Abstract

   Hepatic fibrosis is a common disease with high morbidity and mortality
   rates. The complex and poorly understood mechanisms underlying hepatic
   fibrosis represent a significant challenge for the development of more
   effective therapeutic strategies. MKP5 is a potential regulator of
   multiple fibrotic diseases. However, its precise role and mechanism of
   action in hepatic fibrosis remains unclear. This study identified a
   reduction in MKP5 expression in fibrotic liver tissues of mice treated
   with CCl[4] and observed that MKP5 knockout mice exhibited a more
   pronounced development of hepatic fibrosis. In addition, RNA-seq data
   indicated activation of protein processing in the endoplasmic reticulum
   signalling pathway in fibrotic liver tissues of mice lacking MKP5.
   Mechanistically, MKP5 inhibits the activation of hepatic stellate cells
   (HSCs) and hepatocyte apoptosis through the regulation of the IRE/XBP1
   pathway. Based on these findings, we developed PLGA-MKP5 nanoparticles
   coated with a mesenchymal stem cell membrane (MSCM). Our results
   demonstrated that MSCM-PLGA-MKP5 was most effective in attenuating
   hepatic inflammation and fibrosis in murine models by modulating the
   IRE/XBP1 axis. This study contributes to the current understanding of
   the pathogenesis of hepatic fibrosis, suggesting that the targeted
   delivery of MKP5 via a nano-delivery system may represent a promising
   therapeutic approach to treat hepatic fibrosis.

Graphical abstract

   [44]graphic file with name 12951_2024_3029_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-024-03029-8.

   Keywords: Hepatic fibrosis; MKP5; HSCs activation; Hepatocytes
   apoptosis; Nano-delivery system, Mesenchymal stem cell membrane

Introduction

   Hepatic fibrosis (HF) is a pathological process of repair and
   remodelling that can result from several factors, including viral
   infection, alcohol abuse, non-alcoholic steatohepatitis, cholestasis,
   and autoimmune diseases. It is a common precursor to cirrhosis and
   hepatocellular carcinoma [[45]1, [46]2]. HF has emerged as a
   significant public health concern because of its high prevalence and
   adverse clinical outcomes. Given the current dearth of definitive
   pharmaceuticals targeting HF and incomplete understanding of its
   pathogenesis, it is imperative to explore the molecular mechanisms
   underlying its prevention and treatment.

   It is widely accepted that inflammatory disorders are the primary
   initiators of HF. During this phase, macrophages and other cells
   secrete inflammatory cytokines and release TGF-β, which activates HSCs.
   These cells are characterised by cell proliferation and chemotaxis as
   well as a transition of cell morphology to myofibroblast-like cells.
   Concurrently, hepatocytes are damaged or undergo death, releasing
   damage-related molecular patterns (DAMPs), which destroy the normal
   liver structure and ultimately help to form HF [[47]2–[48]5]. Recent
   studies have identified endoplasmic reticulum stress (ERS) as a key
   factor in the pathogenesis of HF, with evidence that regulating ERS may
   offer a promising avenue for improving HF outcomes [[49]6, [50]7].
   Moreover, the specific inhibitor of IRE1a, STF-083010, has been
   demonstrated to mitigate CCl[4]-induced HF by upregulating the
   expression of miR-122 [[51]8]. It has been demonstrated that, during
   HF, ERS induces hepatocyte apoptosis [[52]9] and activates HSCs
   [[53]10], promoting the progression of HF. Both inflammation and ERS
   play pivotal roles in the development of HF. Consequently, modulation
   of these processes through the regulation of associated gene expression
   represents a pivotal strategy in the treatment of HF.

   Dual-specific mitogen-activated protein kinase phosphatases (MKPs)
   inactivates the mitogen-activated protein kinase (MAPK) pathway by
   selectively dephosphorylating serine/threonine residues [[54]11]. Our
   previous studies have demonstrated that MKP5 exerts inhibitory effects
   on obesity-induced inflammation [[55]12] and ERS-mediated type 1/2
   diabetes [[56]13, [57]14]. MKP5 has been linked to the modulation of
   pulmonary and cardiac fibrosis [[58]15, [59]16], indicating a potential
   involvement of MKP5 in the pathogenesis of metabolic disorders.
   However, the regulatory role of MKP5 in HF remains unclear. Therefore,
   it is imperative to explore the involvement of MKP5 in HF and to
   elucidate the molecular mechanisms involved in the prevention and
   treatment of HF.

   The complex aetiology of HF presents a significant challenge for the
   development of effective therapeutic options. Nucleic acid, as a
   universal genetic component encoding proteins, offer a promising avenue
   for treating a wide range of diseases, particularly compared to
   traditional drug targets [[60]17]. Nevertheless, clinical applications
   of nucleic acid drugs remain constrained by their inherent instability,
   intrinsic obstacles, and lack of selective targeting [[61]18]. In
   recent years, an increasing number of studies have demonstrated that
   chemical modification of nucleic acid drugs or the utilisation of
   carrier-delivery systems can enhance the therapeutic efficacy of
   nucleic acid drugs for disease treatment. Using nanoparticles (NPs) for
   drug delivery has emerged as a novel approach to treat diseases. It has
   the potential to enhance the stability of nucleic acid molecules,
   prevent their degradation by nucleases, and improve their effectiveness
   in penetrating cells [[62]19]. Nevertheless, NPs exhibit certain
   limitations regarding their biostability, targeting, and rapid
   clearance from the body, which present challenges for their use in
   clinical applications. Accordingly, a biomimetic approach utilising
   cell membranes has been identified as a potential solution [[63]20,
   [64]21]. Compared to other cell membranes, the mesenchymal stem cell
   membrane (MSCM) represents a novel carrier with active targeting,
   favourable biocompatibility, and immune escape characteristics of MSC.
   This characteristic can effectively mitigate the risk of immune
   rejection and enhance the treatment safety. Furthermore, MSCM protects
   nucleic acid drugs from degradation by nucleases in the bloodstream,
   prolonging their circulation [[65]22]. The coating of NPs with MSCM
   represents an efficient method for selective targeting, enabling drugs
   to enter specific cells, improving treatment accuracy. This approach
   has a wide range of applications in tumour therapy, inflammatory
   diseases, and tissue regeneration [[66]23].

   This study aimed to investigate the role of MKP5 in HF. Initially, a
   notable decline in MKP5 expression was observed in murine fibrotic
   liver tissue. Subsequently, MKP5 knockout mice were generated, and the
   absence of MKP5 resulted in a more pronounced CCl[4]-induced HF.
   Moreover, our results demonstrate that MKP5 plays a regulatory role in
   TGF-β1-induced HSC activation and hepatocyte apoptosis through the
   IRE/XBP1 pathway. Based on these findings, we developed a biomimetic
   PLGA-MKP5 coated with MSCM. This delivery system exhibited enhanced
   efficacy of MKP5 in targeting damaged liver tissue and demonstrated a
   pronounced anti-fibrotic effect in CCl[4]-induced HF without overt
   toxicity. This provides a promising strategy for effective
   anti-fibrotic therapy.

Results

MKP5 knockout exacerbates CCl[4]-induced hepatic fibrosis

   To investigate the relationship between MKP5 and HF, a CCl[4]-induced
   HF mouse model was established. After 8 weeks of CCl[4] stimulation,
   the liver exhibited structural abnormalities characterised by fibrotic
   scarring and substantial collagen deposition (Fig. S1A). Furthermore,
   fibronectin, the primary component of the extracellular matrix (ECM),
   and α-smooth muscle actin (α-SMA), a fibrosis marker protein, exhibited
   a notable increase in expression in CCl[4]-stimulated liver tissue
   relative to that in control mice (Fig. S1B). The results demonstrated
   that mice exhibited signs of HF following CCl[4] stimulation. It is
   noteworthy that the protein and mRNA levels of MKP5 were significantly
   decreased in HF liver tissues (Fig. [67]1A–C, S1C). Moreover, the
   relationship between MKP5 and collagen deposition, as well as the liver
   function markers alanine aminotransferase (ALT) and aspartate
   aminotransferase (AST), was investigated in mice with HF. The results
   demonstrated that an increase in collagen deposition and the levels of
   AST and ALT in the liver tissue were accompanied by a decrease in MKP5
   levels (Fig. [68]1D), indicating the potential involvement of MKP5 in
   the regulation of CCl[4]-induced HF.

Fig. 1.

   [69]Fig. 1
   [70]Open in a new tab

   MKP5 knockout exacerbates CCl[4]-induced hepatic fibrosis. The WT and
   KO mice were subjected to CCl[4]-induce hepatic fibrosis. A, B The
   protein level of MKP5 in liver tissues was determined by western
   blotting (A), and statistical analysis was performed using ImageJ
   software (B). C The mRNA level of MKP5 in liver tissues was detected by
   real-time PCR. (n = 6/group). D Line graphs and histograms summarising
   MKP5 expression samples compared to an indicator of collagen
   deposition, and line graphs and histogram summarising MKP5 expression
   compared to an indicator of ALT and AST in liver tissues (n = 3/group).
   E Liver weight and ratios of liver weight to body weight (M:
   n = 6–8/group, F: n = 9–10/group). F Levels of serum AST, ALT, MDA, and
   GSH, as well as Hyp in the liver (M: n = 6–8/group, F: n = 6–10/group).
   G Representative staining of HE, Masson, and Sirus Red in liver
   tissues. Scale bar: 100 μm. H Statistical analysis of Sirus Red and
   Masson staining of liver tissues. I Protein levels of MKP5,
   fibronectin, and α-SMA in liver tissue were estimated using western
   blotting, and statistical analysis was performed using ImageJ software.
   (M: n = 6–8/group, F: n = 6/group). J The mRNA levels of MKP5, ACTA2,
   CoL1A1, and Fn1 were analysed using real-time PCR (n = 4–6/group).
   *P < 0.05, **P < 0.01, ***P < 0.001

   To further investigate the role of MKP5 in HF, mice deficient in MKP5
   were constructed (Fig. S1D, S1E). Subsequently, CCl[4]-induced HF
   models were established using both female and male mice. There were no
   significant differences in body weight between the male and female mice
   (Fig. S1F). However, liver weight and the ratio of liver weight to body
   weight increased significantly following CCl[4] induction, and these
   increases were further pronounced in the CCl[4]-treated groups after
   MKP5 knockout (Fig. [71]1E). As illustrated in Fig. [72]1F, liver
   functional indicators, ALT and AST in serum and the hydroxyproline
   (Hyp) content in liver, were markedly elevated in CCl[4]-treated
   wild-type (WT) and KO mice compared with control mice. Moreover, the
   absence of MKP5 resulted in more pronounced exacerbation of these liver
   function markers in male mice, with a similar trend observed in female
   mice, albeit to a lesser extent. Malondialdehyde (MDA), an end-product
   of lipid peroxidation, serves as an indirect indicator of free radical
   damage to the liver. Glutathione (GSH), a crucial antioxidant, reflects
   the hepatic antioxidant capacity [[73]24]. In female mice,
   administration of CCl[4] resulted in an increase in MDA content and a
   decrease in GSH content in liver tissues. These changes were further
   exacerbated by MKP5 knockout. Similar trends were observed in male
   mice, although the differences were not statistically significant
   (Fig. [74]1F). In addition, the liver tissue of the KO-CCl[4] group
   exhibited a darker and harder consistency, with a notable increase in
   the number of diffuse fine granular lesions on the surface and within
   the sections, compared to that in the WT-CCl[4] group (Fig. S1G).
   Haematoxylin and eosin (H&E) staining revealed that the WT-CCl[4] group
   exhibited structural destruction of liver tissue, disorder of hepatic
   lobules, deviation or loss of central veins, and vacuolation of
   hepatocytes when compared to the WT-Control group. Masson and Sirus Red
   staining revealed markedly elevated collagen deposition in the liver
   tissue of the WT-CCl[4] group compared to that in the WT-Control group,
   which was further accentuated in CCl[4]-treated KO mice (Fig. [75]1G,
   H)). The CCl[4]-induced increase in α-SMA and fibronectin expression in
   KO mice was significantly higher than that in WT mice (Fig. [76]1I).
   Similarly, the transcription levels of the fibrosis-related genes
   ACTA2, CoL1A1, and Fn1 in CCl[4]-induced KO mice were significantly
   higher than those in WT mice (Fig. [77]1J). Similar results were
   observed in female mice. These findings suggest that MKP5 knockout
   enhances the development of CCl[4]-induced HF.

MKP5 knockout promotes HSCs activation during hepatic fibrosis

   The activation of HSCs represents a pivotal event in the pathogenesis
   of HF. Upon activation, HSCs undergo transformation into
   myofibroblasts, leading to excessive deposition of collagen-based ECM
   in the liver [[78]5]. As illustrated in Fig. [79]2B, the expression of
   α-SMA (a marker of activated HSCs) was elevated in the liver tissues of
   WT-CCl[4] mice and was further augmented in KO-CCl[4] mice. This
   suggests that HSCs were activated following CCl[4] induction and that
   MKP5 knockout exacerbated the activation of HSCs. Of particular note,
   the co-localisation of MKP5 and α-SMA was extensive in the liver
   tissue, indicating that MKP5 may regulate HF through its effect on HSC
   activation (Fig. [80]2A). In addition, an increase in the proliferation
   of HSCs was observed following CCl[4] treatment, and MKP5 knockout
   further promoted proliferation, as evidenced by the co-localisation of
   α-SMA and Ki67 (a marker of cell proliferation) (Fig. [81]2B). To gain
   further insight into the function of MKP5 in HSC activation, primary
   HSCs were isolated from WT and KO mice and treated with TGF-β1. The
   CCK-8 results demonstrated that MKP5 knockout enhanced the
   TGF-β1-induced proliferation of primary HSCs compared to the WT group
   (Fig. [82]2C). Similarly, inhibition of MKP5 expression resulted in a
   more pronounced TGF-β1-induced proliferation of LX-2 cells. Conversely,
   MKP5 overexpression resulted in the inhibition of LX-2 cell
   proliferation (Fig. [83]2D). As illustrated in Fig. [84]2E, TGF-β1
   elevated the expression of ECM proteins in primary HSCs from KO mice,
   including collagen Iα1, fibronectin and α-SMA, compared to WT mice. The
   results demonstrated notable discrepancies between the experimental
   groups (Fig. S2A). The results of real-time polymerase chain reaction
   (PCR) also indicated that the mRNA levels of the fibrosis-related genes
   ACTA2, CoL1A1, and Fn1 in TGF-β1-induced HSCs from KO mice were
   significantly higher than those in WT mice (Fig. [85]2G). Similarly,
   inhibition of MKP5 expression resulted in the up-regulation of
   ECM-related proteins (Fig. [86]2F, S2B) and genes (Fig. [87]2H) in LX-2
   cells induced by TGF-β1, exacerbating LX-2 cell activation. Conversely,
   overexpression of MKP5 reversed TGF-β1-induced activation of LX-2
   cells, as previously described (Fig. [88]2F, H, S2C). In HF, activated
   stellate cells migrate and undergo morphological transformation into
   fibroblasts. In this study, TGF-β1 induced the migration of LX-2 cells,
   and the inhibition of MKP5 expression further promoted TGF-β1-induced
   LX-2 cell migration. Conversely, MKP5 overexpression in LX-2 cells
   reversed TGF-β1-induced migration (Fig. [89]2I, J). Furthermore, the
   effect of MKP5 on the TGF-β1-induced LX-2 cell cycle was investigated.
   As illustrated in Fig. [90]2K, L and S2D, MKP5 overexpression resulted
   in the arrest of the TGF-β1-induced LX-2 cell cycle in the S phase due
   to the up-regulation of the protein expression of S phase-specific
   factors, including CyclinA2 and CDK2. Consequently, this regulatory
   mechanism inhibits cell proliferation. In conclusion, the absence of
   MKP5 promoted HSC activation during HF.

Fig. 2.

   [91]Fig. 2
   [92]Open in a new tab

   MKP5 knockout promotes HSCs activation during hepatic fibrosis. A The
   co-localisation of MKP5 and α-SMA in liver tissue was visualised by
   immunofluorescence (IF) staining. The nuclei were stained with DAPI.
   Scale bar: 100 μm. B IF staining was performed to visualise the
   co-localisation of α-SMA and Ki67 in the liver tissue. The nuclei were
   stained with DAPI. Scale bar: 100 μm. C Primary HSCs isolated from WT
   and KO mice were treated with TGF-β1 and cell proliferation was
   assessed using the CCK-8 assay. D LX-2 cells were infected with
   retrovirus to either overexpress or knock down MKP5 and then treated
   with TGF-β1. The proliferation of LX-2 cells upon TGF-β1 treatment was
   determined using a CCK-8 assay. E, G Expression levels of
   fibrosis-related proteins and genes in primary HSCs were analysed by
   western blotting (E) and real-time PCR (G). F, H–J LX-2 cells were
   infected with retrovirus to either overexpress or knock down MKP5, and
   subsequently treated with TGF-β1. Protein levels of Collagen Iα1,
   fibronectin, and α-SMA were detected by western blotting (F). The mRNA
   levels of ACTA2, Fn1, and CoL1A1 were determined by real-time PCR (H).
   Transwell inserts (I), and wound healing (J) were used to evaluate the
   migration of LX-2 cells, and quantification was performed using ImageJ
   software. Scale bar: 100 μm. K, L LX-2 cells were infected with a
   retrovirus to overexpress MKP5 and then treated with TGF-β1. The cell
   cycle was quantified by Annexin V-FITC/PI staining (K). Protein levels
   of CDK2 and CyclinA2 were measured by western blotting and the blots
   from three repeated experiments were quantified using ImageJ
   software(L). *P < 0.05, **P < 0.01, ***P < 0.001

MKP5 knockout aggravates hepatocyte apoptosis and epithelial-to-mesenchymal
transition (EMT) during hepatic fibrosis

   Hepatocytes constitute the primary parenchymal cell type in the liver
   during HF, ultimately resulting in hepatocyte apoptosis. In this study,
   we observed a reduction in the number of albumin-positive cells in
   CCl[4]-induced liver tissues (Fig. [93]3A). Furthermore, MKP5
   expression declined following CCl[4] treatment, which is in accordance
   with the observations presented in Fig. [94]1A and S1C. Of particular
   note, significant co-localisation between MKP5 and albumin was
   observed, which suggests that MKP5 may regulate HF by affecting
   hepatocyte injury (Fig. [95]3A). In addition, a greater number of
   transferase dUTP nick-end labelling (TUNEL)-positive hepatocytes was
   observed in CCl[4]-stimulated WT mice, indicating that CCl[4] induced
   hepatocyte apoptosis and that MKP5 knockout exacerbated this process
   (Fig. [96]3B). Moreover, Annexin V-fluorescein (FITC)/propidium iodide
   (PI) results demonstrated that MKP5 knockout significantly enhanced
   TGF-β1-induced apoptosis in primary hepatocytes compared to that in WT
   mice (Fig. [97]3C). Conversely, MKP5 overexpression reversed
   TGF-β1-induced apoptosis in AML12 cells (Fig. [98]3D). Subsequently,
   the expression levels of apoptosis-related proteins were evaluated. As
   illustrated in Fig. [99]3E, the protein levels of cleaved Caspase3,
   cleaved Caspase8 and cleaved Caspase9 were markedly elevated in KO mice
   following TGF-β1 treatment compared to WT mice. Conversely, MKP5
   overexpression led to a notable reduction in the expression of
   apoptotic proteins in TGF-β1-induced AML12 cells (Fig. [100]3F). Upon
   stimulation by pro-fibrotic factors, hepatocytes undergo EMT, causing
   the loss of their characteristic epithelial morphology and acquisition
   of a fibroblast-like phenotype, accompanied by increased secretion of
   the ECM and collagen [[101]25]. Our findings revealed that TGF-β1
   induction resulted in a notable reduction in E-cadherin protein
   expression in KO primary hepatocytes compared to WT mice. Concurrently,
   there was an observable increase in the levels of N-cadherin and
   Vimentin (Fig. [102]3G). Similarly, the inhibition of MKP5 expression
   exacerbated EMT induced by TGF-β1 in AML12 cells. Conversely, the
   overexpression of MKP5 resulted in the up-regulation of E-cadherin
   protein expression and downregulation of N-cadherin and Vimentin
   protein levels in TGF-β1-induced AML12 cells, attenuating
   TGF-β1-induced EMT transformation (Fig. [103]3H). These findings
   indicate that MKP5 knockout may contribute to the exacerbation of
   hepatocyte apoptosis and EMT during HF.

Fig. 3.

   [104]Fig. 3
   [105]Open in a new tab

   MKP5 knockout aggravates hepatocytes apoptosis and EMT during hepatic
   fibrosis. A Representative double-staining of MKP5 and albumin in liver
   tissues. The nuclei were stained with DAPI. Scale bar: 100 μm. B IF
   staining for Albumin and TUNEL staining in the liver tissue. The nuclei
   were stained with DAPI. Scale bar: 100 μm. C, D The ratios of apoptotic
   cells in primary hepatocytes (C) and MKP5 overexpression in AML12 cells
   (D) treated with TGF-β1 were quantified by flow cytometry using an
   Annexin V-FITC/PI staining kit. E, F The protein levels of cleaved
   Caspase3, cleaved Caspase8, and cleaved Caspase9 in primary hepatocytes
   (E) and AML12 cells (F) were measured via western blotting, and
   quantification was performed using ImageJ software. G, H Protein levels
   of E-cadherin, N-cadherin, and Vimentin in primary hepatocytes (G) and
   AML12 cells (H) were detected by western blotting. The data were
   quantified using the ImageJ software. *P < 0.05, **P < 0.01,
   ***P < 0.001

MKP5 knockout aggravates the hepatic inflammatory response during hepatic
fibrogenesis

   To elucidate the regulatory mechanism of MKP5 in HF, RNA sequencing was
   employed to analyse differentially expressed genes (DEGs) in the livers
   of CCl[4]-induced WT and KO mice. A total of 566 DEGs were identified,
   with 376 upregulated and 190 downregulated genes (Fig. [106]4A).
   Functional classification and pathway enrichment analysis of the DEGs
   were performed using the Kyoto Encyclopedia of Genes and Genomes
   (KEGG). Ours results demonstrated that there was a notable activation
   of pathways linked to inflammation, including MAPK and TNF signalling
   pathways (Fig. [107]4B). Furthermore, Gene set enrichment analysis
   (GSEA) also revealed that MAPK, TGF-β, and chemokine pathways were
   enriched, and there was a significant positive correlation with MKP5
   KO-modulated genes (MAPK: NES = 1.6477109, P = 0.0, FDR < 0.03797602;
   TGF-β: NES = 1.5691792, P = 0.006369427, FDR < 0.06341224; Chemokine:
   NES = 1.4961016, P = 0.0013513514, FDR < 0.124319231) (Fig. [108]4C).
   In addition, the heat map clustered a notable increase in the
   expression of numerous inflammatory cytokine- and chemokine-related
   genes (Fig. [109]4D). Immunohistochemical staining of liver tissues
   revealed a significant increase in CD68 expression in KO mice after
   CCl[4] treatment compared to WT male and female mice (Fig. [110]4E).
   MKP5 knockout resulted in a notable intensification of macrophage
   infiltration, which contributed to the aggravation of CCl[4]-induced
   hepatic inflammation. The MAPK pathway is intimately associated with
   the inflammatory process and may play a role in the regulation of liver
   inflammation during HF [[111]26]. To validate the activation of MAPK
   pathway, hepatic proteins implicated in this signalling cascade were
   analysed. Elevated levels of phosphorylated ERK, JNK, and P38 were
   observed in the livers of WT mice following CCl[4] treatment.
   Furthermore, MKP5 knockout resulted in further augmentation of the
   phosphorylation levels (Fig. [112]4F). To further investigate the
   impact of MKP5 on hepatic inflammation during the development of HF,
   the levels of inflammatory cytokines in the serum of mice were
   evaluated using enzyme-linked immunosorbent assay (ELISA). It was
   observed that the levels of IL-6 and TNF-α were significantly increased
   after CCl[4] induction in both male and female mice, and that the
   knockout of MKP5 further exacerbated the levels of inflammatory
   cytokines (Fig. [113]4G). Subsequently, the inflammation-related genes
   from RNA sequencing were further validated in mouse liver tissues using
   real-time PCR. It was confirmed that the expression of pro-inflammatory
   cytokines (IL-1β, IL-6, TNF-α, TGF-β1, and IL-18) and chemokines
   (CXCL1, CXCL11, CXCL2, CXCL10, CCL2, CCL3, and CCL4) was significantly
   increased in CCl[4]-induced KO mice compared with WT mice. Conversely,
   the expression of the anti-inflammatory cytokine (IL-10) was
   significantly decreased. Furthermore, the expression of the
   pro-fibrotic cytokine TGF-β1 was significantly increased in
   CCl[4]-induced KO mice (Fig. [114]4H), consistent with the finding in
   Fig. [115]4C. During HF, pro-fibrotic factors trigger an inflammatory
   response in HSCs and hepatocytes. As illustrated in Fig. [116]4I,
   TGF-β1 treatment markedly elevated the levels of pro-inflammatory
   cytokines in primary HSCs from WT mice, as indicated by an increase in
   the mRNA expression of IL-6, IL-1β, TNF-α, and MCP-1. Moreover, these
   genes were further increased in TGF-β1-induced primary HSCs from KO
   mice. In addition, inhibition of MKP5 expression aggravated the
   TGF-β1-induced inflammatory response in LX-2 cells, whereas MKP5
   overexpression reduced the expression of inflammatory cytokines in LX-2
   cells (Fig. [117]4J). Similar results were observed in hepatocytes. The
   transcription levels of inflammation-related genes IL-6, IL-1β, TNF-α,
   and MCP-1 in TGF-β1-induced primary hepatocytes from KO mice were
   significantly higher than those in WT mice (Fig. [118]4K). Inhibition
   of MKP5 expression exacerbated the inflammatory response induced by
   TGF-β1 in AML12 cells. Conversely, overexpression of MKP5 resulted in
   downregulation of the inflammatory genes induced by TGF-β1 in AML12
   cells (Fig. [119]4L). These results suggest that MKP5 may exert a
   regulatory effect on hepatic inflammation during HF via the MAPK
   pathway.

Fig. 4.

   [120]Fig. 4
   [121]Open in a new tab

   MKP5 knockout aggravates the hepatic inflammatory response during
   hepatic fibrogenesis. A Volcano plot illustrating the distribution of
   DEGs between WT-CCl[4] and KO-CCl[4] mouse liver tissues. B KEGG
   analysis highlighting the differential pathways in the liver tissue
   between WT-CCl[4] and KO-CCl[4]. C GSEA was performed on CCl[4]-treated
   WT and KO mice to analyse the TGF-β, MAPK, and chemokine signalling
   pathways. D Heat map illustrating the DEGs between WT-CCl[4] and
   KO-CCl[4] mouse liver tissue. E Representative staining of CD68 in
   liver tissues and statistical analyses. Scale bar: 100 μm. F Protein
   expression of the MAPK pathway in male liver tissue was measured by
   western blotting, and quantification was performed using Image J
   software. (n = 6/group). G Serum levels of IL-6 and TNF-α were
   estimated using ELISA (M: n = 4–5/group, F: n = 6–9/group). H mRNA
   levels of inflammatory factors and chemokines in CCl[4]-treated male WT
   and KO mice (n = 6/group). I–L mRNA levels of IL-6, IL-1β, TNF-α, and
   MCP-1 were detected in primary HSCs (I), LX-2 cells (J), primary
   hepatocytes (K), and AML12 cells (L) following treatment with TGF-β1
   using real-time PCR. * P < 0.05, ** P < 0.01, *** P < 0.001

MKP5 regulates hepatic fibrosis through the IRE/XBP1 pathway

   The RNA sequencing data from the liver of CCl[4]-KO/WT mice revealed
   that a pronounced activation of the protein processing in the
   endoplasmic reticulum pathway, which emerged as the most significantly
   affected pathway (Fig. [122]4B). A compiled list of protein processing
   genes in the endoplasmic reticulum signalling pathway demonstrated that
   top-ranking change genes, including the molecular chaperones Hsp40,
   Dnajb4, Dnajb11, Dnaja1, Dnajc10, and Ern1 (the coding gene of ERS
   kinase IRE1α). Ern1 was identified the second most significant gene
   among the DEGs in terms of P value (Fig. [123]5A).

Fig. 5.

   [124]Fig. 5
   [125]Open in a new tab

   MKP5 regulates hepatic fibrosis via the IRE/XBP1 pathway. A List of the
   top six DEGs between WT-CCl[4] and KO-CCl[4] mouse liver tissues. B–D
   The phosphorylation level of IRE1α was measured in liver tissue (B),
   LX-2 (C), and AML12 cells (D) using western blotting, followed by
   quantification using ImageJ software (n = 6/group). Up: MKP5
   overexpression, down: inhibition of MKP5 expression. E–G The mRNA
   levels of XBP1s were detected in liver tissue (E), LX-2 (F), and AML12
   cells (G) by real-time PCR. Left: MKP5 overexpression, right:
   inhibition of MKP5 expression. H–K LX-2 cells were pre-treated with
   Kira6 (0.5 µM) for 1 h and stimulated with the corresponding inhibitor
   and TGF-β1 for 24 h, V: Veh, K: Kira6, T: TGF-β1, K + T:
   Kira6 + TGF-β1. The levels of p-IRE1α and fibrosis-associated proteins
   were measured using western blotting (H). The mRNA levels of XBP1s (I);
   fibrosis-related genes ACTA2, Fn1, and CoL1A1 (J); and
   inflammation-associated genes IL-6, IL-1β, TNF-α, and MCP-1 (K) were
   determined using real-time PCR. (L-O) AML12 cells were pre-treated with
   Kira6 (1 µM) for 1 h and stimulated with the corresponding inhibitor
   and TGF-β1 for 24 h. The levels of p-IRE1α, apoptosis-associated
   proteins (L), and EMT-related proteins (N) were detected using western
   blotting. The mRNA levels of XBP1s (M), IL-6, IL-1β, TNF-α, and MCP-1
   (O) were measured by real-time PCR. *P < 0.05, **P < 0.01, ***P < 0.001

   The ERS plays a pivotal role in the development of HF. In the event of
   an ERS, IRE1α dissociates from GRP78, leading to activation of its
   kinase and RNase activities. Subsequently, a 26-nucleotide intron
   within XBP1 is cleaved, resulting in the production of activated
   spliced XBP1 (XBP1s) [[126]27]. Our findings revealed that the
   expression of p-IRE1α was elevated in the liver tissue of
   CCl[4]-induced WT mice and that the phosphorylation level of IRE1α was
   markedly increased following MKP5 knockout (Fig. [127]5B). Similarly,
   inhibition of MKP5 expression resulted in a further increase in p-IRE1α
   expression in TGF-β1-stimulated LX-2 and AML12 cells. Conversely,
   overexpression of MKP5 reduced the elevated expression of p-IRE1α
   induced by TGF-β1 (Fig. [128]5C, D). As a downstream factor of IRE1α,
   XBP1 was spliced in the liver tissue of WT mice after CCl[4] induction,
   and the level of XBP1s increased further after MKP5 knockout
   (Fig. [129]5E). In TGF-β1-stimulated LX-2 and AML12 cells, inhibition
   of MKP5 expression resulted in a further increase in the levels of XBP1
   splicing, whereas overexpression of MKP5 led to a reduction in the mRNA
   level of XBP1s (Fig. [130]5F, G). To further elucidate the role of
   IRE1α in HSC activation, hepatocyte apoptosis, and EMT, we pre-treated
   LX-2 and AML12 cells infected with Retro-pSIREN or Retro-shMKP5 with
   Kira6 (an inhibitor of the IRE1α kinase) for one hour before TGF-β1
   treatment. These results demonstrated that Kira6 effectively suppressed
   the phosphorylation of IRE1α and the mRNA level of XBP1s in LX-2 and
   AML12 cells (Fig. [131]5H, I,L, M, S3A, S3B). In Retro-pSIREN-infected
   LX-2 cells, TGF-β1 induced a significant increase in the expression of
   fibrosis-associated proteins, including α-SMA, collagen Iα1, and
   fibronectin. Furthermore, inhibition of MKP5 expression led to a
   further increase in the expression of the aforementioned ECM proteins,
   which is consistent with the results shown in Fig. [132]2F. It is
   noteworthy that Kira6 markedly alleviated TGF-β1-induced overexpression
   of α-SMA, collagen Iα1, and fibronectin in MKP5 knockdown LX-2 cells
   (Fig. [133]5H). Furthermore, real-time PCR results demonstrated that in
   MKP5 knockdown LX-2 cells, Kira6 markedly diminished the elevated
   expression of TGF-β1-induced fibrosis-associated genes, including
   ACTA2, CoL1A1, and Fn1, as well as inflammation-related genes such as
   IL-6, IL-1β, TNF-α, and MCP-1 (Fig. [134]5J, K). These findings
   indicated that MKP5 exerts regulatory control over TGF-β1-induced HSC
   activation and inflammatory responses via the IRE/XBP1 pathway.
   Similarly, in MKP5 knockdown AML12 cells, Kira6 was observed to
   markedly reduce the elevated expression of TGF-β1-induced apoptotic
   proteins, including cleaved Caspase3, cleaved Caspase8, and cleaved
   Caspase9 (Fig. [135]5L, S3B), as well as EMT-related proteins such as
   N-cadherin and Vimentin (Fig. [136]5N, S3C). Furthermore, the
   expression of inflammatory mediators, including IL-6, IL-1β, TNF-α, and
   MCP-1, were also diminished by Kira6 (Fig. [137]5O). Our data
   demonstrated that MKP5 exerts regulatory effects on HSC activation,
   hepatocyte apoptosis, and EMT through the IRE/XBP1 pathway, which
   ultimately attenuates HF progression.

Preparation and physicochemical characterisation of nano-formulation

   The use of cell membrane coatings in biomimetic designs has become
   increasingly prevalent in recent years. In this study, MSCM was
   obtained from MSCs. Subsequently, a gene-carrying nanoparticles
   (PLGA-MKP5) coated with MSCM was prepared (Fig. [138]6A). Agarose gel
   electrophoresis was performed to determine the optimal ratio of PLGA to
   MKP5. The results demonstrated that, at a mass ratio of 2:1 of PLGA to
   MKP5, complete complexes were formed with no observed band tailing
   (Fig. [139]6B). Transmission electron microscopy revealed that the MSCM
   coating was effectively applied to the surface of the PLGA-MKP5
   nanoparticles, exhibiting a spherical core–shell structure comparable
   to that observed in its non-targeted counterparts (Fig. [140]6C).
   Furthermore, the protein structure remained intact, resembling that of
   the native MSCM proteins (Fig. [141]6D, E). Dynamic light scattering
   (DLS) analysis demonstrated that the particle size of MSCM-PLGA-MKP5
   (131 nm) was larger than that of PLGA-MKP5 (119 nm), with a surface
   charge of -3.08 mV (Fig. [142]6F). Subsequently, the biosafety of
   biomimetic nano-formulation were evaluated in LX-2 and AML12 cells. As
   illustrated in Fig. [143]6G, there was no significant change in cell
   viability for both LX-2 and AML12 cells with varying concentrations of
   nano-formulation compared to the untreated group. This suggests that
   the nano-formulation did not induce any cytotoxic effects on cells. The
   cellular uptake of the nanoparticles was further evaluated using laser
   confocal microscopy, which demonstrated a markedly elevated red
   fluorescence intensity of MSCM-PLGA-MKP5 compared to PLGA-MKP5 in both
   cell types (Fig. [144]6H). Furthermore, the protein expression
   (Fig. [145]6I,J) and mRNA levels (Fig. [146]6K) of MKP5 were
   significantly higher in the MSCM-PLGA-MKP5 and PLGA-MKP5 groups than in
   the phosphate buffered saline (PBS) group in the LX-2 and AML12 cells.
   The expression of MKP5 was significantly higher in the MSCM-PLGA-MKP5
   group than in the PLGA-MKP5 group. These findings suggest that a viable
   method for MKP5 delivery has been established.

Fig. 6.

   [147]Fig. 6
   [148]Open in a new tab

   Preparation and physicochemical characterisation of nano-formulation. A
   Schematic showing preparation of the nano-formulation. B The binding
   efficiency of PLGA to MKP5 was measured. M: Marker, 1: PLGA, 2: MKP5,
   3–11: MKP5: PLGA = 10:1, 5:1, 2.5:1, 2:1, 1:1, 1:2, 1:2.5, 1:5, 1:10. C
   The morphology of PLGA-MKP5, MSCM-PLGA-MKP5, and MSCM were observed by
   TEM. Scale bar: 100 nm. D The protein bands of MSCM-PLGA-MKP5 and MSCM
   were detected using SDS-PAGE. E Evaluation of CD45 and CD90 expression
   on MSCM-PLGA-MKP5 by flow cytometry. CD45, a negative marker of MSC;
   CD90, a positive marker of MSC. F Particle size and potential of
   PLGA-MKP5, and MSCM-PLGA-MKP5. G Viability of LX-2 and AML12 cells
   treated with MSCM-PLGA-MKP5 was assessed using the CCK-8 assay. H
   Cellular uptake of PLGA-MKP5 and MSCM-PLGA-MKP5 in LX-2 and AML12 cells
   was assessed by confocal microscopy. Scale bar: 50 μm. I–K The
   expression levels of MKP5 in LX-2 cells and AML 12 cells were measured
   using western blotting (I, J) and real-time PCR (K). *P < 0.05,
   **P < 0.01

In vivo toxicity and biodistribution of nano-formulation

   It is imperative that the biosafety of nano-formulation be evaluated
   before assessing their efficacy. The in vivo toxicity and
   biodistribution of the nano-formulation were evaluated. As illustrated
   in Fig. [149]7A, mice that received tail vein injections of the
   nanoparticles demonstrated a consistent increase in weight over time
   compared to the PBS group, with no statistically significant difference
   in body weight among all groups. Similarly, no significant differences
   were observed in liver weight or liver weight/body weight among the
   groups (Fig. [150]7B). Furthermore, histological examination using HE
   staining demonstrated no discernible changes in the morphology of the
   five primary organs (heart, liver, spleen, lung, and kidney tissues)
   (Fig. [151]7C). The functionality of the liver and kidneys of mice
   belonging to all groups was evaluated, and no discernible liver or
   kidney injury was observed among the groups (Fig. [152]7D). These
   findings suggest that the bionic nano-formulation did not induce
   systemic toxicity at the administered dose, confirming their favourable
   biosafety profile. Furthermore, to gain additional insight into the in
   vivo targeting effect of the nano-formulation, Cy5.5-labelled PLGA-MKP5
   and MSCM-PLGA-MKP5 were prepared and intravenously injected into mice
   with HF. The major organs were imaged ex vivo by using an in vivo IVIS
   imaging system. Twelve hours after injection, accumulation of
   Cy5.5-labelled PLGA-MKP5 and MSCM-PLGA-MKP5 were observed in the
   injured liver. The fluorescence intensity was greater in both the
   MSCM-PLGA-MKP5 and PLGA-MKP5 groups than in the PBS control group
   (Fig. [153]7E). Notably, the fluorescence intensity in the
   MSCM-PLGA-MKP5 group was significantly higher than that in the
   PLGA-MKP5 group, with statistical analysis indicating approximately
   twice the strength of PLGA-MKP5 (Fig. [154]7E). The fluorescence
   intensities of isolated tissues, including the heart, liver, spleen,
   lungs, and kidneys, were evaluated. In accordance with the findings of
   the in vivo imaging, a markedly elevated fluorescence signal was
   observed in the liver compared to other organs (Fig. [155]7F),
   suggesting a greater propensity for nano-formulation accumulation in
   the livers of HF mice. Moreover, the MSCM-modified nano-formulation
   displayed favourable biocompatibility and targeting capabilities.

Fig. 7.

   [156]Fig. 7
   [157]Open in a new tab

   In vivo toxicity and biodistribution of the nano-formulation. A Body
   weight was monitored over a 15-day period following intravenous
   treatment with PBS, MKP5, PLGA-MKP5, and MSCM-PLGA-MKP5 (n = 6/group).
   B Liver weight and the ratio of liver weight to body weight were
   measured in each group of mice (n = 6/group). C Major organs were
   collected on day 15 and assessed by HE staining. Scale bar: 100 μm
   (n = 3/group). D Liver and kidney function, including ALT, AST, BUN,
   UA, and CRE levels, were determined on day 15 (n = 4–5/group). E
   Biodistribution of PLGA-MKP5 and MSCM-PLGA-MKP5 were detected using an
   In Vivo Imaging System. F Biodistribution of isolated organs after
   PLGA-MKP5 and MSCM-PLGA-MKP5 administration was detected using an In
   Vivo Optical System. *P < 0.05, **P < 0.01, ***P < 0.001

Nano-formulation regulate CCl[4]-induced hepatic fibrosis through the
IRE/XBP1 pathway

   An HF model was constructed and treated with MSCM-PLGA-MKP5 to evaluate
   the anti-fibrotic effect of the nano-formulation (Fig. [158]8A). As
   shown in Fig. S5A, S5B, administration of PLGA-MKP5 and MSCM-PLGA-MKP5
   treatment significantly upregulated the protein and mRNA expression of
   MKP5 compared to the CCl[4]-PBS group. Moreover, MSCM-PLGA-MKP5
   exhibited a more pronounced evlevation in MKP5 expression, indicating a
   greater propensity for nano-formulation accumulation in the livers of
   HF mice. The liver weight and liver weight/body weight ratio in the
   CCl[4]-PBS group were significantly higher than those in the control
   group. Conversely, down-regulation trends were observed in the
   MSCM-PLGA-MKP5 and PLGA-MKP5 groups compared with the CCl[4]-PBS group.
   The results of body weight analysis indicated no significant
   differences among the groups (Fig. [159]8B). Furthermore, the liver
   function of the mice was examined, and it was observed that the levels
   of ALT and AST in the serum and Hyp in the liver tissues gradually
   decreased in both the CCl[4]-PLGA-MKP5 and CCl[4]-MSCM-PLGA-MKP5 groups
   compared to those in the CCl[4]-PBS group. The MSCM-PLGA-MKP5 group
   exhibited the most pronounced effects (Fig. [160]8C). Moreover,
   histological examination revealed structural damage to the liver
   tissue, including disruption of the hepatic lobule, deviation of the
   central vein, and vacuolation of the hepatocytes, in the CCl[4]-PBS
   group compared to the PBS group. Masson's trichrome and Sirus Red
   staining revealed markedly elevated collagen deposition in the
   CCl[4]-PBS group compared with the PBS group. However, these
   abnormalities were reversed in CCl[4]-treated PLGA-MKP5 mice and were
   even more pronounced in CCl[4]-MSCM-PLGA-MKP5 mice (Fig. [161]8D, S4A).
   Compared to the control group, the expression of fibrosis-related
   proteins α-SMA and fibronectin (Fig. [162]8E, S4B) and the
   transcriptional levels of fibrosis-related genes ACTA2, CoL1A1, and Fn1
   (Fig. [163]8F) were markedly increased in the CCl[4]-PBS group.
   PLGA-MKP5 and MSCM-PLGA-MKP5 demonstrated a reduction in the levels of
   fibrosis-related proteins and mRNA induced by CCl[4], with a more
   pronounced effect observed in the MSCM-PLGA-MKP5 group. In addition,
   PLGA-MKP5 markedly diminished the transcription levels of
   CCl[4]-induced inflammatory cytokines, including IL-6, IL-1β, TNF-α,
   TGF-β1, and MCP-1. Moreover, MSCM-PLGA-MKP5 further significantly
   reduced the mRNA levels of these inflammatory cytokines and chemokines
   compared to PLGA-MKP5, thereby alleviating the hepatic inflammatory
   response during HF (Fig. [164]8G). These findings indicated that
   PLGA-MKP5 and MSCM-PLGA-MKP5 mitigated CCl[4]-induced HF, with the
   effect of MSCM-PLGA-MKP5 being more pronounced. Subsequently, the
   expression of p-IRE1α and XBP1s in the liver of mice was examined,
   revealing that MSCM-PLGA-MKP5 significantly reduced the phosphorylation
   level of IRE1α and the mRNA level of XBP1s (Fig. [165]8H, I). This
   observation was consistent with the results of previous studies
   (Fig. [166]5C–G). These results demonstrate that MSCM-PLGA-MKP5 exerts
   a suppressive effect on CCl[4]-induced HF via the IRE/XBP1 pathway.

Fig. 8.

   [167]Fig. 8
   [168]Open in a new tab

   Nano-formulation inhibits CCl[4]-induced hepatic fibrosis through the
   IRE/XBP1 pathway. A Schematic illustrating the procedure of the animal
   experiments. B Assessment of body weight, liver weight, and the liver
   weight/body weight ratio in mice (n = 5–7/group). C The serum levels of
   AST, ALT, and Hyp levels in the liver tissue of mice were evaluated
   (n = 4–6/group). D Morphology and collagen deposition in liver tissue
   were examined using HE, Masson’s, and Sirus red staining. Scale bar:
   100 μm. E Protein levels of fibronectin and α-SMA in the liver tissue
   were analysed by western blotting (n = 6/group). F, G The mRNA levels
   of fibrosis-associated genes ACTA2, CoL1A1, and Fn1 (F), as well as
   inflammation-related genes IL-6, IL-1β, TNF-α, TGF-β1, and MCP-1 (G)
   were analysed by real-time PCR (n = 4/group). H The protein expression
   of p-IRE1α was determined by western blotting, and quantification was
   performed using the ImageJ software. (n = 6/group). I The mRNA level of
   XBP1s was estimated by real-time PCR (n = 4/group). *P < 0.05,
   **P < 0.01, ***P < 0.001

Discussion

   HF is a reversible pathological process characterised by the excessive
   accumulation of ECM and a persistent inflammatory response [[169]1].
   Various cell types are involved in the pathogenesis of fibrosis
   [[170]28]. HSCs are regarded as the principal effector cells in HF and
   are the primary source of pro-fibrotic cytokines and ECM [[171]1,
   [172]29]. Under physiological conditions, HSCs are located in the Disse
   space and contain vincristine lipid droplets, thus exhibiting a
   quiescent phenotype [[173]30]. In their quiescent state, HSCs perform
   the functions of pericytes and serve as primary storage sites for
   vitamin A [[174]31]. In the event of liver damage, HSCs undergo
   activation and transition from a quiescent state to an activated state.
   HSC activation is initiated by the release of DAMPs from the
   hepatocytes [[175]32]. Hepatocytes constitute approximately 80% of
   cells in the liver and are responsible for the primary functions of
   this organ. They represent the primary targets of several
   hepatotoxicants, alcohol metabolites, hepatitis viruses, and bile acids
   [[176]2]. A study has indicated that in adult mice, hepatocytes undergo
   EMT in response to stimulation by pro-fibrotic growth factor TGF-β,
   which provides direct evidence for the involvement of hepatocytes in HF
   formation [[177]33]. Hepatocyte-mediated apoptotic vesicles stimulate
   Kupffer cells (KCs) to secrete fibrotic cytokines and promote HSC
   activation through the interaction of toll-like receptors (TLR) with
   DNA [[178]34]. It has recently been demonstrated that MKP5 plays a role
   in the regulation of pulmonary and cardiac fibrosis [[179]15, [180]16].
   In this study, MKP5 knockout resulted in a worsening of CCl[4]-induced
   HF, accompanied by an increase in the expression of ECM proteins.
   Conversely, overexpression of MKP5 inhibited HSC activation and
   hepatocyte apoptosis, as indicated by TGF-β1. This suggests that MKP5
   plays a crucial role in the pathogenesis of HF. One study has
   demonstrated that inflammation is a crucial factor in the development
   of fibrosis. Furthermore, the presence of damaged hepatocytes alone may
   not be enough to directly activate HSCs and cause fibrosis [[181]3]. In
   instances of liver damage, inflammatory cytokines such as TGF-β, TNF-α,
   IL-1β, and PDGF induce HSC activation, prompting a transition from a
   quiescent state to myofibroblast-like cells [[182]5]. Activated HSCs
   subsequently migrate to the site of injury and secrete ECM, ultimately
   resulting in the formation of fibrous scar tissue [[183]35, [184]36].
   MKP5, a phosphatase of the MAPK pathway that is closely related to
   inflammation, has been demonstrated in previous studies to regulate the
   inflammation of adipocytes and islets [[185]12, [186]13]. Moreover,
   another study demonstrated that the expression of a range of
   pro-inflammatory cytokines was elevated in MKP5-deficient mice due to
   T-cell activation [[187]37]. In accordance with the findings, our
   results indicate that the absence of MKP5 exacerbates CCl[4]-induced
   hepatic inflammation by upregulating the expression of inflammatory
   cytokines and chemokines in the liver tissue, HSCs, and hepatocytes.
   This study demonstrated the role of MKP5 in the amelioration of hepatic
   inflammation and fibrosis. Nevertheless, the regulatory function of
   MKP5 in the interaction between hepatocytes and HSCs in HF remains
   unclear and requires further investigation.

   ERS is intimately linked to a plethora of ailments and acts as a nexus
   for autophagy, inflammation, and apoptosis signalling pathways, which
   actively facilitate the advancement of HF [[188]38]. It has been well
   established that ERS is activated in a CCl[4]-induced HF model [[189]6,
   [190]39, [191]40]. Following the onset of ERS, the number of unfolded
   proteins that bind to GRP78 increases. This results in the separation
   of IRE1α, PERK, and ATF6 transmembrane proteins, which are typically
   inactivated by GRP78 binding. This process initiates an UPR, which
   plays a pivotal role in the pathogenesis of hepatic inflammation and
   fibrosis [[192]41]. UPR signalling has been observed in HSCs in
   response to both liver injury in vivo and fibrosis stimulation in vitro
   [[193]42, [194]43]. Moreover, one study demonstrated that ROS
   generation of reactive oxygen species disrupts ER homeostasis in HSCs,
   triggering the UPR and ultimately leading to HSC activation and
   fibrosis [[195]44]. It has been demonstrated that ERS plays a
   significant role in the pathogenesis of metabolic liver disease and HF
   [[196]45]. Furthermore, ERS has been shown to promote HSC activation in
   a CCl[4]-induced HF model [[197]46]. Given the abundance of hepatocytes
   in the ER, which is essential for their physiological function,
   ERS-induced hepatocyte death contributes to the progression of liver
   injury and HF [[198]47]. When ERS persists in cells, three activated
   transmembrane proteins on the ER membrane trigger apoptosis.
   ERS-mediated hepatocyte apoptosis represents a pivotal step in the
   activation of HSCs [[199]48, [200]49]. This study demonstrates that ERS
   activation is responsible for HSCs activation and hepatocyte apoptosis
   and that this ERS activation is further intensified by MKP5 knockout.
   As one of the three major pathways of ERS, IRE/XBP1 has been reported
   to be involved in regulating the development of HF [[201]8, [202]50].
   Some studies have demonstrated that the IRE1α/XBP1 pathway is activated
   during TGF-β1-induced HSC activation, which promotes fibrosis
   development [[203]9, [204]51]. Furthermore, 4μ8C (a noncompetitive
   inhibitor that blocks IRE1α kinase and ribonucleic acid endonuclease
   signalling) has been shown to reduce TGF-β-induced expression of sXBP1,
   Bip, α-SMA, Collagen Iα1, and CTGF [[205]42, [206]43, [207]52]. Our
   findings support the hypothesis that the IRE/XBP1 pathway is activated
   during TGF-β1-induced HSCs activation. The same results were observed
   for TGF-β1-induced hepatocyte apoptosis. Recently, a study has
   demonstrated that ERS promotes endothelial cell injury through the
   Nox4/MKP3 interaction [[208]53]. Importantly, several studies have
   suggested that MKP5 plays a regulatory role in diabetes and apoptosis
   of epithelial cells, as well as dysfunction of tight junction through
   the ERS pathway [[209]14, [210]54]. Our findings demonstrate that MKP5
   exerts an inhibitory effect on HSCs activation, hepatocyte apoptosis,
   and inflammation through regulation of the IRE1α/XBP1 pathway,
   attenuating the progression of HF. These findings suggest that
   targeting MKP5 is a pivotal strategy for the development of
   anti-fibrotic therapies.

   The current treatment for HF primarily encompasses surgical and
   pharmacological interventions, with the latter including antiviral
   therapy, interferon, malotilate, lamivudine, vitamin E, silymarin, and
   polyene phosphatidylcholine drugs [[211]2, [212]55]. However,
   anti-liver fibrotic medications may elicit a range of adverse effects,
   such as headaches, nausea, vomiting, and other digestive discomforts.
   Prolonged use of the same medication may lead to the emergence of drug
   resistance which diminishes therapeutic efficacy and increases the risk
   of disease recurrence and progression [[213]56]. Nucleic acid therapy
   has recently become a topic of interest in the scientific community. In
   contrast to conventional drug therapies that typically target proteins,
   gene-based nucleic acid therapies can induce therapeutic effects by
   regulating gene expression and targeting specific cells in the target
   tissue [[214]57]. Gene therapy currently enables precise targeting of
   specific genes or signaling pathways in the HF process, thus achieving
   more accurate drug delivery and minimizing toxic side effects.
   Nevertheless, nucleic acids have been employed as therapeutic agents is
   challenging because of their susceptibility to degradation by
   nucleases, propensity to elicit immune activation, and unfavourable
   physicochemical properties that impede cellular uptake [[215]57].
   Nano-delivery systems are novel nucleic acid drug delivery systems that
   employ NPs as drug carriers to facilitate precise delivery of drugs to
   the targeted lesion site via electrostatic interactions [[216]58].
   PLGA, a synthetic polymer with good biocompatibility and
   biodegradability, is widely employed as a drug delivery carrier for
   transporting proteins, peptides, and viral DNA [[217]59]. Cellular
   uptake results demonstrated that PLGA-MKP5 effectively transported MKP5
   to the liver, facilitating its activity. NPs function as efficacious
   delivery systems that protect nucleic acid drugs from degradation,
   enhance drug stability, and prolong their half-lives. However, the
   passive targeting effect of NPs is inadequate [[218]17], and
   non-specific cellular uptake also significantly reduces the efficiency
   of nano-delivery systems. The liver contains diverse cell types such as
   resident KCs that produce strong non-specific uptake of NPs,
   accelerating endocytosis and clearance of NPs, which ultimately results
   in lower doses reaching the target cells [[219]60]. Coating synthetic
   NPs with biomimetic cell membrane materials effectively preserves the
   proteins and polysaccharides present on the cell membrane surface,
   imparting NPs with surface properties similar to natural cells. This
   aids in evading recognition by the immune system and subsequent
   clearance, leading to prolonged circulation in the body and increased
   accumulation of drugs or therapeutic agents at the target sites. The
   membranes of red blood cells, platelets, stem cells, and macrophages
   have been identified as effective natural carriers [[220]61]. MSCM, an
   important component of stem cells, inherits various biological
   properties such as cell recognition, low immunogenicity, signal
   transduction and homing ability. These properties enable the NPs
   modified by MSCM to accurately identify target cells in drug delivery
   processes for versatile applications in tumour therapy, inflammatory
   diseases, tissue regeneration and other fields [[221]22, [222]61,
   [223]62]. This study describes the development of an MSCM-coated
   nano-formulation that does not cause systemic toxicity. The
   accumulation of the MSCM-PLGA-MKP5 in CCl[4]-induced liver tissues
   increased the local therapeutic concentration of MKP5. Moreover, the
   MSCM-PLGA-MKP5 demonstrated the ability inhibited the expression of
   IRE/XBP1 pathway indicators in fibrotic liver tissues, enhancing the
   anti-hepatic fibrosis effect and preventing the development of HF. The
   in vivo results demonstrated that MSCM-PLGA-MKP5 enhanced the
   therapeutic effect of MKP5 on HF and affirmed its clinical potential in
   the treatment of HF. However, the passive targeting effect of
   MSCM-coated nano-formulation remains inadequate, necessitating further
   research to improve the targeting specificity and optimize therapeutic
   outcomes. While this study addressed some challenges of
   nano-formulation development for translation purposes, limitations such
   as uncertainty regarding long-term effects remain. Although the cell
   membrane-coated NPs were able to escape the initial clearance by the
   immune system, their long-term stability and safety have not been fully
   validated. It is essential for future studies to rigorously assess
   whether the prolong administration of MSCM-coated NPs may lead to any
   adverse effects on organisms and conduct additional research on
   biosafety and long-term efficacy.

   In conclusion, the results of this study demonstrated that MKP5
   functions as an anti-fibrotic factor in HF, inhibiting the activation
   of HSCs and preventing hepatocyte apoptosis. A systemic knockout of
   MKP5 exacerbates hepatic inflammation and HF in mice. Moreover,
   administration of the MSCM-coated gene-carrying bionanomimetic NPs
   PLGA-MKP5 to damaged liver tissues via the nano-delivery system offers
   protection against CCl[4]-induced HF. Our findings indicate that MKP5
   may be a promising target to treat HF. The use of MSCM-PLGA-MKP5
   represents a novel strategy for effective anti-fibrotic therapy.

Conclusion

   The current study reveals that the biomimetic MSCM-PLGA-MKP5
   ameliorates CCl[4]-induced hepatic inflammation and fibrosis. MKP5
   suppresses HSC activation and hepatocyte apoptosis during HF through
   modulating the IRE/XBP1 pathway. In conclusion, our findings
   demonstrate that MKP5 may play a crucial role in regulating HF, and
   that targeting MKP5 could potentially serve as a therapeutic strategy
   for managing hepatic inflammation and HF.

Materials and methods

Animals

   MKP5 knockout (KO) mice were generated by Cyagen Biosciences using
   CRISPR/Cas9. Briefly, the targeting gDNA-MKP5 exon 2 vector was
   transferred into the uterus of pseudopregnant C57BL/6J mice via
   high-flux electroporation of fertilised eggs for KO mouse production.
   Genotyping was performed by PCR with the following Table [224]1
   primers.

Table 1.

   Primers for mouse genotyping
   Primer sequence (5′–3′)
   F1:5′-GACAGGAGGTCTCTTACCGTTTC-3′
   F2:5′-AGGCTCTGTGTTAAGTCCATCG-3′
   R1:5′-GTTAGAAGAAGGCTGCCACTTG-3′
   [225]Open in a new tab

   Male C57BL/6J WT mice aged 6–8 weeks were purchased from Beijing Vital
   River Laboratory Animal Technology Company (Beijing, CN).

Mouse models of CCl[4]-induced hepatic fibrosis

   All 8–9 week-old WT and KO mice were randomly allocated to four groups:
   WT: Male: n = 7–8, Female: n = 9–10; KO: Male: n = 6–7, Female:
   n = 9–10. CCl[4] (Macklin, CN) was dissolved in corn oil (Macklin, CN)
   at a ratio of 1:9 and intraperitoneally (i. p.) administered to mice at
   a dose of 1 mL/kg body weight twice per week for 8 weeks, control mice
   were injected with equal doses of corn oil.

Preparation and physicochemical characterisation of MSCM-PLGA-MKP5

   One millilitre of dichloromethane containing 20 mg of PLGA-PEI
   nanoparticles (Xi’an Qiyue Biotechnology Company, CN) was used as the
   oil phase and added dropwise to 10 mL of ultrapure water for
   emulsification for 4 h. The internal aqueous and oil phases were
   combined by ultrasonic treatment for 40 s to prepare the primary
   lotion. The external emulsion was added to 3 mL of an external aqueous
   phase containing 2% (w/v) sodium cholate and sonicated in an ice bath
   for 2 min to form a double emulsion suspension. Dichloromethane was
   evaporated for 4 h at room temperature, followed by mixing of MKP5 and
   PLGA-PEI at different mass ratio, and incubation for 30 min.

   MSCs were cultured in α-MEM (Gibco, USA) supplemented with 10% foetal
   bovine serum (FBS) (Gibco, USA), 1% (v/v) penicillin–streptomycin
   solution (Biosharp, CN), and glutamine (HyCyte, CN) at 37 °C in a 5%
   CO[2] incubator in the following experiments. To extract the MSC
   membrane (MSCM), 1 × 10^8 MSCs were suspended in 2 mL of PBS and 1%
   HALT protease inhibitor cocktail (Thermo Fisher Scientific, USA). Cells
   were enucleated using a dispersant, followed by ultracentrifugation
   (28,000 rpm, 30 min, 4 °C) with a discontinuous sucrose density
   gradient dissolved in HEPES buffer (Thermo Fisher Scientific, USA) at
   concentrations of 55%, 40%, and 30% (w/w). After collection, MSCM was
   quantified using the BCA method. Subsequently, MSCM vesicles were
   formed by passing through a single pass of 0.2 μm avanti polar lipids
   (MERCK, USA) to extrude polycarbonate porous membranes. Finally,
   MSCM-PLGA-MKP5 was prepared by mixing 1 mg of PLGA-MKP5 with 0.2 mg
   MSCM, and passing the mixture through a filter membrane nine times.
   Evaluation of CD45 and CD90 expression on MSCM-PLGA-MKP5 by flow
   cytometry. The size and zeta potential of PLGA-MKP5 and MSCM-PLGA-MKP5
   were measured using a Zetasizer Nano-ZS instrument (Malvern, UK). The
   morphologies of PLGA-MKP5, MSCM and MSCM-PLGA-MKP5 were visualised
   using TEM (Tokyo, Japan).

In vivo toxicity of nano-formulation

   All WT mice were randomly assigned to four groups (n = 6/group) and
   received intravenous injections of PBS, MKP5, PLGA-MKP5, and
   MSCM-PLGA-MKP5. On the 15th day, mice were euthanised. The
   histopathology of the major organs (heart, liver, spleen, lungs, and
   kidneys) was assessed by HE staining. Liver and renal functions were
   also measured, including ALT, AST, BUN, UA and CRE levels, which were
   purchased from the Nanjing Jiancheng Bioengineering Institute (CN).

Biological distribution of nano-formulation

   WT mice were intraperitoneally (i.p.) administered 10% CCl[4] (1 mL/kg
   body weight) twice a week for 8 weeks (n = 4/group). In addition, mice
   were intravenously injected with PLGA-MKP5 or MSCM-PLGA-MKP5. After
   12 h post-injection, the biodistribution of the Cy5.5-labelled
   PLGA-MKP5 and MSCM-PLGA-MKP5 were analysed using an IVIS spectrum
   system (PerkinElmer, USA) (748 nm/780 nm).

In vivo efficacy of nano-formulation

   After intraperitoneal injection of CCl[4] for six weeks, CCl[4]-treated
   WT mice were randomly divided into the following groups: CCl[4]-PBS
   (n = 5), CCl[4] + PLGA-MKP5 (n = 6), and CCl[4] + MSCM-PLGA-MKP5
   (n = 7). The mice received intravenous injections of PBS or PLGA-MKP5
   or MSCM-PLGA-MKP5 every other day, and continuing for 2 weeks.

Ethics statement

   All mice were housed in specific pathogen-free, humidity- and
   temperature-controlled microisolator cages with a 12-h light/dark cycle
   at Jilin University. All animal experiments were conducted in
   accordance with the Guidelines for Experimental Animals of Jilin
   University and approved by the Animal Experiment Ethics Committee of
   Jilin University (approval number: 20240038). When the study ended, the
   mice were euthanised by carbon dioxide (CO[2]) asphyxiation.

Blood biochemical assay

   Mouse serum was obtained by centrifuging at 3000 rpm for 15 min. The
   levels of serum ALT, AST, MDA, and GSH, as well as the Hyp content in
   liver tissues, were measured using commercially available detection
   kits from the Nanjing Jiancheng Bioengineering Institute (CN),
   according to the manufacturer's instructions.

Enzyme-linked immunosorbent assay (ELISA)

   The levels of IL-6 and TNF-α in the serum were quantified using an
   ELISA kit (Proteintech, CN) following the manufacturer’s instructions.
   All assays were performed according to the manufacturer’s instructions.

Hematoxylin–eosin (HE), Masson and Sirus red staining

   We commissioned Wuhan Servicebio Technology to perform paraffin
   embedding and sectioning of the liver tissue. Liver paraffin sections
   were stained with HE, Masson’s trichrome (Solarbio, CN), and Sirus Red
   (Biotopped, CN) according to the manufacturer’s instructions. To
   analyse hepatic collagen distribution, fibrotic septa randomly selected
   from the right and left liver lobes of six individual mice per group
   were assessed. Images were captured using Image-Pro Plus software
   (version 6.0; Media Cybernetics, USA). The extent of collagen staining
   was quantified as a percentage of the stained area in each liver
   section.

Immunohistochemistry

   Liver paraffin sections were de-paraffinised and rehydrated for
   immunohistochemical staining for CD68 (Abclonal, CN). Briefly, paraffin
   sections were incubated with 3% H[2]O[2] for 30 min to remove
   endogenous peroxidases. Subsequently, the slides were subjected to
   antigen retrieval in a preheated antigen repair solution (Beyotime, CN)
   for 30 min. After blocking with 5% BSA, the slides were incubated
   overnight at 4 °C with anti-CD68 antibody (Proteintech, CN), followed
   by incubation with HRP-conjugated secondary antibodies (CST, USA).
   Finally, the liver tissue sections were visualised using a DAB
   horseradish peroxidase chromogenic kit (ZSGB-BIO, CN) and
   counterstained with haematoxylin. Image analysis software Image-Pro
   Plus 6.0 was used to quantify positively stained areas.

Immunofluorescence

   Frozen tissues were fixed in cold acetone for 10 min, followed by
   30-min of incubation with 0.5% Triton X-100 to permeate the cell
   membrane. Subsequently, the tissues were blocked with 2% goat serum
   (Beyotime, CN) and then incubated with anti-MKP5 (Abclonal, CN),
   anti-α-SMA (Santa Cruz, USA), anti-albumin (Affinity, USA), and Ki67
   (BIOSS, USA) antibodies overnight at 4 °C, followed by incubation with
   fluorescence-conjugated secondary antibodies for 1 h. DAPI (Beyotime,
   CN) was used to stain cell nuclei. Fluorescent images were captured
   using a laser scanning confocal microscope (Nikon, Japan) and analysed
   using Image-Pro Plus software (version 6.0) to measure the areas of the
   positively stained sites.

Isolation of primary hepatocytes and HSCs

   Primary hepatocytes were isolated from male mice using a modified
   two-step collagenase perfusion protocol, as previously described
   [[226]63]. Briefly, the mice were anaesthetised, and a V-shaped
   incision was made to expose the internal organs. The livers were washed
   with D-Hank’s (Biosharp, CN) balanced salt solution containing EDTA
   (0.5 mM) and perfused with collagenase IV (Sigma-Aldrich, USA) in
   D-Hank’s balanced salt solution. Following perfusion, the livers were
   removed from the mice and the digested hepatocytes were dispersed in
   Dulbecco’s modified Eagle’s medium (DMEM)-free medium (Corning, USA)
   and filtered through 100 mesh sieves to remove any undigested debris.
   The filtrate was centrifuged at 50 ×g for 5 min at 4 °C. The
   hepatocytes in the precipitate were washed three times with a DMEM-free
   medium and harvested for subsequent analysis.

   Primary HSCs were isolated, as described above. Following perfusion,
   digestion, and centrifugation, liver parenchyma cells were removed, and
   the remaining non-parenchymal cells (NPC) were collected. The NPC was
   then re-resuspended in 25% Percoll (GE Healthcare Biosciences AB, USA)
   and layered between the bottom 75% Percoll buffer and top PBS. After
   centrifugation at 1 400 ×g for 20 min at 4 °C, the HSCs fraction was
   obtained at the interface between the top and intermediate layers. Cell
   viability was assessed using trypan blue staining.

Retrovirus packaging

   The MKP5 gene fragment was cloned into the pLNCX2 vector by gene
   cloning to construct a recombinant retroviral vector containing the
   target gene fragment. Briefly, the MKP5 gene with the NotI/Bgl2 site
   was inserted into the NotI/Bgl2 cloning site of the pLNCX2 retroviral
   vector via NotI/Bgl2 digestion to obtain the recombinant retroviral
   vector pLNCX2-MKP5. Recombinant MKP5 overexpression retroviruses were
   generated by co-transfecting pLNCX2-MKP5, pVSV-G envelope vector, and
   p-Gag-pol packaging vector into HEK-293 T cells (HyCyte, CN) using
   Lipofectamine 3000 (Thermo Fisher Scientific, USA). Transfected
   HEK-293 T cells were incubated for 48 h before viral collection.
   Subsequently, the virus was concentrated, and its titre was determined
   for further experiments. The restriction site that interfered with the
   expression of MKP5 was BamHI/EcoRI, and the vector was pSIREN, which
   was in accordance with those for MKP5 overexpression, as described
   above. The retrovirus overexpressing of MKP5 was designated as
   Retro-PLNCX2/Retro-MKP5, and the retrovirus suppressing MKP5 expression
   was denoted as Retro-pSIREN/Retro-shMKP5.

Cell lines and treatments

   LX-2 (HyCyte, CN), AML12 (HyCyte, CN), and HEK-293 T (HyCyte, CN) cells
   were cultured in DMEM supplemented with 10% FBS (Gibco, USA), 100 U/mL
   penicillin, and 0.1 mg/mL streptomycin (Gibco, USA) at 37 °C with 5%
   CO[2]. LX-2 and AML12 cells were infected with the virus
   (Retro-PLNCX2/Retro-MKP5 or Retro-pSIREN/Retro-shMKP5) for 24 h,
   followed by stimulation with 10 ng/mL TGF-β1 (Sino Biological, CN) for
   an additional 24 h. For experiments involving IRE inhibition, LX-2 and
   AML12 cells infected with the virus were preincubated with Kira6 (MCE,
   USA) for 1 h before stimulation with TGF-β1 plus the aforementioned
   inhibitors for another 24 h. For experiments involving cellular uptake,
   LX-2 and AML12 cells were treated with PLGA-MKP5 or MSC-PLGA-MKP5 for
   24 h and photographed using a fluorescence microscope (Nikon, Japan).

Cell viability and proliferation assay

   Cell viability and proliferation were assessed using Cell Counting
   Kit-8 (Sigma, USA) according to the manufacturer’s instructions.
   Briefly, cells were seeded in 96-well plates at a density of 5 × 10^3
   cells/well. For the cell proliferation experiments, LX-2 cells were
   infected with the virus for 24 h and then stimulated with TGF-β1
   (10 ng/mL) for an additional 24 h. For the cell viability experiments,
   LX-2 and AML12 cells were exposed to PLGA (25, 50, 75, and 100 μg/mL)
   for 24 h. Subsequently, the cells were incubated with CCK-8 solution
   (Promega, USA) at 37 °C for 1 h before measuring the absorbance of each
   well at 490 nm.

Transwell invasion and scratch-wound healing assay

   After infecting LX-2 cells with Retro-MKP5 or Retro-shMKP5 for 24 h,
   the cells were stimulated with 10 ng/mL TGF-β1 for an additional 24 h.
   For the Transwell invasion assay, the treated cells were digested and
   cultured in a petri dish with a pore size of 8 μm for 24 h, fixed,
   stained, and photographed using a light microscope. For the
   scratch-wound healing assay, the cells were scraped using a sterile
   20 μL pipette head, and the cell fragments were washed with PBS.
   Subsequently, the cells were cultured for 24 h. Finally, scratched
   areas were captured at both time points (0 and 24 h) and analysed using
   ImageJ Pro Plus 6.0.

Transferase dUTP nick-end labelling (TUNEL) assay

   Terminal deoxynucleotidyl TUNEL staining was performed using a TUNEL
   apoptosis kit (Yeasen, CN) according to the manufacturer's
   instructions. The area of positive staining was quantified using ImageJ
   Pro Plus 6.0.

Flow cytometry

   An Annexin V-FITC/PI kit (Elabscience, CN) was used to assess
   apoptosis. Briefly, AML12 cells were harvested after 24 h of TGF-β1
   treatment, washed twice with PBS, suspended in Annexin V-FITC and
   PI-labelled solution, and incubated at room temperature for 15 min.
   Subsequently, the ratio of apoptotic cells was quantified using flow
   cytometry and analysed using FlowJo software (Stanford University,
   USA).

   A PI kit (Solarbio, CN) was used to assess the cell cycle. LX-2 cells
   treated with TGF-β1 were collected and fixed at 4 °C for 2 h by adding
   precooled 70% ethanol. The samples were then centrifuged at 1500 rpm
   for 15 min at 4 °C, and the precipitates were collected and washed
   thrice with PBS. Subsequently, the cells were resuspended in a PI
   labelling solution and incubated on ice for 15 min. Cell cycle
   distribution was determined using flow cytometry and was analysed using
   FlowJo software.

RNA preparation and quantitative real-time PCR

   Total RNA was extracted using TRIzol (Invitrogen, USA) and reverse
   transcribed with oligonucleotide primers following the manufacturer’s
   recommended protocol. Subsequently, real-time PCR was performed using
   SYBR Green Master Mix and ROX on a StepOnePlus system (Thermo Fisher
   Scientific, USA). The primers for detecting mRNA levels were
   synthesised by Sangon Biotech, and their sequences are as following
   Table [227]2.

Table 2.

   Primers for Quantitative Real-time PCR
   Genes     Primer Sequence (5′–3′)
   GAPDH     F:5′-GGAGCCATCTTTGAGCCTTCA-3′
             R:5′-GAACCAAACTGAGGAATGGATCT-3′
   m ACTA2   F:5′-CGGGAGAAAATGACCCAGATT-3′
             R:5′-GGACAGCACAGCCTGAATAGC-3′
   m CoL IA1 F:5′-GCACGAGTCACACCGGAACT-3′
             R:5′-AAGGGAGCCACATCGATGAT-3′
   m Fn1     F:5′-GAAACCTGCTTCAGTGTGTCTG-3′
             R:5′-TTGAATTGCCACCATAAGTCTG-3′
   m MKP5    F:5′-TTAGACGACAGGGTAGTAGT-3′
             R:5′-GCTGGATGAGGGCATATA-3′
   m IL-6    F:5′-CCACTTCACAAGTCGGAGGCTTA-3′
             R:5′-GCAAGTGCATCATCGTTGTTCATAC-3′
   m IL-1β   F:5′-GCAACTGTTCCTGAACTCAACT-3′
             R:5′-ATCTTTTGGGGTCCGTCAACT-3′
   m TNF-α   F:5′-AACGCCCTCCTGGCCAA-3′
             R:5′-GCAAATCGGCTGACGGTG-3′
   m TGF-β1  F:5′-TTGCTTCAGCTCCACAGAGA-3′
             R:5′-TGGTTGTAGAGGGCAAGGAC-3′
   h IL-6    F:5′-TGAGAGTAGTGAGGAACAAG-3′
             R:5′-CGCAGAATGAGATGAGTTG-3′
   h MCP-1   F:5′-GATCTCAGTGCAGAGGCTCG-3′
             R:5′-TGCTTGTCCAGGTGGTCCAT-3′
   h TNF-α   F:5′-CTCTCTCTAATCAGCCCTC-3′
             R:5′-AGGACCTGGGAGTAGATGA-3′
   h IL-1β   F:5′-CCTGAGCTCGCCAGTGAAA-3′
             R:5′-ATCCAGAGGGCAGAGGTCCA-3′
   [228]Open in a new tab

   The relative mRNA levels normalised to GAPDH were calculated according
   to the comparative 2^−∆∆Ct method.

Western Blotting

   Cells and liver tissues were lysed using RIPA buffer (EpiZyme
   Biotechnology, CN) supplemented with a protease inhibitor (Roche,
   Switzerland). Protein concentrations were determined using a
   bicinchoninic acid (BCA) protein assay kit (EpiZyme Biotechnology, CN)
   and adjusted for immunoblotting. Subsequently, the protein samples were
   separated using 10% SDS-PAGE and transferred to polyvinylidene fluoride
   (PVDF) membranes. Membranes were blocked with 5% BSA in Tris-buffered
   saline containing Tween 20 (TBST). The membranes were then incubated
   with primary antibodies overnight at 4 °C, followed by incubation with
   HRP-coupled secondary antibodies. Finally, the protein bands were
   visualised using an enhanced chemiluminescence (ECL) luminescent liquid
   (YEASEN, CN) on a Multi FluorChem imaging system (Tanon, CN).

   The primary antibodies used in this study included MKP5 (Abclonal, CN),
   Collagen Iα1(Rockland, USA), α-SMA (Santa Cruz, USA), Fibronectin
   (Santa Cruz, USA), cleaved Caspase3 (CST, USA), cleaved Caspase8
   (Abclonal, CN), cleaved Caspase9 (Abclonal, CN), E-cadherin (Affinity,
   USA), N-cadherin (Affinity, USA), Vimentin (CST, USA), p-IRE1α
   (Affinity, USA), IRE1α (CST, USA), p-P38 (CST, USA), P38 (CST, USA),
   p-JNK (CST, USA), JNK (CST, USA), p-ERK (CST, USA), ERK (CST, USA),
   β-Tubulin (Affinity, USA) and β-actin (Abclonal, CN).

RNA sequencing analysis

   RNA sequencing analysis was conducted on fibrotic liver tissues from
   both the WT and KO mice. Total RNA was extracted using the
   aforementioned method, and RNA sequencing was performed by Shanghai
   Jiyan Biotechnology (CN). The data obtained were subsequently analysed
   using Shanghai NovelBio (CN). Principal component analysis was
   performed using the Stats and ggplot2 packages in R (version4.0).
   Hisat2 served as an RNA mapping algorithm to determine the sequence
   location in the genome by comparing the filtered sequence (clean data)
   to the reference genome (mm10_Ensembl100) corresponding to the
   sequenced species (Taxonomy ID:10090). Algorithm software was utilised
   for normalisation treatment, with P < 0.05 was employed to identify the
   DEGs. │NES│ > 1, FDR < 0.25, and P < 0.05 was subjected to GSEA
   analyse. The selected DEGs were subjected to GO and KEGG enrichment
   analyses, and the Fisher test was used to calculate the significance
   level (P value) for GO and KEGG analyses.

Statistical analysis

   All data are presented as the mean ± standard deviation (SD) and were
   analysed using GraphPad Prism9 software. Statistical analyses were
   performed using one-way ANOVA variance, followed by Tukey’s test for
   multiple groups. Differences between two groups were analysed using a
   two-tailed unpaired student’s t-test. Statistical significance was set
   at P < 0.05.

Supplementary Information

   [229]Additional file 1.^ (2.3MB, docx)

Acknowledgements