1 Introduction
Schistosomiasis is a neglected disease caused by trematodes of the genus
Schistosoma and has high morbidity rates in South-east Asia, Africa, and South America [
1,
2]. Schistosomiasis in China is caused by infection with
Schistosoma japonicum [
3]. The pathophysiology of
S. japonicum is due to adult worms residing in the mesenteric veins and excreting eggs that are deposited in the liver, producing granulomatous lesions and subsequently causing periportal fibrosis [
1,
4]. This egg-induced liver fibrosis is the primary cause of mortality and morbidity associated with this chronic disease [
5]. Praziquantel can effectively kill the schistosomes to treat
S. japonicum infection; however, when the causative agent is removed, liver fibrosis still progresses [
6]. Therefore, further elucidation of the molecular mechanisms underlying the reversibility of liver fibrosis in schistosomiasis is needed.
As shown by our group and others, hepatic progenitor cells (HPCs) proliferate and differentiate into either hepatocytes or biliary epithelial cells in response to injury to promote liver regeneration [
7,
8]. Upon liver injury, activated HPCs release cytokines, including osteopontin (OPN), and growth factors (connective tissue growth factor (CTGF) and platelet-derived growth factor (PDGF)), which determine the characteristics of non-parenchymal cells such as hepatic stellate cells (HSCs) [
7,
9]. Recent studies have shown that the expansion of HPCs suppressed the development of hepatic granulomas and fibrosis and further promoted liver regeneration in
S. japonicum-infected mice [
10]. This study suggested an important schistosomiasis-induced connection between HPCs and HSCs that regulates the fibrotic process. Nevertheless, the underlying mechanism remains unclear.
Extracellular vehicles (EVs) are nanoscale membrane-derived vesicles that can transport various substances including DNA, miRNA, and proteins between neighboring cells and distant cells, as a form of cellular communication [
11,
12]. Interestingly, in schistosomiasis, both the pathogen and infected host make and release EVs into the extracellular environment, and both factors likely play a role in disease pathogenesis [
13]. In HPCs, EVs can be induced to maintain liver cell viability and reduce the production of reactive oxygen species (ROS) through cellular crosstalk [
14]. Furthermore, EVs derived from HPCs have shown suppressive effects on fibroblast metabolic activity [
15]. However, whether EVs released by HPCs can regulate liver fibrosis in schistosomiasis is unclear. We hypothesized that EVs released from HPCs can be controlled by specialized pathways and that the engagement of EVs may affect the liver fibrotic process by targeting HSCs in schistosomiasis.
Here, we discovered that HPCs were activated and attenuated the activation of HSCs in S. japonicum-induced liver fibrosis. Cell coculture of HPCs (primary HPCs and the HPC cell line LE/6) and HSCs (primary HSCs and the HSC cell line HSC-T6) was performed. For further verification, we performed proteomic analysis to identify the precise mechanism in HPCs. We also established the in vitro and in vivo models to examine schistosomiasis-related liver fibrosis and the relationships between HPC-derived EVs and HSCs upon pathogen stimulation. Taken together, our results helped elucidate how HPCs regulate the activation of HSCs and affect the progression of liver fibrosis induced by S. japonicum.
2 Materials and methods
2.1 Patient samples
All samples were obtained from Tongji Hospital of Huazhong University of Science and Technology (HUST, Wuhan, China). Liver specimens were collected from 36 patients with chronic schistosomiasis. As controls, biopsy specimens from normal portions of the liver were obtained from 17 patients with hepatic carcinoma or hepatic hemangioma. These samples were used for Western blot analysis and immunohistochemistry (IHC) experiments. Serum samples for EV isolation and detection were obtained from 17 patients with chronic schistosomiasis and 17 healthy people. Liver fibrosis in the 17 patients with chronic schistosomiasis was diagnosed pathologically, and other liver diseases associated with drugs, alcohol, or hepatitis were excluded. All patients provided signed informed consent forms. The study was approved by the Ethics Committee of Tongji Hospital (TJ-IRB20231140) (HUST, Wuhan, China), and sample tissues were utilized in accordance with the appropriate regulations.
2.2 Animals, parasites, and infection
Male C57BL/6J mice weighing 20 g (4 weeks) and male Sprague-Dawley (SD) rats weighing 90 g (4 weeks) were purchased from Weitonglihua Bioscience Co. (Wuhan, China), and were housed in the animal center of Tongji Hospital. All animal experiments were conducted in accordance with the guidelines of the Committee of Animal Care of Huazhong University of Science and Technology, and the study was approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology. SD rats were each infected percutaneously with 100 ± 10 S. japonicum cercariae obtained from the Ecological Station of Oncomelania hupensis in Gongan County, Hubei Province. Mice were exposed percutaneously to 20 ± 5 S. japonicum cercariae. For EV administration, 2 × 109 CON-EVs or soluble egg antigen (SEA)-EVs/rat were intraperitoneally injected twice a week for 3 weeks post 5 weeks after S. japonicum infection. For the pharmacological alteration of autophagy, the rats or mice were intraperitoneally injected with chloroquine (CQ, an autophagy inhibitor) (60 mg/kg) and trehalose (Tre, an autophagy promotor) (1 g/kg) every other day for 4 weeks after 4 weeks of S. japonicum infection. Liver samples were collected and analyzed for the subsequent experiments.
2.3 Isolation of primary HSCs and HPCs
Primary HSCs were prepared from mice using a two-step collagenase-pronase protocol for perfusion of livers as previously described [
16,
17]. In brief, the mouse livers were cut up
in situ with pronase (P5147, Sigma‒Aldrich), collagenase IV (BS165, Biosharp), and deoxyribonuclease (BS137, Biosharp). Next, the dissected livers and the cell suspension were pooled, filtered with a cell strainer (70 mm), and transferred into 50 mL falcon tubes. Nonparenchymal cells (NPCs) were collected from the supernatant after centrifugation at 50×
g for 5 min. Three minutes after the red cell lysis buffer was added (Servicebio), NPCs were centrifuged at 700×
g for 6 min and washed with phosphate-buffered saline (PBS) before being subjected to a density Percoll gradient (25% on top of 33%) (Percoll, 17089102, Cytiva) at 1380×
g for 20 min without braking. Isolated HSCs were maintained in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS) and 1% antibiotics (100 IU/mL penicillin G and 100 mg/mL streptomycin) and cultured at 37 °C. For evaluation of the purity of the HSCs isolated from the livers, desmin-positive cells were manually counted and are expressed as the percentages of total cells. Approximately 90% of the cells were desmin-positive HSCs, as shown by a Leica DM4000 B LED microscope (Fig. S1A).
Primary HPCs from healthy mice and
S. japonicum-infected mice (0 week, 4 weeks, 6 weeks, and 8 weeks) were isolated as described previously [
18]. Briefly, the liver tissue was dissociated to obtain NPCs after hepatocytes were removed. After HSCs were isolated from the NPCs, the remaining cells were incubated with 10 µL of anti-mouse EpCAM microbeads per 10
7 total cells for 30 min at 4 °C. Then, the cells were sorted by magnetic-activated cell sorting (MACS) using a MACS separator and washed three times with buffer following the manufacturer’s instruction (Miltenyi). EpCAM
+ cells were collected through positive selection for further experiments. For determination of the purity, EpCAM
+ cells were stained with BV421-conjugated anti-EpCAM antibody (563214, BD Biosciences, USA) and analyzed with a CytoFLEX-3 system (BD Bioscience, USA). The results showed that the EpCAM
+ progenitor cell populations had a purity of approximately 95.1% (Fig. S1B). Isolated EpCAM
+ cells were seeded in collagen-coated dishes containing DMEM/F12 medium with 10% FBS and 1% penicillin/streptomycin. EpCAM
+ progenitor cells from 7-day
in vitro culture showed stable expression of the hepatic stem cell markers CK19 and SOX9 (Fig. S1C).
2.4 Cell lines and culture
The hepatic stem-like epithelial cell line LE/6 was a gift from Dr. Nelson Fausto [
19]. These cells were cultured in DMEM with Ham’s F-10 (1:1) (Invitrogen, Waltham, MA, USA) supplemented with 1 μg/mL insulin (Promotor, Wuhan) and 0.5 μg/mL hydrocortisone (MedChem XExpress, MCE). HSC-T6 cells were acquired from Procell and maintained in DMEM. All cells were supplemented with 10% FBS. All reagents used to treat cells are listed in Table S1.
2.5 Cellular fractions and Western blot analysis
Nuclear and cytoplasmic fractions of HPCs were prepared by using a commercial kit (Thermo Scientific, 78833). Cells or tissues were lysed using RIPA buffer supplemented with complete ultra-protease/phosphatase inhibitor (Roche, Basel, Switzerland). The lysates were centrifuged at 12 000× g for 15 min, and the cellular debris was discarded. Proteins were resolved by SDS-PAGE and transferred onto Immobilon-PSQ polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA). The membranes were blocked in 5% milk for 1 h and incubated with primary antibodies at 4 °C overnight. Analysis was performed using a Bio-Rad GelDoc system (Bio-Rad, Hercules, CA, USA). The antibodies used are listed in Table S1.
2.6 Histopathology and fibrosis measurement
Slices from paraffin-embedded liver tissues from human, mouse or rat liver samples were used for IHC staining following the instructions of the SP IHC Kit (ZSGB-BIO, Beijing, China). Immunostaining was performed after deparaffinization, rehydration, peroxidase quenching with 3% H2O2 and heat-mediated antigen retrieval. The sections were incubated with primary CK19, OV-6, and ɑSMA antibodies overnight at 4 °C and then with HRP-conjugated secondary antibodies. Finally, 3′-diaminobenzidine (DAB, ZSGB-BIO) was used for visualization. Slices from paraffin-embedded liver samples were used for HE staining and Sirius red staining, and images were acquired by a Leica DM4000 B LED microscope (Leica Microsystems, Wetzlar, Germany). The granuloma size was determined by HE staining of sections using CaseViewer 2 software (3DHISTECH, Budapest, Hungary).
For CK19-EpCAM-SOX9-p62 immunofluorescence staining, sections were treated in the same manner as for immunohistochemistry before blockage, as described above. After blocking with 1% bovine serum albumin (BSA) for 1 h, the sections were incubated with anti-CK19, anti-EpCAM, anti-SOX9, and anti-p62 antibodies overnight at 4 °C. Images were acquired by a Leica DM4000 B LED microscope.
2.7 Immunofluorescence analysis
Cells were fixed in 4% paraformaldehyde for 20 min at room temperature and then permeabilized with 0.1% Triton X-100 in PBS for 10 min. The cells were blocked in PBS with 5% BSA for 30 min at room temperature and then incubated with anti-CD63, anti-LC3B, and anti-LAMP1 antibodies overnight at 4 °C. After 3 washes in PBS for 5 min each, the sections and cells were then incubated with the indicated Alexa Fluor–conjugated secondary antibodies. DAPI was used to detect nuclei. Images were acquired using a laser-scanning confocal microscope.
For analysis of autophagy degradation, cells were grown on glass chamber slides overnight and then transfected with GFP-RFP-LC3 lentivirus for 24 h. After transfection, the cells were exposed to SEA (20 μg/mL) and CQ (20 μmol/L) for 24 h. The cells were then rinsed twice with PBS and fixed in 4% paraformaldehyde for 20 min at room temperature. Images were acquired using a laser-scanning confocal microscope.
For LysoTracker Red staining, live cells were incubated with 20 nmol/L LysoTracker Red for 50 min and with Hoechst 33342 (Solarbio, Wuhan, China) for 10 min. The cells were washed in PBS for 1 min and then fixed by 4% paraformaldehyde for 20 min at room temperature. Images were acquired using a laser-scanning confocal microscope.
2.8 RNA extraction and quantitative real-time PCR (qRT-PCR)
Total RNA was extracted from frozen tissues or cells using RNAiso plus (TAKARA, 9109) following the manufacturer’s instructions. The purity and quality of the RNA were assessed by a NanoDrop system (Thermo Fisher Scientific, USA). The RNA (1 μg) was reverse-transcribed either using a PrimeScript RT reagent kit (TAKARA, RR047A) or a miRNA first-strand synthesis kit (TAKARA, cat#638313). ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711-02) was used for the qRT-PCR analysis. GAPDH and U6 served as a housekeeping gene. The primers used for mature miRNAs were designed by Sangon Biotech. Other primer sequences used in this study are shown in Table S2.
2.9 Cytotoxicity assay
Cell Counting Kit-8 (CCK-8, Dojindo Crop, Japan) was used for cytotoxicity assays according to the manufacturer’s instructions [
20]. LE/6 cells were uniformly seeded in a 96-well plate, and the medium was changed every day. Next, CCK-8 solution was added into each well for 2 h at 37 °C. Finally, the absorbance of the lysates was read at 450 nm using a microplate reader.
2.10 TMT-based quantitative proteomics and data analysis
Proteomic analysis using tandem mass tag-LC-MS/MS was performed by Applied Protein Technology (Shanghai, China). LE/6 cells were harvested and prepared for protein extraction and digestion. After centrifugation, the supernatants were processed for filter-aided sample preparation (FASP) digestion using trypsin, and then, 100 g of the peptide mixture from each sample was labeled using TMT reagent according to the manufacturer’s instructions (Thermo Scientific). The TMT-labeled digested samples were separated into 18 fractions using the Pierce High pH Reversed-Phase Peptide Fractionation Kit (Thermo Fisher Scientific, 84868). Each fraction was subjected to LC-MS/MS analysis on a Q Exactive HF-X Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific) for 60 min. The MS raw data for each sample were searched using the MASCOT engine (Matrix Science, London, UK; version 2.2) embedded into Proteome Discoverer 1.4 software for identification and quantitation. The protein sequences were processed for Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and subsequent enrichment analyses. All identified proteins were selected using a false discovery rate (FDR) threshold of < 0.01. The mass spectrometry proteomic data have been deposited in the ProteomeXchange Consortium via the iProX partner repository [
21,
22] with the data set identifier PXD046558.
2.11 EV isolation, purification, and depletion
EVs were isolated using an ultracentrifugation method [
23]. Cell culture medium was harvested every 24 h from unstimulated or SEA-stimulated HPCs and stored at −80 °C until isolation of EVs. HPC-conditioned medium (CM) and serum were centrifuged for 10 min at 300×
g, 10 min at 2000×
g, and 30 min at 10 000×
g at 4 °C to remove cell debris and aggregates. The supernatant was transferred into new tubes and ultracentrifuged at 120 000×
g for 2.5 h by a Beckman centrifuge. EVs were resuspended in PBS in equal volumes for nanoparticle tracking analysis (NTA), Western blotting (WB), transmission electron microscopy (TEM) or EV coculture. The remaining cell culture supernatant was collected and defined as EV-depleted CM for further studies.
2.12 Internalization of HPC-EVs
Purified HPC-EVs were used for EV internalization into HSC-T6 cells. EVs were resuspended in PBS containing the fluorescent tracing dye PKH26 (MangkangBio) and were labeled according to the manufacturer’s instructions. PKH26-labeled HPC-EVs were incubated with serum-starved HSCs for 12 h in a 6-well tissue culture slide. The cells were washed with cold PBS and fixed/permeabilized with 4% paraformaldehyde. The cells were imaged with a laser scanning confocal microscope.
For EV in vivo tracing studies, LE/6 cell-derived CM was centrifuged to isolate and collect EVs. Then, 1 mmol/L DiR (KeyGEN, Nanjing, China) was used to stain equal amount of EVs (2 × 109 per mouse) with a final concentration of 5 μmol/L at room temperature for 30 min, according to the manufacturer’s protocol. DiR + EVs were washed with PBS and then isolated by ultracentrifugation at 120 000× g for 2.5 h at 4 °C. Rats (n = 2–3) received a single injection of 2 × 109 DiR + EVs per rat via the tail vein or 100 μL of saline. After 24 h, the distribution of DiR-EVs in rats was analyzed using the SPECTRAL Lago X imaging system.
2.13 miRNA sequencing and data analysis
The total RNA from LE/6-EVs was extracted using a miRNeasy Mini kit (Qiagen, cat. no. 217004) according to the manufacturer’s protocol. For library preparation and sequencing, a total amount of 250 pg–10 ng RNA per sample was used as input material using the SMARTer Stranded Total RNA-Seq Kit (TaKaRa Bio USA, Inc.) following the manufacturer’s recommendations, and index codes were added to attribute sequences to each sample. The reads were used to detect known miRNAs and new miRNAs predicted by comparison with known miRNAs from the miRbase and Human Genome (GRCh38) databases, respectively.
2.14 miRNA transfection
miRNA transfection was performed in the conditioned medium using Lipofectamine 3000 as per the manufacturer’s specifications. Three mimics (miR-1306-3p, miR-493-3p, and miR-34a-5p; RiboBio, Guangzhou, China) and a control mimic, the miR-1306-3p inhibitor and a control inhibitor were used. In 6 wells, 20 nmol/L mimics and inhibitors were used with 5 × 105 HSC-T6 cells, and 48 h post-transfection, the cells were harvested and used for subsequent experiments.
2.15 Dual-luciferase reporter assay
A luciferase reporter containing the full-length 3′UTR of Foxo3 (wild-type and mutant-type) was constructed by RiboBio. HSC-T6 cells were transfected in a 24-well culture plate with 300 ng reporter plasmids encoding the wild-type or mutated 3′UTR of Foxo3 and with 20 nmol/L control mimic or miR-1306-3p mimic for 48 h using Lipofectamine 3000 (Invitrogen, Carlsbad, CA). The luciferase activity was measured using the dual-luciferase reporter assay system (Promega).
2.16 Statistical analysis
GraphPad Prism 8 software was used for statistical analysis. All results were expressed as the mean ± SEM of at least 3 independent experiments followed by one-way ANOVA or t-tests. The difference was considered significant for a P value lower than 0.05.
3 Results
3.1 Activation of HPCs during S. japonicum-induced liver fibrosis
First, we investigated whether HPCs were involved in S. japonicum-induced liver fibrosis. The liver specimens from patients with or without S. japonicum infection were collected and assessed by H&E and Sirius red staining (Fig.1). Both the human HPC markers CK19 and OV-6 and the fibrotic marker ɑSMA (actin alpha 2, smooth muscle) (Fig.1) were assessed by IHC staining, and the results showed that the levels of CK19, OV-6, and ɑSMA were significantly elevated in the liver tissues with S. japonicum infection. IF staining of mouse liver sections (Fig.1) also showed a marked increase in the mouse progenitor markers CK19, EpCAM, and SOX9 in the fibrotic livers infected with S. japonicum compared with those of the noninfected mice, with a high level of colocalization among CK19, EpCAM, and SOX9. These results suggested that the HPCs were activated during the S. japonicum-induced liver fibrosis.
3.2 Effects of HPCs on HSCs activation in vitro
Transformation of quiescent HSCs into myofibroblasts is the key event in the progression of liver fibrosis. To further determine whether activated HPCs have a regulatory effect on HSCs during schistosomiasis, we isolated HPCs from mice infected without or with S. japonicum at 4 weeks, 6 weeks, and 8 weeks, and collected the CM from the HPCs. The CM from mice with different durations of S. japonicum infection was subsequently cocultured with primary murine HSCs, and then, the expression of ɑSMA was determined (Fig.2). The expression of ɑSMA was downregulated in HSCs in a time-dependent manner (Fig.2), and CM from the mice infected at 8 weeks with S. japonicum showed the most significant inhibition of HSCs activation.
We next applied an
in vitro model using the HPC cell line LE/6 and the rat HSC cell line HSC-T6 [
19]. A widely used antigenic preparation from schistosomes, soluble egg antigen (SEA), was used as the stimulator to mimic
S. japonicum infection
in vitro, and HSC-T6 cells were cocultured either with LE/6 cells or CM from LE/6 cells stimulated with SEA or PBS for 24 h. The mRNA levels of ɑSMA and collagen type I alpha 1 chain (COl1ɑ1), as HSC-T6 activation markers, were analyzed. We found that CM from LE/6 cells stimulated with SEA significantly attenuated the activation of HSCs (Fig.2). Interestingly, the coculture of LE/6 cells and HSC-T6 cells stimulated with SEA did not have the same results, indicating that a critical mediator secreted by HPCs upon SEA stimulation induced the suppressive effect on HSCs. We therefore further conducted proteomic analysis to identify changes in the protein expression of HPCs under SEA stimulation through high-resolution LC-MS/MS. We identified 244 altered proteins in LE/6 cells between the control and SEA stimulation groups (Fig.2). The differentially expressed genes were visualized using heatmap analysis (Fig.2) and a volcano plot (Fig.2). Pathway enrichment analysis and overlap analysis using the KEGG pathway database were conducted based on the differentially expressed proteins identified by proteomic analysis (Fig.2). We observed an enrichment of genes in phagosome-related pathways and neurological disorders, in which dysregulation of autophagy played an important role [
24,
25]. Thus, we speculated that the autophagy regulation induced by SEA on HPCs might attenuate HSCs activation.
3.3 Impaired autophagy and lysosomal dysfunction in HPCs upon SEA stimulation
To assess the changes of autophagy activity, we examined two autophagic markers, LC3B and sequestosome-1 (p62), in LE/6 cells with and without SEA stimulation. CQ was used as a positive control of impaired autophagy. We found that both the LC3B II/I ratio and p62 were significantly increased in the HPCs with SEA or CQ stimulation, indicating impaired autophagy degradation (Fig.3). As LC3B is associated with autophagosome development and maturation [
26], we used an mRFP-GFP-LC3B reporter construct and transfected it into LE/6 cells to detect degrading versus non-degrading autophagosomes. When autophagy degradation is impaired, the RFP fluorescence is constant, and the GFP signal will not be extinguished, leading to an accumulation of LC3B-positive yellow autophagosomes. As predicted, either SEA stimulation or administration of CQ (positive control) increased the GFP-mRFP autophagosome pool (yellow color), leading to an increased yellow/red ratio in HPCs (Fig.3 and 3C). Next, cell viability was measured by CCK-8 assays at 0, 1, 2, 3, 4, or 5 days after treatment with SEA or CQ. The results showed that both SEA and CQ were not cytotoxic, suggesting that impaired autophagy in HPCs had no significant effect on cell viability (Fig. S2). Together, the results suggested that SEA stimulation resulted in defective autophagy degradation in HPCs.
Lysosomes are involved in two stages of autophagy degradation: autophagosome-lysosome fusion and lysosomal proteolysis [
24,
27,
28]. Impaired lysosomal function is considered the leading cause of autophagic flux blockage, and we next assessed the lysosomal function during SEA-mediated autophagic events in HPCs. We first tested the autophagosome-lysosome fusion in HPCs with SEA stimulation by analyzing the colocalization of LC3B and lysosomal-associated membrane protein 1 (LAMP1; a lysosome marker) by confocal microscopy. Rapamycin (10 µmol/L), an autophagy inducer, was used as a positive control, and bafilomycin A1 (10 nmol/L), which inhibits autophagosome-lysosome fusion, was used as a negative control. The results showed that SEA stimulation did not affect the colocalization of LC3B and LAMP1, indicating a normal autophagosome-lysosome fusion process (Fig.3). Therefore, we next investigated whether SEA stimulation regulated the proteolytic activity of lysosomes in HPCs. Lysosome acidification was evaluated by LysoTracker Red staining, an acidic lysosome indicator. CQ (20 µmol/L) slowed lysosomal acidification and thus was used as a positive control [
29]. SEA stimulation significantly decreased the total number of acidic lysosomes, indicating impaired lysosome acidification in HPCs (Fig.3). We further examined four subunits of the enzyme responsible for the acidification of lysosomes, known as the lysosomal vacuolar-type ATPase or lysosomal proton pump (
Atp6v0c (ATPase H + transporting V0 subunit c),
Atp6v1h (ATPaseH + transporting V1 subunit H),
Atp6v0e1 (ATPase H + transporting V0 subunit e1),
Atp6v0b (ATPase H + transporting V0 subunit b)). We found that SEA repressed the mRNA transcription of all four subunits in HPCs (Fig.3). To further evaluate lysosomal proteolysis, we measured the main proteolytic enzymes cathepsin D (CTSD) and cathepsin B (CTSB). We found that SEA inhibited the conversion of pro-CTSD/pro-CTSB into mature CTSD/CTSB, which is indispensable for the maintenance of lysosomal proteostasis (Fig.3). In addition, we examined the transcription factors of the TFE family, which target various genes involved in the activation of autophagy after translocation from the cytoplasm to the nucleus [
28]. We found that the protein level of transcription factor EB (
Tfeb), a master gene for lysosomal biogenesis [
30], was markedly decreased in the nuclear fractions of HPCs upon SEA stimulation (Fig.3). Together, these results demonstrated that SEA stimulation indeed impaired lysosomal function in HPCs. Finally, we further aimed to confirm that SEA stimulation impaired autophagic degradation by regulating lysosomal dysfunction in HPCs. We found that the increased level of P62 induced by SEA stimulation was returned to the normal level when trehalose (Tre, 1 mmol/L), an autophagy promotor, was applied to HPCs (Fig.3), indicating that the impaired autophagy degradation induced by SEA was reversed by restoring lysosomal function. In conclusion, the results demonstrated that SEA stimulation impairs autophagy degradation by inducing lysosomal dysfunction in HPCs.
3.4 Blocking autophagy ameliorates HPCs-associated fibrosis during S. japonicum infection
First, to investigate whether autophagy modulates HPC-mediated HSC activation following SEA stimulation, we cocultured HSC-T6 cells with CM from LE/6 cells stimulated with SEA, CQ or Tre for 24 h and determined the expression of αSMA in HSC-T6 cells. CQ stimulation of HPCs resulted in the same attenuated activation of HSCs as SEA stimulation, while the decreased level of αSMA induced by SEA stimulation was returned to the normal level in the Tre group (Fig. S3A), suggesting that blocking autophagy reverses HPC-associated HSC activation.
Then, to confirm the role of HPCs in the progression of liver fibrosis induced by S. japonicum infection through autophagy regulation, we determined the autophagy level in HPCs in vivo. We found that p62 was highly expressed and colocalized with CK19 in fibrotic liver tissue from S. japonicum-infected mice (Fig.4), and the protein levels of LC3B and P62 were substantially increased in primary HPCs from the infected mice (Fig.4). Thus, these results suggested that impaired autophagy degradation occurred in HPCs during S. japonicum-induced liver fibrosis. Therefore, we further explored whether drugs that modulate autophagy can affect the progression of liver fibrosis induced by S. japonicum infection. We administered CQ and Tre by i.p. injection every other day for a total of 8 weeks in mice infected with S. japonicum to either inhibit or promote autophagy (Fig.4). We found that when autophagy was impaired by CQ, the liver function of the mice was ameliorated and the ALT and AST levels were reduced (Fig.4). As shown by H&E staining, Sirius red staining (Fig.4), and WB of αSMA and Col1ɑ1 protein levels (Fig.4), CQ effectively restrained liver fibrosis during S. japonicum infection, including reducing granuloma size and collagen deposition (Fig.4). Conversely, Tre exacerbated the fibrosis in the mice with S. japonicum infection. Furthermore, the primary HPCs from the S. japonicum-infected mice treated with CQ or Tre were extracted and cocultured with primary HSCs from WT mice. The increased expression of p62 in the CQ group and decreased expression in the Tre group verified the success of the autophagic modulators in HPCs (Fig. S3B). As expected, the expression of the fibrotic marker ɑSMA was significantly decreased in the CQ group but increased in the Tre group (Fig.4). Taken together, these results demonstrated that disrupting autophagy effectively restrained HPC-associated liver fibrosis induced by S. japonicum infection.
3.5 HPCs inhibit S. japonicum-induced liver fibrosis by secreting anti-fibrotic EVs
Studies have suggested that impaired autophagy degradation or lysosomal dysfunction is the regulatory mechanism for EV processing, formation, and release [
23,
31–
33]. Therefore, we wondered whether the quantity and quality of EVs from HPCs were altered and thus involved in the progression of liver fibrosis induced by
S. japonicum infection. EVs were isolated from LE/6 cells using a differential ultracentrifugation method and were assessed for quantity and quality with several techniques recommended by the International Society for Extracellular Vesicles (ISEV), including TEM, NTA, and WB. TEM images revealed that the EVs had a typical cup-shaped morphology and typical bilayer membrane-like structures (Fig.5). The WB results (Fig.5) showed that the EVs expressed the specific markers CD63, Alix, and TSG101, and the absence of calnexin (a negative marker of EV) confirmed the purity of the EVs. The NTA results (Fig.5) demonstrated that the yield of EVs was increased with SEA stimulation, and the diameters of the EVs were approximately 100 nm. Taken together, these results confirmed the purity and quantity of the isolated EVs and showed that SEA increases the production of EVs derived from HPCs.
To further investigate whether the EVs derived from HPCs stimulated with SEA mediate the antifibrotic effects after coculture with HSCs, we applied either supernatant with depletion of EVs (Fig.5) or an EV release inhibitor (GW4869) (Fig.5). The antifibrotic effects of HPCs were abolished when EVs were depleted (Fig.5) or following addition of GW4869 (Fig.5). ɑSMA expression was significantly downregulated in HSC-T6 cells cocultured with EVs from LE/6 cells upon SEA stimulation (SEA-EVs) compared with those without SEA stimulation (CON-EVs) (Fig.5). We confirmed that the EVs were largely taken up by recipient HSC-T6 cells by labeling the EVs with the red lipid membrane dye PKH26 (Fig. S4A). Thus, the results suggested that the EVs are the key substances of CM from SEA-activated HPCs that mediate the antifibrotic effect.
To explore the role of HPC-derived EVs in vivo, we established a S. japonicum infection rat model. Rats were infected with S. japonicum and subsequently given either CON-EVs or SEA-EVs by intraperitoneal (i.p.) injection (Fig. S4B). We first measured the liver localization of EVs following i.p. injection of rats with a single dose of DiR-labeled CON/SEA-EVs (2 × 109). Twenty-four hours later, we confirmed that most of the fluorescent signal was localized in the liver (Fig. S4C). Then, the effects of HPC-derived EVs on the progression of liver fibrosis caused by S. japonicum infection were assessed. We found that the administration of SEA-EVs markedly reduced the pathological damage in the liver compared with that in the CON-EV group, as indicated by diminished levels of ALT and AST (Fig.5). H&E staining, Sirius red staining, and IHC staining of αSMA (Fig.5) revealed that SEA-EVs inhibited liver fibrosis in the rats infected with S. japonicum. Moreover, a significant reduction in extracellular matrix (ECM) deposition (Fig.5), ameliorated liver function, and reduced granuloma size in liver tissues were observed in the SEA-EV-treated rats. We further assessed the expression of the fibrotic markers ɑSMA and Col1ɑ1 in the liver by WB and found a remarkable reduction in these markers (Fig.5). Collectively, these results showed that HPCs inhibited the progression of liver fibrosis induced by S. japonicum by secreting EVs to attenuate the activation of HSCs.
3.6 Impaired autophagy degradation reduced the secretion of miRNAs in HPC-derived EVs
To investigate how EVs derived from HPCs affected the progression of
S. japonicum-induced fibrosis, we analyzed the EV cargo. Small RNAs account for a large proportion of cargoes in EVs and are critical post-transcriptional regulators in cell-to-cell communication [
13,
34]. Among them, miRNAs are the most studied and their regulatory roles in host–pathogen interactions are increasingly clear. Therefore, we performed miRNA sequencing of SEA-EVs and CON-EVs derived from LE/6 cells. Principal component analysis (PCA) revealed that SEA altered the miRNA profiles of EVs (Fig.6). Among the 1218 miRNAs identified, 36 miRNAs with significantly altered expression between the SEA-EVs and CON-EVs were identified (Fig.6 and 6C). Then the target genes of the differentially expressed miRNAs were analyzed by KEGG pathway analysis (Fig. S5A), and the results showed that many pathways differed between the two groups. Overall, the miRNA analysis revealed major differences in the EV cargo between the CON-EVs and SEA-EVs, which suggested that regulation of EV cargo of HPCs is involved in the liver fibrosis induced by
S. japonicum infection.
We further compared the expression of these identified 36 miRNAs between donor and recipient cells. Three miRNAs—miR-1306-3p, miR-493-3p, and miR-34a-5p—showed downregulated expression in the recipient HSC-T6 cells, showing the same trend as that in the EVs derived from donor LE/6 cells (Fig.6, 6E, and S5B). Therefore, we selected these three miRNAs as our study focus. Studies have reported that the regulatory effect of autophagy on EV cargo and the autophagy inhibitor CQ induced distinct extracellular vesicle populations [
35,
36]. To further examine the regulatory effect of autophagy on these three miRNAs in donor cells, we applied CQ. The expression of miR-1306-3p, miR-493-3p, and miR-34a-5p was significantly downregulated in EVs but upregulated in donor LE/6 cells when autophagic degradation was impaired by SEA or CQ. Nevertheless, restoration of autophagy by treatment with Tre eliminated the differences in the expression of the three miRNAs between EVs and their donor cells (Fig.6 and 6G). Thus, the results indicated that altered autophagy affects the secretion of specific miRNA-enriched EVs from their donor cells during
S. japonicum infection-induced liver fibrosis.
EVs are produced by intraluminal vesicles budding from the limiting membrane of multivesicular bodies (MVB) [
12,
37,
38]. To further study the ability of HPCs to selectively secrete miRNAs in EVs and their intracellular route, we developed a method to visualize their delivery to the lumen of MVBs. We used a constitutively active mutant of Rab5-GTPase, Rab5Q79L, which results in the accumulation of caveola-derived vesicles docked on early endosome [
39,
40]. The delivery of miRNAs from the cytoplasm to MVBs can be visualized as the enlarged endosomes are detected by confocal microscopy in LE/6 cells with overexpression of GFP-Rab5Q79L. miR-1306-3p, miR-493-3p, and miR-34a-5p were labeled with Cy3 dye and transfected into HPCs overexpressing GFP-Rab5Q79L. We found that the proportion of MVBs containing individual miRNA was substantially decreased in the HPCs treated with SEA compared with that of the control group. Additional CQ increased this reduction, while Tre prevented the decrease induced by SEA (Fig.6). Thus, we confirmed that impaired autophagic degradation inhibits intraluminal filling of MVBs with miR-1306-3p, miR-493-3p, and miR-34a-5p in HPCs stimulated with SEA. To verify the effect
in vivo, we extracted EVs from primary HPCs in mice infected with
S. japonicum and examined the expression of miR-1306-3p, miR-493-3p, and miR-34a-5p. Consistent with the results
in vitro, all three miRNAs were significantly decreased in the EVs derived from HPCs of the mice infected with
S. japonicum, and i.p. injection of CQ further decreased the expression of the three miRNAs while treatment with Tre increased the expression (Fig.6). Collectively, these results demonstrated that impaired autophagy in HPCs regulates the cargo of EVs and effectively controls the secretion of specific miRNAs in EVs during
S. japonicum infection.
3.7 Active effect of miR-1306-3p on HSCs
To clarify the involvement of miR-1306-3p, miR-493-3p, and miR-34a-5p in HSC activation, we transfected the three miRNA mimics into HSC-T6 cells in vitro (Fig.7). We observed that each miRNA activated HSCs with increased expression of ɑSMA, Col1ɑ1, connective tissue growth factor (CTGF), and fibronectin (Fig.7). We previously showed that the expression of ɑSMA in HSC-T6 cells was suppressed by SEA-EVs; nevertheless, the suppressive effects were reversed with SEA-EVs derived from LE/6 cells transfected with each individual miRNA mimic or all three miRNA mimics (Fig.7 and S6). These results suggested that miR-1306-3p, miR-493-3p, and miR-34a-5p may be the causative factors in the EVs that control crosstalk between HPCs and HSCs during the fibrotic process.
We further chose miR-1306-3p, which exhibited the most obvious reduction in EVs, to explore its target genes in HSCs through bioinformatics analysis. Using miRNA target prediction programs (Targetscan Release 3.1 and miRanda) allowed us to predict the target genes involved in hepatic fibrosis, and 18 genes were identified to be the putative targets of miR-1306-3p (Fig.7). Among them, forkhead box O3 (
Foxo3) has been reported to exhibit antifibrotic effects [
41–
43], and its associated genes were shown to be involved in FoxO signaling through KEGG enrichment pathway analysis (Fig. S4A). We further confirmed the expression of
Foxo3 in HSC-T6 cells, and its level was significantly reduced after treatment with SEA-EVs (Fig.7). Moreover, the expression of
Foxo3 was downregulated when HSC-T6 cells were transfected with miR-1306-3p mimic but restored when the miR-1306-3p inhibitor was applied (Fig.7 and 7G), demonstrating the negative regulation of miR-1306-3p on
Foxo3. Then, we performed a luciferase assay to confirm whether miR-1306-3p directly targeted the 3′UTR of
Foxo3 through its cognate site. As shown in Fig.7, miR-1306-3p significantly reduced the luciferase activity of the wild type
Foxo3-3′UTR reporter gene but not the mutant type
Foxo3-3′UTR reporter gene, which indicated that
Foxo3 was the direct target gene of miR-1306-3p via specific binding to the 3′UTR of
Foxo3 mRNA. Taken together, these findings indicated that the miR-1306-3p activates HSCs by targeting the
Foxo3 gene.
4 Discussion
The role of HPCs in
S. japonicum infection-associated liver fibrosis has not been fully elucidated. In this study, we showed that the interaction between HPCs and HSCs affects liver fibrosis during
S. japonicum infection. Eggs excreted by the adult worms in mesenteric veins will migrate by portal blood to the liver, induce granulomatous lesions and subsequently cause periportal fibrosis [
1,
4]. Under these conditions, HPCs located in the canals of Herring may receive stimulation by antigens released from eggs while they penetrate the portal system. In the current study, we therefore used SEA, the crude extract obtained from the mature egg, to explore the effect of HPCs following
S. japonicum egg stimulation. For the first time, we identified a new mechanism involving modulation of autophagy in HPCs to regulate the release of antifibrotic EVs in schistosomiasis. We also identified a new role of impaired autophagy and lysosomal dysfunction in HPCs upon SEA stimulation, which subsequently promotes the antifibrotic EV release. These EVs contained fewer profibrotic miRNAs, including miR-1306-3p, miR-493-3p, and miR-34a-5p, resulting from inhibition of autophagy, which affects sorting these three miRNAs into MVBs. Furthermore, we showed that these three miRNAs released from HPC-EVs mediated the inhibitory role in HSC activation (Fig.8).
Autophagy is a conserved lysosomal degradation process essential for cellular homeostasis and adaption to various stressors [
44–
46]. Autophagy can eliminate damaged organelles or proteins to exert protective roles in liver-related diseases, and hepatocytes play a central role in this process [
47]. Hepatocyte-specific autophagy-related 7 (Atg7) or autophagy-related 5 (Atg5) knockout mice showed induced hepatocyte death and inflammation [
48]. In our study, proteomic analysis and phenotype experiments revealed that autophagy was involved in HPC activation (Fig.2). Indeed, we observed that SEA stimulation impaired autophagic degradation in HPCs to limit HSC activation (Fig.3), indicating that autophagy was not protective in HPCs. As shown in Fig. S2, impaired autophagy had no significant effect on cell viability in HPCs, suggesting that autophagy may not always cause cell death and inflammation. Some landmark studies showed that loss of autophagic function in HSCs reduced fibrogenesis and matrix accumulation without detrimental cell deaths [
49,
50]. In liver sinusoidal endothelial cells (LESCs), the autophagy inhibitor 3MA maintained the fenestration state and relieved CCl4-induced fibrosis [
51]. All these observations suggest that impaired autophagy does not only induce abnormal liver physiology. Autophagy has a context-dependent impact on fibrosis and displays different functions depending on the cell type [
47]. In schistosomiasis, autophagy has been reported to play essential roles in the modulation of not only the pathogen’s homeostasis but also inflammation, injury, or fibrosis in the host [
52,
53]. The application of autophagy inhibitor bafilomycin negatively affected worm fitness and egg production of adult
S. mansoni [
54,
55]. Recent studies have suggested that by controlling autophagy in different cell types in the host, including macrophages and HSCs, liver fibrosis could be prevented in
Schistosoma-induced liver damage [
46]. The elevation of LC3B, a marker of autophagic activity, was correlated with the ɑSMA level in livers of
S. japonicum-infected mice, suggesting the interaction between autophagy and HSC activation [
56]. In our study, we observed that SEA stimulation impaired autophagic degradation in HPCs by preventing lysosomal acidification and proteostasis. Here, we used the autophagy inhibitor CQ, which is also as a lysosomal inhibitor, and the autophagy promotor Tre in an
S. japonicum infection model
in vivo (Fig.4). In this model, CQ significantly inhibited the progression of fibrosis in the livers of
S. japonicum-infected mice, while Tre promoted liver fibrosis. Our results showed that application of CQ not only inhibited the degree of liver fibrosis in the advanced stage of schistosomiasis, but also delayed the progression of granuloma formation, suggesting the potential feasibility of CQ in clinics to treat
S. japonicum infection.
EVs are nanoscale membrane-derived vesicles that can transport various substances including DNA, miRNA, and proteins between neighboring cells and distant cells, as a form of cellular communication [
11,
12,
38]. EVs derived from HPCs have shown suppressive effects on fibroblast metabolic activity [
15]. Furthermore, EVs can be induced from HPCs to maintain liver cell viability and reduce the production of reactive oxygen species (ROS) through cellular crosstalk [
14]. In our study, we observed that the EVs were the main substance of CM from HPCs with a suppressive effect on HSC activation (Fig.5). The EVs derived from HPCs after SEA stimulation significantly inhibited the progression of
S. japonicum-induced fibrosis
in vivo, which provided a direction for the development of antifibrotic treatment. As miRNAs constitute one of the most important exosomal cargoes to regulate host–parasite interactions, we conducted miRNA sequencing analysis of EVs during
S. japonicum infection. We verified that reduced sorting of miR-1306-3p, miR-493-3p or miR-34a-5p into EVs resulted in the antifibrogenic effects of HPC-EVs. We first showed that inhibition of autophagy, especially lysosomal dysfunction, was involved in the transfer of miR-1306-3p, miR-493-3p, and miR-34a-5p into a specific cellular compartment, namely MVBs (Fig.6). Our data showed minimal differences in intracellular expression of these three miRNAs in HPCs, while a profound difference was observed in EVs. Using GFP-Rab5Q79L to visually localize MVBs, we found that impaired autophagic degradation could significantly decrease the packaging of miR-1306-3p, miR-493-3p, and miR-34a-5p within MVBs. Future studies will be focused on the detailed molecular and biological mechanisms of how autophagy regulates the sorting of miRNAs into MVBs in HPCs during
S. japonicum infection.
In our study, we confirmed that three miRNAs in the EVs derived from HPCs induce the activation of HSCs during
S. japonicum infection. Among miR-1306-3p, miR-493-3p, and miR-34a-5p, miR-1306-3p was selected to examine its putative targets due to its significant reduction in EVs upon SEA stimulation, and we identified its downstream target,
Foxo3, a suppressor of fibrosis, by bioinformatics analysis [
41,
57,
58]. Interestingly, previous research indicated that
Foxo3 modulates HSC activation dependent on different antigenic preparations from schistosomes. For example, when using SEA, the activation of the
Foxo3a pathway in HSCs was observed, whereas in contrast, the application of recombinant
S. japonicum egg antigen P40 (rSjP40) could inhibit
Stat5 and
Foxo3a expression [
59,
60]. Thus, we hypothesize that the antifibrotic effect of HPC-delivered miRNAs on
Foxo3 signals is a special immunologically-driven response to eggs during
S. japonicum infection. Future studies will surely be conducted to dissect the detailed molecular mechanisms.
The present study has some limitations. Specific regulation of autophagy in HPCs in vivo to verify its ability to inhibit liver fibrosis was not conducted in the present study, although our results revealed that CQ and Tre i.p. injection restrained or exacerbated liver fibrosis induced by S. japonicum infection. The EVs released by HPCs might influence many other cells in livers in addition to HSCs and may be involved in the fibrosis induced by S. japonicum infection. Finally, only miRNA sequencing was performed in EVs in the present study, while the EV cargo contained many proteins and other small RNAs, including snRNAs, tRNAs, and piwi-interacting RNAs, in addition to miRNAs.
This study revealed that suppression of autophagy in HPCs has critical and protective role in limiting HSC activation and reducing S. japonicum-induced fibrosis. Autophagy decreased the intraluminal filling of EVs with miRNAs, including miR-1306-3p, miR-493-3p, and miR-34a-5p, in HPCs during S. japonicum infection to mediate HSC activation. This mechanism could be of particular interest and might set the stage for targeting autophagy inhibition in HPCs as a therapeutic strategy for schistosomiasis.
PDF
(9573KB)
