20-HETE synthesis inhibition attenuates traumatic brain injury-induced mitochondrial dysfunction and neuronal apoptosis via the SIRT1/PGC-1α pathway: A translational study

Abstract

Objectives: 20-hydroxyeicosatetraenoic acid (20-HETE) is a metabolite of arachidonic acid catalysed by cytochrome P450 enzymes and plays an important role in cell death and proliferation. We hypothesized that 20-HETE synthesis inhibition may have protective effects in traumatic brain injury (TBI) and investigated possible underlying molecular mechanisms.

Materials and methods: Neurologic deficits, and lesion volume, reactive oxygen species (ROS) levels and cell death as assessed using immunofluorescence staining, transmission electron microscopy and Western blotting were used to determine post-TBI effects of HET0016, an inhibitor of 20-HETE synthesis, and their underlying mechanisms.

Results: The level of 20-HETE was found to be increased significantly after TBI in mice. 20-HETE synthesis inhibition reduced neuronal apoptosis, ROS production and damage to mitochondrial structures after TBI. Mechanistically, HET0016 decreased the Drp1 level and increased the expression of Mfn1 and Mfn2 after TBI, indicating a reversal of the abnormal post-TBI mitochondrial dynamics. HET0016 also promoted the restoration of SIRT1 and PGC-1α in vivo, and a SIRT1 activator (SRT1720) reversed the downregulation of SIRT1 and PGC-1α and the abnormal mitochondrial dynamics induced by 20-HETE in vitro. Furthermore, plasma 20-HETE levels were found to be higher in TBI patients with unfavourable neurological outcomes and were correlated with the GOS score.

Conclusions: The inhibition of 20-HETE synthesis represents a novel strategy to mitigate TBI-induced mitochondrial dysfunction and neuronal apoptosis by regulating the SIRT1/PGC-1α pathway.

Keywords: 20-HETE; SIRT1; apoptosis; mitochondrial dysfunction; traumatic brain injury.

FIGURE 1

Non‐targeted metabolic profiling of damaged brain tissue after TBI. A, B, The clustering of orthogonal partial least‐squares discriminant analysis (OPLS‐DA). C, Volcano plot of up‐ and downregulated metabolites in sham and TBI groups. D, Heat map of the top 20 up‐ or downregulated metabolites in the two groups. E, Twenty‐seven KEGG pathways were significantly different between the two groups. n = 6 for each group

FIGURE 2

Induction of CYP4A expression and increase in 20‐HETE after TBI. A, B, Western blots and analysis of CYP4A levels in the perilesional cortex after TBI. C, Concentration of free 20‐HETE in perilesional cortex after TBI measured using ELISA. D, Distribution of NeuN⁺/CYP4A⁺ cells in the perilesional cortex 24 h after TBI. Green, CYP4A; red, NeuN. Scale bars: 100 μm. E, Quantitative analysis of fluorescence intensity of CYP4A (fold over Sham). n = 6 for each group. *P < .05 and **P < .01 vs sham group. Values are presented as the mean ± SD

FIGURE 3

HET0016 reduces lesion volume and improves neurological outcome after TBI. A, The concentration of free 20‐HETE in the perilesional cortex after treatment with various doses of HET0016 (1, 1.5 and 2 mg/kg) 24 h after TBI. n = 6 for each group. B, Nissl staining of brain sections after TBI. n = 6 for each group. C, Intraperitoneal injection of HET0016 (1.5 mg/kg) reduced lesion volume. n = 6 for each group. D, HET0016 reduced brain oedema 24 h after TBI. n = 6 for each group. E, Kaplan‐Meier survival curves indicating lack of significant differences between the vehicle and HET0016 treatment groups. F, HET0016 reduced the neurological deficit score on day 3 and day 7 after TBI. n = 9 for each group. G, HET0016 improved the corner turn test performance of mice on day 3 and day 7 after TBI. n = 12 for each group. H, HET0016 increased latency to fall in the wire‐hanging test on day 3 and day 7 after TBI. n = 12 for each group. **P < .01 vs sham group, ^& P < .05 and ^&& P < .01 vs TBI + vehicle group. Values are presented as the mean ± SD

FIGURE 4

HET0016 alleviates oxidative stress and neural apoptosis after TBI. A, Representative images of DHE staining of the perilesional cortex 24 h after TBI, indicating the ROS levels. Scale bar: 100 μm. B, C, The effects of HET0016 on mitochondrial MnSOD activity and MDA levels. D, E, Representative images and statistical analysis of TUNEL staining of the perilesional cortex 24 h after TBI. Scale bar: 100 μm. F, G Representative Western blots and statistical analysis of the levels of Nrf2, Bax, Bcl‐2 and cleaved caspase‐3. n = 6 for each group. **P < .01 vs sham group, ^& P < .05 and ^&& P < .01 vs TBI + vehicle group. Values are presented as the mean ± SD

FIGURE 5

Effects of HET0016 on the ultrastructure of neurons in the perilesional cortex 24 h after TBI, as detected using electron microscopy. A–D, Representative ultrastructure of neurons in each group. A1‐D1, Magnified images of the ultrastructure of neurons shown in panels A‐D. Scale bar: 1 μm (A–D) and 0.25 μm (A1‐D1). n = 6 for each group

FIGURE 6

HET0016 reverses mitochondrial dysfunction via the SIRT1/PGC‐1α signalling pathway after TBI in vivo. A, C, Western blots and statistical analysis of the levels of Drp1, Mfn1 and Mfn2 in each group. B, C, Western blots and statistical analysis of the levels of mitochondrial and cytoplasmic cytochrome c in each group. D, Normalized ATP levels. E, Mitochondrial complex I, II, III and IV activities (percentage of sham). F, G, Western blots and statistical analysis of the levels of SIRT1 and PGC‐1α in each group. n = 6 for each group. *P < .05 and **P < .01 vs sham group, ^& P < .05 and ^&& P < .01 vs TBI + vehicle group. Values are presented as the mean ± SD

FIGURE 7

20‐HETE alters normal mitochondrial morphology and function via the SIRT1/PGC‐1α pathway in vitro. A, C, Western blots and statistical analysis of the levels of SIRT1 and PGC‐1α in each group. B, C, Western blots and statistical analysis of the levels of Drp1, Mfn1 and Mfn2 in each group. D, E Western blots and statistical analysis of the levels of mitochondrial and cytoplasmic cytochrome c in each group. F, Representative MitoTracker fluorescence images illustrating mitochondrial morphology. Scale bar: 10 μm. G, Representative MitoSOX fluorescence images of mitochondria‐derived ROS. Scale bar: 50 μm. H, Representative fluorescence staining of JC‐1 aggregates (red)/JC‐1 monomers (green) illustrating the MMP. Scale bar: 10 μm. n = 6 for each group. **P < .01 vs control group, ^& P < .05 and ^&& P < .01 vs 20‐HETE + vehicle group. Values are presented as the mean ± SD

FIGURE 8

Plasma 20‐HETE levels in patients with TBI correlate with long‐term outcome. A, Plasma 20‐HETE levels are increased in patients with poor outcome compared to those in patients with good outcome (P < .001). B, ROC analysis of the discriminative ability of plasma 20‐HETE levels for risk of poor outcome in patients with TBI. Area under the curve: 0.799; optimal cut‐off point: 254.26 pg/mL; sensitivity: 81.1%; specificity: 77.1%. C, Linear correlation between plasma 20‐HETE levels and short‐term neurological outcome assessed using GOS scores of patients with TBI. r = −.488, P < .001. Values are presented as the mean ± SD

FIGURE 9

The proposed mechanism of 20‐HETE–induced alterations of mitochondrial dynamics in neurons. 20‐HETE is synthesized from AA by CYP4A, which is inhibited by HET0016. The altered mitochondrial dynamics induced by 20‐HETE activate oxidative stress and subsequently result in neuronal apoptosis mediated by inhibition of the SIRT1/PGC‐1α signalling pathway