Adult healthy C57BL/6 mice (7–8 weeks, 18–22 g) were obtained from Cavens Biogle Model Animal Research Co., Ltd. (Soochow, China). All mice were kept under controlled temperature of 24–26°C, relative humidity 40–60% and 12 h/12 h light-dark cycle with food and water available ad libitum. Animal care and handling were approved by the Institutional Animal Care and Use Committee of Soochow University (Soochow, China). All experiments were performed in accordance with the Guide for the Care of Use of Laboratory Animals published by the US National Institute of Health.

Total RNA was extracted employing Trizol reagent kit (Invitrogen, United States). Then, mRNA was enriched and fragmented and reverse transcripted into cDNA. After second-strand cDNA were synthesized, purified cDNA with QiaQuick PCR extraction kit (Qiagen, Netherlands), end repaired, poly (A) added, and ligated to Illumina sequencing adapters. PCR amplified, and sequenced using Illumina HiSeq2500 by Gene Denovo Biotechnology Co., (Guangzhou, China). FPKM (fragment per kilobase of transcript per million mapped reads) was calculated using StringTie software. RNAs differential expression analysis was performed by DESeq2 (Love, Huber and Anders 2014) between two different groups and by edgeR (Robinson, McCarthy and Smyth 2010) between two samples. The genes/transcripts with the parameter of false discovery rate (FDR) below 0.05 and absolute fold change ≥2 were considered differentially expressed genes/transcripts.

The renal I/R model was established as previous described (Liu et al., 2019). Briefly, renal I/R was induced by occluding bilateral renal hilum for 45 min by a microvascular clip and followed by 48 h reperfusion. Human Embryonic Kidney (HEK) 293T cells (ATCC, Rockville, United States) were grown in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Cells were incubated at 37°C in an atmosphere of 5% CO₂, and grown to 80% confluence before use. For oxygen glucose deprivation/reoxygenation (OGD/R) treatment is the same as the description of our previous research (Yang et al., 2021b). Briefly, HEK293T were cultured in an anaerobic chamber containing 95% N₂ and 5% CO₂ for 6 h and then reoxygenated for 6 h to induce OGD/R condition. Cells were pretreated with Dex (1 μM, 24 h incubation before OGD/R) or/and ACSL4 overexpression vector transfection before exposure to OGD/R, as mentioned in this study.

The plasmid was created and synthesized by RiboBio (Guangzhou, China). Before transfection, cells were transplanted to 6-well or 96-well plates and grown to a confluence of 30%. Gp-transfect Mate (GenePharma, China) was used according to the manufacturer’s protocol. The ACSL4 plasmid was subcloned into the pcDNA3.1 vector to overexpress ACSL4. All plasmids were incubated with the transfection media for 48 h before OGD/R.

Serum and cellular supernatants were collected for the LDH activity assay (A020, JianCheng, China) according to the manufacturer’s instructions. The optical density value for each well at 450 nm absorbance was measured by Spectra Max 190 plate reader. Experiments were repeated at least three times.

To measure cell viability, 6 × 10³ cells per well were seeded in 96-well plates and treated as indicated, after which the medium in each well was replaced with 100 µL fresh medium containing 10% Cell Counting Kit-8 reagent (Apex BIO, K1018). After incubation for 1 h at 37°C according to the instructions of the kit, plate was read by a Spectra Max 190 plate reader at a wavelength of 450 nm.

Phosphate-buffered saline (PBS) was used for ischemic renal tissue homogenized. Collecting the supernatant after centrifugation. The content of iron was determined by the Iron Assay Kit (A039, JianCheng, China) according to the manufacturer’s instructions. ACSL4 was confirmed facilitate the esterification of arachidonoyl and adrenoyl into phosphatidylethanolamine (PE), a process closely related to ferroptosis (Kagan et al., 2017), so PE was used to evaluate the level of ACSL4 activity (Li et al., 2019). PE kits (ab241005, Abcam) were used according to the manufacturer’s instructions.

A lipid peroxidation assay kit (A106, Jiancheng, China) was used to test the lipid peroxidase (LPO) level in lysates of tissue following the manufacturer’s instructions. Briefly, lipid peroxide reacts with chromogenic reagents under the condition of 45°C for 60 min and produces a stable chromophore with a maximum absorption peak at 586 nm.

Lipid ROS was measured using the live-cell analysis reagent BODIPY 581/591 C11 (27,086, Cayman). Treated cells were incubated with BODIPY (5 μM) for 1 h at 37 °C in 24-well cultures. After incubation, cells were harvested and washed with PBS, and then resuspended in 500 μL PBS. Images were acquired under an IX83 fluorescence microscope (ECLIPSE Ts2R-FL, Nikon). Cell fluorescence was acquired on a Flow Cytometer (Cytomics^TMFC500, Beckman) and analyzed with FlowJo software (FlowJo^TMv10, United States).

The whole protein of mice renal tissues or HEK293T cells was extracted using the RIPA lysate buffer (Beyotime, China). Protein concentration was measured by a bicinchoninic acid reagent kit (Beyotime, China). The proteins were separated in 8% or 10% gels using a sodium dodecyl sulphate-polyacrylamide gel electrophoresis and then transferred onto polyvinylidene fluoride membranes (Millipore, Bedford, United States). The membranes were blocked with 5% non-fat milk at room temperature for 2 h and then incubated at 4°C overnight with the corresponding primary antibodies: GPx4 antibody (P02794, Abways), ACSL4 antibody (O60488, Abways); COX2 (ab62331, Abcam); β-tubulin antibody (86,298, Cell Signalling Technology). Next, the membranes were washed and incubated at room temperature for 2 h with the secondary antibodies: Goat anti-rabbit IgG antibody (CW0103S, CoWin), Goat anti-mouse IgG antibody (CW0102S, CoWin). The bands were detected using an enhanced chemiluinescence kit (NCM, China) under a luminescent imaging workstation (Tanon5200, China). The protein intensity was analyzed with the ImageJ software and normalized to β-tubulin.

The renal tissue was subjected to RNA extraction, reverse transcription, and then real-time quantitative reverse transcription-PCR as described before (Peng et al., 2020). The following primer sets (Sangon Biotech Inc, China) were used in the process: mouse ACSL4 (forward, 5′-CAA TAG AGC AGA GTA CCC TGA G-3′; reverse, 5′-TAG AAC CAC TGG TGT ACA TGA C-3′), mouse IL-6 (forward, 5′-TGA TGC ACT TGC AGA AAA CAA TCT GA-3′; reverse, 5′-AGC TAT GGT ACT CCA GAA GAC CAG AGG-3′), mouse TNF-α (forward, 5′-TGA TCG GTC CCC AA A GGG ATG; reverse, 5′-TTG GTG GTT TGC TAC GAC GTG G-3′), mouse β-tubulin (forward, 5′-CAG CGA TGA GCA CGG CAT AGA C; reverse, 5′-CCA GGT TCC AAG TCC ACC AGA ATG-3′). human IL-6 (forward, 5′-ATG AAC TCC TTC TCC ACA AGC GC -3′; reverse, 5′-GAA GAG CCC TCA GGC TGG ACT G -3′), human TNF-α (forward, 5′-TGA TCG GTC CCC AA A GGG ATG; reverse, 5′-TTG GTG GTT TGC TAC GAC GTG G-3′), human β-tubulin (forward, 5′-CAG CGA TGA GCA CGG CAT AGA C; reverse, 5′-CCA GGT TCC AAG TCC ACC AGA ATG-3′).

After 48 h of reperfusion, the anesthetized mice were killed, the kidneys were washed with precooled PBS (pH 7.4), cut into 2 mm × 2 mm x 2 mm block tissue and then fixed in 100 mM PBS (pH 7.0–7.5) containing 2.5% glutaraldehyde (Servicebio, China) at room temperature for 2 h, and then transferred to 4°C for storage overnight. Ultrathin sections (70 nm) were cut, stained with uranyl acetate and lead citrate, and viewed under a transmission electron microscope (TEM; HT7700; Japan). Five fields for each sample was randomly chose and counted the mitochondria with ferroptotic features. The number of ferroptotic mitochondria per field in each sample was quantified.

Malondialdehyde (MDA) production was determined using a lipid peroxidation assay kit (A003, Jiancheng, China) based on the method of thiobarbituric acid reactive substances (TBARS) to reflect the degree of lipid peroxidation in vivo. Mice renal was homogenized and the supernatant was collected for GSH analysis using a GSH assay kit (A06, Jiancheng, China).

Concentrations of serum creatinine and BUN were detected by Automatic biochemical instrument (Siemens ADVIA 1800 Chemistry System, Germany). Hematoxylin and Eosin (H&E)-stained kidney specimens were used to evaluate kidney injury, the renal sections were fixed in 4% paraformaldehyde and embedded in paraffin. Five millimeter slices were cut for H&E staining and then observed under an optical microscope (Olympus, Japan). Renal injury score was evaluated by an established grading by Zhou, W et al. as described before (Zhou et al., 2000). Renal injury was graded as fellow: normal (0 score); <10% necrosis (1score); 10–25% (2 score); 25–75% (3 score); and >75% (4).

Liproxstatin-1(Lip-1, B4987, APExBIO, United States), a ferrptosis inhibitor, was administrated i.p. at a concentration of 10 mg/kg 1 h before ischemia induction, in accordance with previous study protocols (Li et al., 2019). Rosiglitazone (ROSI, S2556,Selleck), a classic peroxisome proliferator-activated receptor-γ agonist that has been used for ACSL4 inhibition, was administered intravenously at a concentration of 0.4 mg/kg 1 h before ischemia induction (Doll et al., 2017; Li et al., 2019). Atipamezole (ATI, abs816081, absin), a α2-AR antagonist, was administered intraperitoneally at a concentration of 250 μg/kg 0.5 h before I/R treatment, and just the renal pedicle clamp was released. Dexmedetomidine (21,090,731, Yangtze River Pharmaceutical group, China) was administered intraperitoneally at a concentration of 50 μg/kg in two time points, including 30 min before I/R and just the renal pedicle clamp was released. Sham mice underwent all the procedures except renal ischemia. Blood and kidney samples were collected 48 h after reperfusion.

Data were analyzed by GraphPad Prism 8.0 software (GraphPad Inc, La Jolla, United States) and expressed as means ± S.D. Statistical significance was determined by using Student’s t-test in two groups. ANOVA was used to compare more than two group. p < 0.05 was considered statistically significant.

The renal I/R injury significantly increased serum creatinine (Scr) and blood urea nitrogen (BUN) levels compared with the sham group, Dex administration effectively reduced the increased Scr and BUN levels (Figures 1A,B), which is consistent with the previous study that Dex has renal protection, but the exact protective mechanism is not clear, so we implemented RNA-seq to find the target gene of Dex. Principal component analysis (PCA) was performed with R package models (http://www.rproject.org/) in this experience. PCA is largely used to reveal the structure/relationship of the samples/datas (Figure 1C). The RNA-seq analysis showed 1901 up-regulated after renal I/R compared with the sham group, and 1398 down-regulated genes after Dex administration compared with the I/R group, Among them, 1047 genes have attracted our attention, as shown in the Venn diagram (ǀlog2FCǀ>1, and FDR<0.05, Figures 1D–F). We focused on the ACSL4, a key isozyme for polyunsaturated fatty acids metabolism that dictates ferroptosis sensitivity. RNA-seq result showed higher ACSL4 expression in the I/R group than those in the sham group and then significantly decreased with Dex administration ( Figure 1G; Table1). The qRT-PCR (Figure 1H) and western blot (Figure 1I) results reveal that Dex significantly reversed the increased expression of ACSL4 following renal I/R injury.

Some core factors, such GPx4, COX2, were confirmed as key and valid proteins in ferroptosis regulation (Li et al., 2019) (Chen et al., 2021). Therefore, to evaluate ferroptosis sensitivity after ischemia in the renal, we determined the expression levels of these proteins under I/R condition. The decreased GPx4 and increased COX2 protein expression in the I/R group were confirmed by western blot analysis (Figure 2A). LPO and MDA assay indicated that lipid peroxidation was higher in the I/R group than those in sham group (Figures 2B,C). LDH level was also increased after I/R injury (Figure 2D). TEM was used to investigate the morphological feature of ferroptosis. The mitochondrial morphological feature of ferroptosis was assessed, and the results showed that shrunk mitochondrial, reputed outer mitochondrial membrane, and the disappearance of mitochondrial cristae in the random field in the I/R group (Figure 2E). In addition, tissue GSH level was reduced in the I/R group (Figure 2F). And renal I/R led to an accumulation of tissue iron levels (Figure 2G). Our results also showed IL-6 mRNA and TNF-α mRNA, the critical indicators that involved in inflammation, were significantly increased during I/R (Figure 2H). Taken together, our result indicated that ferroptosis and inflammation may be contributed to renal I/R injury.

In order to confirm the role of ferroptosis in renal I/R injury, Liproxstatin-1 (Lip-1), a specific ferroptosis inhibitor, was administrated. Lip-1 treatment significantly reversed the increased COX2 and decreased GPx4 protein expression induced by renal I/R injury (Figure 3A). The histopathological scores showed that renal I/R resulted in severe tubular damage, included widespread degeneration of tubular architecture, tubular dilation, tubular cell swelling, pyknotic nuclei, and luminal congestion, which could be significantly improved with Lip-1 treatment (Figure 3B). Lip-1 reduced the increased serum LDH level during I/R injury. (Figure 3C). Furthermore, Lip-1 also reduced MDA (Figure 3D) and LPO (Figure 3E), and increased the GSH levels (Figure 3F), indicating an improvement in lipid peroxidation and oxidation resistance.

ACSL4, a critical contributor and regulator of ferroptosis, determined the sensitivity of ferroptosis. Rosiglitazone (ROSI), a peroxisome proliferator-activated receptor-γ (PPAR-γ) activator, could inhibit ACSL4 activity. After ROSI treatment, ACSL4 activity was inhibited in the I/R + ROSI group compared with in the I/R group (Figure 4A). Meanwhile, the decreased GPx4 and increased COX2 protein expression in the I/R group were also reversed by ROSI treatment (Figure 4B). The relative histopathological score was significantly lower in the I/R + ROSI group than that in the I/R group (Figure 4C). ROSI treatment significantly decreased LPO, LDH and MDA level, as well as increased GSH level, in the kidney tissue during renal I/R injury (Figure 4D–G). Besides, the enhanced TNF-α mRNA and IL-6 mRNA levels induced by I/R were effectively reduced by ROSI treatment (Figures 4H,I). These results manifested that the suppression of ACSL4 activity alleviated ferroptosis-induced kidney damage and inflammation during renal I/R.

To assess whether ACSL4 is the mediator of Dex exerting its effects, HEK293T cells were subjected to OGD/R. OGD/R significantly enhanced ACSL4 and COX2 expression and suppressed GPx4 expression (Figure 5A). OGD/R fortified the level of LPO (Figure 5B), suggesting that OGD/R contributes to lipid peroxidation. OGD/R increased cell death observed in the CCK-8 assay (Figure 5C). LDH release was also increased with OGD/R treatment (Figure 5D). In addition, OGD/R treatment increased inflammatory cytokines in HEK293T cells (Figure 5E). A distinct decrease in LDH release and increase in cell viability was observed in cells after lip-1 treatment (Supplementary Figure 1A, 1B). Then, HEK293T cells were transfected with ACSL4 overexpression plasmid for 48 h. ACSL4 overexpression effectively blocked the decreased ACSL4 in HEK293T cells during OGD/R with Dex treatment. Moreover, ACSL4 overexpression increased COX2 expression and decreased GPx4 expression (Figure 5F). As ACSL4 has been confirmed to aggrandize cell sensitivity to ferroptosis, we further evaluated the effects of ACSL4 on HEK293T cell ferroptosis. ACSL4 overexpression could partially reversed the protection of Dex on the inhibition of cell death determined by CCK-8 assay (Figure 5G). Dex decreased the density of green fluorescence represented by BODIPY581/591 C11 staining after OGD/R and moderately reversed by ACSL4 overexpression (Figure 5H). Lipid peroxidation analysis using BODIPY demonstrating that Dex significantly decreased cell ferroptosis and ACSL4 overexpression could reversed the protection of Dex to OGD/R-induced ferroptosis (Figure 5I). ACSL4 overexpression blocked the declined LDH (Figure 5J), and Dex effectively decreased lipid peroxidation as indicated by the decrease in LPO level (Figure 5K). Furthermore, ACSL4 overexpression abolished the decreased inflammatory cytokines (Figure 5L, 5M) in HEK293T cells during OGD/R with Dex treatment. These results demonstrated that Dex inhibited ferroptosis and inflammation induced by OGD/R, and this effect was partially reversed by ACSL4 overexpression, suggesting that Dex might regulate ferroptosis and inflammation by targeting ACSL4 during OGD/R.

Base on the important role of ACSL4 signaling during I/R injury, we further investigated that whether ACSL4 signaling were involved in Dex-elicited protection against renal I/R injury. ATI administration effectively blocked the decreased ACSL4 and COX2 protein expression and increased GPx4 level suffering from Dex treatment during renal I/R injury (Figure 6A). ATI treatment declined the histopathological score (Figure 6B), as well as the reversion of increased GSH in kidney tissue induced by Dex treatment during I/R injury (Figure 6C). MDA production was higher in the I/R + Dex + ATI group than that in the I/R + Dex group (Figure 6D). ATI treatment significantly increased LPO and LDH levels (Figures 6E,F). ATI also attenuated the anti-inflammation effect of Dex (Figures 6G,H). In brief, ATI treatment significantly reversed the Dex effects on the inhibition role of ACSL4 during renal I/R injury, which suggest that Dex attenuates ferroptosis-mediated I/R injury and inflammation by downregulating ACSL4 signaling via α2-AR.