Hepatocyte-specific TMEM16A deficiency alleviates hepatic ischemia/reperfusion injury via suppressing GPX4-mediated ferroptosis

Figures

• 

  Fig. 1. Hepatic TMEM16A expression is increased during I/R liver injury. Levels of TMEM16A (TM) A mRNA and B protein in liver samples harvested from patients pre- and post-hepatectomy, with β-actin as the inner control (n = 14). **P < 0.01 versus pre-hepatectomy. C Western blot analysis of TMEM16A protein expression in human liver samples obtained post-hepatectomy (n = 30). Pearson correlation analysis between post-hepatectomy hepatic TMEM16A protein expression and D serum ALT levels (r2 = 0.4701, P < 0.01) or E serum AST levels (r2 = 0.4821, P < 0.01). F Hepatic TMEM16A protein levels in mice challenged with the sham operation or 90 min of ischemia followed by reperfusion at the indicated times (n = 4). *P < 0.05, **P < 0.01 versus sham. G Western blot analysis of TMEM16A protein expression in primary hepatocytes subjected to 4 h of hypoxia followed by reoxygenation for the indicated times (n = 4). *P < 0.05, **P < 0.01 versus control. H Representative immunofluorescence of TMEM16A (red) and HNF4 (green) expression in liver sections of mice challenged with the sham operation or I/R (90 min of ischemia and 24 h of reperfusion) (n = 4). Nuclei were stained with DAPI (blue). Representative…

• 

  Fig. 2. Hepatocyte-specific TMEM16A deficiency protects against hepatic I/R injury. A Representative H&E staining and B quantitation of necrotic areas (upper) and Suzuki histological scores (bottom) of liver tissues from TMFlox and TMLKO mice subjected to 90 min of hepatic ischemia followed by reperfusion for the indicated periods (n = 6–8). #P < 0.05 versus corresponding TMFlox. C Serum ALT and AST levels in TMFlox and TMLKO mice from indicated groups (n = 6). #P < 0.05, ##P < 0.01 versus corresponding TMFlox. D Representative H&E staining of liver sections from AAV-Con- or AAV-TM-treated mice subjected to sham operation or 90 min of hepatic ischemia followed by reperfusion for indicated periods. E Quantification of necrotic areas (upper) and Suzuki histological scores (bottom) (n = 6–8). #P < 0.05 versus corresponding AAV-Con. F Serum ALT and AST levels in AAV-Con- and AAV-TM-treated mice from the indicated groups (n = 6). #P < 0.05 versus corresponding AAV-Con. Data were presented as the mean ± SD.

• 

  Fig. 3. Hepatocyte-specific TMEM16A deficiency ameliorates inflammation during hepatic I/R injury. A Representative immunofluorescence staining and quantification of infiltrating CD3+, LY6G+, or CD68+ cells (red) in the livers of TMLKO or AAV-TM mice and their counterpart control mice subjected to 90 min of hepatic ischemia followed by 24 h of reperfusion (n = 6). #P < 0.05 versus TMFlox or AAV-Con. B Flow cytometry histogram showing neutrophils and macrophages in liver samples from indicated mice subjected to 90 min of ischemia and 24 h of reperfusion (n = 6). #P < 0.05, ##P < 0.01 versus TMFlox I/R or AAV-Con I/R. Levels of C serum and D hepatic mRNA levels of pro-inflammatory factors (TNF-α, IL-1β, Il-6, MCP-1, and CXCL-1) in TMLKO or AAV-TM-treated mice and their counterparts subjected to 90 min of hepatic ischemia followed by 24 h of reperfusion (n = 6). **P < 0.01 versus TMFlox sham or AAV-Con sham; #P < 0.05, ##P < 0.01 versus TMFlox I/R or AAV-Con I/R. E TNF-α, MCP-1, and CXCL-1 levels in hepatocytes isolated from TMLKO or AAV-TM-treated mice and their counterpart control mice after H/R challenge (4 h of hypoxia and 12 h of reoxygenation) (n = 6). **P < 0.01 versus TMFlo or…

• 

  Fig. 4. TMEM16A deletion protects against hepatic I/R-induced ferroptosis. A Representative TUNEL immunofluorescence staining and quantification of TUNEL-positive cells in liver sections from TMFlox, TMLKO, AAV-Con-treated, and AAV-TM-treated mice subjected to 90 min of hepatic ischemia followed by 24 h of reperfusion (n = 6). ##P < 0.01 versus TMFlox I/R or AAV-Con I/R. B LDH release and C CCK-8 assay results of hepatocytes isolated from TMLKO mice or AAV-TM-treated mice and their counterpart control mice subjected to 4 h of hypoxia followed by reoxygenation (n = 6). #P < 0.05 versus corresponding TMFlox or AAV-Con. D Hepatic GPX4 protein levels in livers of mice challenged with the sham operation or 90 min of ischemia followed by 24 h of reperfusion (n = 6). **P < 0.01 versus TMFlox sham or AAV-Con sham; #P < 0.05 versus TMFlox I/R or AAV-Con I/R. E Representative transmission electron microscopy images and quantification of mitochondria length in primary hepatocytes isolated from indicated groups after 4 h of hypoxia and 12 h of reoxygenation (n = 6). ##P < 0.01 versus TMFlox H/R or AAV-Con H/R. F Representative images of 4-HNE immunohistochemical staining in liver samples from mice…

• 

  Fig. 5. TMEM16A regulates hepatic I/R injury in a GPX4-dependent manner. A Lipid ROS levels of primary hepatocytes isolated from indicated mice treated with RSL3 (3 μmoL/L) for 8 h followed by 4 h of hypoxia and 12 h of reoxygenation (n = 6). Red, reduced form of C11-BODIPY; a green, oxidized form of C11-BODIPY. *P < 0.05, **P < 0.01 versus TMFlox DMSO; ##P < 0.01 versus TMLKO DMSO. B Levels of pro-inflammatory factors (TNF-α, MCP-1, and CXCL-1) and C LDH release in indicated hepatocytes treated with RSL3, Erastin (10 μmoL/L), or Ferrostatin-1 (Fer, 2 μmoL/L) for 8 h followed by 4 h of hypoxia and 12 h of reoxygenation (n = 6). *P < 0.05, **P < 0.01 versus TMFlox DMSO or AAV-Con DMSO; ##P < 0.01 versus TMLKO DMSO or AAV-TM DMSO. D Representative 4-HNE immunohistochemical staining of liver samples from TMLKO and TMFlox mice injected with AAV-GPX4 shRNA via tail vein two weeks prior to I/R surgery (n = 6). E Representative H&E staining of liver sections, necrotic area quantification (area outside the contour indicates the injured part in the last two pictures on the right side), and F serum ALT and G AST levels in indicated mice (n = 6). *P < 0.05, **P < 0.01 versus TMFlox AAV-Scr shRNA;…

• 

  Fig. 6. TMEM16A directly binds with GPX4 and facilitates its ubiquitination. A Western blotting results of GPX4 protein expression in hepatocytes treated with 4 h of hypoxia and 12 h of reoxygenation, followed by incubation with cycloheximide (CHX, 100 µg/mL) for the indicated durations (n = 6). #P < 0.05, ##P < 0.01 versus TMFlox H/R or AAV-Con H/R. B GPX4 protein expression in hepatocytes isolated from AAV-TM-treated mice after 4 h of hypoxia and 12 h of reoxygenation, followed by treatment with MG-132 (10 µmol/L) or chloroquine (CQ, 20 µmol/L) for 4 h (n = 6). GPX4 protein expression in hepatocytes isolated from C TMLKO and D AAV-TM-treated mice and their control littermates after 4 h of hypoxia and 12 h of reoxygenation (n = 4). Immunoprecipitation was performed with an anti-GPX4 antibody followed by western blot analysis of GPX4 ubiquitination. E Flag and HA expression in HEK293T cells co-transfected with vectors expressing Flag-TMEM16A and HA-GPX4 (n = 6). Immunoprecipitation (IP) was performed with an anti-Flag antibody (upper panel) or anti-HA antibody (lower panel), followed by western blotting with an anti-HA antibody or anti-Flag antibody. F Results of IP assay showing…

• 

  Fig. 7. The interaction between TMEM16A and GPX4 is indispensable for TMEM16A function in hepatic I/R injury. A Representative immunohistochemical staining of 4-HNE in liver sections of AAV-Con-, AAV-TM-, and AAV-TM-M-treated mice subjected to 90 min of hepatic ischemia followed by 24 h of reperfusion. Hepatic levels of B MDA, C PTGS2, and D serum iron content (n = 6). **P < 0.01 versus AAV-Con I/R; ##P < 0.01 versus AAV-TM I/R. E Representative H&E staining, necrotic area quantification, and Suzuki histological score in liver samples, and F serum ALT and AST levels in indicated mice followed by 90 min of ischemia and 24 h of reperfusion (n = 6–8). **P < 0.01 versus AAV-Con I/R; ##P < 0.01 versus AAV-TM I/R. G Western blot analysis and quantification of levels of hepatic GPX4 protein and proteins related to the NF-κB signaling pathway from the indicated groups (n = 6). *P < 0.05, **P < 0.01 versus AAV-Con I/R; ##P < 0.01 versus AAV-TM I/R. Data were presented as the mean ± SD.