XBP1 deficiency promotes hepatocyte pyroptosis by impairing mitophagy to activate mtDNA-cGAS-STING signaling in macrophages during acute liver injury

2.1. Patients and specimens

Five patients with ALI and five control patients with haemangioma without ALI undergoing partial liver resection (Supplementary Table 1) in the First Affiliated Hospital of Nanjing Medical University were enrolled in the current study. Liver biopsy specimens from patients with ALI or peri-tumor normal tissues from control patients were collected. Informed consent was obtained from each patient. All procedures that involved human samples were approved by the Ethics Committee of the Affiliated Hospital of Nanjing Medical University (Institutional Review Board approval number 2020-SRFA-220).

2.2. Animals

Hepatocyte-specific Xbp1 knockout (HKO) mice were generated by breeding Xbp1-LoxP with Alb-Cre mice (Nanjing Biomedical Research Institute of Nanjing University). Briefly, homozygous Xbp1^loxP/loxP mice were first bred with homozygous Alb-Cre mice, and the heterozygous offspring (for both Xbp1 and Cre) were back-crossed with homozygous Xbp1^loxP/loxP mice. Mice were housed and maintained under a 12-h light/12-h dark cycle with ad libitum access to water and standard chow with supplements under specific pathogen-free conditions. All animals were treated humanely, and all animal procedures met the relevant legal and ethical requirements according to the protocols (number NMU08-092) approved by the Institutional Animal Care and Use Committee of Nanjing Medical University.

2.3. Mouse acute liver injury induced by TAA and LPS

Male Xbp1^loxP/loxP (WT) and HKO mice (8-week-old) were intraperitoneally injected with 300 mg/kg TAA (Sigma, Saint Louis, MO, USA) or lipopolysaccharide (LPS, 5 mg/kg, Sigma, USA) dissolved in phosphate-buffered saline (PBS). The same volume of PBS was intraperitoneally injected into the control mice. Mice were sacrificed 24 h after TAA treatment or 6 h after LPS challenge. In some experiments, mice were intraperitoneally administered chloroquine (CQ, 60 mg/kg, Sigma, Saint Louis, MO, USA), MitoTEMPO (20 mg/kg, Sigma), or AC-YVAD-cmk (8 mg/kg, Sigma) 1 h before TAA or LPS injection. Mice were intraperitoneally injected with 1 mg/kg of tunicamycin (TM, Sigma) and sacrificed 6 h after injection.

2.4. Hepatocellular function assay

Blood samples were centrifuged to obtain serum, and the levels of aspartate aminotransferase (ALT) and alanine aminotransferase (AST) were measured using an automatic chemical analyzer (Olympus Company, Tokyo, Japan).

2.5. Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay

Sections of paraffin-embedded hepatic tissue were washed with PBS and fixed with 4% polyformaldehyde and then incubated with a TUNEL reaction mixture (Roche, Basel, Switzerland). Briefly, slides were rinsed twice with PBS. Next, 50 $\mu$ l TUNEL reaction mixture was added to samples, which were incubated for 60 min at 37$°C$ in a humidified atmosphere in the dark. Slides were rinsed three times with PBS. Samples were analyzed in a drop of PBS under a fluorescence microscope at this state, using an excitation wavelength in the range of 450–500 nm and detection in the range of 515–565 nm (green).

2.6. Histopathology and immunohistochemical staining

Liver tissue sections (4-$\mathrm{\mu }$ m-thick) were stained using haematoxylin and eosin (H&E) to observe inflammation and tissue damage using light microscopy. F4/80 and STING were identified by immunofluorescence using a primary rat anti-mouse F4/80 (1:100, ab11101, Abcam) and STING (1:200, #13647S, Cell Signaling Technology) antibodies, and secondary biotinylated goat anti-rat IgG (1:100, BA-1000, Vector) according to the manufacturer's instructions. Positive cells were counted blindly at 10 HPE/section.

2.7. Bone marrow-derived macrophage (BMDM) culture

Bone marrow cells were isolated from the mouse femurs and tibias as previously described [15]. BMDMs were then re-plated and cultured overnight in new culture dishes for further experiments.

2.8. Isolation of liver cells

Primary hepatocytes and liver intrahepatic macrophages were isolated by a two-stage collagenase perfusion method as described previously [15]. Primary hepatocytes were stimulated with lipopolysaccharide (LPS; 1 $\mathrm{\mu }$ g/mL, 6 h) followed by treatment with nigericin (20 $\mathrm{\mu }$ M, 1 h) in the presence or absence of NLRP3 inhibitor MCC950 (10 $\mathrm{\mu }$ M) and treated with TM (2 $\mu$ g/ml) for 3 h to induce ER-stress. In some experiments, cells were pre-treated with chloroquine (CQ, 10 $\mathrm{\mu }$ M), MitoTEMPO (50 $\mathrm{\mu }$ M), or AC-YVAD-cmk (25 $\mathrm{\mu }$ g/mL) for 1 h prior to TAA (70 $\mathrm{\mu }$ M, 6 h) with or without MCC950 (10 $\mathrm{\mu }$ M) administration. BMDMs and intrahepatic macrophages were treated with or without mtDNA isolated from primary hepatocytes for 12 h, and the cells and supernatants were harvested for further analysis.

2.9. Measurement of ROS levels

Two kinds of fluorescent probes were used to detect intracellular ROS in liver tissues and cells, namely the ROS-sensitive probe 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA), and dihydroethidium (DHE). For DHE staining, cryosections were incubated with 10 $\mathrm{\mu }$ M DHE at 37$°C$ for 30 min. For ROS detection in cultured cells, the cells were seeded in a 96-well plate and incubated with H₂DCFDA. Fluorescence was detected by a confocal microscope system and flow cytometry at 488 nm excitation and 525 nm emission.

2.10. Measurement of malondialdehyde (MDA), superoxide dismutase (SOD) and ROS levels and the ratio of GSH/GSSG

MDA, GSH, GSSG, SOD and ROS levels were estimated using commercial kits. Liver tissues were washed with PBS, homogenized in lysis buffer and sonicated. After being sonicated, the lysed tissue was centrifuged (10,000×g, 10 min) to remove debris and retain the supernatant. A microplate reader was used to measure the levels of MDA, GSH, GSSG, SOD and ROS in the supernatant. In addition, MDA, GSH, GSSG, SOD, and ROS levels were normalized according to the protein concentration.

2.11. Adeno-associated virus (AAV)-mediated gene transfer

For knockdown cGAS or overexpression of PINK1, the AAV vectors carrying cGAS shRNA or control shRNA and AAV overexpressed PINK1 or AAV-enhanced green fluorescent protein (eGFP) (Shanghai GeneChem) were intravenously injected into the tail vein of 8-week-old WT or HKO male mice at a dose of 4 $\times$ 1,0¹¹ plaque-forming units per mouse.

2.12. Lentiviral knock down of cGAS in BMDMs

BMDMs were prepared and infected with control or cGAS-siRNA lentiviruses (Shanghai GeneChem) for 72 h at 37 °C in the presence of 4 mg/ml polybrene. BMDMs were then washed and cultured in fresh medium for 24 h before mtDNA stimulation for further analysis.

2.13. mtDNA isolation and quantification

DNA was extracted from 200 $\mathrm{\mu }$ l of cells culture supernatant of untreated and treated primary murine hepatocytes using the QIAmp DNA mini kit (QIAGEN, Duesseldorf, Germany) according to the manufacturer's instructions. Real-time polymerase chain reaction (qPCR) was performed for the quantification of mtDNA. mtDNA was quantified using mouse mt-ATP6 primers/Taqman 5′ FAM-3′ MGB probe (Bio-Rad, Hercules, CA, USA). The mitochondrial lysate of primary hepatocytes was used to isolate mtDNA using a mitochondrial DNA isolation kit (Abcam) according to the manufacturer's instructions. Total DNA was isolated using the DNeasy Blood and Tissue Kit (QIAGEN) following the manufacturer's instructions. RT-qPCR analysis was performed using a sequence detection system (ABI Prism 7000, Applied Biosystems, Foster City, CA, USA/Invitrogen, Carlsbad, CA, USA) with an SYBR green 1-step kit (Invitrogen). To evaluate the possibility of BMDMs engulfing mtDNA released from hepatocytes, mtDNA isolated from primary murine hepatocytes was tagged with Cy5-dCTP (Amersham Cy5-dCTP, GE Healthcare) using PCR, as described by the manufacturer (BioPrime DNA Labeling system, Life Technologies). The Cy5-labeled DNA PCR was cleaned with a DNeasy Mini Spin Column (QIAGEN) before the DNA concentration was measured using a Nanodrop (Thermo Scientific).

2.14. Mitochondria and cytosolic dsDNA measurements

The PicoGreen dsDNA Reagent and Kit (Invitrogen, no. P11496) and MitoTracker Red CMXRos (Invitrogen, no. M5712) were used at 0.3% v/v and 100 nM in serum-free DMEM for 30 min at 37 $°C$ to co-localize dsDNA and mitochondria. To determine the mitochondrial membrane potential, cells were stained with tetramethylrhodamine methyl ester (100 nM; Invitrogen, no. I34361) for 15 min. Stained cells were then examined using a confocal microscope system (LSM880, Carl Zeiss, Inc.) equipped with a 63 × 1.2 NA objective lens (Carl Zeiss, Inc.).

2.15. Immunofluorescence staining

LC3 in primary hepatocytes and F4/80 and STING in macrophages were identified using immunofluorescence with rabbit anti-mouse LC3B mAb (1:200, #83506S, Cell Signaling Technology), rat anti-mouse F4/80 mAb (1:200, ab6640, Abcam), and rabbit anti-mouse STING mAb (1:200, 19851-1-AP, Proteintech). Thereafter, they were incubated with secondary Cy3-conjugated goat anti-rabbit IgG (1:100, AP132C, Sigma) and FITC-conjugated goat anti-rat IgG (1:100, AP136F, Sigma) according to the manufacturer's instructions. 4′,6‐diamidino‐2‐phenylindole and Hoechst 33,342 (Thermo Scientific, no. 62249) were used for nuclear staining. To assess pyroptosis, active caspase-1 was measured using the FLICA reagent FAM-YVAD-FMK (FAM-FLICA Caspase-1 Assay Kit, ImmunoChemistry Technologies, USA) according to the manufacturer's instructions, and propidium iodide (PI) was used to mark cells with membrane pores. Stained sections were imaged with a confocal microscope system LSM880 (Zeiss) and analyzed using Zen software (Zeiss).

2.16. Western blot analysis

Proteins were extracted from liver tissue or cells using ice-cold lysis buffer (50 mM Tris, 150 mM Nacl, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate and 1% Triton-100). Proteins (20 $\mathrm{\mu }$ g/sample) were subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride nitrocellulose membranes (Bio-Rad). Anti-rabbit cGAS (1:1000, #31659S), STING (1:1000, #13647S), P-TBK1 (1:1000, #5483S), P-IRF3 (1:1000, #29047S), P–NF–$\mathrm{\kappa }$ B (1:1000, #3033S), NLRP3 (1:1000, #15101S), cleaved caspase-1 (1:1000, #89332S), P62 (1:1000, #23214S), LC3B (1:1000, #43566S) and anti-mouse $\mathrm{\beta }$-actin (1:1000, #3700S) antibody were from Cell Signaling Technology; Anti-rabbit XBP1 (1:1000, ab37152) and GSDMD (1:1000, ab209845) antibody were from Abcam.

2.17. Quantitative real-time-PCR

Total RNA was purified from cells using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. Reverse transcription into cDNA was performed using a Transcriptor First‐Strand cDNA Synthesis kit (Roche). Quantitative real‐time PCR was performed using SYBR green (Roche) on a StepOnePlus Real-Time PCR System (Applied Biosystems). Quantitative PCR was repeated three times. The expression levels of target genes were normalized to that of hypoxanthine phosphoribosyltransferase.

2.18. Transmission electron microscopy

Transmission electron microscopy of hepatocytes was performed according to the manufacturer's instructions. The sections were further stained with 0.3% lead citrate and imaged using an electron microscope (HITACHI, Tokyo, Japan; 4000 × or 30,000 × magnification).

2.19. Enzyme-linked immunosorbent assay (ELISA)

Cytokine/chemokine levels (IL-1β, IL-18, MCP-1, and CXCL-10) were measured using ELISA kits according to the manufacturer's protocols (Thermo Fisher Scientific).

2.20. Statistical analyses

Data are expressed as the mean ± SEM. Two-group comparisons were performed using two-tailed student's t-tests, comparisons between multiple groups were performed using one-way or two-way ANOVA with Tukey's multiple comparisons. All statistical analyses were performed using STAT software, version 11.0.

3.1. Hepatocyte-specific XBP1 deficiency aggravated TAA-induced ALI

We first examined XBP1 protein levels in the liver after TAA or PBS treatment. Significantly increased protein and mRNA levels of XBP1 were observed in the livers after TAA treatment (Fig. 1A and Supplementary Fig. 1A). To better understand the functions of XBP1 in regulating hepatocyte cell injury following TAA treatment, we used the Cre-LoxP system to create hepatocyte-specific XBP1 KO mice. In these HKO mice, XBP1 expression was significantly decreased in liver tissues and deficient in hepatocytes, although similar in liver non-parenchymal cells (Fig. 1B). Hepatocyte XBP1 depletion aggravated TAA-induced liver injury as evidenced by higher levels of serum ALT and AST (Fig. 1C), exacerbated architectural destruction (Fig. 1D), and increased hepatocellular cell death (Fig. 1E). The liver proinflammatory immune response was exacerbated in XBP1 HKO mice as shown by increased proinflammatory cytokine/chemokine gene expression induction and protein expression (Supplementary Figs. 1B and C). Much more F4/80-positive macrophages were significantly more abundant in the livers of HKO mice than that in those of WT controls after TAA treatment (Supplementary Fig. 1D). Summarily, these data suggested that XBP1 deficiency in hepatocytes aggravated TAA-induced ALI and macrophage-related intrahepatic inflammation.

3.2. XBP1 deficiency promoted mtDNA release from TAA-stressed hepatocytes to activate macrophage STING signaling

STING-mediated macrophage proinflammatory response was reported in a variety of liver diseases [14]. Thus, we hypothesized that macrophage STING signaling functions in regulating TAA-induced ALI. Intrahepatic macrophages were isolated from the livers after TAA treatment, and the STING signaling pathway was analyzed. Macrophages from HKO mice showed enhanced activation of the STING signaling pathway in macrophages, as evidenced by enhanced STING staining (Fig. 2A) and increased protein levels of cGAS, STING, P-TBK1, P-IRF3, and p–NF–kB (Supplementary Fig. 3A).

We previously found that mtDNA release from hepatocytes contributes to STING activation of macrophages [16]. Next, we investigated if enhanced macrophage STING activation was caused by increased mtDNA release from XBP1 deficient hepatocytes. To test this hypothesis, BMDMs were stimulated with isolated mtDNA or supernatant from TAA-treated primary hepatocytes. Results showed that mtDNA-Cy5 was engulfed by BMDMs early at 4 h and peaked at 12 h post-co-culture of mtDNA-Cy5 with BMDMs (Fig. 2B). Dose-dependence of mtDNA stimulation in activating cGAS-STING signaling (Supplementary Fig. 2A) and the subsequent proinflammatory response of macrophages (Supplementary Figs. 3B and C) were also observed, indicateing that increased mtDNA release from hepatocytes might contribute to enhanced macrophage STING activation, as observed in HKO mice.

Thus, we next measured the extracellular release of mtDNA from hepatocytes in WT and HKO mice after TAA treatment. Primary hepatocytes were isolated from HKO mice after TAA treatment and cultured in vitro, and then the supernatant levels of mtDNA were analyzed. The results showed that hepatocytes from HKO mice released increased levels of mtDNA in the supernatant (Fig. 2C and D). Stimulation of BMDMs with the above supernatant effectively activated STING signaling in macrophages. BMDMs stimulated with the supernatant of hepatocytes from HKO mice showed enhanced STING signaling (Fig. 2E).

We also evaluated the direct effect of TAA treatment on mtDNA modification, which may further modulate macrophage STING activation. BMDMs stimulated with the same dose of mtDNA isolated from primary hepatocytes with or without TAA pre-treatment showed a similar level of cGAS-STING activation (Supplementary Fig. 3B) and subsequent proinflammatory response (Supplementary Fig. 3C). Moreover, treatment of mitoTEMPO, a specific ROS scavenger of mitochondria, decreased mtDNA level in the cell supernatant of cultured primary hepatocytes post TAA treatment (Supplementary Fig. 3D), indicating an important role of mtDNA, but not other dsDNA, in the activation of STING signaling. Collectively, these results indicated that XBP1 deficiency promoted extracellular mtDNA release from TAA-stressed hepatocytes to activate macrophage STING signaling.

3.3. Hepatocyte-derived mtDNA promoted macrophage STING activation in a cGAS-dependent manner

cGAS activation induced by dsDNA functions in the synthesis of cGAMP and subsequent activation of the STING signaling pathway. To further examine the role of the mtDNA-cGAS-STING signaling axis in regulating proinflammatory macrophage response, we blocked macrophage cGAS activation by AAV-cGAS-shRNA in vivo (Supplementary Fig. 4A). Immunohistochemistry analysis of STING in liver tissues showed that inhibition of cGAS activation in macrophages decreased intrahepatic STING activation (Fig. 3A). Furthermore, blockade of cGAS activation reversed the upregulation of STING signaling pathway activation (Fig. 3B) and proinflammatory cytokine and chemokine production (Fig. 3C) of intrahepatic macrophages by hepatocyte XBP1 depletion. Functionally, in vivo blockade of cGAS activation in macrophages abrogated the role of hepatocyte XBP1 depletion in aggravating TAA-induced liver injury, as evidenced by the analysis of serum levels of ALT and AST, pathological examination, and TUNEL staining (Supplementary Figs. 4B–D). Reduced infiltration of F4/80-positive macrophage infiltration (Supplementary Fig. 4E) and decreased intrahepatic inflammation (Supplementary Fig. 4F) were also observed after cGAS inhibition.

Subsequently, we blocked cGAS activation in BMDMs after mtDNA stimulation in vitro. Consistent with the above in vivo findings, cGAS inhibition significantly decreased STING signaling activation in BMDMs after mtDNA activation, as evidenced by Western blot analysis and STING staining (Fig. 3D and E). Collectively, these results indicated that cGAS signalling played an essential role in regulating macrophage STING activation by mtDNA released from hepatocytes.

3.4. XBP1 deficiency facilitated mtDNA extracellular release by inducing hepatocyte pyroptosis

Pyroptosis is a lytic programmed cell death characterised by rapid plasma membrane rupture and the release of proinflammatory intracellular contents. We investigated whether hepatocyte pyroptosis could be induced by TAA and its role in promoting the extracellular release of mtDNA. Primary hepatocytes were isolated from WT and HKO mice and then treated with TAA or LPS combined with nigericin. Both TAA and LPS treatment with nigericin induced primary hepatocytes pyroptosis, which was exacerbated by XBP1 depletion. MCC950, a pyroptosis inhibitor, abrogated the role of hepatocyte XBP1 depletion in aggravating hepatocytes pyroptosis (Fig. 4A). The NLRP3/caspase-1/GSDMD signaling axis of pyroptosis in hepatocytes was also activated by TAA treatment and enhanced by XBP1 depletion (Fig. 4B). Enhanced caspase-1 staining was observed in XBP1 deficient hepatocytes (Fig. 4C). To further determine the functional importance of pyroptosis in regulating the extracellular release of mtDNA, YVAD-cmk was used to block caspase-1 activation and subsequent hepatocellular pyroptosis. As expected, blockade of caspase-1 activation effectively inhibited the cleavage of the N-terminal domain of GSDMD (Fig. 4D). Positive staining of cytosolic dsDNA was observed in the area around the mitochondria after TAA treatment, which was inhibited by the caspase-1 blockade. XBP1 depletion enhanced peri-mitochondrial dsDNA staining and was reversed by caspase-1 blockade (Fig. 4E). Semiquantitative analysis of cytosolic mtDNA established that XBP1 depletion enhanced caspase-1-mediated pyroptosis leading to increased cytosolic mtDNA accumulation (Fig. 4F). Furthermore, increased levels of mtDNA were observed in the supernatant of cultured XBP1-deficient hepatocytes after TAA treatment, which could be reversed by caspase-1 inhibition as well (Fig. 4G). Thus, these results suggested that XBP1 depletion promoted pyroptosis in hepatocytes induced by TAA treatment, facilitating mtDNA cytosolic accumulation and extracellular release.

3.5. Mitochondrial ROS production was essential for XBP1 deficiency to promote hepatocyte pyroptosis

ROS play an important role in regulating TAA-induced liver injury and GSDMD activation. We analyzed the effects of mitochondrial ROS on hepatocyte pyroptosis. Evidently, TAA treatment induced intrahepatic oxidative response as shown by analyzing MDA, GSH/GSSG, SOD, and ROS levels in liver tissues, which were enhanced by XBP1 depletion (Fig. 5A–E). Increased mitochondrial membrane permeability was observed in XPB1-deficient hepatocytes (Fig. 5F). MitoTEMPO, a specific mitochondrial ROS scavenger significantly reduced ROS expression in hepatocytes following TAA treatment (Fig. 5G–I). Furthermore, ROS scavenging by MitoTEMPO inhibited the activation of NLRP3-caspase-1-GSDMD signaling in hepatocytes and abrogated the role of XBP1 depletion in promoting pyroptosis signaling activation (Fig. 5J). These findings suggested that mitochondrial ROS played an essential role in mediating the activation of NLRP3-caspase-1-GSDMD signaling and was essential for XBP1 deficiency to promote pyroptosis in hepatocytes.

3.6. Impaired mitophagy contributed to ROS-mediated pyroptosis of XBP1 deficient hepatocytes and subsequent macrophage STING activation during TAA-induced ALI

The interplay between ER stress and mitophagy has recently been reported. Impaired mitophagy may result in the accumulation of damaged mitochondria and the dysregulation of free radical scavenging. We investigated if increased mitochondrial ROS expression was caused by impaired mitophagy in XBP1 deficient hepatocytes. TAA treatment promoted mitophagy activation, as shown by the enhanced LC3 staining in primary hepatocytes from WT mice. Contrastingly, no significant difference in LC3 staining was observed in hepatocytes from HKO mice (Fig. 6A). LC3 immunofluorescence staining of liver tissues confirmed the results (Supplementary Fig. 5A). Electron microscopy analysis demonstrated that acute TAA treatment induced the appearance of mitophagosomes, and XBP1 depletion blocked mitophagosome formation (Fig. 6B). We observed that there were higher levels of LC3-II and lower levels of p62 in hepatocytes isolated from WT mice post-TAA treatment, compared to corresponding levels in hepatocytes isolated from HKO mice. After TAA treatment, enhanced activation of mitophagy signaling by Pink1 and Parkin was observed in WT hepatocytes after TAA treatment (Fig. 6C). Autophagic flux was then analyzed, and results showed that CQ treatment caused a further increase in LC3-II, only in WT and not in XBP1-deficient hepatocytes after TAA treatment, indicating that XBP1 activation promoted autophagic flux of hepatocytes during TAA-induced injury (Fig. 6D).

To further investigate the role of mitophagy in regulating mtDNA release from hepatocytes and subsequent STING activation of macrophages, CQ was used to block mitophagy activation in vivo. Evidently, mitophagy blockade resulted in an increased oxidative response in WT hepatocytes, whereas no significant effects were found in XBP1 deficient hepatocytes (Fig. 6E–H). An increase in the extracellular release of mtDNA was observed by mitophagy inhibition in WT hepatocytes (Fig. 6I). Enhanced NLRP3-caspase-1-GSDMD activation in hepatocytes, cGAS-STING activation in macrophages, and intrahepatic inflammation were found in WT mice after CQ treatment (Fig. 6J, Supplementary Figs. 5B and C). Mitophagy inhibition also aggravated TAA-induced liver injury in WT mice, as evidenced by serum levels of ALT and AST and pathological examination (Supplementary Figs. 5D and E).

To determine the critical role of PINK1 signaling in regulating mitophagy, we overexpressed PINK1 in XBP1 HKO mice by AAV-PINK1 (Supplementary Fig. 6A). Overexpression of PINK1 increased protein levels of PINK1/Parkin and LC3-II but decreased protein levels of P62 in hepatocytes from XBP1 HKO mice (Supplementary Fig. 6B), leading to attenuated liver injury (Supplementary Figs. 6C–E).

These data suggested that mitophagy inhibition by XBP1 deficiency aggravated the mitochondrial injury and hepatocyte pyroptosis, ultimately leading to increased hepatocellular mtDNA release and enhanced macrophage STING activation during TAA-induced liver injury.

3.7. Activation of XBP1-mediated hepatocellular mitophagy and pyroptosis and macrophage STING signaling pathway in patients with ALI

To test the clinical significance of XBP1 in regulating hepatocellular mitophagy and pyroptosis and macrophage STING activation during ALI, we collected liver tissues from patients with or without ALI, and the above signaling pathways were analyzed. Significantly increased protein levels of XBP1s were observed in livers from patients with ALI (Fig. 7A). ALI induced hepatocellular mitophagy and pyroptosis as indicated by enhanced immunofluorescence staining of LC3, cleaved-caspase-1, and GSDMD-NT in ALB-positive hepatocytes (Fig. 7B). Furthermore, enhanced staining of cGAS, STING, and P-TBK1 was observed in CD68-positive macrophages as well (Fig. 7C). These data suggested the critical role of XBP1 in regulating mitophagy and pyroptosis of hepatocytes and STING signaling of macrophages during human ALI.

3.8. Hepatocyte-specific XBP1 deficiency activated macrophage STING signaling in LPS-induced acute liver injury model

In order to verify the generality of above findings of hepatocyte XBP1 in regulating macrophage STING signaling, another mouse model of liver injury induced by LPS were applied. Similarly, the protein and mRNA levels of XBP1s were increased in the livers after LPS challenge (Supplementary Figs. 7A–B). Hepatocyte XBP1 depletion aggravated LPS-induced liver injury as evidenced by higher levels of serum ALT and AST (Supplementary Fig. 7C), exacerbated architectural destruction (Supplementary Fig. 7D), and increased hepatocellular cell death (Supplementary Fig. 7E). Enhanced activation of macrophages STING (Supplementary Fig. 7F) and decreased activation of hepatocyte mitophagy signaling (Supplementary Fig. 7G) were observed in HKO mice after LPS treatment. Thus, these data suggested that XBP1 deficiency promotes hepatocyte pyroptosis by impairing mitophagy to activate mtDNA-cGAS-STING signaling of macrophages during acute liver injury with different causes.