Resolvin D1 attenuates liver ischaemia/reperfusion injury through modulating thioredoxin 2-mediated mitochondrial quality control

Abstract

Background and purpose: Liver ischaemia and reperfusion (IR) injury is a sterile inflammatory response involving production of ROS. Mitochondrial homeostasis is maintained by mitochondrial quality control (QC). Thioredoxin (TRX) 2 is a key mitochondrial redox-sensitive protein. Resolvin D1 (RvD1), a specialized pro-resolving lipid mediator, exerts anti-inflammatory and antioxidant activities. We investigated mechanisms of RvD1 protection against IR-induced oxidative damage to the liver, focusing on TRX2-mediated mitochondrial QC.

Experimental approach: Mice underwent partial warm IR. RvD1 was administered 1 h before ischaemia and immediately prior to reperfusion. Human liver carcinoma HepG2 cells were exposed to hypoxia/reoxygenation and transfected with TRX2 siRNA. Immunohistochemistry, Western blotting and enzyme assays were used to follow changes in mitochondrial structure and function.

Key results: RvD1 attenuated hepatocellular damage following IR, assessed by serum aminotransferase activities and histology. RvD1 reduced mitochondrial swelling, lipid peroxidation and glutamate dehydrogenase release. Impaired activities of mitochondrial complexes I and III were restored by RvD1. RvD1 enhanced expression of the mitophagy-related protein, Parkin and inhibited accumulation of PTEN-induced putative kinase 1. RvD1 restored levels of mitochondrial biogenesis proteins including PPARγ coactivator 1α, nuclear respiratory factor 1 and mitochondrial transcription factor A and mtDNA level. RvD1 attenuated the increase in levels of the mitochondrial fission-related protein, dynamin-related protein 1. IR reduced TRX2 levels while increasing TRX2 association with TRX-interacting protein. RvD1 attenuated these changes. The regulatory effects of RvD1 on mitochondrial QC were abolished by TRX2 knockdown.

Conclusions and implications: We suggest that RvD1 ameliorated IR-induced hepatocellular damage by regulating TRX2-mediated mitochondrial QC.

Figure 1

Effect of RvD1 on hepatocellular damage in liver IR. Mice received i.p. injections of RvD1 (15 μg·kg⁻¹) 1 h prior to ischaemia and again immediately before reperfusion. (A) Serum ALT and AST levels were measured by spectrophotometric assay. (B) Histological lesions were graded using Suzuki score. The necrotic area and inflammatory cell infiltration are indicated by arrows. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the vehicle‐treated sham group; ⁺ P < 0.05, significantly different from the vehicle‐treated IR group.

Figure 2

Effect of RvD1 on mitochondrial dysfunction in liver IR. Mice received i.p. injections of RvD1 (15 μg·kg⁻¹) 1 h prior to ischaemia and again immediately before reperfusion. (A) Mitochondrial swelling, (B) serum GDH activity, (C) mitochondrial respiratory complex I and III activities and (D) hepatic ATP level were measured. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the vehicle‐treated sham group; ⁺ P < 0.05, significantly different from the vehicle‐treated IR group.

Figure 3

Effect of RvD1 on mitochondrial oxidative stress in liver IR. Mice received i.p. injections of RvD1 (15 μg·kg⁻¹) 1 h prior to ischaemia and again immediately before reperfusion. (A) Mitochondrial lipid peroxidation, (B) MnSOD protein level in whole liver extracts and (C) GSH/GSSG ratio were measured. Protein expression was adjusted to β‐actin as a loading control. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the vehicle‐treated sham group; ⁺ P < 0.05, significantly different from the vehicle‐treated IR group.

Figure 4

Effect of RvD1 on mitochondrial QC‐related protein levels in liver IR. Mice received i.p. injections of RvD1 (15 μg·kg⁻¹) 1 h prior to ischaemia and again immediately before reperfusion. Western blot analysis was performed to measure the protein levels of (A) PINK1 and Parkin, (B) PGC1α, NRF1 and TFAM, and (C) DRP1 and MFN2 in whole liver extracts. Protein expression was adjusted to β‐actin as a loading control. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the vehicle‐treated sham group; ⁺ P < 0.05, significantly different from the vehicle‐treated IR group.

Figure 5

Effect of RvD1 on TRX2 protein level and TRX2‐TXNIP association in liver IR. Mice received i.p. injections of RvD1 (15 μg·kg⁻¹) 1 h prior to ischaemia and again immediately before reperfusion. Western blot analysis was performed to measure the levels of (A, B) TRX2, TXNIP protein, (C) association of TRX2‐TXNIP and (D) TRX1 protein in whole liver extracts. Protein expression was adjusted to β‐actin as a loading control. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the vehicle‐treated sham group; ⁺ P < 0.05, significantly different from the vehicle‐treated IR group.

Figure 6

Effect of TRX2 gene silencing on the regulatory effect of RvD1 on inflammatory response. HepG2 cells were transfected with TRX2 siRNA 24 h before H/R. HepG2 cells were treated with 0.05 μM of RvD1 30 min before H/R. (A) TRX2 protein expression was measured to confirm the efficacy of TRX2 siRNA. Western blot analysis was performed to measure the protein levels of (B) TNF‐α and (C) IL‐6 in cell lysates. Protein expression was adjusted to β‐actin as a loading control. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the control group; ⁺ P < 0.05, significantly different from the H/R group; ^# P < 0.05, significantly different from the RvD1‐treated H/R group.

Figure 7

Effect of TRX2 gene silencing on the regulatory effect of RvD1 on mitochondrial QC. HepG2 cells were transfected with TRX2 siRNA 24 h before H/R. HepG2 cells were treated with 0.05 μM of RvD1 30 min before H/R. Western blot analysis was performed to measure the protein levels of (A, B) PINK1 and Parkin, (C–E) PGC1α, NRF1 and TFAM, and (F) DRP1 in cell lysates. Protein expression was adjusted to β‐actin as a loading control. All values are expressed as mean ± SEM (n = 8 per group). ^* P < 0.05, significantly different from the control group; ⁺ P < 0.05, significantly different from the H/R group; ^# P < 0.05, significantly different from the RvD1‐treated H/R group.