thromboxane A2 [21]. The PTGS pathway is frequently dysregulated in cancers [22]. PGI 2 synthase (PTGIS), also known as PGIS or CYP8A1, is a member of family 8 (CYP8) in the cytochrome P450 superfamily. PTGIS catalyzes the conversion of PGH 2 to PGI 2 [23]. Altered PTGIS plays an important role in preventing tumor growth and progression [24][25][26][27]. In addition, PTGIS plays an anti-inflammatory role in colorectal cancer [28], and prostaglandin G/H synthase 2 involved in the initial response to LPS stimulation [29]. However, whether PTGIS is involved in regulating macrophage polarization in ALD has not yet been determined. Therefore, in this study, we investigated whether PTGIS is involved in regulating macrophage polarization in ALD.

Murine model of ALD
Six-to-eight-week-old male C57BL/6j mice (18-22 g weight) were purchased from the Laboratory Animal Center of Anhui Medical University. All animal procedures were reviewed and approved by the Institutional Animal Experimental Ethics Committee. Mice were housed in a temperature-controlled room (22°C). After a week of adaptive breeding, mice were randomly divided into control diet (CD) and EtOH-fed groups (eight mice per group).
A mouse model of chronic plus single binge ethanol gavage was used as previously described [30]. For chronic alcohol administration, in the first 5 days, all mice were fed a liquid control diet. Thereafter, EtOH-fed mice were fed a liquid diet containing 5% v/v ethanol for 10 days, whereas the CD-fed mice were fed isocaloric maltose-dextrin. On day 16, EtOH-fed mice were administrated a single dose of ethanol gavage (5 g/kg body weight, 20% ethanol), while CD-fed mice were administrated an isocaloric dose of dextrin maltose gavage. Nine hours after gavage, the mice were euthanized, and blood and liver tissue samples were collected for further analysis.

Isolation of primary macrophages
Liver macrophages were isolated following a previously established method [31]. Briefly, a 20-G catheter was inserted through the mouse portal vein, and the inferior vena cava was cut. The liver was perfused with PB (40 mL of PBC diluted in 1 L ddH 2 O). The PBC solution preparation method was as follows: NaCl (103.75 g), KCl (6.25 g), and Hepes (28.70 g) were dissolved in 350 mL H 2 O; 1 M NaOH (75 mL) was added, and ddH 2 O was subsequently added to make up a volume of 500 mL. Next, the liver was perfused with digestion buffer (type IV collagenase [

Histopathology
The specimen was sequentially fixed in 10% formalin for 24-48 h and embedded in paraffin for histopathological analysis. Thin sections (5 μM) were stained with hematoxylin and eosin staining (H&E) for histopathological analysis. The stained tissues were scanned using Panoramic MIDI (3D HISTECH, Hungary) and viewed using Case-Viewer slice software.

Oil-Red O staining
Fresh liver tissues were immersed in freezing medium and stored at −80°C. The liver tissues were sliced, washed with PBS three times (5 min), and stained with Oil Red O reagent according to the manufacturer's protocol (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). The stained tissues were scanned using Panoramic MIDI (3D HISTECH, Hungary) and viewed using CaseViewer slice software.

Immunohistochemistry (IHC)
IHC was performed on paraffin sections to examine the expression of F4/80 and PTGIS using a microwave-based antigen retrieval technique. Sections were incubated with rabbit anti-F4/80 (1:50 dilution) and anti-PTGIS (1:200 dilution) overnight at 4°C. After incubation with the secondary antibody and 3, 30-diaminobenzidine tetrahydrochloride, the slides were counterstained with hematoxylin. Non-immune rabbit IgG instead of the primary antibody was used for the negative control (NC). The stained tissues were examined with an inverted fluorescence microscope (OLYMPUS IX83, Tokyo, Japan) or scanned using Panoramic MIDI (3D HISTECH, Hungary) and viewed using CaseViewer slice software.
Transfection of PTGIS overexpression plasmid constructs and PTGIS siRNA into RAW264.7 cells Overexpression or knockdown of PTGIS was assessed through western blotting and real-time quantitative PCR (RT-qPCR) analysis after transfection of RAW264.7 cells with PTGIS overexpression plasmids or siRNA. All siR-NAs and plasmid constructs were obtained from Gene-Pharma Co., Ltd. Cells (3 ×10 5 /mL) were seeded in six-well plates and transfected with the PTGIS plasmid or siRNA and control constructs were mixed with lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. Cells were incubated with Opti-MEM at 37°C and 5% CO 2 for 6 h. Subsequently, cells were cultured in DMEM containing 10% FBS. The PTGIS overexpressing or knockdown cells were treated with LPS (1 μg/mL) and IFNγ (10 ng/mL) for 24 h to polarize the M1 phenotype, or were cultured with IL-4 (15 ng/mL) for 24 h to induce M2 macrophages. The cells were subsequently harvested for western blotting, RT-qPCR, or other analyses.

ELISA assay
The concentrations of IL-1β, MCP-1, IL-10, and TGF-β in the serum and culture supernatant were determined using Joyee mouse ELISA kit (Joyee Biotechnics Co.,Ltd, Anhui, China) according to the manufacturer's introductions.
Optical density values were measured at 450 nm. Three independent experiments were performed, and each sample was quantified with three replicates.

Immunofluorescence
To determine the expression of PTGIS in vivo, immunofluorescence staining was performed. Briefly, freshly dissected liver tissues were embedded in opti-mum cutting temperature compound, and the sections (8 μm thick) were cut with a cryotome Cryostat (Leica, CM 1950, Germany). Frozen tissue sections were incubated with 10% BSA for 1 h at room temperature (RT) to block nonspecific staining. After rinsing, the slides were incubated with primary antibodies for PTGIS (1:200 dilution) overnight at 4°C. The slides were then washed twice with PBS (5 min) and incubated with FITC (green)-conjugated secondary antibody (1:50 dilution) for 1.5 h at RT in the dark. RAW264.7 cells were fixed and permeabilized in 4% paraformaldehyde and 0.2% TritonX-100 in PBS for 10 min. Nonspecific binding was blocked with 10% BSA for 1 h at RT. Subsequently, the cells were incubated with primary antibodies for PTGIS (1:200 dilution) overnight at 4°C. Sections were washed twice with PBS and incubated with fluorescein-labeled secondary antibody at a dilution of 1:50 for 1.5 h at RT in the dark. Slides were mounted in mounting media with DAPI for 5 min at RT. After washing twice with PBS, the slides were covered with DABCO and the stained tissues were examined with an inverted fluorescence microscope (OLYMPUS IX83, Tokyo, Japan). Fluorescence density was measured using Image Pro Plus 6.

Recombinant adeno-associated virus-mediated PTGIS overexpression in mice
Mouse PTGIS over-expression plasmid labeled with green fluorescent protein (GFP) was obtained from Hanbio Biotechnology Co., Ltd. (Shanghai, China). The PTGIS plasmid was packaged with recombinant adeno-associated-virus for overexpression of PTGIS in vivo. Male C57BL/6j mice (age, six-to-eight weeks, weight, 18-22 g) were housed at the Animal Experimental Center for one week to adapt to the environment. The tails of mice were wiped with alcohol to expand the tail vein for injection. Mice were slowly injected with 100 μL of recombinant adeno-associatedvirus-packaged PTGIS over-expression plasmid at a concentration of 1 × 10 11 v.g/mL/mouse through the tail vein using a 0.5 mL insulin syringe. One week later, mice were fed either control diet or EtOH diet.

RNA sequencing (RNA-seq) and computational analysis
Briefly, RAW264.7 cells transfected with pEX3 and pEX3-PTGIS plasmid were collected and lysed with TRIzol reagent (Invitrogen, USA) according to the manufacturer's instructions. RNA was sequenced using the Illumina HisSeq 1000 at Novogene in Beijing (https://en.novogene.com). Raw data were analyzed using Illumina Casava 1.8, and three types of sequence reads-reads with adapter, no exact base information, or low-quality reads-were filtered and removed. RNA (3 μg) was used for library preparation with the NEBNext® UltraTM RNA Library Prep Kit for Illu-mina® (NEB, USA). The reference genome and gene model annotation files were downloaded directly from the genome website. The index of the reference genome was built using Hisat2 v2.0.5, and paired-end clean reads were aligned to the reference genome using Hisat2 v2.0.5. Differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq2 R package (1.16.1). The threshold we used to screen upregulated or downregulated mRNAs was |log 2 fold-change | > 1 and p value < 0.05.

Dual-luciferase reporter assay
The binding site of 3′-UTR-PTGIS and miRNA-140-3p.1 was detected via the dual-luciferase reporter assay. Briefly, the wt-PTGIS and mut-PTGIS sequences were inserted into the KpnI and HindIII sites of the pSI-Check2 promoter vector (Hanbio Biotechnology Co., Ltd., Shanghai, China) in a dual-luciferase reporter assay. Finally, the 293 T cells were grown in 96-well plates until they reached 50-70% confluence. Subsequently, 0.16 μg PTGIS plasmid was cotransfected with mmu-miR-140-3p.1 mimics/NC. The dualluciferase assay was carried out 48 h after transfection. The experiment was independently performed three times.

Statistical analysis
Data are expressed as the mean ± s.e.m. One-way analysis of variance and the Newman-Keuls post hoc test (Prism 8.3 GraphPad Software, Inc, San Diego, CA, USA) were used to analyze the results. p < 0.05 was considered statistically significant.

Alcohol consumption causes liver injury, steatosis, and inflammation
Mice with chronic ethanol feeding were characterized by liver injury. To investigate the degree of liver injury induced by ethanol consumption, we performed H&E staining and Oil Red O staining, measured serum ALT, AST, TG, and T-CHO levels, and determined the body weights and liver-to-body weight ratios. Histopathological analysis showed extensive steatosis, liver cell cord derangement, and intercellular space dilatation in EtOH-fed mice (Fig. 1A). Serum ALT, AST, TG, and T-CHO levels were substantially elevated in the EtOH-fed mouse group (Fig. 1B). As shown in Fig. 1C, body weight decreased in the early stage and subsequently increased in the EtOH-fed group. However, the liver-to-body weight ratio was elevated in EtOH-fed group (Fig. 1D). We further investigated macrophages infiltration in liver tissues using IHC analysis. Intriguingly, chronic alcohol feeding caused a marked elevation in the total number of macrophages in C57BL/6j mice, as assessed via F4/80 immunostaining (Fig. 1A), associated with an obvious increase in M1-polarized macrophages and a moderate rise of M2-polarized macrophages ( Fig. 1E and F). The in vitro results were consistent with the in vivo results (Fig. 1G). The results indicate that chronic alcohol consumption leads to liver injury, steatosis, and significant macrophage M1 polarization in vivo and in vitro.

Chronic alcohol consumption decreases PTGIS expression both in vivo and in vitro
To identify the expression changes of PTGIS induced by chronic alcohol consumption, western blotting, RT-qPCR, and IHC analyses were performed. As shown in Fig. 1H-J, PTGIS expression significantly decreased in the EtOH-fed group. As shown in Fig. 1G, PTGIS mRNA expression was significantly decreased in RAW264.7 cells cultured with EtOH and LPS. The immunosignals of PTGIS in EtOH-fed mice substantially decreased (Fig. 1J). The immunofluorescence signal of PTGIS significantly decreased in M1-polarized RAW264.7 cells and increased in M2polarized RAW264.7 cells (Fig. 1K). These results indicate that chronic alcohol consumption decreased PTGIS expression both in vivo and in vitro.

Recombinant adeno-associated virus-mediated overexpression of PTGIS protects against chronic alcohol consumption-induced liver injury, steatosis, and inflammation in vivo
The expression of PTGIS, both at protein and mRNA levels, was significantly downregulated by EtOH consumption both in vivo and in vitro. We investigated whether PTGIS-targeted therapy could prevent chronic alcohol-induced liver injury in vivo. The eGFP signal in the liver transfected with the PTGIS overexpression plasmid indicated a successful infection of the virus (Fig. S1A). PTGIS expression was substantially increased in primary macrophages isolated from recombinant adenoassociated viral vector (rAAV8)-PTGIS-transfected liver compared to rAAV8-empty-vector-treated liver tissue at both the protein and mRNA levels ( Fig. 2A, B). As shown in Fig. 2C, overexpression of PTGIS in liver tissues alleviated alcohol-induced liver injury (H&E staining) and inhibited liver steatosis (Oil Red O staining) and macrophage infiltration (F4/80 staining) compared to empty-virus-treated mice in the EtOH-fed group. Forced expression of PTGIS dramatically elevated the bodyweight and suppressed the liver-to-body weight ratio of mice compared to the empty-virus-infected group in the EtOH-fed group (Fig. S1B, S1C). As shown in Fig. 2D, E, forced PTGIS expression in vivo could inhibit M1 polarization and promote macrophage switch to the M2 phenotype. The ELISA results were consistent with those of the RT-qPCR analysis (Fig. 2F). These results indicated that forced expression of PTGIS in vivo inhibited M1 polarization and promoted M2 polarization and may be a potential therapeutic target for chronic alcohol-induced liver injury.

The influence of altered PTGIS expression on M1-polarized macrophages
To investigate the influence of PTGIS on M1 polarization, we performed loss-and gain-of-function-experiments. The transfection efficiency of PTGIS plasmid and PTGIS-siRNA is shown in Fig. 4A and E. As shown in Fig. 4B and C, PTGIS overexpression plasmid significantly elevated PTGIS expression at mRNA and protein levels in M1polarized RAW264.7 cells. Forced PTGIS expression inhibited inflammatory gene expression in M1-polarized macrophages ( Fig. 4C and D). As shown in Fig. 4F and G, PTGIS-siRNA inhibited PTGIS expression at the mRNA and protein levels in M1-polarized macrophages. Furthermore, PTGIS silencing induced the elevation of inflammatory gene expression in M1-polarized macrophages ( Fig. 4H-I). Interestingly, IL-6 levels decreased in the PTGIS overexpression group and increased in the PTGIS silencing group in M1-polarized macrophages (Fig. 4C, D,  H, I). IL-6 acts as a pro-inflammatory cytokine in various conditions [32]. However, Miller et al. demonstrated that the activation of IL-6/STAT3 has a hepatoprotective effect in IL-10 −/− mice [33]. Forced PTGIS expression in RAW 264.7 cells without LPS and IFNγ or IL-4 treatment inhibited macrophage M1 polarization and promoted M2 polarization (Fig. S3A), whereas PTGIS silencing in RAW264.7 cells without LPS and IFNγ or IL-4 treatment promoted M1 polarization but inhibited M2 phenotype (Fig. S3B). These results illustrated that forced PTGIS expression inhibits M1 and promotes M2 polarization as well as enhances the hepatoprotective cytokine IL-6 expression.

The effect of altered PTGIS expression on M2-polarized macrophages
To investigate the influence of PTGIS on M2 polarization, we performed loss-and gain-of-function-experiments. The PTGIS plasmid significantly increased PTGIS expression at the mRNA and protein levels in M2-polarized macrophages (Fig. 5A, B), and forced PTGIS expression significantly elevated anti-inflammatory gene expression (Fig. 5C, D). Furthermore, PTGIS-siRNA significantly inhibited PTGIS expression in M2-polarized RAW264.7 cells (Fig. 5E, F), and PTGIS silencing induced decreased expression of antiinflammatory genes (Fig. 5G, H). These results indicated that PTGIS could promote M2 macrophage polarization.

PTGIS may regulate macrophage polarization via the JAK/STAT pathway
To further evaluate the mechanism of PTGIS in regulating macrophage polarization, we forced PTGIS expression in RAW264.7 cells cultured with LPS and EtOH and subsequently isolated total mRNA for RNA-seq (Fig. 6A). As shown in Fig. 6B, we detected 11,244 genes in the PTGIS overexpression group and 11,123 genes in the pEX3transfected group. As shown in the volcano plot, 290 mRNAs were upregulated and 99 mRNAs were downregulated in the RAW264.7 cells transfected with the PTGIS plasmid compared to those of the pEX3-transfected group (|log2FC | > 1, p < 0.05) (Fig. 6C). Gene ontology analysis revealed that PTGIS overexpression affected many genes associated with inflammatory responses, chemokine activity, chemokine receptor activity, and cytokine activity and so on (Fig. 6D). KEGG pathway analysis revealed that forced PTGIS expression could alter JAK-STAT signaling, NOD-like receptor signaling, PI3K-AKT signaling, cytokine-cytokine receptor interaction, IL-17 signaling, etc. (Fig. 6E). As the JAK/STAT pathway plays an important role in macrophage polarization, we focused on the influence of forced PTGIS expression on the JAK/STAT pathway. The proteins interacting with JAK were predicted by the STRING database (Fig. 6F). The overlap between RNA-seq and the STRING database was IL-6 ( Fig. 6G). Furthermore, we assessed the effects of altered PTGIS expression on JAK/STAT signaling via western blotting analysis (Fig. S4A-S4D). The results revealed that forced PTGIS expression inhibited JAK/ STAT1 activation in M1-polarized macrophages and promoted JAK/STAT6 activation in M2-polarized macrophages ( Fig. S4A and S4C). PTGIS silencing promoted JAK/STAT1 activation in M1-polarized macrophages and inhibited JAK/STAT6 activation in M2-polarized macrophages ( Fig. S4B and S4D). These results suggest that PTGIS might regulate macrophage polarization through JAK/STAT signaling.

Discussion
Alcohol consumption is a leading risk factor for death and disability worldwide. A systematic analysis of the global burden of alcohol revealed that the level of consumption that minimizes heath loss is zero [1]. ALD appears as a wide range of disorders, ranging from simple liver steatosis to more severe forms of liver injury, including alcoholic hepatitis, liver fibrosis, cirrhosis, and hepatocellular carcinoma [33]. Ethanol metabolism-associated oxidative stress, glutathione depletion, abnormal methionine metabolism, malnutrition, ethanol-mediated induction of gut endotoxin leakage, and subsequent activation of macrophages play important roles in the pathogenesis of ALD [34][35][36][37][38][39]. In this study, we illustrated for the first time, to our knowledge, that PTGIS decreased in a chronic alcohol-consuming mouse model, and overexpression of PTGIS in vivo alleviated liver steatosis, inflammatory cell infiltration, and macrophage switch to the M1 phenotype (Fig. 7).
Ethanol increases fatty acid synthesis in hepatocytes via the upregulation of sterol regulatory element-binding protein 1c [40], a transcription factor that promotes fatty acid   B Veen diagram showed mRNAs detected in pEX3 and PTGIS overexpression plasmid transfected group. C Volcano plot was presented. D, E GO and KEGG analysis of altered mRNAs in pEX3 and PTGIS overexpressed transfected group. F The proteins interacted with JAK were predicted by STRING database. G The overlap between RNA-seq and STRING was IL-6. synthesis via the upregulation of lipogenic genes and the inhibition of PPARα [41,42], a nuclear hormone receptor that controls the transcription of a range of genes involved in free fatty acid transport and oxidation. In addition, ethanol can affect the activities of enzymes involved in fat metabolism by inhibiting AMP-activated protein kinase, which reduces fat metabolism and fatty liver [43,44]. In this study, we demonstrated that PTGIS overexpression in vivo alleviated fatty acid accumulation in the liver, as determined by Oil Red O staining (Figs. 1A and 2C).
Alcohol consumption not only causes enteric dysbiosis and bacterial over-growth [45] but also increases gut permeability and the translocation of bacteria-derived LPS from the gut to the liver [46]. LPS interacts with Toll-like receptor 4 (TLR4) to activate the MyD88-independent signaling pathway, leading to the production of proinflammatory cytokines, including IL-1β and TNF-α [47][48][49]. Interestingly, activation of TLR4 and complement factors also causes macrophages to produce hepatoprotective cytokines, such as IL-6, and anti-inflammatory cytokines, such as IL-10 [33,50]. In this study, the results indicated that chronic alcohol consumption induces M1-and M2-polarized macrophages to increase simultaneously ( Fig. 1E-G) and inhibit PTGIS expression in vivo and in vitro (Fig. 1H-K). Overexpression of PTGIS could alleviate liver injury in vivo (Figs. 2C, S1B, and S1C). Forced PTGIS expression inhibited macrophage M1 polarization and promoted M2 polarization both in vivo and in vitro (Figs. 2D-F, 4, and 5). A range of evidence demonstrated that the JAK/STATs signaling pathway plays a pivotal role in regulating macrophage polarization. RNA-seq analysis revealed that that the genes regulated by PTGIS were enriched in the JAK/STAT pathway ( Fig. 6D and E). Forced PTGIS expression inhibited JAK/STAT1 activation in M1-polarized macrophages and promoted JAK/STAT6 activation in M2-polarized macrophages ( Fig. S4A and S4C). PTGIS silencing promoted JAK/ STAT1 activation in M1-polarized macrophages and inhibited JAK/STAT6 activation in M2-polarized macrophages ( Fig. S4B and S4D). The overlap between the altered genes associated with PTGIS and the genes interacted with JAKs was IL-6 ( Fig. 6G). The activation of STAT1 (pSTAT1) and STAT3 (pSTAT3) plays a pivotal role in controlling liver steatosis and liver inflammation [51]. STAT3, which is activated by IL-6 plays an important role in hepatoprotection, liver regeneration, and glucose homeostasis by inducing a variety of anti-apoptotic and mitogenic proteins [51]. Miller et al. demonstrated that deletion of IL-6 or hepatic STAT3 increased STAT1 activation in HFD-fed IL-10 −/− mice [33]. IL-10 is a welldocumented anti-inflammatory cytokine and IL-6 is a proinflammatory cytokine [52]. However, IL-6 can act as an anti-inflammatory cytokine by inhibiting the activation of STAT1 [33]. Therefore, we hypothesized that chronic alcohol consumption leads to liver injury and inflammation and forced PTGIS expression can inhibit M1 macrophage polarization by inhibiting the activation of STAT1 through IL-6, and IL-6 acts as an anti-inflammatory cytokine on this occasion.
A range of evidence over the last decade indicates a pivotal role of miRNAs as key mediators of macrophage differentiation, infiltration, and activation. Initial studies have indicated the property of miRNAs to regulate the magnitude of the innate response, participating as integral components of feedback loop regulatory mechanisms, which significantly shape the inflammatory response [53]. In addition, recent studies have demonstrated that miRNAs regulate macrophage differentiation and polarization [54]. In short, a complex and highly regulated network of miR-NAs regulating inflammatory pathways by targeting components of TLR signaling and influencing downstream inflammatory cytokine production has been defined. In this study, we demonstrated that the regulation of PTGIS in macrophage polarization could be influenced by miRNA-140-3p.1 (Fig. 3).
Macrophages are hindered by their versatility, and can adapt their phenotype (to become pro-inflammatory or antiinflammatory) in response to molecular signals in healthy or injured livers. Historically, the plasticity of macrophages has been described using the terms classically activated (M1 type) and alternatively activated (M2 type). However, this nomenclature is too simple to describe the polarization phenotypes of macrophages driven by many different stimuli in the microenvironment. Ly-6C is a cell surface glycoprotein widely used to identify functionally discrete murine circulating monocyte populations: Ly-6C hi monocytes and Ly-6C lo . Ly-6C hi monocytes are recruited early to inflammatory environments and are thought to be proinflammatory, whereas Ly-6C lo monocytes are a patrolling cell type and can supplement resident tissue macrophages [55,56]. Further studies are needed to investigate the influence of altered PTGIS expression on differential Ly-6C expression in chronicALD.
In conclusion, as shown in Fig. 7, we demonstrated that PTGIS expression was downregulated in chronic alcoholconsuming mice via elevated miR-140-3p.1 expression. Forced PTGIS expression in vivo and in vitro can inhibit macrophage switch to the M1 phenotype and promote M2 polarization. RNA-seq results showed that PTGIS might regulate macrophage polarization through the JAK/STAT pathway, and IL-6 might be involved in inhibiting the activation of STAT1. This study revealed that PTGIS might be a therapeutic target for ALD.

Data availability
The datasets used and analyzed in this study are available from the corresponding author on reasonable request.

Compliance with ethical standards
Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license 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 license, visit http://creativecommons. org/licenses/by/4.0/.