Estradiol-mediated small GTP-binding protein GDP dissociation stimulator induction contributes to sex differences in resilience to ferroptosis in takotsubo syndrome Authors: Ti Wang, Ting Xiong, Yuxue Yang, Xiwei Chen, Ziwei Ma, Bangyun Zuo, Dong Ning, Beibei Zhou, Ruilong Song, Xuesong Liu, Daxin Wang Published: 2023 ABSTRACT AbstractBackgroundDeclining beneficial cardiovascular actions of estradiol (E2) have been associated with disproportionate susceptibility to takotsubo syndrome (TTS) in postmenopausal women. However, the underlying mechanisms between E2 and this marked disproportion remain unclear. SmgGDS (small GTP-binding protein GDP dissociation stimulator), as a key modulator of cardiovascular disease, plays protective roles in reducing oxidative stress and exerts pleiotropic effects of statins. Whether SmgGDS levels are influenced by E2 status and the effect of SmgGDS on sex differences in TTS are poorly understood.MethodsClinical data were reviewed from TTS inpatients. Echocardiography, immunofluorescence, and immunohistochemistry were performed together with expression analysis to uncover phenotypic and mechanism changes in sex differences in TTS-like wild-type (WT) and SmgGDS± mice. HL-1 cardiomyocytes were used to further examine and validate molecular mechanisms.ResultsIn 14 TTS inpatients, TTS had a higher incidence in postmenopausal women as compared to premenopausal women and men. In murine TTS, female WT mice exhibited higher cardiac SmgGDS levels than male WT mice. Ovariectomy reduced SmgGDS expression in female WT mice similar to that observed in male mice, whereas E2 replacement in these ovariectomized (OVX) female mice reversed this effect. The physiological importance of this sex-specific E2-mediated SmgGDS response is underscored by the disparity in cardiac adaptation to isoproterenol (ISO) stimulation between both sexes of WT mice. E2-mediated SmgGDS induction conferred female protection against TTS-like acute cardiac injury involving ferritinophagy-mediated ferroptosis. No such cardioprotection was observed in male WT mice and OVX female. A causal role for SmgGDS in this sex-specific cardioprotective adaptation was indicated, inasmuch as SmgGDS deficiency abolished E2-modulated cardioprotection against ferritinophagy and aggravates TTS progression in both sexes. Consistently, knockdown of SmgGDS in HL-1 cardiomyocytes exacerbated ferroptosis in a ferritinophagy-dependent manner and abrogated the protective role of E2 against ferritinophagy. Mechanistically, our findings revealed that SmgGDS regulated E2-dependent cardioprotective effects via AMPK/mTOR signaling pathway. SmgGDS deficiency abolished E2-conferred protection against ferritinophagy through activating AMPK/mTOR pathway, while treatment with recombinant SmgGDS in HL-1 cells significantly mitigated this pathway-associated ferritinophagy activity.ConclusionsThese results demonstrate that SmgGDS is a central mediator of E2-conferred female cardioprotection against ferritinophagy-mediated ferroptosis in TTS.Keywords: Takotsubo cardiomyopathy, Hormones, Sex, Ferroptosis, Ferritinophagy With an estimated all-cause death rate of 5.6 % per patient year and incidence of major adverse cardiovascular events of 9.9 % per patient year, takotsubo syndrome (TTS) is now diagnosed with increased frequency and known to be an important cause of acute cardiac emergencies, rather than a benign self-limiting condition as previously believed [1,2]. Typified by nonobstructed coronary arteries, TTS is an acute cardiac syndrome with features of typical apical dyskinesia and basal hyperkinesis that lead to myocardial stunning and acute heart failure (HF) [3,4]. To date, with the absence of evidence-based clinical trials or specific guidelines to define the optimal management of TTS, therapies known to be effective for acute coronary artery disease and HF are frequently adopted to treat TTS [5]. Epidemiological studies yield consistent results for the large sex disparity in TTS, showing that most patients (80%–90 %) are women in the sixth decade of life, and the risk for women who develop TTS after the age of 55 years increases 5-fold [[6], [7], [8]]. Although men clearly account for the minority of reported cases, it has been reported that males with TTS have more serious myocardial injury and a higher incidence of complications and death in the acute period than their female counterparts [[9], [10], [11]]. The pathophysiologic mechanisms of this susceptibility and seemingly paradoxical observation in patients may be explain by the delicate balance between resting sympathetic tone and microvascular function [12]. The precise reason for sex differences in TTS remains unknown, but it is believed that estrogen deprivation occurring postmenopause is proposed as an important mechanism involved in sympathetically mediated microvascular ischemia in TTS. Animal models of TTS have largely reproduced the cardioprotection against epinephrine-induced stress of females [13]. In rats, ovariectomy decreased myocardial contraction but improved upon estradiol (E2) replacement, suggesting that female sex conferred cardiac protection at least in part E2-dependent during stress. However, few preclinical studies have simultaneously investigated both sexes for an in-depth understanding of sex-dependent molecular mechanisms underlying the response to cardiac injury, particularly during the acute phase of TTS. A central difference in myocardial response to metabolic insult between sexes is that compared to males, females show beneficial cardiometabolic plasticity not solely as the consequence of the difference in circulating E2 levels, but rather the profoundly different interaction between cardiomyocytes and E2 [14]. Considering the metabolic and energetic impairment in the acute period of TTS [15], it can be hypothesized that differences in pathways mediating cardiac metabolism might be central in explaining the sex dimorphism of TTS. Since a line of evidence reveals a close interaction between E2 and iron metabolism, intact ovarian function has been shown to confer female resistance to ferroptosis in kidneys [16,17]. On the other hand, ferritinophagy-mediated ferroptosis activation has been shown in acute isoproterenol (ISO)-stressed cardiomyocytes [18], which has partly initiated the idea that E2-mediated ferroptosis response in a ferritinophagy-dependent manner might be an important contributor to sex-specific differences in ISO-induced acute takotsubo hearts. This ferroptosis activation in hearts might be relevant to multiple signaling pathways such as AMP-activated protein kinase (AMPK)-dependent, p53-dependent, and Nrf2-dependent pathway [19,20]. Most notably, AMPK-mammalian target of Rapamycin (mTOR) pathway, as one of the most signaling pathways mediated autophagy initiation with energy stress, is causally linked to the activation of ferritinophagy [20,21]. Whether AMPK/mTOR pathway-regulated ferritinophagy is involved in the progression of TTS remains to be further revealed. Small GTP-binding protein GDP dissociation stimulator (SmgGDS) is a guanine nucleotide exchange factor that interacts with the C-terminal polybasic region of small GTPases between Ras and Rho families, promoting the GDP/GTP exchange reactions [22]. The intracellular metabolic roles of SmgGDS have been demonstrated in protecting aortic smooth muscle cells from angiotensin II-induced nicotinamide adenine dinucleotide phosphate oxidase activation and ROS production [23]. In addition, both intracellular and extracellular SmgGDS work as an important molecular switch for the pleiotropic effects of statins through inhibition of Rac1 [24]. What is more, we have demonstrated that atorvastatin exhibits protective effect on ISO-mediated HF and remodeling through mitigation of ferritinophagy-mediated ferroptosis [25]. Thus, as evidence suggesting the interaction between E2 and statins and the similarities in actions and benefits becomes available [26], we hypothesize that SmgGDS works as a key molecule for the cardioprotective effects of E2. In the present study, we first demonstrate that sex-specific E2-mediated SmgGDS induction to acute ISO exposure confers female cardioprotection against ferritinophagy-mediated ferroptosis. We further reveal that SmgGDS deficiency abolishes this E2-modulated protection and aggravates both sexes of ferritinophagy activation, while treatment with recombinant SmgGDS significantly attenuates the progress of ferritin degradation in ISO-stressed cardiomyocytes. In addition, we identify that the AMPK-mTOR axis is a critical signaling pathway for achieving SmgGDS in mediating female sex against ferritinophagy. This work sets SmgGDS as a central mediator of sex-specific differences in cardiac adaptation and TTS susceptibility. Methods Data availability The expanded methods are described in the Supplemental Material. All original data that support the results of this study are available from the corresponding author upon reasonable request. Study participants Our retrospective study screened 14 patients diagnosed with TTS from four major third level grade A hospitals (Taizhou people's Hospital, Clinical Medical College of Yangzhou University, Fujian Provincial Hospital and Chinese People's Liberation Army General Hospital) between December 2014 and December 2022. Each patient underwent a careful assessment of history and physical examination. The presentation of patients mimicked acute coronary syndrome with acute-onset chest pain, transient ST elevation, and mild increase of cardiac enzymes. After exclusion of coronary artery diseases, myocardial infarction and myocarditis, 14 patients were studied in this investigation. The study was approved by the Clinical Medical College of Yangzhou University (NO: 2022ky331). Animals Male and female C57BL/6J mice were purchased from Comparative Medicine Center of Yangzhou University (Jiangsu, China). SmgGDS ± mice were purchased from GemPharmatech (Jiangsu, China) and were backcrossed with C57BL/6J mice. Given the perinatal lethality of global loss of SmgGDS (SmgGDS−/−) in mice [27], SmgGDS ± mice were used in the present study. The genotypes of all mice were identified by using primers specific for the SmgGDS gene (RAP1GDS1) (5′-AGTATTCTCTGTGACGTGTGTGTTCAG-3′ and 5′-CAATTCTACTGCCTTCCCTCTGTC-3′) and wild-type (WT) gene (5′-CATTCCTGTCTTCAGAGGTTGGAAA-3′ and 5′-GAGTGGACAGGACACAGTATAATCCTAG-3′). Littermate SmgGDS+/+ were used as controls. Then agarose gel electrophoresis was used for identification. The identification results are presented in Supplementary Fig. S1E. Mice were 16–18 weeks old at the commencement of experimental modeling of TTS. All animal experiments were performed in accordance with experimental protocols approved by the Institutional Animal Care and Use Committee of the Yangzhou University Medical College based on the Animal Research (NO: YXYLL-2022-58). Ovariectomy and E2 treatment Bilateral ovariectomy (OVX) or sham operation was performed at 8–10 weeks of age after anesthesia with isoflurane. One week after OVX surgery, E2 (5μg/mouse) dissolved in 0.1 mL corn oil or 0.1 mL corn oil was injected subcutaneously twice per week for 4 weeks, as described previously. Mice were submitted to establish the model of TTS 1 week after the end of the E2 treatment. TTS model To establish the TTS model, mice underwent intraperitoneal injection of ISO (400 mg/kg) for 2 h, as described previously [15,28]. 2 h after ISO injection, cardiac function was evaluated by transthoracic echocardiography. Subsequently, the mice were sacrificed by cervical dislocation under anesthesia and the whole heart tissue was harvested immediately for further study. Echocardiography The experimental mice were anaesthetized with 1.5 % isoflurane and echocardiography was performed using a VEVO 2100 ultrasound imaging system (Visual Sonics, Canada) at 2 h post-ISO or saline. An optimal parasternal long axis view was acquired to trace the morphological changes of left ventricular (LV) apical myocardium. Then the probe was rotated 90° to obtain the parasternal short-axis view for evaluating LV function. The echocardiographic parameters including ejection fraction% and fraction shortening% were calculated for three cycles on average. Immunofluorescence staining Mouse cardiac tissues of both sexes after ISO or saline injection were first fixed, then apical segments were embedded in paraffin and cut into 5 μm thick slices. HL-1 cardiomyocytes in 24-well plates were washed by cold PBS and fixed with 4 % paraformaldehyde after ISO or PBS treatment. The tissues and cells were incubated overnight with primary anti-SmgGDS (Santa Cruz Biotechnology, sc-390,003, 1:50) and α-Actinin (Cell Signaling Technology, 3134, 1:25) at 4 °C. Secondary antibodies were used for 2 h and were all from Boster (Goat anti-Rabbit IgG, FITC, BA1105, 1:100; Goat anti-Mouse IgG, DyLight 594, BA1141, 1:500). The immunofluorescence images were viewed using fluorescence confocal microscope (Leica, TCSSP8STED, Germany). Sections or cells incubated with only fluorescent conjugated secondary antibodies were used as negative controls. Histological analysis Murine hearts were fixed in 4 % paraformaldehyde overnight and embedded in paraffin. 5 μm sections of the heart transversely sectioned from apical segment were stained with hematoxylin–eosin (H&E). For immunohistochemistry, the heart sections were incubated with primary antibodies, including CD45 (Servicebio, GB11066, 1:1000), MCP-1 (Servicebio, GB11199, 1:500), PTGS2 (ABclonal, A1253, 1:50) and FTH1 (ABclonal, A1144, 1:50) overnight at 4 °C, and then followed by incubation with the immunohistochemistry commercial kit (ZSGB-Bio, PV-9001). Finally, the sections stained with DAB Kit (ZSGB-BIO, ZLI-9018) before counterstaining with hematoxylin. As a negative control, sections were incubated with species-matched biotinylated IgG instead of the primary antibodies. Prussian blue staining Cardiac sections (5 μm-thick) were dewaxed and soaked in Perls Stain (Solarbio, G1422) for 25 min, washed for 2 min. Dye cell nucleus with the nuclear solid red dye solution for 5 min and washed. The slices were routinely dehydrated and sealed in neutral gum. The accumulation of iron was detected by the light microscope (Olympu, Japan). Generation of SmgGDS-knockdown cells by lentiviral transfection To knock down the SmgGDS gene, HL-1 cardiomyocytes were transfected with lentiviruses containing shRNA targeting mouse SmgGDS (sh-SmgGDS) or negative control shRNA (sh-NC) according to the manufacturer's protocol for 48 h. Subsequently, stable transformants were selected in medium containing 4 μg/mL puromycin for 2 days. After 2–3 passages in the presence of maintained puromycin concentration, HL-1 cardiomyocytes were verified by Western blot and used for further experiments. The verification of transfection efficiency is shown in Supplementary Fig. S1F. Cell culture and treatment HL-1 cardiomyocytes, which can reflect the biological hallmarks of adult cardiomyocytes [15], were purchased from Procell (Wuhan, China) and grown in DMEM with phenol red supplemented with 10 % fetal bovine serum (FBS) at 37 °C in a 5 % CO2 humidified incubator. The culture medium was changed every day. HL-1 cardiomyocytes were starved and pretreated with different concentrations (1, 5, and 10 mM) of ISO for 2 h to induce an in vitro TTS model as described previously [15]. HL-1 cells were pretreated with recombinant SmgGDS protein (1 μg/mL) for 24 h or Fer-1 (10 μM) for 30 min or 3-MA (5 μM) for 12 h followed by ISO stimulation or co-treated with ISO and MHY1485 (10 μM). For E2 treatment, HL-1 cells were starved with 2.5 % charcoal stripped FBS (c.s. FBS) in supplemented DMEM without phenol red, subsequently incubated with E2 (1 μmol/L) in phenol-red free DMEM with 10 % c. s. FBS for 24 h. Treatment with ICI 182780 (1 μmol/L), an ER-inhibitor, was performed with E2 treatment for 24 h before ISO stimulation. After the treatment with each drug, HL-1 cells were harvested for staining, and gene and proteins expression were investigated. Detection of ROS The HL-1 cardiomyocytes were loaded with DCFH-DA fluorescence probe (Beyotime Biotechnology, S0033S, 5 μM) at 37 °C for 30 min to detect the level of ROS. The evaluation of lipid ROS production in vitro was performed by BODIPY™581/591C11 (Thermo Scientific, D3861). HL-1 cardiomyocytes were washed with PBS and incubated with BODIPY fluorescence probe (10 μM) for 30 min at 37 °C according to the manufacturer's instructions. Fluorescence images were acquired with the fluorescence microscope (ZEISS, Germany) and quantified by ImageJ. Measurement of malondialdehyde (MDA) in cardiac tissue and HL-1 cardiomyocytes The levels of MDA in the hearts and HL-1 cells were detected by colorimetric determination using a Lipid Peroxidation MDA Assay Kit (Beyotime, S0131S) according to the manufacturer's protocols. Samples were measured the absorbance at OD 532 nm. Detection of Intracellular Fe2+ To detect the level of cellular Fe [2]+, HL-1 cardiomyocytes were treated with ISO (10 mM) for 2 h, then washed with PBS and incubated with FerroOrange (Dōjindo Laboratories, F374, 1 μM) dispersed in serum-free medium for 30 min at 37 °C in 5 % CO2. Fluorescence images were obtained on a ZEISS AXIOVERT A1 Microscope. The relative fluorescence intensity was analyzed by ImageJ. Western blot analysis Total proteins were extracted from heart tissues and HL-1 cardiomyocytes. A total of 25 μg of protein was loaded on SDS-PAGE gel and transferred to 0.22 μm polyvinylidene difluoride (PVDF) membrane. The membranes were blocked with 5 % skim milk in tris-buffered saline containing 0.1 % Tween20 (TBST) for 2 h at room temperature, then immunoblotted with the primary antibodies at 4 °C overnight, following incubating for 2 h at 37 °C with horseradish-peroxidase-conjugated goat anti-mouse (Biodragon, BF03001, 1:5000) or anti-rabbit IgG antibody (Biodragon, BF03008, 1:5000). The regions containing proteins were captured by the enhanced chemiluminescence system (Clinx, China) and quantified by ImageJ Software. Real-time polymerase chain reaction Total RNA was isolated from the left ventricles using Trizol kit (Tiangen, DP424, China). Equal amounts of RNA (1000 ng) were converted to cDNA using HiScript Q RT SuperMix kit (Vazyme, R122-01, China), and quantitative real-time polymerase chain reaction (RT-PCR) on the Applied Biosystems (USA) was performed using SYBR Fast qPCR Mix kit (ABclonal, RK21203, China). The mRNA levels of target genes were normalized to murine GAPDH. The primer sequences used in this study were shown in Supplementary Table S1. Estradiol assay Serum E2 levels were measured by radioimmunoassay with a commercially ELISA kit (Elabscience, E-OSEL-M0008) with a sensitivity of 1.88 pg/mL according to the manufacturer's instructions. Statistical analysis Statistical analysis was performed with IBM SPSS 26.0 software. Clinical data were presented as means ± standard error of mean (SEM). In preclinical experiments, results were shown as mean ± standard deviation by using GraphPad Prism 9.0 software. For comparisons between 2 groups, an unpaired Student's 2-tailed t-test was performed. One-way ANOVA (followed by Tukey post hoc test) was performed to compare a single variable in multiple groups, and 2-way ANOVA (followed by Dunnett post hoc test) was used for 2-factor analysis. The survival rates were assessed using Kaplan-Meier analysis. A value of P < 0.05 was considered statistically significant. Results Sex-specific E2-Mediated SmgGDS response in TTS Of the 14 enrolled patients, all were women. The mean age was 63 ± 5. Prior cardiovascular history was uneventful, except for 10 patients with cardiovascular-related risk factors. All patients presented sudden onset of symptoms mimicking an acute coronary syndrome, including acute onset of chest pain and even cardiogenic shock, and 8 of them had electrocardiographic changes with ST-elevation or T-wave inversion. Laboratory results suggested that troponin T and NT-proBNP levels were elevated (troponin T: 2.3 ± 0.9 ng/mL; NT-proBNP: 5093 ± 1584 pg/mL). Meanwhile, the ejection fraction was substantially reduced at admission (46.9 ± 9.7 %). Preceding the TTS, eight patients (57.1 %) stated they experienced a stressful incident, of which 3 (37.5 %) was emotional stressor versus 5 (62.5 %) physical stressor (Table S2). Overall, except for two young women, the rest were post-menopausal, suggesting a large sex disparity in TTS predisposition and potentially indicating the protection of female-specific hormones. To explore the underlying mechanisms of female hormonal effects on sex differences in TTS and whether SmgGDS is involved, we first confirmed the expression of SmgGDS in mouse hearts and HL-1 cardiomyocytes by immunofluorescence and colocalized with cardiomyocyte marker α-actinin (Fig. S1A). It suggested that SmgGDS expressed in heart and HL-1 cardiomyocytes. Furthermore, we examined the SmgGDS expression in ISO-induced TTS mice with different gender, and so did the ISO-treated HL-1 cardiomyocytes. After ISO treatment, immunofluorescence staining of the cardiac tissue revealed that SmgGDS was decreased in male mice compared with hearts from saline-treated mice, however, SmgGDS was increased in female mice (Fig. 1A). A similar decrease in cardiac SmgGDS protein and mRNA levels was observed in ISO-induced TTS male samples but an increase in female samples (Fig. 1B and C), suggesting a potential role of SmgGDS in sex-dependent myocardial responses. Meanwhile, SmgGDS protein was gradually decreased in ISO-treated HL-1 cardiomyocytes compared with control HL-1 cells in a concentration-dependent manner, which could steadily reduce SmgGDS expression at the dose of 10 mM (Fig. S1B and Fig. 1A). Considering the importance of E2 deprivation occurring postmenopause in mediating sex-dependent responses in TTS, we next assessed the effect of E2 on myocardial SmgGDS expression following acute ISO treatment. We observed that depletion of endogenous E2 by ovariectomy resulted in an attenuated SmgGDS similar to that observed in male mice after ISO stimulation, whereas E2 pretreatment reversed SmgGDS levels in heart tissue in OVX females following ISO administration (Fig. 1D and E). Additionally, repletion of E2 significantly reduced cardiac injury in ovariectomized (OVX) female mice compared to their untreated counterparts after ISO treatment (Fig. 1F). These results suggest that SmgGDS expression was regulated in E2-dependent manner in TTS. SmgGDS expression is mediated by sex-related E2 in TTS. A, Immunofluorescence staining of SmgGDS in both male and female hearts injected with saline (mControl) or isoproterenol (ISO) (mTTS) for 2 h, and in HL-1 cardiomyocytes challenged by PBS (cPBS) or ISO (cTTS) for 2 h. B, Western blot analysis of SmgGDS protein in the hearts of both sexes injected with saline or ISO for 2 h. Males, n = 10; Females, n = 8. C, The mRNA levels of SmgGDS in cardiac tissues of male and female mice. Males, n = 5; Females, n = 5. D, Western blot analysis of SmgGDS protein in the hearts of sham-operated and ovariectomized (OVX) female mice treated with saline or ISO for 2 h n = 5, each. E, Western blot analysis of SmgGDS expression in the hearts of saline or ISO-injected OVX female mice pretreated with or without E2. n = 5, each. F, Representative images of hematoxylin-eosin staining in the heart tissues of placebo-pretreated and 17β-estradiol-pretreated OVX female mice following saline or ISO injection for 2 h. Scale bars, 20 μm ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01, ***P < 0.001 comparing saline vs ISO treatment. SmgGDS deficiency abolishes the sex-specific protection from acute cardiac injury in TTS mice To determine whether this sex-specific SmgGDS response gives rise to sex disparity of physiological features, cardiac function after 2 h of saline or ISO injection was assessed in age-matched WT (SmgGDS+/+) mice and littermate SmgGDS+/- mice in both sexes. Analysis of echocardiography revealed that acute treatment with ISO caused apparent typical apical akinesis but no LV dysfunction and even enhanced LV function after 2 h in female WT mice compared with saline-injected mice (Fig. 2A and B). On the contrary, ISO-treated male mice showed apparent typical apical akinesis and developed lower ejection fraction% and fraction shortening% than those in saline-injected male group (Figs. S2A and S2B). However, this sex difference in cardiac function after ISO stimulation was abolished in SmgGDS ± mice with alterations in severer LV dysfunction in both sexes, although the magnitude of cardiac function was still more worsen in male SmgGDS ± mice (Fig. 2A and B, and Figs. S2A and S2B). In addition, following ISO exposure, female SmgGDS ± mice showed less protection in that mortality and histological cardiac injury than in female WT mice with TTS (Fig. 2C and D); this aggravation of cardiac injury was similar to that observed in male SmgGDS ± mice with ISO injection (Figs. S2C and S2D). Among the evaluated inflammatory infiltration, ISO-treated WT females had no changes in immunohistochemical staining of CD45 and RT-PCR analyses of tumor necrosis factor α (TNFα) expression at the apex of the heart which were significantly evident in SmgGDS ± females after ISO treatment (Fig. 2D and E), equalizing the sex differences in monocyte chemoattractant protein 1 (MCP-1) and TNFα responses between WT males and WT females (Figs. S2D and S2E). The inflammatory markers of MCP-1 and IL (interleukin)-6 were increased in both ISO-injected WT sexes, and the expression of them was further higher upon SmgGDS deletion in both male and female SmgGDS ± hearts (Fig. 2D and E, and Figs. S2D and S2E). These findings suggest that this sex-specific effects on myocardial protection after ISO exposure are, at least in part, SmgGDS-dependent. Therefore, the sex-specific SmgGDS response in TTS leads to less acute cardiac injury in female mice than in male mice. SmgGDS knockdown exacerbates TTS in female mice. A, Representative echocardiographic images from female WT and SmgGDS± mice following saline or ISO injection. B, Echocardiographic parameters including left ventricle ejection fraction% and fraction shortening% in female WT and SmgGDS± mice after saline or ISO injection. Saline, WT, n = 6; Saline, SmgGDS+/−, n = 5; ISO, WT, n = 6; ISO, SmgGDS+/−, n = 5. C, The 2-h survival curve in female WT and SmgGDS± mice injected with saline or ISO. n = 17, each. D, Representative images of hematoxylin-eosin staining (upper), and immunohistochemistry staining for CD45 (middle) and monocyte chemoattractant protein 1 (MCP-1) (lower) in the hearts of female WT and SmgGDS± mice injected with saline or ISO for 2 h, and CD45 and MCP-1 levels in each group was quantified. n = 4, each. Scale bars, 100 μm and 50 μm in hematoxylin-eosin staining, and 20 μm in CD45 and MCP-1 immunohistochemistry. E, The mRNA levels of TNF-α and IL-6 in myocardial tissues of each group. n = 4, each. ###P < 0.001; *P < 0.05, ***P < 0.001 comparing saline vs ISO treatment. SmgGDS deficiency abolishes the sex-specific protection from ferritinophagy-mediated ferroptosis in TTS mice As apoptosis and pyroptosis are predominantly attributable for acute cardiac injury, we examined the myocardial expression of apoptosis-related proteins (Bcl-2, Bax and cleaved caspase-3)) and pyroptosis-related proteins (NLRP3 and cleaved N-terminal GSDMD) in cardiac tissues of both sexes via Western blot analysis. ISO treatment markedly decreased the expression of Bcl-2 in both sexes of TTS mice compared with their WT counterparts. However, the expression of Bax did not change in female WT mice after ISO stimulation. Notably, in the male ISO group, levels of both Bax and cleaved caspase-3 even decreased. In addition, ISO administration did not alter NLRP3 and cleaved N-terminal GSDMD levels (Figs. S3A and S3B). These results imply that sex-specific effects on myocardial protection after ISO exposure are not mediated by cardiomyocyte apoptosis and pyroptosis. Considering the essential role of ovarian hormone in female ferroptosis resilience [17], we then examined the expression of several ferroptosis biomarkers to further elucidate the potential mechanism by the sex disparity of acute cardiac injury in TTS. We first observed that the protein expression of cardiac PTGS2 remained unaltered while the mRNA level increased nor were SLC7A11 and GPX4 protein expression reduced, and even an increase in GPX4 occurred, in female WT mice after ISO stimulation (Fig. 3A, B and 3E). However, we observed an increase in PTGS2 expression and a decrease in SLC7A11 and GPX4 levels, which are indicative of ferroptosis activation, in the male ISO group (Figs. S3C, S3D and S3G). More importantly, knockdown of SmgGDS attenuated the female protection against ferroptosis in mouse hearts. Increased expressions of PTGS2 and decreased expressions of SLC7A11 and GPX4 were exhibited in female SmgGDS ± mice after ISO treatment (Fig. 3A, B and 3E), although with greater changes in magnitude in male SmgGDS ± mice (Figs. S3C, S3D and S3G). These alterations were accompanied by an exacerbated accumulation of iron, as indicated by Prussian blue staining, in both sexes of ISO-induced SmgGDS ± animals compared with ISO-treated WT mice. Notably, iron deposition was only observed in male WT TTS mice, aligning with the aforementioned results (Fig. 3C and Figs. S3E). We also measured the cardiac level of MDA generated during lipid peroxidation, a key process driving ferroptosis. In the present study, MDA was found no significant difference in both female and male WT group, with or without ISO injection, while knockdown of SmgGDS significantly aggravated MDA accumulation in mouse hearts in both sexes ISO groups compared with their control mice (Fig. 3D and Fig. S3F). Thus, our present findings indicate that SmgGDS deficiency attenuates the protective effect observed in females against TTS-like acute cardiac injury, involving ferroptosis. SmgGDS knockdown exacerbates ferritinophagy-mediated ferroptosis in female mice. A, The mRNA level of PTGS2 in cardiac tissues of each group. n = 4, each. B, Representative images of immunohistochemistry staining for PTGS2 and FTH1 protein in cardiac tissues of each group and quantification. Scale bars, 20 μm. C, Representative images of Prussian blue staining for iron deposition in the hearts of each group and quantification. n = 4, each. Scale bars, 50 μm. D, The cardiac MDA level in female WT and SmgGDS± mice after 2 h of saline or ISO injection. n = 3, each. E, Western blot analysis of PTGS2, P62, SLC7A11, FTH1, GPX4 and LC3B-II/LC3B–I expression in female WT and SmgGDS± mice after saline or ISO treatmen. n = 4, each. ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01, ***P < 0.001 comparing saline vs ISO treatment. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) In light of the critical role of autophagic ferritin degradation, known as ferritinophagy, in contributing to cardiac ferroptosis and subsequent heart failure [29,30], we next tested whether the female sex against ferroptosis directly related to SmgGDS is mediated in a ferritinophagy-dependent manner. Immunohistochemistry staining did not show decreased expression of cardiac ferritin-H (FTH1), and even an increase in FTH1 protein expression evidenced by Western blot, in female WT hearts after ISO treatment compared with saline-injected control mice (Fig. 3B and E). However, both results of immunohistochemistry staining and Western blot showed significantly reduced FTH1 in male WT hearts (Figs. S3D and S3G). In addition, knockdown of SmgGDS remarkably reduced FTH1 expression in both sexes but, again, at a lower level in males (Fig. 3B and E, and Figs. S3D and S3G). Consistent with this, while ISO exposure had no marked effect on autophagy activation in female WT mice, and even increased P62 protein expression, more dramatic upregulation of LC3B-II/LC3B–I ratio and downregulation of P62 were in male WT mice following ISO injection than those of the WT controls (Fig. 3E and Fig. S3G). Similarly, SmgGDS deficiency abolished the female protection against autophagy and led to a higher LC3B-II/LC3B–I ratio and lower P62 expression in both sexes of TTS mice compared with their WT counterparts (Fig. 3E and Fig. S3G). Overall, these results demonstrate that SmgGDS deficiency diminishes the female protection against ferroptosis in a ferritinophagy-dependent manner in TTS. SmgGDS deficiency abolishes E2-Mediated protection from ferritinophagy in TTS mice To investigate the importance of female sex hormone E2 in mediating SmgGDS-dependent cardiac protection in female mice, we next assessed the effect of E2 on cardiac function in sham-operated and OVX female SmgGDS ± mice and their littermates. In female WT mice, saline-injected OVX mice did not show signs of cardiac dysfunction. However, ovariectomy resulted in pronounced apical akinesis and a rapid decrease in cardiac function after ISO treatment (Fig. 4A and B). Importantly, improved LV functional was observed in ISO-induced WT hearts pretreated with E2 after OVX (Fig. 4B), whereas knockdown of SmgGDS abolished the protective effect of E2 pretreatment on myocardial function accompanied by typical apical akinesis (Fig. 4C and D), suggesting the critical role of SmgGDS in mediating E2-initiated acute cardiac protection in TTS. Furthermore, no death occurred in WT control mice after OVX, whereas the mortality rate in the ISO-treated OVX group was 90 % (Fig. 4E). When comparing the placebo-treated and E2-treated OVX mice following ISO administration, repletion of E2 significantly blunted ISO-induced increased mortality, but this protective role of E2 was lost in OVX SmgGDS ± mice (Fig. 4E). Consistently, knockdown of SmgGDS abolishes E2-induced protection for cardiac injury during the acute phase of TTS, whereas improved histological cardiac injury was observed in OVX hearts pretreated with E2 (Fig. 4F). Collectively, these data indicate that E2 confers female protection against acute cardiac injury and SmgGDS as a central mediator of this E2-mediated sex disparity in takotsubo hearts. SmgGDS knockdown abolishes the anti-ferritinophagy effect of E2 in TTS. A, Representative echocardiographic images from OVX female WT mice pretreated with or without E2 following acute saline or ISO injection. B, Echocardiographic parameters including left ventricle ejection fraction% and fraction shortening% in OVX WT mice pretreated with or without E2 following saline or ISO injection. Saline, OVX, n = 4; Saline, OVX + E2, n = 4; ISO, OVX, n = 4; ISO, OVX + E2, n = 5. C, Representative echocardiographic images from E2-pretreated OVX female WT and SmgGDS± mice following ISO administration. D, Echocardiographic parameters including left ventricle ejection fraction% and fraction shortening% in E2-pretreated OVX WT and SmgGDS± mice after ISO injection. n = 5, each. E, The survival curve in OVX WT mice pretreated with or without E2 following ISO injection and SmgGDS± mice following ISO injection. n = 10, each. F, Representative images of hematoxylin-eosin staining in the hearts of E2-treated OVX WT and SmgGDS± mice after injected with ISO. Scale bars, 20 μm. G, Western blot analysis of PTGS2, P62, SLC7A11, FTH1, GPX4 and LC3B-II/LC3B–I protein in hearts of sham-operated and OVX female mice treated with saline or ISO. n = 4, each. H, Western blot analysis of P62, FTH1 and LC3B-II/LC3B–I expression in the hearts of E2-pretreated OVX WT and SmgGDS± mice after ISO treatment. n = 5, each. ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01, ***P < 0.001 comparing saline vs ISO treatment. To further identify the contribution of E2 in female protection against ferritinophagy, we next investigated the effect of TTS-like acute cardiac injury in OVX mice. Compared with sham-operated female mice, OVX mice developed pronounced ferritinophagy activation in the ISO-induced TTS model as indicated by an increased PTGS2 expression and LC3B-II/LC3B–I ratio, and decreased SLC7A11, GPX4, P62, FTH1 expression (Fig. 4G). More importantly, SmgGDS deficiency attenuated the effect of E2 pretreatment on myocardial ferritinophagy in OVX mice during the acute phase of TTS compared with their WT counterparts (Fig. 4H). These results indicate the importance of SmgGDS in mediating E2-initiated acute cardiac protection from ferritinophagy in vivo. SmgGDS deficiency attenuates E2-modulated protection against ferritinophagy activation in ISO-induced HL-1 cardiomyocytes Based on SmgGDS as a central mediator in E2-mediated cardiac iron homeostasis in TTS mice, we then down-regulated SmgGDS in HL-1 cardiomyocytes by transfection with lentiviruses containing shRNA targeting mouse SmgGDS (sh-SmgGDS) or negative control shRNA (sh-NC) to decipher the intracellular metabolic roles of SmgGDS. In line with the in vivo findings, knockdown of SmgGDS remarkably augmented the level of MDA in ISO-treated HL-1 cardiomyocytes (Fig. 5A). Most importantly, we found more remarkable reduction of SLC7A11 and GPX4 expression in sh-SmgGDS HL-1 cells after ISO treatment (Fig. 5B). Likewise, LC3B-II/LC3B–I ratio was significantly higher while the levels of FTH1 and P62 were significantly lower in sh-SmgGDS HL-1 cells than in sh-NC HL-1 cells after ISO administration (Fig. 5B). Because of the crucial effect of Fe2+-mediated lipid peroxidation on ferroptosis, we next detected intracellular Fe2+ in ISO-induced HL-1 cardiomyocytes. A larger amount of Fe2+ was shown in sh-SmgGDS HL-1 cardiomyocytes stained with FerroOrange than that of control cells after ISO treatment (Fig. 5C). Furthermore, both ferrostatin-1 (Fer-1), a ferroptosis-specific inhibitor, and 3-methyladenine (3-MA), an autophagy inhibitor, reversed elevated ROS levels determined by DCFH-DA staining in sh-SmgGDS HL-1 cells under ISO administration (Fig. S4A). In addition, increased lipid ROS induced by ISO was also attenuated in sh-SmgGDS HL-1 cells by Fer-1 and 3-MA (Fig. 5D). These in vitro observations are consistent with the results in vivo that SmgGDS deficiency aggravated ferritinophagy-mediated myocardial ferroptosis in ISO-induced HL-1 cardiomyocytes. Next, we evaluated the extracellular metabolic role of SmgGDS in TTS to further clarify the potential effect of SmgGDS. After pretreatment with recombinant SmgGDS protein for 24 h in HL-1 cells, extracellular SmgGDS significantly mitigated the increased ROS production analyzed by DCFH-DA staining under ISO treatment (Fig. 5E). Moreover, consistent with the lower ROS status, pretreated with recombinant SmgGDS protein in HL-1 cells attenuated higher levels of lipid ROS induced by ISO (Fig. S4B). For further verification, the results were evidenced at protein levels. After pretreatment with recombinant SmgGDS protein, the downregulation of SLC7A11 and GPX4 expression and the upregulation of PTGS2 expression were reversed in ISO-induced HL-1 cardiomyocytes (Fig. 5F). Similarly, increased LC3B-II/LC3B–I ratio and decreased FTH1 and P62 were mitigated by recombinant SmgGDS protein under ISO administration (Fig. 5F). Overall, these results indicate that both intracellular and extracellular SmgGDS play cardioprotective roles against ferroptosis in a ferritinophagy-dependent manner. SmgGDS modulates E2-mediated anti-ferritinophagy in ISO-induced HL-1 cardiomyocytes. A, MDA in sh-NC and sh-SmgGDS HL-1 cardiomyocytes treated with PBS or ISO (10 mM) for 2 h n = 3, each. B, Western blot analysis of P62, SLC7A11, FTH1, GPX4 and LC3B-II/LC3B–I in sh-NC and sh-SmgGDS HL-1 cardiomyocytes after treatment with PBS or ISO (10 mM) for 2 h n = 3, each. C, Representative images of FerroOrange staining for intracellular Fe2+ in HL-1 cardiomyocytes of each group and quantification. n = 4, each. D, Representative images of intracellular lipid ROS by BODIPY probe in PBS or ISO-challenged sh-NC and sh-SmgGDS HL-1 cardiomyocytes pretreated with or without Fer-1 (10 μM) or 3-MA (5 μM); the fluorescence intensity in each group was quantified by ImageJ. n = 4, each. E, Representative images of DCFH-DA staining in PBS or ISO-challenged HL-1 cardiomyocytes pretreated with or without glutathione S-transferase (GST)-tagged recombinant SmgGDS protein (GST-SmgGDS, 1 μg/mL); the fluorescence intensity in each group was quantified by ImageJ. n = 4, each. F, Western blot analysis of PTGS2, SLC7A11, GPX4, FTH1, P62 and LC3B-II/LC3B–I expression in HL-1 cardiomyocytes pretreated with or without GST-SmgGDS (1 μg/mL) for 24 h followed by ISO stimulation. n = 3, each. G, Western blot analysis of SmgGDS, PTGS2, P62, SLC7A11, FTH1, GPX4 and LC3B-II/LC3B–I expression in ISO-challenged sh-NC and sh-SmgGDS HL-1 cardiomyocytes pretreated with or without E2. n = 4, each. ns, P ≥ 0.05; #P < 0.05, ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01, ***P < 0.001 comparing PBS vs ISO treatment. Additionally, to further determine whether E2 exerts its anti-ferroptosis effects via SmgGDS in vitro, both sh-NC and sh-SmgGDS HL-1 cardiomyocytes were pretreated with E2 for 24 h before ISO stimulation. We found that pretreatment with E2 markedly augmented SmgGDS levels and conferred protection from ferritinophagy in HL-1 cardiomyocytes exposed to ISO, while these effects were blocked by co-pretreatment with the nonselective ER antagonist ICI 182780 (Fig. S4C). More importantly, knockdown of SmgGDS attenuated the protective role of E2 against ferritinophagy, indicating that E2 modulated the process of ferroptosis, at least in part, via SmgGDS in cardiomyocytes (Fig. 5G). SmgGDS deficiency abolishes E2-modulated protection against ferritinophagy activation through AMPK/mTOR signaling pathway Recent studies have shown that the AMPK-mTOR pathway is actively involved in autophagy-dependent ferroptosis and the pathogenesis of acute cardiac injury [31,32]. Here, to elucidate the molecular mechanisms for E2-modulated protection from ferritinophagy via SmgGDS, we first assessed the profiles of AMPK/mTOR pathway in TTS mice. Interestingly, ISO-injected SmgGDS ± male mice revealed a marked increase in phosphorylated AMPK Thr172 and a decrease in phosphorylated mTOR S2448 compared with ISO-injected WT male mice (Fig. 6A). Next, we detected the possible effect of AMPK/mTOR signaling pathway in ISO-administrated HL-1 cardiomyocytes. Consistently, we observed more phosphorylated AMPK Thr172, as well as less phosphorylated mTOR S2448 in sh-SmgGDS HL-1 cells compared with sh-NC HL-1 cells after ISO treatment (Fig. 6B). Furthermore, extracellular SmgGDS significantly diminished ISO-induced phosphorylated AMPK Thr172 increase and phosphorylated mTOR S2448 decrease in HL-1 cardiomyocytes (Fig. 6C). In addition, we found that upregulated AMPK phosphorylation and downregulated mTOR phosphorylation under ISO administration showed a significant tendency to ameliorate in sh-NC HL-1 cardiomyocytes after E2 pretreatment, whereas there was no marked improvement in E2-pretreated sh-SmgGDS HL-1 cells (Fig. 6D). These results suggested that E2 regulated AMPK-mTOR pathway through SmgGDS in cardiomyocytes. SmgGDS regulates E2-mediated anti-ferroptosis through AMPK/mTOR pathway. A, Western blot analysis of p-mTOR, mTOR, p-AMPK, and AMPK expression in WT and SmgGDS± mice treated with saline or ISO for 2 h n = 3, each. B, Western blot analysis of p-mTOR, mTOR, p-AMPK, and AMPK in sh-NC and sh-SmgGDS HL-1 cardiomyocytes challenged by PBS or ISO for 2 h. C, Western blot analysis of p-mTOR, mTOR, p-AMPK, and AMPK in HL-1 cardiomyocytes pretreated with or without GST-SmgGDS (1 μg/mL) for 24 h followed by ISO stimulation. n = 3, each. D, Western blot analysis of p-mTOR, mTOR, p-AMPK, and AMPK in ISO-challenged sh-NC and sh-SmgGDS HL-1 cardiomyocytes pretreated with or without E2. n = 3, each. E, Western blot analysis of P62, FTH1, and LC3B-II/LC3B–I in ISO-challenged sh-NC and sh-SmgGDS HL-1 cardiomyocytes co-treated with or without MHY1485. n = 3, each. ns, P ≥ 0.05; #P < 0.05, ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01, ***P < 0.001 vs without ISO treatment. Finally, given SmgGDS diminished the activation of ferritinophagy and AMPK/mTOR pathway, we aimed at confirming the relationship between ferritinophagy and AMPK/mTOR signaling pathway in vitro. In the present study, we found a reduced LC3B-II/LC3B–I ratio and increased expressions of FTH1 and P62 when the AMPK/mTOR pathway was blocked by the mTOR activator, MHY1485, in ISO-treated sh-SmgGDS HL-1 cardiomyocytes (Fig. 6E). Taken together, these results suggest that SmgGDS in mediating E2-modulated myocardial ferritinophagy resistance could be achieved via AMPK/mTOR pathway in ISO-induced TTS. Discussion The results of present study are the first to our knowledge to demonstrate that SmgGDS knockdown abolishes E2-conferred female protection against acute cardiac injury in TTS and exacerbates TTS development, which involves dysfunction of HL-1 cardiomyocytes and enhanced ferroptosis activity in a ferritinophagy-dependent manner. Here, we confirm a marked sex difference in the prevalence of TTS in the study patients. This was reproduced in our preclinical TTS model which showed striking cardioprotection in intact female mice compared with OVX female and male mice. The cardioprotection was attributable to E2-mediated SmgGDS response, as protection was lost in (1) OVX female and male mice in vivo, (2) SmgGDS ± mice, (3) E2-pretreated SmgGDS ± OVX mice, (4) downregulated SmgGDS in HL-1 cardiomyocytes, and (5) E2-pretreated HL-1 cardiomyocytes in the presence of SmgGDS knockdown, while pretreatment with E2 replicated cardiac protection in a SmgGDS-dependent manner. Moreover, resistance to cardiac ferritinophagy-mediated ferroptosis in female as opposed to male mice was attributable to increased SmgGDS expression modulated by E2. Upregulation of SmgGDS and improvement of ferritinophagy activation were replicated in ISO-induced HL-1 cells pretreated with E2, further consolidating the link between SmgGDS and female sex-exerted anti-ferritinophagy effects. Taken together, these findings underscore SmgGDS as a key mediator of E2-conferred anti-ferritinophagy protection from TTS. A consistent observation in all series of TTS has been the marked sex difference in prevalence, with 90 % women versus 10 % men. Indeed, approximately 90 % of the cases are postmenopausal women with a mean age of 67–70 years, although patients can occur in any age group [7,11]. The results from our clinical data show a higher prevalence of TTS in postmenopausal women, which is consistent with the epidemiological status, consolidating a marked sex difference in the prevalence of TTS. Although the precise reason of the marked disproportion remains unknown, it is believed that the declining beneficial effects of E2 on cardiovascular function predisposes postmenopausal women to TTS [12]. This effect was replicated in a murine TTS model with female mice showing protection against cardiac injury through E2-mediated SmgGDS response. After ISO treatment, female mice showed increased SmgGDS expression, while male mice demonstrated less cardiac SmgGDS expression than the control group. Removal of the ovaries, the major endogenous source of E2, significantly decreased SmgGDS levels, while E2 supplement in OVX female mice reversed this effect. Thus, the sex-specific SmgGDS response following ISO administration is mediated by E2. The physiological importance of this sex disparity is underscored in cardiac functional adaptation after ISO stimulation in intact female mice, which exhibited reserved cardiac function, but not male mice, which demonstrated reduced cardiac function. However, knockdown of SmgGDS abolished this sexual dimorphism, and aggravated cardiac dysfunction and mortality in both sexes, accompanied by increased cardiomyocyte necrosis and inflammatory cells infiltration, indicating that protective cardiac adaptation in intact female mice was causally associated with SmgGDS expression. These results suggest that the sex-specific E2-mediated SmgGDS response resulted in sex disparity of myocardial responses to acute injury. More recently, female resilience against ferroptosis in kidneys through ovarian-dependent mechanisms is subsequently highlighted in a model of renal injury and repair [17,33]. Here, we identified a marked sex-specific protection against ferroptosis in female hearts compared with male hearts that was evident as attenuated iron deposition and ferroptosis activation, even increased anti-ferroptosis responses, although no significant differences in MDA levels were detected under control and TTS conditions. The latter finding highlighted that SmgGDS deficiency blunted the female resilience against ferroptosis, showing that the hearts of both female and male SmgGDS ± mice suffered from more severe ferroptosis after ISO treatment. This sex-specific SmgGDS-conferred female resistance to ferroptosis may explain cardioprotective adaptation observed in female mice. Since activated ferritinophagy was found to mediate ferroptosis in diverse models of cardiovascular diseases [18,29,30], we probed whether this sex-specific ferroptosis response to pathological stress is in a ferritinophagy-dependent manner. As a result, ferroptosis resilience in female mice was predominantly attributable to anti-ferritinophagy in ISO-stressed TTS, yet this protection is abrogated by SmgGDS deficiency, showing that ferritin degradation and autophagy activation were exacerbated in both male and female SmgGDS ± hearts. Taken together, our findings demonstrate a central role of SmgGDS in the female-specific cardiac adaptation against ferritinophagy-mediated ferroptosis in TTS. Since the sex disparity in SmgGDS expression mediated by E2 has been identified in our TTS model, further inspecting the role of E2 in the relationship between SmgGDS and ferritinophagy-mediated cardiac dysfunction is needed. Based on previously published works, the dose of E2 and the regimen (twice per week) selected for this study have been certified to better mimic the physiological state [[34], [35], [36]]. The physiological importance of E2 is reflected in the findings that ovariectomy deteriorated cardiac function and increased mortality in OVX female mice with TTS, which exhibited marked reduction of SmgGDS, but E2 replacement reversed these effects in OVX animals, which demonstrated relatively higher post-ISO SmgGDS expression. While SmgGDS deficiency abrogated E2-modulated cardioprotection, and worsened heart function and the progression of TTS. Thus, E2-mediated SmgGDS induction during the acute phase of TTS resulted in cardioprotective adaptation in intact female mice but not in OVX female and male mice. The association of E2 with ferroptosis activation in TTS remains unclear. As a major endogenous hormone of ovaries, E2 has been considered a protective factor against ferroptosis, ensuring cellular redox balance [16]. Moreover, exogenous/endogenous estrogen has been suggested to relieve excess iron-related disorders in extragonadal tissues [37,38]. Ovariectomy compromises female resilience to ferroptosis and ferroptotic stress in mouse kidneys and worsens kidney function [17]. Consistently, our data show that female mice are resistant to ferroptosis at the acute stage of TTS, while ovariectomy promotes ferroptosis in a ferritinophagy-dependent manner, leading to compromised compensatory anti-ferritinophagy in female mice. Furthermore, in line with the proven cardioprotective effects of E2 on acute cardiac injury abrogated by SmgGDS deficiency, knockdown of SmgGDS induced enhancement in cardiac ferritinophagy activity in E2-treated OVX mice. These results suggest SmgGDS plays an important role in E2-exerted anti-ferritinophagy effects and TTS pathogenesis. Although these findings point to the importance of SmgGDS in E2-derived acute cardiac protection from ferritinophagy, the in vivo phenotype of the global SmgGDS ± mouse cannot be attributed to a particular type of cardiac cell. This cardioprotection of SmgGDS was therefore further consolidated in sh-SmgGDS HL-1 cardiomyocytes. In vitro, sh-SmgGDS HL-1 cells suffered more severe ferritinophagy-mediated ferroptosis compared with sh-NC HL-1 cells under ISO stimulation. In light of the secretory property of SmgGDS, the present study also provides insights into the extracellular role of SmgGDS that underlies ferritin degradation in cardiomyocytes. Indeed, treatment with recombinant SmgGDS to upregulate extracellular SmgGDS has been demonstrated to reduce intracellular Rac1 expression and ROS levels in cardiac fibroblasts, which may partly explain the pleiotropic effects of statins on cardiac hypertrophy and fibrosis [24]. Consistently, our findings indicated that recombinant SmgGDS reduced ISO-induced lipid peroxidation status and ferritinophagy-mediated ferroptosis in HL-1 cardiomyocytes, comparable to those of Fer-1 and 3-MA in ISO-stressed sh-SmgGDS HL-1 cells, showing that both intracellular and extracellular SmgGDS could protect against ferritinophagy. Further inspection of the role of E2 in vitro showed that pretreatment with E2 upregulated SmgGDS expression and conferred protection from ferritinophagy in ISO-treated HL-1 cells, while the protection of E2 was attenuated by SmgGDS deficiency. This suggests that the protective role of E2 is, at least in part, attributable to increased SmgGDS. In addition, the effects of E2 were inhibited by the ER antagonist ICI 182780. Thus, these data support that E2 increases SmgGDS expression by an ER-dependent mechanism, thereby attenuating ISO-induced ferritinophagy activity. The specific mechanism by which E2 mediates SmgGDS induction in the acute phase of TTS is a subject of future investigation. Given that AMPK, which functions as a pivotal energy metabolic switch, governs various physiological and pathological cellular events in response to hypoxia, ischemia and oxidative stress, AMPK phosphorylation has been shown to regulate cell ferroptosis through inhibiting mTOR signaling to block system xc− activity [39,40]. Moreover, the study carried out by Nef et al. in takotsubo cardiomyopathy patients further proved that mTOR activation might protect cardiomyocytes from death as an adaptive mechanism [41]. Following this possibility, we focused on the role of AMPK-mTOR to elucidate the potential pathway for the knockdown of SmgGDS-induced iron metabolism disorder. In the current study, we found SmgGDS ± mice showed reduced expression of phosphorylated mTOR S2448 and increased expression of phosphorylated AMPK Thr172 after ISO stimulation. Similarly, phosphorylated AMPK Thr172 was increased and phosphorylated mTOR S2448 was reduced in sh-SmgGDS HL-1 cardiomyocytes compared with sh-NC HL-1 cells under ISO treatment. Furthermore, the effects of ISO-induced phosphorylated AMPK Thr172 increase and phosphorylated mTOR S2448 reduction were mitigated by recombinant SmgGDS in HL-1 cells. These results implicate that both intracellular and extracellular SmgGDS may work for cardioprotective effects through AMPK/mTOR pathway in TTS. More importantly, we found that E2 decreased AMPK phosphorylation and increased mTOR phosphorylation, which were abolished by SmgGDS deficiency, consistent with previous findings that loss of mTOR function disturbed E2-mediated adaptive cellular responses of female cardiomyocytes. Hence, our findings demonstrate that SmgGDS modulates the protective effects of E2 through AMPK/mTOR pathway. Additionally, one of the modulating pathways of autophagy activation leading to ferroptosis is the AMPK/mTOR signaling pathway [21,42], in agreement with a previous study reported that AMPK and mTOR promoted uncontrolled lipid peroxidation as well as oxidized lipids involved in autophagy‐dependent ferroptosis [43,44]. Moreover, it has also been reported that ferritinophagy activation participates in response to the zinc oxide nanoparticles-induced vascular injury in an AMPK-dependent manner [21,45]. In this study, we presented evidence that SmgGDS knockdown exacerbated ISO-induced ferritinophagy in HL-1 cardiomyocytes, while this increased ferritinophagy activation was attenuated when AMPK-mTOR pathway was blocked by MHY1485. Taken together, these results suggest that SmgGDS regulates E2-dependent cardioprotective effects against ferritinophagy via AMPK/mTOR pathway in the development of TTS. Although our data strongly suggest that SmgGDS has a central role in sex-specific E2-modulated cardioprotection against ferritinophagy in TTS, the present study has several limitations. Compared with the used TTS model in male mice demonstrating remarkable cardiac dysfunction at the acute stage of TTS, surviving female mice 2 h post-ISO show enhanced cardiac function instead. Since the pathological state of the current study does not include the recovery phase of TTS, when patients also have persistent structural, functional, and metabolic changes after TTS [46,47], it would be interesting to know whether this enhanced cardiac adaptation is a key point for female resilience against acute stress and also for the future surviving state; additional studies to determine whether initial E2-mediated SmgGDS induction is a transient adaptative response or even persists in the subsequent course of the disease are needed. In addition, our studies focus on estradiol withdrawal post-ovariectomy in OVX mice. However, there are not only a series of other hormonal changes after ovariectomy, such as luteinizing hormone and follicle-stimulating hormone, but also non-hormonal changes, such as increased sympathetic nerve activity, renin activity and elevated angiotensin II levels [[48], [49], [50]]. Therefore, we cannot exclude other factors from being involved in the development of TTS. It remains to be further explored whether these factors mediate the changes of SmgGDS and subsequent cardioprotective mechanisms. For the mechanism study, our research focuses on the pro-ferritinophagy effect of SmgGDS deficiency through AMPK/mTOR pathway in TTS, while more detailed molecular mechanisms by which E2 influences SmgGDS expression have not yet been identified. It will be interesting to uncover whether E2 mediates SmgGDS induction by binding to specific estrogen receptors (ERs), such as ERα, ERβ, and GPER, or even by triggering a cascade of genomic pathways upon binding to ERs. Similarly, specific dependent pathways for ferritin degradation have not been identified. Considering the critical role of nuclear receptor coactivator 4 (NCOA4), a cargo receptor that promotes iron release from ferritin storage, in the process of ferritinophagy triggering ferroptosis [51,52], further studies might be needed to determine whether the effect of SmgGDS deficiency on ferritinophagy is dependent or even independent of NCOA4 in TTS. Moreover, given that SmgGDS deficiency in these mice is not restricted to the cardiovascular tissues, cardiomyocyte-specific deletion of SmgGDS may provide information for further elucidating the roles of intracellular SmgGDS in cardiomyocytes. To summarize, we provide convincing evidence demonstrating that SmgGDS plays critical roles in E2-exerted anti-ferritinophagy effects during the acute phase of TTS. SmgGDS deficiency abolishes E2-modulated cardiac protection against ferritinophagy and aggravates the progress of TTS, while both intracellular and extracellular SmgGDS exert crucial cardioprotective effects against ferritinophagy through AMPK/mTOR pathway. This novel, sex-specific, hormone-mediated stress response suggests that E2-mediated SmgGDS induction is critical for young females against acute TTS. More importantly, SmgGDS downregulation mediated by declining E2 may represent a potential mechanism by which postmenopausal females may increase the sex gap in susceptibility to TTS. Pharmacologically targeting SmgGDS may represent a promising therapeutic strategy for the acute phase of TTS and the prevention of disease progression. Sources of funding This work was supported by grants from National Natural Science Foundation of China General Program (No.82170508) and Yangzhou University International Academic Exchange Fund. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments We thank the Taizhou people's Hospital, Clinical Medical College of Yangzhou University, Fujian Provincial Hospital and Chinese People's Liberation Army General Hospital for help with patient data. Nonstandard Abbreviations and Acronyms takotsubo syndrome heart failure estradiol isoproterenol AMP-activated protein kinase mammalian target of Rapamycin Small GTP-binding protein GDP dissociation stimulator ovariectomized monocyte chemoattractant protein 1 tumor necrosis factor α interleukin-6 ferritin-H estrogen receptors Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.redox.2023.102961.Clinical PerspectiveWhat Is New?•Sex-specific estradiol (E2)-mediated small GTP-binding protein GDP dissociation stimulator (SmgGDS) induction leads to the sex disparity of cardiac adaptation in the development of isoproterenol (ISO)-induced takotsubo syndrome (TTS).•SmgGDS is necessary for the cardioprotective effects of E2 against ferritinophagy-mediated ferroptosis in ISO-induced TTS.•SmgGDS deficiency abolished E2-exerted anti-ferritinophagy effects and aggravates ferritinophagy activation through AMPK/mTOR pathway.What Are the Clinical Implications?•SmgGDS downregulation mediated by declining E2 may represent a potential mechanism by which postmenopausal females may increase the sex gap in susceptibility to TTS.•SmgGDS may serve as a predictive biomarker for both the diagnose and prognosis of TTS.•SmgGDS would be a promising therapeutic target for acute TTS treatment and to prevent the progression of TTS. Clinical Perspective What Is New? •Sex-specific estradiol (E2)-mediated small GTP-binding protein GDP dissociation stimulator (SmgGDS) induction leads to the sex disparity of cardiac adaptation in the development of isoproterenol (ISO)-induced takotsubo syndrome (TTS). Sex-specific estradiol (E2)-mediated small GTP-binding protein GDP dissociation stimulator (SmgGDS) induction leads to the sex disparity of cardiac adaptation in the development of isoproterenol (ISO)-induced takotsubo syndrome (TTS). •SmgGDS is necessary for the cardioprotective effects of E2 against ferritinophagy-mediated ferroptosis in ISO-induced TTS. SmgGDS is necessary for the cardioprotective effects of E2 against ferritinophagy-mediated ferroptosis in ISO-induced TTS. •SmgGDS deficiency abolished E2-exerted anti-ferritinophagy effects and aggravates ferritinophagy activation through AMPK/mTOR pathway. SmgGDS deficiency abolished E2-exerted anti-ferritinophagy effects and aggravates ferritinophagy activation through AMPK/mTOR pathway. What Are the Clinical Implications? •SmgGDS downregulation mediated by declining E2 may represent a potential mechanism by which postmenopausal females may increase the sex gap in susceptibility to TTS. SmgGDS downregulation mediated by declining E2 may represent a potential mechanism by which postmenopausal females may increase the sex gap in susceptibility to TTS. •SmgGDS may serve as a predictive biomarker for both the diagnose and prognosis of TTS. SmgGDS may serve as a predictive biomarker for both the diagnose and prognosis of TTS. •SmgGDS would be a promising therapeutic target for acute TTS treatment and to prevent the progression of TTS. SmgGDS would be a promising therapeutic target for acute TTS treatment and to prevent the progression of TTS. Contributor Information Xuesong Liu, Email: seanxsl712@163.com. Daxin Wang, Email: daxinw2002@sina.com, daxinw2016@126.com. Supplementary data The following are the Supplementary data to this article. Multimedia component 1 Multimedia component 2 Data availability Data will be made available on request. References Associated Data Supplementary Materials Multimedia component 1 Multimedia component 2 Data Availability Statement The expanded methods are described in the Supplemental Material. All original data that support the results of this study are available from the corresponding author upon reasonable request. Data will be made available on request.