Curcumin reduces renal damage associated with rhabdomyolysis by decreasing ferroptosis-mediated cell death

ABSTRACT: Acute kidney injury is a common complication of rhabdomyolysis. A better understanding of this syndrome may be useful to identify novel therapeutic targets because there is no specific treatment so far. Fer- roptosis is an iron-dependent form of regulated nonapoptotic cell death that is involved in renal injury. In this study, we investigated whether ferroptosis is associated with rhabdomyolysis-mediated renal damage, and we studied the therapeutic effect of curcumin, a powerful antioxidant with renoprotective properties. Induction of rhabdomyolysis in mice increased serum creatinine levels, endothelial damage, inflammatory chemokines, and cytokine expression, alteration ofredox balance (increased lipidperoxidation and decreasedantioxidant defenses), and tubularcelldeath. Treatment with curcumin initiated before or after rhabdomyolysis induction ameliorated all these pathologic and molecular alterations. Although apoptosis or receptor-interacting protein kinase (RIPK)3–mediated necroptosis were activated in rhabdomyolysis, our results suggest a key role of ferroptosis. Thus, treatment with ferrostatin 1, a ferroptosis inhibitor, improved renal function in glycerol-injected mice, whereas no beneficial effects were ob- served with the pan-caspase inhibitor carbobenzoxy-valyl-alanyl-aspartyl-(O-methyl)- fluoromethylketone or in RIPK3-deficient mice. In cultured renal tubular cells, myoglobin (Mb) induced ferroptosis-sensitive cell death that was also inhibited by curcumin. Mechanistic in vitro studies showed that curcumin reduced Mb-mediated in- flammation and oxidative stress by inhibiting the TLR4/NF-kB axis and activating the cytoprotective enzyme heme oxygenase 1. Our findings are the first to demonstrate the involvement of ferroptosis in rhabdomyolysis-associated renal damage and its sensitivity to curcumin treatment. Therefore, curcumin may be a potential therapeutic ap- proach for patients with this syndrome.—Guerrero-Hue, M., Garc´ıa-Caballero, C., Palomino-Antol´ın, A., Rubio- Navarro, A., Va´zquez-Carballo, C., Herencia, C., Mart´ın-Sanchez, D., Farre´-Alins, V., Egea, J., Cannata, P., Praga, M., Ortiz, A., Egido, J., Sanz, A. B., Moreno, J. A. Curcumin reduces renal damage associated with rhabdomyolysis by decreasing ferroptosis-mediated cell death. FASEB J. 33, 000–000 (2019). www.fasebj.org

KEY WORDS: acute kidney injury • myoglobin • kidney • oxidative stress

ABBREVIATIONS: 4-HNE, 4-hydroxynonenal; 7-AAD, 7-aminoactinomycin D; AKI, acute kidney injury; BUN, blood urea nitrogen; CCL2, C-C motif chemokine ligand 2; Fer-1, ferrostatin 1; GSH, glutathione; Havcr1, Hepatitis A Virus Cellular Receptor 1; HK-2, human proximal tubular epithelial; HO- 1, heme oxygenase 1; ICAM-1, intercellular adhesion molecule 1; KIM-1, kidney injury molecule 1; Lcn2, Lipocalin-2; Mb, myoglobin; MCT, proximal murine tubular epithelial cell; MDA, malondialdehyde; MLKL, pseudokinase mixed-lineage kinase domain-like; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide; MyD88, myeloid differentiation primary response 88; Nec-1s, necrostatin 1s; NGAL, neutrophil gelatinase-associated lipocalin; NLRP3, nod-like receptor protein 3; Nrf2, nuclear factor (erythroid-derived 2)-like 2; RIPK, receptor-interacting protein kinase; ROS, reactive oxygen species; SnPP, tin protoporphyrin; TBS, Tris-buffered saline; zVAD, carbobenzoxy-valyl-alanyl-aspartyl-(O-methyl)- fluoromethylketone Rhabdomyolysis is characterized by skeletal muscle dam- age, as reported in severe trauma, prolonged ischemia, in- tense exercise, and some metabolic disorders, among others
(1). Acute kidney injury (AKI) is a common complication of rhabdomyolysis that increases patient mortality (2). Cur- rently, there is no specific treatment for rhabdomyolysis- induced AKI – only palliative care (3). Therefore, a better understanding of the pathogenesis of rhabdomyolysis may help to identify novel therapeutic targets.

Rhabdomyolysis causes the release of myoglobin (Mb) and other muscle-cell components to the bloodstream. Mb is freely filtered by glomeruli and reabsorbed by proximal tubules, promoting oxidative stress, inflammation, and cell death, thus leading to AKI (4, 5). Mb-mediated oxi- dative stress changes mitochondrial membrane perme- ability, resulting in cytochrome c release, caspase-3 activation, and tubular cell apoptosis (6–9). It has been suggested that, besides apoptosis, other types of cell death may be involved in rhabdomyolysis-induced AKI (10). Necroptosis is a caspase-independent death pathway de- pendent on the receptor-interacting protein kinase (RIPK) 1 and pseudokinase mixed-lineage kinase domain-like (MLKL) and characterized by phosphorylation of RIPK3 and MLKL (11–14). Ferroptosis is another form of regu- lated nonapoptotic cell death, mediated by iron accumu- lation and lipid peroxidation (15–17). Ferroptosis has been involved in renal damage but there is no information about its specific role in rhabdomyolysis-associated AKI, where iron accumulation and lipid peroxidation have al- ready been described (3, 18).

Curcumin [1,7-bis(4-hydroxy-3-methoxyphenyl)- 1,6-heptadiene-3,5-dione], a phenolic compound from Curcuma longa, is a powerful antioxidant and anti- inflammatory molecule that protects from renal dam- age in experimental AKI, such as cisplatin nephrotoxicity or ischemia-reperfusion injury, or in chronic kidney in- jury, as reported in subtotal nephrectomy (19–22). The beneficial actions of curcumin depend on direct antiox- idant effects related to scavenging reactive oxygen spe- cies (ROS) (23, 24). Moreover, curcumin induces the expression of the antioxidant enzyme heme oxygenase 1 (HO-1) via activation of the nuclear factor (erythroid- derived 2)-like 2 (Nrf2) transcription factor (25). HO-1 catabolizes Mb-derived heme to biliverdin, carbon monoxide, and iron, which is subsequently stored in ferritin (26). Therefore, the HO-1–ferritin system de- creases cellular exposure to heme-derived products.

Inhibition of oxidative stress was protective in experi- mental rhabdomyolysis (6, 27). We hypothesized that Mb- derived iron accumulation and lipid peroxidation may injure tubular cells by activating ferroptosis and thereby contribute to AKI progression in rhabdomyolysis. Fur- thermore, we hypothesized that curcumin may reduce ferroptosis by inhibiting oxidative stress via activation of the Nrf2–HO-1 axis. In this sense, previous studies show that the Nrf2 pathway protects against ferroptosis in he- patocellular carcinoma cells (16). Thus, in this study we aimed to investigate the implication of ferroptosis in rhabdomyolysis-associated renal injury and the possible beneficial effect of curcumin. To further clarify the role of ferroptosis in rhabdomyolysis-associated renal damage,we used the ferroptosis inhibitor ferrostatin 1 (Fer-1) in a murine model of rhabdomyolysis and evaluated the im- plication of other cell death pathways by using the pan- caspase inhibitor carbobenzoxy-valyl-alanyl-aspartyl-(O- methyl)- fluoromethylketone (zVAD) and RIPK3-deficient mice. Finally, we characterized renal alterations (oxidative stress, inflammation, and cell death) and the molecular pathways involved in rhabdomyolysis injury and ex- plored whether ferroptosis-mediated damage is pre- vented by curcumin treatment both in vivo and in cultured tubular cells.

MATERIALS AND METHODS
Animal model

To induce rhabdomyolysis, C57BL/6J mice (male, 12 wk old, n = 50) were intramuscularly injected in each thigh caudal muscle with 10 ml/kg of 50% glycerol ($99.5% m/v) or saline as a control. Mice were dehydrated for 16 h before glycerol injection. In a first set of experiments, 250 ml of curcumin (1 g/kg) per vehicle was injected intraperitoneal 24 h before rhabdomyolysis induction and on the same day of glycerol injection. In other studies, curcumin was administrated 30 min or 3 h after glycerol injection to analyze its therapeutic value once rhabdomyolysis was induced. In all cases, mice were euthanized 24 h after glyc- erol administration and blood, urine, and kidney samples were collected. In other experiments, RIPK3 knockout mice [provided by Kim Newton and Vishva Dixit (Genentech, San Fransisco, CA, USA) were studied, and, additionally, C57BL/6J mice were in- jected intraperitoneally with 5 mg/kg Fer-1 (Santa Cruz Bio- technology, Dallas, TX, USA), 10 mg/kg zVAD (Bachem, Bubendorf, Switzerland), or vehicle (DMSO) 30 min before glycerol injection. Mice were anesthetized for all procedures with isoflurane 2–3% v/v or ketamine-xylazine. Anesthetized mice were perfused with saline; 1 kidney was frozen in liquid nitrogen for RNA and protein isolation, and the other kidney was fixed in 4% paraformaldehyde, embedded in paraffin, and used for im- munohistochemistry. Blood was collected in serum tubes and stored at —80°C until used. All experiments were conducted in accordance with the Directive 2010/63/EU of the European Parliament and were approved by a local Institutional Animal Care and Use Committee.

TUNEL assay

Cell death was quantified using a TUNEL assay. DNA frag- mentation was detected using a kit from MilliporeSigma (S7110; Burlington, MA, USA). The number of green-colored cells per field was counted in renal tissue at 3400 magnification. At least 10 areas in the cortex per tissue section were randomly selected. TUNEL is a method to detect DNA breakage in individual cells, as reported in apoptosis, but it may also label cells having DNA damaged by other means than in the course of apoptosis, in- cluding ferroptosis (10).

Cell culture

Proximal murine tubular epithelial cells (MCTs) were cultured in RPMI 1640 (R0883; MilliporeSigma) supplemented with 10% decomplemented fetal bovine serum (F7524; MilliporeSigma), glutamine (2 mM; G7513; MilliporeSigma), and penicillin- streptomycin (100 U/ml; P0781; MilliporeSigma) in 5% CO2 at 37°C (28). Human proximal tubular epithelial (HK-2) cells [American Type Culture Collection (ATCC), Manassas, VA, USA] were grown on the same media as the MCT cells plus 5 mg/ ml insulin, 5 mg/ml transferrin, 5 ng/ml sodium selenite, and 5 ng/ml hydrocortisone (29). Cells were stimulated with Mb (1–10 mg/ml; M1882; MilliporeSigma) in the presence or absence of curcumin (10 mM; MilliporeSigma) and the HO-1 inhibitor tin protoporphyrin (SnPP; 1 mM), the glutathione (GSH) peroxidase 4 inhibitor RSL3 (750 ng/ml; Selleck Chemicals, Houston, TX, USA), the RIPK1 activity inhibitor necrostatin-1s (Nec-1s; 30 mM; MilliporeSigma), the pan-caspase inhibitor zVAD (25 mM; BD Biosciences, San Jose, CA, USA), and the ferroptosis inhibitor Fer- 1 (1 mM; Santa Cruz Biotechnology).

Histopathology and immunohistochemistry

Histological studies were performed in paraffin-embedded 3-mm kidney sections. Hematoxylin and eosin–stained sections were examined by a nephropathologist blinded to the nature of the samples who scored signs of tubular damage (loss of brush border, tubular dilation, edema, and tubular death) on a semi- quantitative scale from 0 to 4.

Immunofluorescence and immunohistochemistry studies were performed in paraffin-embedded tissue sections or cultured cells as previously described in Moreno et al. (30). Primary anti- bodies were rabbit anti-mouse 4-hydroxynonenal (4-HNE; 1:100 dilution; ab46545; Abcam, Cambridge, MA, USA), rabbit anti- mouse HO-1 (1:100 dilution; ADI-OSA-150-D; Enzo Life Sciences, Farmingdale, NY, USA), rabbit anti-mouse ferritin (1:800 di- lution; ab86247; Abcam), anti-p-NFkB p65 (1:100 dilution; sc- 136548; Santa Cruz Biotechnology), and anti–intercellular adhe- sion molecule 1 (ICAM-1; 1:100; ab124760; Abcam), anti-mouse anti–lotus tetragonolobus lectin (1:100; B-1325-2; Vector Labora- tories, Burlingame, CA, USA). The biotinylated secondary anti- bodies were applied for 1 h. Avidin-biotin peroxidase complex (Vectastain ABCkit; PK-7200; Vector Laboratories) was added for 30 min. Sections were stained with 3,39-diaminobenzidine or 3- amino-9-ethyl carbazol (S1967; Agilent Technologies, Santa Clara, CA, USA) and counterstained with hematoxylin. Images were taken witha Nikon Eclipse E400 microscope (Nikon, Tokyo, Japan) and Nikon ACT-1 software (Nikon). In immunofluores- cence studies, tubular cells were fixed in 4% paraformaldehyde and permeabilized in 0.2% Triton X-100/PBS, washed in PBS, and incubated with rabbit anti-mouse HO-1 (1:100 dilution; ADI- OSA-150-D; Enzo Life Sciences) or anti-Nrf2 (1:100; sc-722; Santa Cruz Biotechnology), followed by Alexa Fluor 488 secondary antibody (1∶200; A11090; Thermo Fisher Scientific, Waltham, MA, USA). Nuclei were stained with DAPI.

ELISA

Concentrations of murine C-C motif chemokine ligand 2 (CCL2) in cell culture supernatants were determined by ELISA (MJE00B; R&D Systems, Minneapolis, MN, USA) following the manufac- turer’s instructions.

RNA extraction and real-time PCR

Total RNA from kidneys or cultured cells was isolated by the Trizol method with Tri Reagent (TR 118; Molecular Research Center, Cincinnati, OH, USA) and reverse-transcribed with a High Capacity cDNA Archive Kit (Thermo Fisher Scien- tific); real-time PCR was performed on an ABI Prism 7500 PCR system (Thermo Fisher Scientific) using the DDCt method. Target gene expression levels were normalized with 18S ribosomal RNA expression. Expression of target genes was analyzed using Taqman gene expression assays for murine kidney injury molecule 1 (KIM-1; Mm00506686_m1), Lipocalin-2 (Lcn2) (Mm01324470_m1), CCL2 (Mm00441242_ m1), TNF-a (Mm00443258_m1), HO-1 (Mm00516005_m1), ferritin (Mm03030144_g1), endothelin-1 (Mm00438656_m1), MLKL (Mm01244222_m1), RIPK3 (Mm00444947_m1), IL-6 (Mm004 46190_m1), ICAM-1 (Mm00516023_m1), TGF-b (Mm01178819_m1), TLR4 (Mm00445273_m1), myeloid differentiation primary response 88 (MyD88) (Mm00440338_m1), and IL-1b (Mm00434228_m1) (Thermo Fisher Scientific), and designed probes for caspase 1 (59-GGAAGTATTGGCTTCTTATTGG-39 and 59-GGGACATTAAACGAAGAATCC-39) and nod-like receptor protein 3 (NLRP3; 59-GTCTAATTCCAGCATCTG- TAG-39 and 59-TTCAATCTGTTGTTCAGCTC-39).

Western blot

Protein isolation was performed by homogenizing tissue samples with lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 0.2% Triton X-100, 0.3% NP-40, 0.1 mM PMSF, and 1 mg/ml pepstatin A). Proteins were then separated by 10% SDS- PAGE under reducing conditions. After electrophoresis, samples were transferred to PVDF membranes (IPVH00010; Milli- poreSigma). Membranes were blocked with 5% skimmed milk in Tris-buffered saline (TBS)/0.5% v/v Tween 20 for 1 h, washed with TBS/Tween, and incubated with primary antibodies. Anti- bodies were diluted in 5% milk TBS/Tween. Membranes were washed with TBS/Tween and then incubated with appropriate horseradish peroxidase–conjugated secondary antibody (1:2000; GE Healthcare, Waukesha, WI, USA). After washing with TBS/ Tween, blots were developed with the chemiluminescence method (ECL Luminata Crescendo, WBLUR0500; MilliporeSigma). Blots were then probed with mouse monoclonal anti–a-tubulin anti- body (1:5000, T6199; MilliporeSigma) and levels of expression were corrected for minor differences in loading.

Cell viability and cell death assay

Cell viability was determined using the 3-(4,5-dimethylthiazol-2- yl)-2,5 diphenyltetrazolium bromide (MTT; MilliporeSigma) colorimetric assay. After treatment, cells were incubated with 1 mg/ml MTT in PBS for 1 h at 37°C. The resulting formazan crystals were dried and dissolved in DMSO. Absorbance (in- dicative of cell viability) was measured at 570 nm in a microplate reader (BMG Labtech, Cary, NC, USA).Cell death was assessed by flow cytometry of cells stained with an annexin V 7-aminoactinomycin D (7-AAD) kit (BD Bio- sciences). This test is usually interpreted as identifying apoptotic cells (annexin V+ cells), necrotic cells (7-AAD positive cells), and late apoptosis and necrosis (annexin V and 7-AAD double- positive cells).

GSH assay

To quantify free GSH, we used monochlorobimane, a non- fluorescent dye that forms a highly fluorescent adduct when conjugated with GSH by GSH S-transferase (31, 32). Renal tissue sections were resuspended in 50 ml potassium phosphate buffer (100 mM, pH = 7.4) and mechanically disaggregated by sonica- tion (3–10 s, power 2) in ice. The reactionstarted bythe addition of monochlorobinane (100 mM) and GSH S-transferase (0.5 U/ml)
(32) in a final volume of 100 ml of potassium phosphate buffer (100 mM, pH = 7.4). The reaction was monitored in a Fluostar Optima microplate reader (BMG Labtech) at excitation and emission wavelengths of 410 and 485 nm, respectively, over 1 h at room temperature. All measurements were done in triplicate. In the absence of GSH S-transferase, the rate of adduct formation was very slow; therefore, the use of GSH S-transferase eliminates

Lipid peroxidation assay

Lipid peroxidation was assessed by quantifying malondial- dehyde (MDA) content, using a specific MDA kit (10009055; Cayman Chemicals, Ann Arbor, MI, USA), following the manufacturer’s instructions.

Statistics

Data are expressed as means ± SEM. Data comparisons between experimental groups at each time point were analyzed with a 1- way ANOVA test and least significant difference post hoc test. Values of P< 0.05 were considered significant. Statistical analysis was performed using SPSS v.19.0 (IBM SPSS, Chicago, IL, USA) statistical software.

RESULTS

Curcumin improves functional and histologic renal damage caused by rhabdomyolysis

Glycerol-injected mice showed an increase in blood urea nitrogen (BUN) and serum creatinine levels, as compared with controls (Fig. 1A, B). In line with the decline of renal function, histologic analysis reported signs of severe kid- ney injury in mice with rhabdomyolysis, such as loss of brush border of the tubular epithelium, tubular cell death, tubular lumen dilatation, and interstitial edema (Fig. 1C, D). Gene expression studies disclosed increased mRNA levels of the tubular injury biomarkers KIM-1 (Havcr1) and neutrophil gelatinase-associated lipocalin (NGAL; Lcn2) after glycerol administration, confirming the differences observed in renal function (Fig. 1E, F). These functional, histological, and molecular parameters associated with AKI were decreased by curcumin treatment, suggesting a protective effect of this molecule against rhabdomyolysis- related renal damage.

Curcumin reduces inflammation and endothelial damage associated with rhabdomyolysis in vivo and in vitro

We have previously demonstrated that glycerol adminis- tration produces a kidney inflammatory response (33). Here, we observed that curcumin reduced CCL2, TNF- a, and TGF-b1 mRNA expression induced by rhabdo- myolysis (Fig. 2A, B). Previous studies have shown that rhabdomyolysis is associated with endothelial damage (6, 34). We observed an increased mRNA expression of the markers of endothelial damage ICAM-1 and endothelin-1 in mice with rhabdomyolysis, whereas curcumin reduced expression of these molecules (Fig. 2C). Double-immnunofluoresce analysis of kidney ICAM-1 and lectin, a proximal tubular cell marker, confirmed these results, showing increased ICAM-1 expression in peritubular capillaries (Fig. 2D). NGAL is a renal injury marker, produced by tubular cells and related to inflammation (35). To study the molecular mechanisms involved in the protection of curcumin against rhabdomyolysis, we performed experiments in cultured tubular MCT cells stimulated with Mb. As we observed in vivo, Mb increased NGAL (Lcn2) mRNA ex- pression in cultured tubular cells, an effect that was pre- vented by curcumin (Fig. 2E). A similar response was observed when we analyzed the mRNA expression of different inflammatory markers, such as TNF-a, IL-6, and CCL2 (Fig. 2F–H). In the same way, Mb-induced CCL2 secretion was reduced in curcumin-treated cells (Fig. 2I) markers in our experimental model. Rhabdomyolysis promoted kidney lipid peroxidation, as determined by enhanced levels of reactive aldehydes such as MDAand 4- HNE, an effect that was reduced by curcumin treatment (Fig. 4A, B). This data may be related to either a reduced ROS production or lower heme content in curcumin- treated mice. To address this issue, we first quantified heme levels in kidneys from our model. Rhabdomyolysis increased Mb-derived heme concentration in the kidneys but no differences were observed after curcumin admin- istration (Supplemental Fig. S1A). We next analyzed the expression of 2 key antioxidant proteins involved in heme catabolism, such as HO-1 and ferritin (Fig. 4B–F). Rhab- domyolysis induced the mRNA and protein expression of HO-1 and ferritin, an effect that was reduced in glycerol- injected mice treated with curcumin (Fig. 4B–F). In cul- tured MCT cells, curcumin also decreased Mb-mediated HO-1 and ferritin mRNA and protein expression, in- dicative of a lower oxidative stress (Fig. 5A, B). Because curcumin induces Nrf2 activation (25), we hypothe- sized that the protective effects of curcumin were me- diated by a rapid activation of the Nrf2/HO-1 axis to increase cellular antioxidant defenses and to prevent Mb-mediated harmful effects. To validate this hypoth- esis, we performed short time experiments in cells stimulated with Mb, curcumin, or the combination of both molecules (Fig. 5C). It is important to note that curcumin was added 16 h before Mb stimulation. Cur- cumin pretreatment induced HO-1 protein expression, so this antioxidant protein was up-regulated before Mb addition. At 3 h of Mb stimulation, HO-1 expression was higher in cells pretreated with curcumin than in cells stimulated with Mb alone, indicative of a higher protection. However, by 6 h, HO-1 expression was lower in cells costimulated with Mb and curcumin, suggesting a lower oxidative stress, as was observed in vivo. In a further step, we analyzed the possible anti-oxidant effect of curcumin on Mb-stimulated cells. As reported in Fig. 5D and E, Mb increased the production of hydrogen peroxide, an effect partially reduced by curcumin administration. These protective effects were mediated by HO-1 because costimulation with curcu- min and SnPP (an HO-1 inhibitor) restored Mb- mediated ROS production (Fig. 5D, E).

A reduction in antioxidant defenses may also be associated with an enhanced oxidative stress. Therefore, we determined the effect of curcumin on GSH, the most abundant endogenous antioxidant (38). Rhabdomyolysis decreased renal GSH levels, an effect that was abolished by curcumin (Supplemental Fig. S1B). Altogether, these re- sults suggest that the beneficial effects of curcumin on rhabdomyolysis-mediated oxidative stress are related to both a reduction in ROS production and maintenance of the antioxidant defenses of the kidney.

Curcumin reduces renal ferroptosis, a key player involved in rhabdomyolysis-associated renal damage

Oxidative stress may induce tubular cell death (39–41), so we explored Mb-mediated cell death in MCTs and experimental rhabdomyolysis. Mb increased cell death (positive 7-AAD and annexin V staining) and this effect was reduced by curcumin in an HO-1–dependent way (Fig. 6A, B). We next performed TUNEL staining to analyze tubular cell death in our experimental model. As reported in Figs. 6C and 6D, we observed increased.

In our study, those mice with rhabdomyolysis that were pretreated with curcumin showed a lower kidney HO-1 protein and mRNA expression than nontreated mice with rhabdomyolysis. Similar results were obtained in tubular cells preincubated with curcumin and further exposed to Mb. These data may be associated with a rapid curcumin- mediated HO-1 induction to prevent further oxidative stress. Thus, our in vitro results show that curcumin in- duces a rapid HO-1 expression, thus increasing HO-1 levels before Mb-mediated HO-1 induction. In non– curcumin-treated cells, HO-1 levels at 3 h were lower than in curcumin-stimulated cells; however, the opposite trend was observed at 24 h. These results indicate a rapid curcumin-mediated HO-1 activation that may protect the kidney against Mb injury in early stages, which would to explain the reduced HO-1 levels observed in the experi- mental model, because, at later stages, the reduction in oxidative stress precludes the maintenance of high HO-1 levels.

Inflammation is involved in rhabdomyolysis- induced AKI (34), and recent evidences show a key role of the TLR4/NF-kB pathway in this background (59, 60). Our results confirm a high kidney TLR4 ex- pression and NF-kB activation in rhabdomyolysis both in vitro and in vivo. In this line, we observed that Mb induced the expression and secretion of proin- flammatory cytokines and chemokines, including CCL-2, IL-6, and TNF-a. However, this proinflammatory re- sponse was inhibited by curcumin. The precise molecular mechanism involved in the anti-inflammatory role of curcumin is not well characterized, but it is likely related to inhibition of TLR4-homodimerization, as previously in- dicated in Youn et al. (36). Komada et al. (61) recently re- ported that the NLRP3 inflammasome pathway mediates rhabdomyolysis-induced AKI, and curcumin has been proposed to inhibit NLRP3 inflammasome activation and IL-1b production (62). For that reason, we examined whether the NLRP3 inflammasome pathway was in- volved in curcumin renoprotective effects. However, we did not observe augmentation of inflammasome compo- nents in mice with rhabdomyolysis (unpublished results). This disparity of results may be explained because in the study of Komada et al., the authors only observed aug- mentation of NLRP3 and ASC protein levels 72 h after rhabdomyolysis induction (61), whereas our study was performed 24 h after glycerol administration.

In summary, our study demonstrates for the first time that ferroptosis, a programmed cell death de- pendent on iron, plays a key role in renal damage as- sociated with rhabdomyolysis both in a mouse model of rhabdomyolysis-induced AKI and in Mb-stimulated tubular cells. Moreover, we show the beneficial effects of curcumin in renal damage associated with rhabdo- myolysis, decreasing functional and structural injury, lipid peroxidation, inflammation, endothelial damage, and tubular cell death, specifically ferroptosis. Finally, we identified HO-1 as a key pathway involved in the protective effects of curcumin, whereas NF-kB and ERK are deactivated by this antioxidant molecule. Therefore, curcumin may be a potential therapeutic approach for Ferrostatin-1 patients with rhabdomyolysis.