Orexin-A alleviates ferroptosis by activating the Nrf2/HO-1 signaling pathway in traumatic brain injury

Reagents and materials

OXA was purchased from MedChemExpress (Monmouth Junction, NJ, USA), western blot and immunofluorescence primary antibodies from Abcam (Boston, MA, USA), and secondary antibodies from Proteintech (Wuhan, China), the malondialdehyde (MDA) content detection kit, superoxide dismutase (SOD) activity detection kit, and reduced glutathione (GSH) content detection kit from Solarbio Life Sciences (Beijing, China), Prussian blue staining kit from Servicebio (Wuhan, China), and hematoxylin and eosin (HE) staining kit from Leagene Biotechnology (Beijing, China).

Animals

Animal experiments were all checked with approval from Ethics Committee of the First Affiliated Hospital of Nanchang University (Approval No. CDYFY-IACUC-202301QR057). Animal procedures were all performed strictly according to National Institutes of Health’s Guide for the Care and Use of Laboratory Animals. Specific Pathogen Free (SPF) male adult Sprague Dawley (SD) rats (6–8 weeks, weigh 250–300 g) supplied by Zhejiang Weitong Lihua Experimental Animal Technology Co., Ltd. (License No. SCXK (Zhejiang) 2019-0001) were selected for the experiment. Experimental animals were all placed in the same condition (room temperature 25°C, humidity 50%, 12-hour light/dark cycling) with water and food freely, and 7-day adaptation to the laboratory condition prior to the experiment.

Construction of the TBI rat model and intracerebroventricular (ICV) injection

TBI rat model was constructed via a modified Feeney’s approach [22]. First, the animals experienced anesthesia with pentobarbital sodium (50 mg/kg, intraperitoneal injection) with their heads fixed in stereotactic frame, and a constant temperature heating pad was utilized for maintaining the body temperature. After disinfection of surgical area, a median incision (10 mm length) was created on scalp, and skin and periosteum were stripped. Then, at 3 mm posterior to coronal suture and 2.5 mm to the left of midline, one round bone window (5 mm diameter) was drilled with a cranial drill. The cranial disc was carefully removed without damaging the dura mater. Next, a T-shaped lance was placed outside the dura mater, and a 40 g weight was allowed to free-fall along the guide tube from 20 cm height to hit T-shaped lance, resulting in a compression depth of 2–3 mm without penetrating the dura mater, causing local brain contusion and laceration. Immediately after the impact, the T-shaped lance was removed from the injury position to avoid secondary injuries. The sham group experienced identical treatment as TBI group except for no impact on cerebral cortex. The stereotactic coordinates of the lateral ventricle were as follows: with the location of the bregma as the center, 1.5 mm to the side, 1.1 mm to the direction of posterior fontanelle, and a depth of 4.5 mm. After TBI modeling was completed, 10 μl of OXA (OXA dissolved in saline under concentration of 0.75 μg/μL, with an injection dose of 30 μg/kg) or 10 μL of 0.9% NaCl was injected with a speed of 2 μL/min.

Experimental design

There were three parts to the experimental design of this study.

Part 1: First, we aimed at confirming expression trend of ferroptosis-associated molecules at diverse time points after TBI. The rats were haphazardly grouped into (n = 6 in each group): sham, 12 h, 24 h, 48 h and 72 h groups. Western blot was then performed.

Part 2: For exploring OXA influence on ferroptosis and neuronal damage after TBI in SD rats and its related pathways, rats were haphazardly grouped into (n = 6 in each group): sham, TBI, TBI+OXA, and TBI + normal saline (NS) groups. Western blot, real-time PCR, iron assay, Perls’ blue, Nissl, HE and FJB staining, immunohistochemistry, mNSS score test, were evaluated at indicated time-points following TBI.

Part 3: To further demonstrate that the biological effects of OXA depend on the Nrf2/HO-1 signaling pathway and to compare the effects of OXA and Lip-1, we used ML385 to inhibit the action of Nrf2. The rats were randomly divided into 5 groups (n = 6 in each group): TBI group, OXA group, Lip-1 group, OXA+ML385 group, Lip-1+ML385 group. Western blot and Hematoxylin-eosin (HE) staining were then performed.

After TBI modeling was completed, 10 μl of OXA (OXA dissolved in saline, with an injection dose of 30 μg/kg) was injected with a speed of 2 μL/min. 1 hour before modeling, the Nrf2 inhibitor ML385 (ML385 dissolved in DMSO, with an injection dose of 30 mg/kg) or equivalent volume of DMSO was injected intraperitoneally. 1 hour after TBI, the inhibitor of ferroptosis Lip-1 (Lip-1 dissolved in DMSO, with an injection dose of 10 mg/kg) or equivalent volume of DMSO was injected intraperitoneally.

Quantitative polymerase chain reaction (qPCR)

Total RNA was extracted via TRizol reagent (Invitrogen, Carlsbad, CA, USA), and 800 ng of total RNA was reverse-transcribed into cDNA via Takara PrimeScript RT kit (Takara, Kyoto, Japan). Subsequently, qPCR was done via CFX96 Real-Time PCR Detection System. The mRNA levels were normalized to GAPDH to internal control. The mRNA expression was quantified using the 2^−ΔΔCt method. The calculation formula is as follows: ΔΔCt = ΔCt (experimental group) −ΔCt (control group); ΔCt (experimental group) = Ct (the target gene of the experimental group) −Ct (internal reference gene in the experimental group); ΔCt (control group) = Ct (the target gene of the control group) −Ct (internal reference gene in the control group); Fold Change = 2^−ΔΔCt. We used one-way ANOVA to compare the statistical differences between the groups of samples. Primer sequence is detailed in Table 1.

Biochemical assays

A lipid peroxidation, GSH, SOD, ROS, and tissue iron assay kits were used to evaluate the relative MDA, GSH, SOD, ROS, and iron ion concentrations in the cortex around the injury area, respectively. MDA levels were shown as nmol/mg protein, GSH levels as μMol/g protein, SOD levels as U/mg protein, ROS levels as fluorescence intensity/mg protein, and iron ion levels as μMol/g protein.

Hematoxylin-eosin staining

Hematoxylin-eosin (HE) staining mainly observes the internal morphological structure of cells. The hematoxylin staining solution is alkaline, mainly staining the chromatin in the nucleus and the nucleic acids in the cytoplasm with purple-blue color. Eosin is an acidic dye, mainly staining the components of the cytoplasm and extracellular matrix with red color [23]. At 24 h after TBI, after completing behavioral testing, rats experienced anesthesia with pentobarbital sodium (50 mg/kg, intraperitoneal injection). Brain tissues were separated using cardiac perfusion followed by 24-hour fixing with 4% paraformaldehyde under ordinary temperature. After gradient dehydration with 75%, 95%, and 100% ethanol, the tissues were embedded in paraffin and cut into 5 μm-thick slices. Then, paraffin-embedded brain tissues experienced 6-minute and 3-minute staining by hematoxylin and eosin separately. Images were acquired via optical microscope.

Nissl staining

Nissl body is distributed in the cytoplasm of neurons. Nissan contains RNA and is easily stained with alkaline dyes such as toluidine blue and tar violet [24]. Paraffin-embedded tissues were sliced into 5-μm-thick sections, experiencing dewaxing in xylene, rehydration in 100%, 95%, and 75% ethanol, 3-minute staining in toluidine blue and twice washing by distilled water (several seconds each time). After gradient dehydration with 75%, 95%, and 100% ethanol, slices experienced clearing in xylene and sealing by neutral resin before monitoring, imaging, and analysis via microscope.

Perls staining

Perls stain, also known as the hemin stain, is a very classic histochemical reaction and a sensitive and traditional method for displaying trivalent iron in tissues. The principle of the staining process is that potassium ferrocyanide solution separates trivalent iron ions from proteins with dilute hydrochloric acid, and the trivalent iron reacts with potassium ferrocyanide to form an insoluble blue compound, namely trivalent iron ferrocyanide, also known as Prussian blue [25]. Paraffin-embedded slices experienced dewaxing and rehydration as section 2.8 and were subjected to Perls staining in the dark using a Prussian blue staining kit. The sections were sequentially incubated with the Perls working solution, incubation solution, and enhancing solution and washed thrice with distilled water. After dehydrating in gradient ethanol solutions, slices experienced clearing in xylene, sealing by neutral resin and observation via optical microscope.

Immunohistochemistry

Paraffin slices were prepared, dewaxed, and rehydrated as in section 2.12 and experienced 30-minute immersing in preheated blocking permeating solution (40 ml PBS, 120 μl TritonX-100 and 400 μl 30% H₂O₂) (RT avoiding light) for reducing activity of endogenous peroxidase. Slices experienced thrice washing with PBS, immersing in 0.01 M sodium citrate buffer (pH 6.0) and 4-minute heating in microwave oven on high heat until boiling followed by being removed and cooled to room temperature which were repeated twice before slices experienced thrice washing by PBS. Slices experienced 30-minute incubation by normal goat serum under 37°C, and then incubation by anti-GPX4 (1:2000, Abcam, ab71495) and anti-prostaglandin peroxidase synthase 2 (PTGS2) (1:50, Abcam, ab9787) antibodies all night under 4°C. Following PBS washing, slices experienced 2-hour incubation by secondary antibody (goat anti-rabbit, 1:5,000, Zsgb Bio, China) under ordinary temperature. Images were acquired via optical microscope.

Western blotting

After the behavioral assessment, the contusion cortex proteins around the injury edge were extracted (3 × 3 × 5 mm (width × deep × length)) and protein concentration was confirmed via Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA). Protein levels of ferroptosis-associated proteins GPX4, PTGS2, SLC7A11, and pathway-related proteins HO-1, NQO1, and Nrf2 were determined using western blotting. The primary antibodies for western blotting included anti-GPX4 (1:1000, Abcam, ab32503), anti-PTGS2 (1:1000, Abcam, ab59348), anti-SLC7A11 (1:1000, Cell Signaling Technology, USA, 8242), anti-NRF2 (1:500, Abcam, ab214185), anti-HO-1 (1:500, Abcam, ab175449), anti-NQO1 (1:1000, Novusbio, USA, NB10056565), as well as anti-GAPDH (1:2000, Proteintech, 60004-1-lg). After sodium dodecyl sulfate-polyacrylamide gel electrophoresis, proteins of all samples were transferred to nitrocellulose filter membrane. For binding primary antibody to secondary antibody, membrane experienced incubation by secondary antibody horseradish peroxidase-labeled goat anti-rabbit or goat anti-mouse for 60 min under ordinary temperature. Western blot results were analyzed using enhanced chemiluminescence.

Behavioral test

Before treatment with OXA and at 24 h after the injury, two experimentally blind individuals evaluated the rats using modified neurological severity score (mNSS). mNSS test owns ten diverse components to evaluate rats’ motor functions (muscle statuses and unusual movements), sensation (vision, touch, and proprioception), balance and reflex function. The score ranged from 0 to 18 (0, normal; 1–6, mild injury; 7–12, moderate injury; 13–18, serious damage; and 18, most severe neurological deficits).

Statistical analysis

Continuous data (normal distribution) are shown as mean ± standard deviation. Great data difference among groups was confirmed through one-way analysis of variance with Tukey’s post hoc analysis. For mNSS data (non-normal distribution), great difference among groups was evaluated via Kruskal–Wallis test. Statistical analysis was done via GraphPad Prism 8.0 software. P < 0.05 was of statistical significance.

Data availability statement

Data utilized for supporting research discoveries remain obtainable from corresponding author as required.

Changes in ferroptosis-associated proteins after TBI over time

For investigating ferroptosis function in pathophysiological process of TBI, expression levels of ferroptosis-associated proteins in the contused brain tissues around the edge of the injury at diverse time points were detected via western blotting. The expressions of SLC7A11 and GPX4 were analyzed. At 24 h after TBI, the overall trend of SLC7A11 and GPX4 decreased at first and then increased. It is worth noting that it was at the 24 h time point after TBI at which all these proteins showed statistical differences (GPX4: F (4,10) = 63.91, P < 0.001; SLC7A11: F (4,10) = 20.67, P < 0.001, Figure 1A–1D). Oxidative stress is an important pathophysiological process in the development of TBI. GPX4 can protect cell membranes and mitochondria from oxidative damage. Downregulation of SLC7A11 can indirectly inhibit the activity of GPX4 by inhibiting the cysteine metabolic pathway, leading to decreased intracellular cysteine levels and GSH biosynthesis depletion, resulting in accumulation of lipid peroxides and ultimately inducing cell death through ferroptosis. In this study, it was found that both GPX4 and SLC7A11 decreased at 24 hours after TBI, indicating that the degree of ferroptosis was most pronounced at this time and oxidative stress was most severe. Subsequently, as vascular spasm resolved, GPX4 and SLC7A11 began to rise.

Figure 1. The trend of expression of SLC7A11 and GPX4 in injured peripheral cortex of TBI rats over time. (A, B) Representative WB bands of GPX4, SLC7A11, β-Tubulin and GAPDH in the injured peripheral cortex at a specified time point. (C, D) Comparison of relative expression levels of GPX4 and SLC7A11 in injured peripheral cortex of each rat group. Findings are shown as mean ± standard deviation. ^**P < 0.01 vs. sham, ^***P < 0.001 vs. sham.

OXA attenuates TBI-induced ferroptosis

First, we explored the effect of OXA on the markers of ferroptosis in the sham group. It was found that there was no significant change in the ferroptosis marker proteins in the sham group after intervention with OXA. (GPX4: F (2, 2) = 17.75, P = 0.1067; SLC7A11: F (2, 2) = 13.58, P = 0.1371, Figure 2A–2C). To determine whether OXA can rescue TBI-induced ferroptosis, we conducted Perls staining on tissue slices 24 h after TBI and determined iron content in cortex via tissue iron assay kit. Perls staining showed iron deposition around the injured brain tissues (Figure 3A). The findings revealed number of iron-positive cells observed in OXA treatment group was less relative to TBI and saline treatment groups (Fe²⁺: F (3, 8) = 9.161, P < 0.05, Figure 3B). In addition, immunohistochemical staining revealed relative to TBI and saline treatment groups, OXA treatment diminished PTGS2 expression and raised GPX4 expression in cortex 24 h following TBI (Figure 3C). Excessive lipid peroxidation, the primary biological characteristic of ferroptosis, has a close relationship with oxidative stress injury as well as ROS production. Therefore, we used biochemical analysis to evaluate the levels of MDA, GSH, SOD, and ROS in TBI rats’ brain tissue. Findings revealed relative to sham group, TBI and saline treatment groups exhibited significantly raised levels of MDA and ROS but significantly diminished levels of GSH and total SOD activity, indicating significant oxidative stress damages to brain tissue. OXA could reverse these changes (GSH: F (3, 8) = 15, P < 0.01, ROS: F (3, 8) = 12, P < 0.01, SOD: F (3, 8) = 21.33, P < 0.01, MDA: F (3, 8) = 16, P < 0.01, Figure 3D–3G), indicating that OXA could effectively protect brain tissue from oxidative stress damage after TBI. Next, we continued to measure the protein and mRNA levels of several ferroptosis biomarkers through western blotting and qPCR, including SLC7A11, PTGS2, and GPX4. Findings of western blot analysis coincided with qPCR. We found that relative to sham group, TBI and TBI+NS groups exhibited significantly increased mRNA and protein levels of PTGS2 but significantly decreased those of SLC7A11 and GPX4 at 24 h (protein:GPX4: F (3, 8) = 17.35, P < 0.001; SLC7A11: F (3, 8) = 24.62, P < 0.001; PTGS2: F (3, 8) = 11.95, P < 0.001; mRNA: GPX4: F (3, 8) = 53, P < 0.001; SLC7A11: F (3, 8) = 12, P < 0.001; PTGS2: F (3, 8) = 47, P < 0.001, Figure 4A–4G). OXA administration after TBI reversed these changes. In addition, to compare the effects of OXA and ferroptosis inhibitor, we used Lip-1 to intervene in TBI rats. The results showed that there was no significant difference in the inhibitory effect of OXA and Lip-1 on ferroptosis (SLC7A11, F (4, 10) = 82.23, P < 0.001; PTGS2, F (4, 10) = 11.05, P < 0.01; Figure 2D–2F). To further compare the protective effects of OXA and Lip-1 on the brain, we performed HE staining and found that the protective effect of OXA was better than that of Lip-1 (Figure 2G). In conclusion, above discoveries advice TBI results in changes in biomarkers of ferroptosis in injured peripheral cortex, while OXA treatment can effectively alleviate TBI-induced ferroptosis.

Figure 2. The effect of OXA on the sham group rats and the effect of ML385 on OXA. (A–C) The protein expression levels of GPX4 and SLC7A11 in sham group rats after intervention with OXA. (D–F) Comparison of the effects of OXA and Lip-1 on ferroptosis -related proteins SLC7A11 and PTGS2. (G) Typical HE staining images: comparison of the protective effects of OXA and Lip-1 on the brain. Statistical data are shown in mean ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001.

Figure 3. OXA alleviates TBI induced Ferroptosis and oxidative stress. (A) Perls staining showed iron deposition around the injured brain tissues of each rat group 24 hours after TBI. (B) Cortical non-heme iron was measured in sham, TBI, TBI+OXA, and TBI+NS groups. (C) Representative immunohistochemical staining of GPX4 and PTGS2. (D–G) Biochemical analysis was used to assess contents of GSH, SOD, MDA, and ROS in injured peripheral cortex of each group 24 hours following TBI. Findings are shown as mean ± standard deviation. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^nsp > 0.05.

Figure 4. Changes in markers related to ferroptosis following OXA treatment. (A) Expression of GPX4, SLC7A11, and PTGS2 in injured peripheral cortex of rats following TBI was found via western blotting. (B–D) Western blot quantitative data for different groups. (E–G) Relative mRNA levels of GPX4, SLC7A11, and PTGS2 in injured peripheral cortex of rats in each group 24 hours after TBI. Findings are shown as mean ± standard deviation. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001.

OXA alleviates ferroptosis in TBI rat models via activating the Nrf2/HO-1 pathway

Nrf2/HO-1 is main signaling pathway in ferroptosis regulation. For elucidating possible mechanism of OXA’s protective functions, western blot analysis was utilized for measuring relative protein expression of Nrf2/HO-1 signaling pathway in surrounding tissues of the injury 24 h after TBI. Our results revealed relative to those in sham group, levels of Nrf2, NQO1, and HO-1 significantly diminished 24 h after TBI, and there existed nongreat difference between TBI and TBI+NS groups. OXA treatment increased expression of Nrf2, HO-1, and NQO1 after TBI (Nrf2: F (3, 8) = 19.89, P < 0.001; NQO1: F (3, 8) = 21.77, P < 0.001; HO-1: F (3, 8) = 18.66, P < 0.001; Figure 5A–5D). In addition, to further demonstrate that OXA relies on the Nrf2/HO-1 signaling pathway to alleviate ferroptosis, we also used the Nrf2 inhibitor ML385 to intervene in TBI rats. The results showed that the effects of both OXA and Lip-1 could be blocked by ML385, but ML385 had a stronger blocking ability on Lip-1 (SLC7A11, F (4, 10) = 82.23, P < 0.001; PTGS2, F (4, 10) = 11.05, P < 0.01; Figure 2D–2F). Therefore, we concluded that OXA may restrain ferroptosis through Nrf2/HO-1 signaling pathway activation.

Figure 5. OXA influences on Nrf2/HO-1 signal pathway in injured peripheral cortex in TBI model rats. (A) Typical western blot bands. (B–D) Quantification of relative expression levels of HO-1, NRF2 and NQO1. Statistical data are shown in mean ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001.

OXA improves functional outcomes and decreases lesion volume and rescues neuronal damage in the TBI mouse model

For farther studying OXA function in neuronal injury after TBI, HE, Nissl, and fluoro-Jade B (FJB) stainings were done. Discoveries of HE staining showed brain tissue in the sham group exhibited rich neuronal structures, with intact cell morphology and structure, a close arrangement between cells, and normal cell volume. In TBI and TBI+NS groups, structure remained loose, inflammatory cells were significantly infiltrated, and extensive vacuolar changes were observed. However, OXA treatment significantly improved these conditions (Figure 6A). Nissl and FJB staining were used to observe neuronal injury. Relative to sham group, TBI and TBI+NS groups displayed cytoplasmic contraction or nuclear pyknosis, and OXA treatment attenuated these histopathological changes (Figure 6B). Brain slices were stained by FJB for determining the neuronal degeneration 24 h following TBI, and it was discovered TBI increased number of degenerated neurons in cerebral cortex, while OXA greatly reduced number of degenerated neurons (F (3, 8) = 25.05, P < 0.001, Figure 7A, 7B). These data suggest that OXA can prevent TBI-induced ipsilateral cortical neuron damage. In addition, we performed the mNSS test for evaluating OXA influence on neurological deficits before treatment with OXA and 24 h following TBI. Before treatment with OXA, there was no significant difference in mNSS scores between the groups (Figure 7C). At 24 hours after TBI, relative to sham group, TBI group owned greater score and difference between TBI and TBI+NS groups was not found. However, OXA treatment significantly reduced the mNSS score, indicating that OXA owns protection influence on neuronal injuries (Figure 7D).

Figure 6. OXA can alleviate brain tissue damage after TBI. (A) Typical HE staining images of cortex in TBI rats of each group. (B) Representative Nissl stained images of cortex in TBI rats of each group.

Figure 7. OXA inhibits neuronal degeneration in the contused peripheral cortex after TBI and improves neural function after TBI. (A) Typical images of Fluoro Jade B (FJB) staining of injured peripheral cortex 24 hours after injury. (B) Quantitative analysis showed number of FJB positive cells in TBI group remained greater in contrast with sham operation group. OXA effectively reduces number of degenerated neurons in contusion surrounding cortex following TBI. Data are shown as mean ± standard deviation. (C) Neurological functions were analysed via mNSS before treatment with OXA. There was no significant difference in mNSS scores between the TBI, TBI+OXA and TBI+NS group. Data are shown as median (P25, P75). (D) Neurological functions were analysed via mNSS 24 hours following TBI. Score in either TBI or TBI+NS group was tremendously greater relative to sham operation group, and OXA treatment reduced neurological deficit score. Data are shown as median (P25, P75), ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, ^nsp > 0.05.