DHFR silence alleviated the development of liver fibrosis by affecting the crosstalk between hepatic stellate cells and macrophages

Abstract Liver fibrogenesis is a dynamic cellular and tissue process which has the potential to progress into cirrhosis of even liver cancer and liver failure. The activation of hepatic stellate cells (HSCs) is the central event underlying liver fibrosis. Besides, hepatic macrophages have been proposed as potential targets in combatting fibrosis. As for the relationship between HSCs and hepatic macrophages in liver fibrosis, it is generally considered that macrophages promoted liver fibrosis via activating HSCs. However, whether activated HSCs could in turn affect macrophage polarization has rarely been studied. In this study, mRNAs with significant differences were explored using exosomal RNA‐sequencing of activated Lx‐2 cells and normal RNA‐sequencing of DHFR loss‐of‐function Lx‐2 cell models. Cell functional experiments in both Lx‐2 cells and macrophages animal model experiments were performed. The results basically confirmed exosomes secreted from activated HSCs could promote M1 polarization of macrophages further. Exosome harbouring DHFR played an important role in this process. DHFR silence in HSCs could decrease Lx‐2 activation and M1 polarization of M0 macrophages and then alleviate the development of liver fibrosis both in vitro and vivo. Our work brought a new insight that exosomal DHFR derived from HSCs had a crucial role in crosstalk between HSCs activation and macrophage polarization, which may be a potential therapeutic target in liver fibrosis.

macrophages (M0) depending on local microenvironment polarize into classically activated macrophages (M1) or alternatively activated macrophages (M2) with specific functions. 7 M1 macrophages mediate tissue damage and initiate inflammatory responses, while M2 macrophage is an essential player in the resolution of inflammation, so M1/M2 macrophage balance polarization governs the fate of an organ in inflammation or injury. 8 Exosomes harbour various biomolecules, including lipid, proteins, DNAs, mRNAs and non-coding RNAs, which can be transferred to recipient cells and modulate their functions 9,10 Exosomes are known as important mediators of cell-cell communication. 11 In liver, various types of cell can either release or receive exosomes in cell-cell communication, including HSCs and macrophage. Exosomes involve in various aspects of liver physiology and pathology and in the progress of liver diseases, like HSCs activation and participating in liver fibrosis. 12 Exosomes derived from qHSCs or aHSCs possess anti-fibrotic and pro-fibrotic properties respectively. 13 Exosomes derived from natural killer cells can inhibit HSCs activation and liver fibrosis. 14,15 Macrophage also could take up exosomes from other cells which leading to macrophages activation, 16 polarization which directly influence macrophage function. [17][18][19][20][21][22] However, whether aHSCs-derived exosomes could affect macrophage polarization and further regulate the progression of liver fibrosis have not been fully studied.
Our research found aHSCs-derived exosomes promoted M1 macrophage polarization and inhibited M2a polarization. Then, exosome-RNA-sequencing (Exo-RNA-seq) was conducted and bioinformatic analysis results showed that dihydrofolate reductase (DHFR) was one of the most significant up-regulated genes in exosomes derived from activated Lx-2 cells when compared with quiescent Lx-2 cells which suggested it could participate in liver fibrosis. 23 DHFR is a crucial target in anticancer drug development which plays key roles in the nucleic acid biosynthesis. Through TCGA online database retrieval, DHFR was found to be significantly up-regulated in LIHC primary tissues and it was negatively correlated with the overall survival rate of LIHC patients. However, there is no relevant research on whether DHFR is involved in the regulation of fibro- In brief, our results brought a new insight that aHSCs-derived exosomal DHFR has a crucial role in crosstalk between HSC activation and macrophage polarization, which may be a potential therapeutic target in liver fibrosis.

| Ethics statement
Animal and human studies were approved by the ethics committee of Xiangya Second Hospital of Central South University (Project number: 2020161; 2020R113). All selected patients were treated based on the Liver Disease Treatment Guide. 24 Additionally, the patients' consent and the approval were all obtained .
During the process of study, the ethical standards of the Committee on Human Experimentation and the Helsinki Declaration were strictly obeyed. Liver tissues from patients with liver fibrosis and cirrhosis were diagnosed by liver biopsy. Both tumour tissues and bile samples were collected from surgical operations.
Animal care in the research was conducted according to the Guide for the Care and Use of Laboratory Animals enacted by the US National Institutes of Health and was approved by the Animal Research Committee of Center of Central South University.

| Lx-2 cell activation and macrophage polarization
Lx-2 cells were cultured in DMEM medium containing 2% foetal bovine serum for 24 h. After starvation for 12 h, Lx-2 cells were treated with 5 ng/ml TGFβ for another 48 h.

| Exosome uptake labelling
To confirm exosomes were absorbed by receptor macrophages, PKH67 mixer (Cat.no CGLDIL, Sigma, in Diluent C) was used to label exosomes from 1.5 × 10 6 cells according to the manufacturer's instructions. The stained cells were observed and photographed with a fluorescence confocal microscope (Zeiss LSM880).

| RNA-sequencing
As to Exo-RNA-seq, total RNAs were extracted from the exosomes of Lx-2 culture supernatant. Besides, RNAs were extracted from exosomes from activated (Lx-2-TGFβ) and quiescent Lx-2 cells (Lx-2-C), and then the RNA-seq and relative bioinformatics analysis were performed in Kaitai Biotech and Science & Well biotech. The differentially expressed genes (DEGs) between exosomes derived from Lx-2-TGFβ and Lx-2-C groups were selected by the difference multiples (|log2FoldChange| > 1) and significance (p < 0.05). The subsequent RNA-seq in DHFR-silenced Lx-2 cells was also implemented according to the above methods.

| Western blot assay
For the sake of detecting the expression of fibroblast activation markers and inflammatory markers of macrophages, Western blot was carried out as previously described. 25 Amount of the protein of interest was normalized to the densitometric units of β-actin.
Antibodies used were shown in Supplementary Material.

| Cell survival assay
In order to evaluate the viability of Lx-2 cells after DHFR silence, the Cell Counting Kit-8 (CCK-8) assay was used to assess the cell survival ability until 96 h as previously described (Cat.no K1018; APExBio). 26 To evaluate the viability of M0 macrophages absorbed different exosomes, CCK-8 was performed again at 24 and 48 h respectively.

| H&E staining and Masson Trichrome staining
6-8-week aged male C57BL/6 mice were purchased from the Promab biotech Co. Ltd. Mice were randomly divided into 3 groups (n = 6), and all mice were maintained under standard conditions and diet. Experimental hepatic fibrosis was induced by CCl-4 (10% in olive oil, 2 ml/kg, twice a week for 6 weeks). Paraformaldehyde-fixed liver tissue sections were stained with haematoxylin and eosin (H&E) and Masson's Trichrome staining to evaluate liver fibrosis. Collagen deposition was quantitatively analysed using the following formula: collagen area/total area × 100%.

| Statistical analysis
All tests were set up in triplicate. Statistical analysis was performed and presented with Graphpad Prism 7.0 software (GraphPad Software). Differences between two independent groups were evaluated by Student's t tests. Differences for multiple comparisons were calculated using one-way ANOVA. p < 0.05 was considered significant differences. Data are presented as mean ± SD.

| Exosomes derived from activated Lx-2 induced M1 macrophage polarization of M0 macrophage
To evaluate the effect of exosomes derived from aHSCs on mac- showed that the content of exosome marker proteins CD81 and TSG101 in exosomes was significantly higher than those in cell lysates. Calnexin, as a negative control, had no significant expression in exosome samples ( Figure 1D). Exosome morphology was examined using TEM which were showed membrane-bounded, spherical shape vesicles with a size range of 50-100 nm ( Figure 1E) and was consistent with the typical morphology of exosomes reported previously.
In order to investigate the effect of exosomes secreted by acti- The results showed that the expression level of DHFR increased gradually from liver fibrosis to liver cirrhosis and then to liver cancer at least in the limited collected samples ( Figure 2E). Subsequently, DHFR expression was verified again; in collected bile samples, it was found that DHFR enhanced obviously in bile samples of liver fibrosis patients when compared with bile samples from normal volunteers ( Figure 2F).

| DHFR silence in Lx-2 alleviated activation of HSCs and M1 macrophage polarization of M0 macrophage
To investigate whether DHFR plays a role in the regulation of Lx-2 activation, DHFR was knocked-down in Lx-2 by siRNA interference.

| INPP5D was one of the downstream negative regulators of DHFR in Lx-2 cells
In order to clarify the downstream regulators of DHFR in Lx-2 cells,  Finally, a comprehensive picture of the molecular mechanisms elucidated in this study was drawn in Figure 6E. But the mechanism of exosomes derived from aHSCs involved in liver fibrosis is poorly understood. 31  in liver fibrogenesis, the potential mechanism of DHFR regulating INPP5D expression was not fully explored in this study. Although we tried to explore this potential mechanism both through bioinformatics analysis or literature retrieval, unfortunately, DHFR is neither a transcriptional regulator nor an epigenetic regulator because of which there is little relevant information. Only the following information is available for reference: DHFR silence resulted in inhibition of folate metabolism then led to enhancement of activators of PI3K signalling. 38 Meanwhile, INPP5D was found to be one of the subsequent co-activators after PI3K activation. 39 In addition, the regulatory mechanism between DHFR and INPP5D would be involved more complex network, similar to competitive RNA and so on, which needs further scientific exploration.

| DISCUSS ION
The data from TCGA database about the DHFR in liver primary tumour samples and its expression on liver cancer patients' survival were analysed. The results indicated DHFR was obviously up-regulated in tumour group and its expression was negatively correlated with survival probability of liver cancer patients. DHFR expression was tested in liver fibrosis tissues, liver cirrhosis and liver cancer respectively for the first time. It was found DHFR was gradually enhanced from liver fibrosis, liver cirrhosis to liver cancer in terms of the overall trend. These results provided a new understanding for DHFR in the development from liver fibrosis to liver cancer, maybe DHFR will become a promising therapeutic target in liver fibrosis, as well as a predicted and prognosis biomarker of liver cancer.

ACK N OWLED G EM ENTS
We are also very grateful to Ms. Weiwei (director of Hunan s Science & Well Biotechnology Limited) for all her help in the experiment implementation and manuscript compilation.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request.